OnCo · weekly issue
OnCo weekly, 2026-11-17
1574 changelog entries, 4 regulatory events, 2 upcoming dates, 4062 corrections. Generated from the corpus; every item links to its page and, through it, to the primary source. Not medical advice.
What changed this week
New pages, new data and fixes, from the changelog.
- 17 November 2026: Second accuracy pass on the cancer pages: 31 cancers re-read claim by claim against the trials, labels and guidelines they cite, with about 2,700 corrections, among them approvals since withdrawn or granted, trials shown as awaited after they had reported, and survival gains that later analyses did not confirm; and the last thirteen care protocols
- 17 November 2026: Fourteen more care protocols, for side effects that had none: high blood pressure and protein in the urine on VEGF drugs, heart rhythm changes, bleeding on BTK inhibitors, raised liver enzymes, lung inflammation on antibody-drug conjugates, fluid retention, high blood sugar, low magnesium, high phosphate, thyroid changes, differentiation syndrome, capillary leak, tumour flare and constipation; 226 protocols in all
- 17 November 2026: Second accuracy pass on the original 39 country pages, among them the United States, China, India, Japan, Germany, Brazil and the United Kingdom: about 660 claims corrected after each was read against its source again
- 17 November 2026: Second accuracy pass on live country pages, last batch for waves 5 to 12: corrections to Austria, Serbia, Denmark, Finland, Norway, Ireland, Hungary, Belarus, Bulgaria and North Korea, each claim read against its source again
- 17 November 2026: Second accuracy pass on live country pages, continued: corrections to eighteen more pages, among them Haiti, Senegal, Slovenia, Mali, Papua New Guinea, Burkina Faso, Costa Rica, Paraguay, Belgium, Cuba, Czechia, Greece and Portugal, each claim read against its source again
- 17 November 2026: Second accuracy pass on live country pages: corrections to 53 pages after every claim was read against its source again, including what the IAEA radiotherapy directory lists for 36 countries, WHO's HPV vaccination data where pages had said there was none, and sources pages had called unreadable that do answer
- 17 November 2026: The first weekly gap fill read from FDA sources lands, each fact checked against the FDA list and device records: the Guardant360 Liquid CDx blood test approved on 19 May 2026 with its seven drug pairings, and four companion tests linked to the biomarkers they read
- 17 November 2026: More of the graph is joined up: 1,251 confirmed links between people and the trials whose papers they wrote, matched by ORCID or full name and affiliation; who pays for 21 tests in the US and UK, from Medicare national coverage decisions and NICE guidance; fifty warning-card lines now name the protocol that manages them; and eight wrong side-effect links are removed
- 17 November 2026: Twenty-two more care protocols close the gaps in the census, among them stoma care, drain care at home, transplant conditioning, veno-occlusive disease, lung biopsy, intrathecal chemotherapy in children, end-of-life care for children, advance care planning and stopping smoking after a diagnosis; 212 protocols in all, with a table of what is covered and what is still open
- 17 November 2026: Shasqi in depth: its clinical programme SQ3370, its trial and the click-activated chemistry behind it, from the company's own pages, the registry and the published papers; the weekly gap-fill job gets the tools it needs to read its allowed sources; and the command-line tool and MCP server are rebuilt to know about care protocols
- 17 November 2026: A contributor's education work arrives, 215 new records on biomarkers, immune tests, specimens, RNA and clinical evidence, with every source checked; and a backlog of corrections lands: new FDA approvals including pimicotinib, four generic icons removed from logos, mislabelled trials fixed, radiotherapy counts for Madagascar and Angola, the two Cuban lung cancer vaccines, and 33 records renamed for their subject
- 17 November 2026: Second accuracy pass on cancer pages: merge the newest main, and bring six one-line descriptions up to date with facts this pass verified (daraxonrasib approval, Claudin 18.2 treatments, sacituzumab govitecan first-line approvals)
- 17 November 2026: Second accuracy pass on the remaining care protocols: the fourteen written on 17 November, radiation safety, line care, procedures, enhanced recovery after surgery, follow-up and survivorship, each re-read against its guidelines and labels, with every correction logged
- 17 November 2026: Second accuracy pass on care protocols, batch twelve: the last 13 protocols re-read against their sources (1,273 claims, 54 wrong, 260 unsupported, 17 unverifiable), so all 226 now have a second reader; doses and figures added during the pass re-read independently in 50 more files, with one survey count corrected and missing scalp-cooling contraindications added; advice on when to report bleeding restored from Cancer Research UK
- 17 November 2026: Second accuracy pass on cancer pages, fourteen more high-burden cancers (blood cancers, oesophageal, kidney, endometrial, ovarian, melanoma, glioblastoma, myeloma and the skin cancer family): 2,189 claims re-read against their sources, 337 wrong, 162 unsupported and 116 unverifiable corrected or labelled
- 17 November 2026: Second accuracy pass on care protocols, batch eleven: 17 protocols on radiotherapy delivery and quality assurance, long-term follow-up, nutrition, lymphoedema and support re-read against their sources (1,682 claims, 61 wrong, 312 unsupported, 16 unverifiable)
- 17 November 2026: Second accuracy pass on care protocols, batch ten: 18 protocols on enhanced recovery after surgery, follow-up, fatigue, distress, consent and daily life re-read against their sources (1,745 claims, 64 wrong, 330 unsupported, 40 unverifiable)
- 17 November 2026: Second accuracy pass on cancer pages, ten high-burden cancers (lung types, two breast types, head and neck, stomach, thyroid, liver, bladder, cervical): 1,187 claims re-read against their sources, 244 wrong, 71 unsupported and 67 unverifiable corrected or labelled
- 17 November 2026: Second accuracy pass on care protocols, batch nine: 18 protocols on biopsies and procedures, fertility, breast follow-up and planning ahead re-read against their sources (1,771 claims, 48 wrong, 316 unsupported, 26 unverifiable); batch eight's true counts were 2,195 claims, 53 wrong, 459 unsupported and 24 unverifiable
- 17 November 2026: Second accuracy pass on care protocols, batch eight: 22 protocols on procedures, contrast, radiotherapy side effects, heart surveillance and infection precautions re-read against their sources (2,188 claims, 51 wrong, 461 unsupported, 23 unverifiable)
- 17 November 2026: Second accuracy pass on cancer pages, lymphoma: 1,850 claims re-read against their sources, 220 wrong, 140 unsupported and 53 unverifiable corrected or labelled
- 17 November 2026: Second accuracy pass on care protocols, batch seven: 19 radiation-safety and vascular access protocols re-read against their sources (1,939 claims, 59 wrong, 410 unsupported, 136 unverifiable); the independent dose re-read finished across 64 files, with one wrong dose corrected, plerixafor in kidney impairment
- 17 November 2026: Second accuracy pass on cancer pages, prostate: 2,322 claims re-read against their sources, 206 wrong, 117 unsupported and 40 unverifiable corrected or labelled
- 17 November 2026: Second accuracy pass on care protocols, batch six: the 14 side-effect protocols added on 17 November and the radiotherapy dose reference re-read against their sources (1,835 claims, 64 wrong, 407 unsupported, 12 unverifiable); doses added in the first half re-read independently in 21 files with none wrong; sourced advice on when to seek urgent help restored to five emergency protocols; four drug links no source supports removed
- 17 November 2026: Second accuracy pass on 104 care protocols, the emergencies, immunotherapy side effects, cell therapy and everything with drug doses first: 413 wrong and 1,623 overstated claims corrected against the guidelines and labels, among them the fluoropyrimidine dose table for DPYD carriers, the vaccine advice for patients, and steroid doses for a CAR-T side effect; newer guideline editions noted or built in
- 17 November 2026: Second accuracy pass on cancer pages, lung: 2,575 claims re-read against their sources, 385 wrong, 108 unsupported and 65 unverifiable corrected or labelled
- 17 November 2026: Second accuracy pass on care protocols, batch five: 23 transplant, paediatric, pharmacy and safety protocols re-read against their sources
- 17 November 2026: Second accuracy pass on cancer pages, colorectal: 2,741 claims re-read against their sources, 578 wrong, 132 unsupported and 45 unverifiable corrected or labelled
- 17 November 2026: Second accuracy pass on care protocols, batch four: 20 dosing, palliative and symptom protocols re-read against their sources
- 17 November 2026: Second accuracy pass on cancer pages, pancreatic: 1,908 claims re-read against their sources, 298 wrong, 78 unsupported and 47 unverifiable corrected or labelled
- 17 November 2026: Second accuracy pass on care protocols, batch three: 20 protocols on CAR T care, opioids, antiemetics, anticoagulation and antimicrobial prophylaxis
- 17 November 2026: Second accuracy pass on cancer pages, triple-negative breast cancer: 1,385 claims re-read against their sources, 198 wrong, 54 unsupported and 27 unverifiable corrected or labelled
- 17 November 2026: Second accuracy pass on care protocols, batch two: 20 immunotherapy and cell-therapy protocols re-read against their sources
- 17 November 2026: Trial records brought into line with the registry: 206 stale statuses refreshed, fourteen phases corrected, trials that shared a registration number separated, and a terminated trial no longer shown as active; and 785 institution records read against their sources again, with 22 corrections
- 17 November 2026: Institution records re-read against their sources, last batch: the records behind 43 more country pages, the log of all 24 corrections, and Belgium's note about its medicines agency, whose pages answered again
- 17 November 2026: Institution records re-read against their sources, second batch: the Newsweek hospitals in Europe, the census of 83 smaller countries, and the records behind 36 more country pages, among them Indonesia, Turkey, Australia, Poland, Pakistan, Nigeria, Colombia, Belgium and Sweden
- 17 November 2026: Second accuracy pass on cancer pages, gallbladder: 1,008 claims re-read against their sources, 148 wrong, 61 unsupported and 39 unverifiable corrected or labelled
- 17 November 2026: chore: draft change proposals that need review (#441)
- 17 November 2026: chore: EU regional rows read from the EMA register pages (#440)
- 17 November 2026: chore: draft records for gaps in the freshness issue, for review
- 16 November 2026: Fourteen more care protocols, for side effects that had none: high blood pressure and protein in the urine on VEGF drugs, heart rhythm changes, bleeding on BTK inhibitors, raised liver enzymes, lung inflammation on antibody-drug conjugates, fluid retention, high blood sugar, low magnesium, high phosphate, thyroid changes, differentiation syndrome, capillary leak, tumour flare and constipation; 226 protocols in all
- 16 November 2026: Second accuracy pass on care protocols, batch one: 21 emergency and cell-therapy protocols re-read against their sources
- 16 November 2026: Differentiation syndrome protocol: drop a field the protocol record does not have
- 16 November 2026: Trials whose registry status changed say so in their plain-language and Chinese lines, and the recruiting-site file follows the refreshed registry entries
- 16 November 2026: Care protocols for seven more side effects: high blood pressure and urine protein on VEGF-pathway inhibitors, QT prolongation, atrial fibrillation and bleeding on BTK inhibitors, differentiation syndrome, capillary leak syndrome, tumour flare, and constipation
- 16 November 2026: Institution records re-read against their sources, first batch: the Newsweek hospitals outside Europe and the records behind twenty country pages, among them Brazil, Spain, Mexico, the Philippines, Viet Nam, Ukraine, Iran and Taiwan; ten claims corrected
- 16 November 2026: Known faults in institution and country records corrected against their sources: IPO Coimbra's registry, Myanmar's screening programme, Guatemala's registry, Hamad's founding year and accreditation dates, Kuwait's cancer centre, unsourced funders for four countries, the countries trial count's note, and the FDA companion diagnostics link
- 16 November 2026: Plain-language lines and translations that repeated the faults: KontRASt-02's status, seven trial phases, and Francisco Berrospi Espinoza's post at INEN, whose head is now Wilder Calmet Berrocal
- 16 November 2026: Trial pages show only their own registration, and statuses and phases follow the registry as read on 17 November 2026: MOUNTAINEER, START-B and FOxTROT no longer carry another study's sites, the two SEQUOIA trials are two records, KontRASt-02 is terminated, and CAMBRIA names both of its registrations
- 16 November 2026: Care protocols for four metabolic side effects: high blood sugar on PI3K, AKT and mTOR inhibitors, low magnesium on cetuximab and panitumumab, high phosphate on FGFR inhibitors, and underactive thyroid on VEGFR-targeted kinase inhibitors
- 16 November 2026: Care protocols for three side effects that had none: raised liver enzymes on targeted drugs, interstitial lung disease on antibody-drug conjugates and targeted drugs, and fluid retention and pleural effusion on kinase inhibitors
- 16 November 2026: Second accuracy pass on the original 39 country pages, among them the United States, China, India, Japan, Germany, Brazil and the United Kingdom: about 660 claims corrected after each was read against its source again
- 16 November 2026: Equecabtagene autoleucel: one-line description shortened to fit, facts unchanged
- 16 November 2026: Second accuracy pass, the original country pages, batch four: Colombia, South Africa, Romania, Ethiopia, Algeria, Sweden, Switzerland, Malaysia, Israel and Taiwan
- 16 November 2026: Second accuracy pass, the original country pages, batch three: Vietnam, Thailand, Iran, Ukraine, Nigeria, the Philippines, Argentina, Egypt and the Netherlands
- 16 November 2026: Second accuracy pass, the original country pages, batch two: Canada, Spain, South Korea, Turkey, Australia, Poland, Mexico, Pakistan and Bangladesh
- 16 November 2026: Second accuracy pass, the original country pages, batch one: China, the United States, India, Japan, Brazil, Russia, Germany, Indonesia, the United Kingdom, Italy and France
- 16 November 2026: Second accuracy pass on live country pages, last batch for waves 5 to 12: corrections to Austria, Serbia, Denmark, Finland, Norway, Ireland, Hungary, Belarus, Bulgaria and North Korea, each claim read against its source again
- 16 November 2026: Second accuracy pass, last batch: Austria, Serbia, Denmark, Finland, Norway, Ireland and North Korea complete the forty pages of waves 5 to 8
- 16 November 2026: Second accuracy pass on live country pages, continued: corrections to eighteen more pages, among them Haiti, Senegal, Slovenia, Mali, Papua New Guinea, Burkina Faso, Costa Rica, Paraguay, Belgium, Cuba, Czechia, Greece and Portugal, each claim read against its source again
- 16 November 2026: CIMAvax-EGF: one-line description shortened to fit, facts unchanged
- 16 November 2026: Second accuracy pass, batch four: Belgium, Cuba, Czechia, Greece, Portugal, Hungary, Belarus and Bulgaria
- 16 November 2026: Second accuracy pass on the last eight live country pages (wave 11): Haiti, Senegal, Slovenia, Mali, Papua New Guinea, Burkina Faso, Costa Rica and Paraguay
- 16 November 2026: Second accuracy pass on live country pages: corrections to 53 pages after every claim was read against its source again, including what the IAEA radiotherapy directory lists for 36 countries, WHO's HPV vaccination data where pages had said there was none, and sources pages had called unreadable that do answer
- 16 November 2026: Casablanca cancer centre: one-line description shortened to fit, with the fuller wording kept in the summary
- 16 November 2026: Second accuracy pass, batch three: Venezuela, Sri Lanka, Syria, Kazakhstan, Cambodia, Ecuador, Zimbabwe, Chile and Tunisia, and HPV vaccination figures re-read on the earlier pages
- 16 November 2026: Second accuracy pass, batch two: Ghana, Saudi Arabia, Peru and Nepal, and the IAEA radiotherapy directory read for all sixteen pages checked so far
- 16 November 2026: Second accuracy pass on twelve live country pages: DR Congo, Tanzania, Kenya, Myanmar, Sudan, Uganda, Iraq, Afghanistan, Yemen, Morocco, Angola and Uzbekistan
- 16 November 2026: Second accuracy pass on twelve more live country pages (wave 10): Chad, Lebanon, Bolivia, Zambia, Puerto Rico, Uruguay, Lithuania, Niger, Cameroon, Georgia, Azerbaijan and Bosnia and Herzegovina
- 16 November 2026: Twenty country pages now say what the IAEA's radiotherapy directory lists, and two say what WHO's HPV data hold
- 16 November 2026: Second accuracy pass on ten more live country pages (wave 9): Singapore, New Zealand, Croatia, Slovakia, Mozambique, Malawi, Madagascar, the Dominican Republic, Guatemala and Côte d'Ivoire
- 16 November 2026: Second accuracy pass on ten live country pages (wave 12): Guinea, Honduras, Armenia, Somalia, El Salvador, Jordan, Rwanda, Moldova, Latvia and Panama
- 16 November 2026: The first weekly gap fill read from FDA sources lands, each fact checked against the FDA list and device records: the Guardant360 Liquid CDx blood test approved on 19 May 2026 with its seven drug pairings, and four companion tests linked to the biomarkers they read
- 16 November 2026: Review of the 17 November gap fill: Guardant360 Liquid CDx gets a row in the tests box of the eight readouts it measures, the BRACAnalysis CDx row on the prostate HRR readout says it covers BRCA1 and BRCA2 only, the FDA companion diagnostic list is cited at its new address, and the gap audit no longer reads the SP142 and SP263 PD-L1 assays as one product
- 16 November 2026: More of the graph is joined up: 1,251 confirmed links between people and the trials whose papers they wrote, matched by ORCID or full name and affiliation; who pays for 21 tests in the US and UK, from Medicare national coverage decisions and NICE guidance; fifty warning-card lines now name the protocol that manages them; and eight wrong side-effect links are removed
- 16 November 2026: Twenty-two more care protocols close the gaps in the census, among them stoma care, drain care at home, transplant conditioning, veno-occlusive disease, lung biopsy, intrathecal chemotherapy in children, end-of-life care for children, advance care planning and stopping smoking after a diagnosis; 212 protocols in all, with a table of what is covered and what is still open
- 16 November 2026: Shasqi in depth: its clinical programme SQ3370, its trial and the click-activated chemistry behind it, from the company's own pages, the registry and the published papers; the weekly gap-fill job gets the tools it needs to read its allowed sources; and the command-line tool and MCP server are rebuilt to know about care protocols
- 16 November 2026: Who pays for 21 tests, from Medicare national coverage determinations and NICE guidance, and 32 named gaps
- 16 November 2026: Author lists of the 818 papers that report a trial, read from Europe PMC, and who among them has a page here
- 16 November 2026: The National Cancer Institute and Cancer Research UK records carry their abbreviations as aliases
- 16 November 2026: Fifty warning-card lines name the protocol that manages them
- 16 November 2026: The scalp cooling record says in its first line what the cap is for, preventing or reducing hair loss, so a question about hair loss still finds it beside the longer care protocol
- 16 November 2026: The coverage table records that Europe PMC holds no open copy of the two interventional radiology standards on blood thinners before a biopsy, so that row stays open
- 16 November 2026: A contributor's education work arrives, 215 new records on biomarkers, immune tests, specimens, RNA and clinical evidence, with every source checked; and a backlog of corrections lands: new FDA approvals including pimicotinib, four generic icons removed from logos, mislabelled trials fixed, radiotherapy counts for Madagascar and Angola, the two Cuban lung cancer vaccines, and 33 records renamed for their subject
- 16 November 2026: Protocols page weight: the live page weighs 157,107 bytes, over the 155,000 ceiling, so the ceiling stays and the note says so
- 16 November 2026: The onco-cc and onco-mcp packages go to 0.1.1, so a rebuild that knows the protocol kind can be published
- 16 November 2026: The recorded Ask rubric is the figure measured after merging main: 0.6765
- 16 November 2026: The gap-fill workflow installs bubblewrap and socat, so the fetch tool can start inside the command sandbox
- 16 November 2026: Evidence keeps its scope: conjugate payload pages show each measurement with the exact chemical form it was made on and how it was tested, drug pages show what a conjugate is made of, assay approvals say who authorised what and for which use, and a missing measurement reads as a gap in our records and not as a no; a contributor's work, reviewed and put into plain language
- 16 November 2026: Shasqi in depth: its one clinical programme SQ3370 with the stopped phase 1/2a trial and its results, the click-activated prodrug technology, seven resolved papers, and a company record corrected to say the trial was terminated and carrying sourced leadership, financing and grants
- 16 November 2026: The plain-language summaries of the new care protocols and their upgrade ideas are tightened to the 400-character limit every record keeps
- 16 November 2026: The glossary size recorded for the Wikipedia-link gauge is brought up to the live count of 1,685 terms
- 16 November 2026: Giredestrant's United States row now records its approval; the two Cuban vaccines are filed as established, as nimotuzumab is, and their trial papers link to them; stale trial-site and country entries removed
- 16 November 2026: The two registrations of the PILDI colon surgery trial stay as two records for now, each saying they are one study: merging them adds a redirect line that waits for the owner's approval
- 16 November 2026: Corrections log: the sixteen faults fixed from the review backlog on 16 November 2026, each with its source
- 16 November 2026: Thirty-three paper records named as questions or how-sentences are renamed for their subject, so a record's name no longer competes with the questions readers ask; papers whose published title is itself a question keep it
- 16 November 2026: The cancer blood clot treatment protocol now links enoxaparin and dalteparin, the two heparins its own guideline names; twelve Newsweek-ranked hospitals whose sites still refuse a script stay unwritten and are named
- 16 November 2026: CIMAvax-EGF and racotumomab, Cuba's two registered lung cancer vaccines, get product pages written from the Cuban regulator's registration entries and the two randomised trials
- 16 November 2026: Nine records corrected from their sources: Angola's cancer institute no longer shows a paper's DOI as its website; the University of Puerto Rico cancer centre is filed under Puerto Rico; the technology tags on King Hussein Cancer Center and King Faisal Specialist Hospital each rest on a page of the hospital's own, and one that had none is removed; IPO Lisboa no longer claims a regional registry that the 2017 law folded into the national one; the Senegal and Nigeria breast cancer cohort's mean age is given for the 378 women it describes; and the Madagascar, Angola and Somalia pages carry the IAEA's radiotherapy counts and the Somali ministry's statement that the country has no medicines regulator yet
- 16 November 2026: ESPAC-4 no longer shows ESPAC-3's sites, sponsor and dates: the ISRCTN record lists a ClinicalTrials.gov number that belongs to ESPAC-3, the fetcher now refuses that cross-reference, and the Greek and Hungarian pages name ESPAC-3 as the trial that ran there; SIRveNIB's own registration (NCT01135056), publication and result join SARAH's on their shared record
- 16 November 2026: Four trial records corrected against ClinicalTrials.gov: a children's bladder-drug trial that was never a cancer trial is removed with its false links to childhood cancers and to the CIC gene; two Cushing's disease trials now say what they tested and why they sit under pituitary tumours; the PILDI colon surgery trial, held twice under two registration numbers, is one record; and the vulvar electrochemotherapy study carries the registry's own entry (no phase, status unknown, 20 participants) in place of phase 2 and recruiting
- 16 November 2026: Four records lose a web host's stock icon shown as their logo (Apmonia Therapeutics, Loki Therapeutics, Cansera, Japan Psycho-Oncology Society), and the logo fetcher now refuses any new file that is byte-for-byte a known generic icon: the WordPress and Vercel defaults, the HostGator mascot, the European Commission emblem and grey placeholders
- 16 November 2026: FDA approval notices are counted once each: the fetcher now identifies a notice by its date and title, so FDA's move of the notices to a new address no longer doubles the list, and the refreshed list brings the four approvals of 5 to 16 November 2026 (subcutaneous amivantamab in head and neck cancer, pimicotinib, durvalumab with enfortumab vedotin, giredestrant) onto their records with the FDA's own words
- 16 November 2026: Twenty-two new care protocols close gaps the census found, among them stoma care, drain care at home, transplant conditioning, veno-occlusive disease, children's central lines and end of life care, advance care planning and ReSPECT; three existing protocols gain hair care during scalp cooling and label dose rules for skin reactions
- 16 November 2026: A coverage table places all 251 subjects of the care protocol census against the protocols OnCo holds: 182 had their own record, 51 were covered inside another, and 18 were not covered
- 16 November 2026: Ask's retrieval measurement is recorded after the contributor batch: recall 0.4365 against a floor of 0.4293, floors unchanged
- 16 November 2026: Translated summaries for pathological complete response and radiotherapy follow their rewritten English text, and two Chinese summaries that no longer match are withdrawn
- 16 November 2026: chore: draft records for gaps in the freshness issue, for review
- 16 November 2026: Teaching records now cite the primary papers and guidance only: 22 links to the contributor's own teaching notes are removed, 66 placeholder source labels carry their paper's real title, and the short definition of biomarker returns to its plainer wording
- 16 November 2026: Payload, drug and test pages say what was measured in words a patient can follow
- 16 November 2026: The RAS pathway page keeps its plain explanation, and a sourced link no longer switches off the drug highlighting
- 16 November 2026: Teaching concepts across biology, tests, treatments and trial evidence get a contributor's audit: 137 new records and about 500 revised ones, each with its sources, and the corrections they record
- 16 November 2026: Weekly gap-fill automation, with the model's fetching limited to an allow-list of regulators, registries and societies
- 16 November 2026: Antibody-drug conjugate payloads get deeper coverage from a contributor: 34 new records and 31 supplemented ones that keep free drugs, attached linker-drugs and released forms apart, each measurement with its experiment and source
- 16 November 2026: Glossary links now respect capitals, so ordinary words such as platinum, select, open and decision no longer link to trials and acronyms that share their spelling, about twelve thousand false links fewer; with the weekly refresh of trial counts, universe lists, logos, the fact check and two EU register rows
- 16 November 2026: Names with a digit link in any case again, and the ordinary words "verify" and "alliance" stop linking
- 16 November 2026: Eighty more countries get deep dives researched by agents and checked claim by claim by a second agent, the menu shows one Countries entry instead of a line per country, Ask reads first any medicine a question names by name or brand, and the ivonescimab record now states that the FDA issued a complete response letter, announced on 15 November 2026
- 16 November 2026: Names in capitals of any length link only where written in capitals, and bracket qualifiers in record names stop linking ordinary words
- 16 November 2026: Nine source links in the immune-concepts round now name the paper instead of showing only a PubMed number
- 16 November 2026: Ivonescimab: the FDA issued a complete response letter on the US application, announced 15 November 2026
- 16 November 2026: Vedolizumab gets its EU approval row, read from the EMA register, and one RP-6306 trial loses a subtype link its text does not support
- 16 November 2026: chore: render this week's issue to public/newsletter/
- 16 November 2026: chore: weekly fact check, audit, and provenance refresh
- 16 November 2026: chore: refresh public/votes.json from Discussions reactions
- 16 November 2026: chore: propose this week's landmark oncology papers for review
- 16 November 2026: chore: refresh trial counts from ClinicalTrials.gov
- 16 November 2026: chore: draft change proposals that need review (#411)
- 16 November 2026: chore: EU regional rows read from the EMA register pages
- 15 November 2026: chore: draft change proposals that need review (#410)
- 14 November 2026: chore: draft change proposals that need review (#409)
- 13 November 2026: The search health check now keeps a bounded cache, so it needs under 1.5 GB of memory instead of nearly 3 GB and keeps passing as the corpus grows
- 13 November 2026: chore: draft change proposals that need review (#407)
- 12 November 2026: Search health audit holds a bounded word cache, so the gate fits in memory as the corpus grows
- 12 November 2026: 269 cancer institutions for under-covered countries, from national centres and radiotherapy units to registries, regulators and payers, and oncology hospital ranks checked against the published Newsweek 2026 list, which corrects several ranks and removes three that the list does not contain
- 12 November 2026: Logo gauge counts the new institutions
- 12 November 2026: Ask finds the brachytherapy page and trastuzumab deruxtecan again
- 12 November 2026: Country institutions stand on their own until the country deep dives arrive
- 12 November 2026: Menu: one Countries entry instead of a line per country deep dive
- 12 November 2026: Drug pages answer whether a medicine is available in your country, with approval, who pays and help with the cost, and country and institution pages list the trials recruiting there, counted only where that site's own registry status says recruiting
- 12 November 2026: chore: refresh pulse, abstract and citation snapshots (#405)
- 12 November 2026: chore: weekly roadmap registry check (public/roadmap-watch.json)
- 12 November 2026: chore: refresh institution research output from OpenAlex (#404)
- 12 November 2026: chore: draft change proposals that need review (#403)
- 11 November 2026: Newsweek 2026 oncology ranks on 268 institutions, with 75 new records for ranked hospitals OnCo did not hold
- 11 November 2026: Cancer institutions for the 83 countries that held fewer than three: national centres, radiotherapy centres, registries, regulators and payers
- 11 November 2026: Newsweek 2026 oncology ranks corrected against the published list, and unsourced superlatives removed from the ranked records
- 11 November 2026: Glossary links no longer fire on ordinary words such as has, cure, prior and bleeding, and the care protocols now list the supportive medicines their steps name, so each medicine's page shows the protocols that use it
- 11 November 2026: The care protocols now list the everyday medicines their own steps name, so 40 of the 43 medicines added in the November supportive care fill show the protocols that use them: blood thinners on the clot and endoscopy protocols, anti-sickness drugs, opioids, antibiotics, antivirals and the immune-toxicity rescue drugs
- 11 November 2026: Glossary acronyms of two to five capital letters now link only where the text writes them in capitals, so "has", "must", "fit" and "let" stop linking to HAS, MUST, FIT and LET
- 11 November 2026: Glossary links no longer fire on ordinary words: "has" stops linking to health technology assessment, and 57 other aliases that named something else leave the hover matcher or are qualified
- 11 November 2026: Forty-three everyday medicines the care protocols name now have records (blood thinners, anti-sickness drugs, pain relief, antibiotics and antivirals among them), with three line types and thirteen glossary terms; and three China approvals now rest on the makers' own exchange announcements, with two indications corrected
- 11 November 2026: The availability table says each payer once and keeps its rows short on a phone
- 11 November 2026: Drug pages say where a drug is available and country and institution pages list the trials recruiting there
- 11 November 2026: Three China approvals now rest on the makers' own exchange announcements, marked as reported by the company with the regulator notice not read
- 11 November 2026: Care protocols arrive: 190 sourced protocols for every part of cancer care, each step with its reason, source, grade and quote, best practice set beside named variants, linked from cancer, drug, device, side-effect, red-card and first-60-days pages; and every country in GLOBOCAN now has a page, with France's overseas departments on the France page
- 11 November 2026: Forty-three medicines used in cancer care, three vascular access devices and thirteen glossary terms join the corpus, each from its label or a guideline
- 11 November 2026: The scalp cooling protocol leaves the anagen effluvium term to the hair loss pages, so Ask still finds the scalp cooling record first for whether cooling prevents hair loss
- 11 November 2026: Endometrial cancer lists fertility preservation, as the fertility preservation record already names it, so the fertility protocol stays on its page
- 11 November 2026: The side-effects page shows how each symptom is managed: the care protocol for the symptom, and a treatment's own protocol beside the products it names
- 11 November 2026: First-60-days checklist items link the care protocol they are about: fertility referral, vaccinations, port or PICC care, scalp cooling, the DPD test, the temperature rule and follow-up
- 11 November 2026: Red cards for breast, bowel, lung, pancreatic, prostate, skin, lymphoma and triple-negative breast cancer link the protocol that manages each emergency
- 11 November 2026: Breast, bowel, ovarian, pancreatic, stomach and AML pages list the ports, PICC lines, chemotherapy pumps and DPYD testing their treatment uses, so port care and line care appear on them
- 11 November 2026: Protocols name the medicines, devices, glossary terms and upgrade ideas their steps are about, and the port protocol links a proposal to test masks and head-turning at port access
- 11 November 2026: A cancer page reaches a protocol through a treatment only where the protocol names that treatment as its own, and a steroid or other supportive medicine brings only supportive-care protocols
- 11 November 2026: The protocols index carries ten light rows, and weighs 148 KB rather than 464 KB
- 11 November 2026: Cancer pages list only the protocols that apply to them, and every protocol summary fits the short-summary limit
- 11 November 2026: The protocols index pages its table now that there are more than sixty, and a similar page no longer repeats a protocol the page already lists
- 11 November 2026: Two pairs of care protocols written in parallel now read as one each, and neighbouring protocols point at each other
- 11 November 2026: Eighteen supportive care and survivorship protocols: diarrhoea, nausea, skin, eyes, hair, fatigue, exercise, pain, lymphoedema, distress, sleep, sex, hot flushes and survivorship care plans
- 11 November 2026: Ask: a medicine the question names by name or brand is read first
- 11 November 2026: Three protocol-upgrade ideas from the pharmacy and vascular access protocols: carboplatin GFR estimate outcomes, vesicants through midlines, and central line discard volume
- 11 November 2026: Nineteen more care protocols for radiotherapy, radiation safety, biopsies and scans, and ten protocol upgrades
- 11 November 2026: Protocol for central line and port care in the community and at home
- 11 November 2026: Protocols for looking after yourself during treatment: bleeding precautions, cancer tablets at home, sex and contraception, skin and nails, sun, driving and travel, and diet and alcohol; two upgrade ideas
- 11 November 2026: Protocol grades show only what each source prints
- 11 November 2026: Protocol for dose banding of anticancer drugs
- 11 November 2026: Protocol for central line and port removal
- 11 November 2026: Protocol for midline catheters in cancer care
- 11 November 2026: Protocol for ambulatory chemotherapy pumps at home; insertion protocol no longer links the tip-location record
- 11 November 2026: Country deep dives: keep the merged waves inside their tests
- 11 November 2026: Protocols: blood thinners around surgery, anti-VEGF drugs before and after operations, sentinel node tracers and blue dye, and lowering the risk of arm lymphoedema
- 11 November 2026: Protocol: enhanced recovery after radical cystectomy, from the EAU 2026 bladder cancer guideline with the ERAS Society guideline and its 2025 surgeons' update
- 11 November 2026: Protocols: enhanced recovery after oesophagectomy, gastrectomy, Whipple, cytoreduction with HIPEC and head and neck free flap surgery; syringe drivers, pleural catheter drainage at home and malignant ascites
- 11 November 2026: Protocol for aseptic compounding of hazardous anticancer drugs in the pharmacy
- 11 November 2026: Protocols: enhanced recovery after colorectal, gynaecological cancer, lung and breast reconstruction surgery, each saying which ERAS edition could be read
- 11 November 2026: Protocol for hepatic dose adjustment of anticancer drugs, including the FDA's September 2026 draft guidance
- 11 November 2026: Two survivorship and assessment protocols: breast screening after chest radiotherapy at a young age, and geriatric assessment before systemic therapy
- 11 November 2026: Protocol for rapid drug desensitisation after an allergic reaction to a platinum, taxane or antibody
- 11 November 2026: Two more protocols: heart risk assessment and monitoring around heart-toxic cancer treatment, and nutrition screening and support
- 11 November 2026: Protocols for a blocked central line or port and for blood sampling from a central line
- 11 November 2026: Protocol for consent and education before anticancer treatment
- 11 November 2026: Four more supportive care protocols: G-CSF by febrile neutropenia risk, chemotherapy nerve damage, transfusion thresholds, and anaemia with ESAs and iron
- 11 November 2026: Protocol for inserting a PICC, tunnelled line or implanted port for anticancer treatment
- 11 November 2026: Fifteen more care protocols: eight oncological emergencies and seven steps in CAR T-cell and transplant care
- 11 November 2026: Protocol upgrade idea: start an antimuscarinic when dying is recognised to prevent the death rattle, on the SILENCE trial, linked from the last-days protocol
- 11 November 2026: Protocol for siting and checking a peripheral cannula for anticancer treatment, from the 2024 Japanese extravasation guideline, CDC, epic3 and UK network guidance
- 11 November 2026: Nineteen care protocols for preventing infection and for children with cancer, each step quoted from the current guideline
- 11 November 2026: Protocol: routine tests before planned surgery, from NICE NG45, by size of operation and fitness
- 11 November 2026: Protocols for body-size-based dose calculation (including full doses in obesity), drug, herbal and food interaction review, and medicines reconciliation
- 11 November 2026: Palliative sedation protocol: describe refractory suffering without the word the tone rules reserve
- 11 November 2026: Sixteen care protocols for radiotherapy and for scans, scopes and tests
- 11 November 2026: Three more supportive care protocols: fertility preservation before treatment, bone health on hormone therapy, and dental assessment before bone-modifying drugs
- 11 November 2026: Protocol: mouth care at home during chemotherapy and head and neck radiotherapy, from the MASCC/ISOO mucositis guideline and the NCI patient guide
- 11 November 2026: Protocol: enhanced recovery after liver surgery, from the ERAS Society 2022 recommendations with each element's evidence level and grade, and NICE NG180
- 11 November 2026: Protocols for carboplatin dose calculation and renal dose adjustment of anticancer drugs, from the 2025 international ADDIKD guideline and KDIGO 2024
- 11 November 2026: Protocol: palliative sedation for refractory suffering, from the revised EAPC framework (2024) and the Korean last-days guideline
- 11 November 2026: Ten care protocols for immunotherapy side effects: thyroid, diabetes, heart, nerves, kidneys, joints, eyes and blood, plus restarting treatment and tapering steroids
- 11 November 2026: Protocols: blood clot prevention around cancer surgery and surgical site infection prevention, from NICE NG89 and NG125 and the WHO global guidelines, with the points where they disagree set side by side
- 11 November 2026: Five supportive care and follow-up protocols: antiemetics by emetic risk, oral mucositis prevention, outpatient clot prevention, and follow-up after breast and colorectal cancer
- 11 November 2026: Protocols: strong opioids for cancer pain, opioid switching and conversion, and care in the last days of life, each read from NICE, WHO and national guidelines with every quote checked against the source text
- 11 November 2026: Three proposals to test practices the protocols found unmeasured: video second checks before chemotherapy at home, closed-system transfer devices, and antiseptic caps on long-term lines; linked from their protocols
- 11 November 2026: Protocol for oral anticancer medicines at home: prescribing, teaching, adherence support and safe handling
- 11 November 2026: Protocol for choosing a venous access device for anticancer treatment
- 11 November 2026: Country deep dives, wave 10: pages for Chad, Lebanon, Bolivia, Zambia, Puerto Rico, Uruguay, Lithuania, Niger, Cameroon and Georgia
- 11 November 2026: Ten more places read from their own sources: registries, payers, regulators and screening for Chad, Lebanon, Bolivia, Zambia, Puerto Rico, Uruguay, Lithuania, Niger, Cameroon and Georgia
- 11 November 2026: Protocol for intrathecal chemotherapy safety and giving vinca alkaloids only from a minibag
- 11 November 2026: Care protocols for checkpoint inhibitor skin reactions and infection prevention after CAR T-cell therapy
- 11 November 2026: feat: country pages for Afghanistan, Bulgaria, Norway, Finland, North Korea, Cambodia, Ireland, Zimbabwe, Angola and Tunisia
- 11 November 2026: Ten more countries read from their own sources: Afghanistan, Bulgaria, Norway, Finland, North Korea, Cambodia, Ireland, Zimbabwe, Angola and Tunisia
- 11 November 2026: Hazardous drug protocol: plain wording for USP chapter 800 so the page renders it as written
- 11 November 2026: Country deep dives, wave 9: pages for Slovakia, New Zealand, Croatia, Mozambique, Malawi, Madagascar, the Dominican Republic, Guatemala, Côte d'Ivoire and Singapore
- 11 November 2026: Ten more countries read from their own sources: registries, payers, regulators and screening for Slovakia, New Zealand, Croatia, Mozambique, Malawi, Madagascar, the Dominican Republic, Guatemala, Côte d'Ivoire and Singapore
- 11 November 2026: Country pages for Azerbaijan, Bosnia and Herzegovina, Haiti, Senegal, Slovenia, Mali, Papua New Guinea, Burkina Faso, Costa Rica and Paraguay, listed in the navigation in every language
- 11 November 2026: Ten more countries read from their own sources: registries, payers, regulators, screening and radiotherapy for Azerbaijan, Bosnia and Herzegovina, Haiti, Senegal, Slovenia, Mali, Papua New Guinea, Burkina Faso, Costa Rica and Paraguay
- 11 November 2026: Protocol for safe handling of hazardous anticancer drugs where they are given: protective equipment, closed systems, spills, waste and body fluids
- 11 November 2026: Care protocols for tumour lysis syndrome and bispecific antibody step-up dosing
- 11 November 2026: feat: country deep dives for Rwanda, Guinea, Moldova, Latvia, Honduras, Armenia, Somalia, Jordan, El Salvador and Panama
- 11 November 2026: Ten more countries read from their own sources: Rwanda, Guinea, Moldova, Latvia, Honduras, Armenia, Somalia, Jordan, El Salvador and Panama
- 11 November 2026: Protocol for PICC and tunnelled central line care: dressings, cleaning the connector, flushing and living with the line
- 11 November 2026: Country deep dives, wave 5: pages for Belgium, Morocco, Iraq, Peru, Myanmar, Chile, Portugal, Greece, Kenya and Czechia
- 11 November 2026: Ten more countries read from their own sources: registries, payers, regulators and screening for Belgium, Morocco, Iraq, Peru, Myanmar, Chile, Portugal, Greece, Kenya and Czechia
- 11 November 2026: Protocol for a suspected central line infection in a person with cancer: paired cultures, first antibiotics, and removing or keeping the line
- 11 November 2026: Ten care protocols for oncological emergencies, immunotherapy side effects and CAR T-cell toxicity
- 11 November 2026: Protocols for treating a blood clot in a person with cancer, and for a clot around a central line
- 11 November 2026: Protocols for infusion and allergic reactions during anticancer treatment, and for premedication before taxanes, antibodies and other reaction-prone drugs
- 11 November 2026: Country deep dives, wave 6: pages for Sudan, the Democratic Republic of the Congo, Hungary, Austria, Tanzania, Uzbekistan, Saudi Arabia, Uganda, Venezuela and Sri Lanka
- 11 November 2026: Ten more countries read from their own sources: registries, payers, regulators and screening for Sudan, the Democratic Republic of the Congo, Hungary, Austria, Tanzania, Uzbekistan, Saudi Arabia, Uganda, Venezuela and Sri Lanka
- 11 November 2026: Country deep dives, wave 7: pages for Ghana, Denmark, Nepal, Kazakhstan, Ecuador, Cuba, Yemen, Belarus, Serbia and Syria
- 11 November 2026: Ten more countries read from their own sources: registries, payers, regulators and screening for Ghana, Denmark, Nepal, Kazakhstan, Ecuador, Cuba, Yemen, Belarus, Serbia and Syria
- 11 November 2026: Protocols for DPD testing before fluorouracil and capecitabine, for prescribing and pharmacy verification of anticancer treatment, and for the checks at the chair before it is given
- 11 November 2026: Outperform Cancer is no longer called a nonprofit, since neither its site nor the IRS says so; it is now typed as media and education. The Chan Zuckerberg Initiative is shown as both a research institute and a philanthropy, with its own account of its structure
- 11 November 2026: Three care protocols read from their guidelines: accessing and caring for an implanted port, chemotherapy leaking out of the vein by drug class, and the first hour of suspected neutropenic sepsis
- 11 November 2026: A new record kind for care protocols: how a part of cancer care is carried out, step by step, each step with its reason, its source and the grade the source gives it
- 11 November 2026: Every country in GLOBOCAN has a page: 144 automatic country pages beside the 39 deep dives, France's overseas departments on the France page, and a filterable list of every country on /countries/
- 11 November 2026: Clarify conjugate sensing, redesign and transcription evidence
- 11 November 2026: Clarify target engagement and causal validation evidence
- 11 November 2026: Extend biomarker and study-comparison concepts
- 11 November 2026: Expand immune escape reading in education draft
- 11 November 2026: chore: weekly citation link check (#392)
- 11 November 2026: chore: refresh universe lists (NCI drugs and types, NCI centres, OECI, NHS alliances, NLM journals, OpenAlex, KEGG, ChEMBL, ClinicalTrials.gov, FDA OCE)
- 11 November 2026: chore: check regional approvals against the EMA register (#390)
- 11 November 2026: chore: refresh FDA approvals feed (OCE notifications, openFDA drugsfda) (#389)
- 11 November 2026: chore: draft change proposals that need review (#388)
- 11 November 2026: chore: EU regional rows read from the EMA register pages (#387)
- 10 November 2026: Clarify immune editing processes and evidence scope
- 10 November 2026: Add specimen allocation and RNA quality teaching
- 10 November 2026: Complete immune assay interpretation and repertoire measurement concepts
- 10 November 2026: Clarify specimen conditions for molecular detection and quantitation
- 10 November 2026: Expand immune assay measurements and interpretation boundaries
- 10 November 2026: Clarify blood sampling, variant origin and quantitative evidence
- 10 November 2026: Expand model context and intended-use teaching
- 10 November 2026: Clarify immunogenicity and efficacy teaching examples
- 10 November 2026: Expand evidence interpretation with scoped worked examples
- 10 November 2026: The site follows the device's light or dark setting until a reader chooses, an accessibility check runs on every ship, and the faults it found are fixed: unnamed timeline links, links hidden in expandable rows, mislabelled body and resistance buttons, low-contrast colours and small targets
- 10 November 2026: feat: the theme follows the device by default, and accessibility is checked in CI
- 10 November 2026: Drug labels that came from word fragments are corrected (37 antagonists no longer read as agonists, fluoropyrimidines no longer as degraders), companion-diagnostic rows link their devices, and the open-source catalogue keeps projects whose GitHub page went away
- 10 November 2026: A kept open-source record cites a page that answers today: paquo and tiffslide now cite PyPI, without the GitHub-only star counts
- 10 November 2026: paquo and tiffslide stay in the open-source catalogue from PyPI after their GitHub repositories went away
- 10 November 2026: Ten nonprofits move from companies to institutions, among them the Jaime Leandro Foundation and Chan Zuckerberg Initiative, and two drugs held twice become one record each: GFH375 with VS-7375, AZD0120 with GC012F
- 10 November 2026: Open-source catalogue: fuck-cancer, a patient brief skill for AI agents, joins the patient-facing projects
- 10 November 2026: Companion diagnostic rows carry their device's record, and the readout table links it
- 10 November 2026: Static redirect pages for the retired drug ids and the moved nonprofits
- 10 November 2026: Redirect the retired drug ids and the moved nonprofits' company routes
- 10 November 2026: Merge GFH375 with VS-7375 and AZD0120 with GC012F, and move ten nonprofits from company to institution records
- 10 November 2026: Classifier patterns read whole words: antagonists are no longer agonists, benzimidazoles no longer IMiDs
- 10 November 2026: Expand cited binder and manufacturing learning concepts
- 10 November 2026: Add validation and quantification reading guides
- 10 November 2026: Expand spatial assay interpretation at existing owners
- 10 November 2026: Clarify RAS signalling evidence boundaries
- 10 November 2026: chore: weekly maintenance (logos, structures, trial links, acronyms, orphan links)
- 10 November 2026: Clarify RNA construct design and manufacturing evidence
- 10 November 2026: chore: draft change proposals that need review (#385)
- 10 November 2026: chore: EU regional rows read from the EMA register pages (#384)
- 9 November 2026: Extend existing antigen-presentation owners with evidence distinctions
- 9 November 2026: Expand repertoire, dominance and spreading evidence boundaries
- 9 November 2026: Retain chemical species and assay context across evidence consumers
- 9 November 2026: Distinguish ADC chemical species and add tubulin cargo designs
- 9 November 2026: Clarify cell-product release evidence and manufacturing clocks
- 9 November 2026: Restore review captures and disclose expired temporary artifacts
- 9 November 2026: chore: weekly fact check, audit, and provenance refresh (#383)
- 9 November 2026: chore: render this week's issue to public/newsletter/ (#382)
- 9 November 2026: chore: refresh public/votes.json from Discussions reactions (#381)
- 9 November 2026: chore: refresh trial counts from ClinicalTrials.gov (#379)
- 9 November 2026: chore: draft change proposals that need review (#378)
- 9 November 2026: chore: EU regional rows read from the EMA register pages (#377)
- 8 November 2026: Preserve legacy cargo identities and avoid ordinary-word glossary links
- 8 November 2026: Explain dose coupling in fixed-composition dual-payload conjugates
- 8 November 2026: Clarify payload identities and evidence scopes
- 8 November 2026: Preserve scoped observations in readers and queries
- 8 November 2026: Add scoped evidence to oncology records
- 8 November 2026: Expand sourced immune, specimen and product evidence concepts
- 8 November 2026: Expand concepts for measurements, models and clinical evidence
- 8 November 2026: Expand sourced antibody-conjugate payload concepts beyond TOP1
- 8 November 2026: chore: draft change proposals that need review (#367)
- 8 November 2026: chore: EU regional rows read from the EMA register pages (#366)
- 7 November 2026: chore: draft change proposals that need review (#365)
- 7 November 2026: chore: EU regional rows read from the EMA register pages (#364)
- 6 November 2026: chore: draft change proposals that need review (#361)
- 6 November 2026: chore: EU regional rows read from the EMA register pages (#360)
- 5 November 2026: Add Elicit market map and twelve oncology research vendors
- 5 November 2026: The Rare Cancer Research Foundation and its Pattern.org tissue donation programme, with 56 more rare-cancer foundations and coalitions from sarcoma and chordoma to ocular melanoma, NUT carcinoma and DIPG
- 5 November 2026: Rare-cancer foundations: the Rare Cancer Research Foundation, Pattern.org and 56 more nonprofits
- 5 November 2026: Cassian Yee of MD Anderson, with his programmes, advisory roles, key papers on adoptive T-cell therapy, the trials he leads, and the organisations in his profile
- 5 November 2026: Cassian Yee of MD Anderson, with eight papers on endogenous T-cell therapy, nine trials from ClinicalTrials.gov, the term itself, and the companies and funders the faculty profile names
- 5 November 2026: Twenty-one more country pages (Indonesia to Thailand, Viet Nam to Taiwan), the machines page built from its records, searches that find nothing can start research privately, and device and contrast-agent records
- 5 November 2026: Seven product changes wait at /admin/
- 5 November 2026: chore: weekly roadmap registry check (public/roadmap-watch.json)
- 5 November 2026: chore: refresh institution research output from OpenAlex (#359)
- 5 November 2026: chore: draft change proposals that need review (#358)
- 5 November 2026: chore: EU regional rows read from the EMA register pages (#357)
- 4 November 2026: The shared chrome no longer carries the regional approvals table or one-page styles
- 4 November 2026: Country deep dives, wave 4: pages for Viet Nam, the Philippines, Ethiopia, South Africa, Colombia, Argentina, Ukraine, Romania, Malaysia, Algeria and Taiwan
- 4 November 2026: Eleven more countries read from their own sources: registries, payers, regulators and screening for Viet Nam, the Philippines, Ethiopia, South Africa, Colombia, Argentina, Ukraine, Romania, Malaysia, Algeria and Taiwan
- 4 November 2026: The timeline's year scrubber draws its pills with short classes, 91 KB lighter and unchanged on screen
- 4 November 2026: The census fill rounds: hospices and Maggie's centres, psycho-oncology and fertility services, helplines, patient coalitions and patient guides, 87 reporting and staging systems, 103 graded complementary products, 27 hereditary syndromes and germline tests, vaccines and screening, donor registries, biobanks, laws, payers, open-source projects, veterinary drugs and prizes
- 4 November 2026: Devices and contrast agents: ports, PICCs and pumps, stoma products, breast implants and prostheses, lymphoedema pumps, embolic beads, biopsy and localisation systems, pleural and peritoneal catheters, stents, laboratory analysers, Barrett's ablation, cholangioscopy and automated breast ultrasound, and 25 contrast agents and dyes
- 4 November 2026: The health gauges count the records the census fill rounds added
- 4 November 2026: Twenty US proton centres, 31 stem cell donor registries and cord blood banks, 20 biobanks including BBMRI-ERIC and NCI TARGET, the nine CDMRP cancer programmes and 22 more funding programmes, four European right-to-be-forgotten statutes, Medicare's negotiated cancer drug prices and 16 Medicare coverage rulings
- 4 November 2026: Prevention and screening gaps: hepatitis B and Chinese HPV vaccines, sixteen national HPV programmes from WHO data, nineteen FDA screening tests, three ER and PR staining kits, and twenty-seven national screening programmes
- 4 November 2026: Hereditary cancer syndromes and germline tests: 27 syndrome pages from Orphanet and the NHS test directory with guideline-cited management, and 13 named hereditary cancer tests with their FDA or Genetic Testing Registry status
- 4 November 2026: Patient coalitions and patient education: 33 European and international patient groups, NCCN Guidelines for Patients on 77 guides, 112 NCI PDQ patient summaries and 309 ONS/NCODA patient education sheets linked from their records
- 4 November 2026: The four FDA-approved dog cancer medicines and the Comparative Oncology Trials Consortium, the bodies behind the Brupbacher, Pezcoller-AACR, King Faisal, Gairdner and Sjoberg prizes, and the AACR, ASCO, Lasker, Coley, Szent-Gyorgyi, Griffuel and Deutscher Krebspreis awards on the records that give them, with their laureates linked
- 4 November 2026: Five product changes wait at /admin/: header alignment, country wave 3, faster search, the machines page and the search-miss event
- 4 November 2026: Complementary products: 103 graded records from Memorial Sloan Kettering's About Herbs and NCI PDQ, including milk thistle, black cohosh, low-dose naltrexone, PC-SPES, the Gonzalez regimen, kratom, cesium chloride, vitamins and minerals, medicinal mushrooms and herbal cancer cures
- 4 November 2026: Hospices, end-of-life care, navigation, fertility preservation, psycho-oncology, genetics services and helplines: 31 Maggie's centres, the nine Marie Curie hospices, Sue Ryder and Hospice UK; advance care planning documents, hospice services and nine assisted dying laws read from their statutes; patient navigation and the Medicare navigation codes; six fertility preservation services and schemes; IPOS and its 28 national societies; the seven NHS genomic regions and the NSGC directory; and 23 cancer helplines and information services in 20 countries with the ICISG member list
- 4 November 2026: Reporting systems, response criteria and staging systems: 87 glossary records from the November gap census, each citing its defining paper or society page
- 4 November 2026: The virotherapy page's step classes move from globals.css into a stylesheet only that page loads
- 4 November 2026: Open-source catalogue: 91 projects from the November gap census, pathology first, with the stale Sybil, GISTIC2 and Rcan skips corrected
- 4 November 2026: The Aa button is named for what its menu now holds: reading, language and display settings, in all nine languages
- 4 November 2026: The page-weight script weighs the page it is given: without --base the first path named was dropped and every recorded page was weighed instead
- 4 November 2026: The virotherapy page comes back under its weight ceiling, and the timeline and the technology pages stop scrolling sideways on a 320 px phone
- 4 November 2026: Census corrections: four more state wig-cover laws, ten NCCN guideline links that pointed at the wrong or a retired guideline, the UKONS and CTCAE links and the CTCAE v6 bands, Bravnetsa's September approval, the lung ninth edition, and aliases that sent readers to the wrong page
- 4 November 2026: Empty files registered for the November gap fill of hospices, end-of-life care, navigation, fertility preservation, psycho-oncology, genetic counselling and helplines
- 4 November 2026: The Aa menu gains text size, reading spacing and a reduce motion switch
- 4 November 2026: Gap census: 174 kinds of oncology object measured against outside lists, with 3,313 sourced gaps, eight kinds with no record and the corrections found on the way
- 4 November 2026: 268 cancer institutions measured against the IRS and Charity Commission registers, AACI, the Children's Oncology Group, Japan's designated cancer hospitals, India's National Cancer Grid and UICC, including the American Cancer Society itself
- 4 November 2026: National cancer societies: 27 UICC members from Denmark, Norway, Finland, Belgium, New Zealand, South Africa, Singapore, Hong Kong, Japan, Korea, the Philippines, Egypt and Latin America, the registered names of fourteen centres the corpus held, and the institution lists described in the data sources
- 4 November 2026: Search answers names from a small first file and says "Loading search…" until it can, and an exact name now comes first
- 4 November 2026: Machines page lists every machine the corpus marks, by family, with its chain of development
- 4 November 2026: Accessibility: survival-chart arms told apart by line and shape as well as colour, readable accent buttons in dark mode, the header fits at 320 px, and interactive drawings reachable by screen readers
- 4 November 2026: The ten wave 3 country pages read against their sources: 1,895 claims checked, 206 faulty (10.9 per cent), every one fixed or labelled
- 4 November 2026: Record header: the status chip and the "Also called" line share one baseline, with the same space under the lede on every record
- 4 November 2026: The British Thoracic Oncology Group is held as a nonprofit company beside the other groups, and new charity records drop a tag that had no description
- 4 November 2026: New centres keep their own trials: sponsor names for Loyola, Mass General Brigham, Mott, UCSF Benioff, UPMC Children's and St. Jude India resolve to their records, and UICC links the American Cancer Society itself
- 4 November 2026: UK cancer charities and Japan's designated cancer hospitals: 46 charities from the Charity Commission register and Maggie's, 39 prefectural hub and genomic medicine core hospitals from the health ministry's lists with their own English names, and the UICC member directory as a list
- 4 November 2026: Nineteen RNA tests, every radiotherapy machine line dated from its FDA record with scalp cooling, cryocompression and infusion chairs, Invictus Therapeutics and Invictus Oncology, and a daily check that the live search finds every record
- 4 November 2026: Country deep dives, wave 3: pages for Indonesia, Pakistan, Türkiye, Bangladesh, Iran, Mexico, Egypt, Poland, Nigeria and Thailand
- 4 November 2026: Ten more countries read from their own sources: registries, payers, regulators and screening for Indonesia, Pakistan, Türkiye, Bangladesh, Iran, Mexico, Egypt, Poland, Nigeria and Thailand
- 4 November 2026: Cancer charities and paediatric centres: 45 US charities from the IRS register including the American Cancer Society and Blood Cancer United, 44 Children's Oncology Group member programmes and 36 National Cancer Grid members in India, each linked from the cancer, network or parent record it belongs to
- 4 November 2026: Invictus: the two cancer companies of that name, Invictus Therapeutics in Sydney and Invictus Oncology in Delhi
- 4 November 2026: Radiotherapy machines, chemotherapy chairs and scalp cooling: every model line dated from its FDA record, five roadmaps of how each machine family developed, and four improvement ideas
- 4 November 2026: Search misses: a private research queue read from PostHog at run time, counts only in its output
- 4 November 2026: Search: count a settled search that finds nothing, with consent, as one filtered PostHog event
- 4 November 2026: A daily search health check: records missing from the live index, how far the site is behind main, and every typed name ranked by the site's own search
- 4 November 2026: AACI member centres: 25 academic cancer centres and the association itself, five large US cancer centres from the IRS list, and the legal names of eleven centres the corpus already held; the Jaime Leandro Foundation's record now carries its IRS filing, how its vaccine process works and who it is named for
- 4 November 2026: RNA tests: nineteen tests that read tumour RNA in tissue, urine and blood, a liquid RNA technology, and the proposal to read tumour RNA beside tumour DNA in one residual disease blood draw
- 4 November 2026: The command-line tool is onco-cc, since the npm name onco belongs to someone else, and four EU approvals read from EMA's register
- 4 November 2026: The CLI is published as onco-cc, because the npm name onco belongs to an unrelated package and npx onco would run it
- 4 November 2026: Institution coverage: a proper name at the start of a longer name counts, a description at its end does not
- 4 November 2026: Institution universe lists: US cancer charities (IRS), UK cancer charities (Charity Commission), AACI, Children's Oncology Group and National Cancer Grid members, with a coverage script
- 4 November 2026: Country coverage: every GLOBOCAN 2024 country ranked by cases, deaths, research output and population, and the sixty without a page
- 4 November 2026: EU authorisation rows for elinzanetant, Lumeblue, gallium-68 edotreotide and epoetin beta, read from the EMA pages, with epoetin beta quoting its cancer indication
- 4 November 2026: The subclone residual disease record now says the 57 mutations came from exome sequencing, and states the design cut-off and molecule counts that limit naming a subclone
- 4 November 2026: GLOBOCAN 2024 across the site: every figure from IARC's 2024 estimates, the edition named from the data, and a map legend for every 2024 data-method code
- 4 November 2026: Ultra-sensitive residual disease: eight more MRD assays including China's, the methods that read clones and phenotype from blood, TROP2 PET tracers, eleven trials and the limit on telling subclones apart; and 47 portfolio companies the monitors found
- 4 November 2026: Three new MRD trials are linked from their assay, drug and technology records
- 4 November 2026: Residual disease tests with their stated detection limits, the subclone attribution bottleneck, TROP2 PET tracers and the trials that act on a breast cancer blood result
- 4 November 2026: Electra Therapeutics and ipsoprubart are one record each, and the papers-snapshot gauge counts the drugs and targets added this week
- 4 November 2026: 83 oncology companies and 68 products found by the startup monitors in SEC filings, FDA and EMA orphan designations and NCI small-business grants, and GLOBOCAN 2024 waiting at /admin/
- 4 November 2026: Curadel and CPI-008 rest on the EMA designation and the trial registry only, since the company site refuses automated reads
- 4 November 2026: 47 oncology companies from investor and accelerator portfolios, with 32 pipeline products and 41 registry trials
- 4 November 2026: 83 oncology companies and 68 products found by the startup monitors in SEC filings, FDA and EMA orphan designations and NCI small-business awards
- 4 November 2026: 282 pipeline products for 156 companies and 14 new targets now that the engine pages are lighter, 130 unsupported links removed, survival calculators linked for issue 43, and the NHS Cancer Alliances list read from ONS
- 4 November 2026: chore: refresh universe lists (NCI drugs and types, NCI centres, OECI, NHS alliances, NLM journals, OpenAlex, KEGG, ChEMBL, ClinicalTrials.gov, FDA OCE) (#348)
- 4 November 2026: chore: refresh preprint tracker from Europe PMC (#347)
- 4 November 2026: chore: check regional approvals against the EMA register (#346)
- 4 November 2026: chore: refresh FDA approvals feed (OCE notifications, openFDA drugsfda) (#345)
- 4 November 2026: chore: draft change proposals that need review (#344)
- 3 November 2026: Add sourced RGCC International evidence coverage
- 3 November 2026: Alleanza Contro il Cancro keeps its own sponsor name rather than pointing at a consortium record, as the one-home rule for cooperative groups requires
- 3 November 2026: Pipeline products keep a cancer link only where a registered trial, approval or paper supports it: ten trials gain the cancers their ClinicalTrials.gov conditions or titles name, 130 unsupported product-cancer links from the company pipeline fill are removed, four trials drop cancers their registry entries never name, and MDS answers to its singular name
- 3 November 2026: Eleven pipeline products name the protein they are linked to as their sources do (TROP2, FCGR2B, CD3 for CLSP-1025, PIK3CG, BCMA from the KLN-1010 registry title, MYH9, MET, MEK1/2, BCR::ABL1); CBX-250 loses a CD3 link no source states
- 3 November 2026: Paper identifier snapshot covers the 43 PubMed ids added with the country pages; every one resolves to the paper its record describes, DOI and year agreeing
- 3 November 2026: PRODUCT (needs owner approval): the engine's interleukin-15 pattern matches IL-15 as a word, not inside the product alias cytoTIL15
- 3 November 2026: PRODUCT (needs owner approval): the eligibility pattern for androgen receptor pathway inhibitors matches enzalutamide as a word, and deutenzalutamide by name rather than inside it
- 3 November 2026: Trials sponsored by the Uganda Cancer Institute's UCI-Fred Hutch centre count for the Kampala institute, and Alleanza Contro il Cancro's English name no longer resolves to the US Alliance group
- 3 November 2026: Members test follows the tests fill: BostonGene Tumor Portrait and FoundationOne Heme rows now resolve to their own records
- 3 November 2026: Remove a duplicated companies key on AVZO-103 left by a merge; both copies named Avenzo
- 3 November 2026: Prognosis calculators from Memorial Sloan Kettering and the National Cancer Institute are listed on /software/ as link-outs with their validation papers, and the NHS Cancer Alliance list is read from the ONS geography after NHS England's page began refusing scripts
- 3 November 2026: A trial-linked TROP2 record listed its company twice, which TypeScript rejects; the duplicate key is removed
- 3 November 2026: The ten new country pages quote GLOBOCAN 2024 beside the 2024 table, and France's gap with GLOBOCAN is no longer explained by skin cancer
- 3 November 2026: GLOBOCAN's all-cancers figure includes non-melanoma skin cancer: the UK and Israel pages and the default label say so
- 3 November 2026: Pages read the GLOBOCAN edition from the data file instead of naming 2022, and the cancer map explains every 2024 data-method code in IARC's words
- 3 November 2026: Maintenance scripts name the GLOBOCAN edition they read rather than 2022
- 3 November 2026: Research-ranking rates for France, Canada, Italy, the Netherlands, Korea, Australia, Spain, Brazil, Sweden and Switzerland move to GLOBOCAN 2024
- 3 November 2026: Cancer figures move to GLOBOCAN 2024: IARC's new edition is fetched for 186 countries, and every burden line, country card, research-ranking rate and geography layer that cited GLOBOCAN 2022 or 2020 is re-read from it
- 3 November 2026: Accuracy round two checks every figure on the country pages and the 60 most-linked cancer pages against its source: 365 corrections, 37 dead DOIs replaced, and NICE's October reversal on trastuzumab deruxtecan for HER2-low breast cancer
- 3 November 2026: Drugs now link what they bind rather than what they act on downstream, an investor and a trials group are no longer credited as makers, and stale company histories are corrected from registry and SEC records
- 3 November 2026: Every figure on the country pages and the 60 most-read cancer pages read against its source: 23 per cent needed a fix, and four checks now stop the commonest faults coming back
- 3 November 2026: Review-list judgements from labels, registries and filings: drugs link the receptor they bind, an investor and a trials group no longer pose as makers, sponsors resolve to their own centres, ten company pages corrected, and two duplicate products cross-referenced with a merge plan
- 3 November 2026: Accuracy round one fixes wrong paper ids, trials and side-effect figures; 28 FDA companion diagnostics and 11 tumour tests, FDA, EMA and WHO coverage lists, about 270 wrong links corrected, and deals and regimens linked to their records
- 3 November 2026: Readout pages keep tests and medicines apart: devices name their readouts, test cancer links rest on coverage, trials or papers, and three devices are credited to their FDA applicants
- 3 November 2026: The ten new country pages were checked against their sources, with 151 faulty claims fixed and every funding row corrected
- 3 November 2026: The open drug engine pages carry the same look in short classes, so the small-molecule page drops from 468,848 to 411,819 bytes of markup
- 3 November 2026: Fourteen proteins named by new pipeline medicines now have target pages, including PLK4, HPK1, PARG, LSD1, CDC7 and KIF18A, and sixteen small molecules move from Not placed into the engine grid
- 3 November 2026: Every drug side-effect rate now names its source: the last 143 unsourced rows read against FDA labels and trial papers, and the count held at zero
- 3 November 2026: Company pages that showed no products now list them: 282 pipeline products from 156 biotech and cell-therapy companies, each confirmed on the company's own pipeline page, filing or the registry, with 273 registry trials; thirteen existing products gain the company that discovered or co-develops them
- 3 November 2026: Keep GC012F and AZD0120 as separate records for now: merging them adds a site redirect, which waits for the owner
- 3 November 2026: Seven regulator and knowledge-base universe lists, FDA oncology approvals since 2020 as dated events, the missing FDA companion diagnostic rows, and nine medicines the EMA, FDA and WHO lists hold
- 3 November 2026: A link-sense audit finds links that resolve but mislead: myeloma loses a prostate vaccine and three imaging agents, 76 bowel cancer trials stop reading "bevacizumab (glioblastoma use)", and sponsor names stop matching by fragment
- 3 November 2026: TL;DRs for the OncoMate test and Promega spell out microsatellite instability, and TruSight Oncology 500 reads as a test kit
- 3 November 2026: Deals, regimens, isotopes, sponsors, software and open-source rows now link the records they name, with twelve deal parties and ten deal assets given sourced records
- 3 November 2026: A read-only gap audit ranks what reader pages leave out, and page sections gathered from the corpus wait at /admin/ for approval
- 3 November 2026: Tumour profiling tests and companion diagnostic devices get their own pages: FoundationOne Heme, Caris MI Profile, OncoExTra, OmniSeq INSIGHT, Tumor Portrait, Altera, TruSight Oncology 500, Strata Select, Tempus xE, NeoGenomics Neo Comprehensive and PanTracer, and 28 FDA companion diagnostic devices
- 3 November 2026: The first measured accuracy sample: 182 claims read against their sources, the errors fixed, and guards for the patterns found
- 3 November 2026: The Signatera merge test follows the readout record, which no longer lists the assay row
- 3 November 2026: Readout, technology and company sections gather their members from the corpus by a stated rule, one item per product
- 3 November 2026: A read-only gap audit lists what reader pages leave out, ranked by pages affected
- 3 November 2026: Every Medicare-covered MRD blood test now has its own page, including Foundation Medicine's FoundationOne MRD and Personalis NeXT Personal, with Oncodetect, Guardant Reveal and RaDaR corrected from their sources
- 3 November 2026: Eight more ctDNA residual disease tests get sourced records (FoundationOne MRD, NeXT Personal, Haystack MRD, NavDx, Labcorp Plasma Detect, Precise MRD, TrueMRD, Tempus xM), and Oncodetect, Guardant Reveal and RaDaR ST are re-sourced
- 3 November 2026: Ten country deep dives wait at /admin/ for approval, and the proposal script records its preview address
- 3 November 2026: Weekly watch for new oncology startups across fifteen sources and for landmark cancer papers, the Nature lung Treg interception paper, and the GLOBOCAN all-cancers note corrected to include non-melanoma skin cancer
- 3 November 2026: GLOBOCAN's all-cancers figures include non-melanoma skin cancer; say so
- 3 November 2026: Country deep dives for Australia, Spain, Brazil, Sweden and Switzerland, each read from its own registry, payer and regulator
- 3 November 2026: A weekly landmark-paper watch proposes new oncology papers for review, and the Nature regulatory T cell interception paper is added
- 3 November 2026: Startup monitors read portfolios, SEC filings, HKEX 18A, orphan designations and NCI small-business awards into the freshness issue
- 3 November 2026: 36 multi-cancer early detection tests, the Prometheus gaps filled, rare lymphomas linked to their trials, sixteen companies brought up to date and imaging agents given approval rows
- 3 November 2026: France, Canada, Italy, the Netherlands and South Korea, each read from its own national sources
- 3 November 2026: Add Selvana and Jiyu Bio, freshen sixteen large companies, name nine drugs' targets and give imaging agents regional approval rows
- 3 November 2026: Lymphoma: trials and papers for the entity pages the taxonomy added, and two corrections
- 3 November 2026: OpenOnco joins the open-source catalogue (PR 290), and the Open Medical Registry stays listed while its repository is private
- 3 November 2026: Fill the second round of Prometheus gaps: nine technologies, two drugs, eighteen companies, two facilities, the tumour atlas network and cDC1 cross-presentation
- 3 November 2026: Extend the multi-cancer test census to China, Taiwan and India, and add two new US Breakthrough designations
- 3 November 2026: chore: weekly maintenance (logos, structures, trial links, acronyms, orphan links) (#339)
- 3 November 2026: Pathway diagrams draw clean right-angle routes with rounded corners: no line runs through a box and every label is shown in full
- 3 November 2026: chore: refresh literature snapshot from Europe PMC (#338)
- 3 November 2026: chore: refresh universe lists (NCI drugs and types, NCI centres, OECI, NHS alliances, NLM journals, OpenAlex, KEGG, ChEMBL, ClinicalTrials.gov, FDA OCE) (#337)
- 3 November 2026: chore: draft change proposals that need review (#335)
- 3 November 2026: chore: EU regional rows read from the EMA register pages (#334)
- 2 November 2026: Draw pathway diagrams with routed, rounded edges that avoid boxes
- 2 November 2026: The weekly refresh reads Even One's new page again and opens an issue when a source goes stale or a portfolio company is missing
- 2 November 2026: Cover brain-to-cancer neural circuit research
- 2 November 2026: chore: render this week's issue to public/newsletter/ (#333)
- 2 November 2026: chore: weekly fact check, audit, and provenance refresh (#332)
- 2 November 2026: chore: refresh public/votes.json from Discussions reactions (#331)
- 2 November 2026: chore: refresh trial counts from ClinicalTrials.gov (#330)
- 2 November 2026: chore: draft change proposals that need review (#329)
- 2 November 2026: chore: EU regional rows read from the EMA register pages (#328)
- 1 November 2026: The weekly universe refresh reads Even One's new page, and opens an issue for stale sources and portfolio companies OnCo lacks
- 1 November 2026: To-do: country deep dives, wave 2
- 1 November 2026: To-dos: Selvana, every source for new oncology startups, and a weekly watch for landmark cancer papers
- 1 November 2026: chore: monthly identifier enrichment (HGNC, Wikidata) (#327)
- 1 November 2026: chore: refresh HTA decisions and survival snapshots (#326)
- 1 November 2026: chore: draft change proposals that need review (#325)
- 1 November 2026: chore: EU regional rows read from the EMA register pages (#324)
- 31 October 2026: chore: draft change proposals that need review (#323)
- 31 October 2026: chore: EU regional rows read from the EMA register pages (#322)
- 30 October 2026: Tests, imaging agents and devices are called what they are, and 83 of them stop claiming to kill cancer cells
- 30 October 2026: Tests, imaging agents and devices are no longer called drugs or treatments
- 30 October 2026: Logos for 218 of the 262 new oncology companies, from each company's own site icon
- 30 October 2026: Preserve PostHog ingestion token and anonymous profile flag
- 30 October 2026: Fix consented Google tag commands and prepare private PostHog page counts
- 30 October 2026: chore: draft change proposals that need review (#320)
- 30 October 2026: chore: EU regional rows read from the EMA register pages (#319)
- 29 October 2026: 262 oncology companies new to OnCo, and 82 existing ones brought up to date with this year's acquisitions, approvals and listings
- 29 October 2026: Exact Sciences records the completed Abbott purchase, with Abbott's own release
- 29 October 2026: 262 oncology companies from 2024 to 2026 added, and 82 existing companies brought up to date with their acquisitions, approvals and closures
- 29 October 2026: Every multi-cancer early detection test we could verify, 27 of them worldwide, with their real performance, regulatory status and an honest grade
- 29 October 2026: Every multi-cancer early detection test we could find: 27 tests, 13 new companies, 57 registered studies and 26 papers
- 29 October 2026: Organisations without a logo show OnCo's own initials tile instead of a stock icon, and a profile for Dr Melissa Taylor at UCSF
- 29 October 2026: Organisations without a logo show OnCo's own initials tile instead of a stock icon from Google's favicon service
- 29 October 2026: A profile for Dr Melissa Taylor, breast medical oncologist at UCSF
- 29 October 2026: Eleven people stop showing a stranger's photo, 28 gain a properly licensed portrait, and 77 logos get sharp
- 29 October 2026: Portraits: 28 added, 11 wrong faces removed; logos: 77 sharper files
- 29 October 2026: Portrait and logo fetchers: stricter identity checks, ORCID matching, site icons
- 29 October 2026: 108 records that nothing linked to now appear on the pages they belong on, and six genes stop being called HER2 amplicon neighbours
- 29 October 2026: Six genes from other chromosomes stop being listed as HER2 amplicon neighbours
- 29 October 2026: The immune and microenvironment pages now show the trials acting on them: 88 for T-cell exhaustion, 28 for macrophages, 22 for adenosine
- 29 October 2026: Give 108 orphan records a way in from the records they belong to
- 29 October 2026: Alligator Bioscience's page stops claiming a single-arm trial showed longer survival
- 29 October 2026: Connect the immune and microenvironment mechanisms to the agents and trials acting on them
- 29 October 2026: To-do: a profile for Melissa Taylor, once the owner confirms which one
- 29 October 2026: chore: refresh pulse, abstract and citation snapshots (#318)
- 29 October 2026: chore: weekly roadmap registry check (public/roadmap-watch.json)
- 29 October 2026: chore: refresh institution research output from OpenAlex (#317)
- 29 October 2026: chore: draft change proposals that need review (#316)
- 29 October 2026: chore: EU regional rows read from the EMA register pages (#315)
- 28 October 2026: chore: refresh universe lists (NCI drugs and types, NCI centres, OECI, NHS alliances, NLM journals, OpenAlex, KEGG, ChEMBL, ClinicalTrials.gov, FDA OCE) (#314)
- 28 October 2026: chore: refresh preprint tracker from Europe PMC (#313)
- 28 October 2026: chore: check regional approvals against the EMA register (#312)
- 28 October 2026: chore: refresh FDA approvals feed (OCE notifications, openFDA drugsfda) (#311)
- 28 October 2026: chore: draft change proposals that need review (#310)
- 28 October 2026: chore: EU regional rows read from the EMA register pages (#307)
- 27 October 2026: chore: weekly maintenance (logos, structures, trial links, acronyms, orphan links) (#309)
- 27 October 2026: chore: draft change proposals that need review (#308)
- 26 October 2026: Performance ceilings lowered to what the live site measures after round two
- 26 October 2026: Pages stop freezing on phones: the home page's blocking time falls from two seconds to almost nothing, and it downloads a third as much
- 26 October 2026: Record pages on a wide screen stop downloading themselves a second time
- 26 October 2026: Home and drug pages stop blocking the phone for two to three seconds, and the home page halves in size
- 26 October 2026: Seven duplicate ideas become one each, a proton paper stops being listed twice, and a cohort study is no longer called a randomised trial
- 26 October 2026: Eight Sleep's Hot Flash Mode, graded honestly: one preprint by company staff in menopausal women, nothing yet in cancer
- 26 October 2026: Record the Ask recall measurement for the 26 October merges
- 26 October 2026: One proton RBE review and six care ideas each had two pages; they are merged, and a lung cancer study is no longer called a randomised trial
- 26 October 2026: chore: weekly fact check, audit, and provenance refresh (#306)
- 26 October 2026: chore: render this week's issue to public/newsletter/ (#305)
- 26 October 2026: chore: refresh public/votes.json from Discussions reactions (#304)
- 26 October 2026: chore: refresh trial counts from ClinicalTrials.gov (#303)
- 26 October 2026: chore: draft change proposals that need review (#302)
- 26 October 2026: chore: EU regional rows read from the EMA register pages (#301)
- 25 October 2026: chore: draft change proposals that need review (#300)
- 25 October 2026: chore: EU regional rows read from the EMA register pages (#299)
- 24 October 2026: chore: draft change proposals that need review (#298)
- 24 October 2026: chore: EU regional rows read from the EMA register pages (#297)
- 23 October 2026: chore: draft change proposals that need review (#296)
- 23 October 2026: chore: EU regional rows read from the EMA register pages (#295)
- 22 October 2026: A measured performance budget: Lighthouse on seven page types under real throttling, with a ratchet
- 22 October 2026: Pages load faster: first visits stop loading twice, the cancer map stops downloading the whole corpus, and drawings wait until the page has settled
- 22 October 2026: chore: weekly roadmap registry check (public/roadmap-watch.json)
- 22 October 2026: chore: refresh pulse, abstract and citation snapshots (#294)
- 22 October 2026: chore: refresh institution research output from OpenAlex (#293)
- 22 October 2026: chore: draft change proposals that need review (#292)
- 22 October 2026: chore: EU regional rows read from the EMA register pages (#291)
- 21 October 2026: The ship chain retries once when Vercel refuses a deploy without creating one
- 21 October 2026: How precise cancer care is: a sourced page test by test, beam by beam and dose by dose, and 37 protocol upgrades that could improve care now
- 21 October 2026: chore: refresh preprint tracker from Europe PMC (#289)
- 21 October 2026: chore: check regional approvals against the EMA register (#288)
- 21 October 2026: chore: refresh universe lists (NCI drugs and types, NCI centres, OECI, NHS alliances, NLM journals, OpenAlex, KEGG, ChEMBL, ClinicalTrials.gov, FDA OCE) (#287)
- 21 October 2026: chore: refresh FDA approvals feed (OCE notifications, openFDA drugsfda) (#286)
- 21 October 2026: chore: draft change proposals that need review (#285)
- 21 October 2026: chore: EU regional rows read from the EMA register pages (#284)
- 20 October 2026: Twenty-two cheap, evidence-backed changes to everyday cancer care are graded by how proven and how widely used they are, and two pages stop saying symptom tracking makes people live longer
- 20 October 2026: Two dosing rows marked as oversold carry their evidence grade
- 20 October 2026: Five precision rows marked as oversold now carry the evidence grade that says how far
- 20 October 2026: How precise a cancer drug dose is today: ten matrix rows and four records on blood levels, genes and kidney function
- 20 October 2026: Pegulicianine says what its trial showed, and the PD-L1 calibration idea cites the assay comparison it names
- 20 October 2026: Fifteen ways to deliver radiotherapy and surgery better, from letting patients see their own breath-hold to four weeks of clot prevention after bowel surgery
- 20 October 2026: Fifteen precision rows for blood tests and genomic matching say how precise each test is today and what acting on it has been shown to buy
- 20 October 2026: Radiotherapy precision rows: margins, contouring, image and surface guidance, breath-hold, motion, MRI-guided adaptation, radiosurgery, proton range, FLASH and dose painting
- 20 October 2026: A page for how precise cancer care is, test by test and dose by dose, with a roadmap and a check that every row is sourced
- 20 October 2026: Precision matrix for surgery, imaging and pathology: thirteen sourced rows on margins, sentinel nodes, robots, PSMA PET, lung screening and slide scoring, and a record for checking breast margins during the operation
- 20 October 2026: Skeleton for the precision and protocol-upgrades run: matrix types, one file per facet, registrations and tags
- 20 October 2026: chore: weekly maintenance (logos, structures, trial links, acronyms, orphan links) (#283)
- 20 October 2026: chore: refresh literature snapshot from Europe PMC (#282)
- 20 October 2026: chore: draft change proposals that need review (#281)
- 20 October 2026: chore: EU regional rows read from the EMA register pages (#280)
- 20 October 2026: Exclude navigation pages before CLI and MCP search limits
- 19 October 2026: 252 lung cancer trials stop being filed under small-cell, and six radioligand trials stop being called FAP-2286
- 19 October 2026: The sidebar stops reprinting the Connected section, and ordinary words like why and infection stop linking to unrelated glossary pages
- 19 October 2026: Trials that have closed stop saying they are recruiting, in English, plain language and Chinese
- 19 October 2026: The audit contradictions it took a day to clear can no longer come back unnoticed
- 19 October 2026: Ten cancer pages now say what the failed trials they cite actually found
- 19 October 2026: A de-escalation trial reads as one, and a product approved in 1990 stops hiding it
- 19 October 2026: An offline API copy in a folder with a space in its name is read correctly by the CLI and the MCP server
- 19 October 2026: chore: weekly fact check, audit, and provenance refresh (#277)
- 19 October 2026: chore: render this week's issue to public/newsletter/ (#276)
- 19 October 2026: chore: refresh public/votes.json from Discussions reactions (#275)
- 19 October 2026: chore: refresh trial counts from ClinicalTrials.gov (#274)
- 19 October 2026: chore: draft change proposals that need review (#273)
- 19 October 2026: chore: EU regional rows read from the EMA register pages (#272)
- 18 October 2026: chore: draft change proposals that need review (#271)
- 18 October 2026: chore: EU regional rows read from the EMA register pages (#270)
- 18 October 2026: Fix offline API file URL decoding
- 17 October 2026: chore: draft change proposals that need review (#266)
- 17 October 2026: chore: EU regional rows read from the EMA register pages (#267)
- 16 October 2026: The weekly literature refresh stops discarding its own work, and stops recording throttled replies as zero papers
- 16 October 2026: Nineteen immune markers get a page, and the microenvironment records link to the agents aimed at them
- 16 October 2026: chore: EU regional rows read from the EMA register pages (#265)
- 15 October 2026: chore: refresh pulse, abstract and citation snapshots (#264)
- 15 October 2026: chore: weekly roadmap registry check (public/roadmap-watch.json)
- 15 October 2026: chore: refresh institution research output from OpenAlex (#263)
- 15 October 2026: chore: EU regional rows read from the EMA register pages (#262)
- 14 October 2026: The command palette stops showing results for a query you have cleared
- 14 October 2026: Concept search recovers after a blip, a stale feed stops claiming it is fresh, and the chain keeps what landed while it worked
- 14 October 2026: Concept search recovers after a blip, and a stale feed stops claiming it is fresh
- 14 October 2026: chore: refresh preprint tracker from Europe PMC (#261)
- 14 October 2026: chore: refresh universe lists (NCI drugs and types, NCI centres, OECI, NHS alliances, NLM journals, OpenAlex, KEGG, ChEMBL, ClinicalTrials.gov, FDA OCE) (#260)
- 14 October 2026: chore: check regional approvals against the EMA register (#259)
- 14 October 2026: chore: EU regional rows read from the EMA register pages (#258)
- 13 October 2026: chore: weekly maintenance (logos, structures, trial links, acronyms, orphan links) (#257)
- 13 October 2026: chore: draft change proposals that need review
- 13 October 2026: chore: EU regional rows read from the EMA register pages (#255)
- 12 October 2026: Fifty-eight UCSF breast studies we were missing, and the PARP breast gaps closed
- 12 October 2026: chore: render this week's issue to public/newsletter/ (#254)
- 12 October 2026: chore: weekly fact check, audit, and provenance refresh (#253)
- 12 October 2026: chore: refresh public/votes.json from Discussions reactions (#252)
- 12 October 2026: chore: refresh trial counts from ClinicalTrials.gov (#251)
- 12 October 2026: chore: draft change proposals that need review
- 12 October 2026: chore: EU regional rows read from the EMA register pages (#250)
- 11 October 2026: chore: EU regional rows read from the EMA register pages (#249)
- 10 October 2026: Complete UCSF adult female breast-study coverage with 58 registry records
- 10 October 2026: Connect PARP evidence records and preserve supportive approval scope
- 10 October 2026: Fill PARP breast trial and evidence gaps with scoped eligibility
- 10 October 2026: chore: EU regional rows read from the EMA register pages (#243)
- 9 October 2026: Correct the prevalence figures in the cBioPortal roadmap note
- 9 October 2026: Seven contributor changes: table filters stop colliding, dossiers say where you are and fold, prevalence opens as a list
- 9 October 2026: The bot's proposal list, and cBioPortal ingestion scoped on the roadmap
- 9 October 2026: chore: draft change proposals that need review
- 9 October 2026: chore: EU regional rows read from the EMA register pages (#241)
- 8 October 2026: Weekly idea votes from Discussions, and the bot's proposal list
- 8 October 2026: The new trials page gets a measured weight ceiling, pinned because its table is paged
- 8 October 2026: A page for finding a trial you could still join, filtered by cancer, country and who it is open to
- 8 October 2026: An unreviewed page says so in three words at the foot, not five sentences in the margin
- 8 October 2026: A trial page now says who can join and where, and a page has one way to correct it
- 8 October 2026: The site no longer loads unstyled after a deploy
- 8 October 2026: The site loaded unstyled after a deploy, and the star count comes back
- 8 October 2026: Land dossier jumps again when a table pages in rows mid-scroll
- 8 October 2026: Refit other columns while a column is being resized
- 8 October 2026: The TROPION-Breast01 record says what the FDA actually did
- 8 October 2026: Link dossier navigation only to sections the target has
- 8 October 2026: chore: refresh pulse, abstract and citation snapshots (#240)
- 8 October 2026: chore: weekly roadmap registry check (public/roadmap-watch.json)
- 8 October 2026: chore: refresh institution research output from OpenAlex (#239)
- 8 October 2026: chore: draft change proposals that need review
- 8 October 2026: chore: EU regional rows read from the EMA register pages (#238)
- 7 October 2026: Keep open header tips from widening table columns
- 7 October 2026: Respect row content minimums when sizing table columns
- 7 October 2026: Include the visible mobile language label in its accessible name
- 7 October 2026: Cover similar page ranking and lazy export behavior
- 7 October 2026: Compute similar pages per record during rendering
- 7 October 2026: Keep review images outside the repository
- 7 October 2026: Refresh resize review screenshots after accessibility fixes
- 7 October 2026: Provide initial accessible column resize values
- 7 October 2026: Show column resize dividers on hover and focus
- 7 October 2026: Fit and resize shared table columns with visual comparisons
- 7 October 2026: Require side-by-side screenshots in product pull requests
- 7 October 2026: Follow React lint rules in disclosure rendering tests
- 7 October 2026: Make dossier sections collapsible with prevalence closed initially
- 7 October 2026: Highlight the current section in dossier navigation
- 7 October 2026: Run CI gates and static export in parallel without weakening checks
- 7 October 2026: Align CI render gates and avoid duplicate page checks
- 7 October 2026: Fix table filter names and badge accessibility
- 7 October 2026: Expire fact-check cache entries and record source retrieval dates
- 7 October 2026: Thirteen contributed fixes: nine fetchers that keep what they already had, four that let search recover
- 7 October 2026: Revert "Expire older-day OpenAlex work caches before dating research snapshots (#213)"
- 7 October 2026: Eleven contributed TROP2 and PARP pull requests, and the trials they brought
- 7 October 2026: AGENTS.md: never resolve a contributor's pull request by union
- 7 October 2026: Eleven contributed records land, and the trials they brought get an inbound edge
- 7 October 2026: The page-weight ratchet counts the right thing now
- 7 October 2026: chore: refresh preprint tracker from Europe PMC (#230)
- 7 October 2026: chore: refresh universe lists (NCI drugs and types, NCI centres, OECI, NHS alliances, NLM journals, OpenAlex, KEGG, ChEMBL, ClinicalTrials.gov, FDA OCE) (#229)
- 7 October 2026: chore: check regional approvals against the EMA register (#228)
- 7 October 2026: chore: refresh FDA approvals feed (OCE notifications, openFDA drugsfda) (#227)
- 7 October 2026: chore: EU regional rows read from the EMA register pages (#226)
- 7 October 2026: Expire older-day OpenAlex work response caches
- 7 October 2026: Retain survival snapshots when a source site fails
- 7 October 2026: Retain institution research after required request failures
- 7 October 2026: Preserve congress abstract snapshots after failed pages
- 7 October 2026: Release saved import capacity before growing lists
- 7 October 2026: Preserve saved lists when an import cannot persist
- 7 October 2026: Type the preprint request test cases consistently
- 7 October 2026: Retain preprints when search responses are incomplete
- 7 October 2026: Keep citation snapshot when an OpenAlex batch fails
- 7 October 2026: Retain pulse items when a source returns a non-feed response
- 7 October 2026: Restore the page origin after closing keyboard shortcuts
- 7 October 2026: Scope recovery assertions to lexical search errors
- 7 October 2026: Preserve paper snapshots after malformed search responses
- 7 October 2026: Restore command palette opener focus on dismissal
- 7 October 2026: Retry concept search after temporary index failures
- 7 October 2026: Require verified rows before advancing the EMA check stamp
- 7 October 2026: fix: retry browser Ask index after temporary failures
- 7 October 2026: Recover search results and inline answers after index failures
- 7 October 2026: Reject malformed browser record envelopes before caching
- 7 October 2026: fix: retain EMA snapshots after unreadable refresh input
- 7 October 2026: Report browser record failures and permit Ask retries
- 7 October 2026: fix: scope modality table URL state
- 7 October 2026: Preserve FDA snapshots until both sources finish completely
- 6 October 2026: Deepen PARP1 programmes, biomarker evidence and clinical trial states
- 6 October 2026: Clarify current US PARP labels and withdrawn ovarian treatment
- 6 October 2026: Expand TROP2 clinical and historical programmes from primary registries
- 6 October 2026: Reconcile TROP2 regulatory labels and FDA trial outcomes
- 6 October 2026: Refresh TROP2 trial landscape from current registry records
- 6 October 2026: Expand TROP2 clinical programs and early evidence
- 6 October 2026: Add primary-sourced preclinical TROP2 ADC programs
- 6 October 2026: Add cited KH815 and ASP2998 TROP2 dual-payload programs
- 6 October 2026: Seven trims to the top of a record page, and a review card that says what is true
- 6 October 2026: Cite breast ADC labels for the TROP2 testing statement
- 6 October 2026: About 19,500 pages stop opening with the least useful true thing they could say
- 6 October 2026: Correct LCB84 modality and cite its phase 1/2 registry record
- 6 October 2026: Separate TROP2 copy-number amplification from IHC expression
- 6 October 2026: Update TROP2 biomarker with current US breast indications
- 6 October 2026: Correct TROPION-Breast01 approval statement and cite final survival results
- 6 October 2026: Seven things leave the top of a record page
- 6 October 2026: Three more things leave the top of a record page
- 6 October 2026: Four things leave the top of a record page
- 6 October 2026: Reopen retained search queries from the keyboard
- 6 October 2026: Recover search box queries after index load failures
- 6 October 2026: Keep trial feed stale when a snapshot date is invalid
- 6 October 2026: fix(factcheck): count unresolved helper record identities
- 6 October 2026: Read explicit marker results before bare sign fallback
- 6 October 2026: fix(factcheck): reject ambiguous ids across data files
- 6 October 2026: Share marker suffix parsing across trial criteria and profiles
- 6 October 2026: Keep protocol identity separate from title context
- 6 October 2026: Expose retained trial dates in modality and feed status views
- 6 October 2026: Require complete marker predicates before applying trial exclusions
- 6 October 2026: Filter semantic candidates before score allocation
- 6 October 2026: Preserve complete trial titles and identifier boundaries
- 6 October 2026: Disclose retained snapshot dates in trial aggregates
- 6 October 2026: Keep ambiguous eligibility marker clauses unresolved
- 6 October 2026: Require trial-specific identity before registry patches
- 6 October 2026: fix(mcp): retain hidden values in field comparisons
- 6 October 2026: chore: weekly maintenance (logos, structures, trial links, acronyms, orphan links)
- 6 October 2026: chore: EU regional rows read from the EMA register pages (#154)
- 5 October 2026: chore: render this week's issue to public/newsletter/ (#153)
- 5 October 2026: chore: weekly fact check, audit, and provenance refresh (#152)
- 5 October 2026: chore: refresh public/votes.json from Discussions reactions
- 5 October 2026: chore: refresh trial counts from ClinicalTrials.gov (#149)
- 5 October 2026: chore: EU regional rows read from the EMA register pages (#128)
- 5 October 2026: fix(mcp): compare complete field values instead of summaries
- 5 October 2026: Propagate Ask record retrieval failures in CLI and MCP
- 5 October 2026: Preserve qualified biomarker names when matching trials
- 5 October 2026: Keep trial biomarker polarity scoped to each result
- 5 October 2026: Hold trial fact-check patches until registry identity is resolved
- 5 October 2026: Preserve trial index entries across failed refresh requests
- 5 October 2026: fix(search): filter record kinds before limiting retrieval
- 5 October 2026: Patch only direct literal fields in fact-check corrections
- 5 October 2026: fix(cli): resolve record URLs with query strings and anchors
- 4 October 2026: The sources come off the sidebar and go to the foot of the page
- 4 October 2026: Arrow keys reach the results, and a brand name finds its drug
- 4 October 2026: Arrow keys reach the results, a brand name finds its drug, and a prefix finds what starts with it
- 3 October 2026: chore: EU regional rows read from the EMA register pages (#127)
- 2 October 2026: How recovery is measured, and who is reached by it
- 2 October 2026: Name the England quality of life survey as a gap rather than describe it
- 2 October 2026: How recovery is measured, and who is actually reached by it
- 2 October 2026: The recovery front gets its hub, its history and its open problems
- 2 October 2026: Recovery: the front's hub rewritten, the history of survivorship, and the research agenda
- 2 October 2026: Rehabilitation, reconstruction, and a matrix of what comes back
- 2 October 2026: What comes back after treatment: a sourced recovery matrix, treatment by treatment
- 2 October 2026: Cancer rehabilitation and reconstruction: the services that put function back, and the gap between them and what people are offered
- 2 October 2026: Recovery: transplant, second cancers, the mind, and the markdown nobody was rendering
- 2 October 2026: Graft-versus-host disease, late effects after transplant, and the immune system afterwards
- 2 October 2026: Fear of recurrence, mood, body image, work, money and sleep after cancer
- 2 October 2026: Second cancers after treatment, and the screening that follows
- 2 October 2026: Childhood and young adult survivorship: the cohorts that made the field
- 2 October 2026: Childhood, adolescent and young adult survivorship: the cohorts, the burden, and the follow-up that governs care
- 2 October 2026: Six summaries that described the page instead of the thing
- 2 October 2026: Six TL;DRs were talking about the page instead of the subject
- 2 October 2026: Recovery and rejuvenation: a new front, and the hair spike the owner asked for
- 2 October 2026: Biological ageing after cancer treatment, and what is sold against it
- 2 October 2026: Recovery and rejuvenation: the body, system by system
- 2 October 2026: AGENTS.md learns the five days since it was written
- 2 October 2026: The home page stops drawing a flow the corpus does not have
- 2 October 2026: Hair: the number for your own regimen, and an honest grade for everything sold for regrowth
- 2 October 2026: The home page stops drawing a flow the corpus does not have, and two labels come off
- 2 October 2026: A twentieth front: Recovery and Rejuvenation
- 2 October 2026: Lymphoma, and a home page with less above the grid
- 2 October 2026: The hero stops promising a plain-English TL;DR
- 2 October 2026: The home page loses the record count, the JSON chip and the body pill above the grid
- 2 October 2026: chore: EU regional rows read from the EMA register pages (#125)
- 1 October 2026: Lymphoma: the biology of the germinal centre, and what a person actually decides
- 1 October 2026: Lymphoma: the UK and NHS pathway, and the world geography layer
- 1 October 2026: Lymphoma evidence: 29 trials, 42 papers, a roadmap and eight open questions, each resolved against a registry or Europe PMC
- 1 October 2026: Nineteen lymphoma entity pages, a rewritten non-Hodgkin hub, and both 2022 classifications quoted where they disagree
- 1 October 2026: The oldest three DLBCL rows retire, because the record now says the same thing three times
- 1 October 2026: Lymphoma molecular layer: the antigens and what they cost, the lesions, the classifications, resistance and the tests
- 1 October 2026: Lymphoma, the person: six decisions, six red cards, three first-sixty-days lists
- 1 October 2026: The weekly check now fails when a page grows past what a reader downloaded last week
- 1 October 2026: A ceiling on what a reader actually downloads, measured on the live site
- 1 October 2026: Lirafugratinib is approved, not phase 2, and the bot stops proposing one combination's approval for another
- 1 October 2026: chore: monthly identifier enrichment (HGNC, Wikidata) (#124)
- 1 October 2026: chore: refresh HTA decisions and survival snapshots (#123)
- 1 October 2026: chore: refresh pulse, abstract and citation snapshots (#122)
- 1 October 2026: chore: weekly roadmap registry check (public/roadmap-watch.json)
- 1 October 2026: chore: refresh institution research output from OpenAlex (#121)
- 1 October 2026: chore: draft change proposals that need review
- 30 September 2026: A name typed in full comes first, and people have faces in the search
- 30 September 2026: /prevalence/ opens on a list, because the grid is 95 per cent empty
- 30 September 2026: Four verified approval rows from the proposals queue, and two structure keys that stopped the build
- 30 September 2026: Page titles, second attempt: a noun phrase naming the thing and the dimension that orders it
- 30 September 2026: The unattended gates run the tests the way the ship chain does, so a weekly data refresh can merge itself again
- 30 September 2026: The sources come off the sidebar and go to the foot of the page, grouped by who published them
- 30 September 2026: chore: refresh preprint tracker from Europe PMC
- 30 September 2026: chore: refresh universe lists (NCI drugs and types, NCI centres, OECI, NHS alliances, NLM journals, OpenAlex, KEGG, ChEMBL, ClinicalTrials.gov, FDA OCE)
- 30 September 2026: chore: check regional approvals against the EMA register
- 30 September 2026: chore: weekly citation link check
- 30 September 2026: chore: refresh FDA approvals feed (OCE notifications, openFDA drugsfda)
- 30 September 2026: chore: draft change proposals that need review
- 29 September 2026: Lymphoma treatment rows, and red cards that no longer name a drug the reader is not taking
- 29 September 2026: Lymphoma treatment: 70 standard-of-care rows across 21 diseases, each citing its trial
- 29 September 2026: The software of oncology, and the top of a record page
- 29 September 2026: chore: weekly maintenance (logos, structures, trial links, acronyms, orphan links)
- 28 September 2026: Every cancer-page section named for what is in it, in nine languages
- 28 September 2026: The block inside Symptoms and diagnosis is not called that too, and four pill labels follow the new names
- 28 September 2026: The page that shows what is waiting, with the queue read from the open pull requests
- 28 September 2026: A command for a person to type is not a change to the product
- 28 September 2026: The queue is a rendering of the open pull requests, refreshed on every ship
- 28 September 2026: The admin page waiting in its own queue
- 28 September 2026: The page that shows what is waiting, and says plainly what it is not
- 28 September 2026: The record page top waiting in the queue
- 28 September 2026: The top of a record page is the record again: a family strip of six, two paragraphs of summary, the clinic names readable, the model panel last
- 28 September 2026: The software map waiting in the queue
- 28 September 2026: The map a product proposal belongs in: 136 pieces of oncology software, each row saying whether anything outside the company backs it up
- 28 September 2026: The title review waiting in the queue
- 28 September 2026: Thirty page titles that named a format, could belong to any site, or needed a term first
- 28 September 2026: Product changes now wait for the owner, and the plan for the cancer page is written
- 28 September 2026: The approval mechanism belongs on main; the renames stay in the queue
- 28 September 2026: The cancer page measured before it is redesigned, and a plan the owner can approve part by part
- 28 September 2026: The first product change waiting in the queue
- 28 September 2026: Product changes wait for the owner; data ships
- 28 September 2026: Each section of a cancer page is now named for what is inside it, and the years index says which years it holds
- 28 September 2026: The front page links to 140 sections and shows almost none of them
- 28 September 2026: The prevalence matrix: 23 columns wide and growing with the corpus
- 28 September 2026: Visual first, and the measurement that would be misleading if taken naively
- 28 September 2026: Every page title, with the count of how many are one or two words
- 28 September 2026: The machine links, and the admission that seven patches are not a design
- 28 September 2026: The years index title, and the question of how it differs from the timeline
- 28 September 2026: The section names and the wall of links, with what each section holds and how many links a page carries
- 28 September 2026: The section called Evidence is called Trials and papers, which is what is in it
- 28 September 2026: Three more record-page asks on the roadmap, with what the print control actually does
- 28 September 2026: The fourth ask: the also-known-as line is the smallest type on the page
- 28 September 2026: Three owner asks on the roadmap: what a reader meets first on a record page
- 28 September 2026: Five weekly runs merged, the agent rules written down with the scripts they point at, and a card that can name no record no longer counts as one that lost one
- 28 September 2026: A card that can name no record is not a card that lost one
- 28 September 2026: Provenance regenerated with the private name redacted after the weekly run reintroduced it
- 28 September 2026: A locally installed skill directory is not part of this project
- 28 September 2026: The rules an agent needs, in the file every agent reads, with the scripts they point at now in the repository
- 28 September 2026: chore: render this week's issue to public/newsletter/
- 28 September 2026: chore: weekly fact check, audit, and provenance refresh
- 28 September 2026: chore: refresh public/votes.json from Discussions reactions
- 28 September 2026: chore: refresh trial counts from ClinicalTrials.gov
- 28 September 2026: chore: draft change proposals that need review
- 27 September 2026: A total review of fatalistic language: 153 phrases rewritten, the survival guard now says what a median is, and the red-flag card leads with calling
- 27 September 2026: A cancer with no NHS pathway page now says so in development as it does in production
- 27 September 2026: A word that is the source's is shown as the source's
- 27 September 2026: The half a keyword search cannot find: order, framing and the numbers printed as verdicts
- 27 September 2026: Say what the cancer does, not how to feel about it: fatalistic language out of OnCo's own voice, every quotation left exactly as its author wrote it
- 27 September 2026: Three dated fields the corpus lacked: accelerated approvals with their confirmations, a sourced first-description year on targets, and a trial start date
- 27 September 2026: Adenoid cystic carcinoma described without the word relentless
- 25 September 2026: Three dates the timeline was missing: every accelerated approval and what became of it, when each target was first described, and when each trial opened
- 25 September 2026: Measure what a reader downloads, not what the tests can render
- 25 September 2026: A record for every year and one timeline over all of them, with ten questions asked of the corpus and two it cannot answer
- 25 September 2026: Oncolytic virotherapy in full: four approved replicating viruses and not one randomised survival benefit, and the 2024 self-experiment read from its case report
- 25 September 2026: A record for every year in oncology, one timeline over all of them, and what the shape of the dates shows
- 25 September 2026: The for-me budget measures its own tiles, not the chrome every page carries
- 25 September 2026: Oncolytic virotherapy as a subject, not scattered records: the interferon defect, the approvals and what they really showed, and the virologist who treated herself
- 25 September 2026: Clicking a value in a table now filters by it: thirteen cells across eight tables, a year facet on trials, and a guard that a filter link can never come back empty
- 25 September 2026: Click the year to filter the table: thirteen cells across eight tables become filters
- 25 September 2026: Three owner asks on the roadmap: cells as filters, a record per year, and oncolytic virotherapy
- 25 September 2026: Six countries in depth, each read from its own registry and regulator: the United States, Germany, Japan, the United Kingdom, Israel and Russia
- 25 September 2026: Wave one of the country deep dives is done: six countries
- 25 September 2026: Four country pages said the figure excluded non-melanoma skin cancer; it never did
- 25 September 2026: The benchmarking partnership is hosted by Cancer Research UK, not founded by it
- 25 September 2026: Israel: a committee decides in public, once a year, inside a fixed sum, which cancer drugs the state will pay for
- 25 September 2026: One label over four health services: the United Kingdom deep dive, and the cost-per-QALY threshold the corpus had wrong
- 25 September 2026: Japan reads tenth in our own count and fifth in the world, so the Japan page is written from Japanese sources
- 25 September 2026: Russia read from Russian sources: the Herzen register and what it does not count, who pays, and the Eurasian approvals route
- 25 September 2026: Two country deep dives: the United States, where the question is who pays, and Germany, where a drug is paid for on the day it is approved and assessed afterwards
- 25 September 2026: The first two country deep dives: the United States and Germany
- 25 September 2026: In America the pill has a ceiling and the drip does not, and 239 accelerated approvals show what happens when nobody checks
- 25 September 2026: A red-card pattern can no longer match inside an unrelated word
- 25 September 2026: Compugen, which found PVRIG by predicting it, and the paper that first measured what a liposome does in a patient
- 25 September 2026: Germany pays for a new cancer drug on the day it goes on sale and asks what it is worth afterwards
- 25 September 2026: A blood-clot warning for a myeloma drug was reaching every fluoropyrimidine page, because the pattern matched inside the word pyrimidine
- 25 September 2026: Remove the scratch script used to survey what the corpus held for Israel
- 25 September 2026: Israel: the people and papers behind the country page, and the term for what a founder variant is
- 25 September 2026: The skin review: a page about a cream warned a reader about a blood clot, and an appraisal that refused a drug was recorded as one that funds it
- 25 September 2026: A second front page at /v2 for review, and table filters that find what you asked for: family cancers, phases in time order, search ranked by what you typed
- 25 September 2026: Country deep dives on the roadmap, ranked by what the corpus holds and by what that ranking gets wrong
- 25 September 2026: Filtering a table by a cancer family finds its subtypes, the phase filter runs in time order, and the facet search ranks what you typed
- 25 September 2026: The front page opens on the list again; the graph is one click away
- 25 September 2026: Four page-render tests were measuring how busy the machine is, not the code
- 25 September 2026: Four launch-list items were already done: tick them so nobody does them twice
- 25 September 2026: The answer to which page is mine now sits on the page a reader lands on, not the last tab
- 25 September 2026: Skin cancer on the NHS: nearly a third of England's urgent cancer referrals are for skin, and the commonest skin cancer is excluded from the standard that measures them
- 25 September 2026: Two layers of the skin round wrote the same trial: the ratchet caught it and the merge tool fixed it
- 25 September 2026: Record which of the skin layer's sources a program cannot open, checked one URL at a time
- 25 September 2026: The skin cancer NHS pathway, and the two exclusions that mean the published figures are not what they look like
- 25 September 2026: A reader on a breast subtype page now gets the cord compression card, and never two sepsis cards
- 25 September 2026: Skin cancer: what a person with a basal cell or squamous cell cancer faces, the family tree given its reason, and the counting rule that hides the commonest cancer there is
- 25 September 2026: The breast family review: the family page duplicates nothing, but the two receptor pages it routes to said everything twice
- 25 September 2026: What a person with a basal cell or squamous cell skin cancer actually faces, and a rule that lets a row cite the leaflet that carries its sentence
- 25 September 2026: The treatment of the keratinocyte cancers: the margin, Mohs surgery, the creams and the light, radiotherapy, the hedgehog pathway and the transplant recipient
- 25 September 2026: Three review lessons from the skin layer: a citation that does not contain the claim passes every gate
- 25 September 2026: The rows about living with a skin cancer now cite the leaflet that actually contains the sentence, and the photodynamic therapy figures move to the cream the trials used
- 25 September 2026: The skin family given its routing, the keratinocyte cancers given their subtypes and staging, and the counting rule that governs every figure on them
- 25 September 2026: A patient leaflet may be the source for a row about living with a disease, if the row names who wrote it
- 25 September 2026: What a person with a basal cell or squamous cell skin cancer actually faces: the choice between surgery and its alternatives with the trial figures, the operation on the face, Mohs as a day, and the ones that are not cured
- 25 September 2026: The family roll-up does not assume who is reading, and stops printing the cancer's name twice in one sentence
- 25 September 2026: Family cancer pages now reach what their subtypes hold: 3,683 trials, 1,079 medicines and 3,534 centres that no family page could reach before
- 25 September 2026: A family page now shows the trials, medicines and centres its subtypes hold, grouped by the subtype each came from
- 25 September 2026: The breast family page given its surgery, its NHS pathway and its shared decisions, and 83 duplicate records retired onto the ones that carry the reading
- 25 September 2026: The prostate review found an appraisal that refused a drug recorded as merely restricting it, and the breast family page given the routing a reader needs
- 25 September 2026: Breast cancer in the NHS: the screening programme the site never had, the family history service, both referral routes, and four wrong or stale funding claims corrected
- 25 September 2026: One thing, one id: 83 duplicate records merged, with the rule, the helper and the ratchet the three reviews asked for
- 25 September 2026: Breast cancer, the shared treatment spine: 22 trials on the operation, the axilla, radiotherapy and reconstruction, read at twenty and twenty-five years
- 25 September 2026: Breast cancer, living with it: the surgery choice with the twenty-year figures, the axilla and lymphoedema, fertility, sensation, hair, and four red cards
- 25 September 2026: The breast family page: the routing a reader needs in one click, the shared story of grade, stage and the receptor tests, and the classification edition nobody had noticed changed
- 25 September 2026: Prostate review: the appraisal that refused abiraterone was recorded as one that restricted it, eleven papers written twice, and the screening trials the whole section rests on
- 25 September 2026: The lung review: 33 lines the layers had written twice folded away, four contradictory figures resolved against their sources, and two duplicate-key bugs the lung data exposed
- 25 September 2026: The lung review: a placeholder idea rewritten from the UK page's own gaps, 33 duplicated base lines folded, three papers that existed twice merged, and two duplicate-key bugs the lung data exposed
- 25 September 2026: Prostate cancer in depth: 57 trials from their publications, the receptor landscape from eighteen deposits, the March 2026 screening recommendation, and four wrong NHS funding rows corrected
- 25 September 2026: Wire the two prostate pathology terms into the papers that found them
- 25 September 2026: Prostate molecular: the androgen receptor's five escape routes, the repair genes the four PARP labels disagree about, and what a PSMA PET cannot tell you
- 25 September 2026: The explained page budget measures its furniture, not how much the corpus knows
- 25 September 2026: Prostate glossary: fifteen new terms, with the screening figures, the two endpoint definitions and PI-RADS against the NICE Likert scale
- 25 September 2026: Prostate cancer on the NHS: the screening committee's March 2026 recommendation from its own coversheet, the risk in Black men with its cohorts, and the 62-day standard prostate fails worst
- 25 September 2026: Prostate treatment from its trials: 57 randomised comparisons, what each option costs a man, and the four NICE refusals
- 25 September 2026: Prostate cancer: five orthogonal classifications settled, the Cambridge Prognostic Groups added, and the decisions written with the trial figures rather than the word side effects
- 25 September 2026: Register the twelve prostate glossary drafts and link them from the decisions that use them
- 25 September 2026: Prostate cancer evidence: the 44 papers that made the field, a roadmap in twelve eras, seven ideas and nine open problems
- 25 September 2026: Prostate cancer, living with it: the ProtecT trade-off in numbers, hormone therapy for the long run, and cord compression named as the emergency
- 25 September 2026: An agent that has only the origin can now find the corpus: MCP and llms descriptors under .well-known, licence and provenance on every API response
- 25 September 2026: Prostate cancer has five classifications at once, and only one of them makes pages
- 25 September 2026: How to read the UK bodies that answer a script with a wall
- 25 September 2026: Lung cancer on the NHS: the screening programme from its own protocol, 34 centres, 21 appraisals and 8 refusals, the deprivation gradient from four national sources
- 25 September 2026: The UK lung pathway hangs off the family, not one branch of it
- 25 September 2026: Lung cancer gets its UK and NHS layer: the screening programme with its risk thresholds, the deprivation gradient in four nations, 108 funding decisions and the reflex testing clock
- 25 September 2026: Lung cancer given its trials, its molecular landscape and its evidence: 36 trials written from source, three landscapes, 44 papers, a roadmap of eleven eras
- 25 September 2026: A standard-of-care row for the antibody-drug conjugates and bispecifics entering lung cancer
- 25 September 2026: Lung molecular layer: adenocarcinoma, squamous and small-cell landscapes read separately, PD-L1 by assay, why tumour mutational burden failed, and resistance mechanism by mechanism
- 25 September 2026: Approval dates cut back to what the FDA notification page shows
- 25 September 2026: Lung cancer evidence: 44 landmark papers from Doll and Hill to tarlatamab, a roadmap in eleven eras with a 2032 registry watch, and seven open questions
- 25 September 2026: Registry routing reads the title and conditions, not the eligibility text
- 25 September 2026: Roadmap page quoted the old review gauge label
- 25 September 2026: Write down the three gauges no script can move, and what the owner holds
- 25 September 2026: Lung registry snapshot gets a breadth rule, later lines get their trials, and Ask keeps its floor
- 25 September 2026: Lung cancer given a family and a patient side: five tiers under lung-cancer, ten decision rows, 86 appointment questions, six red-card sets
- 25 September 2026: Europe PMC snapshots for 1,545 records, 42 logos, and a collection that had borrowed a charity's mark
- 25 September 2026: Take the backlink fill back out: it costs 0.002 of Ask recall, and the floor does not move
- 25 September 2026: Lung cancer: the patient side, written so a person who smoked and a person who never smoked are both addressed
- 25 September 2026: Read the three new lung pages in the dev server and drop one redundant alias
- 25 September 2026: Three more lung glossary terms the other layers will need, and an outlook paragraph on the family page
- 25 September 2026: Glossary terms: four Wikipedia links found, 58 absences recorded as checked
- 25 September 2026: Lung taxonomy settled and the core layer written: the family now hangs together, and lung cancer is described from its sources
- 25 September 2026: Lung cancer treatment layer: 884 registry trials, 31 landmark trials and the England access rows
- 25 September 2026: A test that fails when a gauge's denominator moves and its label does not
- 25 September 2026: Backlink fill for 1,002 weak records: two acronym collisions fixed first
- 25 September 2026: Provenance was blind to the generated files: 1,486 records get their commit line back
- 25 September 2026: Three gauges were measuring the wrong population; fix the questions, not the targets
- 25 September 2026: Colorectal reviewed against its sources: duplicate treatment rows folded, eleven NICE errors corrected, RAS wild-type recorded as a biomarker, 1,406 links checked
- 25 September 2026: Review the colorectal deep spike: fold the duplicated rows, settle the contradictory figures, fix eleven NICE errors and add the RAS wild-type readout
- 25 September 2026: Colorectal cancer deep spike complete (core, 710 trials, molecular landscape, evidence and roadmap); red cards re-sourced against the live toolkit; pages no longer grow with what is known about a reference; weekly jobs merge only when the gates pass
- 25 September 2026: Roadmaps and dossiers made robust to corpus growth: a reference costs the same whatever is known about it, the dossier export is a file, and no bot merges a tree the tests have not seen
- 25 September 2026: Check every cited URL and fix the one that was dead
- 25 September 2026: Register the colorectal treatment spike, its helper files and the sponsor supplement with the audit
- 25 September 2026: De-duplicate the NICE NG151 link across the three colorectal patches, and the CMS paper on its own term
- 25 September 2026: Thirteen glossary terms the evidence, living-with and molecular layers asked for, and the eight living-with drafts registered
- 25 September 2026: Colorectal: the historical trials the evidence and molecular layers cite, and targets for ten pipeline drugs
- 25 September 2026: Colorectal cancer deep spike, round one: the UK and NHS pathway across four nations with the national audit indicators, decisions, first 60 days, red cards and the adjuvant chemotherapy aid
- 25 September 2026: Red flags re-sourced: the UKONS triage tool found at its new URL, nine thresholds quote-checked against it, six redirects fixed, soft 404s caught by the weekly link check
- 25 September 2026: Colorectal: company links for the trials the registry names, and the sponsor backfill that keeps the rankings honest
- 25 September 2026: Colorectal molecular layer: 38-row genomic landscape from eight cBioPortal cohorts, sidedness as biology, the adenoma and serrated routes, CMS, immunity and ctDNA, with 47 paper records
- 25 September 2026: Colorectal cancer, the UK and NHS layer: the screening programme and its four-nation differences, the faecal immunochemical test that now governs referral, the National Bowel Cancer Audit, 13 NICE appraisals with the 2026 bevacizumab reversal, and the UK trial legacy
- 25 September 2026: Colorectal cancer deep spike: the evidence, the roadmap and the ideas
- 25 September 2026: Colorectal core, second pass: sidedness and consensus molecular subtype supplements, the screening standards source, and three claims tightened against their sources
- 25 September 2026: Colorectal: anti-EGFR rechallenge and the microsatellite-stable immunotherapy attempt
- 25 September 2026: Colorectal trials: cross-links into the records already in the corpus, and two corrections
- 25 September 2026: Colorectal cancer, living with it, care and decisions: ten decision rows, 74 appointment questions, a 19-item first 60 days checklist, six red-card sets and a NICE NG151 adjuvant decision aid
- 25 September 2026: Fold the duplicates the base colorectal record and spike carried: five history entries, seven biomarker lines, three state-of-the-art points and three open problems written twice
- 25 September 2026: Colorectal treatment layer: 38 hand-written trials from MOSAIC to ALTAIR, 661 registry trials, verified NICE and FDA approval rows, palliation
- 25 September 2026: Colorectal core layer: sourced epidemiology, the early-onset rise, risk, referral, screening, staging, pathology and prevention on the colorectal record; four subtype pages and eleven glossary terms
- 24 September 2026: LAUNCH: pancreatic spike and the cooperative-group home ticked
- 24 September 2026: Pancreatic reviewed against its sources: duplicates folded, figures settled, 1,006 links checked, mobile overflow fixed
- 24 September 2026: Pancreatic deep spike review: duplicates merged across seven files, daraxonrasib approval and NICE terminations settled, 13 paper duplicates supplemented, 47 redirected URLs rewritten, seven glossary terms and two journals, UK and tool pages lay out at 390 px; docs/PANCREATIC-QA.md
- 24 September 2026: Pancreatic cancer deep spike, round four: the molecular landscape with allele-level KRAS shares, subtype classes, precursor sequence and 82 papers
- 24 September 2026: Pancreatic molecular layer: the genomic landscape, subtypes, precursors, stroma and monitoring markers with their cohorts
- 24 September 2026: Pancreatic cancer deep spike, round three: 30 trials with outcomes, 284 registry trials, verified NICE and label rows, fifteen standard-of-care rows; Ask ranking weights curated records
- 24 September 2026: Ask concept index: inverse document frequency over curated records only; extractive recall 0.408 to 0.427, floor raised to 0.42
- 24 September 2026: Pancreatic treatment layer gates: cooperative groups under companies, pancrelipase flagged supportive, deep-dive tag described, stroma-enzyme schematic rule
- 24 September 2026: Pancreatic cancer deep spike, round two: the record extended with sourced epidemiology, risk, diagnosis and surveillance, six subtypes, 22 terms, 34 papers and a twelve-era roadmap
- 24 September 2026: Pancreatic core: classical versus basal-like (GATA6) and clinical benefit response terms for the evidence layer, high-risk individual folded into familial pancreatic cancer, disease-free survival note on the EFS term; the other requested ids resolve to existing records
- 24 September 2026: Pancreatic treatment layer: ESPAC-1 record, shorter TL;DRs, duplicate label links removed, explained-page budget note
- 24 September 2026: Pancreatic evidence: UNICANCER is a cooperative-group company, listed under companies
- 24 September 2026: Pancreatic treatment layer: 32 hand-written trials, 284 registry trials, drug approvals by regulator, standard-of-care export
- 24 September 2026: Pancreatic core: register the seven living-with glossary drafts (enzyme draft folded into pancreatic-enzyme-replacement, new-onset diabetes alias repointed), drop URLs the UK and living patches already link, subtype rows cite the neoadjuvant term rather than PRODIGE 24 so /explained/ stays under budget, audit list and stale zh summary updated
- 24 September 2026: Pancreatic cancer evidence: 34 Europe PMC paper records, a roadmap from Whipple 1935 to the 2031 registry watch, nine ideas
- 24 September 2026: Pancreatic cancer deep spike, round one: the UK and NHS pathway with the national audit findings, decisions, first 60 days, red cards and the first-treatment aid
- 24 September 2026: Pancreatic core layer: sourced epidemiology, risk factors, presentation and emergency share, diagnosis, resectability and TNM 8, pathology and precursors, surveillance programmes and outlook on the pancreatic record; six WHO subtype pages; thirteen glossary terms; five supplements
- 24 September 2026: Pancreatic cancer living with it, care and decisions: ten decision rows, 90 hand-written questions, first 60 days checklist, six red-card sets with the 111 and 999 wording, first-treatment decision aid from NICE NG85, sourced living-with notes on 26 side-effect and procedure records
- 24 September 2026: Pancreatic cancer UK and NHS page: NG12 diabetes rule, 28/31/62-day standards, NPaCA 2026 findings, 23 HPB hubs, NICE TA476/TA440/TA750/TA1052 and SMC decisions verified, PERT audit gap, EUROPAC and Precision-Panc legacy, four-nation registry figures
- 24 September 2026: Cooperative groups have one home: 35 trial groups move from institutions to cooperative-group companies with redirects
- 24 September 2026: One home for cooperative groups: 35 consortium institutions become cooperative-group companies (row 251)
- 24 September 2026: TNBC reviewed against its sources: figures reconciled, duplicates folded, 822 links checked
- 24 September 2026: Roadmap watch row: text column shrinks and wraps on phones
- 24 September 2026: TNBC QA: de-duplicate the six patches, correct KEYNOTE-522, BRCA share, NG101 1.3.6, TA992 and TROPION-Breast05, fix four dead links, roadmap and UK page fit 390 px; docs/TNBC-QA.md
- 24 September 2026: Wave 4: 99 cancer entity pages under the taxonomy rule; roadmaps page their eras; subtype pages rank below their parents in Ask
- 24 September 2026: Ask OnCo retrieval: subtype capped below its parent, provenance weight shared by both stages; extractive recall 0.403 to 0.411, floor held at 0.41
- 24 September 2026: LAUNCH: decisions and shipping of 25 Sept 2026
- 24 September 2026: Section planner test allows eight paged sections for NSCLC after the lung histologies; wave 4 pages weigh 0.7 in the Ask index
- 24 September 2026: Wave 4 cancer pages, family 7: sixteen pituitary, thymoma, salivary, cervical, vulvar and vaginal entities; roadmap Progress row
- 24 September 2026: Cancer hubs with section sub-pages, TNBC deep spike complete (core, treatments and 235 trials, molecular landscape), Ask recall repaired, route progress bar and edge caching, Methods and models glossary with 129 terms
- 24 September 2026: Roadmap pages: eras collapse behind Show era pills fed by /api/v1/roadmaps/<id>.json; watch table paged past 20; 300 KB markup budget
- 24 September 2026: Ask OnCo retrieval: drop function words and rank by kind and name in the word search; extractive recall 0.329 to 0.416, floor 0.41 with a ratchet
- 24 September 2026: Wave 4 cancer pages, family 6: fourteen rare tumour entities; cancer map SVG budget 320 KB
- 24 September 2026: TNBC treatment: reword a code comment to the page name
- 24 September 2026: TNBC treatment: LOTUS, PAKT, SCAN-B and ZEST detail, registry trials linked to subtype pages, test fixtures for the new rows
- 24 September 2026: Wave 4 cancer pages, family 5: sixteen lymphoma and leukaemia entities from WHO-HAEM5
- 24 September 2026: EntityDetail: remove the conflict markers committed with the architecture merge, drop the unused Tip import
- 24 September 2026: TNBC molecular layer: sourced genomic landscape, Lehmann and Burstein subtypes, ancestry and founder mutations, ctDNA studies, assay-specific PD-L1 and HER2-low thresholds, 55 landscape papers
- 24 September 2026: Wave 4 cancer pages, family 4: twenty colon, kidney and testicular entities; sibling linker so no new page arrives orphaned
- 24 September 2026: TNBC core: register the eight living-with glossary terms drafted by the living file, drop the brand-name aliases that hijacked Ask, trim two TL;DRs and a jargon phrase
- 24 September 2026: TNBC treatment deep dive: trial and drug records, approvals by regulator with NICE appraisal numbers, 229 registry trials, standard-of-care rows
- 24 September 2026: record-fold test: TNBC ceiling follows the new layers until the hub architecture lands, then drops to 300 KB
- 24 September 2026: CanSim terms, methods group: 58 "Methods and models" glossary terms with one-line definitions and why they appear in cancer papers; gap document gets a Done column
- 24 September 2026: CanSim terms, oncology group: 70 sourced glossary terms and the cancer-ai-vocabulary hub, 11 HGNC-sourced gene targets, TCGA cohort codes as cancer aliases, 12 data-source rows, Methods and models category
- 24 September 2026: Loading experience: navigation progress bar, route skeletons, intent prefetch on dense lists, pages hot at the edge, above-the-fold checks
- 24 September 2026: Decisions page holds 390 px: referenced-record and guideline pills truncate with tooltips, grid children may shrink
- 24 September 2026: TNBC core layer: sourced epidemiology, risk, presentation, diagnosis, pathology, staging, screening and outlook on the tnbc record; twelve subtype pages; nineteen glossary terms
- 24 September 2026: Modality hubs for twelve formats, TNBC deep spike round one (UK and NHS pathway, evidence and roadmap, decisions, first 60 days, red cards, after-chemotherapy aid), supportive care flag on 55 drugs, 22 cooperative-group records
- 24 September 2026: Geography section fits 390 px: grid items may shrink and source labels wrap
- 24 September 2026: Wave 4 cancer pages, family 3: fourteen breast histologies under breast-cancer from the WHO 2019 breast classification
- 24 September 2026: TNBC living with it, care and decisions: eleven decision rows, 77 hand-written questions, first 60 days checklist, three red-card sets with the 111 and 999 wording, after-chemotherapy decision aid, sourced living-with notes on 20 side-effect and procedure records
- 24 September 2026: TNBC UK and NHS layer: pathway page for tnbc with the screening caveat, NG12 and NG101 steps, every NICE and SMC decision including the TA992 refusal against SMC2608, R208 and M234 tests, eight UK-open trials with sites, TNT, POSH/OlympiA and PARTNER legacy, registry figures read from PHS, WCISU and NICR downloads, Breast Cancer Now and three researcher records, named gaps
- 24 September 2026: WIP: tnbc-core spike draft before merging main
- 24 September 2026: Cooperative groups as company records:
cooperative-groupcompany type with browser facet, glyph pill and tooltip; 22 group records (GORTEC, German CLL Study Group, French Sarcoma Group, AGO, AIO, MRC CTU, HORG, DPCG, DBCG, DAHANCA, GERCOR, ISG, SSG, EMN, IFM, ALFA, AMC, OCOG, KGOG, SARC, FIL, FILO) linked to 41 corpus trials as sponsor; sponsor aliases carry their ids and the IFM and KGOG mis-resolutions are fixed; the trials fetchers link and never give products to them - 24 September 2026: Supportive care medicines:
supportivedrug flag on 55 records, "Supportive care" pill with glyph and tooltip, Purpose facet on /drugs/, excluded from the home treatments count, the product rankings and the treatment references on standard-of-care rows; EMA and FDA fetchers flag candidates by a stated indication rule - 24 September 2026: Cancer records as a hub with ten sections: registry, section pages, shared strip, sections.json, budgets and deep-link tests
- 24 September 2026: Wave 4 cancer pages, family 2: ten lung histologies under nsclc from the WHO 2021 thoracic classification
- 24 September 2026: Front navigation: the kinds drawn as a linked graph, the body map as the entry point on phones; ship chain builds on Vercel only
- 24 September 2026: TNBC evidence: 45 papers verbatim from Europe PMC (guidelines, receptor and basal-like discovery, BRCA1 founder mutations, disparities, subtypes, pCR and RCB, platinum, CREATE-X, OlympiA updates, PD-L1 assays, TILs, TROPION-Breast02, CAPItello-290, HER2-low, ctDNA), tnbc-roadmap with 14 eras and a registry watch list to 2033, nine ideas
- 24 September 2026: Modality lens: /modalities/ and twelve format hubs assembling what the corpus records per shape of medicine
- 24 September 2026: Home page: the counts grid becomes a living graph of the kinds, with the body map as the entry point on phones
- 24 September 2026: Wave 4 cancer pages, family 1: entity pages under the seven parents without subtype records, taxonomy rule in docs/CANCER-PAGES.md
- 24 September 2026: LAUNCH: the ship chain no longer builds locally; Vercel builds and the chain verifies the alias
- 24 September 2026: LAUNCH: issue and PR sweep of 24 Sept, checkpoints, target statuses, ivermectin
- 24 September 2026: Ivermectin: the trial evidence so far, with the label toxicity table and the papers under expressions of concern stated
- 24 September 2026: Ivermectin evidence page: drug record with label toxicity table, four registry trials, 22 papers with expressions of concern and retraction stated, compact fenbendazole and mebendazole records, per-file citation test
- 24 September 2026: Checkpoint families mapped with immune and cell-cycle hubs, 26 new targets; 31 reader proposals applied as 23 trial records and 27 papers; Cancersim terms gap filed
- 24 September 2026: Cancersim terms gap: 378 of 504 terms not yet in OnCo, listed by bucket with the decision the methods terms need; IMPROVEMENTS row 250
- 24 September 2026: Target specificity and distribution on 904 targets with where each is found, gallbladder decision aids and comparison, world map of gallbladder cancer rates, six EU register rows decided by hand, PALD1 rewritten
- 24 September 2026: Issue 98: LITESPARK-011 (NCT04586231) readout written onto the record: Lancet 2026 dual primary progression-free and overall survival rows and toxicity as stated in the abstract, ASCO GU 2026 response, duration and treatment-related toxicity rows marked as conference data, Motzer 2026 paper record linked, plain-English summary, simple sentence and Chinese TL;DR
- 24 September 2026: Issue 97: INSITE (NCT03686215) trial record for pegulicianine with the NEJM Evidence 2023 coprimary endpoints and the existing Smith paper linked, pegulicianine.trials wired; status mixed because the sensitivity threshold was not met
- 24 September 2026: Issue 96: HBI0101 / NXC-201 multiple myeloma phase 1 (NCT04720313) trial record with the Haematologica 2023 response, survival and cytokine release endpoints, Asherie 2023 paper record, nxc-201-car-t wired to the trial and to multiple myeloma
- 24 September 2026: Issue 95: ScreenTrustCAD (NCT04778670) trial record for Lunit INSIGHT MMG with the Lancet Digital Health 2023 paired-reader detection endpoints, Dembrower 2023 paper record, lunit-insight-mmg.trials wired
- 24 September 2026: Issue 94: PSMA-PreRP (NCT03368547) trial record for Illuccix and 68Ga-PSMA-11 with the JAMA Oncology 2021 nodal accuracy endpoints, Hope 2021 paper record, illuccix.trials and ga68-psma-11.trials wired
- 24 September 2026: Issue 80: Vistogard Studies 1 and 2 (NCT01432301) trial record and Cancer 2017 paper; uridine-triacetate.trials wired with label link
- 24 September 2026: Issue 79: Study 0201 (NCT00003991) trial record and Blood 2006 paper; histamine-dihydrochloride.trials wired
- 24 September 2026: Issue 78: Study 201 (NCT00168064) trial record and JAMA Dermatology 2013 paper; mechlorethamine wired to the trial and to cutaneous-t-cell-lymphoma
- 24 September 2026: Issue 77: Study 2005-01 (NCT00358501) trial record and Blood 2016 paper; defibrotide.trials wired with label link
- 24 September 2026: Issue 76: CT041-ST-01 (NCT04581473) enriched in place with Lancet 2025 outcome and paper; satricabtagene-autoleucel.trials wired, false rivoceranib link and stale skip dropped
- 24 September 2026: Issue 75: MSB-GVHD001 (NCT02336230) trial record and BBMT 2020 paper; remestemcel-l.trials wired
- 24 September 2026: Issue 93: ROMAN (NCT03689712) trial record for avasopasem manganese with the eClinicalMedicine 2025 severe oral mucositis endpoints, Anderson 2025 paper record, avasopasem-manganese.trials wired; status mixed because the paper records the FDA complete response
- 24 September 2026: Issue 72: FOCUS (NCT02678572) enriched in place with Ann Surg Oncol 2024 outcomes; hepzato.trials wired with paper and label links
- 24 September 2026: Issue 71: eXalt3 (NCT02767804) enriched in place with JAMA Oncology 2021 outcomes; ensartinib cites the paper and label; stale zh TL;DR dropped
- 24 September 2026: Issue 70: Act.In.Sarc (NCT02379845) trial record and Lancet Oncology 2019 paper; nbtxr3.trials wired
- 24 September 2026: Issue 69: ROCKstar (NCT03640481) trial record and Blood 2021 paper; belumosudil.trials wired
- 24 September 2026: Issue 92: VISION (NCT02864992) enriched in place for tepotinib with the Paik NEJM 2020 response, duration and toxicity outcomes and the existing paper linked; capmatinib-tepotinib.trials wired to GEOMETRY mono-1 and VISION; geometry-mono-1 leads capmatinib.trials; simple sentence and Chinese TL;DR follow the record
- 24 September 2026: Issue 68: ARCHER 1050 (NCT01774721) enriched in place with separate Wu 2017 and Mok 2018 rows and papers; dacomitinib cites both
- 24 September 2026: Issue 67: ASCEND-4 (NCT01828099) enriched in place with Lancet 2017 outcomes and paper; ceritinib cites the paper
- 24 September 2026: Issue 66: CK-301-101 (NCT03212404) trial record and JITC 2023 paper; cosibelimab.trials wired, label and paper links replace the generic FDA listing
- 24 September 2026: Issue 65: ALTA (NCT02094573) trial record and JCO 2017 paper; brigatinib.trials wired to ALTA and ALTA-1L
- 24 September 2026: Glossary and record pages: checkpoint disambiguation term owns the bare word, Checkpoint families pills on target, term, pathway, technology and mechanics stage pages
- 24 September 2026: Checkpoint hubs: /checkpoints/ with a pill switch, /checkpoints/immune/ with the interactive synapse, /checkpoints/cell-cycle/ with the cell-cycle ring, class tables, JSON companions, nav entry in nine languages
- 24 September 2026: Checkpoint map: typed taxonomy of immune and cell-intrinsic checkpoints, 26 new targets from HGNC and UniProt, derived rows and tests
- 24 September 2026: Issue 91: Alliance A071401 (NCT02523014) trial record for GSK2256098 with the JCO 2023 six-month progression-free survival endpoints, Brastianos 2023 paper record, gsk2256098.trials wired
- 24 September 2026: Issue 90: PREEMPT CRC (NCT04369053) trial record for SimpleScreen CRC with the JAMA 2025 accuracy endpoints, Shaukat 2025 paper record, simplescreen-crc.trials wired
- 24 September 2026: Issue 89: TRANSCEND (NCT05774756) trial record for setmelanotide with the NEJM 2026 body-mass index and hunger endpoints, Miller 2026 paper record, setmelanotide refreshed with the 19 March 2026 US approval for acquired hypothalamic obesity and a CORRECTIONS row
- 24 September 2026: Issue 88: UX023T-CL201 (NCT02304367) trial record for burosumab with the JBMR 2021 serum phosphorus and fracture-healing endpoints, Jan de Beur 2021 paper record, burosumab.trials wired
- 24 September 2026: Issue 87: PATRICIA (NCT00122681, HPV-008) trial record for the bivalent HPV vaccine with the Lancet 2009 vaccine-efficacy endpoints, Paavonen 2009 paper record, hpv-bivalent-vaccine.trials wired
- 24 September 2026: Issue 85: XM01-22 (ISRCTN08063129) trial record for epoetin theta with the Archives of Drug Information 2011 haemoglobin response and transfusion endpoints, Tjulandin 2011 paper record, epoetin-theta.trials wired
- 24 September 2026: Issue 84: rolapitant Study 3 (NCT01500226, TS-P04834) trial record with the Lancet Oncology 2015 delayed-phase complete response endpoint, Schwartzberg 2015 paper record, rolapitant.trials wired and label links made explicit
- 24 September 2026: Issue 64: PALD1 (paladin) rewritten from the primary papers instead of the proposed GeneCards text; hand-written target records from reader proposals live in src/data/targets-reader-edits.ts and replace the generated gene record of the same id
- 24 September 2026: PR 74 residue decided by hand: six EU rows from the EMA register pages (olaratumab revoked 2019, pixantrone expired 2024, alitretinoin withdrawn 2021, cediranib application withdrawn 2016, ensartinib and senaparib positive CHMP opinions as under review); pixantrone status corrected to withdrawn, IMP4927 identified as senaparib (Sepalna) with PARP target and aliases, cediranib aliases; 15 rows closed as non-cancer medicines or pending
- 24 September 2026: Issue 83: XM22-03 (EudraCT 2009-015999-10) trial record for lipegfilgrastim with the BMC Cancer 2013 cycle 1 neutropenia endpoint, Bondarenko 2013 paper record, lipegfilgrastim.trials wired, id allowlisted as a phase 3 without an NCT
- 24 September 2026: Issue 82: MNTX 302 (NCT00402038) trial record for methylnaltrexone with the NEJM 2008 coprimary laxation endpoints, Thomas 2008 paper record, methylnaltrexone.trials wired and label links made explicit
- 24 September 2026: Gallbladder cancer geography: GLOBOCAN 2022 rates by country and sex for 185 countries, a reusable CancerGeography choropleth with incidence, mortality and sex pills, the high-burden regions with their papers, Chile's GES prophylactic cholecystectomy programme, India's NCRP registry rates and Japan's registry and guideline practice, prevention evidence, India and Chile spotlights, a geography.json companion and a gallbladder section on /countries/in/
- 24 September 2026: Target specificity explainer links the Human Protein Atlas collection record, whose licence now reads CC BY-SA 3.0
- 24 September 2026: Target specificity tests: enums cover every populated value, every row has a note and an https source, the graph carries the fields, HER2 reads tumour-associated and many types with a tumour-agnostic mark, BCR::ABL1 tumour-specific and one type, CD19 lineage antigen, PD-1 immune; browser facet counts match the data; the explainer and the pills render
- 24 September 2026: Target specificity in the pages: pills with glyphs and tooltips on target pages, a "Where it is found" section from the Human Protein Atlas rows, specificity and cancer-type facets on the target browser and the gene hub, and the explainer at /targets/specificity/
- 24 September 2026: Target specificity: blood-cancer family regex no longer swallows "(all types)" names, family names for the merged blood families, own-record sources for targets the atlases do not cover
- 24 September 2026: IMPROVEMENTS: rows 150 to 249, the 100 upgrade ideas of 24 September 2026
- 24 September 2026: Target specificity and distribution: schema, fetcher and data for 904 targets with an approved or clinical-stage drug
- 24 September 2026: Record pages stay inside a phone screen at 390 px; 130 more trials carry posted registry results
- 24 September 2026: Gallbladder decision aids and comparison: polyp aid on the 2022 European guideline, incidental cancer aid on AHPBA, Søreide, Ethun and CAPBIL, gallbladder compared with cholangiocarcinoma and ampullary cancer
- 24 September 2026: Record pages fit a 390 px phone and lose the Tabs key warning: grid children shrink (min-w-0 on Neighbours and What changed cards, *:min-w-0 on EntityDetail grids), GuidelineChip and DrugChip wrappers cap at max-w-full, tables in EntityDetail, InvestorPanels, StartupPanels, BiomarkerDetail, PrevalenceTable, ToxicityTable, AccessTable and ResearchOutput scroll inside ScrollRow with the year bars; each tab's content is a fragment keyed on the tab id so Flight no longer hands Tabs a keyless array; one record of every kind is a width-only check in scripts/mobile-audit.ts
- 24 September 2026: Registry passes over the gallbladder spike: outcomes for every NCT trial without its own, status re-read for all 4,102 NCT trials, key papers for 10 gallbladder trials
- 24 September 2026: LAUNCH: gallbladder deep spike of 24 Sept 2026
- 24 September 2026: Gallbladder cancer deep spike, final round: review reconciled the six facets to their sources, Wales and Northern Ireland figures, Cancer Drugs Fund status, prescription and benefit rules, lighter cancer pages
- 24 September 2026: Gallbladder deep spike review: duplicate terms, trials and history entries become supplements, HER2 and NICE figures reconciled, dead and redirected links fixed, backlinks for the ctDNA roadmap and the chemo-immunotherapy pairing, long relation lists capped with deep links so the cancer page drops from 997 KB to 652 KB, QA checklist in docs/GALLBLADDER-QA.md
- 24 September 2026: Gallbladder UK layer, second pass: Wales and Northern Ireland registry figures with stage, Scotland checked and found not to publish C23, CDF list v1.365 status on every biliary drug, four-nation prescription and special-rules benefit pages, GLH turnaround statements, CAPBIL tied to BILCAP, Hassan Malik record
- 24 September 2026: Gallbladder cancer deep spike, round three: treatments and trials, 131 registry trials, approval rows with NICE numbers, standard-of-care rows
- 24 September 2026: Gallbladder evidence: ARTEMIDE-Biliary01 points at its existing registry record instead of a second trial record
- 24 September 2026: Gallbladder cancer deep spike: the record rewritten with UK and world figures, eight subtypes, molecular landscape, UK and NHS pathway page, 40 papers, roadmap and ideas, decisions, first 60 days and red cards; weekly bot refreshes merged
- 24 September 2026: Gallbladder cancer treatments and trials: surgery, adjuvant, systemic, HER2 and palliation records with 131 registry trials
- 24 September 2026: Gallbladder cancer deep spike, first round: the record rewritten with UK and world figures, eight subtypes, molecular landscape, decisions, first 60 days and red cards
- 24 September 2026: Citation counts for the gallbladder evidence papers
- 24 September 2026: Gallbladder cancer roadmap, ideas, guideline and trial papers, terms and history
- 24 September 2026: Gallbladder cancer evidence: surgical, staging, UK, polyp, prevention, HER2 and ctDNA papers from Europe PMC
- 24 September 2026: UK and NHS page per cancer: /cancers/<id>/uk/, cancer-page pill, sitemap and JSON companion
- 24 September 2026: Gallbladder cancer: UK and NHS layer data, HPB centres, AMMF and UK trialists
- 24 September 2026: Gallbladder cancer record rebuilt with sourced UK, world and US figures, TNM 8th edition in plain words, risk factors with evidence, screening rules and patient FAQ; eight subtype pages and ten glossary terms
- 24 September 2026: Spike patches reach every cancer record wherever it lives, and fail loudly when the cancer id is missing; gallbladder landscape guard test
- 24 September 2026: Spike patches also merge into NCI coverage cancer records
- 24 September 2026: Gallbladder cancer molecular layer: sourced genomic landscape, regional cohorts, premalignant sequence, immune findings, HER2 testing, 30 landscape papers; spike supplements onto records other files own
- 24 September 2026: Gallbladder cancer: cancer-scoped red cards, dated first-60-days checklist, incidental-cancer journey
- 24 September 2026: Pipeline engine: a biparatopic antibody with one declared target resolves both arms at high confidence; zovodotin is a named auristatin payload
- 24 September 2026: Gallbladder cancer living-with spike: decision rows, question set and glossary records
- 24 September 2026: chore: weekly roadmap registry check (public/roadmap-watch.json)
- 24 September 2026: chore: refresh pulse, abstract and citation snapshots (#86)
- 24 September 2026: chore: refresh institution research output from OpenAlex (#81)
- 24 September 2026: chore: EU regional rows read from the EMA register pages (#73)
- 23 September 2026: Top bar fits at 1280 px with a classic scrollbar
- 23 September 2026: LAUNCH: owner asks of 23 Sept 2026
- 23 September 2026: Drugs browser fits its card with capped cells and clearer approval years; NHS-Galleri performance paper, trial outcomes and Galleri test page
- 23 September 2026: Header fits a 1280 px window with a classic scrollbar: star count folds into its icon, tighter nav gaps and 13 px labels between xl and 2xl
- 23 September 2026: NHS-Galleri test-performance paper (Nature Medicine, 22 Sept 2026) as a paper record, trial rebuilt on the registry and the abstract, result on Galleri, MCED and the ctDNA roadmap
- 23 September 2026: Drugs browser fits at 1280 and 1440: capped list cells with a "+N more" pill, Approved as first year plus muted latest with region flags, results tables scroll inside their card only when measured too wide
- 23 September 2026: Biomarker readouts under their parent genes at /biomarkers/, genome hub pages its genes
- 23 September 2026: Gene hub pages its role sections: first 30 genes per role in the HTML, the rest from one tables file, role and evidence deep links over the full set
- 23 September 2026: Proposals bot merges register-verified rows itself, mechanics of cancer rebuilt with 56 stage pages, mobile patterns for views where a control drives a visual
- 23 September 2026: Regenerate target cross-references for the seven new parent targets (HGNC and ChEMBL, 23 Sept 2026)
- 23 September 2026: Dependency map gets a touch path, and wide matrices scroll inside their own box
- 23 September 2026: Biomarker pages: /biomarkers/ browser with gene, measurement, cancer and approval facets; readout pages with a Part of strip; target, drug, term and tumour-test pages link their readouts
- 23 September 2026: Biomarker readouts: 80 records under 40 parent genes, thresholds quoted from FDA labels and the companion diagnostic list
- 23 September 2026: Add the biomarker kind: schema, KIND_META, glyph, colours, nav and kind labels in nine languages
- 23 September 2026: Mechanics atlas: a page per stage and a light journey hub
- 23 September 2026: Cancer gene layer of 1,447 genes, open-source oncology map, evidence-backed combination proposals on the engine, daily spotlight, iodine target correction
- 23 September 2026: LAUNCH: proposals bot, what merges itself and what waits
- 23 September 2026: Nightly proposals: auto branch squash-merged after the gates, review PR labelled needs-review with a reason on every EMA row
- 23 September 2026: Proposals bot decides EU rows from the EMA medicine pages: page status, exact attribution, biosimilar rule, drift check, apply script
- 23 September 2026: Regional approvals io helper: parse and write the hand-written file with a round-trip test, and R() takes a source
- 23 September 2026: Two-column tools get a phone layout: sticky preview, Choose and View pills, inline results
- 23 September 2026: Cancer genes, tier 4: 389 genes associated with cancer without a proven role
- 23 September 2026: Cancer genes, tier 3: 332 drivers called by cohort mutation analysis
- 23 September 2026: Cancer genes, tier 2: 507 genes with clinical evidence on their variants
- 23 September 2026: Cancer genes, tier 1: 219 genes with an approved or late-stage drug recorded against them
- 23 September 2026: Cancer gene layer: target roles, evidence tiers and sources, fetch-cancer-genes script, genome hub and browser facets
- 23 September 2026: Home spotlight rotates by kind each day: one hero record per kind (most connected), build day's set in the HTML, reader's day swapped in from /api/v1/spotlight.json without layout shift
- 23 September 2026: ADC payload fields and setidegrasib note: record only what the records say
- 23 September 2026: Open pipeline engine, tag power search pages and full-width tabs, Edge type and For you filters, API licence wording, For me browser-only notice
- 23 September 2026: Open-source catalogue tests: unique ids, https URLs, SPDX or placeholder licences with notes, every corpus id exists, 200 sources on the generation date, curated entries generated or explained
- 23 September 2026: Open source in oncology page: paged browser with category, licence, language, openness and activity facets, intro on how oncology is done in the open, and cross-links from technology, collection, organisation and data-source pages
- 23 September 2026: Open-source oncology catalogue: schema, curated list of 379 projects and the fetcher that reads licence, stars and dates from the GitHub API
- 23 September 2026: Radioactive iodine no longer linked to SSTR2: I-131 enters thyroid cells through the sodium-iodide symporter, which has no target page yet
- 23 September 2026: Document the open pipeline: how proposals are scored, searched and contributed
- 23 September 2026: Sixty combination proposals, each searched against trials, papers and patents
- 23 September 2026: Schema for combination proposals: a scored, evidence-backed cell of the modular-drug grid
- 23 September 2026: api layout: drop the entries duplicated by the engine merge
- 23 September 2026: Record pages: full-width tab strip above both columns; tag chips open /tagged/ power-search pages
- 23 September 2026: API page licence wording matches the footer and about page; For me says it is browser-only with a signed-in version coming
- 23 September 2026: Open drug engine: every medicine taken apart into its modules, and the permutation grid per format with each cell classified from records
- 23 September 2026: Edge: filter by type or For you, with URL state, per-cancer deep links and per-type feeds
- 23 September 2026: Search ranks cancers and treatments first and groups results by kind
- 23 September 2026: Search ranks cancers and treatments above journals and papers, with kind-grouped dropdowns
- 23 September 2026: Top bar without the tagline, with the weekly bot refreshes merged
- 23 September 2026: Top bar without the tagline
- 23 September 2026: Top bar: drop the tagline beside the wordmark; it took space the navigation needs
- 23 September 2026: chore: refresh universe lists (NCI drugs and types, NCI centres, OECI, NHS alliances, NLM journals, OpenAlex, KEGG, ChEMBL, ClinicalTrials.gov, FDA OCE) (#61)
- 23 September 2026: chore: check regional approvals against the EMA register (#60)
- 23 September 2026: chore: refresh preprint tracker from Europe PMC (#59)
- 23 September 2026: chore: refresh FDA approvals feed (OCE notifications, openFDA drugsfda) (#58)
- 23 September 2026: chore: weekly citation link check (#57)
- 22 September 2026: Health gauges explain people without papers by role and exempt the closed section list
- 22 September 2026: Kind-size gauge holds open kinds only: the 19 treatment fronts are a closed OnCo taxonomy and are named in the note
- 22 September 2026: People-papers gauge counts roles that produce no papers as explained: 1,528 of 1,547 with a note on who is still without
- 22 September 2026: LAUNCH: chains 101 to 104 ticked
- 22 September 2026: Kind browser pages ship 60 rows and load the rest on demand, 402 registry trials for drugs and technologies, glossary terms without a Wikipedia article say so
- 22 September 2026: Kind browsers page their rows: first 60 in the HTML, the rest from /api/v1/tables/kind-<route>.json
- 22 September 2026: Content roadmap: wave 5 trial run recorded, headline numbers refreshed from content:gaps
- 22 September 2026: Wave 5 technologies: 32 technologies without a trial gain 65 registry trials and 7 corpus links from ClinicalTrials.gov
- 22 September 2026: fetch-entity-trials: hold technologies whose aliases name something else, stop words for ex vivo, Catalyst, STRIDE and hormone names
- 22 September 2026: Wave 5 drugs: 142 drugs without a trial gain 337 registry trials and 19 corpus links from ClinicalTrials.gov
- 22 September 2026: 75 glossary terms marked wikipediaChecked 2026-09-22; NO_ARTICLE comment lists retired
- 22 September 2026: fetch-entity-trials: ClinicalTrials.gov trials for drugs and technologies without one (wave 5, second half)
- 22 September 2026: Terms record a checked-and-absent Wikipedia state: wikipediaChecked field, gauge counts it as explained, page says so
- 22 September 2026: fetch-term-wikipedia: Wikipedia links for glossary terms, exact title or redirect match with a sense check
- 22 September 2026: Paper pages for 736 cited DOIs, 1,381 bare references now link to them
- 22 September 2026: Wave 7 batch 4: the last 155 paper pages for cited DOIs, 736 in all, 1,161 citing records linked; Progress row
- 22 September 2026: fetch-cited-papers: a preprint is "posted on bioRxiv", not "published in bioRxiv (preprint)"
- 22 September 2026: Trial statuses follow the registry with dated notes, iobenguane records merged
- 22 September 2026: Wave 7 batch 3: 194 more paper pages for cited DOIs, 581 in all, 1,042 citing records linked; intismeran autogene gains its aliases
- 22 September 2026: Wave 7 batch 2: 192 more paper pages for cited DOIs, 387 in all, 905 citing records linked
- 22 September 2026: Wave 7 batch 1: 195 paper pages for the most-cited DOIs without one, 785 citing records linked
- 22 September 2026: fetch-cited-papers: name citing pages by kind and count only, no year in TL;DR or id
- 22 September 2026: Registry status data: 13 trial status changes from 3,542 ClinicalTrials.gov records (active -> completed 5, active -> recruiting 3, recruiting -> active 2, completed -> withdrawn 2, active -> withdrawn 1) with dated notes and why-stopped text; iobenguane-i-131 (Azedra) merged into i131-mibg with its approval, brand, aliases, links and a redirect stub
- 22 September 2026: Registry status pass: scripts/fetch-registry-status.ts reads overallStatus, whyStopped and the registry dates for every trial with an NCT id and writes src/data/trial-registry-status.ts, applied in src/data/index.ts through applyRegistryStatus with a dated note; result statuses kept, hand-written statuses newer than the registry never downgraded; iobenguane-i-131 listed as a duplicate of i131-mibg in dedupe MERGES
- 22 September 2026: Wave 7 fetcher for cited DOIs: scripts/fetch-cited-papers.ts writes a paper page per DOI a record cites without one
- 22 September 2026: Explained page pages its sections, four duplicate companies and one duplicate drug merged with redirects
- 22 September 2026: Page /explained/ sections at 10 rows: 1,135 KB to 561 KB of markup, budget lowered to 600 KB
- 22 September 2026: LAUNCH: content waves through chain 100
- 22 September 2026: Merge five duplicate records: Orca Biosystems into orca-bio, InxMed (Shanghai) into inxmed, Cellectis S.A. into cellectis, the second Valius Sciences into valius, and the gap-fill nogapendekin alfa inbakicept into nogapendekin-alfa; trials, cancers, drugs, investors, aliases, links and a logo move to the kept records, 30 references repointed, sponsor aliases updated, stale simple text and translations dropped, redirects and static stubs added; biocon/biocon-biologics, jw-pharmaceutical/jw-therapeutics, astrazeneca/fusion-pharma and gtx/oncternal-therapeutics reviewed and kept apart
- 22 September 2026: Duplicate-record finder and merger for companies: scripts/dedupe-companies.ts flags pairs by normalised name (legal suffixes, descriptors and city words stripped), shared website domain, or shared Wikidata or Wikipedia identifier, prints hand-reviewed MERGE and SKIP verdicts, and with --apply moves the loser's fields onto the winner, removes the loser literal, rewrites every reference across src/data (dropping duplicates inside id arrays, renaming or dropping keyed simple, translation and wikidata entries), moves or deletes the logo and registers the redirect build-redirect-stubs.ts expects; institutions and drugs are report-only unless listed as obviously identical; the source-editing helpers in scripts/orphan-links.ts are exported for reuse
- 22 September 2026: Papers for people and ideas, trials for ideas and drugs for companies from wave 6
- 22 September 2026: Content roadmap: wave 6 progress row (people without a paper 318 to 300, ideas without a key paper 1,104 to 1,026, ideas without a trial 1,047 to 1,044, companies without a drug 720 to 718) with what was held back and why
- 22 September 2026: Wave 6 company drugs: 2 companies gain a drug through the ClinicalTrials.gov lead-sponsor field (Impact Therapeutics IMP4927, R-Pharm RPH-051), each verified on the registry record; 75 other sponsors held back because their corpus trials name only comparators with a maker on record
- 22 September 2026: Registry results for 144 trials, 24 more standard-of-care trials with outcomes, papers and standard-of-care rows linked
- 22 September 2026: Wave 6 idea evidence: 78 ideas gain key papers (124 new Europe PMC records, 98 existing reused) and 3 gain trials (4 ClinicalTrials.gov phase 2 or 3 records written, 4 corpus trials linked), matched on the drug or target the idea's text names plus its cancer
- 22 September 2026: heavy-pages test: explained budget follows the corpus after the trial waves, with the paging rule for the next rise
- 22 September 2026: Content gaps: the trial-acronym measure treats the programme stem before a hyphen as a known trial name; roadmap notes the 33 remaining tokens are all in the wave SKIP list
- 22 September 2026: Content wave 3, rare cancers: FIRSTMAPPP, InPACT, GETUG 13, ADIUVO, ADIUVO-2, TROPHIMMUN and BALLAD, 4 with structured outcomes, 4 primary papers, 13 standard-of-care rows linked
- 22 September 2026: Content wave 3, myeloma: DRAMMATIC (SWOG S1803, MRD-guided maintenance after transplant) recorded from the registry and linked from the myeloma maintenance row; NCI sponsor-of-record institution links removed across the wave
- 22 September 2026: Content wave 3, endometrial: RAINBO programme (four molecular-class trials including POLEmut-BLUE de-escalation) recorded from the registry; POLE-ultramutated standard-of-care row linked
- 22 September 2026: Roadmap Progress: wave 7 registry outcomes row (144 of 2,845 ingested trials gain 272 outcome rows from 153 records with posted results; 114 requests)
- 22 September 2026: Wave 7 data: registry outcomes for the first 2845 registry-ingested trials by family burden (144 trials with 272 outcome rows from 153 records with posted results)
- 22 September 2026: Content wave 3, ovarian: AURELIA (bevacizumab with single-agent chemotherapy in platinum-resistant ovarian cancer) with structured outcomes and its JCO 2014 paper; two standard-of-care rows linked
- 22 September 2026: Wave 7 data: registry outcomes for the first 2500 registry-ingested trials by family burden (144 trials with 272 outcome rows from 153 records with posted results)
- 22 September 2026: Content wave 3, head and neck: CONTINUUM (sintilimab added to induction-concurrent chemoradiotherapy in locoregionally advanced nasopharyngeal carcinoma) with structured outcomes and its Lancet 2024 paper; standard-of-care row linked
- 22 September 2026: Wave 7 data: registry outcomes for the first 2000 registry-ingested trials by family burden (111 trials with 209 outcome rows from 117 records with posted results)
- 22 September 2026: Wave 6 people papers: 18 people gain up to three most cited papers each from Europe PMC by author and affiliation (46 new paper records, 6 existing reused), administrators filtered by role before searching, 2 ambiguous names skipped with reasons
- 22 September 2026: Content wave 3, colorectal: CAPP2 (aspirin in Lynch syndrome, 10-year follow-up) with structured outcomes and its Lancet 2020 paper; Lynch syndrome standard-of-care row linked
- 22 September 2026: Wave 7 data: registry outcomes for the first 1500 registry-ingested trials by family burden (85 trials with 171 outcome rows from 91 records with posted results)
- 22 September 2026: Wave 7 data: registry outcomes for the first 1000 registry-ingested trials by family burden (61 trials with 126 outcome rows from 65 records with posted results)
- 22 September 2026: Content wave 3, lung: IMpower010 is the existing registry record nct02486718, which gains the IMpower010 alias, its positive status and disease-free survival outcomes from the registry results section and the Lancet 2021 report; JCOG0802 skipped (UMIN only)
- 22 September 2026: Content wave 3, anal family: ANCHOR, POD1UM-303/InterAACT-2, InterAACT and NCI9673 with structured outcomes, 5 primary papers, 6 standard-of-care rows linked
- 22 September 2026: Explained page: the Open explanation pill carries only its label; glyph and Close state move to CSS; row classes and cross-reference titles shortened
- 22 September 2026: Content wave 3, sarcoma family: ACTG A5263/AMC 066 (paclitaxel against etoposide or bleomycin-vincristine with antiretroviral therapy for AIDS-associated Kaposi sarcoma) with structured outcomes and its Lancet 2020 paper; Progress table rows for gastric, pancreatic and sarcoma
- 22 September 2026: Content wave 3, pancreatic family: 7 standard-of-care trials (ESPAC-3, ESPAC-4, CONKO-001, NORPACT-1, LAP07, ESPAC-5, PRECEDE), 6 with structured outcomes, 6 primary papers, 7 standard-of-care rows linked
- 22 September 2026: LAUNCH: chains 97 and 98 ticked
- 22 September 2026: Wave 7 data: registry outcomes for the first 500 registry-ingested trials by family burden (39 trials with 78 outcome rows from 40 records with posted results)
- 22 September 2026: Wave 7 script: fetch-registry-outcomes.ts copies structured outcomes from the ClinicalTrials.gov results section into a side file merged by index.ts
- 22 September 2026: Content wave 3, gastric family: CLASSIC (adjuvant CAPOX after D2 gastrectomy) with structured outcomes and two primary papers; SKIP list extended for UMIN- and ISRCTN-only trials
- 22 September 2026: Wave 1 trial papers from Europe PMC, wave 3 standard-of-care trials, wave 5 targets and EMA products and sponsors, seven heavy table pages and idea rankings load rows on demand
- 22 September 2026: Paged tables: the heavy StaticTable pages carry 30 rows and fetch the rest from /api/v1/tables/
- 22 September 2026: Content wave 3: structured outcomes for 8 hand-written trials from the ClinicalTrials.gov results section (ARANOTE, CheckMate 915, PACIFIC-2, INTELLANCE-1, ATEMPT, ASPEN, DESTINY-Gastric02, AREN0532); roadmap Progress notes the families not reached
- 22 September 2026: Content wave 3, lymphoma family: 7 standard-of-care trials (FoRT, GALLIUM, RELEVANCE, TRIANGLE, SYMPATICO, BMT CTN 0803, PTLD-1), all with structured outcomes, 7 primary papers, 9 standard-of-care rows linked
- 22 September 2026: Content roadmap: wave 5 progress row (targets 30 to 0, EMA products 12 to 0, sponsor companies 21 to 15) with what was not written and why
- 22 September 2026: Wave 5 sponsor companies: 6 trial sponsors and 2 EMA authorisation holders from Wikidata and the ClinicalTrials.gov sponsor field (Criterium, Artios Pharma, Ambrx, Nerviano Medical Sciences, Qurient, Immunovaccine, Grünenthal, 4SC); drugs and companies waves registered
- 22 September 2026: Wave 5 EMA products: 7 register products written from their EPAR pages (crisantaspase, burosumab, epoetin theta, human normal immunoglobulin, fentanyl, resminostat, Tacquell), 3 matching faults fixed with aliases, vedolizumab excluded
- 22 September 2026: Wave 5 targets: 26 pages for the symbols two or more drug records name without a target page (NRAS, CD28, HLA-A, MPL, SHP2, STAT5, IKZF1, JAK1, CD80, ABL1, FAK, IKZF3, PTCH1, CD137, HIF-1α, IDO1, BRD4, IRF4, B7-H4, TET2, NF1, ACVR1, HOXA9, MEIS1, BIM, BAX)
- 22 September 2026: Content wave 3, childhood family: 9 standard-of-care trials (ANBL1221, ANBL0531, ANBL17P1, AEWS0031, SIOPEL-3, SIOPEL-4, AGCT1531, AHOD1331, EuroNet-PHL-C2), 7 with structured outcomes, 6 primary papers, 11 standard-of-care rows linked
- 22 September 2026: Content wave 3, brain and CNS family: 10 standard-of-care trials (ACNS1123, ACNS0122, ACNS0333, ACNS0121, NOA-08, SJMB12, ACNS1422, IELSG32, MATRix/IELSG43, INTUITT-NF2), 8 with structured outcomes, 8 primary papers, 12 standard-of-care rows linked
- 22 September 2026: Content wave 3, leukaemia family: 18 standard-of-care trials with registry records and structured outcomes, 12 primary papers, standard-of-care rows linked
- 22 September 2026: Fetched paper records link only their trial and journal; wave 1 progress row; citation counts for the new papers
- 22 September 2026: Idea rankings page carries the first 30 rows of each view and fetches the rest per view from /api/v1/ideas/rankings/<view>.json
- 22 September 2026: Prostate trials: key papers from Europe PMC for the prostate family
- 22 September 2026: Skin trials: key papers from Europe PMC for the skin family
- 22 September 2026: Colorectal trials: key papers from Europe PMC for the colorectal family
- 22 September 2026: Breast trials: key papers from Europe PMC for 77 trials, 85 new paper records, 22 long-term follow-ups
- 22 September 2026: Lung trials: key papers from Europe PMC for 125 trials, 134 new paper records, 25 long-term follow-ups
- 22 September 2026: Cancer map DAG, every hand-written table on the shared header filter, explained page loads explainer bodies on demand, 3,056 corpus-only cross-links
- 22 September 2026: GuidelineConcordance keeps the guideline body's website link: a small external-link glyph beside each stance chip in the row's cell (sibling anchor, Tab-reachable, tooltip "Open the body's website"), with the test asserting one per body cell and no nested anchors
- 22 September 2026: Wave 2 corpus-only joins applied: 3,056 links from fields the records already carry, links still to make down from 12,295 to 10,335 Trial to technology 2,743 links on 1,739 trials (modality technologies of the studied agents); technology to trial 211 links, 26 technologies gaining their first trial; idea to trial 133 links on 30 ideas and idea to key paper 46 links on 32 ideas, only where the idea's text names the drug or target in the idea's cancers; drug to key paper 34 links on 25 drugs; drug to trial 35 links where the title names the drug; target drugs arrays 47 links on 21 targets; company to drug 6 links on 5 companies. No EU approval rows written: 59 of the 60 snapshot candidates already had rows and the last is a withdrawn application with no date. No target-to-drug mechanism links: both measured hits were substrings.
- 22 September 2026: Corpus-only joins script: scripts/link-joins.ts plans and applies wave 2 links from fields the records already carry, with every rule stated; content-gaps measures corrected Ten rules, dry run by default and --apply to write through the editor in scripts/orphan-links.ts (append only, never removes a link, sorted and de-duplicated): trial to technology through the studied agents' modality technologies; idea to trial and idea to key paper only where the idea's own text names the drug or target, in the idea's cancers, for actors that trials test, capped at five and three; company to drug through sponsored trials where the drug has no maker; drug to key paper and drug to trial through titles that name the drug; technology to trial; target drugs arrays mirroring drug.targets; target to drug by whole-word mechanism mention; EU approval rows from the EMA register snapshot with the file's A, C, V and W helpers, product name and snapshot date checked.
- 22 September 2026: Header filters for the toolbar facets that had no column: CountryRanking gains a Region column, ScorecardTable Type and Country columns (chips that select the value) and a Regions header filter, PipelineFunnel filters the crowding table by approved product and population completeness, ResearchRanking by parent university and five-year coverage; docs/TABLES.md carries every table's state
- 22 September 2026: Header filters on the client tables that keep their own controls: TrialFinder, TrialFinderGeo, AssistanceBrowser, GuidelineConcordance, ManufacturingMap, InteractionChecker, AutoPulse, SurvivorshipPlan, MarketEstimator, TumorBoard, Navigator and PlanRankings filter from their column headers through useHeaderFilters and FilterHead, sharing state with the existing facet controls where one exists; CaregiverPanel skipped (eight rows, no categorical column)
- 22 September 2026: Trial key papers fetcher: Europe PMC by registry id, one primary and one follow-up per trial, links merged from trial-key-papers-wave1.ts
- 22 September 2026: Header filters on every server-rendered table: StaticTable draws plain rows from the page (links, chips, bars, avatars as small objects) so deals, evidence, funding, universities, payloads, costs, status, freshness, eval questions, survival, cited papers, papers pulse, startup investors, audit, review, China trials and deals, India sites, pathway matrix and dossier trials all filter from their column headers with a N of M count and a Clear pill; docs/TABLES.md records every table's state and the skips under ten rows
- 22 September 2026: Trials in plain words keeps every trial's heading, status, summary and page link in the HTML and fetches the explanation on demand: one file per cancer section, prefetched as the section scrolls near and opened with an Open explanation pill
- 22 September 2026: LAUNCH: owner asks of 22 Sept 2026 and what shipped in chain 96
- 22 September 2026: Column headers filter every browser table, tumour testing redesign with laboratory logos, Edge feed, idea and corpus rankings, explore pagination, content roadmap, PR 54 and 55 rows, signed-in bridge
- 22 September 2026: Cancer map: a layered DAG of every cancer type at /cancers/map/, with organ systems, histology groupings, cancers and subtypes, badges switchable between trials, products, approvals and ideas, an accessible outline, graph.json for agents, and links from the cancer browser and every family strip
- 22 September 2026: Rankings hub: fourteen league tables computed from corpus counts and dates, one page each with basis, coverage and a JSON companion
- 22 September 2026: Content roadmap: pages still to write and cross-connections still to make, measured by scripts/content-gaps.ts; waves 143 to 149 on the roadmap
- 22 September 2026: Edge: one feed of the freshest signals in cancer, new papers, trial results, approvals, withdrawals, law, bot proposals and weekly issues, ranked by recency with a modest importance weight, with Atom and JSON feeds at /edge/feed.xml and /edge/feed.json
- 22 September 2026: Tumour testing: By scope and By sample become card grids with company logos, and the table filters by sample, scope, laboratory and status from the URL query
- 22 September 2026: Idea rankings: six views of the same ideas at /ideas/rankings/, every score from existing fields with its formula shown
- 22 September 2026: Explore, For me and the navigator page their sections: the first 30 rows of each kind ship in the HTML and the rest load per section from static JSON as the reader scrolls or presses Show 30 more; For me and the navigator fetch one file per cancer instead of carrying all 328
- 22 September 2026: Filter from the column header: every filterable column in the results tables is a button with a funnel glyph that opens a popover of value pills (counts, search past eight, Clear, Close; bottom sheet on phones), wired to the same setter and URL as the toolbar facets; sortable headers keep sort on the label and put the filter on a sibling glyph
- 22 September 2026: Four EMA application withdrawals recorded as withdrawn rows, melflufen US withdrawal row restored, FDA Tazverik communication dated 11 May 2026
- 22 September 2026: Bridge to the signed-in site: header greets a signed-in reader, Watch keeps pages on the account with a sign-up offer, a star on every list row, For me can be saved to the account; consent copy
- 22 September 2026: chore: weekly roadmap registry check (public/roadmap-watch.json)
- 22 September 2026: Upkeep runs on its own: weekly logos, structures, trial links and acronyms, monthly identifiers, and the bots' refreshes now deploy without a manual ship
- 22 September 2026: Weekly maintenance and monthly identifier workflows; bot pushes now deploy through the Vercel ignore test
- 22 September 2026: chore: refresh literature snapshot from Europe PMC (#56)
- 22 September 2026: Add FDA safety communication link and SPM rate to epithelioid sarcoma record
- 21 September 2026: Trial phases read as plain labels everywhere: Platform trial and Observational study instead of Phase platform, with old filter links still working
- 21 September 2026: Phase facet test asserts the string key and the shared normaliser
- 21 September 2026: Facet normaliser is a string key resolved in the client, not a function crossing the server boundary
- 21 September 2026: Trial phases read "Observational study" and "Platform trial" instead of "Phase observational", from one phaseLabel helper
- 21 September 2026: A JAMA Oncology 2026 comparison of tissue-free and tumour-informed residual-disease blood tests joins the ctDNA roadmap's sources
- 21 September 2026: c-TRAK TN tissue-free residual-disease assay paper added and linked from the tumour-informed assay term
- 21 September 2026: Trial pages say what their enrolment number counts, randomised, analysed or enrolled, with the registry figure explained on hover
- 21 September 2026: Four new publications found by the registry check are in, including the eight-year ADAURA survival update with its figures, and the check no longer flags screening studies
- 21 September 2026: Registry check: a non-drug interventional study matches the corpus's observational phase, clearing the PATHFINDER 2 flag
- 21 September 2026: Four publications flagged by the roadmap registry check recorded as key papers: the ADAURA eight-year overall survival update (J Thorac Oncol 2026) linked from the trial and the EGFR-mutant and resectable NSCLC pages with its stage II-IIIA and IB-IIIA survival figures added to the trial's outcomes, and the FLAURA2 long-term safety analysis (Lung Cancer 2026), the KRYSTAL-12 plain language summary (Future Oncol 2026) and the TROPION-Breast01 China cohort (ESMO Open 2026) linked from their trials only; citation counts fetched from Europe PMC
- 21 September 2026: Trial enrolment figures now say what they count: 25 landmark trials explain why their randomised or analysed numbers differ from the registry, and five stale figures are corrected from their sources
- 21 September 2026: Enrolment gaps on 30 roadmap trials reviewed one by one against ClinicalTrials.gov and the primary papers: 25 keep the paper's randomised, analysed, treated or registered count and now say so in a new enrolledBasis and enrolledNote, 5 stale or unsourced figures corrected from the registry or the paper; the roadmap watch reports explained gaps separately and its offline mode re-compares against the committed registry records
- 21 September 2026: MIMic-01 trial status corrected to active against ClinicalTrials.gov, the first fix from the new roadmap registry check
- 21 September 2026: Every roadmap now carries a registry check: a weekly script compares each referenced trial with ClinicalTrials.gov and Europe PMC, flags contradictions and overdue readouts, and shows the check date under What to watch
- 21 September 2026: MIMic-01 status and PATHFINDER 2 design corrected against ClinicalTrials.gov, as flagged by the roadmap watch
- 21 September 2026: Roadmap watch for every roadmap: scripts/roadmap-watch.ts compares each roadmap's referenced trials with ClinicalTrials.gov (status, phase, enrolment within 5 percent, moved completion dates), lists Europe PMC papers since asOf, flags watch items past their expected date, writes public/roadmap-watch.json for a weekly Thursday workflow and a "Registry check" line on roadmap pages
- 21 September 2026: A roadmap and timeline for circulating tumour DNA tests, from the 1948 discovery of cell-free DNA to the residual-disease and multi-cancer detection trials reading out next, with a watch list and a script that checks it against the registries
- 21 September 2026: PATHFINDER 2 status and ECLIPSE enrolment corrected against ClinicalTrials.gov, as flagged by the ctDNA watch script
- 21 September 2026: ctDNA tests roadmap: ten eras from the 1948 sighting of DNA in plasma to the 2029 randomised readouts, a "What to watch" list with registry and meeting dates, six trials the corpus lacked (TRACC, MERMAID-1 and -2, BESPOKE, CIRCULATE-US, NCI Vanguard), a tumour-informed versus tumour-naive term, and npm run ctdna:watch to keep it current
- 21 September 2026: Twenty-six landmark papers for the pancreatic cancer pages, from PRODIGE 24 and PREOPANC to the KRAS inhibitor cohorts and the IPMN guidelines, each checked against Europe PMC
- 21 September 2026: Every pancreatic subtype page now rests on two to four landmark papers: 26 new records (PRODIGE 24, ESPAC-4, PREOPANC, ESPAC5, Alliance A021501, LAP07, CodeBreaK 100 and KRYSTAL-1 pancreatic cohorts, POLO and its survival analysis, the BRCA/PALB2 platinum trial, KEYNOTE-158, the NRG1 fusion discovery and zenocutuzumab papers, Fukuoka, Kyoto and European cyst guidelines, and the acinar cell carcinoma and pancreatoblastoma reference series), each DOI checked on Europe PMC and doi.org, with Le 2017 reused for the MSI-high page and citation counts fetched
- 21 September 2026: Seventy-four registry trials linked to the newest pancreatic and nasopharyngeal cancer pages, with a guard that keeps neuroendocrine trials off the pancreatic adenocarcinoma pages
- 21 September 2026: Registry trials now reach the 50 subtype pages added this week: 74 trials linked to the pancreatic disease-state and driver pages and to recurrent or metastatic nasopharyngeal carcinoma, with a neuroendocrine guard so GEP-NET trials stay off the ductal adenocarcinoma pages
- 21 September 2026: Pages load about three times less script: the header no longer drags in the molecule viewer, every language dictionary and the search engine, so slow connections reach a working page far sooner
- 21 September 2026: Bundle split follow-ups: stars JSON out of the guarded data folder, layout test reads dictionaries from the full set, row 140 done
- 21 September 2026: Seven European approvals corrected to withdrawn against the EMA register, the PARTIQoL proton trial gains its reported result, and the KEYNOTE-A18 papers are split into their two Lancet reports
- 21 September 2026: Client JavaScript on every page cut from 3.0 MB to 1.0 MB: the layout no longer pulls the animated schematics, the structure index, all eight chrome dictionaries, zod or MiniSearch
- 21 September 2026: Paganetti 2014 RBE reference points at the real paper (Physics in Medicine and Biology)
- 21 September 2026: Visit counting now asks first: Google Analytics loads only after you allow it, the browser's privacy signal is honoured, and the choice can be changed in the footer
- 21 September 2026: Data follow-ups: seven withdrawn EU authorisations, PARTIQoL readout, KEYNOTE-A18 split into its two Lancet papers
- 21 September 2026: Roadmap row 141 marked done for onco.cc
- 21 September 2026: Google Analytics now loads only after the visitor presses Allow: a consent bar at the foot of the viewport, the choice kept in the browser, Global Privacy Control read as No thanks, an Analytics choice control in the footer and on /privacy/, strings in nine languages
- 21 September 2026: GitHub star count is baked in at build time, so the top-bar badge is full from the first paint and never flickers
- 21 September 2026: Launch doc: two dated account notes marked as removed
- 21 September 2026: Launch doc: old account instructions cut to one-line history notes pointing at me.onco.cc
- 21 September 2026: Sign in/up opens in the same tab: links to me.onco.cc are treated as part of the site rather than external
- 21 September 2026: Sign in/up in the top bar now opens the WorkOS sign-in directly, with no stop on a landing page
- 21 September 2026: Sign in/up on onco.cc now hands you to me.onco.cc, the signed-in side; onco.cc itself keeps no account data, in the browser or the cloud
- 21 September 2026: Sign in/up now links to me.onco.cc; onco.cc holds no account data
- 21 September 2026: Sign in/up now goes straight to the WorkOS sign-in page instead of stopping at the newsletter page
- 21 September 2026: Greptile review config encoding the corpus rules for community pull requests
- 21 September 2026: Your role, cancer choice and preferences now save to your OnCo account so they follow you across devices, and the outlook block is held back pending a wording review
- 21 September 2026: Merge cloud profile sync; outlook block held pending wording review
- 21 September 2026: Community fixes: alphagenome model corrected, epithelioid sarcoma record completed with verified facts, RMC-9805 pancreatic pairing and ubamatamab mechanism corrected, plain-English outlook for ten cancers behind the survival disclosure, and 15 hidden FDA events restored
- 21 September 2026: Cloud profiles stage one: Supabase tables under the WorkOS token, sync, delete, privacy text
- 21 September 2026: PR round 2: alphagenome weights gated, FDA regulatory events surfaced, epithelioid sarcoma withdrawal history
- 21 September 2026: Issues 43, 45, 46, 47: prognosis outlook on ten cancer pages, ubamatamab as MUC16xCD3 bispecific with corrected citation, g12d-plus-pan-ras phase and links from the registry and company release
- 21 September 2026: Roadmap rows 141 and 142: GDPR compliance for accounts and HIPAA readiness
- 21 September 2026: chore: render this week's issue to public/newsletter/ (#52)
- 21 September 2026: chore: refresh public/votes.json from Discussions reactions (#51)
- 21 September 2026: chore: weekly fact check, audit, and provenance refresh (#50)
- 21 September 2026: chore: refresh trial counts from ClinicalTrials.gov (#49)
- 20 September 2026: Fix duplicate entry for 'alphagenome' in model registry
- 18 September 2026: Health data compliance analysis: HIPAA does not apply today, GDPR and state health-data laws do; stay browser-only and prepare encrypted sync
- 18 September 2026: Top bar sign-in works even before the page script loads: it is now a link that upgrades to the direct WorkOS start
- 18 September 2026: Top bar button shortened to Sign in/up in every language, and cached pages refreshed so everyone sees the new header
- 18 September 2026: After sign-in a welcome page asks who you are and which cancer, the top bar shows you are signed in, four community pull requests and four issue fixes land, and secret scanning guards every commit
- 18 September 2026: Fix issues 42, 39, 37, 36: chip ellipsis and titles, daraxonrasib approval, clickable tooltips, urothelial basics
- 18 September 2026: Welcome step as its own page: sign-in leaves /signup/ for /welcome/ or the return path, signed-in header pill, preferences kept with the account profile
- 18 September 2026: Secret scanning: gitleaks on every push and pull request, a local pre-commit hook, and an allowlist for public ids and data prose
- 18 September 2026: Sign up or log in button in the top bar, a medical disclaimer in the footer, and new Terms of Use and Privacy Policy pages
- 18 September 2026: Trastuzumab carries the June 2026 FDA approval of palbociclib with trastuzumab as HER2-positive maintenance, the one bot proposal from PR 35 not already in the corpus
- 18 September 2026: Pages are far lighter: the shared header, footer and cancer list are no longer copied into every page's data, and the sign-in id ships with the code
- 18 September 2026: Terms of use and privacy policy pages, medical disclaimer in the footer, copyright to OnCo
- 18 September 2026: Header: a Sign up or log in pill when signed out; region, language and search controls sit on the centre line
- 18 September 2026: Stop serialising the cancer list and the shared chrome into every page's payload
- 18 September 2026: Sign-in is live: first-time visitors choose whether they are a patient, caregiver, researcher or clinician and can set a cancer, kept only in their browser
- 18 September 2026: Each ship now builds once: Git-triggered Vercel builds are skipped and only the verified deploy runs
- 18 September 2026: First sign-in welcome step: who you are, optional cancer, browser-only profile per WorkOS user
- 18 September 2026: Vercel: skip Git-triggered builds so each ship builds once, from the verified CLI deploy
- 18 September 2026: add: Strand AI (W26) to the YC oncology list
- 18 September 2026: Owner list: proposal for six technology sections to lift the section kind past the size gauge
- 18 September 2026: Every reference DOI on the site checked against Europe PMC: four wrong links corrected, and a checker script added for future waves
- 18 September 2026: DOI checker for citation links: four wrong DOIs corrected and two rare-cancer key papers added
- 18 September 2026: Sixty-six landmark and classification papers for the rare-cancer pages, each checked against Europe PMC, plus two corrected reference links
- 18 September 2026: Two rare-cancer reference links corrected to the verified POD1UM-303 Lancet 2025 and MEK inhibition Nature 2019 papers
- 18 September 2026: Key papers for the 39 rare-cancer subtype pages: 66 verified landmark trials, classifications and guidelines from ANCHOR and ACT II to FIRM-ACT, PATHFINDER, HD21 and FIRSTMAPPP, linked through keyPapers
- 18 September 2026: Fifth Chinese quality pass over the new rare and pancreatic cancer pages: 14 corrections
- 18 September 2026: Four pancreatic Chinese TL;DR fixes synced from the wave export into the dictionary
- 18 September 2026: Chinese TL;DR pass 5: fix 14 rare and pancreatic subtype entries
- 18 September 2026: Eleven pancreatic cancer pages by stage and driver, from resectable and borderline disease to KRAS G12C, BRCA-mutant and MSI-high tumours, each linked from its trials and drugs
- 18 September 2026: Add 11 pancreatic cancer subtype pages: stage, driver and histology
- 18 September 2026: Thirty-nine subtype pages for rarer cancers, from anal and vulvar cancer by HPV status to Hodgkin lymphoma by stage and mastocytosis, plus an IndexNow script for changed pages
- 18 September 2026: Add 39 subtype pages for the rare cancer families that had none: anal, penile, vulvar, vaginal, thymic epithelial, adrenocortical, appendiceal, small bowel, cancer of unknown primary, histiocytoses, Langerhans cell histiocytosis, systemic mastocytosis, Hodgkin lymphoma, gestational trophoblastic neoplasia, pheochromocytoma and paraganglioma, nasopharyngeal and sinonasal cancers, with organ-drawing mappings, reverse links from 76 existing trial and drug records, Chinese TL;DRs registered for fold-zh and parent phrases for the cancer-families test
- 18 September 2026: IndexNow script: submit record pages changed since a date, batched, with a dry run
- 18 September 2026: Target pages with a mutation hotspot map no longer throw a hydration error: chart tooltips are written as single strings, and the page test now checks markup structure
- 18 September 2026: Fix hydration error 418 on target pages: multi-child SVG titles rendered empty by the server
- 18 September 2026: Fourth Chinese quality pass: 22 corrections across researcher, institute, trial and company summaries, including arsenic trioxide and enzalutamide names
- 18 September 2026: Chinese quality pass 4 over 243 folded TL;DRs
- 18 September 2026: Thirty-five key papers gain their DOI and PubMed ids, so every key paper but one now shows its citation count
- 18 September 2026: Add DOIs and PMIDs to 35 key papers that had neither
- 18 September 2026: Drug cards no longer nest a link inside a link, and a permanent test renders every page to forbid nested anchors site-wide
- 18 September 2026: Gene targets carry HGNC, Ensembl, UniProt and NCBI Gene identifiers, and key papers show how often they have been cited
- 18 September 2026: Guard every page against nested anchors; unnest the product card
- 18 September 2026: Roadmap rows 22 and 97 marked done
- 18 September 2026: TRACERx 2023 paper DOI corrected (ends -5, verified at doi.org)
- 18 September 2026: Fill HGNC, Ensembl, UniProt and NCBI Gene ids on single-gene targets
- 18 September 2026: Citation counts for key papers from Europe PMC (improvement 97)
- 18 September 2026: Dependency map hydration error fixed: vendor tiles no longer nest a link inside a link, with a test that forbids nested anchors
- 18 September 2026: Fix React 418 on the dependency map: no anchor inside an anchor
- 18 September 2026: Health pass: 369 paper links for 270 researchers, PubChem structures for 43 drugs, six glossary Wikipedia links, refreshed literature snapshots for all 2,093 tracked records, provenance for every record, and 28 logos
- 18 September 2026: Six more glossary terms gain Wikipedia links; 74 recorded as having no article
- 18 September 2026: Molecules for 43 more small-molecule and peptide drugs via PubChem name lookup
- 18 September 2026: Link 369 key papers to 270 people from existing corpus links, people-papers gauge 73.8% to 79.4%
- 18 September 2026: Trial acronym aliases completed for every registry trial that has one (115 more), so searches by study name find them all
- 18 September 2026: Add 115 registry trial acronyms from the CT.gov acronym field
- 18 September 2026: Eighty more drug companies gain Wikipedia and Wikidata links
- 18 September 2026: Add Wikidata ids and Wikipedia links for 80 more companies
- 18 September 2026: Sixty-three more institutes and 48 drug companies gain Wikipedia and Wikidata links for readers and machines alike
- 18 September 2026: Add Wikidata ids and Wikipedia links for 48 companies
- 18 September 2026: Add Wikipedia links and Wikidata ids for 67 more institutions from the Wikidata API
- 18 September 2026: Sixty-six hospitals and institutes gain Wikipedia and Wikidata links, which also feed the machine-readable sameAs data for agents
- 18 September 2026: Add Wikipedia links and Wikidata ids for 66 institutions from the Wikidata API
- 18 September 2026: Final researcher identifiers pass: 64 ORCID ids and 60 Wikipedia links, taking 373 people to ORCID and 181 to Wikipedia, plus STRASS trial results
- 18 September 2026: Sixty-four more researchers carry ORCID ids and 60 gain Wikipedia links in the fourth identifier pass
- 18 September 2026: Seventy-seven landmark trials now show their results as structured outcomes transcribed from the published abstracts, and 146 more researchers carry ORCID ids
- 18 September 2026: Add structured outcomes to STRASS from its Lancet Oncology 2020 abstract
- 18 September 2026: Add identifiers pass three: 146 ORCIDs and 7 Wikipedia links for people
- 18 September 2026: Add structured outcomes to 77 subtype-wave landmark trials from Europe PMC abstracts
- 18 September 2026: A further 121 researchers carry ORCID ids and 18 gain Wikipedia links, each checked against their ORCID employment record
- 18 September 2026: Add 121 ORCIDs and 18 Wikipedia links in a second identifier pass
- 18 September 2026: Subtype cancer pages now link their sibling subtypes (667 links) and 188 researchers carry ORCID ids with 22 Wikipedia links
- 18 September 2026: Add 188 ORCIDs and 22 Wikipedia links to people records
- 18 September 2026: Forty-one hospitals and institutes added and 78 researchers linked to where they work
- 18 September 2026: Link cancer subtypes to their siblings and name parents in tldrs
- 18 September 2026: Fill institutions for 78 trial investigators and add 41 institution records
- 18 September 2026: Five duplicate person pages merged into one each with redirects, and a duplicate finder for people added
- 18 September 2026: Merge five duplicate person records and add a duplicate-person finder
- 18 September 2026: Update weights and URLs for alphagenome model
- 18 September 2026: A hundred and eleven trial leads and first authors get person pages, linked from the landmark trials and key papers they led, plus 160 more trial acronyms
- 18 September 2026: A further 160 trial acronyms added as aliases to registry trial records
- 18 September 2026: Add 111 trial investigators and lead authors named by the subtype trials and key papers
- 18 September 2026: Trial acronyms (MARIPOSA-2, SPARTAN, CheckMate 9DX and 700 more) added as aliases to registry trial records so searches by the name patients hear find them
- 18 September 2026: Trial acronyms for registry records: scripts/trial-acronyms.ts adds 711 aka aliases
- 18 September 2026: Seventeen more landmark trials named on subtype pages (ATTRACTION-2, MAIA, RESONATE-2, NeoSphere and others) recorded with confirmed registry ids, and acronyms added to three registry records
- 18 September 2026: Third Chinese quality pass: 19 corrections to new drug, company and trial summaries, including wrong drug names and Korean text left in Chinese sentences
- 18 September 2026: Confirm registry ids for 24 subtype-page trials: 17 new records, 3 acronyms on registry records, 4 links filled
- 18 September 2026: Third Chinese quality pass: 19 of 236 TL;DRs from the biomarker, drug, maker and trial waves corrected, and the 73 trial TL;DRs folded into zh.ts
- 18 September 2026: Seventy-three trials named on subtype pages now have their own records with confirmed registry ids, plus logos for the new drug makers
- 18 September 2026: Add 73 trials named by the cancer subtype pages (trials-subtypes-wave)
- 17 September 2026: Seventy-four drug makers added and 210 drug-to-company links filled, so pipeline drugs now show who is developing them
- 17 September 2026: Link 193 company-less drugs to their makers and add 74 maker records
- 17 September 2026: Key papers for 147 subtype pages: 364 pivotal trial, classification and guideline papers, each checked against Europe PMC
- 17 September 2026: Add key papers to 147 cancer subtype pages
- 17 September 2026: Thirty-nine drugs named by the new subtype pages now have their own records, from hormone therapies for pituitary tumours to antibody-drug conjugates in trials
- 17 September 2026: Add drugs named by the cancer subtype pages (drugs-subtypes-wave)
- 17 September 2026: Fifty biomarker glossary terms named by the new subtype pages (molecular groups, risk scores, resistance mutations), each linked from the pages that use it
- 17 September 2026: Add 50 biomarker terms named by the cancer subtype pages
- 17 September 2026: Registry trials linked to 109 subtype pages by driver and stage (1,018 links), with a re-runnable matcher script
- 17 September 2026: Link 1,018 registry trials to cancer subtype pages by driver and disease-state tokens
- 17 September 2026: Plain-English sentences for the 170 new subtype pages and trials, so every record carries one again
- 17 September 2026: Add simple sentences for the subtype-wave cancer pages and trials (part-s)
- 17 September 2026: Seven duplicate trial records folded into their named trials with redirects; second Chinese quality pass corrects 64 TL;DRs from tonight's new pages, including one inverted meaning, and adopts NMPA drug names
- 17 September 2026: Review Chinese TL;DRs of the subtype waves: standard drug names, terms and alignment with the English
- 17 September 2026: Merge duplicate trial records onto their named keepers
- 17 September 2026: Head and neck cancers by HPV status and site: HPV-positive and HPV-negative disease, recurrent or metastatic squamous cell carcinoma, hypopharyngeal cancer, three salivary gland histologies and three oral cavity subsites, tied to KEYNOTE-048, RTOG 1016, De-ESCALaTE, ECOG 3311, PATHOS and NIVOPOSTOP
- 17 September 2026: Head and neck cancers by HPV status and site: HPV-positive oropharyngeal, HPV-negative squamous, recurrent or metastatic state page, hypopharyngeal, adenoid cystic, salivary duct, mucoepidermoid, oral tongue, gingivobuccal (India) and lip, with NIVOPOSTOP, ECOG-ACRIN E3311, PATHOS and CAPTAIN-1st
- 17 September 2026: Neuroendocrine neoplasms by site and grade: small intestinal, pancreatic and lung neuroendocrine tumours, grade 3 NET and extrapulmonary neuroendocrine carcinoma, tied to NETTER-1 and 2, CLARINET, PROMID, RADIANT and CABINET; NETTER-2 refiled under neuroendocrine cancer
- 17 September 2026: Add neuroendocrine subtypes by site and grade
- 17 September 2026: Childhood cancers by type and risk: acute lymphoblastic leukaemia by risk group, Philadelphia status, infancy and relapse, paediatric AML, neuroblastoma by risk group, medulloblastoma by molecular group, and a page for adolescents and young adults
- 17 September 2026: Add paediatric cancers by type and risk group
- 17 September 2026: Sarcomas and bone tumours by type: chondrosarcoma, angiosarcoma, undifferentiated pleomorphic sarcoma, myxofibrosarcoma, alveolar soft part sarcoma, PEComa, epithelioid haemangioendothelioma, nerve sheath tumours, retroperitoneal and extremity soft tissue sarcoma, with STRASS, EORTC 62012, AMPECT and ANGIOTAX
- 17 September 2026: Add sarcoma and bone tumour subtype pages with landmark trials
- 17 September 2026: Cancer pages by state for the common cancers: gynaecological cancers by molecular class and stage, skin cancers by state, lung cancer by driver and stage, breast cancer by state, 44 pages tied to their landmark trials and drugs; the My cancer chip leaves the header until accounts are on
- 17 September 2026: Breast cancer by state: high-risk early and post-CDK4/6 HR-positive, HER2-low, early HER2-positive and HER2-positive brain metastases, early and metastatic triple-negative, inflammatory breast cancer, Paget disease of the nipple and phyllodes tumour, each linked from the trials and drugs that treat them
- 17 September 2026: Lung cancer by driver and stage: 14 subtype pages and 10 landmark trials
- 17 September 2026: Add skin cancers by state: melanoma by mutation, stage and site, advanced cSCC, advanced BCC, DFSP
- 17 September 2026: Gynaecological cancers by molecular class and stage: endometrial cancer by POLE, MMR, p53 and NSMP class, advanced or recurrent endometrial cancer, uterine carcinosarcoma, cervical cancer by stage (early and fertility-sparing, locally advanced, recurrent or metastatic) and ovarian cancer by platinum sensitivity, each tied to its landmark trials and drugs
- 17 September 2026: Blood cancers by state: acute myeloid leukaemia by driver and fitness, acute promyelocytic leukaemia, myeloma from smouldering to relapsed, MDS by risk, CLL by line and Richter transformation, CML by phase and primary myelofibrosis, nineteen pages tied to 111 drugs and trials
- 17 September 2026: Blood cancers by driver, state and phase: nineteen subtype pages
- 17 September 2026: Upper gastrointestinal and liver cancers by state: gastric cancer by HER2, CLDN18.2, PD-L1 and MSI status and early gastric cancer, hepatocellular carcinoma by BCLC stage, intrahepatic and extrahepatic cholangiocarcinoma, and GIST by mutation, each tied to its landmark trials and drugs
- 17 September 2026: Colorectal cancer by state: rectal, MSI-high, BRAF V600E, HER2-amplified, KRAS G12C and early-onset pages with OPRA, PROSPECT, BEACON and other landmark trials; primary mediastinal B-cell lymphoma
- 17 September 2026: Add upper GI and liver subtypes: gastric by biomarker, early gastric, HCC by BCLC stage, cholangiocarcinoma by site, GIST by mutation
- 17 September 2026: Brain tumour subtypes: IDH-mutant astrocytoma, oligodendroglioma, meningioma, brain metastases, paediatric high-grade glioma, spinal cord tumours, vestibular schwannoma and CNS germ cell tumours, each linked from the drugs and trials that treat them
- 17 September 2026: Add colorectal subtype pages and primary mediastinal B-cell lymphoma
- 17 September 2026: Add brain and spinal tumour subtype pages
- 17 September 2026: Plain-English sentences for every remaining record: 1,167 early-phase trials, 138 people, 65 companies, 8 institutions and 3 ideas, so all 11,306 records now carry one
- 17 September 2026: Simple layer: part-r, fourth and final wave to full coverage
- 17 September 2026: Websites for 15 organisations and nine more logos, with archived addresses replaced by live sites; roadmap row 139 records the open dependency map hydration mismatch
- 17 September 2026: Add official websites to 15 sponsor records and fetch nine logos
- 17 September 2026: Roadmap row 139: dependency map hydration mismatch
- 17 September 2026: 969 more plain-English sentences for early-phase trials, key papers and collections, taking the simple layer to 9,930 records; one non-cancer registry record removed
- 17 September 2026: Simple layer wave 3: part-q with 969 sentences (trials, papers, collections)
- 17 September 2026: Orphan trials given content: ten drug matches, 17 sponsor aliases, 36 sponsor companies and 130 trials tied to their makers, two non-cancer records removed; backlinks round: 303 weakly linked records gain true relations from their own text; orphans down to 214
- 17 September 2026: After the trial removals: no dangling references, floor re-run, sponsor wave three Chinese folded
- 17 September 2026: Backlink fill: text-supported relations for 303 weakly connected records
- 17 September 2026: 1,024 more plain-English sentences: 279 companies, 104 institutions and 640 phase 3 trials, taking the simple layer to 8,961 records
- 17 September 2026: Orphan trials: link sponsor-only trial records to drugs, companies, institutions and cancers; 36 new sponsor companies (wave 3); remove two myasthenia gravis records
- 17 September 2026: Papers for 36 more researchers matched by institution; trial snapshots refreshed for all 1,024 products with 136 registry changes detected; paper snapshots refreshed for all 1,924 records
- 17 September 2026: Add simple part-p: 1024 plain sentences for companies, institutions, the theories pathway and phase 3 trials
- 17 September 2026: One papers block per person after the two paper waves met
- 17 September 2026: People papers: 36 more research-trained people via OpenAlex and Europe PMC
- 17 September 2026: docs(corrections): name the reporting clinician on the two glioblastoma rows
- 17 September 2026: Company content and sources for 33 makers, 26 sponsor corrections and papers for twelve researchers; 7,065 reverse links take orphaned trials from 2,467 to 279; Wikipedia links for six terms; structured outcomes with sources for 26 trials; 153 new logos; the schematic gauge counts aliased drawings
- 17 September 2026: Add structured outcomes to 26 reported trials flagged by the trial-outcomes gauge
- 17 September 2026: Schematic gauge counts aliased drawings as specific
- 17 September 2026: Add Wikipedia links to six terms lacking one
- 17 September 2026: Fold a doubled relation line after the sponsor merge
- 17 September 2026: Reverse links for orphaned trials: their drugs, companies (or sponsor via alias), technologies and cancers now list them back; orphan floor 2572 to 384
- 17 September 2026: Link orphan sponsor companies, repoint AAA trials to Novartis, add Europe PMC papers to twelve people
- 17 September 2026: 661 plain-English sentences: every cancer, 201 drugs, 179 technologies, 120 terms, 50 phase 3 trials and 63 targets; 451 reverse links give 126 orphaned drugs and 105 companies their first inbound link, with three maker corrections
- 17 September 2026: Reverse links for orphaned drugs and companies: their companies, trials, cancers, technologies and institutions now list them back; orphan floor 2827 to 2572
- 17 September 2026: Simple layer: part-o with 661 sentences, layer test
- 17 September 2026: Machine translation pipeline for record summaries with hash-checked cache and a report link; 157 Chinese summaries live (every cancer page and nine glossary terms) and two Spanish; model review panel populated for twenty cancers by Fable, each verdict tied to the record's own sources
- 17 September 2026: Every record has its own RDF Turtle file at /api/v1/rdf/<id>.ttl, linked from the page head, the hidden machine links, JSON-LD, the OpenAPI document and llms.txt
- 17 September 2026: Machine-translated summaries: cached per record and language under public/i18n/summaries, shown only while the English hash matches, marked as machine translated with a report link and an English toggle; prioritised batch script with a fetch call to the Messages API and a validator for NCT ids, doses, gene symbols and leftover English; hand-written Spanish and Chinese examples for pCR; coverage script counts them
- 17 September 2026: Per-record RDF: one Turtle file per record at /api/v1/rdf/<id>.ttl, linked from every record page
- 17 September 2026: For me from a real situation: setting, biomarkers, treatments had, country and trial interest drive what is standard, what your biomarkers change, trials that fit, warnings and questions, all from record fields with sources; dependency map renders identically on server and client
- 17 September 2026: Theme stays light even when React recovers from a hydration mismatch; the counts grid fills its rows again with twenty kinds
- 17 September 2026: Fix hydration mismatch on the dependency map
- 17 September 2026: For me from a real situation (roadmap item 101): setting, biomarkers, treatments had, country and trials drive record-backed sections
- 17 September 2026: Machines of oncology, second wave: 23 more machine families from boron neutron capture and FLASH beams to hyperthermia, fluorescence surgery, positioning, gating and dosimetry, with 32 vendors; roadmap rows marked done for the day's work
- 17 September 2026: Decisions you may face: a page per cancer laying out each treatment setting's options, evidence and trade-offs; expert centres as a sortable table of what can be measured, with a near-you marker; your remembered cancer follows you through the header, home, hubs, trials and search; a what-changed page per cancer; red cards for warnings on cancer pages
- 17 September 2026: Machines wave 2: BNCT accelerators, FLASH, electron, hyperthermia, PDT lasers, LITT, focused ultrasound, ECT, HIPEC, endoscopy, skin and cervical imaging, RT guidance and QA, CSTDs
- 17 September 2026: Prostate cancer by state: eight subtype pages from low risk to neuroendocrine, linked from 45 trials and drugs; Tregzi added from the FDA notice and the remaining update bot proposals checked against the registry; Hansoh merged into one record with a redirect
- 17 September 2026: Patient roadmap 102, 104, 106: remembered cancer, per-cancer what changed, red cards
- 17 September 2026: Decision pages per cancer and outcome-based centre tables (roadmap items 107 and 108)
- 17 September 2026: Prostate risk-group and disease-state pages, update bot leftovers, Hansoh merge
- 17 September 2026: Chinese quality pass: 82 TL;DRs corrected (standard NMPA drug names, basket trial term, methotrexate spelling, idioms), account strings in the formal register
- 17 September 2026: Radiation wave 5: LET, RBE and when protons beat IMRT
- 17 September 2026: Chinese quality pass on today's TL;DRs and interface strings: 篮式试验 for basket trial, 甲氨蝶呤 typo, 磁共振直线加速器, established drug names in 61 templated trial lines, natural wording in 30 machine-flavoured sentences, 您 throughout the account strings
- 17 September 2026: First sixty days after diagnosis: a week-by-week guide for every cancer built from its records, and a printable appointment sheet per cancer with questions, words you may hear and treatments you may be offered
- 17 September 2026: Model review panel: schema, batch script and page for AI model commentary on records, with two example panels; diagnostics wave two: 24 tests from RAD51 foci and CTC counts to pharmacogenomics, cardiac markers and low-resource screening, with five companies; institution networks: Parker Institute members and leaders, CRUK, NCI, UNICANCER and DKTK links, and ten weakly linked centres strengthened
- 17 September 2026: Add the first 60 days guide and per-cancer appointment sheets
- 17 September 2026: Mine institution networks (rows 110, 111): PICI, CRUK, NCI, UNICANCER, DKTK members and leaders; strengthen weak institution pages
- 17 September 2026: Diagnostics wave 2: 24 tests around sequencing, 5 companies (roadmap row 119)
- 17 September 2026: Animations pause off screen, when the tab is hidden and on slow phones, with a shared budget of six spinning molecules or drawings at once; the record page sidebar stays inside its column and never reaches the footer
- 17 September 2026: Model panel: AI models comment on records beside the human review path
- 17 September 2026: Search results page as an index: kind chips, did-you-mean, pages strip, Ask OnCo card, related strip; 29 trial design terms and a design picker page; startup requests ranked by urgency and commerciality with every input cited; 51 law and regulation terms by jurisdiction with a law page; translation hygiene: language tags, no-translate on names and doses, a browser translation offer
- 17 September 2026: Animation budget for slower phones: one hook (use-animation-budget) pauses molecules, wireframe schematics, the docking scene, the garden sway, the body map, the resistance map and the steppers when off screen or the tab is hidden, draws a still frame under reduced motion, caps thumbnails at six spinning per kind, animates on tap on Save-Data or four-core devices, defers structure fetches until on screen, lazy-loads the gallery 3D viewer; record page sidebar confined to its grid column and not sticky when taller than the viewport; roadmap rows 127 and 138
- 17 September 2026: Add laws around oncology: 51 regulatory terms and /law/ page (roadmap 137)
- 17 September 2026: Manufacturing in oncology: seven supply chains, 18 technologies and 14 makers; theories of cancer: a hub map and sixteen theory records with status; platform trials at depth: 21 master protocol trials and a hub term; technology dependency map with 119 edges and a viewer; Chinese TL;DRs folded in for every new record
- 17 September 2026: Browser translation and language attributes (roadmap row 128, part three): summaries, field values, data tables, bullets and table bodies carry lang="en" while TL;DRs keep their own language; drug, gene, company, trial and person names, aka lines, NCT ids, gene symbols, database ids, molecule labels, formulas and doses carry translate="no" with the notranslate class; TranslateOffer shows one dismissible line at the top of the main column that switches to a carried language or explains the browser's translator in nine languages; wireframe canvas labels repeated as hidden real text; scripts/translation-coverage.ts measures the share of a page in the chosen language per kind
- 17 September 2026: Rank startup requests by urgency and commerciality (row 112)
- 17 September 2026: Trial designs at depth: 29 new terms, 40 enriched, design picker page (row 136)
- 17 September 2026: Search results page as a proper index (roadmap row 123)
- 17 September 2026: Manufacturing wave: remove the Hitachi remnant
- 17 September 2026: Platform and adaptive trials at depth (roadmap row 124): master-protocol hub term, 21 platform, basket and umbrella trials, STAMPEDE enriched
- 17 September 2026: Theories of cancer: hub map and sixteen theory records in the mechanics section
- 17 September 2026: One Hitachi record (machines wave)
- 17 September 2026: Machines of oncology: 21 machine technologies, 17 vendors and 66 centres with their proton, carbon, MR-linac, Gamma Knife and other rare machines, with a machines page; Free in oncology page with 62 services; tumour sequencing tests compared across 13 laboratories; useful 404 page with nearest matches; section tabs, aligned header icons and unsqueezed tables; machine-readable links, Wikidata sameAs and short titles for search engines and agents; footer credit; navigation starts at the top
- 17 September 2026: Technology dependency DAG viewer (roadmap 134)
- 17 September 2026: Manufacturing wave: seven supply chains, 18 technologies, 14 makers
- 17 September 2026: Merge fixes: single Strata Oncology record, Chinese dictionary without duplicate keys
- 17 September 2026: Machines wave: imaging, radiotherapy, particle, ablation, surgical and laboratory hardware with vendors and centres
- 17 September 2026: Free in oncology and tumour sequencing test pages (roadmap 117, 118)
- 17 September 2026: SEO and agent surface on record pages: titles within 60 chars, Wikidata sameAs, hidden machine-readable landmark
- 17 September 2026: Compact section tabs, table column widths and header icon alignment
- 17 September 2026: Make the 404 page useful: nearest matches, ways in, local miss log
- 17 September 2026: Tables: one shared stand-in for rows with no picture, so no visual slot is ever an empty gap
- 17 September 2026: Header star count stays put on reload; counts grid fills its rows; region menu closes after the tap so it cannot fall through to the profile link; drawer close button without a border; pills never break inside a word; page titles capitalised; landing cards show a corner arrow; off-site links open in a new tab; intoDNA STRIDE and Strata Oncology with sequencing links for testing companies; roadmap rows 115 to 121
- 17 September 2026: Accounts through WorkOS: browser-only sign-in behind the same account functions, recorded in WorkOS, on when the client id is set; README retitled Total information dominance on cancer with an open-repository note and a screenshot gallery; site title to match; pivot modality headers use one classifier so every header links
- 17 September 2026: Search index is fresh on every visit (service worker network-first for the API, cache rule narrowed to snapshots and images); site pages ranked higher in search; models table shows each maker's logo or an initials tile; orphan detector script and ratchet test; hub page linked from digital twins
- 17 September 2026: Apply the update bot's proposals (PR 16): 55 EU approval rows from the EMA register and 28 FDA regulatory events; pancreatic cancer wave: four failed phase 3 programmes, nimotuzumab's NOTABLE approval, elraglusib, their sponsors and cross-links; PDAC and Cloudflare rows on the roadmap
- 17 September 2026: Merge PRs 25, 28 and 29 (CheckMate split, topic cards, WebMCP); mobile region menu stays on screen; one heading per cancer group; bevacizumab NICE note; glioblastoma MGMT and tovorafenib wording; mathematical models hub and site pages in search; long lists load in pages as you scroll
- 17 September 2026: Add progressive WebMCP tools for public oncology search and lookup
- 17 September 2026: fix: filter cancer topic cards by relevance
- 17 September 2026: fix(glioblastoma): split CheckMate 143/498/548, add extent of resection and CCTG CE.6
- 17 September 2026: Pivot headers all link to pages: modality, front, status and phase; new small molecule and cancer vaccine pages; day-long cache headers for snapshots; Cloudflare caching on the roadmap
- 17 September 2026: Changelog sync keeps the pending entry through the build
- 17 September 2026: Changelog kept current from the commit log; linked from the README
- 17 September 2026: Link 51 new technologies back from their drugs, targets and base drawings
- 17 September 2026: Point 27 trials, drugs, targets, pathways and terms at their 44 landmark papers
- 17 September 2026: Link 28 cancer subtype pages from the drugs and trials that treat them
- 17 September 2026: Refresh provenance for this week's records
- 17 September 2026: Cancer links for 20 unlinked trials; Chinese version complete for trials
- 17 September 2026: Watch button confirms each click
- 17 September 2026: Country pages: cancer site labels link to their OnCo pages in both tables
- 17 September 2026: Chinese TL;DRs for the last 87 curated trials and drugs
- 17 September 2026: Mathematical and biophysical models: 22 model technologies, a third type on the models page
- 17 September 2026: Types strip as pills, navigator visuals, account-style signup, linked grid headers
- 17 September 2026: chore: refresh pulse, abstract and citation snapshots (#20)
- 17 September 2026: chore: refresh institution research output from OpenAlex (#17)
- 17 September 2026: chore: draft change proposals from feed snapshots
- 16 September 2026: Add Anocca; biophysics sources as data-source candidates
- 16 September 2026: Prevalence rows for 20 more targets
- 16 September 2026: Prevalence rows for 36 targets
- 16 September 2026: Molecules for 26 more drugs and tracers
- 16 September 2026: Molecules for 29 more small-molecule drugs
- 16 September 2026: Pipelines and missing rows for 29 new cancer pages
- 16 September 2026: Keep the corpus out of the tool pages' JavaScript
- 16 September 2026: Keep the corpus out of the guideline pages' JavaScript
- 16 September 2026: Load translation tables on demand; Chinese TL;DRs for 2,678 registry records
- 16 September 2026: Chinese TL;DRs for the remaining 131 people
- 16 September 2026: Subtype pages: leiomyosarcoma, liposarcoma, synovial sarcoma, marginal zone lymphoma, cutaneous T-cell lymphoma
- 16 September 2026: Subtype pages: kidney, testicular, oesophageal, bladder
- 16 September 2026: Fix accessibility regressions in shared UI components
- 16 September 2026: Subtype pages: thyroid, head and neck, ovarian
- 16 September 2026: Chinese TL;DRs for every company, institution, term and idea
- 16 September 2026: Radiation wave 4 (21 landmark trials) and Chinese TL;DRs for every cancer, target, technology and curated drug
- 16 September 2026: Parent pages for breast, lung, leukaemia and biliary cancer; radiation drugs; audit link report; Chinese TL;DRs
- 16 September 2026: check-links: separate blocked from broken in the report (#15)
- 16 September 2026: Cancer subtype pages (pleural and peritoneal mesothelioma), radiation wave 2, roadmap 112
- 16 September 2026: Mobile search opens into the box; radiation wave 1
- 16 September 2026: Add SciRouter, a prevention-first longevity start-up
- 16 September 2026: Molecules for 31 small-molecule drugs; gumarontinib folded into glumetinib
- 16 September 2026: Essential thrombocythaemia page, aspirin record, vorolanib
- 16 September 2026: Polycythaemia vera: dedicated page, seven pivotal trials, eight terms, two ideas
- 16 September 2026: China deep dive: fifteen trial sponsors as company records
- 16 September 2026: Offline page only on real failures, China wave 1, favicon set
- 16 September 2026: Registry trials wave 6 and products wave 7
- 16 September 2026: Clickable breadcrumbs, pipeline products wave 6, checklist additions
- 16 September 2026: Header signup, trial wave 5 and Ask weighting
- 16 September 2026: trials: wave 4 from ClinicalTrials.gov, 309 recruiting or active industry phase 3 oncology trials each linked to a recorded drug and cancer (295 to their sponsor); wave 3 trials gain sponsor links; search lifts the kind named at the start of a question so a 'drug for' query ranks drugs above trials and people
- 16 September 2026: data: pipeline wave 2 from ClinicalTrials.gov, curated: 27 products missing from the corpus (ten China-approved kinase and checkpoint drugs, Japan's amrubicin and pirarubicin, tabelecleucel, HER2 and GDF-15 antibodies, PSMA and CAIX PET tracers, CD45 radioimmunotherapy and others) with 32 phase 3 trial records and China regional rows; launch checklist gains the agentic and AI accessibility to-do
- 16 September 2026: cancers: the four thin pages filled from sources (epithelioid sarcoma, extragonadal germ cell tumour, metastatic cancer, NUT carcinoma) with state of the art, history and development pipelines; six trials (TIGER, SABR-COMET, the tazemetostat run-in, three NUT carcinoma trials) and the BET inhibitor ZEN-3694; gaps page counts only developers as missing products
- 16 September 2026: site: Google Analytics tag with an honest disclosure on the about page; refresh workflows merge their own pull requests once validate and tests pass (the proposals workflow stays a human gate); deploys the FCCT directory and the nine merged weekly snapshots
- 16 September 2026: chore: draft change proposals from feed snapshots (#1)
- 16 September 2026: chore: refresh pulse, abstract and citation snapshots (#2)
- 16 September 2026: chore: refresh trial counts from ClinicalTrials.gov (#3)
- 16 September 2026: chore: refresh public/votes.json from Discussions reactions (#5)
- 16 September 2026: chore: weekly citation link check (#7)
- 16 September 2026: chore: refresh FDA approvals feed (OCE notifications, openFDA drugsfda) (#8)
- 16 September 2026: chore: refresh preprint tracker from Europe PMC (#9)
- 16 September 2026: chore: check regional approvals against the EMA register (#10)
- 16 September 2026: chore: refresh universe lists (NCI drugs and types, NCI centres, OECI, NHS alliances, NLM journals, OpenAlex, KEGG, ChEMBL, ClinicalTrials.gov, FDA OCE) (#11)
- 16 September 2026: data: the Sijbrandij Foundation and Future of Cancer Care Today as entities with their leads, Sid Sijbrandij linked, and the FCCT directory brought in: 39 suppliers, tools and services and 10 clinics and centres for tissue preservation, tumour profiling, functional drug testing, liquid biopsy, theranostics, vaccines, cell therapy, trial finding and navigation
- 16 September 2026: owner requests: aldoxorubicin and Gemini Therapeutics; anetumab ravtansine, its two mesothelioma trials, the UCL proton therapy trial and an idea on why ADCs have failed in mesothelioma; cancer pages replace the curated Pipeline list with an In development tab built from the whole corpus; Even One Ventures and fourteen portfolio companies with a weekly-refreshed universe list
- 16 September 2026: data: forty-six approved cancer drugs missing from the corpus, curated from every ChEMBL molecule with an oncology indication (historic hormonal and cytotoxic agents, photodynamic and diagnostic agents, growth factors, antiemetics, drugs approved in Japan, China, India or Australia); approval pill no longer wraps
- 16 September 2026: header: profile icon on every page opens an email capture dialog (magic-link sign-in when accounts are on, list endpoint otherwise); weekly-issue signup boxes removed from footer and newsletter page; strings in nine languages
Regulatory events, past 7 days
Dated filings, approvals, letters and label changes recorded on product pages.
- 2026-11-16 · Giredestrant · approval (US): Approved as Wilserdly with everolimus for ER-positive, HER2-negative, ESR1-mutated locally advanced or metastatic breast cancer source
- 2026-11-15 · Ivonescimab · crl (US): Complete response letter, announced by Summit on 15 November 2026 (the letter's own date is not stated): "The CRL indicates that the BLA cannot be approved in its present form, as the FDA determined that substantial evidence of effectiveness is not established in the absence of a statistically significant overall survival (OS) improvement." source
- 2026-11-14 · Ivonescimab · pdufa (US): PDUFA target date source
- 2026-11-13 · Durvalumab · approval (US): FDA approved durvalumab (Imfinzi, AstraZeneca) in combination with enfortumab vedotin-ejfv as neoadjuvant treatment, followed by single agent durvalumab as adjuvant treatment after radical cystectomy, for the treatment of adult patients with muscle invasive bladder cancer (MIBC) who are ineligible for or who have refused cisplatin based chemotherapy. source
Upcoming readouts and decisions, next 30 days
Decisions, advisory committees, expected readouts and congresses from the readout calendar. Expected dates are editorial estimates.
- 2026-12-08 · San Antonio Breast Cancer Symposium 2026 (8 to 11 December) (confirmed) · Triple-negative breast cancer (TNBC), HR-positive / HER2-negative breast cancer, HER2-positive breast cancer · source
- 2026-12-12 · ASH Annual Meeting 2026, New Orleans (12 to 15 December) (confirmed) · Acute myeloid leukaemia, Diffuse large B-cell lymphoma, Multiple myeloma, Chronic lymphocytic leukaemia, Acute lymphoblastic leukaemia · source
Corrections, past 7 days
Confirmed factual corrections to the corpus, with the fixing commit.
- 2026-11-17 · Claudin 18.2: The one-line description gave only the biology and nothing of where the target now stands, although the gastric records had been corrected to show an approved antibody, a cell therapy approved in China in June 2026 and an antibody-drug conjugate with a phase 3 result (Second accuracy pass; the EMA page for zolbetuximab, the registry and report for CLARITY-Gastric 01 and the approval record for satricabtagene autoleucel, each read in raw text for the gastric check) · The description now states those three facts
- 2026-11-17 · Pancreatic cancer: The one-line description said daraxonrasib nearly doubled survival in 2026 and did not say it had been approved (Second accuracy pass; the US approval record for daraxonrasib (NDA 220910, approved 26 August 2026) read through openFDA) · The line now says it was approved in the United States in August 2026
- 2026-11-17 · Triple-negative breast cancer: UK funding of platinum chemotherapy before surgery: The funding row credited the platinum recommendation to appraisal TA851, which covers pembrolizumab, and was then listed as a NICE appraisal decision under guideline number NG101 (Second accuracy pass; NICE guideline NG101 recommendation 1.8.1 and the TA851 page read in raw text) · The row now gives the funding body as NHS England and names NG101 as a guideline recommendation, not an appraisal
- 2026-11-17 · Odronextamab: The record gave the brand as Lynozyfic and said the FDA granted accelerated approval in follicular lymphoma in July 2025. Lynozyfic is linvoseltamab, a different Regeneron antibody for myeloma. Odronextamab has no US approval: 30 July 2025 was the date of a complete response letter (Second accuracy pass; DailyMed and openFDA searched for a US label and approved application, the EMA page for Ordspono, and Regeneron's releases of 26 February 2025, the second quarter of 2025 and the third quarter of 2026 read as raw text from archived copies) · Brand set to Ordspono, the US approval row removed, the July 2025 event changed to a complete response letter, and the summary now says the drug is licensed in the EU only, with a US resubmission for diffuse large B-cell lymphoma accepted and a target date of February 2027. The response rate is given as 31.5 percent in 127 patients
- 2026-11-17 · EPCORE DLBCL-2: The trial was shown as active with results said to have been presented at EHA 2026. Topline results came later, on 5 October 2026, and have not been presented in full (Second accuracy pass; the Genmab and AbbVie announcement of 5 October 2026 read as raw text from an archived copy, and the ClinicalTrials.gov record for NCT05578976) · Status set to positive with the topline result: progression-free survival hazard ratio 0.49, 95 percent interval 0.36 to 0.67, in people with an IPI of 2 to 5. The record says the data are topline, survival is not reported and the combination is not approved. The cancer page, the epcoritamab record and the history line were updated to match
- 2026-11-17 · TRANSFORM: The outcome paired a median event-free survival of 29.5 months, which comes from the three-year follow-up, with the hazard ratio of the primary analysis, and cited the primary paper for both. At the primary analysis the median had not been reached (Second accuracy pass; the full text of the primary analysis (DOI 10.1182/blood.2022018730) read through Europe PMC and the abstract of the three-year follow-up (DOI 10.1200/JCO-25-00399) read through PubMed) · The outcome now gives the three-year figures with their own hazard ratio, 0.375 with interval 0.259 to 0.542, names the primary result of 0.356, and adds three-year overall survival of 63 against 52 percent, which was not significantly different with 66 percent of the standard arm crossing over
- 2026-11-17 · POLARIX: Five-year overall survival rates of 82.3 and 79.5 percent were given, but neither figure is in the abstract or text of the paper cited. The trial and the drug were also called the first improvement on R-CHOP in twenty years, which neither paper says (Second accuracy pass; the abstract and full text for DOI 10.1200/JCO-25-00925 read through Europe PMC and the abstract for DOI 10.1056/NEJMoa2115304) · The two rates were removed and the hazard ratio of 0.85 with interval 0.63 to 1.15 kept and marked not significant. The one-line summaries of the trial and of polatuzumab vedotin now state the progression result and the absence of a survival difference, and the subgroup findings are described as exploratory
- 2026-11-17 · Tafasitamab: The median duration of response in L-MIND was given as about 44 months. In the final analysis it had not been reached, and 44.0 months is the median follow-up. The dosing line left out the day 4 dose in cycle 1, the first-line filing was still shown as planned for the first half of 2026, and the drug was said to have shown the first first-line gain over R-CHOP (Second accuracy pass; the abstract for DOI 10.3324/haematol.2023.283480 read through PubMed, the US Monjuvi label read through openFDA, the frontMIND abstract (DOI 10.1016/S0140-6736(26)00866-4) and Incyte's second quarter 2026 results read from an archived copy) · The duration of response and dosing were corrected, the frontMIND result now carries its interval and the higher rates of serious and fatal adverse events, and the record says submissions were accepted in the second quarter of 2026 and that the use is not on the US label. The same updates were made on the frontMIND trial and the Incyte note
- 2026-11-17 · EPCORE DLBCL-1: Overall survival was described as the sole primary endpoint in the United States, with dual endpoints elsewhere partially met. Neither company release nor the registry says this; the registry lists overall survival as the single primary outcome (Second accuracy pass; the AbbVie and Genmab releases of 16 January 2026 read as raw text and the posted results for NCT04628494 read through the ClinicalTrials.gov API) · The record now says the trial missed its primary endpoint of overall survival, with medians of 11.3 against 10.3 months and a hazard ratio of 0.96 from the posted results, and that the US label still carries the indication under accelerated approval. The same wording was corrected on the epcoritamab and EPCORE NHL-1 records
- 2026-11-17 · Diffuse large B-cell lymphoma: incidence and standard of care: The burden line gave a US incidence of 7 to 8 per 100,000, where SEER gives 5.6. Pola-R-CHP was offered for an IPI of 0 to 1 although the licence and POLARIX start at 2. Four bispecific antibodies were said to be licensed and to give about 40 percent complete remissions, the frontMIND regimen was still pending, a STARGLO complete response letter was cited with no source, and the one-line summary said CAR-T cures about 40 percent of relapsed patients (Second accuracy pass; SEER Cancer Stat Facts for diffuse large B-cell lymphoma, the US labels for Polivy, Columvi, Epkinly, Lunsumio and Monjuvi read through openFDA, the EMA pages, the FDA advisory committee minutes of 20 May 2025 and the ZUMA-7 survival abstract (DOI 10.1056/NEJMoa2301665)) · The burden line now gives the SEER rate for 2019 to 2023, the median age of 67 and five-year relative survival of 64.8 percent for 2016 to 2022, and labels the UK figure as not re-read. The standard of care rows state which uses are licensed, three bispecifics are counted with complete remissions of roughly 30 to 40 percent, the STARGLO lines give the 8 to 1 advisory vote, and the summary gives the four-year progression-free figure from ZUMA-7 in place of a cure rate
- 2026-11-17 · Double-hit / high-grade B-cell lymphoma: The entry said the disease is often treated with DA-EPOCH-R without saying this is unproven, and did not mention the one randomised trial in this group, which closed early for deaths (Second accuracy pass; the NCI PDQ summary on aggressive B-cell lymphoma read as raw text and the abstract for DOI 10.1016/S2352-3026(26)00191-2 read through PubMed) · The entry quotes the NCI wording that intensive therapies are being explored, and reports Alliance A051701: 73 patients, six deaths on treatment with added venetoclax against one, and 24-month overall survival of 52 against 72 percent. An unsourced sentence on CAR-T outcomes was replaced with what the labels say
- 2026-11-17 · SUNMO, GOLSEEK-1 and golcadomide: The SUNMO result line gave no figures although the paper appeared in 2025. GOLSEEK-1 and the golcadomide record still described the trial as recruiting with a readout in 2027 to 2028 (Second accuracy pass; the SUNMO abstract (DOI 10.1200/JCO-25-01957) read through PubMed and the ClinicalTrials.gov record for NCT06356129) · SUNMO now gives median progression-free survival of 11.5 against 3.8 months, hazard ratio 0.41, and response in 70 against 40 percent, and says the pairing is not on the US label. GOLSEEK-1 is shown as no longer recruiting with primary completion estimated for August 2027
- 2026-11-17 · ctDNA monitoring in lymphoma and the IPI: The entry said undetectable ctDNA at the end of treatment predicts cure better than PET in several cohorts, which the two papers cited do not show. The IPI entry described outcomes in the rituximab era with no source (Second accuracy pass; the abstracts for DOI 10.1200/JCO.2018.78.5246 and DOI 10.1038/s41587-021-00981-w and the 1993 IPI paper (PubMed 8141877) read through PubMed) · The ctDNA entry, the related idea and the cancer summary now say what the papers found: early ctDNA response predicted outcome independently of the IPI and interim PET. The IPI entry gives the original five-year survival figures of 73, 51, 43 and 26 percent and says they date from before rituximab
- 2026-11-17 · Lisocabtagene maraleucel: The drug record and the TRANSCEND CLL 004 trial record gave an overall response rate of 47 percent and undetectable residual disease in blood in 64 percent. In the primary analysis set of 49 patients the response rate was 43 percent and the residual disease figure 63 percent. The 47 percent belongs to the 30 patients with del(17p) or a TP53 mutation (Second accuracy pass; the full text for DOI 10.1016/S0140-6736(23)01052-8 read through Europe PMC, and the US label read through openFDA) · Both records now give 43 percent and 63 percent with the denominator of 49, and the drug record adds the label figures of 45 percent response and 20 percent complete response in 65 patients
- 2026-11-17 · Pirtobrutinib: The drug record said the first-line trials BRUIN CLL-313 and CLL-314 were reading out in 2026 to 2027, and the BRUIN CLL-321 trial record called them pending and first-line use an open question. Both trials were published in December 2025 and the FDA approved first-line use on 2 October 2026. The records also gave only the primary analysis of BRUIN CLL-321 (Second accuracy pass; the abstracts for DOIs 10.1200/jco-25-02380, 10.1200/jco-25-02477 and 10.1200/jco-25-00166 read through Europe PMC, the US label read through openFDA, and the FDA approval notices read as raw HTML) · The records now give the CLL-313 hazard ratio of 0.20 against bendamustine and rituximab, the CLL-314 response rates of 87.0 against 78.5 percent, and the final CLL-321 medians of 14.0 against 8.7 months with a hazard ratio of 0.54. The chronic lymphocytic leukaemia page and its first-line row mention the October 2026 approval
- 2026-11-17 · Sonrotoclax: Said the drug was approved in China in late 2025 and that the CELESTIAL-TNCLL trial completes in 2026. China gave conditional approval on 6 January 2026, and the registry puts primary completion of the trial in February 2032 (Second accuracy pass; the abstract for DOI 10.1007/s40265-026-02322-0 read through Europe PMC and the ClinicalTrials.gov record NCT06073821 read through the registry API) · The summary, the approvals row and the chronic lymphocytic leukaemia timeline give 6 January 2026 with the Chinese indication, and the paper is linked
- 2026-11-17 · CELESTIAL-TNCLL: Said results were expected from late 2026 and that primary completion was estimated for November 2026. The registry record, updated in October 2026, lists the trial as active and no longer recruiting, with primary completion estimated for February 2032 and no results (Second accuracy pass; the ClinicalTrials.gov record NCT06073821 read through the registry API) · The one-line summary, the summary and the chronic lymphocytic leukaemia page now give the registry status and date
- 2026-11-17 · ELEVATE-TN: The summary and the registry link used NCT02475097, a number the registry does not hold. The survival hazard ratio of 0.62 was credited to the 2020 Lancet paper, which does not report it (Second accuracy pass; the ClinicalTrials.gov API, and the full text for DOI 10.1182/blood.2024024476 read through Europe PMC) · The summary and link use NCT02475681, and the survival row cites the six-year Blood paper with its P value of 0.0349
- 2026-11-17 · GLOW: Gave an overall survival hazard ratio of 0.487 at four years and undetectable residual disease in marrow of 51.9 against 17.1 percent. Neither figure is in the abstracts that could be read. The 67-month report gives 66-month survival of 79.0 against 60.8 percent with a hazard ratio of 0.46, and the primary report gives best marrow rates of 55.7 against 21.0 percent (Second accuracy pass; the abstracts for DOIs 10.1056/EVIDoa2200006 and 10.1038/s41375-026-03150-7 read through Europe PMC, and for PubMed 37944541 through PubMed) · The result line and summary give the 67-month survival figures, and the residual disease row gives the best marrow rates
- 2026-11-17 · CAPTIVATE: Said the randomised cohort showed no difference in progression-free survival at three years and dated that paper 2024. The paper is from 2021 and reports one-year disease-free survival of 95 percent with placebo against 100 percent with ibrutinib. The four-year progression-free figure of about 79 percent could not be found in a published paper (Second accuracy pass; the abstracts for DOIs 10.1200/jco.21.00807 and 10.1182/blood.2021014488 read through Europe PMC) · The summary gives the one-year figures and the right year, the result line gives the published 24-month rate of 95 percent, and the four-year figure is labelled as not re-read
- 2026-11-17 · Chronic lymphocytic leukaemia: fixed-duration against continuous treatment: The open problems said the two approaches had never been compared and that the CLL17 trial would inform. CLL17 reported in December 2025: three-year progression-free survival was 81.1 percent with venetoclax and obinutuzumab, 79.4 percent with venetoclax and ibrutinib and 81.0 percent with continuous ibrutinib, meeting non-inferiority at an interim analysis (Second accuracy pass; the abstract for DOI 10.1056/nejmoa2515458 read through Europe PMC) · The open problem and the 2025 timeline entry now give the CLL17 result and say survival follow-up is still to come
- 2026-11-17 · BGB-16673: Said the phase 1 response rate of 86 percent came from 66 patients in an update at EHA 2026, and the CaDAnCe-304 trial record said enrolment runs from 2025 to 2027. The cited report describes 67 patients presented at ASH 2025, and the registry gives a September 2025 start with primary completion estimated for April 2028 (Second accuracy pass; the CLL Society page read as raw HTML and the ClinicalTrials.gov record NCT06973187 read through the registry API) · Both records corrected, the registry record linked, and the registry name tacabrutideg noted
- 2026-11-17 · Nemtabrutinib: Said the BELLWAVE-008 trial was recruiting. The registry lists it as active and no longer recruiting (Second accuracy pass; the ClinicalTrials.gov record NCT05624554 read through the registry API) · The summary now gives the registry status
- 2026-11-17 · Zanubrutinib: The one-line summary called it the most prescribed BTK inhibitor in chronic lymphocytic leukaemia, with no source (Second accuracy pass; the abstract for DOI 10.1056/NEJMoa2211582 read through Europe PMC) · The line now states only the head-to-head result on progression-free survival and heart rhythm problems
- 2026-11-17 · Obinutuzumab: Listed 2019 US approvals with venetoclax and with acalabrutinib as approvals of obinutuzumab. The obinutuzumab label lists chronic lymphocytic leukaemia only with chlorambucil, and the 2019 actions are on the venetoclax and acalabrutinib applications (Second accuracy pass; the US label read through openFDA and the approval history read through the drugsfda API) · The approvals row now says which label carries each combination
- 2026-11-17 · Acalabrutinib: Called acalabrutinib with venetoclax the first all-oral fixed-duration regimen without saying where, and gave the February 2026 indication as previously untreated disease without del(17p) or TP53 mutation. Ibrutinib with venetoclax came earlier outside the United States, and the FDA states the indication as adults with chronic lymphocytic leukaemia or small lymphocytic lymphoma (Second accuracy pass; the FDA approval page read as raw HTML and the abstract for DOI 10.1182/blood.2021014488 read through Europe PMC) · The one-line summary says approved in the United States, and the indication follows the FDA wording with the trial population named
- 2026-11-17 · RESONATE, CLL14, CLL13 and AMPLIFY: Four outcome rows cited a paper that does not hold the figure: the six-year medians in RESONATE and the six-year rates in CLL14 were credited to the first reports, the five-year CLL13 rates to the 2023 paper, and the three-year AMPLIFY rates to an FDA page that gives medians (Second accuracy pass; the abstracts and full texts for DOIs 10.1002/ajh.25638, 10.1182/blood.2024024631, 10.1182/blood.2025032160 and 10.1056/NEJMoa2409804 read through Europe PMC) · Each row now cites the paper that reports it. The figures were right and did not change
- 2026-11-17 · Gilteritinib: The summary said the MORPHO trial of gilteritinib after transplant reduced relapse in MRD-positive patients, and that the HOVON 156 trial against midostaurin reported non-superiority for event-free survival. MORPHO missed its primary endpoint (relapse-free survival HR 0.679, 95% CI 0.459 to 1.005, p=0.0518) and only a prespecified MRD-positive subgroup benefited. HOVON 156 is reported as showing longer event-free survival with gilteritinib (median 51.1 against 19.9 months, p=0.05) with similar overall survival. (Second accuracy pass; the MORPHO abstract (PMID 38471061) read through PubMed, and the HOVON 156 result read in the full text of Gangat and Ravandi, Blood Cancer Journal 2026 (PMC13661620), which cites the 2026 European Hematology Association abstract.) · Summary rewritten with both results, a note that the congress abstract itself was not read, and links to the MORPHO paper and the review. The same MORPHO wording was corrected in the FLT3 inhibitor pairing and in two standard-of-care rows of the AML page.
- 2026-11-17 · Quizartinib: The record said the benefit was sustained in the maintenance phase and that the drug is continued for up to 36 cycles after consolidation or transplant. The US label states that quizartinib is not indicated as maintenance after allogeneic transplant because no survival gain was shown there (OS HR 1.62, 95% CI 0.62 to 4.22). (Second accuracy pass; the current Vanflyta label (effective 16 April 2026) read as raw text through openFDA.) · Summary and mechanism step corrected to the label wording, with the hazard ratios after transplant and after consolidation. The QuANTUM-First trial summary now labels the relapse-free survival and MRD statements as not re-read.
- 2026-11-17 · Ziftomenib and KOMET-001: The records gave a complete remission rate of 23% in 112 patients and said a quarter of patients reached complete remission. The US label gives complete remission in 17% and complete remission with full or partial count recovery in 21.4% of 112 patients; the 22% rate, the 33% overall response rate and the 6.6 month median survival belong to the 92-patient phase 2 publication. (Second accuracy pass; the Komzifti label read through openFDA and the JCO 2025 abstract (PMID 40997296) read through PubMed.) · Trial tldr, summary, result line and outcome rows, and the drug summary, now give each figure with its own population and source.
- 2026-11-17 · Enasidenib: The tldr said enasidenib was never approved in Europe after it failed a confirmatory trial. The European application was withdrawn on 6 December 2019, before the phase 3 IDHENTIFY trial was published; the record also dated the withdrawal 1 December 2019 and carried an unsourced statement about market withdrawals in some territories. (Second accuracy pass; the EMA question and answer document EMA/47029/2020 read as text from the PDF, and the IDHENTIFY abstract (PMID 35714312) read through PubMed.) · Tldr and summary rewritten in the right order with the IDHENTIFY figures (median OS 6.5 against 6.2 months, HR 0.86, p=0.23), withdrawal date corrected, unsourced statement removed.
- 2026-11-17 · ALFA-0701: Median event-free survival of 17.3 against 9.5 months (HR 0.56) was credited to the 2012 Lancet report, which gives two-year event-free survival of 40.8% against 17.1% (HR 0.58) and a significant survival gain at that cut. The medians come from the 2019 final analysis and the US label, where overall survival was no longer significantly different. (Second accuracy pass; the Lancet 2012 abstract (PMID 22482940), the Haematologica 2019 full text (PMC6312010) and the Mylotarg label read through PubMed, PMC and openFDA.) · Summary now separates the first report from the final analysis, outcome sources point to the 2019 paper, the final OS hazard ratio (0.81, 95% CI 0.60 to 1.09) is given and both papers are linked.
- 2026-11-17 · VIALE-A: The long-term follow-up was dated 2022 with 24-month survival of 37.5% against 16.7%. The final analysis was published in 2024 and gives 37.5% against 16.9%, with a hazard ratio of 0.58 at 43.2 months of median follow-up. (Second accuracy pass; the abstract for DOI 10.1002/ajh.27246 read through PubMed.) · Figure, year and hazard ratio corrected and the paper linked; the subgroup hazard ratios are labelled as not re-read.
- 2026-11-17 · CPX-351 Study 301: The record said there was no second randomised trial and that UK NCRI AML19 reported comparable remission rates, and the drug record said CPX-351 was still being tested in younger patients. AML19 was a randomised comparison with FLAG-Ida in 189 younger adults and found no overall survival difference (13.3 against 11.4 months), with a response rate of 64% against 76%. (Second accuracy pass; the AML19 abstract (PMID 37171402) read through Europe PMC and the Study 301 full text read through PMC.) · Trial summary, replication line and drug summary rewritten with the AML19 result and a link; an unsupported phrase about benefit being concentrated in transplanted patients was replaced with the paper's exploratory landmark analysis.
- 2026-11-17 · RATIFY and midostaurin: The tldrs said midostaurin tripled median survival, or took it from about two years to more than six. The medians are as published, but the paper itself says the gap may look large because of the shape of the curves and that the hazard ratio of 0.78 reflects the benefit more accurately; four-year survival was 51.4% against 44.3%. The record also claimed benefit across subgroups, where no subgroup result was significant. (Second accuracy pass; the NEJM 2017 full text read through PMC (PMC5754190).) · Both tldrs now lead with the 22% reduction in the risk of death; the summaries carry the authors' caution and the subgroup wording was corrected.
- 2026-11-17 · Tagraxofusp: The summary gave a complete response rate of 57% in untreated patients. The current US label gives 7 of 13 patients, 54%. (Second accuracy pass; the Elzonris label (effective 9 December 2024) read through openFDA.) · Figure corrected with its denominator, and the 27 May 2026 US approval date of pivekimab sunirine added from openFDA.
- 2026-11-17 · Acute myeloid leukaemia: The burden text gave 120,000 cases worldwide a year and survival above 60% and below 10% for two genetic subgroups with no source; the SEER page cited gives none of these. The myelodysplasia-related subtype listed RUNX1 among the defining genes, which is not in the WHO 2022 list of eight. The history said gemtuzumab was withdrawn after toxicity in SWOG S0106, whose report describes no improvement in remission or survival. Two open problems were stale or unsupported: a randomised phase 2 trial of venetoclax duration has reported, and the one-third figure for MEN1 resistance is not in the sources cited. (Second accuracy pass; the SEER Stat Facts page read as raw HTML, the WHO fifth edition paper read through PMC (PMC9252913), the S0106 abstract (PMID 23591789), the Perner Nature 2023 abstract and the 2026 Blood Cancer Journal review read through Europe PMC and PMC.) · Unsourced figures removed and labelled, SEER figures given with their years (33.4% for 2016 to 2022, 22,720 cases and 11,500 deaths expected in 2026), WHO gene list and blast rule corrected, history and open problems reworded.
- 2026-11-17 · Azacitidine: The record listed a 2020 US approval of azacitidine with venetoclax for AML. That indication is on the venetoclax label; the azacitidine injection label lists only myelodysplastic syndromes and juvenile myelomonocytic leukaemia. (Second accuracy pass; the Vidaza, Venclexta and Onureg labels read through openFDA.) · Summary and approval row reworded; the QUAZAR hazard ratio (0.69, 95% CI 0.55 to 0.86) and the exact Onureg indication added.
- 2026-11-17 · del(17p) / TP53 aberration in CLL: The summary gave a five-year progression-free survival of 72% for SEQUOIA arm C, which could not be found in a source read today, and said acalabrutinib with venetoclax is not labelled for this group, although the US acalabrutinib indication is not restricted by it. (Second accuracy pass; the abstract of Tam et al., Blood Advances 2026 (PMID 40829104) read through Europe PMC and the Calquence label read through openFDA.) · Reworded to the figure read (estimated 60-month PFS 70.7% in 127 untreated patients with del(17p) or a TP53 mutation), paper linked, and the AMPLIFY sentence corrected.
- 2026-11-17 · CD22: The summary said inotuzumab was moving into frontline treatment for older adults in Alliance A041501. The registry record for A041501 is a trial in young adults. (Second accuracy pass; ClinicalTrials.gov records NCT03150693 and NCT03959085 read through the registry API.) · Reworded to young adults, with both trials noted as still recruiting.
- 2026-11-17 · ADMIRAL: The summary said the trial converted gilteritinib's accelerated approval to full approval. The FDA notice describes an approval on an interim analysis and does not call it accelerated. (Second accuracy pass; the FDA approval notice read as raw HTML and the approval history read through openFDA.) · Reworded to what the sources show: approval in November 2018 on an interim analysis, final survival analysis added to the label in May 2019.
- 2026-11-17 · AUGMENT-101: The summary said MEN1 resistance mutations emerged in about a third of relapsing patients. The JCO report does not mention MEN1 and the Nature 2023 abstract gives no fraction. (Second accuracy pass; the JCO full text read through PMC (PMC11687943) and the abstract for PMID 36922589 read through Europe PMC.) · Reworded to a described route of acquired resistance, with the grade 3 or higher toxicity rates from the paper.
- 2026-11-17 · Oral decitabine and cedazuridine, ASCERTAIN-V and Taiho Oncology: Four records called the May 2026 regimen the first all-oral regimen for AML. The FDA notice does not make that claim. (Second accuracy pass; the FDA approval notice read as raw HTML and the Inqovi label read through openFDA.) · Reworded to an approved all-oral venetoclax regimen and the single-arm design and confidence interval added.
- 2026-11-17 · AGILE and ivosidenib: The records stopped at the 2022 analysis. The long-term report gives median overall survival of 29.3 against 7.9 months (HR 0.42, 95% CI 0.27 to 0.65) at 28.6 months of median follow-up. (Second accuracy pass; the abstract for DOI 10.1182/bloodadvances.2025016399 read through PubMed.) · Long-term figures and a link added to the trial and drug records.
- 2026-11-17 · myeloMATCH and shorter venetoclax courses: myeloMATCH was said to run across about 200 sites; the registry lists about 360. The idea record on shorter venetoclax courses said only retrospective and small prospective data existed, although a randomised phase 2 trial is reported as favouring 28 days over 14 for survival. (Second accuracy pass; ClinicalTrials.gov NCT05564390 read through the registry API and the 2026 Blood Cancer Journal review read through PMC.) · Site count corrected; the idea record now states the randomised result with a note that the trial report itself was not read.
- 2026-11-17 · Acute lymphoblastic leukaemia: The page said children have a "90% cure", that five-year survival is above 90% in children and 40 to 50% in adults, that ALL was the first disease treated with CAR-T, that blinatumomab consolidation is "for all", that there are two CD19 CAR-T products for ALL, that blinatumomab had full approval for MRD-positive disease in 2018, that AALL1731 was approved in June 2024, that ponatinib is the preferred TKI, and that menin inhibitors are the first targeted therapy for infant leukaemia. (Second accuracy pass; the SEER stat facts page and the NCI PDQ summaries read as raw text, the FDA approval notice of 14 June 2024, the FDA list of converted accelerated approvals, and the current US labels for Blincyto, Tecartus, Iclusig and Revuforj read through openFDA.) · Survival now reads approximately 90% under 15 and more than 75% at 15 to 19 (NCI), 73.2% across all ages for 2016 to 2022 (SEER), with no adult-only figure claimed. Brexucabtagene autoleucel added as the third CAR-T. The MRD approval is given as accelerated on 29 March 2018 and converted on 20 June 2023. The June 2024 approval is credited to E1910 and Study 20120215. Ponatinib is described by its accelerated approval, and revumenib by its label, which starts at 1 year of age.
- 2026-11-17 · D-ALBA (GIMEMA LAL2116): Randomised confirmation was described as pending and the pairing as "phase 3 pending". Survival at a median follow-up of 18 months was written as 18-month survival, disease-free survival was flagged as the primary endpoint, and "only half went to transplant". (Second accuracy pass; the NEJM 2020 and JCO 2024 abstracts read through Europe PMC and PubMed, and the GIMEMA ALL2820 report (NEJM 2026, PMID 42874102) found by search and read through PubMed.) · GIMEMA ALL2820 is now reported: 236 adults randomised, 24-month event-free survival 86% with ponatinib plus blinatumomab against 67% with imatinib plus chemotherapy. The primary endpoint is molecular response (60% after two blinatumomab cycles), 24 of 63 were transplanted, and the final analysis is given as OS 80.7% and DFS 75.8% at a median follow-up of 53 months. The same changes were made on the dasatinib and ponatinib records, the TKI plus blinatumomab pairing and the transplant-free idea.
- 2026-11-17 · ELIANA: The five-year overall survival of 55% was cited to the JCO 2023 paper, which is the three-year update, and the page said most responders were still in remission without further therapy at five years. The tisagenlecleucel record gave "5-year EFS about 44% in responders". (Second accuracy pass; the abstract for DOI 10.1200/JCO.22.00642 read through Europe PMC, and the five-year analysis (DOI 10.1200/JCO-25-01471, JCO 2026) found by search and read through PubMed.) · Three-year figures (EFS 44%, OS 63%, all 79 infused) and five-year figures (relapse-free survival among 70 responders 47% to 51%, OS 55% to 62% depending on how later transplants are censored) are now each credited to the right paper.
- 2026-11-17 · COG AALL1731: The arm sizes were swapped (722 was given for blinatumomab), the trial was said to have led to the June 2024 FDA approval, and the population was called average-risk. (Second accuracy pass; the NEJM abstract read through Europe PMC, the author manuscript read through NCBI PMC, and the FDA approval notice read as raw text.) · 718 were assigned to blinatumomab and 722 to chemotherapy alone. The approval predates the report and rested on E1910 and Study 20120215. The paper's own statistic (restricted mean survival time difference 72 days, 95% CI 36 to 108) is added, and the hazard ratio of 0.39 is labelled as not found in the text that could be read.
- 2026-11-17 · FELIX: Overall remission of 77% and complete remission of 55% were given for all 127 infused patients, two outcomes were cited to a JAMA item that is not the trial report, and "about forty percent event-free at two years" had no source. (Second accuracy pass; the NEJM abstract and full text read through Europe PMC and NCBI PMC, PubMed record 39928344, and a 2026 Leukemia paper read through NCBI PMC.) · The remission rates are now given for the main cohort of 94 patients, the outcomes cite the NEJM paper, and the follow-up claim reads 40.4% in ongoing remission without a transplant or other therapy at a median follow-up of 21.5 months. The UK and EU conditional approvals of 2025 were added to the obecabtagene autoleucel record from the EMA and MHRA pages.
- 2026-11-17 · PhALLCON: The primary endpoint arms were given as 164 and 81 patients, and the summary claimed fewer treatment failures and a "first" randomised comparison. The ponatinib record called it the only BCR::ABL1 inhibitor that covers T315I. (Second accuracy pass; the JAMA abstract (DOI 10.1001/jama.2024.4783) read through PubMed, and the Iclusig and Scemblix US labels read through openFDA.) · The primary analysis is 53 of 154 against 13 of 78, reported at an interim analysis with event-free survival not yet mature. Asciminib is also indicated for T315I, so "only" is removed. The frontline indication is stated as still accelerated on the August 2026 label.
- 2026-11-17 · Inotuzumab ozogamicin: Overall survival in INO-VATE was called statistically significant on the 2-year landmark, the remission rates were given for 326 patients, and Alliance A041501 was described as a trial in older adults. (Second accuracy pass; the NEJM 2016 abstract and the Cancer 2019 final report abstract read through Europe PMC and PubMed, ClinicalTrials.gov NCT03150693, and the Besponsa US label.) · The primary survival analysis did not meet its boundary (HR 0.77, 97.5% CI 0.58 to 1.03); the final report gave 7.7 against 6.2 months and 2-year survival of 22.8% against 10.0%. The remission analysis covered the first 218 patients. A041501 enrols adults aged 18 to 39. The transplant caution now quotes the label: 18 of 79 patients developed veno-occlusive disease.
- 2026-11-17 · ECOG-ACRIN E1910: The summary said the benefit persisted across MRD sensitivity and that the trial changed frontline care worldwide, and a pairing said E1910 was stopped early. (Second accuracy pass; the NEJM abstract read through Europe PMC and the full text through NCBI PMC.) · The result is described as the third interim efficacy analysis at a median follow-up of 43 months, with the age subgroup finding the paper reports. The unsupported phrases are removed.
- 2026-11-17 · Asciminib: The summary line said asciminib was approved for all newly diagnosed chronic myeloid leukaemia in 2024. (Second accuracy pass; the Scemblix US label effective 22 September 2026 read through openFDA.) · The indication is newly diagnosed Ph-positive CML in chronic phase, under accelerated approval.
- 2026-11-17 · Oesophageal cancer: First-line immunotherapy was described without the PD-L1 limit now on US labels, and adenocarcinoma was given as CPS 5 or 1. Current US labels limit pembrolizumab, nivolumab, nivolumab with ipilimumab and tislelizumab in oesophageal cancer to tumours expressing PD-L1 of 1 or more (Second accuracy pass; the current Keytruda, Opdivo, Yervoy and Tevimbra labels read through openFDA) · Summary, standard-of-care rows and state of the art now state the PD-L1 limit, in both the main record and the deeper layer
- 2026-11-17 · Oesophageal cancer: The burden line said about 600,000 cases a year, an older estimate, and US five-year survival was given as 21.6 percent overall and 48.1 percent for localised disease with no diagnosis years (Second accuracy pass; the GLOBOCAN 2024 copy in the repository and the SEER stat facts page read as raw text) · Burden now reads about 494,000 cases and 442,000 deaths in 2024. Survival now reads 22.2 percent overall, 48.6 percent localised, 29.1 percent regional and 5.3 percent distant for diagnoses from 2016 to 2022
- 2026-11-17 · Oesophageal cancer: Izalontamab brengitecan was shown as in trials with approval pending, and the 2026 history entry pointed at the breast cancer trial. China approved it for previously treated squamous oesophageal cancer in July 2026 (Second accuracy pass; the company announcement of 17 July 2026 read as raw text and openFDA searched for a US label) · Rows now say approved in China in July 2026 with no US label, and the history entry points at PANKU-Esophagus01
- 2026-11-17 · Oesophageal cancer: Two trials of immunotherapy with definitive chemoradiation were both called pending. SKYSCRAPER-07 reported in 2026 and missed its primary endpoints (Second accuracy pass; the abstract for DOI 10.1016/j.annonc.2026.07.413 read through PubMed and both registry records read on ClinicalTrials.gov) · Summary, open problems and the organ preservation idea now give the SKYSCRAPER-07 result and say KEYNOTE-975 completed in January 2026 with no results posted
- 2026-11-17 · PANKU-Esophagus01 (BL-B01D1-305): The record gave no figures, named SystImmune and BMS as sponsor, and said global confirmatory trials were being planned, which no source read supports. The trial was published in JAMA on 12 November 2026 and is sponsored by Sichuan Biokin in China (Second accuracy pass; the abstract for DOI 10.1001/jama.2026.20250 read through PubMed, the registry record and the company release) · Added overall survival 9.79 versus 7.20 months, hazard ratio 0.63, and progression-free survival 4.17 versus 1.97 months, hazard ratio 0.50, with the adverse event rates, the sponsor and the China approval
- 2026-11-17 · SANO: The outcome row gave arm sizes of 156 and 118, the one-line summary said survival was the same and that a third of patients have a vanished tumour, and the result line said about a third avoided surgery. The paper reports 198 and 111 patients, two-year survival of 74 versus 71 percent meeting non-inferiority, and says longer follow-up is required (Second accuracy pass; the abstract for DOI 10.1016/S1470-2045(25)00027-0 read through Europe PMC) · Arm sizes and survival figures corrected, the one-line summary reworded to no worse at two years, and the figures the abstract does not give labelled as not re-read
- 2026-11-17 · KEYNOTE-590: The summary said the FDA approved the combination for all patients. That was true on 22 March 2021, but the current US label limits it to tumours with a PD-L1 combined positive score of 1 or more (Second accuracy pass; the FDA approval page and the current Keytruda label read through openFDA) · Summary now gives both the 2021 approval and the current limit
- 2026-11-17 · RATIONALE-306: The summary said the EU approved the regimen in 2024, with no source, and did not give the EU limit. The EU indication requires a PD-L1 tumour area positivity score of 5 percent or more (Second accuracy pass; the EMA product page for Tevimbra and the US label read as raw text) · Summary now gives the US and EU PD-L1 limits as the labels word them
- 2026-11-17 · ESCORT-1st: Overall survival was called the primary endpoint. Overall survival and progression-free survival were co-primary and both were met (Second accuracy pass; the abstracts for PubMed ids 34519801 and 38870932 read through PubMed) · Summary corrected and the final analysis figures added, 15.6 versus 12.6 months, hazard ratio 0.70
- 2026-11-17 · Camrelizumab: The liver cancer survival figure, 22.1 versus 15.2 months, was the interim result, and the record listed two US complete response letters, both blamed on manufacturing. The final analysis gave 23.8 versus 15.2 months, a third letter came in July 2026, and the reasons for the second were not made public (Second accuracy pass; the abstract for DOI 10.1016/S1470-2045(25)00543-1 read through PubMed, with the OncLive report of 13 July 2026 and the Elevar release of 4 November 2026) · Final figures given, the three letters described as the sources describe them, and the October 2026 resubmission with its April 2027 target date added
- 2026-11-17 · CROSS: The summary said 368 patients were randomised. 368 were enrolled and 366 analysed (Second accuracy pass; the abstract for PubMed id 22646630 read through PubMed) · Summary corrected, and the Neo-AEGIS and ESOPEC results added to the comparison line
- 2026-11-17 · KEYNOTE-564: Called the first adjuvant immunotherapy in any solid tumour to show an overall survival gain, and the disease-free survival hazard ratio of 0.72 was credited to the 2021 report. Adjuvant ipilimumab lengthened survival in stage III melanoma in 2016, and the 2021 report gives 0.68. (Second accuracy pass; the abstracts for DOI 10.1056/NEJMoa1611299, 10.1056/NEJMoa2106391 and 10.1056/NEJMoa2312695 read through Europe PMC.) · Claim narrowed to kidney cancer in the summary, the trial record and the renal cell carcinoma page; the 0.72 figure now cites the 2024 update; the preliminary RAMPART result is noted as not yet published in full.
- 2026-11-17 · CheckMate 9ER: An earlier data cut was labelled the final overall survival result (46.5 against 36.0 months, hazard ratio 0.77) and credited to the 2021 paper. (Second accuracy pass; the abstract of the final analysis, DOI 10.1016/j.annonc.2025.09.006, read through Europe PMC.) · Final result written in: 46.5 against 35.5 months, hazard ratio 0.79 (95% CI 0.65 to 0.96) at a median 5.6 years, with the right source. The cabozantinib note now gives 75% for grade 3 or higher events, as the 2021 paper states, instead of about 65%.
- 2026-11-17 · LITESPARK-012: Said to have been reported at ASCO GU 2026 with hazard ratios not disclosed. The result was announced on 21 April 2026 and the full paper has since given the figures. (Second accuracy pass; the abstract for DOI 10.1016/j.annonc.2026.10.1281 read through Europe PMC, and the CURE report and the Merck ASCO GU release read as raw text.) · Date corrected; hazard ratios added (progression 1.08, death 1.04 for the belzutifan triplet), with the number randomised and the paper linked.
- 2026-11-17 · LITESPARK-022: Enrolment given as 1,600, the setting as high risk only, no hazard ratio, and a side-effect comparison that could not be traced. (Second accuracy pass; the FDA approval notice of 12 June 2026 and the US Welireg label read as raw text.) · 1,841 randomised; intermediate-high or high risk; disease-free survival hazard ratio 0.72 (95% CI 0.59 to 0.87) at the first interim analysis; approval dated 12 June 2026; side effects restated from the label and the untraced comparison labelled as not re-read.
- 2026-11-17 · LITESPARK-005: Response rate given as 22.7% where the paper's abstract gives 21.9% at the same analysis as the progression result; no overall survival figure; no mention of belzutifan with lenvatinib. (Second accuracy pass; the PubMed abstract for PMID 39167807, the US Welireg label and the FDA notice of 24 September 2026 read as raw text.) · 21.9% written in; overall survival hazard ratios added (0.88 interim, 0.92 in the label, not significant); LITESPARK-011 and the September 2026 approval added here, in the second-line row and in the history of the renal cell carcinoma page.
- 2026-11-17 · COSMIC-313: The primary progression-free survival result was shown with all 855 patients as its denominator; it was assessed in the first 550 randomised. (Second accuracy pass; the abstract for DOI 10.1056/NEJMoa2212851 and the registry results for NCT03937219.) · Arm sizes corrected to 276 and 274 and the endpoint labelled; the final survival row now cites DOI 10.1016/j.annonc.2026.02.011.
- 2026-11-17 · CheckMate 214: The eight-year intention-to-treat survival figure cited the 2018 primary report, and the final analysis published in 2026 was missing. (Second accuracy pass; the abstracts for DOI 10.1016/j.annonc.2024.07.727 and 10.1016/j.annonc.2026.02.017 read through Europe PMC.) · Source corrected; final analysis at a median 9.3 years added (hazard ratio 0.71 overall, 0.69 in intermediate and poor risk, 0.80 and not significant in favourable risk).
- 2026-11-17 · KEYNOTE-426: Progression-free survival of 15.7 against 11.1 months cited the 2019 report, which gives 15.1 against 11.1. (Second accuracy pass; the PubMed abstract for PMID 30779529 and the full text of DOI 10.1038/s41591-025-03867-5 read through Europe PMC.) · Row relabelled as the five-year result, source corrected and confidence interval added.
- 2026-11-17 · CLEAR (KEYNOTE-581): Described as the longest progression-free survival ever reported in first-line kidney cancer, with no source, and as a survival benefit whose curves separated late. (Second accuracy pass; the abstracts for DOI 10.1056/NEJMoa2035716 and 10.1200/JCO.23.01569.) · Superlative replaced by the medians (23.9 against 9.2 months); survival restated as hazard ratio 0.66 at first report and 0.79 at the final analysis with medians of 53.7 and 54.3 months. Lenvatinib note changed to match.
- 2026-11-17 · Renal cell carcinoma: Burden given as about 430,000 cases a year with no edition; SEER survival for distant disease given without diagnosis years. (Second accuracy pass; the IARC Global Cancer Observatory 2024 world fact sheet and the SEER kidney and renal pelvis page read as raw text.) · GLOBOCAN 2024 figures written in (about 443,000 cases, 145,000 deaths); diagnosis years 2016 to 2022 added.
- 2026-11-17 · Sarcomatoid differentiation (RCC): Median survival on kinase inhibitors alone given as under a year; the trial cited reports 14.2 months on sunitinib. (Second accuracy pass; the abstract for DOI 10.1158/1078-0432.CCR-20-2063 read through Europe PMC.) · Reworded with the trial's figure and its population of 139 intermediate or poor-risk patients.
- 2026-11-17 · Exelixis: STELLAR-303 in colorectal cancer listed as an ongoing phase 3 with no result; it reported in 2025. (Second accuracy pass; the abstract for DOI 10.1016/S0140-6736(25)02025-2 and the registry records.) · Result added (10.9 against 9.4 months, hazard ratio 0.80) and the kidney and neuroendocrine phase 3 trials named.
- 2026-11-17 · Tivozanib: The European approval entry carried no indication. (Second accuracy pass; the EMA page for Fotivda read as raw text.) · Indication and the authorisation date of 24 August 2017 written in.
- 2026-11-17 · RUBY / ENGOT-EN6 / GOG-3031: The short summary said immunotherapy "extended life by more than a year on average". The figure is a difference in medians (44.6 against 28.2 months) at the second interim analysis with 51 percent of deaths recorded, and it was statistically significant only in the whole trial population. In mismatch repair proficient tumours alone the hazard ratio was 0.79 with an interval of 0.60 to 1.04, which is not significant. The record also gave no EU dates (Second accuracy pass; the abstract for DOI 10.1016/j.annonc.2024.05.546 read through Europe PMC, the US Jemperli label read through openFDA, the FDA announcements read as raw HTML, and the EMA procedural steps document for Jemperli) · The summary, result and short summary now give the medians, say the analysis was interim, and state the result by mismatch repair status. Exact US dates (31 July 2023 and 1 August 2024) and EU dates (7 December 2023 and 15 January 2025) are added. The same correction is made in the endometrial cancer summary and state of the art, on the main record and the deeper layer
- 2026-11-17 · NRG-GY018 / KEYNOTE-868: The record said whether mismatch repair proficient patients gain a survival benefit "is the open question", although overall survival was reported in 2025: it was immature and not significant in either cohort (hazard ratio 0.79, interval 0.53 to 1.17, and 0.55, interval 0.25 to 1.19). The mismatch repair proficient arms were given as 294 and 294 patients; the paper has 293 and 295 (Second accuracy pass; the abstract for DOI 10.1038/s41591-025-03566-1 read through Europe PMC and the full text of DOI 10.1056/NEJMoa2302312 read from PubMed Central) · The summary states the reported overall survival result and links the 2025 paper. The arm sizes are corrected and the approval date is given as 17 June 2024
- 2026-11-17 · DUO-E / GOG-3041 / ENGOT-EN10: The record gave the results by mismatch repair status without saying they were exploratory subgroups, left out the weaker result for durvalumab alone in mismatch repair proficient disease (hazard ratio 0.77) and the interim overall survival, which was not significant for durvalumab alone, and gave an EU approval year that could not be confirmed (Second accuracy pass; the abstract and full text for DOI 10.1200/JCO.23.02132 read through Europe PMC, the FDA announcement of 14 June 2024, the US Imfinzi label through openFDA and the EMA pages for Imfinzi and Lynparza) · The summary now gives those results and says what they are. The US approval is stated as mismatch repair deficient disease only, the EU authorisation is described as it stands on the EMA page, and the record says the EU decision date could not be re-read
- 2026-11-17 · KEYNOTE-775 / Study 309: The record said lenvatinib with pembrolizumab is "now largely used after first-line chemo-immunotherapy". The trial excluded women who had already had a PD-1 drug, so it gives no evidence for that use. On the lenvatinib record, monitoring was given as blood pressure weekly for two cycles and thyroid tests every 6 weeks, which is not what the label says (Second accuracy pass; the full text for DOI 10.1056/NEJMoa2108330 read from PubMed Central, the abstract for DOI 10.1200/JCO.22.02152 through Europe PMC and the US Lenvima label through openFDA) · The sentence is replaced by the exclusion and what it means, and the final overall survival analysis is added (hazard ratio 0.65 in all patients, 0.70 in mismatch repair proficient disease). The lenvatinib note now follows the label: blood pressure after 1 week, every 2 weeks for 2 months, then at least monthly, and thyroid function at least monthly
- 2026-11-17 · PORTEC-3: The record stopped at the 5-year analysis and credited the molecular analysis to "de Boer/León-Castillo". The 10-year results were published in 2025: overall survival 74.4 against 67.3 percent (hazard ratio 0.73, interval 0.54 to 0.97), with the clearest gain in p53-abnormal tumours (52.7 against 36.6 percent) and none apparent in POLE-mutated or mismatch repair deficient tumours. The first author of the molecular paper is León-Castillo (Second accuracy pass; the abstracts for DOI 10.1016/S1470-2045(25)00379-1 and DOI 10.1200/JCO.20.00549 read through Europe PMC) · The 10-year figures and both papers are added, the 2019 analysis is described as post hoc, and the attribution is corrected
- 2026-11-17 · Endometrial cancer molecular classes: The shares of the four classes were given with no cohort (POLE about 7, mismatch repair deficient 25 to 30, p53-abnormal about 15, no specific profile about 50 percent), and the adjuvant trials were all described as ongoing. In a population-based series of 452 women the shares were 9.3, 28.1, 12.2 and 50.4 percent; among the high-risk women of PORTEC-3 they were 12, 33, 23 and 32 percent. PORTEC-4a reported in 2026: treatment chosen by molecular profile was non-inferior to standard vaginal brachytherapy and spared 46 percent of that group any adjuvant treatment (Second accuracy pass; the abstracts for DOI 10.1093/annonc/mdy058, DOI 10.1200/JCO.20.00549, DOI 10.1016/S1470-2045(25)00612-6 and DOI 10.1136/ijgc-2022-004039 read through Europe PMC, and the full text for DOI 10.1038/nature12113) · The glossary entry and the subtype list name the cohorts. PORTEC-4a is described as reported in the glossary entry, the idea record, the endometrial cancer summary, the adjuvant row and the open problems, and RAINBO's results are given as expected from 2028 rather than 2027 to 2030
- 2026-11-17 · Endometrial cancer: adjuvant immunotherapy and maintenance length: The high-risk adjuvant row said pembrolizumab or dostarlimab is added for stage III to IV disease. In fully resected high-risk disease the KEYNOTE-B21 trial found no gain in disease-free survival from adding pembrolizumab (hazard ratio 1.02, interval 0.79 to 1.32). The first-line row said maintenance runs "up to 2-3 years" for all three antibodies; durvalumab continues until progression. A 2009 history entry said the LAP2 trial showed laparoscopic surgery equivalent to open surgery, when that report covered feasibility and complications (Second accuracy pass; the abstracts for DOI 10.1016/j.annonc.2024.08.2242 and DOI 10.1200/JCO.2009.22.3248 read through Europe PMC and the US labels through openFDA) · The adjuvant row now limits chemo-immunotherapy to disease meeting the RUBY or NRG-GY018 entry criteria and states the KEYNOTE-B21 result. The first-line row gives each drug's maintenance length. The history entry is reworded. A trial of a hormone tablet in fibroids and endometriosis (NCT05862272) is removed from the page's trial list
- 2026-11-17 · DESTINY-PanTumor02: The record called the April 2024 decision "the first tumour-agnostic HER2 approval" with no source for "first", and the endometrial cancer page quoted an 85 percent response rate without saying it comes from 13 women. The approval is an accelerated one for previously treated tumours and was still accelerated on the label dated May 2026 (Second accuracy pass; the full text for DOI 10.1200/JCO.23.02005 read through Europe PMC, the FDA announcement of 5 April 2024 as raw HTML and the US Enhertu label through openFDA) · "First" is removed, the cohort sizes (40 women, 13 with the strongest staining) are stated, and the endometrial page now reads 11 of 13 women and says the approval is accelerated and for previously treated disease
- 2026-11-17 · Selinexor: The record listed a US approval in diffuse large B-cell lymphoma, which was removed from the label in April 2026. It gave 60 mg weekly as the dose in the endometrial trials, but SIENDO used 80 mg, and it described the EU authorisation as conditional although it became a full authorisation on 18 July 2022. The XPORT-EC-042 record lacked the hazard ratio and trial size (Second accuracy pass; the US Xpovio label and approval history read through openFDA, the abstract for DOI 10.1200/JCO.22.02906 through Europe PMC, the EMA page for Nexpovio and the Karyopharm announcement of 30 July 2026 as raw HTML) · The lymphoma approval is recorded as removed, both trial doses are given, and the EU note has both dates. XPORT-EC-042 keeps its negative status and now carries the hazard ratio 0.76 (interval 0.51 to 1.12, one-sided p 0.0791), 236 patients in the analysed population and 257 enrolled, and says these are company topline figures
- 2026-11-17 · Progestins: The record credited NRG-GY019, an ovarian cancer trial, as an endometrial trial of an aromatase inhibitor with a CDK4/6 inhibitor, and said such treatment gives durable control "with minimal toxicity"; in the PALEO trial severe side effects affected 67 percent against 30 percent. It gave a complete response of 70 to 80 percent within 12 months and relapse of about 30 percent for fertility-sparing treatment; the pooled reviews give regression in 76.2 percent, relapse in 35.4 to 40.6 percent and a live birth rate of 28 percent. The megestrol dose was given as 160 mg daily where the label says 40 to 320 mg (Second accuracy pass; the registry record for NCT04095364, the abstracts for DOI 10.1016/j.ygyno.2024.12.003, DOI 10.1016/j.ajog.2012.08.011 and DOI 10.1016/j.ygyno.2012.01.003 through Europe PMC, and the megestrol label through openFDA) · The trial is removed, PALEO's result and side effects are stated, and the fertility-sparing figures are replaced with the two reviews' own numbers here and on the fertility-sparing treatment record, which now links both reviews and says the guideline's full text could not be re-read
- 2026-11-17 · FIRES and SENTOR: The record said sentinel node mapping finds spread as reliably as removing all nodes "with far less lymphoedema", and the endometrial page said it works "without missing metastases". Neither study measured lymphoedema, and FIRES missed about 3 percent of node-positive cases (Second accuracy pass; the abstracts for DOI 10.1016/S1470-2045(17)30068-2 and DOI 10.1001/jamasurg.2020.5060 read through Europe PMC) · The lymphoedema claim is removed and the 97 percent detection figure is stated in its place
- 2026-11-17 · Chemotherapy with PD-1 or PD-L1 blockade, first-line endometrial cancer: The record said the combination "roughly triples progression-free time" in mismatch repair deficient tumours and called it "the most reliably replicated regimen in gynaecological oncology"; neither statement is in the trial reports. The lenvatinib pairing gave about 13 percent as the response to a PD-1 drug alone, where the GARNET report gives 15.4 percent (Second accuracy pass; the trial abstracts read through Europe PMC and PubMed, including DOI 10.1016/S1470-2045(24)00334-6 for AtTEnd and DOI 10.1158/1078-0432.CCR-22-3915 for GARNET) · The summary now gives the range of the reported reductions in progression risk (about 60 to 70 percent), adds the AtTEnd trial, says only RUBY has shown a significant overall survival gain, and drops the superlative. The response rate is corrected
- 2026-11-17 · Melanoma: The page said half of people with advanced melanoma now live ten years, and that about half of those given nivolumab with ipilimumab are alive at ten years. In CheckMate 067 ten-year overall survival was 43% with the combination, 37% with nivolumab and 19% with ipilimumab; 52% is melanoma-specific survival. A care row called the combination the option with the highest long-term survival, although the trial was not designed to compare the two nivolumab arms. (Second accuracy pass; the full text of the final ten-year report (DOI 10.1056/NEJMoa2407417, PMC12080919) read through Europe PMC.) · The summary line, the state of the art, the history row and the first-line care row now give 43% overall and 52% melanoma-specific survival and say the trial was not designed to compare the nivolumab arms.
- 2026-11-17 · Melanoma: Several figures and labels were wrong or out of date. BRAF and NRAS frequencies were given as about half and a quarter with no cohort; the 2002 discovery paper was said to have found BRAF V600E in about 50% of melanomas when it reported BRAF mutations in 66%. The CheckMate 204 brain response of 56% was the 2018 interim figure. Two ESMO benefit scores were wrong. Lifileucel was called the first cell therapy for a solid tumour. The intismeran phase 3 was described without its published result. (Second accuracy pass; the TCGA full text (PMC4580370), the abstract of Davies 2002 (PMID 12068308), the final CheckMate 204 abstract (PMID 34774225), the ESMO 2025 cutaneous melanoma guideline full text (PMC7618628), the FDA approval summary for lifileucel (PMID 40699950), the Tudriqev and Amtagvi US labels on openFDA and the NEJM abstract for INTerpath-001 (PMID 43013837).) · TCGA figures given with the cohort (52% BRAF, 28% NRAS of 318 tumours); the history row corrected to 66%; the brain response updated to 53.5% at three years; ESMO scores corrected to 4 and 3 in the first line and to A for pembrolizumab and no evaluable benefit for nivolumab after surgery; lifileucel described as the first tumour-derived T-cell therapy approved by the FDA; lifileucel and RP1 marked as accelerated approvals; the intismeran result added (24-month recurrence-free survival 82.2% against 70.0%); United States incidence and death rate trends added with their years.
- 2026-11-17 · COMBI-AD: The record credited the final analysis to JCO 2024 with more than ten years of follow-up, said the survival analysis was hampered by crossover and gave no survival result. The final report is in the New England Journal of Medicine: median follow-up 8.33 and 6.87 years, and the overall survival difference was not statistically significant (hazard ratio 0.80, 95% CI 0.62 to 1.01, P = 0.06). The claim of the longest follow-up of any adjuvant melanoma trial had no source. (Second accuracy pass; the abstract for DOI 10.1056/NEJMoa2404139 read through PubMed, and the ESMO 2025 guideline full text (PMC7618628). The ASCO abstract page refused a script.) · Summary, short description and result rewritten with the hazard ratios and the non-significant survival result; the V600K subgroup and the ESMO advice against adjuvant targeted therapy for V600K disease added; the ten-year percentages labelled as not re-read; the adjuvant care row on the melanoma page corrected to match.
- 2026-11-17 · RELATIVITY-047: Progression-free survival of 10.1 against 4.6 months was cited to the three-year paper, which gives 10.2 months and a hazard ratio of 0.79; the figures are from the 2022 primary analysis. The four-year update was said to bring the confidence bound for survival below 1, which is the three-year report. Severe side effects were given as about 21% when the papers give 18.9% and then 22.0%. The survival result was quoted without saying its formal test was not statistically significant, and the five-year analysis of October 2026 was missing. (Second accuracy pass; abstracts for DOI 10.1056/NEJMoa2109970, 10.1016/j.ejca.2025.115547 and 10.1038/s41591-026-04756-1 read through PubMed, the full text of DOI 10.1200/JCO.24.01124 (PMC12054981), and the ESMO 2025 guideline.) · Each figure tied to its report; the non-significant formal survival test stated; four-year and five-year results added; the European licence limit to PD-L1 below 1% added.
- 2026-11-17 · SWOG S1801: The record gave an event-free survival hazard ratio of 0.58. The New England Journal of Medicine paper reports no hazard ratio for that endpoint, only a log-rank P value of 0.004. A three-year follow-up was described as sustaining the benefit with no readable source. (Second accuracy pass; the full text of DOI 10.1056/NEJMoa2211437 (PMC10410527) read through Europe PMC, and searches of PubMed and Europe PMC for a later report.) · The hazard ratio removed from the result and the outcome row and the P value given; the non-significant early survival comparison added; the three-year follow-up labelled as not re-read.
- 2026-11-17 · DREAMseq (ECOG-ACRIN EA6134): The record said the trial settled the order of immunotherapy and targeted therapy and that the Italian SECOMBIT trial reached the same conclusion. Only 73 of 265 patients reached the second treatment, the ESMO guideline says the sequence question was not completely answered, and SECOMBIT was a non-comparative trial at 37 sites in nine countries. (Second accuracy pass; the full text of DOI 10.1200/JCO.22.01763 (PMC9839305), the SECOMBIT abstracts (PMID 36049147 and 38167503) and the ESMO 2025 guideline full text.) · Short description and summary reworded to what the trial showed, with the two-year survival and second-line response figures; SECOMBIT described as non-comparative with its figures; the sequencing pairing and the melanoma page reworded to match.
- 2026-11-17 · COLUMBUS: The record said the seven-year update showed the longest median survival of the class, with less fever and photosensitivity than dabrafenib with trametinib. The update claims the longest follow-up, not the longest survival, and no head-to-head comparison exists. (Second accuracy pass; the abstract of the seven-year update (PMID 38723373) and the Mektovi US label on openFDA.) · Reworded to the trial's own results: median survival 33.6 against 16.9 months, seven-year survival 27.4% against 18.2%, fever in 18% on the label; tolerability comparisons described as cross-trial. The binimetinib record corrected to match.
- 2026-11-17 · C-144-01: Lifileucel was described as the first cell therapy for a solid tumour, the 153 patients were called the pivotal cohort, and the response rate was sourced to the company home page. (Second accuracy pass; the five-year analysis (DOI 10.1200/JCO-25-00765, full text PMC12622280), the FDA approval summary (PMID 40699950), the Amtagvi label on openFDA and registry record NCT05727904.) · Reworded to the first tumour-derived T-cell therapy approved by the FDA; the 153 patients described as two pooled cohorts with the label figure of 31.5% in 73 patients; median survival of 13.9 months added; the approval described as still accelerated; sources replaced with the papers.
- 2026-11-17 · IMCgp100-202: The phase was recorded as 2 although the trial is reported as phase 3, and the summary said the result showed that falls in circulating tumour DNA matter more than tumour shrinkage, which the sources read do not say. (Second accuracy pass; the PubMed abstracts for DOI 10.1056/NEJMoa2103485 and 10.1056/NEJMoa2304753 and the Kimmtrak US label on openFDA.) · Phase changed to 3 with a note that the registry lists phase 2; medians, the 9.1% response rate and the three-year result (21.6 against 16.9 months, hazard ratio 0.68) added.
- 2026-11-17 · PRAME: The figure of about 90% of cutaneous melanomas was cited to a paper about kidney cancer. (Second accuracy pass; the cited PMC id resolved through Europe PMC to an eLife paper on clear cell renal cancer; the PRAME immunohistochemistry series (PMID 30045064, PMC6631376) read as its abstract.) · Figures replaced with 83% of primary and 87% of metastatic melanomas in a 400-tumour series and the source corrected.
- 2026-11-17 · Cobimetinib: Cobimetinib was called the first drug approved for histiocytic neoplasms. Vemurafenib was approved for Erdheim-Chester disease, a histiocytic neoplasm, in 2017. The triplet with atezolizumab was described as a cobimetinib approval; it is carried on the atezolizumab label. (Second accuracy pass; the Cotellic, Zelboraf and Tecentriq US labels and approval histories read through openFDA.) · Short description and summary reworded; the 2020 approval row marked as carried on the atezolizumab label.
- 2026-11-17 · ctDNA-guided adjuvant therapy in stage II-III melanoma: The proposed test was described as the UK DETECTION design without saying that the DETECTION trial closed early. (Second accuracy pass; registry record NCT04901988 read through the ClinicalTrials.gov API: terminated after 8 patients because the design needed to change with the standard of care.) · The closure stated in the summary and the test field.
- 2026-11-17 · KEYNOTE-716: The record gave no figures and rested its long-term claim on a conference poster that cannot be read. NADINA carried an unsourced claim to be the first phase 3 of its kind and said NCCN and ESMO list it as preferred. (Second accuracy pass; the KEYNOTE-716 abstracts (PMID 35367007, 38452313 and 40198940), the NADINA abstract (PMID 38828984) and the ESMO 2025 guideline full text. NCCN requires a login and was not read.) · KEYNOTE-716 now gives the hazard ratios and 48-month rates at a median follow-up of 52.8 months, with the poster labelled as not re-read; NADINA lost the unsourced claim and now quotes the ESMO recommendation and its caveats.
- 2026-11-17 · ROSELLA / GOG-3073: The Lancet link opened a different trial, IMforte in small-cell lung cancer. The overall survival hazard ratio of 0.69 was the interim figure and had no interval. The record did not say that every woman in the trial had already had bevacizumab or that the US approval requires it (Second accuracy pass; the DOI behind the link resolved through Europe PMC, the abstracts for PMID 40473448 and PMID 41974149 read through PubMed, and the Lifyorli label read through openFDA) · The record links the two ROSELLA papers in the Lancet, gives the final overall survival result of 16.0 against 11.9 months with a hazard ratio of 0.65 (95% CI 0.51 to 0.83), and gives the approval date of 25 March 2026 with the prior bevacizumab requirement
- 2026-11-17 · Ovarian cancer: niraparib and PARP maintenance: The page offered niraparib as first-line maintenance for tumours without homologous recombination deficiency and PARP maintenance after relapse for HRD-positive tumours. The US niraparib label was limited to HRD-positive tumours in March 2026, and the current US olaparib, niraparib and rucaparib labels limit maintenance after relapse to BRCA-mutated disease. The history said later-line labels were narrowed, when the treatment indications were withdrawn on 26 March 2024 (Second accuracy pass; the current Zejula, Lynparza and Rubraca labels read through openFDA and the Federal Register notice 2024-06299 read as raw text) · The summary, the standard-of-care rows, the history note and the short record in the cancer list now state the current label limits and the withdrawal. The PRIMA record says the same and gives the final overall survival intervals
- 2026-11-17 · SOLO-1: The seven-year survival difference was described as a clear overall survival gain and as the basis of the 2018 approval. The paper calls it clinically meaningful but not statistically significant by the prespecified criteria (P = .0004 against a required P below .0001) and the analysis was descriptive. The approval rested on progression-free survival. A line saying roughly half of patients were progression-free at five years had no source (Second accuracy pass; the abstracts for DOI 10.1200/JCO.22.01549 and DOI 10.1016/S1470-2045(21)00531-3 read through PubMed) · The summary, result and replication lines state the significance result, and the five-year line now gives the published medians of 56.0 against 13.8 months. The PAOLA-1 record and the olaparib with bevacizumab pairing now give the whole-population overall survival result, a hazard ratio of 0.92 that was not significant, beside the exploratory HRD-positive result
- 2026-11-17 · Luveltamab tazevibulin: The drug was shown in phase 3 with the REFRaME-O1 trial ongoing and an interim analysis expected in 2026. The registry lists the trial as terminated because the sponsor deprioritised the drug, with an end date of August 2025 (Second accuracy pass; the ClinicalTrials.gov record for NCT05870748 read through the registry API) · The drug is marked withdrawn, the summary gives the registry's reason, and it is out of the ovarian pipeline list and the page summary. Upifitamab rilsodotin, whose two later trials the sponsor terminated, is also out of the pipeline list
- 2026-11-17 · RAINFOL-01 (Rina-S): The phase 3 RAINFOL-02 trial was described as enrolling with completion expected in 2027, and REJOICE-Ovarian01 as reading out in 2026 to 2027. The registry lists RAINFOL-02 as active and not recruiting with primary completion in August 2026 and no results posted, RAINFOL-01 as recruiting, and REJOICE-Ovarian01 with primary completion estimated for February 2028 (Second accuracy pass; the ClinicalTrials.gov records for NCT06619236, NCT05579366 and NCT06161025 read through the registry API) · The three trial records and the rinatabart sesutecan drug record give the registry status as read on 17 November 2026. The endometrial response rate now carries its patient numbers
- 2026-11-17 · HIPEC at interval cytoreductive surgery: A Korean trial was described as supportive. The trial of 184 women found no difference in progression-free or overall survival, and a benefit only in the subgroup having interval surgery (Second accuracy pass; the abstract for DOI 10.1001/jamasurg.2022.0143 read through PubMed) · The pairing and the OVHIPEC-1 record say the trial was negative overall with a positive subgroup, and the ten-year OVHIPEC-1 figures are given with a link
- 2026-11-17 · Bevacizumab: The note gave hypertension in 42 percent of women, 17 percent at grade 3 or higher, for the GOG-0218 arm that had bevacizumab throughout. The US label gives 32 percent and 10 percent for that arm (Second accuracy pass; the Avastin label read through openFDA and the abstract for DOI 10.1056/NEJMoa1104390 read through PubMed) · The note gives the label figures and names the 2.6 percent figure as gastrointestinal-wall disruption needing medical intervention
- 2026-11-17 · Pegylated liposomal doxorubicin: The summary called it the comparator arm in MIRASOL and KEYNOTE-B96. KEYNOTE-B96 gave weekly paclitaxel in both arms, and in MIRASOL it was one of three chemotherapies the investigator could choose (Second accuracy pass; the Elahere label read through openFDA and the abstract for PMID 41974150 read through PubMed) · The summary names it as one of the three comparator chemotherapies in MIRASOL only
- 2026-11-17 · Letrozole: The note said the NRG-GY019 trial supports endocrine maintenance in low-grade serous ovarian cancer. The trial has finished recruiting and has not reported (Second accuracy pass; the ClinicalTrials.gov record for NCT04095364 read through the registry API) · The note says the support comes from retrospective series and that the randomised trial has not reported
- 2026-11-17 · Rucaparib: The sources for the 2020 accelerated approval and its 2025 conversion pointed at the FDA page for retifanlimab in Merkel cell carcinoma. The 2018 maintenance approval was typed as a conversion and carried the wording of the 2016 treatment indication (Second accuracy pass; the FDA pages for the two rucaparib prostate approvals read as raw text, one through the Internet Archive, and the dates checked against Drugs@FDA) · The two events link the rucaparib pages and the 2018 event is an approval with the maintenance indication
- 2026-11-17 · Karyopharm Therapeutics: The summary said selinexor is approved in multiple myeloma and diffuse large B-cell lymphoma. The US label lists the lymphoma indication as removed in April 2026 (Second accuracy pass; the Xpovio label read through openFDA) · The summary says the lymphoma accelerated approval was removed from the US label in April 2026
- 2026-11-17 · Ovarian cancer: stage at diagnosis and survival: The page said three-quarters of women are diagnosed at stage III to IV, with no source, and gave survival of about 50 percent and about 32 percent without diagnosis years. The US registry page gives 54 percent diagnosed after distant spread and 18 percent after regional spread, with five-year relative survival of 52.0 percent overall and 31.5 percent for distant disease, for diagnoses from 2016 to 2022 (Second accuracy pass; the SEER Cancer Stat Facts page for ovarian cancer read as raw HTML) · The summary, the burden line and the open problems give the registry figures with their years. The high-grade serous entry now names the Cancer Genome Atlas series behind the BRCA and repair-defect frequencies
- 2026-11-17 · Risk-reducing and opportunistic salpingectomy: The summary said population data from British Columbia show falling incidence after the policy was adopted. The cited kind of evidence is a cohort of 25,889 people who had the operation, among whom no serous ovarian cancers arose where 5.27 were expected (Second accuracy pass; the abstract for DOI 10.1001/jamanetworkopen.2021.47343 read through PubMed) · The summary states the cohort result with its follow-up and links the paper
- 2026-11-17 · Tovorafenib: The FIREFLY-1 response rate of about 51% was credited to RANO-HGG criteria; the EU approval was described as regardless of BRAF alteration type; the first-line phase 3 was called LOGGIPY-2, a name that does not exist (Second accuracy pass; the FDA approval notice and the current US label read through openFDA, the EMA page for Ojemda, and the ClinicalTrials.gov record NCT05566795) · Response rate given as 51% by RAPNO-LGG criteria in 76 patients at approval and 53% on the 2025 label; EU conditional authorisation dated 20 April 2026 with its BRAF fusion, rearrangement or V600 requirement; trial renamed LOGGIC/FIREFLY-2 here and on the Day One page, with its registry status
- 2026-11-17 · Focused-ultrasound blood-brain barrier opening: Said phase 1/2 trials show 4 to 6 times higher brain concentrations of carboplatin, temozolomide and albumin-bound paclitaxel (Second accuracy pass; the abstracts for DOI 10.1016/S1470-2045(23)00112-2 and DOI 10.1038/s41598-018-36340-0 read through PubMed and Europe PMC) · Now gives the measured figures from the 17-patient phase 1 trial, 3.7-fold for albumin-bound paclitaxel and 5.9-fold for carboplatin, and says the temozolomide study gave no figure; the same correction made on the glioma page and the ultrasound and antibody-drug conjugate idea
- 2026-11-17 · Rindopepimut: The ACT IV survival figures of 20.1 and 20.0 months were given as the result for all 745 patients (Second accuracy pass; the abstract for DOI 10.1016/S1470-2045(17)30517-X read through PubMed) · The figures are now stated as the primary analysis in the 405 patients with minimal residual disease, with the hazard ratio 1.01 and its interval, on the drug and the ACT IV trial page; the loss of EGFRvIII at recurrence is labelled as not re-read
- 2026-11-17 · Temozolomide: Called the only chemotherapy proven to extend life in glioblastoma, and gave median survival of about 23 months for MGMT-methylated and about 13 months for unmethylated tumours with no source that states them (Second accuracy pass; the abstracts for DOI 10.1056/NEJMoa043331, DOI 10.1016/S1470-2045(09)70025-7 and DOI 10.1093/neuonc/5.2.79 read through PubMed) · The word only removed and the carmustine wafer trial named; MGMT figures replaced with those in the Hegi paper, 21.7 against 15.3 months in methylated tumours, on the drug and the MGMT term; five-year survival of 9.8% against 1.9% added
- 2026-11-17 · INDIGO: Called the first targeted-therapy win in low-grade brain tumours and said vorasidenib delayed radiotherapy and chemotherapy by years (Second accuracy pass; the abstracts for DOI 10.1056/NEJMoa2304194 and DOI 10.1016/S1470-2045(25)00472-3 read through PubMed, the US label through openFDA, and the FDA notice of 16 March 2023 for dabrafenib with trametinib) · Narrowed to IDH-mutant low-grade glioma, since a targeted pair was approved for childhood low-grade glioma in March 2023; by years replaced with the hazard ratio of 0.26 for time to next intervention; the 2025 follow-up report added
- 2026-11-17 · EF-14: The 2015 interim and 2017 final analyses were not told apart, and the page did not say what had become of later randomised trials of tumour treating fields (Second accuracy pass; the abstracts for DOI 10.1001/jama.2017.18718, DOI 10.1001/jama.2015.16669 and DOI 10.1093/noajnl/vdag106 read through PubMed, and the ClinicalTrials.gov record NCT04471844) · Interim and final results separated with their intervals; added that TRIDENT (981 patients) reached primary completion in February 2026 with no published result found on 17 November 2026, and that a 66-patient randomised phase 2 of early use missed its primary endpoint
- 2026-11-17 · DCVax-L: Said almost all placebo patients crossed over and that most neuro-oncologists regard the evidence as inconclusive, neither of which the paper states, and carried a September 2026 status (Second accuracy pass; the abstract for DOI 10.1001/jamaoncol.2022.5370 read through PubMed, the company releases of 21 December 2023 and 8 October 2026 read in Internet Archive copies, and the NICE page GID-TA10143) · Reworded to what the paper reports, 232 randomised to vaccine and 99 to placebo with the comparison made against external controls; UK application dated 20 December 2023; status re-dated to the 8 October 2026 release, which still calls the vaccine experimental; same change on the Northwest Biotherapeutics page
- 2026-11-17 · CAR-T for glioma: Early studies were summarised without their sample sizes, a 50% disease-control figure had no denominator, and one programme was dated 2024-26 with no later report (Second accuracy pass; the abstracts for DOI 10.1038/s41591-024-02875-1, DOI 10.1038/s41586-024-08171-9, DOI 10.1056/NEJMoa2314390, DOI 10.1126/scitranslmed.aaa0984 and DOI 10.1038/s41591-025-03745-0 read through PubMed) · Each study now carries its number of patients and its result, for example 29 of 58 evaluable patients with stable disease or better and a median survival of 7.7 months in the 65-patient trial, and the page says all are uncontrolled phase 1 studies
- 2026-11-17 · Laser interstitial thermal therapy: Said case series suggest survival comparable to resection, which the cited series does not compare (Second accuracy pass; the abstract for DOI 10.1093/neuros/nyy375 read through Europe PMC) · Replaced with the series result, median survival 11.5 months after treatment in 54 patients, and a statement that no randomised comparison with resection exists; citation year corrected to 2019
- 2026-11-17 · Glioma and glioblastoma: Said tovorafenib was the first targeted drug for BRAF-altered childhood glioma; credited the Nordic trial with a finding only NOA-08 reports; listed dabrafenib with trametinib only for relapsed disease; gave no evidence for re-irradiation or lomustine at recurrence (Second accuracy pass; the FDA notices of 16 March 2023 and 23 April 2024, and the abstracts for DOI 10.1016/S1470-2045(12)70164-X, DOI 10.1016/S1470-2045(12)70265-6, DOI 10.1056/NEJMoa1707358 and DOI 10.1200/JCO.22.00164 read through PubMed and Europe PMC) · Summary line reworded to BRAF fusions; the two elderly trials described separately; the childhood low-grade glioma row corrected; added that neither bevacizumab with lomustine nor re-irradiation with bevacizumab lengthened survival in randomised trials
- 2026-11-17 · Multiple myeloma: The page said belantamab mafodotin was withdrawn in 2022 and offered belantamab with bortezomib or with pomalidomide at first relapse. It called iberdomide a drug in trials, said the FDA accepted MRD as an approval endpoint in 2024, dated the IMROZ and CEPHEUS labels to 2025, and gave the anito-cel decision only as December 2026. (Second accuracy pass; the FDA list of withdrawn accelerated approvals, the FDA approval pages for belantamab (23 October 2025), iberdomide (13 August 2026), daratumumab (6 November 2025 and 27 January 2026) read as raw text, the US Blenrep label through openFDA, the EMA page for Blenrep, and the Arcellx annual report from an Internet Archive copy.) · The US withdrawal is dated February 2023. The first-relapse row now says the US label is belantamab with bortezomib and dexamethasone after two or more lines, with the pomalidomide combination authorised in the EU only, and adds iberdomide with daratumumab (accelerated approval August 2026). The 2024 MRD entry is an advisory vote, with the first approval on MRD in August 2026. History now carries the November 2025 smouldering myeloma approval and the January 2026 CEPHEUS approval. The anito-cel date is 23 December 2026.
- 2026-11-17 · Daratumumab: The US approval for high-risk smouldering myeloma was dated 2026 and described as 2025-26, and the record had no US approval for daratumumab with VRd in transplant-ineligible patients. (Second accuracy pass; the FDA approval pages of 6 November 2025 and 27 January 2026 read as raw text, the Darzalex Faspro label through openFDA, and the supplement dates on drugsfda.) · The smouldering myeloma approval is dated 6 November 2025 and the CEPHEUS approval 27 January 2026, in the summary, the approvals list and the regulatory events, each with its FDA page.
- 2026-11-17 · CEPHEUS: The summary said the regimen was approved in 2025 alongside IMROZ-era labelling. (Second accuracy pass; the FDA approval page of 27 January 2026 read as raw text.) · The summary now gives the US approval of 27 January 2026, for transplant-ineligible patients only, and the FDA's statement that effectiveness is not established in patients who refused transplant.
- 2026-11-17 · Iberdomide: The short definition still said iberdomide was in phase 3, although the record carried its approval. The summary said the maintenance trial had met its primary endpoint and that the relapsed-disease trial was ongoing. (Second accuracy pass; the FDA approval page and Zenbexus label, the EXCALIBER-RRMM abstract for DOI 10.1016/S1470-2045(26)00450-X read through PubMed, the sponsor's release of 8 October 2026 from an Internet Archive copy, and registry record NCT05827016.) · The definition and summary now give the accelerated approval of 13 August 2026 on MRD-negative complete response (41 percent against 21 percent), the company-reported progression-free survival result, and that EXCALIBER-Maintenance has not reported.
- 2026-11-17 · Mezigdomide: The summary said both phase 3 trials were ongoing. (Second accuracy pass; the SUCCESSOR-2 abstract for DOI 10.1016/S0140-6736(26)01088-3 read through PubMed and registry record NCT05519085.) · The summary now reports SUCCESSOR-2 (median progression-free survival 18.0 against 8.3 months, hazard ratio 0.48) with its excess of neutropenia and infection, and says SUCCESSOR-1 is still recruiting.
- 2026-11-17 · Anitocabtagene autoleucel: The FDA decision date was given as 27 December 2026, the pivotal study population as four or more prior lines, and the response rate as 97 percent. (Second accuracy pass; the Arcellx annual report for 2025 read as raw text from an Internet Archive copy of the SEC filing, and the eligibility text of NCT05396885. The news page previously cited refused the request.) · The date is 23 December 2026, the population is three or more prior lines (fourth line or later), and the figures are 112 of 117 responding (96 percent) with a complete response in 74 percent at the October 2025 data cut. The iMMagine-1 trial record carries the same corrections.
- 2026-11-17 · Linvoseltamab: The response rate of 70 percent and complete response rate of 45 percent were credited to all 117 patients and called the highest of the class. They are the US label's figures for 80 patients with four or more prior lines. (Second accuracy pass; the abstract for DOI 10.1200/JCO.24.01008 through Europe PMC, the Lynozyfic label from DailyMed, the FDA approval page and the EMA page.) · The 117 patients are now given 71 percent and 50 percent, the label population is named for 70 and 45 percent, the unsourced superlative is removed, and the EU conditional authorisation is dated 23 April 2025. The LINKER-MM1 trial record is corrected to match.
- 2026-11-17 · Talquetamab: Response rates of about 73 percent at both doses and 65 percent after prior T-cell redirection did not match the phase 1-2 report, and MonumenTAL-3, 5 and 6 were all described as phase 3 trials in progress. (Second accuracy pass; the abstracts for DOI 10.1016/S2352-3026(24)00385-5 and 10.1056/NEJMoa2604657 read through Europe PMC and PubMed, and registry records NCT05461209 and NCT06208150.) · Rates are now 74, 69 and 67 percent. MonumenTAL-3 is reported as positive, MonumenTAL-5 as withdrawn before enrolling, and MonumenTAL-6 as closed to enrolment. The MonumenTAL-1 trial record, which cited the phase 1 paper for phase 2 figures, now cites the phase 1-2 paper with 375 patients.
- 2026-11-17 · Elranatamab: Monitoring said 48 hours in hospital after each of the first two step-up doses. (Second accuracy pass; the Elrexfio label read through openFDA.) · Corrected to 48 hours after the first step-up dose and 24 hours after the second, with a note that the approval is still accelerated and neither confirmatory trial has reported.
- 2026-11-17 · DREAMM-7: The overall survival hazard ratio of 0.58 was cited to the 2024 NEJM paper, which reports only 18-month survival, and the summary said the drug was re-approved in the US and EU without saying the US label is narrower. (Second accuracy pass; the abstract for DOI 10.1016/S1470-2045(25)00330-4 read through PubMed, the FDA approval page, the Blenrep label and the EMA page.) · The survival result is cited to the Lancet Oncology 2025 second interim analysis with its p value. The summary gives the February 2023 US withdrawal, the EU authorisation of July 2025 and the US approval of 23 October 2025 with bortezomib and dexamethasone after two or more lines. DREAMM-8 no longer claims a survival trend, since its survival data were immature.
- 2026-11-17 · MajesTEC-3: The primary result was cited to a news site and the hazard ratio given as approximate with no interval. (Second accuracy pass; the abstract for DOI 10.1056/NEJMoa2514663 read through PubMed and the FDA approval page of 5 March 2026.) · The result is cited to the NEJM paper with hazard ratio 0.17 (95 percent interval 0.12 to 0.23) and arm sizes. The summary adds the overall survival hazard ratio of 0.46 and the conversion of teclistamab monotherapy to traditional approval.
- 2026-11-17 · CARTITUDE-1: A response rate of 97.9 percent was cited to the 2021 Lancet paper, which reports 97 percent. (Second accuracy pass; the abstracts for DOI 10.1016/S0140-6736(21)00933-8 and 10.1200/JCO.22.00842 read through Europe PMC.) · The figure is now cited to the two-year follow-up paper and labelled as such.
- 2026-11-17 · IMROZ: The p value was given as 0.0005 and the interval was not marked as a 98.5 percent interval. PERSEUS carried a p value of under 0.0001 where the paper states under 0.001. (Second accuracy pass; the abstracts for DOI 10.1056/NEJMoa2400712 and 10.1056/NEJMoa2312054 read through Europe PMC.) · Both p values are corrected to under 0.001 and the IMROZ interval is labelled.
- 2026-11-17 · iStopMM: The record said about 3,700 people with MGUS were randomised and that results were early. (Second accuracy pass; the abstract for DOI 10.1200/JCO-25-02771 read through Europe PMC.) · Now 75,422 screened and 3,541 randomised, with the 2026 result: far more smouldering myeloma found, earlier diagnosis, no difference in active malignancy, survival still to come.
- 2026-11-17 · High-risk cytogenetics (myeloma): The definition was credited to an IMS 2024 consensus, required a TP53 mutation with del(17p), omitted one category and added an ultra-high-risk tier the source abstract does not contain. (Second accuracy pass; the abstract for DOI 10.1200/JCO-24-01893 read through Europe PMC.) · Rewritten to the four categories of the IMS and IMWG consensus published in 2025, with the paper linked.
- 2026-11-17 · R-ISS / R2-ISS staging: Median survival of about four to five years for stage III was given with no cohort. (Second accuracy pass; the abstracts for DOI 10.1200/JCO.2015.61.2267 and 10.1200/JCO.21.02614 read through Europe PMC.) · Replaced with the published figures and their cohorts: five-year survival of 82, 62 and 40 percent in the 2015 report, and median survival by R2-ISS stage in the 2022 report, both from patients treated before quadruplet induction.
- 2026-11-17 · CD38 antibody added to PI-IMiD-dex (quadruplet induction): MAIA was counted as one of four trials supporting the quadruplet, although it tested a three-drug regimen without a proteasome inhibitor. The transplant record made a similar slip, listing PERSEUS among trials of transplant against no transplant. (Second accuracy pass; the abstracts for MAIA, its 2026 final survival analysis, PERSEUS, IFM 2009 and DETERMINATION read through Europe PMC.) · The pairing now names three quadruplet trials and cites MAIA separately for its survival benefit (hazard ratio 0.67). The transplant record credits the survival comparison to IFM 2009 and DETERMINATION only.
- 2026-11-17 · MRD negativity (myeloma): The advisory vote of April 2024 was described as applying to MRD negativity at 12 months, which the cited paper does not say. (Second accuracy pass; the abstract for DOI 10.1016/j.jcpo.2025.100588 through Europe PMC and the FDA iberdomide approval page.) · Reworded to the vote as reported, with the first approval granted on MRD (13 August 2026). The smouldering myeloma term now carries the FDA approval of daratumumab on 6 November 2025.
- 2026-11-17 · Odronextamab: The plain-language line and all eight translations said it was approved in the United States for follicular lymphoma. (Second accuracy pass; Regeneron results for the second quarter of 2025 and the third quarter of 2026 and its release of 26 February 2025 read as raw text.) · Now says it is approved in Europe and not in the United States, where the FDA issued complete response letters and a resubmission is under review.
- 2026-11-17 · Glofitamab: The record, the regulatory timeline and the approvals table stated an FDA complete response letter for the STARGLO combination on 2 July 2025 with no source. (Second accuracy pass; Roche results releases for July 2025, October 2025 and January 2026, the FDA advisory committee minutes of 20 May 2025 and the current US label read as raw text. No primary source for the letter was found.) · The statement is kept and labelled as not confirmed on 17 November 2026; the committee vote of 8 to 1 and the absence of a second-line indication on the label are stated. The same wording was corrected on the STARGLO plain-language line, its translations and the confirmatory trial glossary entry.
- 2026-11-17 · EPCORE DLBCL-2: The plain-language line said first results were presented in 2026, and the translations still described the trial as only testing. (Second accuracy pass; the company announcement of 5 October 2026 read as raw text.) · Now says topline results were announced in October 2026 and full results are awaited.
- 2026-11-17 · Tafasitamab: The plain-language line and translations called its 2026 trial the first first-line gain over R-CHOP. (Second accuracy pass; the POLARIX and frontMIND abstracts read through PubMed.) · The claim of being first is removed; the higher rate of serious side effects is stated in the translations.
- 2026-11-17 · Polatuzumab vedotin: The plain-language line and translations called it the first improvement on first-line lymphoma therapy in twenty years. (Second accuracy pass; the POLARIX abstract read through PubMed.) · Now says it lowered the risk of progression with no significant difference in overall survival.
- 2026-11-17 · Diffuse large B-cell lymphoma: The short record gave about 150,000 cases a year with no source. (Second accuracy pass; the SEER statistics page for this lymphoma read as raw text.) · Replaced with the SEER rate of 5.6 new cases per 100,000 a year for 2019 to 2023.
- 2026-11-17 · BRUIN CLL-314: The record said the trial was recruiting and gave no result. (Second accuracy pass; the abstract for DOI 10.1200/JCO-25-02477 read through PubMed and the ClinicalTrials.gov record.) · Marked positive with the published result: response rate 87.0% against 78.5% with ibrutinib in 662 patients, non-inferiority met.
- 2026-11-17 · BRUIN CLL-313: The record said the trial was active with 309 enrolled and gave no result or approval. (Second accuracy pass; the abstract for DOI 10.1200/JCO-25-02380 read through PubMed, the ClinicalTrials.gov record and the FDA approval notifications page.) · Marked positive: 282 patients, progression-free survival hazard ratio 0.199, and US approval for previously untreated disease on 2 October 2026. The approvals table for pirtobrutinib now carries the 2025 and 2026 actions.
- 2026-11-17 · CELESTIAL-TNCLL: The plain-language line and translations said results were expected from late 2026. (Second accuracy pass; the ClinicalTrials.gov record NCT06073821.) · Now says enrolment has closed and primary completion is estimated for February 2032.
- 2026-11-17 · Zanubrutinib: The plain-language line and translations said it beat ibrutinib on efficacy and safety and was the most prescribed BTK inhibitor, which no source read supports. The SEQUOIA record could not source its five-year figure for del(17p) disease. (Second accuracy pass; the corrected record and the abstract for PMID 40829104 read through PubMed.) · Lines aligned to the corrected record. The SEQUOIA record now adds the readable figure, 70.7% progression-free at 60 months in 127 patients with del(17p) or TP53 mutation.
- 2026-11-17 · Enasidenib: The plain-language line and translations said it was never approved in Europe after failing a confirmatory trial, and the approvals table dated the European withdrawal to 2021. (Second accuracy pass; the EMA question and answer document EMA/47029/2020 and the abstract for PMID 35714312.) · The European application was withdrawn on 6 December 2019, before the phase 3 trial reported; all copies corrected.
- 2026-11-17 · KOMET-001: The plain-language line and translations said a quarter of patients reached complete remission, and the NPM1 and KMT2A leukaemia page gave 23 percent. (Second accuracy pass; the US label read through openFDA and the abstract for PMID 40997296.) · Now about a fifth: 22% of 92 in the paper and 21.4% of 112 on the label, counting remission with full or partial blood count recovery.
- 2026-11-17 · Acute myeloid leukaemia in older or unfit patients: The two-year survival in the comparison arm of VIALE-A was given as 16.7 percent. (Second accuracy pass; the abstract for PMID 38343151 read through PubMed.) · Corrected to 16.9 percent, with the final analysis follow-up and hazard ratio.
- 2026-11-17 · RATIFY: Translations and plain-language lines for the trial and for midostaurin said median survival tripled or rose from about two to more than six years. (Second accuracy pass; the abstract for PMID 28644114 read through PubMed.) · Now give the 22% reduction in the risk of death that the paper reports.
- 2026-11-17 · ASCERTAIN-V: Translations, and the plain-language line for Taiho Oncology, called the May 2026 regimen the first all-oral regimen for this leukaemia. (Second accuracy pass; the corrected English records, whose FDA source does not say first.) · The word first is removed.
- 2026-11-17 · QuANTUM-First: The record linked Richard Stone as an investigator. (Second accuracy pass; the author list of the Lancet paper (PMID 37116523) read through PubMed.) · He is not an author; the link is removed.
- 2026-11-17 · ELIANA: The plain-language line and translations said many responders stayed in remission for years. (Second accuracy pass; the abstract for DOI 10.1200/JCO-25-01471 read through PubMed.) · Now says about half of responders were relapse-free at five years.
- 2026-11-17 · Acute lymphoblastic leukaemia: Translations and plain-language lines for the cancer, ponatinib, asciminib and four trials carried claims already corrected in English: first disease treated with CAR-T, the only inhibitor covering T315I, approval for all newly diagnosed chronic myeloid leukaemia, the largest infant trial. Revumenib, approved since November 2024, was listed as pipeline here and on the myeloid leukaemia record. (Second accuracy pass; the US labels for ponatinib, asciminib and revumenib read through openFDA and the corrected English records.) · All copies aligned; revumenib removed from both pipeline lists.
- 2026-11-17 · PANKU-Esophagus01: The outcome rows used press release figures and cited the release. (Second accuracy pass; the abstract for DOI 10.1001/jama.2026.20250 read through PubMed.) · Now the published figures: overall survival 9.79 against 7.20 months, hazard ratio 0.63, and progression-free survival 4.17 against 1.97 months.
- 2026-11-17 · LITESPARK-012: The outcome row said hazard ratios were not disclosed and gave 1,854 enrolled. (Second accuracy pass; the abstract for DOI 10.1016/j.annonc.2026.10.1281 read through PubMed and the ClinicalTrials.gov record.) · 1,688 randomised; medians and hazard ratios for both investigational arms added.
- 2026-11-17 · LITESPARK-011: The record, its plain-language line and the Chinese translation said a final survival analysis was still to come and no regulatory decision had been announced. (Second accuracy pass; the abstract for DOI 10.1016/j.annonc.2026.10.109 read through PubMed and the FDA notice of 24 September 2026.) · Final overall survival added (33.7 against 28.6 months, hazard ratio 0.85, not significant) with the US approval.
- 2026-11-17 · Belzutifan: The record had no approval with lenvatinib and dated the adjuvant approval only to a quarter. The clear cell kidney cancer page omitted the combination from later lines and listed belzutifan as pipeline. (Second accuracy pass; the FDA notices of 12 June and 24 September 2026 read as raw text.) · Both approvals added with dates; the kidney page row and pipeline list corrected.
- 2026-11-17 · Relugolix bone density study: The record and its Chinese translation described a phase 3 trial of relugolix in endometrial cancer. (Second accuracy pass; the ClinicalTrials.gov record NCT05862272, whose conditions are uterine fibroids and endometriosis.) · The cancer link is removed and the text says it is not a cancer trial.
- 2026-11-17 · KEYNOTE-B21: The record showed the trial as active with no result and sat in a pipeline list. (Second accuracy pass; the abstract for DOI 10.1016/j.annonc.2024.08.2242 read through PubMed.) · Marked negative: disease-free survival hazard ratio 1.02 in 1,095 patients, with the mismatch-repair-deficient subgroup result stated as a subgroup.
- 2026-11-17 · Chemo-immunotherapy first line in endometrial cancer: The plain-language line and translations said it roughly triples progression-free time in mismatch-repair-deficient tumours. (Second accuracy pass; the RUBY, NRG-GY018 and DUO-E abstracts read through PubMed.) · Now says it cuts the risk of progression by roughly 60 to 70 percent.
- 2026-11-17 · Sentinel node mapping studies: The plain-language line and translations said mapping was as reliable as removing all nodes, with less swelling. (Second accuracy pass; the FIRES abstract read through PubMed.) · Now gives the 97% sensitivity; the swelling claim, which the study did not measure, is removed.
- 2026-11-17 · POLE ultramutation: PORTEC-4a was described as testing de-escalation, with no result. (Second accuracy pass; the abstract for DOI 10.1016/S1470-2045(25)00612-6 read through PubMed.) · Result added: five-year vaginal recurrence 4.5 against 1.6 percent, non-inferior.
- 2026-11-17 · CheckMate 067: The record, plain-language line and translations said about half of patients on the combination were alive at 10 years. (Second accuracy pass; the full text of the ten-year paper (PMC12080919).) · Ten-year overall survival was 43%; 52% is melanoma-specific survival. Both are now stated.
- 2026-11-17 · INTerpath-001: The record said full data were expected at a late 2026 congress and hazard ratios were pending. The vaccine record said filings were expected. (Second accuracy pass; the abstracts for DOI 10.1056/NEJMoa2609762 and DOI 10.1200/JCO-26-00835 read through PubMed.) · Published results added to the trial, its outcome rows and the intismeran record: 24-month recurrence-free survival 82.2% against 70.0%.
- 2026-11-17 · CheckMate 204: The record gave only the interim figures. (Second accuracy pass; the abstracts for PMID 30134131 and PMID 34774225 read through PubMed.) · Final three-year figures added for both cohorts.
- 2026-11-17 · IMCgp100-202: The phase had been changed to 3 although the registry lists phase 2. (Second accuracy pass; the ClinicalTrials.gov record NCT03070392 and the journal abstract.) · Phase set back to the registry value; the summary says the journal calls it phase 3.
- 2026-11-17 · Lifileucel: The record said EU approval 2025. (Second accuracy pass; the EMA page for Amtagvi and the FDA approval summary (PMID 40699950).) · The EU application was withdrawn on 22 July 2025; the US approval is dated 16 February 2024.
- 2026-11-17 · Luveltamab tazevibulin: The approvals table, the folate receptor alpha target, an open question and a bispecific idea described it as in phase 3 or in development, two of them with a topoisomerase payload. Its trial record had no status. (Second accuracy pass; the ClinicalTrials.gov records NCT05870748 and NCT06619236.) · The phase 2/3 trial was terminated in August 2025 and the payload is a hemiasterlin; all five places corrected and the drug removed from three pipeline lists.
- 2026-11-17 · Upifitamab rilsodotin: The record did not mention that its later studies were stopped. (Second accuracy pass; the ClinicalTrials.gov records NCT05329545 and NCT04907968.) · Both terminated studies added to the summary.
- 2026-11-17 · NMS-03305293 with temozolomide: The record, plain-language line and Chinese translation said the trial was active. (Second accuracy pass; the ClinicalTrials.gov record NCT04910022, updated 17 November 2026.) · Marked as terminated by the sponsor for reasons not related to safety, with 80 enrolled.
- 2026-11-17 · Temozolomide: The plain-language line and translations called it the only chemotherapy proven to extend life in glioblastoma. (Second accuracy pass; the corrected record, which cites a placebo-controlled trial of carmustine wafers.) · The word only is removed. The INDIGO lines were narrowed to IDH-mutant low-grade glioma in the same way.
- 2026-11-17 · Belantamab mafodotin: The record, glossary, plain-language line and translations said it was withdrawn in 2022; the record gave the US indication as after one prior line and credited DREAMM-8 with a survival benefit. (Second accuracy pass; the FDA list of withdrawn accelerated approvals, the FDA notice of 23 October 2025, the US label, the EMA page and the abstracts for PMID 38828951 and PMID 40680754.) · Withdrawn 6 February 2023; US approval is with bortezomib and dexamethasone after two prior lines; the survival benefit is from DREAMM-7.
- 2026-11-17 · Teclistamab: The 2026 US approval row omitted daratumumab and the record did not say the earlier accelerated approval had been converted. (Second accuracy pass; the FDA notice of 5 March 2026 read as raw text.) · Row, summary and approvals table corrected.
- 2026-11-17 · Linvoseltamab: The plain-language line and translations claimed the highest complete response rate of its class. (Second accuracy pass; the abstract for DOI 10.1200/JCO.24.01008 read through PubMed.) · Now says half of patients in the pivotal study reached a complete response.
- 2026-11-17 · Iberdomide: The plain-language line and translations said it was in phase 3 trials, and two pipeline lists still held it. (Second accuracy pass; the FDA notice of 13 August 2026 read as raw text.) · The accelerated approval is stated and the drug removed from both lists.
- 2026-11-17 · Carfilzomib: The approval with dexamethasone alone was dated 2015, and for daratumumab the approval with bortezomib, melphalan and prednisone was dated 2019. (Second accuracy pass; the four FDA approval letters read in Internet Archive copies.) · Dated 21 January 2016 and 7 May 2018; each row split.
- 2026-11-17 · Skin cancer (all types): The summary said melanoma accounts for most skin cancer deaths. Lifileucel was listed as a pipeline drug although it has been approved in the US since February 2024. (Second accuracy pass; the world, UK and US rows of the GLOBOCAN 2024 data the site holds, and the FDA page for Amtagvi read as raw text.) · Summary now says melanoma is about half of the world's recorded skin cancer deaths (58,398 against 58,721 for non-melanoma skin cancer) and most of them in the UK and the US. Lifileucel moved from the pipeline list to the drugs list.
- 2026-11-17 · Why nobody knows how many skin cancers there are: Cancer Research UK's 156,000 cases and 1,200 deaths were given without their years, the all-cancers comparison was given as 403,601 when the page says more than 403,000, and the charity's figure was said to use the lifetime-first method, which its page does not state. A Northern Ireland figure had no readable source. (Second accuracy pass; the three Cancer Research UK statistics pages read by curl; the Northern Ireland Cancer Registry answered 403.) · Years added (2016 to 2018 for cases, 2022 to 2024 for deaths, 2019 and 2021 to 2022 for all cancers), the comparison reworded to the page's own figure, the method claim removed and the Northern Ireland figure labelled as not re-read.
- 2026-11-17 · Keratinocyte cancer: The short summary called basal cell and squamous cell carcinoma the two commonest cancers in the world. The papers cited say the commonest in White populations. (Second accuracy pass; the abstracts for DOI 10.1093/bjd/ljag315 and DOI 10.1111/bjd.17873 read through Europe PMC.) · Reworded to the commonest cancers in White populations.
- 2026-11-17 · The growth pattern on a basal cell carcinoma report: The term said most basal cell carcinomas contain more than one pattern and that there is no evidence on how much of a high-risk pattern matters. The UK dataset says many contain both low-risk and high-risk patterns and that there is no robust evidence. (Second accuracy pass; the Royal College of Pathologists dataset G123 read from its PDF.) · Reworded to many and to no robust evidence.
- 2026-11-17 · How skin carcinoma is staged in Britain: Lymph node categories for the trunk and limbs were said to run on node size alone. They also depend on whether one or several nodes are involved. The ninth edition was called now in force without a source for the date. (Second accuracy pass; the Royal College of Pathologists UICC TNM 9 appendix of 6 November 2025 read from its PDF.) · Reworded to size and number of nodes, and to the appendix date.
- 2026-11-17 · The Brigham and Women's Hospital staging system: A note said the cohorts were referred because their tumours were difficult and that rates in an unselected population are lower than in any of them. Neither statement is in the papers' abstracts. Two percentages were merged into one approximate figure. (Second accuracy pass; the abstracts for PMID 23325457, 24366933 and 30969315 read through PubMed and Europe PMC.) · Note reworded to what the abstracts state (a single academic tertiary care centre, a derivation cohort of high-risk tumours, a call for population-based validation). The comparison now gives 71 and 70 percent of metastases and 85 and 92 percent of deaths.
- 2026-11-17 · What makes a skin cancer high risk: The national audit was said to have found risk status one of the most frequently omitted items in skin cancer reports. The dataset says a frequently omitted item in basal cell and squamous cell carcinoma reports. (Second accuracy pass; the Royal College of Pathologists dataset G124 read from its PDF.) · Reworded to the dataset's own statement.
- 2026-11-17 · Actinic keratosis: A 2009 study of 7,784 lesions was called the largest prospective study, two sentences before a 1988 study of 21,905 lesions that had no link. Two figures credited to the 2017 British guideline could not be re-read. Separately, two source labels in the same file named the wrong authors for a Dutch registry study and a Gorlin syndrome surveillance paper. (Second accuracy pass; the abstracts for DOI 10.1002/cncr.24284 and PMID 2895318 read through Europe PMC and PubMed, and every DOI in the file resolved through Crossref.) · Superlative removed, the 1988 Lancet paper linked by its resolved DOI, the guideline figures labelled as not re-read, and the two author lists corrected (Tokez, Wakkee, Louwman; Guerrini-Rousseau, Smith, Kratz).
- 2026-11-17 · Cutaneous squamous cell carcinoma: The outlook paragraph said the ten-year figures by Brigham stage came from cohorts selected for difficulty and a referral population. They come from all 1,818 primary tumours diagnosed at one hospital from 2000 to 2009. The page and the skin cancer family page also said the Brigham system had been validated only in single-centre cohorts, and that mortality rose over 2013 to 2019. (Second accuracy pass; the abstracts for DOI 10.1200/jco.2012.48.5326, 10.1001/jamadermatol.2013.2456 and 10.1093/bjd/ljag315 read through Europe PMC, and the abstract of PMID 34862598 read through PubMed.) · The cohort is described as it was. The nationwide English validation of 2022 (887 metastatic cases and 887 controls; highest specificity for the Brigham system, positive predictive value 13.2 percent) is added with its link. The death rate trend is dated 2013 to 2022.
- 2026-11-17 · Skin cancer: The page said deaths recorded as non-melanoma skin cancer were rising 4 percent a year, where the source gives the age-standardised death rate. It said the joint audit found risk status one of the most frequently omitted report items, where the dataset says a frequently omitted item. It gave 403,601 for all other cancers, a figure not on the Cancer Research UK page read. It said everything but low-risk basal cell carcinoma is treated in secondary care, omitting the Model 2 practitioner in primary care. (Second accuracy pass; the abstract for DOI 10.1093/bjd/ljag315 read through Europe PMC, the Royal College of Pathologists datasets G123 and G124 read from their PDFs, and the Cancer Research UK incidence pages read as raw text.) · Each sentence reworded to what its source states.
- 2026-11-17 · Basal cell carcinoma: The page said most tumours contain mixed growth patterns, where the dataset says many. It credited the statement that the nodular pattern is the commonest to the pathology dataset, which does not say it. It said the cancer metastasises in well under one in a thousand cases and that millions of lesions are treated, with no source for either. A three-year net survival of 101.9 percent could not be re-read. (Second accuracy pass; the Royal College of Pathologists dataset G123 read from its PDF, the Cancer Research UK basal cell carcinoma page read as raw text, and the abstract for DOI 10.1093/bjd/ljad033 read through Europe PMC (the publisher refused the full text).) · Most changed to many. The nodular statement is credited to Cancer Research UK. The two unsourced figures are replaced with the sources' own words. The survival figure is labelled as not re-read, with the abstract's statement that net survival is above 100 percent.
- 2026-11-17 · Bowen's disease (squamous cell carcinoma in situ): The page called the lower leg in an older woman the classic site and the lower leg the commonest site. Cancer Research UK lists the legs among several common sites, and the Dutch national cohort found the face commonest in women and the scalp or neck in men. The figure of about 10 percent for erythroplasia of Queyrat could not be re-read. (Second accuracy pass; the Cancer Research UK Bowen's disease page read as raw text and the abstract for DOI 10.1001/jamadermatol.2020.1988 read through Europe PMC. The two British guidelines were refused by their publishers.) · The lower leg is described as a common site. A sentence says which figures were confirmed and which were not.
- 2026-11-17 · Inherited syndromes that cause skin cancer: Gorlin syndrome and xeroderma pigmentosum: The Gorlin surveillance paper was cited as Foulkes 2021. Its first author is Guerrini-Rousseau. The term also said many of the xeroderma pigmentosum complementation groups cause progressive neurological disease, where the review says a minority of patients do. (Second accuracy pass; DOI 10.1007/s10689-021-00247-z resolved through Europe PMC and its full text read, and the full text for DOI 10.1186/1750-1172-6-70 read through Europe PMC.) · Citation corrected to Guerrini-Rousseau 2021 and the neurological sentence reworded to the review's words.
- 2026-11-17 · Perineural invasion: The skin note said the eighth edition of TNM had the nerve-calibre and nerve-depth criteria. The UK datasets say only the AJCC version did, and that the UICC version was confined to a named nerve. (Second accuracy pass; the Royal College of Pathologists datasets G123 and G124 read from their PDFs.) · Note rewritten to give both versions.
- 2026-11-17 · ADMEC-O: The record, the Merkel cell carcinoma treatment row and its open problem all said STAMP, the phase 3 trial of adjuvant pembrolizumab, had not reported. It reported in 2025 and again in 2026. The short summary also gave the gain with nivolumab without saying it was not statistically significant. (Second accuracy pass; the American Society of Clinical Oncology 2026 abstracts LBA9505 (STAMP) and LBA9504 (ADAM) read as raw text through Crossref, two 2026 reviews read in full through Europe PMC, and the registry record NCT03712605.) · STAMP is now described as reported in abstracts only: 293 patients, overall relapse-free survival not statistically significant (one-sided p=0.10), relapse-free survival counting only Merkel cell carcinoma events hazard ratio 0.66 (90 per cent confidence interval 0.45 to 0.96). ADAM, of adjuvant avelumab in 100 patients, is added with its hazard ratio of 0.61 (0.32 to 1.16). The short summary now says the ADMEC-O difference was not statistically significant.
- 2026-11-17 · C-POST: The record called C-POST the first randomised adjuvant trial in skin squamous cell carcinoma and did not mention KEYNOTE-630, the phase 3 trial of adjuvant pembrolizumab that was stopped for futility. It also did not say that the adjuvant use is now on the United States label. (Second accuracy pass; the results section of the registry record NCT03833167, three 2026 review abstracts read through PubMed, and the Libtayo label and approval history read through openFDA.) · KEYNOTE-630 added with its registry result (450 patients, recurrence-free survival hazard ratio 0.76, 95 per cent confidence interval 0.53 to 1.10, not significant) and a link to the registry. C-POST is now described as the only randomised adjuvant trial in this disease to meet its primary endpoint. The label change of October 2025 is stated.
- 2026-11-17 · Cutaneous squamous cell carcinoma: The England row and an open problem said adjuvant cemiplimab and cosibelimab had no NICE appraisal, with no sign that either was on its way. The radiotherapy row said only 7 per cent failed first at a distant site, and three passages gave the transplant recipient's risk as 65 to 100 times normal. (Second accuracy pass; the NICE project pages GID-TA11869 and GID-TA11883 read with curl, the TROG 05.01 abstract (PMID 29537906) and the full text of the kidney transplant study (PMC10950183) read through PubMed and Europe PMC.) · The page now says the adjuvant cemiplimab appraisal is in progress with publication expected on 24 March 2027 and that a cosibelimab appraisal has been selected. The TROG sentence now reads as the source does: isolated distant metastasis was the first failure in 7 per cent of each arm. The transplant risk is given as 65 to 250 times, the figure in the linked paper. The TROG 05.01 and TUMORAPA records and the glossary entry on skin cancer after a transplant carry the same wording.
- 2026-11-17 · Sirolimus: The record said sirolimus was the drug that transplant recipients were switched to before cemiplimab in the Dana-Farber study. (Second accuracy pass; the full text of the study (PMC10950183) read through Europe PMC.) · The protocol used sirolimus or everolimus, and the one death followed everolimus. The sentence now says so.
- 2026-11-17 · STEVIE: The short summary said the average person stopped vismodegib after about eight months. The source gives a median of 8.6 months. (Second accuracy pass; the abstract (PMID 29073584) read through PubMed.) · Reworded to say half had stopped within about nine months.
- 2026-11-17 · Sun protection after a skin cancer diagnosis: The short summary said that anyone who is immunosuppressed is recommended SPF 50 every day of the year. The leaflet it rests on makes that recommendation to people with an organ transplant. The record, the skin cancer page and its question set also said the Scientific Advisory Committee on Nutrition recommends a daily 10 microgram vitamin D supplement, where the source says people should consider taking one. (Second accuracy pass; the British Association of Dermatologists transplant leaflet and sun protection fact sheet and the Cancer Research UK page on skin care after skin cancer, each read as raw page text.) · The summary now says organ transplant. The vitamin D sentence now says people are recommended to consider a daily 10 microgram supplement, in all three places.
- 2026-11-17 · Skin cancer: The Mohs surgery row said the local anaesthetic stings for less than 30 seconds and that Mohs takes half a day against minutes for a standard excision. The leaflet says the injection may sting at first and settles within 30 seconds or less, and gives no time for a standard excision. The immunosuppression row and a note described the Brantsch study as 615 squamous cell carcinomas; it followed 615 patients. (Second accuracy pass; the British Association of Dermatologists Mohs leaflet read as raw page text and the Brantsch abstract (PMID 18617440) read through PubMed.) · Both Mohs statements reworded to the leaflet, here and on the Mohs surgery term. The study is now described as 615 patients with squamous cell carcinoma.
- 2026-11-17 · Cutaneous squamous cell carcinoma: The follow-up row said that follow-up checks all of the skin and the lymph nodes nearest the original cancer, credited to Cancer Research UK. The page says the doctor examines the skin and may do a full body skin examination, and does not mention examining lymph nodes. The row also relied on the patient leaflet for what the dermatology guideline says about follow-up. (Second accuracy pass; the Cancer Research UK follow-up page read as raw page text. The guideline itself (DOI 10.1111/bjd.19621) could not be read because the publisher refused the request.) · The sentence now follows the Cancer Research UK page. A note says the one to two year schedule is the leaflet's own summary and that the guideline could not be re-read on 17 November 2026.
- 2026-11-17 · Curettage and cautery: The short summary said the procedure leaves a round pale scar. Neither source linked from the record describes the scar's shape or colour. (Second accuracy pass; the British Association of Dermatologists basal cell carcinoma leaflet and the Cancer Research UK page on types of surgery read as raw page text.) · The phrase was removed.
- 2026-11-17 · Aminolevulinic acid: A note said the gel is approved in the European Union for superficial and nodular basal cell carcinoma without the condition attached to that approval. (Second accuracy pass; the European Medicines Agency page for Ameluz read as raw page text.) · The note now says the approval is for basal cell carcinoma that is unsuitable for surgery.
- 2026-11-17 · The next skin cancer: second primaries after a keratinocyte cancer: The record quoted the 40 and 80 percent five-year figures from the squamous cell carcinoma leaflet but did not link that leaflet, so the figures had no source on the page. (Second accuracy pass; the British Association of Dermatologists squamous cell carcinoma leaflet read as raw page text, where both figures appear.) · The leaflet was added to the record's links.
- 2026-11-17 · Rational treatment selection for Merkel cell carcinoma: The record said the trial had no published result, gave its enrolment as 400, and called it the only randomised trial ever attempted in Merkel cell carcinoma. The 400 was the target. The final report was published in December 2025 and the registry added it on 30 October 2026: 64 people were recruited and only five were randomised, so the trial closed without answering its question. (Second accuracy pass; the ISRCTN16290169 record read through the registry's query interface and the report abstract for DOI 10.3310/EEWD6684 read through Crossref.) · Enrolment set to 64 and the year of reporting to 2025; the short description, summary and result rewritten from the report; the claim narrowed to a randomised comparison of first treatment; the report added as a link and to the registry source label.
- 2026-11-17 · Skin cancer in the UK: NICE appraisals in progress and the product count: The source labels said cosibelimab was expected from NICE on 23 September 2027, that adjuvant pembrolizumab for squamous cell carcinoma was in development, and that NICE's skin cancer list held 43 published or terminated products including 27 appraisals, three terminated. NICE now gives no publication date for cosibelimab, suspended the pembrolizumab appraisal on 23 June 2026 because the company will not submit evidence, and lists 44 published or terminated products including 28 appraisals, four terminated. (Second accuracy pass; the NICE project pages GID-TA11869, GID-TA11883 and GID-TA11582 and the skin cancer product list read as raw text.) · Both labels corrected with the date they were re-read.
- 2026-11-17 · Skin cancer in the UK: source labels: The Cancer Research UK label dated the figure of around 156,000 new non-melanoma skin cancers a year to 2019 and 2021 to 2022; the page labels it 2019, 2016-2018. The link to version 9 of the rare and inherited disease test directory was dead. The page cited for the seven genomic laboratory hubs does not mention them. The All Wales page cited for how NICE guidance applies in Wales answered page not found. (Second accuracy pass; the Cancer Research UK incidence page, NHS England's test directory document and Genomic Medicine Service pages read through its publishing interface, and the All Wales Therapeutics and Toxicology Centre site read directly.) · Years corrected; the directory link moved to version 9.2 of 9 November 2026, in which the four tests cited are unchanged; the hubs citation moved to NHS England's page on the structure of the service; the Welsh citation moved to the centre's live page on how medicines are approved in Wales.
- 2026-11-17 · Skin cancer in the UK: waiting-time figures: England figures stopped at July 2026, Scotland and Northern Ireland at the quarter to March 2026, and the label for the 2023 standards letter still described the rise of the Faster Diagnosis Standard to 80 percent as a future condition. The July figures themselves were correct. The Welsh summary page could not be re-read. (Second accuracy pass; NHS England's July and September 2026 monthly files recomputed, Public Health Scotland's workbooks of 30 June and 29 September 2026, the Northern Ireland workbooks for the quarters ending March and June 2026, and NHS England's Medium Term Planning Framework read through its publishing interface.) · September 2026 added beside July (84,802 of 297,492 urgent referrals were for suspected skin cancer, 28.5 percent; 82.1 percent within 28 days, 85.7 percent within 31 days and 85.8 percent within 62 days); the April to June 2026 Scottish and Northern Irish figures added to their labels, with the revised Northern Irish January to March count; the new 80 percent threshold noted; the Welsh label now says it could not be re-read.
- 2026-11-17 · SCC-AFTER: The record said radiotherapy after surgery for high-risk squamous cell carcinoma had never been tested in a randomised trial and was given without evidence. The registry entry it cites says only that the treatment is sometimes given and that there is no certainty it stops the cancer coming back. (Second accuracy pass; the ISRCTN54806122 record read through the registry's query interface.) · Reworded to the registry's statement and the registered recruitment dates added.
- 2026-11-17 · MoleMate UK Trial: The record said best practice in the trial included the weighted 7-point checklist and that the trial is the reason the NHS asks for training rather than devices. The paper says seven point checklist, and no source read ties the policy to the trial. The results quoted were correct. (Second accuracy pass; the abstract and full text of the BMJ paper (PMID 22763392) read through Europe PMC.) · Checklist wording corrected and the short description now states the finding and the current guidance as separate facts.
- 2026-11-17 · Skin cancer in the UK: the Merkel cell carcinoma trial: The page said in four places that the British randomised trial of surgery against radiotherapy for Merkel cell carcinoma had no published result and no recorded final enrolment, and called it the only randomised trial ever attempted in the disease. The report was published in December 2025 and added to the registry on 30 October 2026: 64 people were recruited, five were randomised, and the trial closed for missing its feasibility target. Other randomised trials in Merkel cell carcinoma are held elsewhere in this corpus. (Second accuracy pass; the ISRCTN16290169 record read through the registry public API, and the abstract for DOI 10.3310/EEWD6684 read through Crossref and PubMed (PMID 41370622).) · The presentation card, the centre note, the legacy story and the named gap now give the reported outcome and cite the report, and the claim that it was the only trial is removed.
- 2026-11-17 · Skin cancer in the UK: England waiting times: England figures stopped at July 2026 and the Faster Diagnosis Standard target was given as 75 percent, rising to 80 percent. September 2026 data are published and show skin performance lower than in July (82.1 percent at 28 days against 87.3, 85.7 percent at 31 days against 88.2, 85.8 percent at 62 days against 87.4), and the threshold is 80 percent from 2026/27. (Second accuracy pass; NHS England monthly combined waiting-times files for April to July and for September 2026 recomputed from the raw CSV, and the Medium Term Planning Framework read through NHS England's content API.) · The timeline and figures now give September 2026 (84,802 of 297,492 urgent referrals, 28.5 percent) with July kept for comparison, the 80 percent threshold, and the planning targets of 94 percent at 31 days and 80 percent at 62 days by March 2027.
- 2026-11-17 · Skin cancer in the UK: Scotland and Northern Ireland waiting times: Scottish and Northern Irish figures were for the quarter to March 2026. The page also said Northern Ireland publishes no 62-day figure by tumour site, and gave 399 of 446 for its 31-day skin figure, which has since been revised to 398. (Second accuracy pass; Public Health Scotland table 1 of 29 September 2026 and the Department of Health workbook for the quarter ending June 2026 (released 1 October 2026) read as raw spreadsheets.) · Scotland now shows 86.8 percent (198 of 228) for melanoma at 62 days. Northern Ireland shows 87.3 percent (364 of 417) at 31 days for skin, 28.4 percent at 62 days for all cancers, and the new tumour-site table: 115 of 272 skin cancers (42.3 percent) treated within 62 days. The Welsh figure is labelled as not re-read because the page refused the request.
- 2026-11-17 · Skin cancer in the UK: appraisals in progress: The funding table said cosibelimab was expected from NICE on 23 September 2027 and that adjuvant pembrolizumab for cutaneous squamous cell carcinoma was in development. NICE now gives no publication date for cosibelimab, and suspended the pembrolizumab appraisal on 23 June 2026 because the company will not submit evidence. The note also called a pending Scottish submission for retifanlimab in anal cancer "advice". (Second accuracy pass; the NICE project pages GID-TA11883, GID-TA11869 and GID-TA11582 and the Scottish Medicines Consortium pages SMC2977 and SMC3019 read as raw HTML.) · The note now gives the committee dates, the suspension and its reason, the pending Scottish submission for adjuvant cemiplimab, and describes the retifanlimab entry as a submission with no meeting date.
- 2026-11-17 · Skin cancer in the UK: non-melanoma incidence years: The figure of around 156,000 new non-melanoma skin cancers a year was dated 2019 and 2021 to 2022, and a named gap said the headline count had moved to those years. Cancer Research UK labels it "2019, 2016-2018" and its overview gives 155,985 cases for 2016 to 2018. The same gap dated the sunbed evidence review to June 2018; the page says 1 October 2018. (Second accuracy pass; the Cancer Research UK non-melanoma skin cancer overview, incidence and risk-factor pages read as raw HTML.) · The period now reads 2016 to 2018 in the figure and in both gaps, and the review dates are corrected.
- 2026-11-17 · Skin cancer in the UK: trials and SINS: The squamous cell carcinoma card said both open UK trials were about immunosuppressed people; SCC-AFTER enrols people with high-risk tumours whatever their immune status. IMPACT was counted as open although it closed to recruitment in September 2025. The imiquimod row said the cream is applied daily, and the SINS story said almost all failures came in the first year and that nobody had run such a trial before. (Second accuracy pass; the ISRCTN records read through the registry public API, the SINS abstracts (PMID 24332516 and 27932240) read through Europe PMC, and the Scottish Medicines Consortium advice document 167/05.) · The card now describes each trial; the trials note says IMPACT has closed to recruitment; the regimen reads five times a week for six weeks; the story says most imiquimod failures came in the first year and drops the unsupported claim.
- 2026-11-17 · STAMP (EA6174): The record said the trial was ongoing and had not reported, gave 280 patients, and said the earlier ADMEC-O trial had shown a recurrence-free survival advantage. STAMP has reported in conference abstracts: 293 patients, and overall relapse-free survival was not significantly better with pembrolizumab. ADMEC-O's hazard ratio had a confidence interval crossing one. (Second accuracy pass; the abstract for DOI 10.1200/JCO.2026.44.17_suppl.LBA9505 read through Crossref, the registry record NCT03712605 read through the ClinicalTrials.gov API, and the ADMEC-O abstract (PMID 37451295) read through PubMed.) · Status changed to mixed with 2025 as the year first reported and 293 enrolled; the summary now gives the one-sided p of 0.10 for overall relapse-free survival, the hazard ratio of 0.66 (90 percent confidence interval 0.45 to 0.96) for relapse-free survival counting only Merkel cell carcinoma events, and says overall survival is not mature, the results are abstracts only, and ADMEC-O's 0.58 was not statistically significant. Chinese summary line updated to match.
- 2026-11-17 · Keratoacanthoma: The short description said the WHO treats keratoacanthoma as a type of squamous cell carcinoma. That was the 2018 edition; the 2025 edition gives it a heading of its own. Two lines on the cutaneous squamous cell carcinoma page repeated the older position. (Second accuracy pass; the table of contents of the WHO skin tumours fifth edition read as text from the IARC file, and the review of that edition (DOI 10.1111/his.15562) read in full through Europe PMC.) · All three passages now say the WHO filed it under squamous cell carcinoma in 2018 and gave it a heading of its own in 2025.
- 2026-11-17 · The Brigham and Women's Hospital staging system: A note said every figure came from one Boston hospital and that validation in a whole population was still awaited. A nationwide English study has done it. (Second accuracy pass; the abstract of Venables and colleagues, British Journal of Dermatology 2022 (PMID 34862598), read through PubMed.) · Note now gives the English result: 887 metastatic cases and 887 controls from 2013 to 2015, the Brigham system with the highest specificity (92.8 percent), positive predictive value (13.2 percent) and c-index (0.84), AJCC 8 with the better negative predictive value (99.2 percent), and a low positive predictive value for all four systems. Source added.
- 2026-11-17 · What makes a skin cancer high risk: The referral rule said every basal cell carcinoma that is not low risk goes to secondary care, and an audit finding was credited to both reporting datasets when only one states it. (Second accuracy pass; the Royal College of Pathologists datasets G123 and G124 read as text extracted from the published files.) · Rule now follows G123: all other basal cell carcinomas must be treated in secondary care or by a Model 2 practitioner in primary care, with that role explained. The audit finding is attributed to G124.
- 2026-11-17 · How skin carcinoma is staged in Britain: A note implied the UICC eighth edition already defined perineural invasion by nerve calibre and depth. The College of Pathologists records that UICC 8 was confined to a named nerve and that it chose to apply the other criteria itself. (Second accuracy pass; dataset G123 and the UK appendix for the ninth edition read as extracted text.) · Note reworded to give the ninth edition definition as the appendix states it and the eighth edition position in the College's words.
- 2026-11-17 · Skin cancer: The first 60 days checklist said a daily vitamin D supplement is recommended in the UK between October and March, and the curettage entries described a round pale scar that no cited source mentions. (Second accuracy pass; the Cancer Research UK pages on skin care after skin cancer and on types of surgery, and the British Association of Dermatologists basal cell carcinoma leaflet, read as page text.) · Vitamin D line now says people are advised to consider a supplement, as the source does. The scar description is removed from the curettage and cautery term and from the low-risk basal cell carcinoma decision aid, which now say only that the wound is not stitched and scabs over.
- 2026-11-17 · Skin cancer in the UK: The four nations table said the Welsh medicines group does not appraise a medicine NICE is appraising, citing a page that no longer exists, and the IMPACT trial of cemiplimab in basal cell carcinoma was still tagged as recruiting. The same dead Welsh link, and a withdrawn version of the NHS genomic test directory, were cited on the breast cancer, triple-negative breast cancer and inherited syndrome pages. (Second accuracy pass; the All Wales Therapeutics and Toxicology Centre page on how medicines are approved read as raw page text, the registry record ISRCTN10511385 read through the ISRCTN interface, and the version 9.2 test directory workbook downloaded and read.) · Welsh row reworded to what the live page says (the group usually assesses medicines NICE has not assessed or selected, and health boards are usually expected to provide a recommended medicine within 60 days). Recruiting tag removed, since the registered recruitment window closed on 1 September 2025. Dead links replaced with the live Welsh page and with version 9.2 of the directory, which marks the cited tests as unchanged.
- 2026-11-17 · Melanoma: Lifileucel was listed among drugs still in development for melanoma although the FDA approved it on 16 February 2024 and the page already gives it as a treatment (Second accuracy pass; the FDA product page for Amtagvi read in raw text) · Removed from the list of drugs in development for melanoma
- 2026-11-17 · HARMONi-3: The trial was marked as reported with a mixed result: an interim analysis in May 2026 was said to have missed statistical significance for progression-free survival. The HARMONi-2 record and the lung cancer summary repeated it. The trial has reported no efficacy result. (Second accuracy pass; Summit's reports for the first and second quarters of 2026 read as raw text from Internet Archive copies, and the registry record NCT05899608. The sponsor describes an early interim look at the squamous cohort only, with a deliberately high bar, seen only by the monitoring committee, which recommended continuing; the primary analysis of that cohort was expected in the second half of 2026.) · Status set to active, the reported year and result removed, the summary rewritten from the sponsor's two reports with links to both, and the HARMONi-2 record and the lung cancer summary corrected.
- 2026-11-17 · Non-small-cell lung cancer: The summary still said an FDA decision on ivonescimab was due on 14 November 2026 and the 2026 history entry said it was under review. Zidesamtinib was given as a first-line option for ROS1 disease, sevabertinib as a drug only for use after earlier treatment, and screening uptake as under 20 percent of eligible people in the US. The short record said adjuvant osimertinib halves the risk of the cancer returning and still listed HER3-DXd among the coming antibody-drug conjugates. (Second accuracy pass; Summit's release of 15 November 2026 (Internet Archive copy), the US labels for Jideytro, Hyrnuo and Hernexeos read through openFDA, the American Lung Association State of Lung Cancer key findings page, the ADAURA abstracts (PMID 32955177 and 37272535) and Merck's release of 29 May 2025.) · The complete response letter is stated; zidesamtinib is placed after a prior ROS1 inhibitor, as its label of 22 July 2026 reads; sevabertinib is shown for untreated patients too, added on 9 September 2026; screening uptake reads 24.4 percent; the osimertinib line now gives the halving of deaths (hazard ratio 0.49); HER3-DXd is removed from the list with its withdrawal stated.
- 2026-11-17 · Krascendo 1: The record said the numbers were pending and due at a 2026 congress. The full results were published in October 2026. (Second accuracy pass; the abstract of the New England Journal of Medicine paper (PMID 42929675) read through PubMed.) · Results added: median progression-free survival 9.8 against 5.7 months (hazard ratio 0.58) and median overall survival 17.7 against 12.0 months (hazard ratio 0.56) in 338 patients, with the note that 156 of 171 control patients received sotorasib and 15 adagrasib. The divarasib record carries the same figures.
- 2026-11-17 · SOHO-01: The approval date was given as 20 November 2025, the response rate as about 64 percent, and the first-line application as still under review. The same appeared on the sevabertinib record. (Second accuracy pass; the openFDA approval history for NDA 219972 and the Hyrnuo label effective 17 September 2026.) · Approval date corrected to 19 November 2025; response rates replaced with the label's (71 percent of 70 previously treated patients, 38 percent of 52 after a HER2 antibody-drug conjugate, 75 percent of 69 untreated patients); the extension to untreated patients on 9 September 2026 added to the trial, the drug and its approvals.
- 2026-11-17 · KRYSTAL-12: The record said the trial supported full approval of adagrasib and that overall survival was not yet significantly different. (Second accuracy pass; the Krazati label effective 31 August 2026 read through openFDA, the Lancet abstract (PMID 40783289) and the registry results.) · The lung indication is stated as still an accelerated approval, and the record now says the primary publication does not report overall survival.
- 2026-11-17 · Pralsetinib: The response rate in untreated patients was an older figure (about 72 percent), and the record said the thyroid indications had been withdrawn in the US. (Second accuracy pass; the Gavreto label effective 30 December 2025 read through openFDA.) · Response rates replaced with the label's (78 percent of 107 untreated patients, 63 percent of 130 after platinum chemotherapy). The record now says the medullary thyroid cancer indication is no longer on the label while the RET fusion-positive thyroid cancer indication remains under accelerated approval.
- 2026-11-17 · Olomorasib: A response rate of 74 percent was credited to first-line patients with PD-L1 of 50 percent or more. It is the rate for all first-line patients. (Second accuracy pass; the abstract for PMID 41319864 read through Europe PMC.) · Corrected to 74 percent first line at any PD-L1 level and 90 percent in first-line patients with PD-L1 of 50 percent or more, and the study named as phase 1/2.
- 2026-11-17 · Dabrafenib + trametinib: The tumour-agnostic indication was given for patients aged 6 and over with no mention that it is an accelerated approval. (Second accuracy pass; the Tafinlar label effective 20 May 2026 and the openFDA approval history.) · The record now says the 2022 approval was accelerated, for previously treated tumours, and that the current label covers patients aged 1 year and over.
- 2026-11-17 · Repotrectinib: Zidesamtinib was described as an alternative with fewer neurologic effects, with no mention that it is licensed only after an earlier ROS1 inhibitor; the NTRK indication was not marked as an accelerated approval. The ROS1 target record had the same gap. (Second accuracy pass; the Jideytro and Augtyro labels read through openFDA and the abstract for PMID 42909821.) · Both records now give zidesamtinib's indication as approved on 22 July 2026, and the NTRK indication is marked accelerated.
- 2026-11-17 · Amivantamab + lazertinib (first-line EGFR NSCLC): The summary line said the combination added more than a year of survival, and the lazertinib record said the median was more than 12 months longer than with osimertinib. The median was not reached in the combination arm. (Second accuracy pass; the MARIPOSA overall survival abstract (PMID 40923797) and Table 24 of the Rybrevant label.) · Replaced with what was measured: hazard ratio for death 0.75, 60 against 51 percent alive at three years, median not reached against 36.7 months.
- 2026-11-17 · KEYNOTE-671: The record called the regimen the first with a survival benefit at its approval in October 2023, and gave one overall survival hazard ratio with no analysis named. (Second accuracy pass; the abstracts for PMID 37272513, 39288781 and 42628840.) · The record now says survival was not yet significantly different at the first analysis and became so at the second interim analysis in 2024, and the result gives both that hazard ratio (0.72) and the five-year one (0.74).
- 2026-11-17 · CROWN: A five-year cumulative incidence of brain progression of about 8 percent was quoted that the cited report's abstract does not give. (Second accuracy pass; the abstract for PMID 38819031 read through PubMed.) · Replaced with the abstract's figure: median time to intracranial progression not reached with lorlatinib against 16.4 months with crizotinib, hazard ratio 0.06, and the analysis named as investigator-assessed and post hoc.
- 2026-11-17 · Low-dose CT lung screening: US screening uptake was given as under 20 percent. (Second accuracy pass; the American Lung Association State of Lung Cancer key findings page, updated 10 November 2026, read as raw text.) · Corrected to 24.4 percent of eligible people, with links to that page and to the USPSTF recommendation.
- 2026-11-17 · DeLLphi-305: The page said the trial had met its survival goal but that the figures were not yet disclosed, described the trial as still testing the idea, and credited the announcement to Amgen and AstraZeneca. The small-cell lung cancer page, the maintenance pairing and the IMforte page repeated that the result was pending. (Second accuracy pass; the abstract of the New England Journal of Medicine report of 24 October 2026 (DOI 10.1056/NEJMoa2609482, PubMed 42929702) read through PubMed and resolved on Crossref, and the Amgen release of 8 September 2026 read as raw text.) · Results added: median overall survival 21.1 against 12.7 months (hazard ratio 0.42, 95% CI 0.32 to 0.54) and progression-free survival 7.4 against 1.9 months at an interim analysis, with the note that the use is not yet on the US label. The announcement is credited to Amgen alone.
- 2026-11-17 · Small-cell lung cancer: The summary line said antibody-drug conjugates had lengthened survival since 2024 and that there had been 30 years without progress; the summary gave ifinatamab deruxtecan a response rate of about 55%, listed lurbinectedin as a second-line option with no note that its US indication was removed, said RYZ101 was in phase 3, called the question of brain scans against preventive brain radiotherapy settled, and dated the topotecan approval to 1996. (Second accuracy pass; the IDeate-Lung01 primary analysis abstract (DOI 10.1200/JCO-25-02142) read through Europe PMC, the ClinicalTrials.gov entries for NCT06203210, NCT05595460 and NCT04155034, the US lurbinectedin label effective 30 September 2026 read on openFDA, and the FDA approval history for topotecan.) · The response rate is 48.2% in 137 patients in a single-arm phase 2 trial and the conjugate is still in phase 3; the two relapsed-disease rows say the US lurbinectedin indication was removed in September 2026; RYZ101 is in phase 1 in this cancer; the brain radiotherapy question is described as unsettled, with the trial closed to recruitment and unreported; topotecan is dated 1998.
- 2026-11-17 · ADRIATIC: The page said 730 patients were randomised to durvalumab or placebo. (Second accuracy pass; the abstract for DOI 10.1056/NEJMoa2404873 read through Europe PMC, the US durvalumab label read on openFDA and the ClinicalTrials.gov record for NCT03703297.) · The 730 patients were randomised across three arms; the reported comparison is 264 on durvalumab against 266 on placebo at an interim analysis, and results for the 200 patients given durvalumab plus tremelimumab were not posted when checked.
- 2026-11-17 · CASPIAN: The three-year survival figures (17.6% against 5.8%) were sourced to the 2019 Lancet paper, which reports only 18-month rates. (Second accuracy pass; the abstracts for DOI 10.1016/S0140-6736(19)32222-6 and DOI 10.1016/j.esmoop.2022.100408 read through Europe PMC.) · The figures are kept and sourced to the 2022 three-year update in ESMO Open.
- 2026-11-17 · ASTRUM-005: The page said serplulimab produced the longest survival of the chemo-immunotherapy trials, did not say the result was an interim analysis, and said the US bridging study ASTRIDE had only completed enrolment. (Second accuracy pass; the JAMA abstract (DOI 10.1001/jama.2022.16464) and the 2026 JAMA Oncology follow-up (PubMed 42240984) read through Europe PMC and PubMed, the Henlius release of 27 October 2026 read as raw text, and the EMA page for Hetronifly.) · The superlative is replaced by the medians with the caveat that trials cannot be compared directly; the interim status and the longer follow-up (15.8 against 11.1 months, 21.9% against 7.2% alive at four years) are stated; ASTRIDE is reported as descriptive (12.4 against 10.2 months, hazard ratio 0.82, 95% CI 0.56 to 1.20).
- 2026-11-17 · Serplulimab: The page said serplulimab had the largest survival gain of any first-line small-cell immunotherapy trial, that an FDA decision was pending, and that it was approved in Korea; the EU approval of February 2025 was sourced to a May 2026 release about two other indications. (Second accuracy pass; the Henlius releases of 24 June 2025, 10 May 2026 and 27 October 2026 read as raw text, the EMA Hetronifly page, NICE TA1167 read on nice.org.uk, and an openFDA search that finds no US label.) · The superlative and the Korea claim are removed, the US status reads not approved with the descriptive ASTRIDE result and no filing date given, the EU approval is dated 3 February 2025 and sourced to the EMA, and the NICE recommendation of 18 June 2026 is added. The immune side-effect percentages are labelled as not re-read because the full text was not available.
- 2026-11-17 · Topotecan: The page said topotecan was approved for relapsed small-cell lung cancer in 1996, and gave a response rate of about 20% and survival of 6 to 8 months without a source. (Second accuracy pass; the FDA approval history on openFDA (first approval 28 May 1996, efficacy supplement 30 November 1998), a 2022 review read in full text through Europe PMC (PMC9330463), and the US topotecan and tarlatamab labels.) · The small-cell approval is dated 1998 with the 1996 first approval noted, on the page and in the small-cell history; the activity figures are replaced by 24% and 5.8 months from the registration trial and 8.3 months from the DeLLphi-304 control arm.
- 2026-11-17 · Lurbinectedin: The dosing line still described use on its own, which is no longer a labelled use in the United States, and cited a label file that could not be read; the page did not carry the NICE outcome or the status of the LAGOON trial. (Second accuracy pass; the US label effective 30 September 2026 read on openFDA, the EMA Zepzelca page, NICE TA1198 read on nice.org.uk and the ClinicalTrials.gov record for NCT05153239.) · Dosing follows the current label and cites it; the summary adds the EU authorisation of 29 May 2026, the NICE appraisal ended without a recommendation on 13 October 2026, and LAGOON listed as completed with no results posted. The PharmaMar page now says the relapsed indication was removed.
- 2026-11-17 · IMforte: The page said the trial was presented at the ASCO 2025 plenary session and that tarlatamab maintenance was still being tested. (Second accuracy pass; the Roche release of 3 October 2025 read as raw text, which cites ASCO abstract 8006, and the DeLLphi-305 report on PubMed.) · Reworded to the ASCO 2025 annual meeting (abstract 8006), and the DeLLphi-305 result is stated.
- 2026-11-17 · CONVERT: The sponsor was given as Cancer Research UK and the EORTC. (Second accuracy pass; the ClinicalTrials.gov record for NCT00433563 and the funding statement in the abstract for DOI 10.1016/S1470-2045(17)30318-2.) · The sponsor is The Christie NHS Foundation Trust; Cancer Research UK and the EORTC are named as funders.
- 2026-11-17 · Prophylactic cranial irradiation vs MRI surveillance: The page said the 1999 meta-analysis showed a survival gain in limited-stage disease and that hippocampal avoidance reduced cognitive harm in both PREMER and NRG CC003. (Second accuracy pass; the abstracts for PubMed 10441603, 34379442 and 40789106 read through Europe PMC, and the ClinicalTrials.gov record for NCT04155034.) · The meta-analysis is described as patients in complete remission; NRG CC003 is reported as missing its primary endpoint while reducing failure on any cognitive test (hazard ratio 0.78); the MAVERICK trial is described as closed to recruitment with no results.
- 2026-11-17 · Shanghai Henlius Biotech: The page said serplulimab received EU approvals in 2025 for extensive-stage small-cell lung cancer, squamous non-small-cell lung cancer and other indications. (Second accuracy pass; the EMA Hetronifly page and the Henlius release of 10 May 2026 read as raw text.) · The EU approvals are February 2025 for small-cell lung cancer and May 2026 for non-squamous non-small-cell lung cancer and oesophageal squamous cell carcinoma.
- 2026-11-17 · Subtype-directed therapy for SCLC (ASCL1 / NEUROD1 / POU2F3 / inflamed): The page said the evidence was preclinical and a stratified trial was only proposed. (Second accuracy pass; the ClinicalTrials.gov record for NCT06769126 and the abstract for PubMed 33482121.) · Added that the biomarker-directed randomised phase 2 PRISM trial (SWOG S2409) has been recruiting since November 2025.
- 2026-11-17 · evERA: The trial was said to have randomised 320 women, with a note that no journal paper had been indexed, and the new drug application was shown as accepted but not decided. (Second accuracy pass; the abstract for DOI 10.1056/NEJMoa2602457 read through PubMed and the FDA approval notice of 16 November 2026 read as raw text.) · The record now gives 373 randomised (183 and 190, with 207 ESR1-mutated), links the New England Journal of Medicine paper, and says the FDA approved giredestrant with everolimus for ESR1-mutated disease on 16 November 2026. The standard-of-care row, the oral SERD term, the history and the summary on the cancer page no longer say filed.
- 2026-11-17 · EMBER-3: The overall survival medians of 34.5 and 23.1 months in ESR1-mutated disease were shown without saying the difference did not reach statistical significance, and the updated progression-free survival interval was 0.46 to 0.75. (Second accuracy pass; the abstract of the updated analysis, DOI 10.1016/j.annonc.2025.11.018, read through PubMed.) · The record and the imlunestrant page now give the hazard ratio of 0.60 (0.43 to 0.86, P 0.0043) and say the boundary for significance was not reached at this interim analysis. The interval is corrected to 0.47 to 0.74, and the 18 September 2026 approval with abemaciclib, limited to ESR1-mutated disease, is added from the FDA notice.
- 2026-11-17 · FOURLIGHT-1: The comparator was given as fulvestrant alone, the result as a hazard ratio not yet published, the drug as causing less neutropenia than dual CDK4/6 inhibitors, the topline date as February 2026 on the atirmociclib page, and FOURLIGHT-3 as still enrolling. (Second accuracy pass; the Pfizer release of 17 March 2026 read as raw text, the abstract for DOI 10.1016/j.annonc.2026.10.1282 read through PubMed, and the ClinicalTrials.gov record for NCT06760637.) · The comparator is now the investigator's choice of fulvestrant alone or everolimus plus exemestane. The result is 9.4 against 6.6 months, hazard ratio 0.60 (0.44 to 0.83) in 263 randomised patients. The neutropenia claim is replaced by the reported 47.2 percent with no comparison arm. The date is 17 March 2026 and FOURLIGHT-3 is shown as fully enrolled with 1,035 patients.
- 2026-11-17 · lidERA: The record said a regulatory submission was planned and gave the result as about 30 percent with an approximate hazard ratio. (Second accuracy pass; the Roche releases of 19 May and 2 June 2026 read as raw text.) · The record now gives hazard ratio 0.70 (0.57 to 0.87, P 0.0014) and three-year rates of 92.4 and 89.6 percent, and says the FDA accepted the adjuvant application under priority review on 2 June 2026 with a decision date of 30 November 2026. The menopausal subgroup percentages are marked as not re-read.
- 2026-11-17 · SERENA-6: Only the interim result was shown, the United States approval was called the first indication of its kind although the European authorisation came first, and the camizestrant page listed no heart rhythm monitoring. (Second accuracy pass; the abstract for DOI 10.1016/S1470-2045(26)00287-1 read through PubMed, the EMA page for Etcamah, and the Etcamah label read from DailyMed.) · The extended analysis is added (16.8 against 9.2 months, hazard ratio 0.45, and a significant gain in second progression-free survival). The European authorisation of 20 July 2026 is stated, the first claim is limited to the FDA, and the label's boxed warning on QT prolongation and its ECG monitoring are on the camizestrant page.
- 2026-11-17 · MONARCH 3: The overall survival interval was 0.62 to 1.03, and both results were cited to the 2017 interim paper, which contains neither. (Second accuracy pass; the abstracts for DOI 10.1016/j.annonc.2024.04.013 and DOI 10.1038/s41523-018-0097-z read through PubMed.) · The interval is corrected to 0.64 to 1.02 and each result now cites the paper that reports it. The cancer page summary no longer says abemaciclib showed a first-line survival gain, since the 13 month difference was not statistically significant.
- 2026-11-17 · MONALEESA-2: The overall survival P value was 0.004, the progression-free survival figures were cited to the 2016 paper that does not contain them, and the summary line called this the first CDK4/6 inhibitor trial to show longer survival. (Second accuracy pass; the abstracts for DOI 10.1056/NEJMoa2114663, DOI 10.1093/annonc/mdy155 and DOI 10.1056/NEJMoa1903765 read through PubMed.) · The P value is the two-sided 0.008 the paper reports, the progression-free survival figures cite the 2018 update, and the summary line now notes that MONALEESA-7 showed a survival gain in 2019.
- 2026-11-17 · PALOMA-2: The updated progression-free survival and the final overall survival were cited to the 2016 primary paper, which reports 24.8 against 14.5 months and no survival result. The same fault affected the overall survival result of TROPiCS-02 and the response rate of INAVO120. (Second accuracy pass; the abstracts for DOI 10.1007/s10549-018-05125-4, DOI 10.1200/JCO.23.00137, DOI 10.1016/S0140-6736(23)01245-X and DOI 10.1056/NEJMoa2501796 read through PubMed.) · Each figure now cites the paper that reports it, and the INAVO120 response rate is labelled as the updated analysis with the primary figures beside it.
- 2026-11-17 · SOFT & TEXT: The summary said adding ovarian suppression to tamoxifen improved overall survival in the chemotherapy cohort with a hazard ratio of about 0.79. (Second accuracy pass; the abstracts for DOI 10.1200/JCO.22.01065 and DOI 10.1016/j.annonc.2026.05.704 read through PubMed.) · The summary now gives the whole-trial hazard ratio of 0.78 (0.60 to 1.01) as not statistically significant and adds the 2026 final analysis, which limits a meaningful survival benefit to high-risk subgroups.
- 2026-11-17 · PALLAS & PENELOPE-B: The combined enrolment was 6,862 and PALLAS was given as 5,796 women, and only the four-year result was shown. (Second accuracy pass; the abstracts for DOI 10.1016/j.annonc.2025.10.003 and DOI 10.1200/JCO.20.03639 read through PubMed.) · Enrolment is 7,003 (5,753 analysed in PALLAS and 1,250 in PENELOPE-B) with a note on the registry figure, and the five-year PALLAS result is added: 84.2 against 82.4 percent, hazard ratio 0.88 (0.77 to 1.01), not significant.
- 2026-11-17 · RxPONDER (SWOG S1007): The summary gave 5,083 women and 3,350 postmenopausal. (Second accuracy pass; the abstract and table 1 of DOI 10.1056/NEJMoa2108873 read through NCBI PMC.) · Corrected to 5,018 participants of 5,083 randomised, 3,353 of them postmenopausal.
- 2026-11-17 · postMONARCH: The summary said the largest gains were in tumours without ESR1 or PIK3CA mutations. (Second accuracy pass; the abstract for DOI 10.1200/JCO-24-02086 read through PubMed.) · Reworded to the paper's finding of a consistent effect with or without those mutations.
- 2026-11-17 · persevERA: The summary said 978 women were randomised and described a numerical improvement of about eleven percent. (Second accuracy pass; the Roche releases of 9 March and 19 May 2026 read as raw text and the ClinicalTrials.gov record.) · Enrolment is 992 as both sources state. The eleven percent phrase is removed and the medians are marked as not re-read, because Roche's releases give no figures.
- 2026-11-17 · EMERALD: The record gave 478 randomised with 239 in each arm. (Second accuracy pass; the full text of DOI 10.1200/JCO.22.00338 read through Europe PMC.) · Corrected to 477 randomised, 239 to elacestrant and 238 to standard therapy.
- 2026-11-17 · CAMBRIA-1 & CAMBRIA-2: Both trials were described as enrolling. The CAPItello-291 record also said final overall survival was awaited with no date. (Second accuracy pass; the ClinicalTrials.gov records for NCT05952557, NCT05774951 and NCT04305496 and Europe PMC searches.) · CAMBRIA-2 is shown as closed with 5,511 enrolled and CAMBRIA-1 as recruiting. CAPItello-291 now says no final survival report was found on 17 November 2026.
- 2026-11-17 · HR-positive / HER2-negative breast cancer: The burden line gave 2.3 million breast cancers a year with no edition. The history row dated the approvals of anastrozole and letrozole to 1998. One row placed gedatolisib under any genotype. The late recurrence term gave half of recurrences after year five and one to two percent a year. (Second accuracy pass; the repository copy of GLOBOCAN 2024, the SEER subtype page, openFDA approval records, the Revtorpyk label, and the abstract and text of DOI 10.1056/NEJMoa1701830.) · The burden is 2.4 million (2,434,087, GLOBOCAN 2024) with the SEER basis of the 70 percent share named. The history row is the 1998 tamoxifen risk-reduction approval with the 1995 and 1997 dates given. Gedatolisib is for disease with no PIK3CA mutation detected. The term gives the paper's 13 to 41 percent 20-year risks.
- 2026-11-17 · APHINITY: The page said overall survival was not significantly different and that no survival benefit had been shown. It also gave the eight-year hazard ratio as 0.78 (0.66 to 0.92) and said the primary endpoint was met at eight years. (Second accuracy pass; the Roche media release of 13 May 2025 read as raw HTML, the abstracts of the 2017 and 2024 reports read through PubMed, and three open papers quoting the 8.4-year analysis read through Europe PMC (the journal page for DOI 10.1200/JCO.23.02505 answered 403).) · The final analysis at a median 11.3 years is now stated: 91.6 against 89.8 percent alive at ten years, hazard ratio 0.83 (0.69 to 1.00), p=0.044, with no benefit in node-negative disease. The eight-year hazard ratio is 0.77 (0.66 to 0.91) and the primary endpoint is dated to the three-year analysis.
- 2026-11-17 · PHERGain: Three-year invasive disease-free survival was given as 95.4 percent; 37.9 percent was called the share of the whole arm that skipped chemotherapy; PHERGain-2 was called a phase 3 trial that was underway; and the short summary said women were cured. The same errors were on the PET-adapted pairing. (Second accuracy pass; the abstracts of the 2021 Lancet Oncology and 2024 Lancet reports and of the 2026 PHERGain-2 report (PMID 42097893) read through PubMed.) · The figure is 94.8 percent (95% CI 91.4 to 97.1). The 37.9 percent is 86 of 227 PET responders, about 30 percent of the 285 in the arm. PHERGain-2 is described as a single-arm phase 2 without PET that reported in 2026. An unsourced sentence about a 2026 congress was removed.
- 2026-11-17 · DESTINY-Breast12: The intracranial response rate of 71.7 percent was shown as the primary endpoint in 263 patients, and the page said responses were higher still in active or untreated brain metastases. (Second accuracy pass; the full text of DOI 10.1038/s41591-024-03261-7 read through Europe PMC.) · The primary endpoint in the brain metastasis cohort was progression-free survival (61.6 percent at twelve months). The response rate applies to 138 patients with measurable brain disease and was 79.2 percent with stable and 62.3 percent with active brain metastases.
- 2026-11-17 · HER2CLIMB-05: The page gave no approval and the cancer page listed tucatinib maintenance without one. (Second accuracy pass; the US Tukysa label (effective 14 October 2026) and the FDA approvals record read through openFDA.) · The FDA approval of tucatinib with trastuzumab and pertuzumab as maintenance treatment on 7 October 2026 is now stated on the trial, the cancer page and its treatment rows. Overall survival is described as not yet mature and the confidence interval is given.
- 2026-11-17 · HER2CLIMB: The page said 48 percent of patients had active brain metastases and gave only the first survival analysis. The linked pairing called the trial phase 3 and the only randomised evidence for brain metastases. (Second accuracy pass; the abstract of the 2020 report, the full text of the brain metastasis analysis (DOI 10.1200/JCO.20.00775) and the abstract of the final survival analysis read through PubMed and Europe PMC, and the registry record read.) · 48 percent had brain metastases at baseline or by history, treated or untreated. The final analysis is added: 24.7 against 19.2 months, hazard ratio 0.73. The pairing now says the trial is registered as phase 2 and that it was the first, not the only, such evidence.
- 2026-11-17 · HER2CLIMB-02: The page said the benefit was mostly in patients with brain metastases, cited a news site, and gave survival as no benefit without saying the analysis was an immature interim one. (Second accuracy pass; the abstract of the primary publication (DOI 10.1016/j.annonc.2025.11.005) read through PubMed.) · The benefit is described as seen across subgroups, the interim survival hazard ratio of 1.23 (0.87 to 1.74) is stated, the higher rate of severe side effects is added and the source is the paper.
- 2026-11-17 · CLEOPATRA: The primary progression-free survival row gave 18.7 months and a hazard ratio of 0.68, which are investigator-assessed figures from the 2015 report, against the 2012 paper as source. The page also called the survival gain the largest ever seen in metastatic breast cancer, with no source. (Second accuracy pass; the abstracts of the 2012, 2015 and 2020 reports read through PubMed, and the US Perjeta label read through openFDA.) · The row now gives the primary result: 18.5 against 12.4 months, hazard ratio 0.62 (0.51 to 0.75). The superlative is replaced by the medians, 57.1 against 40.8 months.
- 2026-11-17 · HERA, NSABP B-31 & NCCTG N9831 (adjuvant trastuzumab): A disease-free survival hazard ratio of 0.60 (0.53 to 0.68) was credited to the 2005 paper, which reported 0.48. (Second accuracy pass; the abstracts of Romond 2005 and Perez 2014 read through PubMed.) · The row is labelled as the update at a median 8.4 years and cites the 2014 paper.
- 2026-11-17 · PERSEPHONE: The sponsor was given as Cancer Research UK, and trastuzumab was said to be stopped for heart toxicity in four percent of the six-month group. (Second accuracy pass; the abstract and full text of DOI 10.1016/S0140-6736(19)30650-6 read through Europe PMC.) · The sponsors are Cambridge University Hospitals NHS Trust and the University of Cambridge, with funding from the National Institute for Health Research. The figure is three percent against eight percent.
- 2026-11-17 · HORIZON-Breast01: The page said the trial randomised 366 patients. (Second accuracy pass; the abstract of DOI 10.1016/S1470-2045(26)00193-2 read through PubMed.) · The primary comparison randomised 287 patients, 142 and 145. The confidence interval (0.15 to 0.34) and follow-up are added.
- 2026-11-17 · Can T-DXd alone cure early HER2-positive disease?: The idea rested on trastuzumab deruxtecan producing complete responses in two thirds of patients and matching the best chemotherapy. (Second accuracy pass; the abstract of the DESTINY-Breast11 report (DOI 10.1016/j.annonc.2025.10.019) read through PubMed.) · Two thirds was the rate for trastuzumab deruxtecan followed by THP. The drug alone gave 43.0 percent against 56.3 percent for standard chemotherapy, and that arm closed early. The idea now says so.
- 2026-11-17 · Trastuzumab rezetecan: The lung cancer response rate was given as 74.5 percent with a progression-free survival of 11.5 months. (Second accuracy pass; the abstract of the HORIZON-Lung phase 2 report (DOI 10.1016/S1470-2045(25)00012-9) read through PubMed.) · The published figure is 69 of 94 patients, 73 percent. The survival figure is not in the abstract and was removed.
- 2026-11-17 · ARX788: A registry link was labelled ACE-Breast-02 although NCT04829604 is the separate ACE-Breast-03 study. (Second accuracy pass; the ClinicalTrials.gov record read through its API.) · The label is corrected, and the brain metastasis study is now described with its figures (response in the brain in 25 percent of 32 patients).
- 2026-11-17 · Trastuzumab cardiotoxicity: The rule for pausing treatment was given as a fall of 16 points or any value below normal. (Second accuracy pass; the US trastuzumab label read through openFDA.) · The label rule is a fall of 16 points or more, or a fall of 10 points or more to below the normal range.
- 2026-11-17 · HER2-positive breast cancer: The history said HER2 was amplified in 25 to 30 percent of breast cancers in the 1987 paper, a treatment row said the guideline update for trastuzumab deruxtecan was pending, and unsourced figures stood on linked drug pages (biosimilar uptake above 80 percent; margetuximab benefit in 158F carriers). (Second accuracy pass; the abstract of Slamon 1987 and of the final SOPHIA analysis read through PubMed; the NCCN guideline needs a login and could not be read.) · The 1987 figure is 30 percent of 189 tumours. The pending wording and the uptake figure were removed, and the margetuximab finding is now the 158FF genotype with the final survival result. Neratinib and pyrotinib summaries gained their overall survival results.
- 2026-11-17 · LiGeR-HN1: The trial was shown as recruiting with interim analyses expected in 2026 to 2027. Enrolment had closed at 699 patients, and on 28 October 2026 the sponsor announced that the first interim analysis met both primary endpoints, with a hazard ratio for death of 0.50 (95% CI 0.37 to 0.67) (Second accuracy pass; the ClinicalTrials.gov record for NCT06525220 and the raw text of the Genmab company announcement of 28 October 2026) · Status, enrolment, sponsor, summary and result now give the topline figures and say they come from a company announcement with full data unpublished. The petosemtamab drug page, the bispecific pairing and the head and neck summary, state of the art and history carry the same update
- 2026-11-17 · Ficerafusp alfa: The dose was given as 1500 mg every 2 weeks. The trial registry, the phase 1/1b paper and the company all give 1500 mg once a week. The paper was also dated 2025 and was published in 2026 (Second accuracy pass; the ClinicalTrials.gov record for NCT06788990, the abstract for DOI 10.1200/JCO-25-02027 read through Europe PMC with its date from Crossref, and a reprint of the Bicara release of 12 January 2026) · Dose corrected to once a week on the drug and on FORTIFI-HN01, the year corrected to 2026, and the response rate of 54 percent in 28 patients with HPV-negative disease added
- 2026-11-17 · KEYNOTE-689: The 682 patients of the whole PD-L1-positive population were shown as the size of the pembrolizumab arm; the arms were 347 and 335. The page also said this was the first new treatment in six years and the first approval since 2019, and that major pathological response predicted excellent outcomes; the cited release says none of these. It said no second phase 3 existed, although NIVOPOSTOP had reported (Second accuracy pass; the US Keytruda label read through openFDA, the Merck releases of 13 June and 29 October 2025, the abstract for DOI 10.1056/NEJMoa2415434 and the open review PMC12832937) · Arm sizes and the hazard ratio of 0.70 at the first interim analysis added, the unsupported claims removed, the European approval of 29 October 2025 added, and the replication line now reports NIVOPOSTOP
- 2026-11-17 · Immunotherapy before surgery rather than with chemoradiation: The page said immunotherapy works before surgery but not alongside chemoradiation. NIVOPOSTOP gave nivolumab with chemoradiation after surgery and improved three-year disease-free survival from 52.5 to 63.1 percent (hazard ratio 0.76). The head and neck page also said such combinations had failed three times without naming the trials (Second accuracy pass; the ASCO Post report of the NIVOPOSTOP presentation read as raw HTML, the abstract for DOI 10.1016/j.ctrv.2025.103018, and the abstracts for KEYNOTE-412 and NRG-HN004 through Europe PMC and PubMed) · The rule now says the failures are in unresected disease treated with chemoradiation, names the trials, and reports NIVOPOSTOP
- 2026-11-17 · Petosemtamab: The response rate for the drug alone was given as about 37 percent; the trial design paper and the company give 36 percent (27 of 75). The page said the registrational trials were enrolling (Second accuracy pass; the full text of DOI 10.1080/14796694.2025.2511470 through Europe PMC and the Merus update for the second quarter of 2025) · Figure corrected, both Breakthrough Therapy designations dated, and the trial status updated
- 2026-11-17 · Head and neck squamous cell carcinoma: The timeline put the Japanese approval of boron neutron capture therapy in 2021; it was 2020. The page said de-escalation had failed in every randomised trial, although the randomised E3311 trial of lower-dose radiotherapy after surgery reported 54-month progression-free survival above 90 percent in both arms. Monalizumab was listed in the pipeline after its phase 3 trial stopped for futility. A link labelled as the NCCN guideline led to a journal paper on ESMO benefit scores (Second accuracy pass; abstracts read through Europe PMC and PubMed for PMID 42624745, DOI 10.1200/JCO-24-02550, DOI 10.1158/1078-0432.CCR-25-0073 and PMC11163648) · The approval moved to 2020, the de-escalation statements narrowed to the three trials of less intensive chemoradiation, monalizumab removed from the pipeline, and the link relabelled
- 2026-11-17 · NRG-HN002 and NRG-HN005: The result gave only the control arm figure and said the trial was stopped. The registry reports two-year progression-free survival of 88.6 and 90.3 percent in the two lower-dose arms against 98.1 percent, and that the phase 3 part did not open (Second accuracy pass; the results section of the ClinicalTrials.gov record for NCT03952585) · All three figures given and the wording corrected
- 2026-11-17 · HERIZON-GEA-01: The record said zanidatamab significantly improved overall survival and that a supplemental licence application was planned for the first half of 2026. Overall survival was significantly longer only in the arm that added tislelizumab (26.4 against 19.2 months, hazard ratio 0.72); with zanidatamab and chemotherapy alone the difference was not significant at the interim analysis (hazard ratio 0.80, 95% CI 0.64 to 1.01). The FDA approved both regimens on 25 August 2026, the one without tislelizumab only for IHC 3+ tumours (Second accuracy pass; the abstract for DOI 10.1056/NEJMoa2517729 read through PubMed, and the Ziihera label and submission list read through openFDA) · The trial, the first-line care row, the state-of-the-art line and the history now give the arm-level survival result and the approval; the main cancer record no longer lists zanidatamab as a later-line treatment
- 2026-11-17 · FORTITUDE-101 and bemarituzumab: The trial and drug records said FORTITUDE-102 was testing bemarituzumab with nivolumab, that no confirmatory trial existed yet and that the regulatory path was uncertain in September 2026. FORTITUDE-102 was stopped for inadequate efficacy and its registry record is marked terminated by the sponsor. The updated FORTITUDE-101 analysis was described without figures (14.5 against 13.2 months, hazard ratio 0.82, 95% CI 0.62 to 1.08), and the share of tumours with any FGFR2b overexpression was given as about 30% where the prescreening studies found 36.5 to 37.8% (Second accuracy pass; the registry record NCT05111626 read through the ClinicalTrials.gov API, the full text for DOI 10.3322/caac.70103 read through Europe PMC, and the abstracts for DOI 10.1200/PO-24-00710 and 10.1016/j.esmoop.2026.107698) · The trial, the drug, the open problem and the subtype line now carry the updated figures and the stopped trial; the main cancer record no longer calls FGFR2b one of three new targets
- 2026-11-17 · Gastric cancer: US labels for first-line PD-1 inhibitors: CheckMate 649 and KEYNOTE-859 were described as FDA approved for all patients whatever the PD-L1 level, with the US and European labels in disagreement. The current US labels for nivolumab, pembrolizumab and tislelizumab all limit first-line use in stomach cancer to tumours expressing PD-L1 of 1 or more. The KEYNOTE-859 figures were also sourced to a 2024 conference update rather than the paper they come from, and CheckMate 649 was called the largest phase 3 in gastric cancer with no source (Second accuracy pass; the Opdivo, Keytruda and Tevimbra labels read through openFDA, the identifier resolved through Crossref, and the abstract for DOI 10.1016/S1470-2045(23)00515-6 read through PubMed) · Both trial summaries, the first-line care rows and the overview state the current US limit, dated to the day the labels were read; the KEYNOTE-859 source is the Lancet Oncology paper; the unsourced superlative is removed
- 2026-11-17 · CLARITY-Gastric 01: The result line said the co-primary endpoints including overall survival were met. Only the survival endpoint in third-line or later disease was met; the other dual primary endpoint, progression-free survival in all patients, did not reach statistical significance. The care row presented sonesitatug vedotin as a second-line option although it has no US label, and the ramucirumab note said the trial had beaten ramucirumab plus paclitaxel, which was one of several comparator choices (Second accuracy pass; the OncLive report of 27 July 2026 read as raw HTML, the registry record NCT06346392 read through the ClinicalTrials.gov API, and an openFDA label search) · The result, the summary, the care row and the ramucirumab note are corrected, and the record says the survival figures could not be read on 17 November 2026
- 2026-11-17 · Gastric cancer: satri-cel approval date: The history gave the Chinese approval of the CAR-T therapy satri-cel as 2025. It was approved in June 2026, after at least two earlier lines of treatment. The five-year CheckMate 649 results were also dated 2025 and were published in 2026 (Second accuracy pass; the full texts for DOI 10.3390/ph19091420 and 10.3389/fimmu.2026.1942381 read through Europe PMC, and the abstract for DOI 10.1016/j.annonc.2026.02.003 read through PubMed) · Both history lists place the approval in 2026, and the care row says it follows two earlier lines
- 2026-11-17 · DESTINY-Gastric04: The summary said the trial was reported at the ASCO 2025 plenary and that it confirmed accelerated approvals from DESTINY-Gastric01 and DESTINY-Gastric02. The ASCO release does not describe a plenary presentation, and the US label carries no accelerated approval wording for the stomach cancer indication. The hazard ratio was given without its interval (Second accuracy pass; the ASCO news release read from its Internet Archive copy, the Enhertu label read through openFDA, and the abstract for DOI 10.1056/NEJMoa2503119 read through PubMed) · The summary says annual meeting and compares with DESTINY-Gastric01 only; the interval 0.55 to 0.90, the P value and the 13.9% rate of lung inflammation are added
- 2026-11-17 · MATTERHORN: The event-free survival figure was sourced to the registry record, which holds no results, and the replication line called MATTERHORN the only positive perioperative immunotherapy trial, with no source, while describing KEYNOTE-585 without its final figures (Second accuracy pass; the Imfinzi label read through openFDA and the abstracts for DOI 10.1016/S0140-6736(26)01254-7 and 10.1200/JCO-25-00486 read through PubMed) · The event-free survival row cites the label and carries its interval, 0.58 to 0.86; the replication line gives the KEYNOTE-585 final analysis (event-free survival hazard ratio 0.81, overall survival 0.86, 95% CI 0.71 to 1.06) and drops the word only
- 2026-11-17 · SELECT: The record said lenvatinib quadrupled the time before the disease grew, and that survival was confounded by crossover without saying whether it differed. The medians were 18.3 and 3.6 months, about fivefold, and overall survival was not significantly different: hazard ratio 0.73, 95% CI 0.50 to 1.07, with 83% of the placebo group crossing over. (Second accuracy pass; the abstract read through PubMed and section 14.1 of the current US Lenvima label read through openFDA.) · The two medians replace "quadrupling", the survival result is stated as not significant with its interval, and the label is linked.
- 2026-11-17 · DECISION: The record gave the progression-free survival gain and said nothing about overall survival. Overall survival did not differ significantly: hazard ratio 0.80, 95% CI 0.54 to 1.19, with 71% of the placebo group crossing over. Fatigue was named as a common side effect; the abstract lists hand-foot skin reaction and diarrhoea. (Second accuracy pass; the full text of the Lancet paper read through NCBI PMC (PMC4366116).) · The survival result and crossover are added to the summary and result, and the side effects follow the abstract.
- 2026-11-17 · LIBRETTO-531: The record said selpercatinib roughly halved the rate of severe adverse events. The paper reports that dose reductions for adverse events halved (38.9% against 77.3%) and fewer people stopped treatment (4.7% against 26.8%); it gives no halving of severe events. The record also did not say the result came from an interim analysis with survival data immature. The vandetanib record quoted a severe-event comparison of 52.8% against 76.4% that neither the abstract nor the label contains. (Second accuracy pass; the abstract read through PubMed and section 14.2 of the current US Retevmo label read through openFDA.) · The tldr and summary give the dose-reduction and discontinuation figures, name the interim analysis and immature survival, and the vandetanib record uses the dose-reduction figure.
- 2026-11-17 · ARROW (thyroid cohorts): The tldr described ARROW as the study behind pralsetinib's thyroid approvals without saying one was withdrawn, and the summary gave only years. The US medullary thyroid cancer indication, approved 1 December 2020, was withdrawn on 20 July 2023; the confirmatory phase 3 trial NCT04760288 was withdrawn with no one enrolled. The current US label keeps only the RET fusion-positive thyroid cancer indication, still under accelerated approval. (Second accuracy pass; the FDA list of withdrawn accelerated approvals read through its Internet Archive copy, the Gavreto label through openFDA and the registry record through the ClinicalTrials.gov API.) · The tldr and summary state the withdrawal with dates, the surviving indication and its accelerated status, and the response rates carry their denominators (15 of 21, 33 of 55, 8 of 9).
- 2026-11-17 · ESTIMABL2: The population was given as pT1a to T1b; the trial enrolled multifocal pT1a or pT1b tumours. The record said the trial proved radioiodine could be skipped without any increase in recurrence; it showed non-inferiority on a composite of events (95.6% against 95.9% at three years, 93.2% against 94.8% at five). The replication line called IoN a pooled analysis; it is a separate randomised trial. ESMO guidance was credited without a source read. (Second accuracy pass; both ESTIMABL2 abstracts read through Europe PMC and PubMed, and the 2025 American Thyroid Association guideline read in full on NCBI PMC (PMC13090833).) · Population, wording and figures corrected, the five-year result added, the guidance sentence now cites Recommendation 32 of the 2025 guideline, and the IoN line is fixed.
- 2026-11-17 · IoN: The record said IoN randomised patients with pT1 to T2, node-negative tumours without adverse features. It enrolled pT1 to pT3 tumours with N0, Nx or N1a nodes; recurrences were commoner in pT3 (9% against 3%) and N1a (13% against 2%) disease, which is why the authors limit the conclusion to pT1 and pT2 tumours with N0 or Nx nodes. (Second accuracy pass; the Lancet abstract read through Europe PMC.) · Setting and summary give the real eligibility and the subgroup recurrence rates, and the five-year figures are exact (97.9% against 96.3%).
- 2026-11-17 · ROAR (anaplastic thyroid cancer cohort): Median overall survival was given as 15 months; the paper reports 14.5 months (95% CI 6.8 to 23.2). The record said a third of anaplastic cancers carry BRAF V600E; the paper cites 10% to 50%. The same 15 months appeared in the thyroid cancer state-of-the-art list. (Second accuracy pass; the Annals of Oncology update read in full through Europe PMC (PMC9338780).) · 14.5 months in the result, summary and state-of-the-art line, and the mutation share is given as the published range.
- 2026-11-17 · ASTRA: The record said ASTRA randomised 401 patients. It enrolled 400 and randomised 233, two to one (155 selumetinib, 78 placebo); the registry lists 233. (Second accuracy pass; the JCO abstract read through Europe PMC and the registry record through the ClinicalTrials.gov API.) · Enrolment is 233, and the summary and result give the counts and the odds ratio (1.07, 95% CI 0.61 to 1.87).
- 2026-11-17 · Thyroid cancer: guideline statements: The standard-of-care rows followed the 2015 American Thyroid Association guideline. The 2025 edition differs: lobectomy should be the first operation for cancers of 2 cm or less, long-term TSH suppression is no longer suggested for low or intermediate risk without recurrence, high-risk radioiodine activity is 3.7 to 5.55 GBq, lenvatinib is the preferred first drug for refractory disease with BRAF-targeted therapy after it, and cabozantinib is second line. The TSH suppression and low-risk terms and the radioactive iodine dose line had the same fault. (Second accuracy pass; the 2025 guideline (Ringel et al., Thyroid 2025) read in full on NCBI PMC, Recommendations 11, 15, 32, 46, 62, 66 to 70 and Table 10, and the Cabometyx label through openFDA.) · Three standard-of-care rows, the two terms, the active surveillance entry and the dose schedule now follow the 2025 text, with the guideline linked.
- 2026-11-17 · Thyroid cancer: overdiagnosis and survival figures: The page said incidence had tripled with no change in mortality, gave a 10-year survival of about 97% for papillary cancer with no source, and gave US five-year survival without diagnosis years. In US SEER data incidence rose from 4.56 to 14.42 per 100,000 between 1974 to 1977 and 2010 to 2013 while incidence-based mortality rose slightly, from 0.40 to 0.46. Five-year relative survival is 98.3% for 2016 to 2022 diagnoses. A history line put BRAF V600E at about 45% of papillary cancers in 2003; the 2003 paper found 28 of 78, 36%. (Second accuracy pass; the JAMA 2017 abstract (PMID 28362912) and the Cancer Research 2003 abstract (PMID 12670889) read through Europe PMC, and the SEER thyroid stat facts page read directly.) · The summary, open problem, overdiagnosis idea, burden line and history line carry the sourced figures with their years; the Korean fifteenfold rise is dated 1993 to 2011.
- 2026-11-17 · TERT promoter mutation: The term said TERT promoter mutations occur in most anaplastic thyroid cancers. The cited paper found them in 51% of anaplastic and poorly differentiated cancers taken together, and in 22.5% of papillary cancers. (Second accuracy pass; Landa et al. 2013 read in full on NCBI PMC (PMC3763971).) · The sentence gives the two figures from the paper.
- 2026-11-17 · Thyroid nodule FNA, Bethesda cytology and molecular classifiers: The entry and the thyroid cancer page said molecular classifiers have halved unnecessary surgery, and that redifferentiation restores iodine uptake in about half of refractory patients. The ThyroSeq v3 study estimated surgery could be spared in up to 61% of indeterminate nodules; in the first selumetinib study uptake rose in 12 of 20 patients and 8 reached the treatment threshold. (Second accuracy pass; the JAMA Surgery 2018, JAMA Oncology 2019 and NEJM 2013 abstracts read through Europe PMC.) · Both claims are reworded to the published figures.
- 2026-11-17 · EMERALD-1: The record gave an overall survival hazard ratio of 0.80 with an interval of 0.57 to 1.11 for EMERALD-1. Those are the interim overall survival figures of a different trial, LEAP-012. The EMERALD-1 paper says its interim overall survival analysis did not reach significance and does not report the figures (Second accuracy pass; the full text for DOI 10.1016/S0140-6736(24)02551-0 read from PubMed Central and the abstract for DOI 10.1016/S0140-6736(24)02575-3 read through Europe PMC) · The hazard ratio and interval were removed, the result line and summary now say the interim analysis was not significant and its figures were withheld, and the pairing record on TACE with systemic therapy no longer describes a trend
- 2026-11-17 · EMERALD-3: The record said 725 patients were randomised and that overall survival was immature and not yet known. The trial was published in September 2026: 760 patients were randomised, and overall survival at the February 2026 cutoff was 39.5 against 34.7 months, hazard ratio 0.84, 95% CI 0.65 to 1.09, not significant (Second accuracy pass; the abstract for DOI 10.1016/S1470-2045(26)00278-0 read through PubMed) · The summary, tldr, result and outcome rows now carry the published progression-free and overall survival figures with the paper as source, and the hub, the TACE technology record and the pairing record say that no TACE combination has shown longer survival, now counting TALENTACE as a fourth trial
- 2026-11-17 · CARES-310: Overall survival was given as 23.8 against 15.2 months with a hazard ratio of 0.62, sourced to the Lancet 2023 paper. The 2023 interim analysis was 22.1 against 15.2 months, hazard ratio 0.62; the final analysis was 23.8 against 15.2 months, hazard ratio 0.64, 95% CI 0.52 to 0.79 (Second accuracy pass; the abstracts for DOI 10.1016/S0140-6736(23)00961-3 and DOI 10.1016/S1470-2045(25)00543-1 read through PubMed) · The trial, drug and company records now give the final figures with the final paper as source and name the interim figures as interim
- 2026-11-17 · Camrelizumab + rivoceranib: The record said all three FDA complete response letters cited the rivoceranib manufacturing site, and showed nothing pending after July 2026. The first letter cited the camrelizumab plant and incomplete clinical-site inspections, the reasons for the second were not made public, and the application was resubmitted on 9 October 2026 with a target action date of 9 April 2027 (Second accuracy pass; the OncLive and Targeted Oncology reports and the Elevar announcements of 10 July, 14 October and 5 November 2026 read as raw HTML) · The drug, company, trial and hub text now describe each letter as reported and give the resubmission and the April 2027 date, with a new regulatory event and link
- 2026-11-17 · CheckMate 9DW: The tldr and summary called 23.7 months the longest median survival yet seen in a first-line liver cancer trial. The final analysis of CARES-310 reported 23.8 months (Second accuracy pass; the abstracts for DOI 10.1016/S0140-6736(25)00403-9 and DOI 10.1016/S1470-2045(25)00543-1 read through PubMed, and the US nivolumab label read through openFDA) · The claim was removed. The record now gives the medians, says the result is from a preplanned interim analysis, and adds the early excess of deaths and the 12 deaths attributed to treatment against three
- 2026-11-17 · IMbrave050: The interim hazard ratio of 0.72 carried an interval of 0.56 to 0.93, where the paper gives an adjusted interval of 0.53 to 0.98, and the summary said the update left a possible signal in subgroups. The updated analysis found a hazard ratio of 0.90, 95% CI 0.72 to 1.12, consistent across subgroups, with overall survival at 1.26, 95% CI 0.85 to 1.87 (Second accuracy pass; the abstracts for DOI 10.1016/S0140-6736(23)01796-8 and DOI 10.1016/j.jhep.2026.01.006 read through PubMed) · The interval was corrected, the subgroup remark removed, and the updated recurrence-free and overall survival figures added as outcome rows
- 2026-11-17 · HIMALAYA: The overall survival hazard ratio of 0.78 carried an interval of 0.66 to 0.92; the paper gives a 96.02% interval of 0.65 to 0.93. The five-year update was called the longest follow-up of any phase 3 in liver cancer, which the paper read does not say (Second accuracy pass; the abstracts for DOI 10.1056/EVIDoa2100070 and PubMed record 40222621 read through Europe PMC and PubMed) · The interval was corrected, the five-year hazard ratio now has its interval of 0.65 to 0.89, and the follow-up claim was replaced by the stated median follow-up
- 2026-11-17 · IMbrave150: Survival of 19.2 against 13.4 months, hazard ratio 0.66, and progression-free survival of 6.9 against 4.3 months were sourced to the 2020 primary paper. They come from the post hoc updated analysis published in 2022; the primary analysis hazard ratio for death was 0.58 (Second accuracy pass; the abstracts for DOI 10.1056/NEJMoa1915745 and DOI 10.1016/j.jhep.2021.11.030 read through PubMed) · The two outcome rows now cite the 2022 update and the result line names both analyses
- 2026-11-17 · Liver transplantation for cancer: The Milan criteria were said to yield about 70% five-year survival, credited to Mazzaferro 1996 in the liver cancer timeline. The paper reports 75% actuarial survival at four years in 48 patients (Second accuracy pass; the abstract for DOI 10.1056/NEJM199603143341104 read through PubMed) · The technology summary and the 1996 timeline note give the four-year figure and the paper is linked
- 2026-11-17 · Sirtex Medical: The record said SIR-Spheres were indicated in the United States only for colorectal liver metastases, with a liver cancer indication still to come. The FDA added unresectable hepatocellular carcinoma to the indication on 1 July 2025 (Second accuracy pass; the FDA premarket approval record P990065 supplement S014 read through openFDA) · The summary gives the July 2025 approval and links the record
- 2026-11-17 · HCC surveillance in cirrhosis: Surveillance was said to detect cancer at an early stage in about 60 to 70% of adherent patients, with no source. The meta-analysis found 70.9% of cancers detected by surveillance were early stage against 29.9% of those found otherwise (Second accuracy pass; the full text for DOI 10.1371/journal.pmed.1001624 read from PubMed Central and the abstract for DOI 10.1002/hep.31309 read through Europe PMC) · The summary gives both figures and the pooled 24% uptake with their sources; the hepatitis term now names the cohorts behind its two 70% figures
- 2026-11-17 · Hepatocellular carcinoma: The main record gave about 900,000 liver cancer cases a year, and the hub text said TACE combinations prolong progression-free survival by about 30%. GLOBOCAN 2024 gives 843,045 cases and 732,489 deaths, and the trials report hazard ratios of 0.77, 0.66 and 0.70, which are reductions in risk, not a 30% longer time (Second accuracy pass; the repository copy of GLOBOCAN 2024 and the trial abstracts read through PubMed) · The burden line names the edition and both counts, and the hub text gives the hazard ratios and adds the non-significant overall survival result from EMERALD-3
- 2026-11-17 · LEAP-012: The final overall survival hazard ratio of 0.98 had no interval, medians or source, and the record did not say the trial was stopped (Second accuracy pass; the results section of NCT04246177 read through the ClinicalTrials.gov API) · Medians of 40.4 against 40.8 months and the interval of 0.76 to 1.26 were added with the registry as source, with the adverse event rates and the termination for business reasons
- 2026-11-17 · EV-304 / KEYNOTE-B15: The record said a filing was expected in 2026 and that the regimen could replace chemotherapy before surgery in fit patients. The FDA approved enfortumab vedotin with pembrolizumab for all patients with muscle-invasive bladder cancer who are candidates for cystectomy on 10 July 2026. The main bladder cancer record, two standard-of-care rows, the pairing record and the timeline still showed the trial as positive and awaiting a label (Second accuracy pass; the FDA approval notice of 10 July 2026 and the Padcev label effective 4 August 2026 read as raw text from fda.gov and openFDA, with the supplement date from the openFDA drugsfda record) · Approval and its date written into the trial, the pairing, the main record and the spike patch, with the FDA notice linked
- 2026-11-17 · CheckMate 274: The record said adjuvant nivolumab extends life and showed an overall survival benefit durable at five years. The five-year hazard ratio of 0.83 has a 95% confidence interval of 0.67 to 1.02, which is not statistically significant, and the earlier interim analyses had not crossed the boundary for significance. The disease-free survival hazard ratio was given as 0.70 (0.57 to 0.85) against a source that states 0.71 (0.58 to 0.86) (Second accuracy pass; the abstract for DOI 10.1016/j.annonc.2025.09.139 read through PubMed and the full text for DOI 10.1200/JCO.24.00340 read through Europe PMC) · Summary, result line, outcomes and the state-of-the-art line now give the interval and say the survival difference is not statistically significant; disease-free survival figures corrected
- 2026-11-17 · POTOMAC: The record said 1,018 patients were randomised to durvalumab plus BCG or BCG alone and called the regimen the first systemic immunotherapy approved in early bladder cancer. The trial had three arms and the approved comparison was 339 against 340 patients. Pembrolizumab alone was already approved for BCG-unresponsive early disease (Second accuracy pass; the Imfinzi label and the FDA notice of 28 May 2026 read as raw text, the abstract for DOI 10.1016/S0140-6736(25)01897-5 read through PubMed, and the Keytruda label read through openFDA) · Arm sizes and the three-arm design stated; the first claim narrowed to disease that has not yet had BCG
- 2026-11-17 · BOND-003: Complete response was given as 75.2 percent in a cohort of 110, the 24-month figure of 41.8 percent was sourced to the Lancet Oncology paper, publication was dated 2025, and the licence application was expected to complete in the fourth quarter of 2026. The paper reports 83 of 110 (75.5 percent) with 115 enrolled and 112 treated, was published in August 2026, and does not give the 24-month figure in its abstract; that figure is from a company update. On 16 November 2026 the company moved completion of the application to the end of March 2027 and reported a positive topline result for the randomised PIVOT-006 trial (Second accuracy pass; the abstract for PMID 42508433 read through PubMed and the CG Oncology releases of 5 September 2025, 27 July 2026 and 16 November 2026 read as raw text from the company site) · Figures, publication year, source of the 24-month figure and filing status corrected in the trial, the cretostimogene drug record, the CG Oncology record, the intravesical therapy record and the spike patch
- 2026-11-17 · QUILT 3.032: The record gave a complete response rate of 71 percent in a cohort of 77 patients. The 71 percent is 58 of 82 patients in the published report; the US label counts 77 patients and gives 62 percent. A company figure for bladder preservation at three years could not be re-read (Second accuracy pass; the abstract for DOI 10.1056/EVIDoa2200167 read through PubMed and the Anktiva label read through openFDA) · Both figures stated with their denominators in the trial and drug records; the company figure replaced by the published 89.2 percent cystectomy avoidance at 24 months among responders
- 2026-11-17 · SunRISe-1: The record said 51 percent remained in complete response at one year, called SunRISe-4 an ongoing phase 3 trial, and gave the Inlexzo schedule as every 12 weeks up to 3 years. The label says 51 percent of responders had a response lasting at least 12 months and sets 8 doses over 6 months then 6 doses every 12 weeks; SunRISe-4 is a completed phase 2 trial (Second accuracy pass; the Inlexzo label read through openFDA, the abstract for DOI 10.1200/JCO-25-01651 read through PubMed, and the registry record NCT04919512 read through the ClinicalTrials.gov API) · Durability wording, dosing schedule and trial phases corrected in the trial and TAR-200 drug records
- 2026-11-17 · EV-303 / KEYNOTE-905: The event-free survival interval was given as 0.28 to 0.56 and the result was called the largest effect yet seen in a perioperative bladder trial. The paper and the label give 0.28 to 0.57, and no source read makes the comparison. The population was described as unable to have cisplatin; it also included patients who declined it (Second accuracy pass; the abstract for DOI 10.1056/NEJMoa2511674 read through PubMed and the Padcev label read through openFDA) · Interval corrected and sourced to the paper, the superlative removed, the population reworded and the approval date of 21 November 2025 added
- 2026-11-17 · NIAGARA: The record said the trial raised the pathologic complete response rate and was the first to show a survival gain from immunotherapy around bladder surgery. Neither the abstract nor the US label reports a pathologic complete response result, and no source read supports the first claim (Second accuracy pass; the abstract for DOI 10.1056/NEJMoa2408154 read through Europe PMC, the Imfinzi label read through openFDA and the FDA notice of 28 March 2025 read as raw text) · The response claim and both first claims removed; overall survival marked as a pre-specified interim analysis
- 2026-11-17 · Erdafitinib: Dosing said to increase to 9 mg if serum phosphate was below 5.5 mg/dL at day 14 and to hold above 7 mg/dL. The current label increases the dose if phosphate is below 9.0 mg/dL at 14 to 21 days, starts a phosphate binder at 7 to 8.99 mg/dL and withholds at 9 mg/dL or above (Second accuracy pass; the Balversa label effective 24 October 2025 read through openFDA) · Dosing, modification and monitoring lines rewritten from the label
- 2026-11-17 · JAVELIN Bladder 100: The primary overall survival result was sourced to the two-year follow-up paper, which reports a different hazard ratio (0.76) (Second accuracy pass; the abstracts for DOI 10.1056/NEJMoa2002788 and DOI 10.1200/JCO.22.01792 read through PubMed) · Source changed to the primary New England Journal of Medicine report
- 2026-11-17 · Bladder & urothelial cancer: The page gave about 600,000 cases a year, ranked bladder cancer eighth worldwide and fourth in US men, called the atezolizumab approval the first blood-test-guided drug approval in any cancer, counted five bladder-sparing options for BCG-unresponsive disease, and listed nivolumab with platinum chemotherapy. GLOBOCAN 2024 gives 635,264 cases and ranks bladder ninth (eighth without non-melanoma skin cancer); the maker describes the approval as the first of a ctDNA-guided therapy; four options are approved and cretostimogene is not; the nivolumab indication is with cisplatin and gemcitabine. The page also lacked the 13 November 2026 approval of durvalumab with enfortumab vedotin (Second accuracy pass; the repository copy of GLOBOCAN 2024, the SEER bladder page, the FDA notices of 15 May and 13 November 2026, the Genentech release of 15 May 2026 and the Opdivo label read as raw text) · Burden, ranking, first claim, option count and regimen wording corrected; the US rank replaced by the SEER figure; the new approval added with its FDA notice
- 2026-11-17 · Bladder preservation for MIBC after perioperative EV + pembrolizumab complete response: SunRISe-4 and MODERN were named as trials testing omission of cystectomy. SunRISe-4 gave treatment before a planned cystectomy and MODERN guides immunotherapy after surgery (Second accuracy pass; the registry records NCT04919512 and NCT05987241 read through the ClinicalTrials.gov API) · The two trials removed as examples, with the reason stated; response rates written in from the trial papers
- 2026-11-17 · EMPOWER-Cervical 1 / GOG-3016 / ENGOT-cx9: The record said the FDA issued a complete response letter in January 2022 over inspection issues and that the drug was declined by the FDA. The sponsors voluntarily withdrew the US application in January 2022 after discussions with the FDA about post-marketing studies. The record also gave only the first survival analysis (Second accuracy pass; two review articles (PMC10697630 and PMC12984288) read as full text from NCBI PMC, the current US cemiplimab label read through openFDA, the EMA product page, and the final analysis abstract for DOI 10.1016/j.ejca.2024.115146 read through PubMed) · The summary and one-line summary now say the application was withdrawn by its sponsors and that the US label has no cervical cancer indication. The final analysis is added (median survival 11.7 against 8.5 months, hazard ratio 0.67, 0.56 to 0.80) and both sources are linked
- 2026-11-17 · KEYNOTE-A18 / ENGOT-cx11 / GOG-3047: The overall survival result (36-month survival 82.6% against 74.8%, hazard ratio 0.67) was given as the trial result without saying it came from the second interim analysis, and no later data were shown. The final analysis posted on the registry gives a hazard ratio for death of 0.73 (0.57 to 0.94) and 36-month survival of 81.8% against 74.4% (Second accuracy pass; the abstract for PMID 39288779 read through PubMed and the results section of NCT04221945 read through the ClinicalTrials.gov API) · The summary, result line and outcome row now mark the published figures as the second interim analysis and give the final registry figures. The statement that CALLA differed in risk population was replaced, since its eligibility was similar
- 2026-11-17 · BEATcc / ENGOT-Cx10 / GOG-3030: Median survival of 32.1 against 22.8 months was given as the final result; the paper reports it as an interim overall survival analysis. The record also said the benefit held irrespective of PD-L1 status and was the longest median survival reported in this setting, neither of which is in the abstract, and that approval differs by region (Second accuracy pass; the abstract for DOI 10.1016/S0140-6736(23)02405-4 read through PubMed, the current US atezolizumab label read through openFDA and the EMA product page) · The summary, result line and outcome row say interim analysis. The two unsupported claims are removed. The record and the first-line row on the cervical cancer page now say atezolizumab has no cervical cancer indication on the US or EU label
- 2026-11-17 · OUTBACK / ANZGOG 0902 / GOG-0274: The hazard ratio for death was given as 0.91 (0.70 to 1.18). The paper reports 0.90 (0.70 to 1.17) (Second accuracy pass; the abstract and full text for DOI 10.1016/S1470-2045(23)00147-X (PMC11075114) read through NCBI PMC) · The outcome row now gives 0.90 (0.70 to 1.17)
- 2026-11-17 · LACC (Laparoscopic Approach to Cervical Cancer): The result quoted a hazard ratio for death of 6.00 from the first report and said only that the final analysis confirmed inferiority. The final analysis gives 4.5-year overall survival of 90.6% against 96.2% and a hazard ratio for death of 2.71 (1.32 to 5.59) (Second accuracy pass; the abstracts for DOI 10.1056/NEJMoa1806395 and DOI 10.1200/JCO.23.02335 read through PubMed) · The summary and result line give both the first and final figures, and the final paper is linked
- 2026-11-17 · SENTICOL III: The trial was described as recruiting about 950 patients and sponsored by ARCAGY-GINECO and ENGOT. The registry shows 989 patients enrolled, recruitment closed, and the University Hospital of Besançon as lead sponsor with the two groups as collaborators. The SENTIX trial it cites has since published its final outcome (Second accuracy pass; NCT03386734 read through the ClinicalTrials.gov API and the SENTIX abstract (PMID 40615729) read through PubMed) · Sponsor, enrolment and status wording corrected, the registry completion date given in place of an expected readout, and the SENTIX result added (two-year recurrence 6.1% against a 7% reference rate)
- 2026-11-17 · KEN SHE (single-dose HPV vaccine): The sponsor was given as the University of Washington; the registry lists Massachusetts General Hospital as lead sponsor. The record said more than 60 countries had adopted a single dose by 2025; WHO counts 93 countries using a one-dose schedule by the end of 2025 (Second accuracy pass; NCT03675256 read through the ClinicalTrials.gov API and the WHO immunization coverage fact sheet of 15 July 2026 read as raw HTML) · Sponsor corrected and the country count updated, here and in the 2025 row of the cervical cancer timeline
- 2026-11-17 · Nonavalent HPV vaccine: The record said Swedish and Scottish data show about 90% and near-100% reductions in women vaccinated at 12 to 13. The Swedish figure (88% lower rate) is for vaccination before age 17 with the quadrivalent vaccine; the Scottish finding of no invasive cancers is for women vaccinated at 12 or 13 with the bivalent vaccine, born 1988 to 1996. Global coverage was given as about 27% in 2023, which is out of date (Second accuracy pass; the abstracts for DOI 10.1056/NEJMoa1917338 and DOI 10.1093/jnci/djad263 read through PubMed and the WHO immunization coverage fact sheet of 15 July 2026) · The sentence now gives each study its own age group, vaccine and cohort. Coverage reads 33% for a first dose in girls in 2025, here and in the single-dose elimination idea, with the WHO page linked
- 2026-11-17 · PD-1 blockade + chemoradiation (locally advanced cervical cancer): The record said CALLA tested durvalumab in a broader, lower-risk population. CALLA enrolled FIGO 2009 stage IB2 to IIB node-positive disease or stage III and above, much the same high-risk group as KEYNOTE-A18 (Second accuracy pass; the abstract for DOI 10.1016/S1470-2045(23)00479-5 read through PubMed) · The record now describes the CALLA population correctly, gives its result (hazard ratio for progression or death 0.84, 0.65 to 1.08) and links the paper
- 2026-11-17 · Cervical cancer: share of the burden in poorer countries: Three passages said 90%, or nine in ten, of cases and deaths in 2024 were in low- and middle-income countries. Neither GLOBOCAN 2024 nor the WHO fact sheet gives a share; both say the highest rates are in those countries. The timeline also said Scotland reported zero cancers in a fully vaccinated cohort, where the study found none in women vaccinated at 12 or 13 (Second accuracy pass; the full text for DOI 10.3322/caac.70090 read through NCBI PMC, the WHO cervical cancer fact sheet of 10 November 2026 read as raw HTML, and the abstract for DOI 10.1093/jnci/djad263) · The burden line, the long summary and the main record now say the highest rates are in low- and middle-income countries, and the Scottish sentences name the age at vaccination and the birth cohort
- 2026-11-17 · KEYNOTE-826: The one-line summary said the regimen added nearly ten months of life. The source reports a difference in median survival (26.4 against 16.8 months at the final analysis), which is not the same as ten months for each woman (Second accuracy pass; the abstract for DOI 10.1200/JCO.23.00914 read through Europe PMC) · The one-line summary and summary now say median survival was extended by nearly ten months and give both medians
- 2026-11-17 · HARMONi-3: The trial outcomes table, the lung treatment row on antibody-drug conjugates and bispecifics, the plain-language line and eight translations said an interim analysis in May 2026 had missed statistical significance for progression-free survival; the HARMONi-2 entry said the global interim had not reproduced the margin; enrolment was given as 1,080 (Second accuracy pass; Summit's reports for the first and second quarters of 2026 read as raw text from their Internet Archive copies, and the ClinicalTrials.gov record for NCT05899608) · All now say no efficacy result has been reported: an early interim look at the squamous cohort was seen only by the monitoring committee, which recommended continuing. Enrolment follows the registry's estimated 1,600
- 2026-11-17 · ROS1-positive non-small-cell lung cancer: The row for disease after crizotinib or entrectinib said zidesamtinib was in trials, and the drug sat in the pipeline list (Second accuracy pass; the Jideytro label and the Drugs@FDA record for NDA 220185 read through openFDA) · The row gives the US approval of 22 July 2026 for use after a prior ROS1 inhibitor, a history entry is added, and zidesamtinib is out of the pipeline lists here and on the lung cancer record
- 2026-11-17 · HER2-mutant non-small-cell lung cancer: The summary said first-line treatment was still chemotherapy with immunotherapy and that sevabertinib had merely followed zongertinib; the first-line row kept both pills to trials (Second accuracy pass; the Hyrnuo and Hernexeos labels and the Drugs@FDA records for NDA 219972 and NDA 219042 read through openFDA) · The summary gives sevabertinib's accelerated approval of 19 November 2025 with the label's response rates and the 2026 widening of both labels to untreated patients (26 February and 9 September); the first-line row lists both pills
- 2026-11-17 · Non-small-cell lung cancer: The pipeline list still held patritumab deruxtecan, whose US application was withdrawn, and a source comment gave US screening uptake as 18.2 percent in 2022 (Second accuracy pass; the Merck release of 29 May 2025 and the American Lung Association State of Lung Cancer page (updated 10 November 2026) read as raw text) · Patritumab deruxtecan is removed from the pipeline lists of this record and of the EGFR-mutant subtype; the comment gives 24.4 percent, the figure the page now states
- 2026-11-17 · DeLLphi-305: The trial outcomes table, the DeLLphi-304 entry, the open question on DLL3, a news item, the plain-language line and eight translations still described the result as undisclosed or the trial as still testing its question (Second accuracy pass; the abstract for PubMed 42929702 (DOI 10.1056/NEJMoa2609482) read through PubMed, and the DOI resolved through Crossref) · All carry the published interim result: median overall survival 21.1 against 12.7 months, hazard ratio 0.42 (95% CI 0.32 to 0.54), and progression-free survival 7.4 against 1.9 months. The response-rate row says the abstract gives no figure
- 2026-11-17 · Small-cell lung cancer: The history dated topotecan's approval for relapsed small-cell lung cancer to 1996, and translations of the summary said antibody-drug conjugates were already moving outcomes (Second accuracy pass; the Drugs@FDA record for NDA 020671 read through openFDA and the full text of PMC9330463) · The year is 1998, the date of the efficacy supplement; 1996 was the first approval, for another cancer. The eight translations now follow the corrected English summary
- 2026-11-17 · Serplulimab: The plain-language line and eight translations of this record and of ASTRUM-005 claimed the largest survival gain, or the longest first-line survival, of any chemo-immunotherapy trial, with no source (Second accuracy pass; the ASTRUM-005 abstract (DOI 10.1001/jama.2022.16464) read through Europe PMC) · The lines give the figures instead: median survival 15.4 against 10.9 months
- 2026-11-17 · Early HER2-positive breast cancer: PHERGain's three-year invasive disease-free survival was given as 95.4 percent, and APHINITY was described without its survival result (Second accuracy pass; the abstract for PubMed 38582092 and the Roche media release of 13 May 2025 read as raw text) · PHERGain reads 94.8 percent. APHINITY's final analysis is added and attributed to the release: ten-year survival 91.6 against 89.8 percent, hazard ratio 0.83 (95% CI 0.69 to 1.00), no benefit in node-negative disease
- 2026-11-17 · Tucatinib: The record carried the October 2026 maintenance approval but not the boxed warning added with it (Second accuracy pass; the Tukysa label effective 14 October 2026 and the Drugs@FDA record for NDA 213411 read through openFDA) · The summary now states the boxed warning for severe and fatal liver injury and the monitoring the label asks for
- 2026-11-17 · Trastuzumab deruxtecan: The approvals list lacked first-line use with pertuzumab and gave no date for the early-stage approvals (Second accuracy pass; the Enhertu label effective 15 May 2026 and the Drugs@FDA record for BLA 761139 read through openFDA) · The first-line approval of 15 December 2025 is added and the early-stage entry is dated 15 May 2026
- 2026-11-17 · Trastuzumab brengitecan: Its only link went to DESTINY-Breast05, a trial of a different drug; the pyrotinib record linked an FDA label search although the drug has no US approval (Second accuracy pass; the ClinicalTrials.gov record for NCT06830889 and the Crossref record for DOI 10.1016/S1470-2045(20)30702-6) · The link now goes to the drug's own phase 3 trial, and pyrotinib links the PHOEBE paper
- 2026-11-17 · Zanidatamab: The record said the drug was approved only for bile duct cancer and that HERIZON-GEA-01 was positive in 2025; it had no gastro-oesophageal approval (Second accuracy pass; the Ziihera label effective 24 August 2026 and the Drugs@FDA record for BLA 761416 read through openFDA, and the abstract for PubMed 42202319) · The US approval of 25 August 2026 is added with its two regimens and the boxed warning for diarrhoea; the trial's figures are given, including that survival was significantly longer only with tislelizumab added. The plain-language line, eight translations and the regional table follow
- 2026-11-17 · Satricabtagene autoleucel: The China approval was dated 2025, in one place September 2025 (Second accuracy pass; the full text of PMC13610619 and PMC13678447 read through Europe PMC) · The approval is dated 22 June 2026 in the drug record, the regional table, the regulatory timeline and the cell therapy roadmap
- 2026-11-17 · FGFR2: The target record and the regional table called FORTITUDE-101 positive on overall survival and said a US application was expected (Second accuracy pass; the full text of PMC13584074 (DOI 10.3322/caac.70103) and the ClinicalTrials.gov record for NCT05111626) · Both say the early survival gain faded with follow-up (hazard ratio 0.82, not significant) and that FORTITUDE-102 was stopped; bemarituzumab and FORTITUDE-101 leave the gastric pipeline list
- 2026-11-17 · LiGeR-HN1: Plain-language lines and translations for the trial, the petosemtamab pairing, the perioperative immunotherapy pairing and KEYNOTE-689 still gave the old claims: no result, immunotherapy failing with all chemoradiation, and a first new treatment in six years. The recurrent disease subtype said the phase 3 was under way (Second accuracy pass; the Genmab announcement of 28 October 2026, the Merck release of 13 June 2025 and the abstract for DOI 10.1016/j.ctrv.2025.103018 read as raw text) · The lines report the interim survival result as a company announcement not yet published, limit the failures to chemoradiation given instead of surgery, and drop the unsourced first claim
- 2026-11-17 · INTERLINK-1: The trial was shown as active with no result, and monalizumab as a phase 3 drug for head and neck cancer with no efficacy results (Second accuracy pass; the abstract for DOI 10.1158/1078-0432.CCR-25-0073 read through Europe PMC, and the ClinicalTrials.gov records for NCT04590963 and NCT05221840) · The trial is marked negative: stopped for futility, median survival 8.8 against 8.6 months, hazard ratio 1.00 (95% CI 0.66 to 1.54). The drug record, the plain-language lines and the translations say so, and monalizumab leaves the head and neck pipeline list
- 2026-11-17 · VOLGA: The trial was shown as active although the FDA approved the regimen on 13 November 2026 (Second accuracy pass; the FDA approval notice of 13 November 2026 read as raw text) · Status is positive with the notice's figures: event-free survival hazard ratio 0.51 (95% CI 0.38 to 0.67) and overall survival hazard ratio 0.58 (95% CI 0.40 to 0.83) against cystectomy alone
- 2026-11-17 · PIVOT-006: The trial was shown as active with no result (Second accuracy pass; the CG Oncology announcement of 16 November 2026 read as raw text) · The record reports the topline result, recurrence hazard ratio 0.32 (95% CI 0.23 to 0.45) at a median of 17 months, and says it rests on a company announcement that is not yet peer reviewed
- 2026-11-17 · Muscle-invasive and advanced bladder cancer: The summary and treatment rows omitted three approved regimens around cystectomy and implied adjuvant nivolumab had a proven survival gain; the drug records for enfortumab vedotin, pembrolizumab and durvalumab lacked the approvals (Second accuracy pass; the FDA notices of 28 March 2025, 21 November 2025, 28 May 2026, 10 July 2026 and 13 November 2026, the Drugs@FDA records read through openFDA, and the abstract for DOI 10.1016/j.annonc.2025.09.139) · The approvals are added with their dates and populations, a row for people who cannot have cisplatin is added, and the summary says the five-year survival difference with nivolumab was not significant (hazard ratio 0.83, 95% CI 0.67 to 1.02). EV-304 and TAR-200, both approved, leave the bladder pipeline list
- 2026-11-17 · Diagnostics roadmap: Two roadmaps, two regulatory timelines and a plain-language line stated as fact that IMvigor011 led to the first ctDNA-guided drug approval (Second accuracy pass; the Genentech release of 15 May 2026 read as raw text. The FDA notice makes no such claim) · Each roadmap and timeline sentence now attributes the claim to the drug's maker, and the plain-language line no longer makes it
- 2026-11-17 · COSMIC-311: The record said the trial was active with no results recorded (Second accuracy pass; section 14.3 of the Cabometyx label read through openFDA and the FDA notice of 17 September 2021) · Status is positive with the label's figures: hazard ratio 0.22 at the primary analysis of 187 patients and median progression-free survival of 11.0 against 1.9 months in the updated analysis of 258; the response-rate endpoint was not met
- 2026-11-17 · Selpercatinib: The summary said only that a July 2026 FDA action was listed among that month's approvals (Second accuracy pass; the Retevmo label effective 11 September 2026 and the Drugs@FDA record for NDA 213246 read through openFDA) · The sentence says what the record shows: an efficacy supplement of 14 July 2026 revised the tumour-agnostic indication, and no indication on the label is marked accelerated
- 2026-11-17 · Elevar Therapeutics: The plain-language line and translations said the FDA had refused the combination three times on manufacturing grounds, and the regional table showed nothing pending (Second accuracy pass; the Elevar release dated 4 November 2026 and the OncLive report of 13 July 2026 read as raw text) · The lines say three refusals, two on manufacturing inspection findings, with the reason for the second not made public; the application was resubmitted on 9 October 2026 with a target date of 9 April 2027
- 2026-11-17 · Intermediate hepatocellular carcinoma (BCLC B): The record said survival was immature in EMERALD-1 and LEAP-012, described EMERALD-3 as durvalumab and tremelimumab with chemoembolisation reporting a benefit, and offered the combinations where approved (Second accuracy pass; the posted results for NCT04246177, the abstract for DOI 10.1016/S1470-2045(26)00278-0, the full text of PMC12282661 and the Imfinzi and Keytruda labels read through openFDA) · LEAP-012 found no survival difference (hazard ratio 0.98) and was terminated; EMERALD-3 added lenvatinib, delayed progression (hazard ratio 0.70) and showed no significant survival difference (hazard ratio 0.84); EMERALD-1 has not reported survival. The row says no US label carries these combinations
- 2026-11-17 · Early hepatocellular carcinoma (BCLC 0 and A): The history dated the IMbrave050 update to 2025 and the trial stayed in the pipeline list (Second accuracy pass; the abstract for DOI 10.1016/j.jhep.2026.01.006 read through Europe PMC) · The update is dated 2026 and the trial, whose benefit was not sustained, is removed from the pipeline list
- 2026-11-17 · IMbrave150: A response rate of 30 against 11 percent was sourced to the 2020 paper, which does not give it (Second accuracy pass; Table 35 of the Tecentriq label read through openFDA; the later report's full text could not be read) · The row gives the label's confirmed response rate at the primary analysis, 28 against 12 percent, with the label as its source
- 2026-11-17 · PD-1 blockade + chemoradiation (locally advanced cervical cancer): The plain-language line and translations said durvalumab was tested in a broader population, and those for EMPOWER-Cervical 1 said the US regulator had refused cemiplimab; the SENTICOL III line gave a results date the registry does not (Second accuracy pass; the CALLA abstract (PubMed 38039991), the full text of PMC10697630 and the ClinicalTrials.gov record for NCT03386734) · The lines say a similar high-risk group, that the makers withdrew the US application, and that enrolment is complete with results not yet reported
- 2026-11-17 · Non-Hodgkin lymphoma (all types): share from B cells: The summary line said about 95 percent of non-Hodgkin lymphoma comes from B cells. That figure, 94.7 percent, is the share of all lymphoma in the UK series, which counts Hodgkin lymphoma on the B-cell side. Among the 4,961 non-Hodgkin lymphomas in the series 308 were T-cell, which leaves 93.8 percent B-cell. The B-cell survival figure of 68.8 percent was also given without saying that group includes Hodgkin lymphoma (Second accuracy pass; the full text for DOI 10.1038/bjc.2015.94 read through Europe PMC) · The summary line now says about 94 percent, and the passages that quote the series say it covers diagnoses from 2004 to 2012 and counts Hodgkin lymphoma with the B-cell lymphomas
- 2026-11-17 · Non-Hodgkin lymphoma (all types): staging, bulk and the E suffix: Bulky disease was defined for every lymphoma as a mass of 10 cm or more or over a third of the chest width. The Lugano classification keeps that definition for Hodgkin lymphoma only and gives no validated size for non-Hodgkin lymphoma. The suffix E was described as a site reached by direct extension, leaving out a single extranodal site with no nodes involved (Second accuracy pass; the full text for DOI 10.1200/JCO.2013.54.8800 read as raw HTML from PubMed Central) · The staging line gives the Hodgkin definition, the suggested sizes for follicular and diffuse large B-cell lymphoma with the note that none is validated, and both meanings of E. The glossary entry on nodal and extranodal lymphoma carries the same wording
- 2026-11-17 · Non-Hodgkin lymphoma (all types): what causes death in follicular lymphoma: Three passages said transformation is the commonest cause of death in follicular lymphoma, with no source. The pooled French and US analysis found lymphoma to be the leading cause of death, and 77 of the 140 deaths from lymphoma, out of 283 deaths in all, were in people whose disease had transformed (Second accuracy pass; the abstract for DOI 10.1200/JCO.18.00400 read through Europe PMC) · The hub and the glossary entries on transformation and on indolent and aggressive lymphoma now give the 77 of 140 figure and link the paper
- 2026-11-17 · Non-Hodgkin lymphoma (all types): bispecific antibodies and ECHELON-2: Bispecific antibodies were said to have arrived in 2023, although mosunetuzumab, named in the same record, was approved in the United States on 22 December 2022. Brentuximab vedotin with chemotherapy was called the only randomised first-line advance in twenty years for T-cell lymphoma, with no source for only (Second accuracy pass; openFDA approval records for mosunetuzumab, epcoritamab and glofitamab, and the abstract for DOI 10.1016/S0140-6736(18)32984-2 read through Europe PMC) · The summary now says 2022 and 2023 with the three approval dates. The ECHELON-2 sentences give its result (hazard ratio 0.71, 0.54 to 0.93, 452 patients) without the unsourced claim, and the paper is linked
- 2026-11-17 · Non-Hodgkin lymphoma (all types): UK figures and the NCI summary: The UK case, death and ten-year survival figures carried no years. The link labelled NCI PDQ adult non-Hodgkin lymphoma treatment now opens the indolent B-cell summary, because NCI has split the adult summary by subtype (Second accuracy pass; Cancer Research UK's statistics page and the NCI page read as raw HTML) · Cases are dated 2019 and 2021 to 2022, deaths 2022 to 2024 and the survival figure 2018. The link label and the guideline stamp now say where the address leads
- 2026-11-17 · The two lymphoma classifications of 2022 (WHO-HAEM5 and ICC): The WHO hierarchy was given as class, family and entity. The paper gives category, family or class, entity or type, and subtype, and says entities are generally ordered from more indolent to more aggressive (Second accuracy pass; the full text for DOI 10.1038/s41375-022-01620-2 read through Europe PMC) · The entry and the hub summary now use the paper's levels and the word generally
- 2026-11-17 · Staging a lymphoma of the stomach or bowel: The Paris category T2 was given as muscularis propria alone; the published table says muscularis propria or subserosa. The 1994 gastrointestinal staging system was cited only to the 2014 Lugano paper, which does not contain it, and the ultrasound accuracy figure was given without the size of the series (Second accuracy pass; the Paris staging letter (DOI 10.1136/gut.52.6.912) read as raw HTML from PubMed Central and the abstracts for PubMed 8075046 and 8287809) · T2 is corrected, the 1994 workshop report and the Paris letter are linked, the series is given as 24 patients, and the entry says the 1994 stage definitions could not be re-read because the abstract does not print them
- 2026-11-17 · Hodgkin lymphoma: taxonomy note: The note said both 2022 classifications state that the subtypes of classic Hodgkin lymphoma have lost most of their prognostic meaning. Only the WHO paper says so; the International Consensus Classification says only that the subtypes are unchanged (Second accuracy pass; the full texts for DOI 10.1038/s41375-022-01620-2 and DOI 10.1182/blood.2022015851) · The statement is now credited to the WHO classification alone
- 2026-11-17 · Gastric MALT lymphoma: The long-term follow-up study behind the antibiotic figures was credited to Fischbach, who is not among its authors. The page also said this was the one lymphoma often cured by a fortnight of antibiotics, that the 1993 Lancet report was the first time an antibiotic had been shown to treat a cancer, and that surgery has essentially no role at any stage. (Second accuracy pass; the abstracts for DOI 10.1200/JCO.2005.02.3903 and DOI 10.1016/0140-6736(93)91409-f read through Europe PMC, and the NCI indolent B-cell lymphoma summary read as raw text.) · The study is credited to Wündisch and colleagues. The summary line now says antibiotics often bring a lasting remission in early disease. The 1993 entry gives what the report states: six patients treated, and no lymphoma on repeat biopsy in five. Surgery is described as not the first treatment, with the NCI summary listing gastrectomy among options when the disease progresses. The translocation figure now says 65 patients were tested and the ultrasound accuracy figure says it came from 24 patients in 1993.
- 2026-11-17 · EBV-positive diffuse large B-cell lymphoma: The page said the 2022 WHO classification dropped the restriction to older people. It also said a European study recommended testing every new case in people over 50, and that the frequency in European and North American populations is several times lower than in Asia. (Second accuracy pass; the WHO fifth edition paper (DOI 10.1038/s41375-022-01620-2) read in full text through Europe PMC, and the abstracts for DOI 10.1016/j.humpath.2009.07.024 and DOI 10.1038/modpathol.2011.62 read through Europe PMC.) · The WHO table shows the previous edition already used the name without the age restriction, and 2022 only shortened it. The study's authors said testing should be considered and left open whether routine testing can be recommended. The frequency is now given as the source states it: 1 to 3 per cent in Western populations.
- 2026-11-17 · Primary large B-cell lymphoma of the testis: The page said survival did not improve after rituximab came into use, while quoting, without a source, a later registry study whose authors report that it did. It also said relapse is most often in the central nervous system. (Second accuracy pass; the abstracts for DOI 10.1200/JCO.2009.22.5896 and DOI 10.1200/JCO.2010.31.4187 read through Europe PMC, and the later registry study found and read on PubMed (PMID 30867741).) · The later study (DOI 10.3892/ol.2019.9953) is now linked, and the page says the two registry studies disagree about rituximab. Relapse in the central nervous system is described as characteristic, with the trial's own count: three of ten relapses.
- 2026-11-17 · T-cell/histiocyte-rich large B-cell lymphoma: The page said the last of the six growth patterns the WHO lists for a related Hodgkin lymphoma, pattern E, resembles this disease. Pattern E is the fifth of six. (Second accuracy pass; Table 6 of the WHO fifth edition paper (DOI 10.1038/s41375-022-01620-2) read in full text through Europe PMC.) · The page now says one of the six patterns, pattern E, and the history entry says one of its two diffuse patterns.
- 2026-11-17 · High-grade B-cell lymphoma with MYC and BCL2 rearrangements (double-hit lymphoma): The page said the cohort in the gene signature study was assembled to contain the cases, called the test that reproduced the result a commercial assay, credited a WHO description to both classifications, and stated intensified treatment as the standard. (Second accuracy pass; the paper for DOI 10.1200/JCO.18.01583 read on its PubMed Central page, the WHO paper read in full text through Europe PMC, and the NCI aggressive B-cell lymphoma summary read as raw text.) · The cohort is described as the paper describes it: people treated with standard immunochemotherapy, drawn from a population-based registry. The assay is called new, as in the paper. The description is credited to the WHO classification alone. Treatment is described as often more intensive, with the NCI wording that such regimens are being explored.
- 2026-11-17 · Ocular adnexal MALT lymphoma: The page said three-quarters of lymphomas around the eye are MALT lymphomas. The series it cites found 67.9 per cent and says that share is higher than in Western countries. (Second accuracy pass; the abstract for DOI 10.1038/s41433-020-01198-y read through Europe PMC.) · The page now gives two-thirds, names the Taiwanese series, and says the share is lower in Western countries by the authors' account.
- 2026-11-17 · Primary effusion lymphoma: The page said the look-alike lymphoma that arises with fluid overload behaves considerably better. The WHO text says only that its prognosis appears fairly favourable. The treatment rows of this and six other aggressive lymphoma pages also cited an NCI summary address that now opens the summary for slow-growing lymphomas. (Second accuracy pass; the WHO fifth edition paper read in full text through Europe PMC, and the NCI address followed with a plain request on 17 November 2026.) · The sentence now carries the WHO wording. The seven aggressive lymphoma pages in this file now cite the NCI aggressive B-cell lymphoma summary.
- 2026-11-17 · Extranodal NK/T-cell lymphoma: The open problems said no randomised comparison of asparaginase regimens exists and that PD-1 blockade had never been tested in a randomised trial. Two randomised comparisons have been published (DDGP against SMILE in 42 patients, and the 256-patient NHL-004 phase 3 trial), and randomised PD-1 trials are registered and recruiting (Second accuracy pass; the abstracts for DOI 10.1158/1078-0432.ccr-16-0153 and DOI 10.3324/haematol.2025.287513 read through Europe PMC, and the ClinicalTrials.gov registry searched through its API) · Both open problems rewritten with the two trials and their results, and the registered trials named as not yet reported; three sources added
- 2026-11-17 · Extranodal NK/T-cell lymphoma: The page said anthracycline chemotherapy does not work at all, must not be used, and that this is the one lymphoma of which that is true, and that delaying radiotherapy worsens the outcome. The sources say the tumour frequently expresses P-glycoprotein, which reduces the efficacy of anthracycline regimens, and that radiotherapy is essential in localised disease with the same survival whether given with or after chemotherapy (Second accuracy pass; the NCI PDQ peripheral T-cell lymphoma summary read as raw HTML and the abstract for DOI 10.1007/s11899-016-0355-9 read through Europe PMC) · The summary line, the summary and the biomarker now say anthracyclines work poorly and why; the radiotherapy sentence carries the PDQ statement
- 2026-11-17 · Adult T-cell leukaemia/lymphoma: The page said Strongyloides hyperinfection is fatal and is prevented by screening, and cited the 2019 consensus report, which does not mention the worm. Two figures, the HTLV-1 prevalence among migrants and the 13 patients in the calcium study, had no source (Second accuracy pass; the full text of the 2019 report read through Europe PMC, the 2009 report read on its PubMed Central page, and the abstracts for DOI 10.3390/v16101526 and DOI 10.1084/jem.172.3.759 read through Europe PMC) · The Strongyloides advice now follows the 2009 report (testing before corticosteroids in people from endemic areas, preventive treatment to be considered) and is cited to it; the two uncited figures were confirmed and their sources added
- 2026-11-17 · ALK-positive anaplastic large cell lymphoma: ECHELON-2 was described as having three-quarters of its patients with systemic anaplastic large cell lymphoma. That was the enrolment target; 316 of 452, 70 percent, had it (22 percent ALK-positive, 48 percent ALK-negative), and ALK-positive patients were eligible only with an International Prognostic Index score of 2 or more. The page also called its median age the youngest apart from Hodgkin and Burkitt lymphoma when it is the lowest in the series (Second accuracy pass; the full text for DOI 10.1016/j.annonc.2021.12.002 read through PubMed Central and the tables of DOI 10.1038/bjc.2015.94 read through Europe PMC) · The share and the eligibility limit are corrected in the treatment row, the summary and the history; hazard ratios added and the five-year update marked exploratory; the age statement corrected; diagnosis years and the confidence interval added to the survival figure
- 2026-11-17 · Breast implant-associated anaplastic large cell lymphoma: The time from implant to diagnosis, 13.16 years, was called a median; the meta-analysis reports a mean. The page also said systemic ALK-negative disease is cured in a minority, which no source read supports (Second accuracy pass; the abstract for DOI 10.1007/s00266-024-03956-9 read through Europe PMC and PubMed) · Corrected to mean; the unsupported comparison replaced with the WHO description; the seroma and no-removal advice now cite the FDA page, which was read as raw HTML
- 2026-11-17 · Primary cutaneous anaplastic large cell lymphoma: The page said up to a quarter of lesions regress without treatment. The consensus recommendations it cites record spontaneous regression in up to 44 percent of patients (Second accuracy pass; the full text for DOI 10.1182/blood-2011-05-351346 read on its PubMed Central page) · Both passages now give up to 44 percent of patients, attributed to the consensus recommendations
- 2026-11-17 · Lymphomatoid papulosis: Mucha-Habermann disease was listed as another name; it is a different condition, pityriasis lichenoides. The page also said staging confirms there is no disease outside the skin, that lesions heal over three to twelve weeks, and that nobody in the published series has died. The consensus recommendations say scans and marrow biopsy are not needed in typical cases, that lesions regress over weeks or months, and the no-deaths finding is one series of 56 patients (Second accuracy pass; the NLM MeSH record read through its lookup API, the full text for DOI 10.1182/blood-2011-05-351346 read on its PubMed Central page and the abstract for DOI 10.1016/S0190-9622(03)02484-8 read through Europe PMC) · Alias removed; the staging, timing and survival statements now follow the sources; the 4 to 25 percent risk of a second lymphoma added; the follow-up row cites the consensus instead of a PDQ page that no longer covers it
- 2026-11-17 · Enteropathy-associated T-cell lymphoma: Median progression-free survival of 3.4 months and overall survival of 7.1 months were credited to 54 patients given anthracycline chemotherapy. Of the 54, 35 had chemotherapy and 19 had surgery alone. The page also called the comparison the best evidence the disease has without mentioning a later prospective study, said a gluten-free diet has not been shown to prevent the lymphoma when NCI PDQ states that it does, and gave no diagnosis years (Second accuracy pass; the abstract for DOI 10.1182/blood-2009-07-231324 read through Europe PMC, the NCI PDQ peripheral T-cell lymphoma summary read as raw HTML and the full text for DOI 10.1002/jha2.70400 read through Europe PMC) · The cohort is described as it was treated, with its years (1994 to 1998); the later phase 2 study is named, with a note that its full text could not be read; the open problem now reports the PDQ statement; the SETD2 sentence gives both WHO and PDQ
- 2026-11-17 · Monomorphic epitheliotropic intestinal T-cell lymphoma: The cohort behind the treatment comparison was called the largest published and drawn from several centres; it is from one hospital in Seoul and makes no such claim. The adjusted overall survival difference was not statistically significant and the page did not say so. The page said no geographic distribution could be verified, and that the disease was named nine years ago (Second accuracy pass; the abstract and full text for DOI 10.1002/jha2.70400 read through Europe PMC and the NCI PDQ peripheral T-cell lymphoma summary read as raw HTML) · Cohort described as single-centre with its arm sizes (22 and 28 of 52 treated); the adjusted overall survival result added (hazard ratio 0.51, p = 0.077); the PDQ statement that it is more common in Asia added; the relative date replaced with the 2016 revision
- 2026-11-17 · T-cell lymphoma pages: NCI PDQ links: Seven T-cell and NK-cell lymphoma pages and five treatment rows cited the NCI PDQ adult non-Hodgkin lymphoma summary, whose address now opens the indolent B-cell summary. Survival figures from the UK subtype series were given without diagnosis years (Second accuracy pass; the old and new PDQ addresses fetched as raw HTML and the statements re-read against the peripheral T-cell and mycosis fungoides summaries; HMRN tables read through Europe PMC) · Links and guideline stamps now point to the peripheral T-cell summary, or the mycosis fungoides summary for that page; the UK survival figures carry September 2004 to August 2012
- 2026-11-17 · Diffuse large B-cell lymphoma: third line and beyond: Selinexor was listed as a treatment option at third line and beyond. Its United States accelerated approval for diffuse large B-cell lymphoma was removed from the label in April 2026 (Second accuracy pass; the current Xpovio label (effective 12 May 2026) read through the openFDA label API, whose recent major changes record the indication as removed) · Selinexor is taken out of the list of options and the row says the indication was removed in April 2026
- 2026-11-17 · Diffuse large B-cell lymphoma: first-line treatment, IPI 2 to 5: Five-year overall survival in POLARIX was given as 82.3 against 79.5 per cent, figures that are not in the abstract or the text of the five-year paper, and the subgroup analyses were called prespecified where the paper calls them exploratory. Tafasitamab and lenalidomide with R-CHOP was called the newer option for IPI 3 to 5 without saying it is not on the United States label, and frontMIND eligibility also allowed an age-adjusted IPI of 2 to 3 in people aged 60 or under (Second accuracy pass; the abstract and full text for DOI 10.1200/JCO-25-00925 read through Europe PMC, the frontMIND abstract (PubMed 42217458) and registry record NCT04824092, and the Monjuvi label read through openFDA) · The survival rates are replaced by the hazard ratio with its interval, 0.85 (0.63 to 1.15), in the row and in the open problem; the subgroups are called exploratory; the frontMIND sentence gives the true eligibility, the interim survival hazard ratio and the adverse event rates, and says the United States label did not include first-line use on 17 November 2026
- 2026-11-17 · What the NHS in England funds for lymphoma: Ibrutinib for Waldenstrom macroglobulinaemia was described as covered by TA491 and TA795; TA795 replaced TA491 and does not recommend it. Pembrolizumab in Hodgkin lymphoma was described as covered by TA540, TA772 and TA967; TA540 did not recommend it and has been replaced by TA967, and TA772 and TA967 both apply from the age of 3 and differ by previous brentuximab vedotin. TA472 and TA110 were cited as current although TA629 and TA243 replaced them. TA1159 was written as if it covered four lymphoma types when it covers two, and the conditions on TA502 and TA513 were missing (Second accuracy pass; the recommendation page of each of the 49 appraisals read as raw text from nice.org.uk, and Cancer Drugs Fund list version 1.410 of 12 November 2026) · Each sentence now gives the current appraisal, whether it recommends the drug, and its conditions
- 2026-11-17 · Watch and wait in lymphoma: The entry said the quality-of-life endpoint of the rituximab against watchful waiting trial did not separate the arms at seven months. The trial reported significantly better mental adjustment to cancer and illness coping scores with maintenance rituximab (Second accuracy pass; the abstract for DOI 10.1016/S1470-2045(14)70027-0 read through PubMed (the Europe PMC copy is truncated), and the NCI PDQ indolent B-cell lymphoma summary) · The sentence now reports the quality-of-life result as published and says overall survival was no different at three years
- 2026-11-17 · Diffuse large B-cell lymphoma: guideline links: Eight standard-of-care rows and one glossary entry cited the NCI PDQ adult non-Hodgkin lymphoma summary, whose address now redirects to the indolent B-cell lymphoma summary (Second accuracy pass; both cancer.gov addresses fetched as raw pages and the aggressive B-cell summary searched for the statements the rows make) · The rows and the regimen entry now link to the aggressive B-cell summary, and the three entries about indolent disease name the indolent summary correctly
- 2026-11-17 · Mantle cell lymphoma: The first-line row for younger patients stopped at the 2024 report of TRIANGLE and called transplant optional with the comparison still open; an open problem said the trial had not answered which arm is best and that the arm without transplant had the shortest follow-up. Ibrutinib was presented as a standard with no mention that it has no mantle cell lymphoma indication in the United States. The SYMPATICO row gave a 69.2 per cent complete response for a TP53-mutated arm, and the row on treatment after a BTK inhibitor lacked the 2026 approvals. Three rows linked the old NCI summary address, which now opens the indolent B-cell summary. (Second accuracy pass; the abstract of the 4.5-year TRIANGLE report (PubMed 42134356) and of its rituximab maintenance analysis (PubMed 42447409), the SYMPATICO primary abstract, final-analysis full text (PMC13462179) and registry results (NCT03112174), the FDA tables of withdrawn, verified and ongoing accelerated approvals and the FDA approval notifications read as raw text from fda.gov, the current Imbruvica label on openFDA, and the NCI mantle cell lymphoma summary.) · Added the 4.5-year results (four-year failure-free survival 82 against 81 per cent with and without transplant, overall survival 88 and 90 against 81 per cent), rewrote the practice sentence and the open problem, stated the withdrawal of ibrutinib's United States approval on 18 May 2023 in three rows, corrected the SYMPATICO arm to untreated patients aged 65 or over or with a TP53 mutation and gave both hazard ratios and the overall survival result, added sonrotoclax (13 May 2026), the conversion of brexucabtagene autoleucel (1 April 2026) and the accelerated status of pirtobrutinib, and pointed the three links at the mantle cell summary.
- 2026-11-17 · Follicular lymphoma: Tazemetostat was listed as an option after two or more lines although its United States approval was withdrawn in 2026. The PI3K inhibitors were said to have largely been withdrawn. Odronextamab was listed under fixed-duration bispecific antibodies. RELEVANCE was described as showing equivalence, and GALLIUM and RELEVANCE were given only at their interim reports. (Second accuracy pass; the FDA table of withdrawn cancer accelerated approvals read as raw text from fda.gov, the ELM-2 abstract (DOI 10.1016/j.annonc.2024.08.2239), the EMA page for Ordspono, the RELEVANCE abstract and its ten-year analysis (PubMed 41915772) and the GALLIUM final analysis (PubMed 37404773) read through PubMed and Europe PMC.) · Stated that the tazemetostat follicular lymphoma approval is listed as withdrawn on 22 June 2026 and removed it from the options; gave the four PI3K withdrawal dates; said odronextamab is continued until progression and gave its European status; reworded RELEVANCE as similar results in a superiority trial and added its ten-year figures (progression-free survival 46.4 against 46.6 per cent); added the GALLIUM seven-year figures and the absence of a survival difference; dated the epcoritamab approval.
- 2026-11-17 · Primary mediastinal B-cell lymphoma: KEYNOTE-170 was given only at its 12.5-month report (response 45 per cent, complete response 13 per cent). The dose-adjusted EPOCH-R result was written as five-year rates. Both rows linked the old NCI summary address, which now opens the indolent B-cell summary. (Second accuracy pass; the abstracts of the 2019 report and the final analysis (PubMed 31609651 and 37130017) and of Dunleavy 2013 (PubMed 23574119) read through PubMed, and the NCI aggressive B-cell summary read from cancer.gov.) · Added the final analysis (response 41.5 per cent, complete response 20.8 per cent, median progression-free survival 4.3 months, four-year survival 45.3 per cent), reworded the EPOCH-R figures as rates over a median five years in a single-arm study of 51 patients, and switched both links to the aggressive B-cell summary. The Burkitt lymphoma relapse row had the same link fault and was switched too.
- 2026-11-17 · Marginal zone lymphoma: Zanubrutinib in MAGNOLIA was said to cause fewer cardiac events than ibrutinib, a comparison the single-arm study did not make, and the row did not say that ibrutinib has lost its marginal zone lymphoma approval. (Second accuracy pass; the MAGNOLIA abstract (PubMed 34526366) read through PubMed and the FDA tables of withdrawn and ongoing accelerated approvals read from fda.gov.) · Reworded to a single-arm study of 68 patients with two cases of atrial fibrillation or flutter, and added that zanubrutinib's approval is still accelerated and that ibrutinib's was withdrawn on 18 May 2023.
- 2026-11-17 · Advanced-stage classical Hodgkin lymphoma: Peripheral neuropathy in SWOG S1826 was given as 28.1 against 54.2 per cent in two rows. ECHELON-1 was called the only first-line trial in advanced Hodgkin lymphoma to show a survival advantage. RATHL was said to show bleomycin could be dropped without loss of control. The page said there was no NICE appraisal of nivolumab with AVD. S1826 was called the first trial to include adolescents and adults, and HD21 the highest progression-free survival reported. (Second accuracy pass; the S1826 full text (PMC11488644) read through NCBI, the abstracts of GHSG HD9 at ten years (PubMed 19704068), RATHL (PubMed 27332902) and HD21 (PubMed 38971175), and the NICE page for appraisal ID6690 read as raw text.) · Neuropathy corrected to 29 against 56 per cent; the ECHELON-1 sentence reworded because HD9 had shown a survival gain for escalated BEACOPP; RATHL said to have fallen just short of its non-inferiority margin; the NICE sentence changed to an appraisal suspended on 19 February 2026 because the company will not make a submission; the two unsourced superlatives reworded.
- 2026-11-17 · Early-stage classical Hodgkin lymphoma: GHSG HD10 was said to give five-year freedom from treatment failure of about 93 per cent and survival of about 97 per cent with two cycles of ABVD and 20 Gy. The 93 per cent is the four-cycle figure. (Second accuracy pass; the abstract of Engert 2010 (PubMed 20818855) read through PubMed.) · Corrected to 91.1 per cent with two cycles against 93.0 per cent with four, not a significant difference, with no significant difference in survival; the survival percentage, which the abstract does not give, was removed.
- 2026-11-17 · Peripheral T-cell lymphoma: Pralatrexate was said to have been approved on a response rate of about 29 per cent; 29 is the number of responders among 109 patients. Autologous transplant in first remission was said never to have had a randomised trial. Six rows, including the angioimmunoblastic one, linked the old NCI summary address, which now opens the indolent B-cell summary. (Second accuracy pass; the current Folotyn and Beleodaq labels read through openFDA, the FDA tables of ongoing and withdrawn accelerated approvals, and the NCI peripheral T-cell lymphoma summary read from cancer.gov.) · Response rate corrected to 27 per cent; the transplant sentence now says there is no randomised trial against no transplant and describes the one randomised comparison of autologous with allogeneic transplant; added that pralatrexate and belinostat are still accelerated approvals and dated the romidepsin withdrawal; added the five-year ECHELON-2 figures; switched the six links to the peripheral T-cell summary.
- 2026-11-17 · Adult T-cell leukaemia/lymphoma: The JCOG9801 survival figures (24 against 13 per cent at three years) were given without saying the difference was not statistically significant. (Second accuracy pass; the abstract of Tsukasaki 2007 (DOI 10.1200/JCO.2007.11.9958) read through Europe PMC.) · Added that the difference was not significant (two-sided P = 0.169).
- 2026-11-17 · Mycosis fungoides: Mechlorethamine gel was said to have cleared index lesions in 58.5 per cent of patients; the figure is a response rate, not clearance. (Second accuracy pass; the abstract of the pivotal trial (PubMed 23069814) read through PubMed.) · Reworded to a response in index lesions.
- 2026-11-17 · TRANSCEND NHL 001: Called lisocabtagene maraleucel the third CAR-T product for lymphoma, and gave the setting as after at least one prior line. The same "third CAR-T product" wording headed the 2020 entry in the non-Hodgkin lymphoma timeline. (Second accuracy pass; the FDA product pages for Tecartus and Breyanzi read as raw HTML (Tecartus approved for mantle cell lymphoma on 24 July 2020, Breyanzi on 5 February 2021) and the ClinicalTrials.gov record NCT02631044, whose large B-cell cohort required at least two lines of systemic therapy or an autologous transplant.) · Now the third CAR-T product for large B-cell lymphoma, in the record and the timeline title, and the setting says at least two prior lines or an autologous transplant.
- 2026-11-17 · ENRICH: Named the funders, Cancer Research UK and Johnson and Johnson Pharmaceuticals, as the sponsor, and said the chemotherapy-free combination "worked better" without giving the survival result. (Second accuracy pass; the full text of DOI 10.1016/S0140-6736(25)01432-1 read through Europe PMC (PMC12549478), which says the trial was sponsored by Plymouth Hospitals NHS Trust and gives a hazard ratio for death of 0.87 (95 per cent confidence interval 0.64 to 1.18).) · Sponsor corrected with the funders named as funders; median progression-free survival (65.3 against 42.4 months) and the non-significant overall survival result added to the summary, result and summary line.
- 2026-11-17 · JULIET: Gave the setting as after two or more lines, or after an autologous transplant. (Second accuracy pass; the ClinicalTrials.gov record NCT02445248 read through the v2 API, which requires at least two lines and a transplant that failed, was not possible or was declined.) · Setting and summary reworded to the registry's eligibility.
- 2026-11-17 · ROSEWOOD: The summary line said the combination roughly doubled the response rate and tripled the time to progression. (Second accuracy pass; the abstract for DOI 10.1200/JCO.23.00775 read through Europe PMC: response 69 against 46 per cent, median progression-free survival 28.0 against 10.4 months.) · The summary line now gives the figures instead of the multiples.
- 2026-11-17 · GHSG HD18: The enrolment note called the registry's 1,500 a target, said 156 patients were ineligible without the split, and said 1,945 were randomised. (Second accuracy pass; the abstract for DOI 10.1016/S0140-6736(17)32134-7 read through Europe PMC (137 ineligible before randomisation and 19 after; 434 and 1,005 patients in the two randomised comparisons) and the ClinicalTrials.gov record NCT00515554, which gives 1,500 as the actual figure.) · Basis changed from randomised to analysed and the note rewritten with the abstract's numbers.
- 2026-11-17 · GOYA: Explained the registry's terminated status in words the registry does not use. (Second accuracy pass; the ClinicalTrials.gov record NCT01287741 read through the v2 API.) · Replaced with the registry's reason, quoted exactly.
- 2026-11-17 · FLYER: Said four cycles caused about a third fewer side effects in one place and roughly a quarter fewer in another. (Second accuracy pass; the abstract for DOI 10.1016/S0140-6736(19)33008-9 read through Europe PMC: 1,330 adverse events against 1,706, which is 22 per cent fewer.) · Both now say about a fifth, with the counts.
- 2026-11-17 · SHINE: The summary line said ibrutinib added about two and a half years before relapse. (Second accuracy pass; the abstract for DOI 10.1056/NEJMoa2201817 read through Europe PMC: 80.6 against 52.9 months, a difference of 27.7 months.) · Now a little over two more years (28 months).
- 2026-11-17 · TROG 99.03: The summary line said adding drug treatment halved the risk of relapse outside the radiotherapy field. (Second accuracy pass; the abstract for DOI 10.1200/JCO.2018.77.9892 read through Europe PMC, which gives a progression-free survival hazard ratio of 0.57 and no separate out-of-field figure.) · Reworded to a reduction of about four tenths in the risk of the lymphoma coming back or growing.
- 2026-11-17 · PRIMA: Described the progression-free survival gain as six extra years before the next treatment; ELARA did not say its figures came from an interim analysis; the EORTC H10 enrolment note lacked the registry figure. (Second accuracy pass; the abstracts for DOI 10.1200/JCO.19.01073 and 10.1038/s41591-021-01622-0 read through Europe PMC and the ClinicalTrials.gov record NCT00433433.) · PRIMA now says six extra years before the lymphoma progresses; ELARA says prespecified interim analysis; the H10 note gives the registry's 1,952.
- 2026-11-17 · JCOG9801: This record and SMILE both said the disease has no cancer record of its own, although each now links to one; the open problems list said adult T-cell leukaemia/lymphoma has had one phase 3 trial, which the cited abstract does not say. (Second accuracy pass; the file read against its own links and the abstract for DOI 10.1200/JCO.2007.11.9958 read through Europe PMC.) · Stale sentence removed from both records; the open problem now says the chemotherapy standard rests on one phase 3 trial.
- 2026-11-17 · POLAR BEAR: Recorded as active and said primary completion was estimated for 28 December 2025, "so the first report is close". (Second accuracy pass; the ClinicalTrials.gov record NCT04332822 read through the v2 API (still recruiting, last updated 19 September 2024) and a Europe PMC search of 2025 and 2026 papers, which found only the 2023 safety abstract on the first 127 patients.) · Status set to recruiting; the sentence now gives the registry state and says no efficacy result could be found on 17 November 2026.
- 2026-11-17 · Mosunetuzumab against rituximab in low tumour burden follicular lymphoma: Described the rituximab arm as four weekly doses and the subcutaneous formulation as permitted; recorded as active. (Second accuracy pass; the ClinicalTrials.gov record NCT06337318 read through the v2 API: four weekly doses and then one every eight weeks to five cycles, subcutaneous after the first dose, against up to eight three-week cycles of mosunetuzumab.) · Setting and summary corrected, the SWOG study number added, status set to recruiting.
- 2026-11-17 · RADAR: RADAR, ARCHED, PRIMA-CNS, FORTplus and the mosunetuzumab with lenalidomide marginal zone trial were recorded as active while still recruiting; the marginal zone record claimed a "first time" for its stratification; the Hodgkin timeline entry for 2030 was titled "RADAR reports". (Second accuracy pass; the ClinicalTrials.gov records NCT04685616, NCT05820841, NCT06830421, NCT05045664 and NCT06006117 read through the v2 API on 17 November 2026.) · Status set to recruiting on all five; the unsupported "first time" removed; the timeline title now says primary completion date.
- 2026-11-17 · NCI dose-adjusted EPOCH-R in Burkitt and MYC-rearranged large B-cell lymphoma: The trial record and its paper record said no randomised trial had ever been confined to MYC-rearranged or double-hit lymphoma. The phase 2 part of Alliance A051701, published in Lancet Haematology in September 2026, randomised 73 people with double-hit lymphoma to this regimen with or without venetoclax and closed early because of deaths on treatment (Second accuracy pass; the abstract for DOI 10.1016/S2352-3026(26)00191-2 read through PubMed, and the registry record NCT03984448 read through the ClinicalTrials.gov API) · Both records now give the trial and its result (six deaths on treatment against one, progression-free survival hazard ratio 1.13, 0.53 to 2.37, two-year overall survival 52 against 72 per cent), say that it did not compare the regimen with R-CHOP, and link the paper and the registry record
- 2026-11-17 · Study 0761-009: mogamulizumab against investigator's choice in relapsed adult T-cell leukaemia/lymphoma: The record described the patients as of Caribbean, African and Latin American origin and explained the low response rate by different subtypes and more prior treatment than the Japanese study. The paper gives neither: it lists the countries, calls the population more ethnically diverse, and names other differences. The same explanation stood on the Japanese phase 2 trial record (Second accuracy pass; the full text for DOI 10.3324/haematol.2018.205096 read through PubMed Central) · The countries replace the statement of origin, and the three records give the authors' own reasons: relapsed but not refractory disease in Japan, a later check of response, and older patients with worse performance status and more bone marrow involvement. A caveat records that the adjusted hazard ratio's lower limit is 0.337 in the abstract and 0.327 in the full text, and that crossover was allowed with no apparent survival difference
- 2026-11-17 · Dose-intensified chemotherapy alone or in combination with mogamulizumab in newly diagnosed aggressive adult T-cell leukaemia-lymphoma: The record said the study made no survival comparison. The full text reports median progression-free survival of 8.5 against 6.3 months and median overall survival not reached in either arm. The enrolment note also omitted that 54 people were enrolled, of whom 53 were treated (Second accuracy pass; the full text for DOI 10.1111/bjh.13338 read through PubMed Central) · The survival figures and follow-up are given and described as descriptive, and the enrolment note on the paper and the trial record says 54 enrolled and 53 treated
- 2026-11-17 · Phase I/II study of concurrent chemoradiotherapy for localized nasal natural killer/T-cell lymphoma: JCOG0211: The summary said the chemotherapy drugs were chosen because none is affected by the multidrug-resistance pump. The authors describe the regimen as multidrug resistance-nonrelated agents and etoposide, setting etoposide apart. Separately, the DDGP against SMILE trial record gave the registry's estimated enrolment of 80 as the number enrolled (Second accuracy pass; the abstract for DOI 10.1200/JCO.2009.23.8295 read through PubMed, and the registry record NCT01501149 read through the ClinicalTrials.gov API) · The summary uses the authors' description, and the DDGP records say the registry figure is an estimate
- 2026-11-17 · Long-term results of the PRIMA study: The summary line said exactly as many people in each group were alive, and the interpretation gave six and a half extra years before the next treatment. The six and a half years is the gain in progression-free survival; the median time to the next lymphoma treatment was not reached with maintenance against 6.1 years with observation. (Second accuracy pass; the abstract (PubMed 31339826) and the open full text (PMC6823890) read through Europe PMC.) · Reworded to about 80 per cent alive in each group at ten years, and to a gain of about six and a half years before progression (10.5 against 4.1 years), with the time to next treatment figures stated.
- 2026-11-17 · ENRICH: ibrutinib and rituximab versus immunochemotherapy in mantle cell lymphoma: A caveat said no overall survival result is reported. The paper reports five-year overall survival of 58 against 55 per cent, hazard ratio 0.87 (0.64 to 1.18). The record also did not say that ibrutinib has no mantle cell lymphoma indication on its United States label. (Second accuracy pass; the open full text (PMC12549478) read through Europe PMC and the Imbruvica label read through openFDA.) · Overall survival added as a finding and the caveat replaced, with the COVID-19 death counts the paper gives; a caveat on the United States label added.
- 2026-11-17 · LyMa: rituximab after autologous stem-cell transplantation in mantle cell lymphoma: The record said maintenance helped people live longer without saying that the long-term follow-up no longer found a significant survival difference: seven-year overall survival 83.2 against 72.2 per cent, hazard ratio 0.63 (0.37 to 1.08), p = 0.088. (Second accuracy pass; the abstracts of the 2017 paper and of the 2024 long-term follow-up (DOI 10.1200/JCO.23.01586) read through PubMed.) · Summary line, findings and interpretation now give the four-year survival result as the primary report and the seven-year result beside it; the follow-up paper is linked.
- 2026-11-17 · EORTC/LYSA/FIL H10: early PET response-adapted treatment in stage I and II Hodgkin lymphoma: The record said the interim scan should be used to intensify treatment, on the five-year gain from 77.4 to 90.6 per cent. At the preplanned ten-year analysis that difference was no longer statistically significant (hazard ratio 0.67, 0.37 to 1.20). (Second accuracy pass; the abstract of the ten-year report (DOI 10.1200/JCO.23.01745) read through PubMed.) · Ten-year results added to the summary line, findings and interpretation, with the scan-negative results that did hold; the report is linked.
- 2026-11-17 · ROSEWOOD: zanubrutinib plus obinutuzumab in relapsed or refractory follicular lymphoma: Only the primary analysis was given, with a caveat that there was no overall survival result. The final analysis reports a shorter median progression-free survival in the combination arm (22.1 against 10.3 months) and an overall survival comparison that was not significant. (Second accuracy pass; the abstract of the final analysis (DOI 10.1007/s11523-026-01268-3) read through PubMed and the Brukinsa label read through openFDA.) · Final analysis added to the caveats and linked; a note added that the United States approval is an accelerated one.
- 2026-11-17 · ZUMA-5: axicabtagene ciloleucel in relapsed or refractory indolent non-Hodgkin lymphoma: The summary line called it the first CAR-T trial in slow-growing lymphoma, which no source read supports, and the interpretation presented the treatment as an option in marginal zone lymphoma although the United States label lists follicular lymphoma only, under accelerated approval. No long-term result was given. (Second accuracy pass; the abstracts of the 2022 paper and the five-year analysis (DOI 10.1200/JCO-25-00668) read through PubMed and the Yescarta label read through openFDA.) · First removed, the label position stated, and the five-year figures (median progression-free survival 62.2 months) added with a link.
- 2026-11-17 · GOYA: obinutuzumab or rituximab plus CHOP in untreated diffuse large B-cell lymphoma: A caveat said the registry termination reflected the end of the development programme; the registry gives a different reason. The final analysis was not mentioned. (Second accuracy pass; the ClinicalTrials.gov record NCT01287741 read through the registry API and the abstract of the final analysis (DOI 10.1186/s13045-020-00900-7) through PubMed.) · Caveat reworded to the registry's own reason, and the five-year result (63.8 against 62.6 per cent, hazard ratio 0.94) added with a link.
- 2026-11-17 · ECHELON-3: brentuximab vedotin combination for relapsed diffuse large B-cell lymphoma: The interpretation called the trial the first demonstration that a CD30-directed conjugate helps where CD30 expression is variable; the paper itself cites an earlier study showing responses across CD30 levels. The approved indication is large B-cell lymphoma, not only diffuse large B-cell lymphoma. (Second accuracy pass; the open full text (PMC11936473) read through Europe PMC and the Adcetris label read through openFDA.) · Replaced with the paper's own statement that almost 70 per cent of patients had CD30-negative tumours and benefit was seen regardless; indication wording corrected.
- 2026-11-17 · inMIND: tafasitamab, lenalidomide and rituximab in relapsed or refractory follicular lymphoma: The interpretation claimed the largest progression-free survival hazard ratio reported in the setting, with no source, and did not give the approval. (Second accuracy pass; the abstract read through PubMed, and the Monjuvi label and the approval history for BLA 761163 read through openFDA.) · Superlative removed; the June 2025 United States indication and its marginal zone lymphoma limitation added.
- 2026-11-17 · ECHO: acalabrutinib plus bendamustine and rituximab in untreated mantle cell lymphoma: A caveat said the investigators had discussed COVID-19 as affecting early mortality, which is not what the paper says, and the approval that followed was missing. (Second accuracy pass; the open full text (PMC12225732) read through Europe PMC and the Calquence label read through openFDA.) · Caveat now gives the paper's figures (33 and 26 COVID-19 deaths, and the censored survival analysis, hazard ratio 0.75, not significant); the United States indication added.
- 2026-11-17 · IELSG-19: chlorambucil, rituximab or both in MALT lymphoma: The record called the trial the only randomised evidence, called survival identical across arms, and said the two-arm comparison was reported before the third arm was unblinded. The abstract says first to the authors' knowledge, about 90 per cent in each arm, and reported before any analysis of the third arm. (Second accuracy pass; the abstract (PubMed 28355112) read through PubMed.) · All three reworded to the abstract's statements.
- 2026-11-17 · TROG 99.03: systemic therapy after involved-field radiotherapy in early-stage follicular lymphoma: The interpretation said the trial is the reason modern practice stages with positron emission tomography, which the paper does not say, and a caveat said ten-year survival exceeded 87 per cent in both arms when one arm was at 87. The extended follow-up was missing. (Second accuracy pass; the abstracts of the 2018 paper and of the extended follow-up (DOI 10.1016/j.ebiom.2024.105468) read through PubMed and Europe PMC.) · Reworded to the association the paper reports (p = 0.056); survival wording corrected; the 11.3-year results added with a link.
- 2026-11-17 · SHINE: ibrutinib plus bendamustine and rituximab in untreated mantle cell lymphoma: The summary line gave the gain as about two and a half years; 80.6 against 52.9 months is 27.7 months. The record did not say that the United States label for ibrutinib no longer lists mantle cell lymphoma. (Second accuracy pass; the abstract read through PubMed and the Imbruvica label read through openFDA.) · Reworded to more than two years, and a caveat on the label added.
- 2026-11-17 · GHSG HD16: PET-guided treatment in early-stage favourable Hodgkin lymphoma: A caveat called the survival curves identical and gave no follow-up beyond 45 months. (Second accuracy pass; the abstract of the follow-up report (DOI 10.1038/s41375-023-02064-y) read through PubMed.) · Reworded, and the 64-month results (94.2 against 86.7 per cent, hazard ratio 2.05) added with a link.
- 2026-11-17 · TRANSCEND NHL 001, ELARA and POLARGO: Three records gave early results with no later figure or label position: five-year survival after lisocabtagene maraleucel, the 29-month ELARA update, and the absence of the POLARGO combination from the United States polatuzumab label. (Second accuracy pass; the abstracts of the later reports (DOI 10.1182/blood.2025032270 and 10.1182/blood.2023021567) read through PubMed and the Polivy label read through openFDA.) · Each added as a caveat or sentence on its own record, with links.
- 2026-11-17 · Lymphoma roadmap: The mantle cell lymphoma step was titled as showing that no way of using a Bruton tyrosine kinase inhibitor extends life, and said nobody had shown any strategy makes patients live longer. (Second accuracy pass; the abstract of the TRIANGLE 4.5-year follow-up (Lancet, May 2026, DOI 10.1016/S0140-6736(26)00362-4) read through PubMed, with the SHINE, ECHO and ENRICH abstracts.) · Title and paragraph rewritten: four-year overall survival was 88 and 90 per cent in the two ibrutinib groups against 81 per cent with transplant alone, and adding the transplant to ibrutinib gave no gain. The statement that survival has not been extended is kept for the three trials in older patients only. Paper linked.
- 2026-11-17 · Lymphoma roadmap: The T-cell step said the randomised evidence amounted to two trials, treated NK/T-cell lymphoma as resting on a single-arm study, and said ALCANZA improved survival in cutaneous T-cell lymphoma as ECHELON-2 had in peripheral T-cell lymphoma. (Second accuracy pass; abstracts for DOI 10.1001/jamaoncol.2022.1968, 10.1016/j.xinn.2023.100426 and 10.1016/S0140-6736(17)31266-7 read through PubMed, and recruiting phase 3 trials listed from ClinicalTrials.gov.) · Added the two randomised asparaginase-regimen comparisons (DDGP against SMILE in 80 treated patients, hazard ratio 0.42 for progression; ESA against MESA in 248 patients). ALCANZA now described as an open-label trial with a response endpoint, 56.3 against 12.5 per cent. Title softened from almost nothing to little.
- 2026-11-17 · Lymphoma roadmap: Said no first-line decision anywhere is made by genetic subtype and that the trial assigning treatment by genetics had not been run. Said circulating tumour DNA flags transformation before it declares itself. (Second accuracy pass; the GUIDANCE-01 abstract (Cancer Cell 2023, DOI 10.1016/j.ccell.2023.09.004) read through PubMed, ClinicalTrials.gov record NCT05351346, and the Scherer 2016 abstract.) · Added GUIDANCE-01 (128 patients, complete response 88 against 66 per cent with treatment guided by a simplified genetic algorithm) and the registered phase 3, whose registry status is unknown. Transformation wording changed to patterns that may predict it.
- 2026-11-17 · Lymphoma roadmap: The link to the NCI adult non-Hodgkin lymphoma summary pointed at an address that now redirects to the indolent B-cell summary. Second-cancer and heart disease figures after Hodgkin lymphoma were given as shares of survivors with no cohort or era. RATHL was said to have shown bleomycin could be dropped, IELSG-19 to be the only randomised first-line trial in marginal zone lymphoma, and POLARIX the only first-line improvement in two decades. (Second accuracy pass; cancer.gov addresses requested directly, and the abstracts for Schaapveld 2015, van Nimwegen 2015, RATHL, IELSG-19, GOYA and POLARIX read through PubMed.) · Three NCI links replace the one. The 48.5 and 50 per cent figures are now 40-year cumulative incidences in five-year survivors treated from 1965. RATHL given with its figures (84.4 against 85.7 per cent, just short of its margin); IELSG-19 described in its authors' words as the first such trial in MALT lymphoma; GOYA marked not significant; POLARIX given with its hazard ratio and no survival difference.
- 2026-11-17 · Lymphoma roadmap: The summary line claimed three firsts for lymphoma that no cited source supports, and the summary dated the first human retrovirus to twenty years after the Burkitt lymphoma virus. POLAR BEAR was described as having no report without saying what had been checked. (Second accuracy pass; PubMed records for the 1958, 1964 and 1980 papers, ClinicalTrials.gov NCT04332822, and PubMed and Europe PMC searches for a POLAR BEAR report.) · Firsts reworded to what the records show. Date corrected to 1980. POLAR BEAR now says the registry record was last updated in September 2024 and no full report was indexed in PubMed on 17 November 2026.
- 2026-11-17 · Assign first-line treatment in diffuse large B-cell lymphoma by genetic subtype: Said the trial that would test genetics-guided first-line treatment had not been run and that nobody had put the pieces in a randomised trial. Said the classifier needs copy number and structural variant data and that its training cohorts were largely of European ancestry. (Second accuracy pass; the GUIDANCE-01 abstract read through PubMed, ClinicalTrials.gov NCT05351346, and the full text of Wright 2020 (PMC8459709) read through NCBI.) · GUIDANCE-01 and the registered phase 3 added; maturity moved to early clinical. LymphGen described as able to run on mutations alone with a loss of one subtype and some accuracy; the ancestry statement replaced by the fact that the paper does not report ancestry.
- 2026-11-17 · Use circulating tumour DNA instead of the interim scan: Said no trial has yet used circulating tumour DNA to change anyone's treatment. (Second accuracy pass; a ClinicalTrials.gov search read through the API, which lists recruiting trials that adapt treatment to the result, among them NCT06745076 in advanced Hodgkin lymphoma.) · Reworded: such trials are recruiting, and none has yet reported that doing so helps. The same correction made on the Scherer 2016 paper.
- 2026-11-17 · Randomised evidence for the T-cell lymphomas: The summary line said adult T-cell leukaemia has had one randomised trial, in 1998, while the text named two, reported in 2007 and 2015. Said diseases without a sponsor have no trials. (Second accuracy pass; abstracts for JCOG9801 (DOI 10.1200/JCO.2007.11.9958), the 2015 mogamulizumab phase 2 (DOI 10.1111/bjh.13338) and the DDGP against SMILE trial read through PubMed.) · Summary line corrected to the two trials and their years. Investigator-led randomised trials in NK/T-cell lymphoma acknowledged.
- 2026-11-17 · Schaapveld 2015: second cancer risk after Hodgkin lymphoma: The summary line said almost half of nearly 4,000 people developed another cancer within forty years. In the cohort 908 of 3,905 had a second cancer at a median of 19.1 years; 48.5 per cent is the estimated cumulative incidence at 40 years. The companion heart disease paper had the same fault (797 of 2,524 with an event; 50 per cent is the 40-year cumulative incidence). (Second accuracy pass; abstracts for DOI 10.1056/NEJMoa1505949 and 10.1001/jamainternmed.2015.1180 read through PubMed.) · Both summary lines rewritten as a 40-year chance in five-year survivors, with the treatment years (1965 to 2000 and 1965 to 1995).
- 2026-11-17 · Schmitz 2018: genetics and pathogenesis of diffuse large B-cell lymphoma: Said a substantial minority of tumours fit none of the four genetic subtypes. The paper estimates the four subtypes cover 46.6 per cent of cases, so most tumours are unclassified. (Second accuracy pass; full text PMC6010183 read through NCBI.) · Caveat and summary line corrected; the survival analysis is now given with its numbers (240 patients, 119 classified).
- 2026-11-17 · LymphGen (Wright 2020): Said the tool needs comprehensive data including copy number and structural variants, that its training cohorts were largely of European ancestry, and that it is the classifier used to select patients in genetics-directed trials. The paper says it runs on mutation data alone with other inputs optional, does not report ancestry, and the one randomised genetics-guided trial used a different algorithm. (Second accuracy pass; full text PMC8459709 read through NCBI and the GUIDANCE-01 abstract through PubMed.) · Three caveats rewritten from the paper, with the share classified (329 of 574, 63.1 per cent). GUIDANCE-01 linked.
- 2026-11-17 · Hans 2004: immunohistochemistry for cell of origin: Said the ROBUST trial selected patients by this algorithm. ROBUST used a gene-expression test. The five-year survival figures of 76 and 34 per cent carried no statement of their era. (Second accuracy pass; ROBUST full text PMC8078325 read through NCBI and the PHOENIX registry record NCT01855750. The publisher refused a request for the Hans full text.) · Caveat corrected. A caveat added that the abstract does not give the treatment, the cases predate September 2003, and the figures are not a guide to current outcomes.
- 2026-11-17 · Casulo 2015: early progression in follicular lymphoma: The summary line counted the two years from first treatment and said survival was half that of everyone else; the paper counts from diagnosis and compares with a reference group. The text said transformation is the commonest cause of early progression, which the paper does not say. (Second accuracy pass; abstracts for DOI 10.1200/JCO.2014.59.7534 and 10.1182/blood.2019000258 read through PubMed.) · Summary line rewritten with 50 against 90 per cent. Transformation statement replaced by the one series read (76 per cent of early progressions among 296 patients given bendamustine and rituximab), with the paper linked. Same change on the roadmap.
- 2026-11-17 · Scherer 2016: circulating tumour DNA in lymphoma: The summary line said the test spotted transformation before it showed; the paper reports patterns allowing potential prediction. Said no trial has acted on a circulating tumour DNA result, and called lymphoma the field where the test works best. (Second accuracy pass; abstract for DOI 10.1126/scitranslmed.aai8545 read through PubMed and a ClinicalTrials.gov search.) · Three statements reworded to the source and the registry.
- 2026-11-17 · The drugs that cure lymphoma, and the places that do not have them: Said rituximab is thirty years old. It was first approved on 26 November 1997. The supply statements carry no source. (Second accuracy pass; the openFDA approval record for BLA 103705.) · Summary line gives the approval year. A sentence says the supply statements were not re-read against a source.
- 2026-11-17 · Poiesz 1980: the first human retrovirus: Two findings were stated without the paper's qualifiers (the magnesium preference held under certain assay conditions; only some protein sizes differed from most primate retroviruses), and the leukaemia was placed in southern Japan and the Caribbean with no source for the Caribbean. (Second accuracy pass; abstracts for DOI 10.1073/pnas.77.12.7415, 10.1073/pnas.78.10.6476 and 10.1002/ijc.2910300302 read through PubMed.) · Qualifiers restored here and on the roadmap, southwestern Japan substituted, and the 1982 Caribbean paper linked.
- 2026-11-17 · Non-Hodgkin lymphoma: geography: The source for Romania as the only endemic focus of HTLV-1 in Europe carried PubMed id 28282882, which opens a paper on citrus flavanones and mouse bone, and named the wrong authors and year (Second accuracy pass; all 77 PubMed ids in the file resolved through PubMed esummary and the titles compared) · Replaced with Cassar and colleagues, PLOS Neglected Tropical Diseases 2024, PubMed id 39106239, whose abstract was read and says both things the page says
- 2026-11-17 · Non-Hodgkin lymphoma: geography: Rituximab for Burkitt lymphoma in East Africa was called the one randomised answer to the treatment gap; the study was not randomised (Second accuracy pass; the abstract and full text for PubMed id 39983056 read through PubMed and Europe PMC, which describe a prospective nonrandomized observational study in patients aged 25 or under) · Both passages now say the comparison was not randomised, give the confidence interval (0.24 to 0.98) and report 8 deaths against 16 rather than halved deaths
- 2026-11-17 · Non-Hodgkin lymphoma: geography: The 153-fold excess of central nervous system lymphoma in people with HIV was called the largest for any cancer in the study; Kaposi sarcoma was 498-fold. The prevention card also credited the diffuse large B-cell and central nervous system lymphoma declines to the 1990 to 2002 AIDS study (Second accuracy pass; the full text of the registry-linkage study (PubMed id 28803888, PMC5669995) and the abstract for PubMed id 16868446 read) · Now the largest for any lymphoma and second to Kaposi sarcoma; each trend is credited to the study that reported it
- 2026-11-17 · Non-Hodgkin lymphoma: geography: Both landmark papers on the t(11;18) translocation were said to contain translocation-positive patients who responded to Helicobacter eradication; the 2001 paper found none, and the two in the 2002 paper later relapsed. The translocation was also said to be commoner in Asia, which the cited review says of MALT lymphoma, not of the translocation. The 2002 paper was listed under another paper's title (Second accuracy pass; abstracts for PubMed ids 11197361 and 11984515 and the full text of the 2023 review (PMC10302523) read) · Sentence rewritten to the 2002 paper's two patients and their relapse; the Asia sentence now describes MALT lymphoma; the title is corrected
- 2026-11-17 · Non-Hodgkin lymphoma: geography: A sentence about high and low socio-demographic index regions was presented as a finding on non-Hodgkin lymphoma and its source was labelled as a non-Hodgkin lymphoma paper; PubMed id 40917004 is a Global Burden of Disease analysis of Hodgkin lymphoma (Second accuracy pass; the title and abstract for PubMed id 40917004 read through PubMed) · The sentence now says Hodgkin lymphoma and the source carries its real title
- 2026-11-17 · Non-Hodgkin lymphoma: geography: Six of the Hodgkin lymphoma incidence top ten were said to change between GLOBOCAN 2022 and 2024; seven do (Second accuracy pass; the 2022 and 2024 factsheet endpoints of the IARC Cancer Today API read for every country, where only Italy, Malta and Lebanon appear in both top tens) · Corrected to seven. Every other GLOBOCAN figure, rank and ratio on the page matched
- 2026-11-17 · Non-Hodgkin lymphoma: geography: About 95 percent of mother-to-child transmission of HTLV-1 was said to come from prolonged breastfeeding; the review says more than 90 percent of cases involve breastfeeding, and 95 percent is the share exclusive formula feeding prevents. The lifetime risk of adult T-cell leukaemia/lymphoma was traced to studies of 1989, 1990 and 2000; the review cites two 1989 studies and one from 2000 (Second accuracy pass; full texts of PubMed ids 37108125 (PMC10138424) and 32547527 (PMC7273189) read through Europe PMC) · Both breastfeeding sentences and both mentions of the study years corrected
- 2026-11-17 · Non-Hodgkin lymphoma: geography: Universal availability of childhood cancer medicines at 9 to 46 percent was credited to low-income countries; the survey grouped low- and lower-middle-income countries, and its source was listed under an invented title. The WHO essential medicines list was said to name five lymphoma indications; it also names small lymphocytic lymphoma. The developing-world share of NK/T-cell lymphoma was set against a much lower developed-world share that the abstract does not give (Second accuracy pass; the abstracts for PubMed ids 35749676 and 27354024 read through PubMed, and the WHO Model List, 24th list, read as text from the PDF) · Income group, title, indication list and comparison wording corrected
- 2026-11-17 · Non-Hodgkin lymphoma: geography: The statement that multicentric Castleman disease carries about fifteen times the expected lymphoma incidence had no source among those cited. Three figures gave a publication year as the period of the data: the Central Australian survey (2014 to 2018), the Japanese survival series (diagnosed 2000 to 2009) and the Peruvian series (1992 to 2023, shown as running to 2026) (Second accuracy pass; Oksenhendler and colleagues, Blood 2002, PubMed id 11895764, found and its abstract read; abstracts for PubMed ids 35903021, 26361794 and 41729993 read) · Source added and the sentence now names the cohort of 60; the three periods corrected; Japan's rates now say they use the world standard population
- 2026-11-17 · Watch and wait in follicular lymphoma: The record credited the 2014 trial of 379 people with showing that overall survival did not differ, a result neither of the trial's abstracts gives. It also said the charity finds that "many" people adapt by treating the lymphoma as a long-term condition and that relatives "often" find the approach harder than the patient, which the page does not say. The same survival wording was in two question sets and the follicular page note. (Second accuracy pass; the abstracts for DOI 10.1016/S1470-2045(14)70027-0 and DOI 10.1016/S2352-3026(25)00034-1 read through Europe PMC and PubMed, the abstract of the 2026 meta-analysis DOI 10.1016/j.clml.2026.09.009 read through PubMed, and the Lymphoma Action active monitoring page read as raw text. The Lancet full text refused a script.) · Survival statement re-sourced to the 2026 pooled analysis of three studies and 1,590 patients (hazard ratio 0.89, 95 per cent confidence interval 0.70 to 1.12) and labelled as not re-read in the trial's full text. Added the three-year figures (88 against 46 per cent not needing new treatment) and the 15-year result (34 per cent of people watched still untreated, median 5.6 years to first treatment). Charity wording restored to "some people" and "might also find", and the helpline hours added.
- 2026-11-17 · Transplant or CAR-T at second line in diffuse large B-cell lymphoma: Said Lymphoma Action describes the transplant route as several weeks in hospital, and that 74 per cent of ZUMA-7 participants had primary refractory disease or other high-risk features. The survival gain was given without its hazard ratio. (Second accuracy pass; the Lymphoma Action autologous transplant page read as raw text, and the abstract for DOI 10.1056/NEJMoa2301665 read through Europe PMC.) · Hospital stay corrected to most people going home two to three weeks after the transplant. The 74 per cent sentence reworded to the report's own description, and the hazard ratio for death (0.73, 0.54 to 0.98) added.
- 2026-11-17 · Escalated chemotherapy or ABVD in advanced Hodgkin lymphoma: The HD21 paragraph gave only the tolerability result and left out that the newer regimen also controlled the disease better, which a reader weighing the two needs. (Second accuracy pass; the abstract for DOI 10.1016/S0140-6736(24)01315-1 read through Europe PMC.) · Added four-year progression-free survival of 94.3 against 90.9 per cent (hazard ratio 0.66, 0.45 to 0.97) and four-year overall survival of 98.6 and 98.2 per cent at a median follow-up of 48 months.
- 2026-11-17 · A clinical trial or standard treatment in lymphoma: Said lymphoma has more trials open than almost any other cancer, with no source. Said Lymphoma TrialsLink has a free enquiry service, that the NHS page calls participation voluntary, and that many trials reimburse travel. None of the three pages cited says so. The first two were repeated in every question set. (Second accuracy pass; the Lymphoma Action taking part and TrialsLink pages and the NHS clinical trials page read as raw text.) · Comparative removed. TrialsLink described in the charity's words as a database and information service. The NHS statement reworded to what the page says: you can leave at any point without giving a reason and without it affecting your care. The travel sentence now points to the charity's suggested question about travel expenses.
- 2026-11-17 · Fatigue after lymphoma treatment: Credited Lymphoma Action with saying the fatigue is not relieved by sleep and continues for months or years, and with running support meetings about fatigue. The charity's page says neither, and its meeting list has no fatigue meeting. The same sentence was on the cancer-related fatigue record. (Second accuracy pass; the Lymphoma Action fatigue and support meetings pages and the Macmillan tiredness page read as raw text.) · The rest and sleep statement and the months or years statement are now credited to Macmillan, which does say them; the Lymphoma Action sentences keep to its own wording.
- 2026-11-17 · The surveillance schedule after lymphoma: Said many hospitals now offer patient-triggered follow-up. (Second accuracy pass; the Lymphoma Action follow-up page read as raw text, which says some hospitals offer a self-management and remote monitoring system, often known as patient-led follow-up.) · Reworded to some hospitals and the charity's term.
- 2026-11-17 · Going back to work after lymphoma: Credited Lymphoma Action's work page with the rule that cancer counts as a disability from diagnosis, which that page does not state, and treated the Equality Act as covering the whole United Kingdom. (Second accuracy pass; the Lymphoma Action work page, the GOV.UK definition of disability page and the Macmillan work and cancer page read as raw text.) · The rule is now credited to GOV.UK, with the link added, and the record notes that Northern Ireland is covered by the Disability Discrimination Act 1995.
- 2026-11-17 · Non-Hodgkin lymphoma: Several questions to ask credited a source with something it does not say: that the NHS carer's assessment does not depend on income or on how much care is given; that Lymphoma Action's list of questions starts with subtype and stage; that the pathologist tests for genetic changes; and that the charity's scan page explains why a residual shadow is not residual lymphoma. (Second accuracy pass; the NHS carer's assessments page and the Lymphoma Action questions, biopsy, diffuse large B-cell lymphoma and CT and PET/CT pages read as raw text.) · Each reason reworded to what its page says: the assessment is free and anyone over 18 can ask the council for one; the charity publishes a list of questions; the sample is tested for particular proteins on the lymphoma cells; scans wait some weeks after treatment because activity can linger where the lymphoma was.
- 2026-11-17 · Immune effector cell-associated neurotoxicity: The lymphoma note said the rule that another adult must be present for four weeks after CAR-T exists because the person affected is often the last to notice, and credited the passage to Lymphoma Action. (Second accuracy pass; the Lymphoma Action CAR-T cell therapy page read as raw text, which gives the possibility of delayed side effects as the reason.) · Reworded to the charity's stated reason.
- 2026-11-17 · Peer support groups: The lymphoma note listed in-person support meetings and meetings on fatigue and stem cell transplants; the watchful waiting note said Lymphoma Action calls active monitoring inappropriate for any aggressive lymphoma. (Second accuracy pass; the Lymphoma Action support meetings, connect with others and active monitoring pages read as raw text.) · Meetings described as online, with the topics the charity lists (active monitoring, relapsed and refractory lymphoma, CAR-T, late effects). The unsupported clause on aggressive lymphoma removed.
- 2026-11-17 · Lymphoma in the UK: pixantrone funding row: The table said England funds pixantrone under NICE TA306 and Scotland does not, and listed it as a drug funded in England but not Scotland. NICE has withdrawn TA306. (Second accuracy pass; the TA306 page read as raw text from nice.org.uk, which now reads "This guidance has been withdrawn because pixantrone (Pixuvri) no longer has a marketing authorisation", with archive.org copies showing the recommendation still live on 29 May 2026.) · Row, its Wales and Northern Ireland cells, its note and the four-nations comparison rewritten to say the guidance is withdrawn.
- 2026-11-17 · Lymphoma in the UK: CAR-T travel rule: The page said that after CAR-T people should stay within about an hour's drive of the centre for about four weeks after discharge, unchanged since 2018, and credited this to the NHS England specification. (Second accuracy pass; NHS England service specification 2101 (PDF, 14 April 2026) read as text, which says approximately 2 hours' travel time for a minimum of 2 weeks after discharge, with 2 to 4 weeks recommended, and a carer available 24 hours a day for 4 weeks after infusion; the Lymphoma Action page read as raw text.) · Rewritten in the timeline and the travel-costs row to give the specification's figures and Lymphoma Action's separately.
- 2026-11-17 · Lymphoma in the UK: Scottish decisions: Three Scottish statuses were stale or missing: lisocabtagene maraleucel at second line was shown as not yet published, acalabrutinib in mantle cell lymphoma as having no date, and the page said no Scottish advice existed for brentuximab vedotin in untreated systemic anaplastic large cell lymphoma. Axicabtagene ciloleucel in follicular lymphoma was described as refused by the SMC and as for two or more lines. (Second accuracy pass; each advice page read as raw text from scottishmedicines.org.uk: SMC2909 accepted 9 November 2026, SMC2929 due 8 February 2027, SMC2310 accepted 18 January 2021, SMC2646 a non-submission for three or more lines; NICE TA894 read for the same population.) · All four corrected; a Scotland cell added for SMC2310, and legacy advice 845/12 and SMC2085 named in the Hodgkin row.
- 2026-11-17 · Lymphoma in the UK: the only CAR-T for mantle cell lymphoma: Brexucabtagene autoleucel was called the only CAR-T licensed for mantle cell lymphoma. (Second accuracy pass; the United States Breyanzi label (effective 20 February 2026) read through openFDA, which carries a relapsed or refractory mantle cell lymphoma indication.) · Reworded to the only CAR-T with a NICE or SMC decision on the page, with the United States position stated.
- 2026-11-17 · Lymphoma in the UK: waiting times: The opening text and a figure note called 59.8 percent the lowest of any named suspected-cancer category in July 2026, although suspected acute leukaemia was lower. July was presented as the latest month. Northern Ireland's 62-day haematological figure was said to be suppressed under a 20-patient threshold. (Second accuracy pass; the NHS England July and September 2026 monthly workbooks, the Public Health Scotland table and the Northern Ireland workbook read cell by cell. Northern Ireland's Table 5 publishes the counts (16 treated, 9 within 62 days) and its 20-patient rule applies to the 95th percentile wait.) · Wording corrected, September 2026 figures added with their source, Northern Ireland reworded in five places, and the Welsh figures labelled as not re-read because gov.wales refused a scripted request.
- 2026-11-17 · Lymphoma in the UK: referral wording: The page said NICE NG12 asks for a very urgent referral for children and young people, in contrast to consider for adults, and that the same symptom could carry no standard at all. A sentence said a fine-needle aspirate is not offered at all, and the diagnostic service requirements were credited to NG47 recommendations 1.1.1 to 1.1.5. (Second accuracy pass; the NG12, NG52 and NG47 recommendation chapters read as raw text from nice.org.uk. Recommendations 1.10.7 and 1.10.9 both begin "Consider a very urgent referral"; NG52 does not mention fine-needle aspiration; the requirements are in NG47 1.1.2.) · Reworded in the presentation and timeline sections.
- 2026-11-17 · Lymphoma in the UK: trials and who sponsored them: RADAR was shown with the registry number of RATHL. University College London Hospitals was called the sponsor of FoRT, RAPID and RADAR and University Hospitals Birmingham the sponsor of BREVITY. GALLIUM was said to have the UK as one of 183 sites. FoRT's two enrolment figures were said to be unreconciled, and RATHL's bleomycin-free arm was said not to have done worse. (Second accuracy pass; ClinicalTrials.gov records NCT04685616, NCT00678327, NCT00943423, NCT01332968 and NCT00310167 and ISRCTN records read through their APIs; the RATHL abstract (DOI 10.1056/NEJMoa1510093) and the FoRT long-term abstract (DOI 10.1016/S1470-2045(20)30686-0) read through Europe PMC.) · RADAR given NCT04685616; sponsors corrected to University College London and the University of Birmingham; GALLIUM corrected to 29 of 183 sites; FoRT explained as 614 target sites in 548 patients; RATHL given its figures and the authors' words that it fell just short of the noninferiority margin.
- 2026-11-17 · Lymphoma in the UK: Hodgkin lymphoma figure notes: Notes said a man of 80 with Hodgkin lymphoma has a worse ten-year outlook than most people with lung cancer, that Hodgkin lymphoma is the cancer with the best long-term survival in the UK, and that its preventable cases are largely down to excess body weight and infection. (Second accuracy pass; Cancer Research UK statistics pages read as raw HTML: lung cancer ten-year survival 11.1 percent against 18.4 percent for men aged 75 to 99 with Hodgkin lymphoma, testicular cancer 96.5 percent against 81.6 percent, and 40 percent of Hodgkin lymphoma cases attributed to Epstein-Barr virus.) · The two comparisons removed and the cause corrected to Epstein-Barr virus.
- 2026-11-17 · Lymphoma in the UK: terminated appraisals and other counts: The funding note counted seven terminated appraisals but named eight. Four CAR-T centres were called the newest with no source. Children's CAR-T was given as under 18. Marginal zone zanubrutinib was called the first targeted option funded across the UK. Maggie's was given 29 centres and ClinicalTrials.gov 102 recruiting UK lymphoma studies. (Second accuracy pass; NICE pages for TA608, TA717, TA774, TA842, TA933, TA978, TA987, TA1186 and TA1001, the NHS England CAR-T commissioning guidance PDF, maggies.org and the ClinicalTrials.gov API read as raw text.) · Recounted as eight with their numbers; unsourced phrases removed; ages corrected to 18 and under; Maggie's corrected to 30; the trial count dated and corrected to 101. The Cancer Drugs Fund sentence now cites list version 1.410 of 12 November 2026.
- 2026-11-17 · Follicular lymphoma: The EZH2 row of the lymphoma molecular layer said tazemetostat is licensed for relapsed follicular lymphoma and called it one of two genotype-selected drug choices. (Second accuracy pass; the FDA table of withdrawn cancer accelerated approvals read as raw HTML from fda.gov, the openFDA record for NDA 211723, and the phase 2 abstract (DOI 10.1016/S1470-2045(20)30441-1) read through Europe PMC.) · Stated that the accelerated approval of 18 June 2020 is listed as withdrawn on 22 June 2026 and that no United States approval now depends on an EZH2 result. Added the trial response rates, 69% of 45 mutant against 35% of 54 wild-type, with their source.
- 2026-11-17 · MALT lymphoma: The t(11;18) row said 47 of the 48 patients whose gastric lymphoma regressed completely were negative for the API2-MALT1 transcript. In the source 47 of 48 is the number at stage IE. (Second accuracy pass; the abstract for DOI 10.1053/gast.2002.33047 read through Europe PMC.) · Replaced with the source figures: the translocation was found in 2 of the 48 complete responders, both of whom later relapsed, and in 42 of the 63 who did not respond.
- 2026-11-17 · Diffuse large B-cell lymphoma: The genetic subtype row said no randomised trial has yet assigned treatment by genetic subtype. The cell of origin row said every randomised attempt to act on cell of origin in first line had failed to change survival. (Second accuracy pass; the abstracts for GUIDANCE-01 (DOI 10.1016/j.ccell.2023.09.004), the PHOENIX subtype analysis (DOI 10.1016/j.ccell.2021.10.006) and the REMoDL-B five-year follow-up (DOI 10.1200/JCO.23.00033) read through Europe PMC.) · Added GUIDANCE-01, a randomised phase 2 trial of 128 patients with complete response 88% against 66%. Reworded the cell of origin line and added the REMoDL-B five-year result: no overall benefit, and in activated B-cell-like disease five-year survival 80% against 67%, hazard ratio 0.58, 95% CI 0.35 to 0.95.
- 2026-11-17 · Diffuse large B-cell lymphoma: Three rows gave the WHO fifth edition double-hit entity a shortened name and one said a double hit or a triple hit is a separate WHO entity without saying that MYC with BCL6 alone is not. Two rows called the 442 RICOVER cases unselected. The Hans row said all 152 cases had a microarray result. (Second accuracy pass; the WHO fifth edition full text (PMC9214472), the abstracts for DOI 10.1182/blood-2012-06-435842 and 10.1182/blood-2003-05-1545 read through Europe PMC, and the registry record NCT00052936.) · Used the full entity name, diffuse large B-cell lymphoma/high-grade B-cell lymphoma with MYC and BCL2 rearrangements, and stated that MYC with BCL6 alone is excluded. Described the 442 cases as de novo lymphomas from a trial of patients aged 61 to 80. Stated that 142 of the 152 Hans cases had a microarray result.
- 2026-11-17 · Non-Hodgkin lymphoma: Resistance rows quoted figures from chronic lymphocytic leukaemia studies without naming the population, including a 73.3% response to pirtobrutinib in 247 patients. The Epstein-Barr virus row said virus-specific T cells are licensed for post-transplant disease after transplant failure. (Second accuracy pass; the abstracts for DOI 10.1056/NEJMoa2300696, 10.1056/NEJMoa1400029, 10.1182/bloodadvances.2022008325 and 10.1158/2159-8290.CD-18-1119 read through Europe PMC, and the EMA page for Ebvallo read as raw HTML.) · Named the population in each resistance row: chronic lymphocytic leukaemia, or for pirtobrutinib chronic lymphocytic leukaemia or small lymphocytic lymphoma. Replaced the licence line with the European Union indication for tabelecleucel: relapsed or refractory disease after at least one prior therapy.
- 2026-11-17 · Angioimmunoblastic T-cell lymphoma: The RHOA row said azacitidine-containing regimens have shown activity in T-follicular-helper lymphoma and did not mention the randomised phase 3 trial that followed. (Second accuracy pass; the ORACLE abstract (DOI 10.1016/S2352-3026(24)00102-9) read through Europe PMC.) · Added that ORACLE, 86 patients, did not meet its primary endpoint: median progression-free survival 5.6 against 2.8 months, hazard ratio 0.63, 95% CI 0.38 to 1.07.
- 2026-11-17 · EZH2 gain-of-function mutation (Tyr646, originally Tyr641): The page said the mutation is the one lymphoma mutation that currently selects a tablet and that the result can open treatment with tazemetostat. The FDA lists the follicular lymphoma accelerated approval of 18 June 2020 as withdrawn on 22 June 2026. (Second accuracy pass; the FDA page of withdrawn cancer accelerated approvals read as raw HTML, and the Tazverik label and approval history read through openFDA.) · Summary line and patient note rewritten to say the United States approval was withdrawn on 22 June 2026; the label's response rates (69% of 42 with the mutation, 34% of 53 without) written in place of an unsourced comparison; FDA links added.
- 2026-11-17 · The germinal centre reaction: Tazemetostat was listed as the only licensed drug acting on the reaction. Its United States follicular lymphoma approval is listed as withdrawn on 22 June 2026. (Second accuracy pass; the FDA page of withdrawn cancer accelerated approvals read as raw HTML.) · Intervention reworded to the past tense with the withdrawal date; FDA link added.
- 2026-11-17 · LymphGen and the genetic clusters of large B-cell lymphoma: The page said no randomised trial has assigned treatment by genetic subtype. GUIDANCE-01, a randomised phase 2 trial of 128 patients published in 2023, did so and reported complete response in 88% against 66%. The share left unclassified and the BTK inhibitor subgroup finding carried no figure or source. (Second accuracy pass; the abstracts for DOI 10.1016/j.ccell.2023.09.004 and 10.1016/j.ccell.2021.10.006 read through Europe PMC and the Wright 2020 full text read through NCBI PMC.) · Sentence replaced with the GUIDANCE-01 result and its limits; 63.1% classified of 574 (against 46.6% in 2018) and the PHOENIX subgroup figures added with their sources.
- 2026-11-17 · Cell of origin in practice: Hans against expression profiling, and what it changes: The page said every first-line randomised attempt had failed without the later REMoDL-B report: at five years there was still no overall benefit, but overall survival in the activated group was 80% with bortezomib against 67% without (hazard ratio 0.58, 95% CI 0.35 to 0.95). (Second accuracy pass; the abstract for DOI 10.1200/JCO.23.00033 read through Europe PMC.) · Sentence and source added, stating it is a subgroup finding in a trial whose main comparison was negative.
- 2026-11-17 · BTK resistance mutations: C481S, and L528W and T474I after the non-covalent inhibitors: The pirtobrutinib result (73.3% response in 247 patients, median progression-free survival 19.6 months) was given without its population. Those patients had chronic lymphocytic leukaemia or small lymphocytic lymphoma, and the page is filed under mantle cell, marginal zone and Waldenstrom lymphoma. T474I had no source. (Second accuracy pass; the abstracts for DOI 10.1056/NEJMoa2300696, 10.1056/NEJMoa1400029 and 10.1056/NEJMoa2114110 read through Europe PMC.) · Population stated for each study; the Wang 2022 resistance series (9 of 55 patients, mutations including T474I and L528W) added with its link.
- 2026-11-17 · BCL2 G101V and the other venetoclax binding-site mutations: The 7 of 15 paired patients were described without saying they had chronic lymphocytic leukaemia, on a page filed under lymphoma. (Second accuracy pass; the abstract for DOI 10.1158/2159-8290.CD-18-1119 read through Europe PMC.) · Population added.
- 2026-11-17 · Double-hit and triple-hit: MYC with BCL2 and BCL6 rearrangement: The summary line defined a double hit as MYC with BCL2 or BCL6, and the WHO entity was misnamed. The WHO fifth edition entity is diffuse large B-cell lymphoma/high-grade B-cell lymphoma with MYC and BCL2 rearrangements, and it excludes cases with MYC and BCL6 rearrangements only. (Second accuracy pass; the full text for DOI 10.1038/s41375-022-01620-2 read through Europe PMC.) · Summary line and the WHO sentence corrected to the entity's name and its exclusion.
- 2026-11-17 · Epstein-Barr virus latency programmes, and why they decide which lymphoma: The page said EBV-specific T cells are licensed, with no source and no jurisdiction. Tabelecleucel is authorised in the European Union (16 December 2022) after at least one prior therapy; openFDA holds no United States label. (Second accuracy pass; the EMA medicine page for Ebvallo read as raw HTML and openFDA queried.) · Sentence reworded to name the medicine and the European Union; EMA link added.
- 2026-11-17 · Cytogenetics and FISH: The lymphoma note, and the MALT lymphoma page, said 47 of the 48 patients whose gastric MALT lymphoma regressed completely after Helicobacter eradication were negative for the API2-MALT1 transcript. 47 of 48 is the number at stage IE. (Second accuracy pass; the abstract for DOI 10.1053/gast.2002.33047 read through Europe PMC, which finds the translocation in 2 of 48 complete-regression cases and in 42 of the 63 non-responders.) · Corrected to 46 of 48 negative, with the two positive cases relapsing and 42 of 63 non-responders positive, on the technology note and on the MALT lymphoma page.
- 2026-11-17 · Tazemetostat: The lymphoma notes on tazemetostat, the FLIPI term, genomic profiling, follicular lymphoma and non-Hodgkin lymphoma said an EZH2 mutation in relapsed follicular lymphoma opens tazemetostat, called it the only genotype-selected tablet in B-cell lymphoma, and credited the gain-of-function finding to Morin 2010, whose abstract reports reduced enzyme activity. (Second accuracy pass; the FDA page of withdrawn cancer accelerated approvals read as raw text, which lists both Tazverik follicular lymphoma indications as withdrawn on 22 June 2026, and the abstracts for DOI 10.1038/ng.518 and DOI 10.1073/pnas.1012525107 read through Europe PMC.) · Each passage now states the withdrawal and its date, the superlative is removed, and the change in substrate preference is credited to Sneeringer 2010 with a link.
- 2026-11-17 · Non-Hodgkin lymphoma: The molecular section said three times that no randomised trial has assigned treatment by a genetic subtype of large B-cell lymphoma. (Second accuracy pass; the abstract for DOI 10.1016/j.ccell.2023.09.004 read through Europe PMC: GUIDANCE-01 randomised 128 patients to R-CHOP or R-CHOP plus an agent chosen by genetic subtyping, with complete response in 88% against 66%.) · The state-of-the-art line, the open problem and the 2020 history entry now name GUIDANCE-01 as the one randomised phase 2 trial, and the five-year REMoDL-B subgroup result is added beside the cell-of-origin statement.
- 2026-11-17 · Non-Hodgkin lymphoma: Resistance figures were given without their disease: the BTK C481S count, the pirtobrutinib response of 73.3% in 247 patients and the BCL2 G101V count are all from chronic lymphocytic leukaemia, and the CD19 escape mechanisms from acute lymphoblastic leukaemia. The ibrutinib note said resistance is on-target in most patients, which the six-patient source cannot carry. (Second accuracy pass; the abstracts for DOI 10.1056/NEJMoa1400029, DOI 10.1056/NEJMoa2300696, DOI 10.1158/2159-8290.CD-18-1119 and DOI 10.1158/2159-8290.CD-15-1020 read through Europe PMC.) · The disease is now named on the resistance term, the pirtobrutinib, venetoclax and ibrutinib notes, the resistance paragraph and three history entries, and the ibrutinib note gives the counts of five and two of six.
- 2026-11-17 · Non-Hodgkin lymphoma: The page said CD20 is not internalised and so carries no conjugate, and the polatuzumab note said CD20 cannot carry a payload into the cell. (Second accuracy pass; the abstract for PubMed id 21768293 read through PubMed, which reports internalisation of rituximab bound to CD20, most evident in chronic lymphocytic leukaemia and mantle cell lymphoma and varying between patients.) · Both passages now say CD20 is not reliably internalised and give the finding, with the paper linked.
- 2026-11-17 · ctDNA monitoring in lymphoma (PhasED-seq, clonoSEQ): The note said no randomised trial has yet acted on a circulating tumour DNA result in lymphoma. (Second accuracy pass; the ClinicalTrials.gov record for NCT06500273 read through the registry API on 17 November 2026: ALPHA3 is a randomised, recruiting trial that enrols patients with residual disease found by PhasED-Seq after first-line treatment.) · The note and the liquid biopsy note now name the trial, its status and that it has not reported.
- 2026-11-17 · Double-hit / high-grade B-cell lymphoma: The term and the 2022 history entry said the WHO fifth edition made high-grade B-cell lymphoma with MYC and BCL2 rearrangements an entity of its own. Two history entries carried unsourced priority claims, and the mediastinal lymphoma page called its CIITA rate the highest of any lymphoma. (Second accuracy pass; the full text for DOI 10.1038/s41375-022-01620-2 read through Europe PMC and for DOI 10.1038/nature09754 through NCBI PMC.) · The WHO change is now described as a renaming that moves MYC and BCL6 cases out, the two priority claims are removed, and the CIITA rate is set beside 15% in classical Hodgkin lymphoma and 3% in diffuse large B-cell lymphoma from the same series.
- 2026-11-17 · Primary CNS lymphoma: The page said primary CNS lymphoma sits almost entirely in one genetic subtype, with no source. (Second accuracy pass; the full text for DOI 10.1016/j.ccell.2020.03.015 read through NCBI PMC, which says primary extranodal lymphomas frequently acquire MYD88 L265P or CD79B mutations and other MCD-defining mutations.) · Reworded to the source's statement and the paper linked.
- 2026-11-17 · High-grade B-cell lymphoma with MYC and BCL2 rearrangements: The page and the diffuse large B-cell lymphoma treatment row said there has never been a randomised trial in double-hit lymphoma. One reported in 2026. (Second accuracy pass; the abstract of Alliance A051701 (Lancet Haematology 2026, DOI 10.1016/S2352-3026(26)00191-2) read through PubMed.) · Both now say that the one randomised trial added venetoclax to dose-adjusted EPOCH-R in 73 patients, did not lengthen progression-free survival (hazard ratio 1.13, 0.53 to 2.37) and closed early after six deaths on treatment against one, and that no randomised trial has compared an intensified regimen with R-CHOP.
- 2026-11-17 · EORTC/LYSA/FIL H10: The trial record, the roadmap, the Hodgkin history and two treatment rows said that intensifying treatment after a positive interim scan works, on the five-year figures alone (77.4 to 90.6 per cent, hazard ratio 0.42). (Second accuracy pass; the abstract of the ten-year follow-up (Federico 2024, DOI 10.1200/JCO.23.01745) and of the 2017 report read through PubMed.) · Every statement now adds that at the preplanned ten-year analysis the difference was no longer statistically significant (hazard ratio 0.67, 0.37 to 1.20), while the loss of tumour control when radiotherapy was omitted after a negative scan held. The follow-up paper is linked from the trial.
- 2026-11-17 · LyMa: The trial summary line said rituximab maintenance after transplant helped people with mantle cell lymphoma live longer, and the roadmap, the mantle cell treatment row and the maintenance glossary entry repeated an overall survival benefit. (Second accuracy pass; the abstract of the seven-year follow-up (Sarkozy 2024, DOI 10.1200/JCO.23.01586) read through PubMed and the 2017 abstract through Europe PMC.) · All four now say the survival advantage seen at four years (89 against 80 per cent) was no longer statistically significant at seven (83.2 against 72.2 per cent, hazard ratio 0.63, 0.37 to 1.08), while seven-year progression-free survival stayed 78.5 against 47.4 per cent.
- 2026-11-17 · TRIANGLE: The record stopped at the 31-month report, said the comparison of ibrutinib with and without transplant was continuing, and gave no overall survival result. (Second accuracy pass; the abstract of the 4.5-year report (Lancet 2026, DOI 10.1016/S0140-6736(26)00362-4) read through PubMed.) · Added the result at a median of 54.9 months: transplant added nothing to ibrutinib (four-year failure-free survival 82 against 81 per cent, hazard ratio 0.86), and four-year overall survival was 88 and 90 per cent in the two ibrutinib arms against 81 per cent with transplant alone (hazard ratios 0.59 and 0.57).
- 2026-11-17 · SYMPATICO: The open-label group was described as untreated patients with a TP53 mutation, and the 69.2 per cent complete response rate as a result in TP53-mutated disease. No survival result was given. (Second accuracy pass; the registry results for NCT03112174 read through the ClinicalTrials.gov API and the full text of the final analysis (British Journal of Haematology 2026, DOI 10.1111/bjh.70620) through PubMed Central.) · The group is now described as previously untreated patients aged 65 or over or with a TP53 mutation, the 69.2 per cent as the rate among all 78 enrolled, and the record says overall survival was not significantly different (hazard ratio 0.83, 0.60 to 1.15).
- 2026-11-17 · SWOG S1826: Peripheral neuropathy of any grade was given as 28.1 against 54.2 per cent. (Second accuracy pass; the full text of the New England Journal of Medicine report (DOI 10.1056/NEJMoa2405888) read through PubMed Central.) · Corrected to peripheral sensory neuropathy in 139 of 482 (29 per cent) with nivolumab and AVD against 266 of 476 (56 per cent) with brentuximab vedotin and AVD.
- 2026-11-17 · Hodgkin lymphoma: The history said 48.5 per cent of 3,905 survivors developed a second cancer and 50 per cent developed cardiovascular disease within 40 years, as if these were counts, and an open problem said every attempt to omit radiotherapy had cost tumour control. (Second accuracy pass; the abstracts of Schaapveld 2015, van Nimwegen 2015, HD16 and H10 read through PubMed.) · Both figures are now given as 40-year cumulative incidences in five-year survivors, the lack of improvement between treatment periods is tied to second solid cancers, and the open problem speaks of the two large trials cited.
- 2026-11-17 · Non-Hodgkin lymphoma: The open problems said no genetic classification had reached a randomised trial of treatment, that a blood test flags transformation before it shows, that T-cell lymphomas have almost no randomised evidence and rest on a trial opened in 1998 and a 38-patient study, and that rituximab is thirty years old. (Second accuracy pass; the abstracts of GUIDANCE-01, Scherer 2016, JCOG9801, the 2015 mogamulizumab phase 2 and the two randomised NK/T-cell trials read through PubMed, and the rituximab approval record through openFDA.) · The lines now name GUIDANCE-01 (128 patients, complete response 88 against 66 per cent), say the blood test shows patterns that may predict transformation, list the randomised T-cell trials that exist, and date rituximab to its 1997 approval.
- 2026-11-17 · ECHELON-2 in the T-cell lymphoma pages: ECHELON-2 was called the only randomised first-line trial in the T-cell family to improve survival, and three-quarters of its patients were said to have systemic anaplastic large cell lymphoma. (Second accuracy pass; the abstract of the DDGP against SMILE trial (JAMA Oncology 2022) read through PubMed and the ECHELON-2 five-year full text through PubMed Central.) · Now one of very few such trials, since a trial in advanced NK/T-cell lymphoma also improved survival (hazard ratio 0.41), and the share with systemic anaplastic large cell lymphoma is 70 per cent (316 of 452).
- 2026-11-17 · Peripheral T-cell lymphoma: NK/T-cell and adult T-cell rows: The NK/T-cell row said CHOP must not be used and every effective regimen contains asparaginase; the adult T-cell leukaemia/lymphoma row said flatly that antiviral therapy does not work in the lymphoma type. (Second accuracy pass; the NCI PDQ peripheral T-cell lymphoma summary read as raw text from cancer.gov and the 2009 international consensus report through PubMed Central.) · Reworded to what the sources say: anthracycline regimens work poorly and asparaginase is built in, and the 2009 consensus report found that people with the lymphoma type did not benefit from zidovudine and interferon.
- 2026-11-17 · Nodal and extranodal lymphoma: Said a marginal zone lymphoma in the stomach is usually cured by a fortnight of antibiotics. (Second accuracy pass; the abstract of Wundisch 2005 (DOI 10.1200/JCO.2005.02.3903) read through PubMed.) · Reworded to say that in early disease it often goes into lasting remission after a course of antibiotics; no source read gives a fortnight.
- 2026-11-17 · IELSG-19: Called the only randomised first-line systemic trial in MALT lymphoma, and the marginal zone treatment row said it was the only randomised evidence. (Second accuracy pass; the abstract (DOI 10.1200/JCO.2016.70.6994) read through PubMed.) · Changed to the authors' own claim, the first randomised trial of first-line treatment to their knowledge, and the word only removed from the treatment row.
- 2026-11-17 · FLYER: The summary line and the roadmap said roughly a quarter fewer side effects with four cycles. (Second accuracy pass; the abstract (DOI 10.1016/S0140-6736(19)33008-9) read through PubMed.) · Corrected to about a fifth fewer: 1,330 adverse events against 1,706.
- 2026-11-17 · TRANSCEND NHL 001: Called lisocabtagene maraleucel the third CAR-T product for lymphoma. (Second accuracy pass; the FDA product pages for Tecartus and Breyanzi read as raw text.) · Changed to the third for large B-cell lymphoma: a product for mantle cell lymphoma was approved on 24 July 2020, before this one on 5 February 2021.
- 2026-11-17 · LymphGen: Said the classifier needs a sequencing panel with copy number and fusion calling. (Second accuracy pass; the full text of Wright 2020 read through PubMed Central.) · Now says it needs mutation data, and that copy number and rearrangement data improve it but can be left out, as its authors designed.
- 2026-11-17 · MYC, BCL2 and BCL6 rearrangement rates in the lymphoma biology pages: Five passages gave the rates as coming from 442 unselected diffuse large B-cell lymphomas, and one said MYC with BCL2 or BCL6 makes a separate WHO entity. (Second accuracy pass; the abstract of Horn 2013 read through Europe PMC, the RICOVER registry record, and the WHO fifth edition paper through PubMed Central.) · The series is now described as 442 de novo cases from the RICOVER trial of patients aged 61 to 80, and the entity as needing a BCL2 rearrangement, with MYC and BCL6 alone excluded.
- 2026-11-17 · CD20: The target page and two lymphoma entries said CD20 is not internalised when an antibody binds it. (Second accuracy pass; the abstract of Lim 2011 (PubMed 21768293) read through PubMed.) · Changed to not reliably internalised: uptake of bound rituximab was most evident in chronic lymphocytic leukaemia and mantle cell lymphoma and uneven within each.
- 2026-11-17 · Circulating tumour DNA in lymphoma: Two entries said no randomised trial has changed treatment on a circulating tumour DNA result, without the trial now doing so. (Second accuracy pass; the registry record for NCT06500273 read through the ClinicalTrials.gov API.) · Now says none has reported, and names ALPHA3, a randomised trial that enrols on a residual disease result and is recruiting.
- 2026-11-17 · BTK inhibitor and venetoclax resistance figures: Figures from chronic lymphocytic leukaemia studies (7 of 15 patients, 73.3 per cent of 247, two of six) were quoted on lymphoma pages and in the resistance table without the disease. (Second accuracy pass; the abstracts of Blombery 2019, Mato 2023 and Woyach 2014 read through PubMed and Europe PMC.) · The population is now stated wherever the figures appear.
- 2026-11-17 · Ibrutinib and pembrolizumab in the UK coverage table: The ibrutinib row listed TA491 for Waldenstrom macroglobulinaemia as if funded, and the pembrolizumab row gave TA540 for Hodgkin lymphoma. (Second accuracy pass; the NICE pages for TA491, TA795, TA540, TA772 and TA967 read as raw text from nice.org.uk.) · Ibrutinib is now shown as not recommended in TA795, which replaced TA491, and pembrolizumab in Hodgkin lymphoma is credited to TA772 and TA967.
- 2026-11-17 · Lymphoma treatment in the UK and the United States: Two glossary entries said no NICE appraisal of nivolumab with AVD exists. (Second accuracy pass; the NICE in-development page GID-TA11908 read as raw text.) · Now says NICE selected the appraisal (ID6690) and suspended it on 19 February 2026 because the company will not make an evidence submission, so there is no recommendation.
- 2026-11-17 · Follicular lymphoma: first 60 days checklist: The checklist said overall survival did not differ in the 379-person trial of early rituximab, which neither trial report gives, and that Lymphoma Action runs a free enquiry service for trials. (Second accuracy pass; the abstract of the 2026 pooled analysis (DOI 10.1016/j.clml.2026.09.009) read through PubMed and the Lymphoma Action trials page read as raw text.) · The survival statement now cites the pooled analysis of three studies and 1,590 patients (hazard ratio 0.89, 0.70 to 1.12), and the service is described in the charity's words as a trial database and information service.
- 2026-11-17 · CAR-T red flag card for lymphoma: The card gave its own reason why another adult must be present for four weeks. (Second accuracy pass; the Lymphoma Action CAR-T page read as raw text.) · Replaced with the charity's stated reason, the possibility of delayed side effects.
- 2026-11-17 · Sepantronium bromide in high-grade B-cell lymphoma (NCT05263583): Listed as active with a primary completion date still to come. (Second accuracy pass; the registry record read through the ClinicalTrials.gov API.) · Status set to completed (14 enrolled, primary completion 12 February 2025, no results posted), with the plain-language and Chinese lines changed to match.
- 2026-11-17 · RADAR in the UK trials table: The RADAR row linked to the record for a different trial, RATHL. (Second accuracy pass; the registry record for NCT04685616 read through the ClinicalTrials.gov API.) · Linked to the RADAR record.
- 2026-11-17 · Breast implant-associated anaplastic large cell lymphoma: The breast glossary entry gave 13.16 years as the median time from implant to diagnosis. (Second accuracy pass; the abstract of the meta-analysis (DOI 10.1007/s00266-024-03956-9) read through PubMed.) · Changed to the mean, as the paper reports.
- 2026-11-17 · Maggie's centres on the lung cancer UK page: Said 29 centres across the UK. (Second accuracy pass; the Maggie's centres page read as raw text, listing 30 UK centres.) · Changed to 30.
- 2026-11-17 · PSA density: The term said NICE NG131 does not set a PSA density threshold and treats density only as something clinicians weigh. (Second accuracy pass; the NG131 recommendations page read as raw text from nice.org.uk, where recommendations 1.2.11 and 1.2.12 name a PSA density greater than 0.15 nanograms per millilitre per millilitre and a PSA velocity greater than 0.75 nanograms per millilitre a year.) · Rewritten to state the two figures, what they are used for (offering a biopsy after a Likert 1 or 2 scan, and the level at which primary care refers back) and the recommendation numbers.
- 2026-11-17 · Ductal adenocarcinoma of the prostate: The page said the NCCN and the Philadelphia consensus conference recommend germline genetic testing for anyone with ductal, intraductal or cribriform morphology whatever the stage, that ductal disease is treated as high-risk disease with no guideline of its own, and that metastatic disease is treated with an androgen receptor pathway inhibitor. None of the sources cited says any of the three. (Second accuracy pass; the full text of Ranasinha 2021 (DOI 10.1002/bco2.60) and Kench 2022 (DOI 10.1111/his.14711) read through Europe PMC, and the NCCN version 3.2024 criteria as stated in Xu 2025 (DOI 10.1002/pros.24918) read through NCBI PMC.) · Germline testing reworded to what the sources say: recommended by the NCCN in high-risk, regional or metastatic disease, to be considered in intermediate-risk disease with intraductal or cribriform histology, and ductal histology not named on its own. Treatment reworded to the review's own finding: the same methods as acinar cancer, no consensus on the best, no randomised trial. Xu 2025 added as a source.
- 2026-11-17 · Intraductal carcinoma of the prostate (IDC-P): The summary line said the NHS offers inherited-cancer gene testing because of this finding, and the text gave a 2019 recommendation as current. The same overstatement appeared on the prostate cancer page and in the grade group notes. (Second accuracy pass; the Royal College of Pathologists dataset G084 (PDF) and NICE NG131 read as raw text, neither of which mentions germline testing, and Kench 2022 and Xu 2025 read in full.) · Reworded to American guidelines treating it as a reason to consider testing, with the 2019 and 2024 NCCN positions distinguished. Added that the UK dataset recommends counting intraductal carcinoma in the Gleason score.
- 2026-11-17 · Extraprostatic extension and seminal vesicle invasion: Said the eighth edition of TNM added microscopic bladder neck involvement to T3a. (Second accuracy pass; section 5.3.10 of the Royal College of Pathologists dataset G084 read from the PDF, which says bladder neck invasion was pT4 in the 2002 TNM and that the seventh edition moved it to pT3a.) · Corrected to the seventh edition.
- 2026-11-17 · Intermediate-risk prostate cancer: Radiotherapy with six months of hormone therapy was credited to NG131 recommendation 1.3.28, which is the recommendation against high-intensity focused ultrasound and cryotherapy. (Second accuracy pass; the NG131 recommendations page read as raw text from nice.org.uk.) · Corrected to recommendations 1.3.21 and 1.3.22.
- 2026-11-17 · Cambridge Prognostic Group (CPG 1 to 5): A quotation from the Royal College of Pathologists dataset had been tidied, and NICE's reason for replacing its three-tier table was given as the table's inability to separate Gleason 3+4 from 4+3, which is the pathologists' point and not NICE's stated reason. (Second accuracy pass; the dataset PDF and the NG131 rationale text read as raw text.) · Quotation restored to the source's words. NICE's own reason added here and on the prostate cancer page: five-tier models predict death from prostate cancer better, and the Cambridge model has been tested in UK populations.
- 2026-11-17 · Prostate cancer: UK incidence, deaths, ten-year survival and US five-year survival were given without the years they apply to. The pathology report was said to give the tumour length in each core. Active surveillance was described as including routine repeat biopsy. (Second accuracy pass; the Cancer Research UK statistics page, SEER Stat Facts, the Royal College of Pathologists dataset and NICE NG131 table 2 read as raw text.) · Years added (cases 2019 and 2021 to 2022, deaths 2022 to 2024, survival 2018, US survival for diagnoses 2016 to 2022). Core item corrected to the longest length of tumour in any one core. Surveillance described as PSA, examination and MRI with a further biopsy if there is concern.
- 2026-11-17 · High-grade prostatic intraepithelial neoplasia (HGPIN): Said isolated HGPIN carries too low a risk to prompt a repeat biopsy, which none of the sources cited says. (Second accuracy pass; NICE NG131 and the Royal College of Pathologists dataset read as raw text.) · Reworded to what they say: the chance of cancer after a negative biopsy is slightly higher if it showed HGPIN, and HGPIN in several cores is a stronger risk factor than a single focus.
- 2026-11-17 · Metastatic hormone-sensitive prostate cancer: Said the ninth edition of TNM asks for a metastasis found on PSMA PET to be written M1(PET). The source gives N1(PET) and stage IV(PET) only. (Second accuracy pass; section 10.1 of Brierley 2026 (DOI 10.1002/ijc.70561) read in full through Europe PMC.) · Reworded to N1(PET) and stage IV(PET).
- 2026-11-17 · Neuroendocrine prostate cancer: The reasoning against routine neuroendocrine stains was credited to the UK dataset when it is the WHO position, and loss of PSA and NKX3.1 staining was presented as a reason the blood PSA stays low, which the dataset does not say. (Second accuracy pass; the Royal College of Pathologists dataset section 10.2 and Kench 2022 read in full.) · Each statement credited to the source that makes it, and the stain finding limited to what the dataset says: the two markers cannot be used to show where such a tumour came from.
- 2026-11-17 · MAGNITUDE: The page said the BRCA cohort "showed a survival benefit at update". The final analysis found no significant overall survival difference in the HRR-positive cohort (HR 0.93) or the BRCA1/2 subgroup (HR 0.79, 95% CI 0.55 to 1.12); only analyses adjusted for baseline factors favoured niraparib (Second accuracy pass; the abstract of the final overall survival analysis (PMID 40328571) and of the primary report (PMID 36952634) read through Europe PMC) · Summary and result now give the final survival figures and say the difference was not significant
- 2026-11-17 · AlphaBreak and AcTION: The page called AlphaBreak and AcTION "the phase 3 trials", gave an enrolment target of 600 and said both were recruiting. AlphaBreak is a randomised phase 2 with 110 men enrolled and no longer recruiting; AcTION is a phase 1 with 101 enrolled (Second accuracy pass; ClinicalTrials.gov API v2 for NCT06402331, NCT04597411, NCT06780670, NCT06855277 and NCT07611110) · Summary and short description rewritten from the registry, naming the phase 3 trials that followed (VECTRA-01, PSMAcTION, AcTFirst)
- 2026-11-17 · PSMAddition: The page said overall survival had not been reported. The registry posted results on 4 November 2026 with an overall survival hazard ratio of 0.84 (95% CI 0.63 to 1.13, p=0.125), not significant; the trial was also published in the Lancet in August 2026 (Second accuracy pass; the results section of NCT04720157 through the ClinicalTrials.gov API and the abstract of PMID 42561994 through Europe PMC) · Summary and result now carry the survival figure and say it is not significant
- 2026-11-17 · CAPItello-281: Overall survival was given only as immature (HR 0.90 at 26.4% maturity). An updated analysis with a data cut-off of 9 March 2026, at 42% maturity, reported HR 0.82 (95% CI 0.68 to 0.99), median 57.4 against 46.5 months, which the authors describe as descriptive (Second accuracy pass; the abstract of the updated survival analysis (PMID 42918424) read through Europe PMC; approval date from openFDA) · Summary and result updated, with the approval date of 12 June 2026
- 2026-11-17 · Prostate cancer: The share of men screened out by TheraP's PSMA and FDG PET criteria was given as 31%. The trial reports 80 of 291 men (27%) excluded after the scans. The screening line also gave only the 16-year ERSPC figure (Second accuracy pass; the abstracts of PMID 38043558 and PMID 41160819 read through Europe PMC) · 27% in both places; the screening line now gives 20% at 16 years, 13% at 23 years and the absolute reduction of 0.22%
- 2026-11-17 · CHAARTED: The result line mixed two data cuts: the overall hazard ratio of 0.61 is from the 2015 primary analysis, while the high-volume (0.63) and low-volume figures are from the 2018 analysis, where the overall hazard ratio was 0.72 (Second accuracy pass; the abstracts of PMID 26244877 and PMID 29384722 read through Europe PMC) · Result now gives each figure with its follow-up
- 2026-11-17 · STAMPEDE: The abiraterone hazard ratio of 0.63 was labelled as metastatic disease; it covers all 1,917 men, 52% of them metastatic, and the metastatic figure is 0.61. The page also said the radiotherapy and enzalutamide comparisons were read in subgroups defined after the fact; the metastatic burden analysis was prespecified in May 2018 before the data were examined (Second accuracy pass; the abstract of PMID 28578639 through Europe PMC and the full text of PMC6269599) · Result and the sentence on subgroups reworded from the papers
- 2026-11-17 · TheraP, Cabazitaxel: An overall survival hazard ratio of 0.97 was quoted; the 2024 report's abstract gives restricted mean survival of 19.1 against 19.6 months (p=0.77) and no hazard ratio, and the full text could not be read (Second accuracy pass; the abstract of PMID 38043558 read through Europe PMC) · Both records now give the restricted mean survival figures
- 2026-11-17 · Pasritamig, KLK2: The phase 3 trial was named KLK2-P3-01, which is not on the registry. There are two phase 3 trials: KLK2-comPAS (NCT07164443, started September 2025) and KLK2-PASenger with docetaxel (NCT07225946, started December 2025) (Second accuracy pass; a ClinicalTrials.gov API search for pasritamig and the abstract of PMID 40450573) · Both records name the registered trials; the phase 1 response is given as 14 of 33 (42%)
- 2026-11-17 · Xaluritamig, XALute, STEAP1: The phase 1 PSA50 of about 49% was described as the result at high or target doses and dated to ESMO 2024; 49% is across all dose cohorts and 59% at target doses, published in Cancer Discovery in 2024. XALute was given as about 675 men; the registry shows 707 enrolled (Second accuracy pass; the abstract of PMID 37861461 through Europe PMC and ClinicalTrials.gov API v2 for NCT06691984) · Figures corrected in all three records
- 2026-11-17 · MEVPRO-1, Mevrometostat: The trial was described as enrolling about 600 men. The registry shows 609 enrolled and the trial active but no longer recruiting (Second accuracy pass; ClinicalTrials.gov API v2 for NCT06551324 and the protocol paper (PMC13203012)) · Status and enrolment updated; the phase 1/2 figures are now exact (14.3 against 6.2 months, 81 men)
- 2026-11-17 · SPLASH, ECLIPSE: SPLASH was said to be sponsored by Lantheus; the registry's lead sponsor is POINT Biopharma, a subsidiary of Eli Lilly. Its interim survival figures, and ECLIPSE's positive result, rest on company statements that could not be re-read (Second accuracy pass; ClinicalTrials.gov API v2 results for NCT04647526 and the record for NCT05204927; the Lantheus release refused a scripted read) · Sponsor corrected, the registry's progression figures added (9.5 against 6.0 months, HR 0.71, 95% CI 0.55 to 0.92), and both records say what could not be confirmed
- 2026-11-17 · LATITUDE: The page said the trial met its primary overall survival endpoint; it had two primary endpoints, overall survival and radiographic progression-free survival, and met both (Second accuracy pass; the abstracts of PMID 28578607 and PMID 30987939 read through Europe PMC) · Summary and result reworded
- 2026-11-17 · ProtecT, Active surveillance: One record said about sixty percent of monitored men were eventually treated and the other said about half. The abstract gives neither; it reports that 24.4% of the monitoring group were alive without any prostate cancer treatment at the end of follow-up (Second accuracy pass; the abstract of PMID 36912538 read through Europe PMC; the full text is not open) · Both records now give the abstract's figure
- 2026-11-17 · Castration-resistant prostate cancer: The page said FDA label language had replaced "hormone-sensitive" since 2026. The Pluvicto, capivasertib and talazoparib labels use the new terms; the enzalutamide, apalutamide and darolutamide labels still say castration-sensitive (Second accuracy pass; current labels read through the openFDA label API) · Sentence reworded to say which labels use which terms
- 2026-11-17 · TALAPRO-2: The survival hazard ratio of 0.62 was shown beside the all-comer figure as if a subgroup of the same 805 men; it comes from a separate cohort of 399 men with HRR-deficient tumours. Among all comers without a known HRR alteration the hazard ratio was 0.88 and not significant (Second accuracy pass; the abstracts of PMID 40683290 and PMID 40683287 through Europe PMC and the Talzenna label through openFDA) · Result and summary now separate the cohorts and state that the US label still requires an HRR mutation
- 2026-11-17 · Number needed to screen (and number needed to diagnose): The page stopped at the 16-year report of the European screening trial and said only that the figures improve with time. A 23-year follow-up was published in October 2025: rate ratio 0.87 (95 percent confidence interval 0.80 to 0.95), one death prevented for every 456 men invited and every 12 men diagnosed (Second accuracy pass; the abstract for DOI 10.1056/nejmoa2503223 read through Europe PMC and PubMed. A correction to that paper published in March 2026 could not be read) · The summary, the short description and the links now carry the 23-year figures, say that the report restates its own 16-year figures as 628 and 18, and name the unread correction
- 2026-11-17 · Bipolar androgen therapy (BAT): RESTORE was quoted as 15 of 21 men (52 percent) responding to enzalutamide rechallenge. Fifteen of 21 is 71 percent; the 52 percent is the intention-to-treat figure over the 29 men who completed bipolar androgen therapy. The page also said both trials came from Johns Hopkins and both excluded men with more than five visceral sites or bone lesions at risk of fracture; TRANSFORMER was run at 17 United States centres and the exclusion is stated for RESTORE (Second accuracy pass; the RESTORE full text (PMC5875180) read through NCBI, and the TRANSFORMER abstract and full text for DOI 10.1200/jco.20.02759) · The denominator is stated, TRANSFORMER is described as a 17-centre trial in men who had progressed on abiraterone, and the exclusion is credited to RESTORE. The same denominator is corrected in the note added to PSA50
- 2026-11-17 · Metastasis-free survival (MFS): The page said the ninth edition of TNM asks for a PSMA PET finding to be recorded as M1(PET); the source gives the suffixes N1(PET) and stage IV(PET) and does not use M1(PET). It also said SPARTAN imaged with bone scan and computed tomography only (Second accuracy pass; the TNM ninth edition review (PMC13432304) read as full text, and the SPARTAN primary outcome description read from ClinicalTrials.gov NCT01946204) · The TNM sentence now gives the suffixes the source gives, and SPARTAN imaging reads bone scan and computed tomography or magnetic resonance imaging
- 2026-11-17 · PI-RADS (Prostate Imaging Reporting and Data System): A note said NICE does not set a PSA density threshold. NG131 names a density above 0.15 as an example of strong suspicion that leads to an offer of biopsy when the Likert score is 1 or 2. The definition of clinically significant cancer gave tumour volume as above 0.5 cubic centimetres where the document says 0.5 or more, and a 93 percent figure from PROMIS and a 65 percent figure from a surgical series were described as the same scan (Second accuracy pass; NICE NG131 recommendations 1.2.3, 1.2.11 and 1.2.12 read as raw text, the PI-RADS v2.1 document from the American College of Radiology, and the PROMIS and Johnson abstracts read through PubMed and Europe PMC) · The note now gives the NG131 density figure and its setting, the volume reads 0.5 cubic centimetres or more, and the two sensitivity figures are attributed to their own studies with their different scanner strengths and definitions. The same caution is added to whole-mount pathology
- 2026-11-17 · Neuroendocrine differentiation in prostate cancer: The reasons for not staining ordinary adenocarcinoma routinely (almost all show some neuroendocrine differentiation, and evidence of a therapeutic or prognostic role is insufficient) were credited to the Royal College of Pathologists dataset. They are the WHO fifth edition's reasons; the dataset says such differentiation in untreated adenocarcinoma is not clinically relevant (Second accuracy pass; Royal College of Pathologists dataset G084 section 10.2 and the full text of Kench 2022 (PMC9542779)) · Each reason is now credited to the source that gives it
- 2026-11-17 · Intermittent androgen deprivation (IAD): One sentence said SWOG 9346 randomised only men whose PSA fell below 4 nanograms per millilitre; the trial randomised men whose PSA fell to 4 or lower (Second accuracy pass; the abstract for DOI 10.1056/nejmoa1212299 read through PubMed) · Now reads 4 nanograms per millilitre or lower
- 2026-11-17 · PSA50 / PSA90 response: The added note said a PSA50 is confirmed by a second value at least four weeks later. The working group paper quotes the original definition as two measurements three to four weeks apart and uses three or more weeks for confirmation (Second accuracy pass; the PCWG2 full text (PMC4010133) for DOI 10.1200/jco.2007.12.4487) · The note now gives the working group's own interval
- 2026-11-17 · Prostate cancer: lutetium-177 PSMA therapy in England: The NICE summary, the PSMAfore and Pluvicto notes and the care row for hormone-relapsed metastatic disease named only TA930 and said no appraisal covered use before a taxane. NICE TA1179, for use after an anti-androgen but not a taxane, was terminated on 16 July 2026 because the company made no evidence submission (Second accuracy pass; the terminated guidance page for TA1179 and the recommendation chapters of 25 other prostate appraisals read as raw HTML from nice.org.uk, and the titles of TA996 to TA1207 scanned) · TA1179 is now stated and linked in all four places, with its date and the reason
- 2026-11-17 · STAMPEDE's arms, and what each one answered: The term said STAMPEDE2 opened in 2023 with three comparisons including niraparib with abiraterone; it opened to recruitment in June 2024 with two, stereotactic radiotherapy and lutetium-177 PSMA-617. It said the trial had asked nine questions, four practice-changing and five not, which no source supports, and that the enzalutamide reading was practice-changing and rested on a subgroup. It said the oestradiol patch cancer results were reported separately, when the locally advanced cohort had reported in April 2026 (Second accuracy pass; ClinicalTrials.gov NCT06320067 and NCT00268476, the STAMPEDE2 trial site read as raw HTML, the full text of DOI 10.1016/S0140-6736(18)32486-3 through Europe PMC, and the abstract for PMID 41880608 (DOI 10.1056/NEJMoa2511781) through PubMed) · The summary line names the comparisons that did and did not change practice, STAMPEDE2 is described as the registry has it, the subgroup sentence gives the low-burden figures (hazard ratio 0.68, 0.52 to 0.90), and the patch result is added (3-year metastasis-free survival 87.1 against 85.9 percent in 1,360 men, non-inferior)
- 2026-11-17 · What androgen deprivation costs: Bone density figures for oestradiol patches (up 7.9 percent against down 3.0 percent) were presented as a result of the 1,694-man PATCH trial. They come from a 74-man substudy, at 2 years, in men still on their allocated treatment. The term also gave no cancer-control result for patches (Second accuracy pass; the abstracts for PMID 26707868, PMID 37973477 and PMID 41880608 read through PubMed) · The substudy size and time point are stated, the three papers are linked, and the 2026 non-inferiority result for locally advanced disease is added
- 2026-11-17 · Prostate cancer: standard of care rows: The high-risk row said a brachytherapy boost doubles freedom from biochemical failure (83 against 62 percent), which the figures do not show; ASCENDE-RT found the external-beam arm twice as likely to fail. The intermediate and high-risk rows credited NICE NG131 but left out that it offers active surveillance for CPG 2 and radical prostatectomy for CPG 4 and 5. The metastatic row gave the first analysis of CHAARTED and the first interim analysis of TITAN as the results (Second accuracy pass; the abstracts for PMID 28262473, PMID 29384722 and PMID 33914595 read through PubMed, and NICE NG131 recommendations 1.3.8 to 1.3.12 read as raw HTML) · The boost is described as halving biochemical failure with no significant survival difference, the NG131 options are stated, CHAARTED carries its long-term figures (hazard ratio 0.72, no benefit in low-volume disease) and TITAN its final hazard ratio of 0.65
- 2026-11-17 · TITAN: The result gave overall survival at 24 months (hazard ratio 0.67) without saying it was the first interim analysis, and the final analysis was missing (Second accuracy pass; the abstracts for PMID 31150574 and PMID 33914595 (DOI 10.1200/JCO.20.03488) read through PubMed) · The result is labelled as interim and the final analysis added: hazard ratio 0.65 (0.53 to 0.79), median not reached against 52.2 months at a median 44.0 months
- 2026-11-17 · Prostate cancer programmes that failed: The term said PROSTVAC and sipuleucel-T after it showed the same for vaccines, though sipuleucel-T came first and extended survival in IMPACT; said bevacizumab quadrupled treatment-related deaths when the figures are 4.0 against 1.2 percent; and called radium-223 a radioligand (Second accuracy pass; the abstracts for PMID 30817251, PMID 20818862 and PMID 22454414 read through PubMed) · The vaccine sentence is rewritten, the death rate reads more than tripled with the figures, and the last word is radium
- 2026-11-17 · Focal and tissue-preserving treatment of prostate cancer: The term named PART and CHRONOS as the UK trials to watch. CHRONOS is recorded as completed after enrolling 100 men, with primary completion in March 2023 and no result found (Second accuracy pass; ClinicalTrials.gov NCT04049747 read through the v2 API and the abstract for PMID 42760023 (DOI 10.1111/bju.70448) read through PubMed) · The sentence now describes PART (ISRCTN17249875, 306 men needed) and says what the registry records for CHRONOS
- 2026-11-17 · How many radiotherapy visits: the prostate fractionation trials: HYPO-RT-PC was said to have randomised 1,180 men. It randomised 1,200, of whom 1,180 formed the per-protocol population analysed (Second accuracy pass; the abstract for DOI 10.1016/S0140-6736(19)31131-6 read through PubMed) · Both numbers are given
- 2026-11-17 · Sipuleucel-T: The note called it the first therapeutic cancer vaccine approved anywhere, which no source read supports (Second accuracy pass; the Europe PMC record for PMID 21471425 (DOI 10.1158/1078-0432.CCR-10-3126)) · The claim is limited to approval by the United States Food and Drug Administration and the paper is linked
- 2026-11-17 · PATCH (Prostate Adenocarcinoma Transcutaneous Hormone): The record said the cancer outcomes were still to come and showed the trial as active. The locally advanced cohort has reported: in 1,360 men, 3-year metastasis-free survival was 87.1 percent with oestradiol patches against 85.9 percent with injections, hazard ratio 0.96 with an upper one-sided 95 percent limit of 1.11, which met non-inferiority, and 5-year overall survival was 81.1 against 79.2 percent (Second accuracy pass; a Europe PMC title search found the report, and its abstract (DOI 10.1056/NEJMoa2511781) was read through Europe PMC) · Summary, result, short summary and outcomes now carry the 2026 result, the paper is linked, the status is positive and the year reported is 2026. The record says no journal report of the metastatic cohort was found on 17 November 2026
- 2026-11-17 · GETUG-AFU 15: The summary said later analyses with longer follow-up found a benefit concentrated in high-volume disease. The long-term analysis found no significant survival benefit overall (hazard ratio 0.88, 0.68 to 1.14) or in high-volume disease (0.78, 0.56 to 1.09, p=0.14). An outcome was labelled grade 3 or worse neutropenia where the paper reports serious adverse events without grades, and the short summary called it the first trial of chemotherapy at the start of hormone therapy with no source (Second accuracy pass; the abstracts for DOI 10.1016/S1470-2045(12)70560-0 and DOI 10.1016/j.eururo.2015.11.005 read through Europe PMC) · Summary and replication note rewritten with the long-term figures, the 2016 paper linked, the outcome relabelled and the short summary reworded to the first of three large docetaxel trials
- 2026-11-17 · COU-AA-302: Overall survival was given only as the interim result that fell short of the efficacy boundary. The final analysis at a median 49.2 months was significant: 34.7 against 30.3 months, hazard ratio 0.81 (0.70 to 0.93), p=0.0033 (Second accuracy pass; the abstract for DOI 10.1016/S1470-2045(14)71205-7 read through Europe PMC after a title search) · Final overall survival added to the summary, result, outcomes and links, and as an attributed finding on the 2013 paper record; the interim figure is labelled interim
- 2026-11-17 · IMPACT (sipuleucel-T): The record called sipuleucel-T the first cancer vaccine approved anywhere, said nothing shrank, credited the trial with prompting the immune-related response criteria, and said the treatment is not available at all. The sources support only the first treatment vaccine approved by the US regulator and a similar time to progression in both arms; the response criteria paper dates from 2009, before this trial was published; and a current United States label exists (Second accuracy pass; the abstracts for DOI 10.1056/NEJMoa1001294 and DOI 10.1158/1078-0432.CCR-10-3126 read through Europe PMC, the NICE TA332 page read with curl, and the openFDA label record for Provenge) · Short summary and summary reworded to what the sources state; the response criteria claim removed; availability limited to the NHS
- 2026-11-17 · CARD: The trial was recorded as phase 3. ClinicalTrials.gov registers NCT02485691 as phase 4 (Second accuracy pass; the ClinicalTrials.gov record read through its API) · Phase changed to 4 and the enrolment note says so
- 2026-11-17 · PREVAIL: The short summary said enzalutamide cut the risk of death by a third. The paper reports a 29 percent reduction (hazard ratio 0.71, 0.60 to 0.84) (Second accuracy pass; the abstract for DOI 10.1056/NEJMoa1405095 read through Europe PMC) · Short summary now gives 81 percent for progression or death and 29 percent for death
- 2026-11-17 · HERO: The summary called major adverse cardiovascular events a secondary endpoint. The registry lists them among other outcome measures, not the secondary ones (Second accuracy pass; the ClinicalTrials.gov record for NCT03085095 read through its API) · Sentence reworded
- 2026-11-17 · PEACE-3 (EORTC 1333): The short summary said fractures were prevented by bone-protecting drugs. The trial's safety analysis reports a fall from 15.6 to 2.6 percent at one year on enzalutamide alone, and the primary report counted fractures in 13.4 and 24.3 percent of the two arms over the whole trial (Second accuracy pass; the abstracts for DOI 10.1016/j.eururo.2024.11.027 and DOI 10.1016/j.annonc.2025.05.011 read through Europe PMC) · Short summary says sharply reduced and the summary gives the figures
- 2026-11-17 · Prostate cancer: history and open problems: The timeline gave PSMAddition's study completion as 11 February 2027 where the registry now lists 1 August 2028. The 2009 screening entry gave only the first ERSPC figures, which the 23-year follow-up has improved to a 13 percent lower mortality with 456 men invited and 12 diagnosed per death prevented. An open problem said Black men had no excess prostate cancer mortality after adjustment in three cohorts, where the paper found a small remaining excess in the registry cohort, none in the Veterans Affairs cohort and a lower hazard in the trial cohort (Second accuracy pass; the ClinicalTrials.gov record for NCT04720157 read through its API, the abstract for DOI 10.1056/NEJMoa2503223 read through Europe PMC, and the abstract for PubMed id 31120534 read through PubMed) · Date corrected, the 23-year result added to the 2009 entry with its DOI, and the open problem reworded to the three cohort results
- 2026-11-17 · COMET-1: The record said Exelixis stopped developing cabozantinib in prostate cancer. The company later ran CONTACT-02, a phase 3 trial of cabozantinib with atezolizumab in 575 men with prostate cancer. (Second accuracy pass; the abstract for DOI 10.1016/S1470-2045(25)00209-8 read through PubMed, with the COMET-2 abstract for DOI 10.1016/j.eururo.2018.11.033.) · The sentence now gives the CONTACT-02 result: progression-free survival improved (hazard ratio 0.65, 0.50 to 0.84) and overall survival did not (0.89, 0.72 to 1.10). COMET-2 is described as stopped early at 119 men with no better pain response. Both papers are linked.
- 2026-11-17 · ERA 223: PEACE-3 was said to have made bone-protecting agents mandatory for everyone, and the fractures in ERA 223 were said to be attributable to the missing bone protection. In PEACE-3 the agents became mandatory from March 2018, more than two years after randomisation began, and neither paper makes the causal statement. (Second accuracy pass; the abstracts for DOI 10.1016/j.annonc.2026.02.009 and DOI 10.1016/j.annonc.2025.05.011 read through PubMed, and the EMA Xofigo page read as raw text.) · The record gives the March 2018 date, names the hazard ratio of 0.76 as the final analysis, reports what the PEACE-3 authors state, says the contraindication is in the European Union, and notes that the ERA 223 abstract does not say whether bone-protecting agents were mandatory.
- 2026-11-17 · CA184-043: The summary line called it the first big immunotherapy trial in prostate cancer. IMPACT, the phase 3 trial of sipuleucel-T, reported in 2010. (Second accuracy pass; the abstract for DOI 10.1056/NEJMoa1001294 read through PubMed.) · The line now reads a large trial of the immunotherapy drug ipilimumab.
- 2026-11-17 · Atrasentan: The endothelin programme was counted as five phase 3 trials when there were six, the pain palliation figures were given in the reverse order of the arms, and the record did not say atrasentan is now an approved kidney medicine. (Second accuracy pass; the abstracts for the three atrasentan and three zibotentan phase 3 trials read through PubMed, the patient-reported outcomes abstract for DOI 10.1186/s41687-018-0054-5, and the openFDA record for NDA 219208.) · Six trials, none meeting its primary endpoint; pain palliation 44.0 percent with atrasentan against 41.7 percent with placebo; and the United States approval of 2 April 2025 for IgA nephropathy, with no cancer indication.
- 2026-11-17 · Custirsen: One phase 2 trial of 42 men was said to have justified both phase 3 trials. That trial gave custirsen to every man after docetaxel; the first-line randomised phase 2 trial was a different one, in 82 men. (Second accuracy pass; the abstracts for DOI 10.1158/1078-0432.CCR-11-0859 and DOI 10.1200/JCO.2009.26.8771 read through Europe PMC and PubMed.) · Both phase 2 trials are described, with the first-line survival figures of 23.8 against 16.9 months.
- 2026-11-17 · PROSTVAC: The phase 2 survival signal was stated as an artefact of comparison against a historical control. The phase 2 trial had a concurrent randomised control arm, and the phase 3 authors say only that its short survival may be the origin of the flawed comparison. (Second accuracy pass; the full text of DOI 10.1200/JCO.18.02031 read through PubMed Central, and the phase 2 abstract for DOI 10.1200/JCO.2009.25.0597.) · Reworded to the authors' statement, with the control arm figures of 16.6 months observed against 20.4 predicted. The phase 2 paper is linked.
- 2026-11-17 · CALGB 90401: Treatment-related deaths of 4.0 against 1.2 percent were described as four times as many and as quadrupling. (Second accuracy pass; the abstract for DOI 10.1200/JCO.2011.39.4767 read through PubMed.) · Now more than three times as many.
- 2026-11-17 · READY: Adverse event counts were given out of the 762 and 760 men randomised. The safety analysis was in the 761 and 757 men treated. (Second accuracy pass; the full text of DOI 10.1016/S1470-2045(13)70479-0 read through PubMed Central.) · Denominators corrected.
- 2026-11-17 · KEYNOTE-641: The summary line said three times as many men had a serious side effect. The figure is for grade 3 or higher treatment-related adverse events, 31.2 against 10.8 percent, which is not the same measure as serious adverse events. (Second accuracy pass; the abstract for DOI 10.1016/j.annonc.2025.05.007 read through Europe PMC.) · Now about three times as many men had a severe treatment-related side effect.
- 2026-11-17 · KEYNOTE-991: The summary line called it the last attempt to make immunotherapy work in prostate cancer, which no source states. (Second accuracy pass; the abstract for DOI 10.1016/j.annonc.2025.05.008 read through Europe PMC.) · Now an attempt to make immunotherapy work by giving it as early as possible.
- 2026-11-17 · KEYNOTE-921: The pooled analysis with CheckMate 7DX was quoted with no source. (Second accuracy pass; the abstracts for DOI 10.1007/s11255-026-05064-w and DOI 10.1016/S1470-2045(25)00566-2 read through PubMed.) · Figures confirmed; both papers linked and the CheckMate 7DX result added (1,030 men, hazard ratios 0.96 and 1.09).
- 2026-11-17 · AFFINITY: The sponsor was given as OncoGenex Technologies; the paper names OncoGenex Pharmaceuticals. (Second accuracy pass; the abstract for DOI 10.1016/S1470-2045(17)30605-8 read through Europe PMC.) · Sponsor corrected.
- 2026-11-17 · ENTHUSE M1: The alias ENTHUSE 14 could not be found in the registry or the paper. (Second accuracy pass; the ClinicalTrials.gov record NCT00554229 read through the registry API.) · Alias replaced with the registry study number D4320C00014.
- 2026-11-17 · DART 01/05 GICOR: The page gave the 5-year result as the trial's answer: status positive, and a summary line saying two more years of hormone therapy still improves survival. The 10-year final results did not confirm it (Second accuracy pass; the abstract for DOI 10.1016/S1470-2045(22)00190-5 read through PubMed) · At 10 years overall survival was 78.4 against 73.3 percent (hazard ratio 0.84, 0.55 to 1.27) and biochemical disease-free survival 70.2 against 62.3 percent (0.84, 0.50 to 1.43), neither significant. The page now gives both data cuts, the high-risk subgroup, and the authors' statement that they could not support the 5-year benefit. Status is mixed
- 2026-11-17 · NRG Oncology RTOG 0521: The page said adding docetaxel improved four-year survival from 89 to 93 percent and stopped there, with status positive. The long-term report found no significant survival difference and concluded docetaxel should not be used (Second accuracy pass; the abstract for DOI 10.1016/j.eururo.2023.04.024 read through PubMed) · At a median 10.4 years the hazard ratio for death was 0.89 (90 percent confidence interval 0.70 to 1.14) with 10-year survival 69 against 64 percent. The summary line, summary, result and outcomes now say so. Status is mixed. The GETUG 12 comparison now names its endpoint, relapse-free survival
- 2026-11-17 · RADICALS-RT: The page named freedom from distant metastases as the primary outcome but gave no result for it, and said two men in three were spared radiotherapy. The long-term report has both (Second accuracy pass; the abstract for DOI 10.1016/j.annonc.2024.03.010 read through PubMed) · Ten-year freedom from distant metastases was 93 percent with adjuvant against 90 percent with salvage radiotherapy (hazard ratio 0.68, 0.43 to 1.07, not significant) and overall survival did not differ. By then 39 percent of the salvage group had had radiotherapy, so the summary line now says about three men in five were spared. The ARTISTIC page, which said long-term outcomes were still awaited, now gives this result
- 2026-11-17 · ERSPC: The page stopped at the 16-year report and said the UK CAP trial found no mortality reduction at 15 years. ERSPC published its 23-year final analysis in October 2025, and CAP found a small reduction at 15 years (Second accuracy pass; the abstracts for DOI 10.1056/NEJMoa2503223 and DOI 10.1001/jama.2024.4011 read through PubMed) · Added the 23-year result: rate ratio 0.87 (0.80 to 0.95), one death prevented per 456 men invited and per 12 diagnosed. CAP is now given as no effect at 10 years and 0.69 against 0.78 percent at 15 years (rate ratio 0.92, 0.85 to 0.99). The PLCO page gains its 17-year result, rate ratio 0.93 (0.81 to 1.08)
- 2026-11-17 · RTOG 92-02: The Gleason 8 to 10 subgroup was called prespecified, the summary line said only those men lived longer, and a note said the trial had no registration (Second accuracy pass; the abstracts for DOI 10.1200/JCO.2003.11.023 and DOI 10.1016/j.ijrobp.2017.02.004 read through PubMed, and the ClinicalTrials.gov record NCT00767286) · The first report describes the subgroup as not part of the original study design, and the page now says so. The update at a median 19.6 years found overall survival also improved with the long course (12 percent relative reduction, p=0.03), which is added. The note now names the registration
- 2026-11-17 · RTOG 94-08: The summary line said four months of hormone therapy helped intermediate-risk men live longer, without the long-term update in which the survival curves converged (Second accuracy pass; the abstract for DOI 10.1016/j.ijrobp.2021.08.031 read through PubMed) · Added: overall survival was 63 against 56 percent at 10 years and 23 percent in both groups at 18 years (hazard ratio 0.94, 0.85 to 1.05), while deaths from prostate cancer stayed lower (8 against 14 percent at 18 years)
- 2026-11-17 · EORTC 22991: The page said overall survival data were not mature. The 12-year report has them (Second accuracy pass; the abstract for DOI 10.1200/JCO.21.00855 read through PubMed) · In the 481 intermediate-risk men, 10-year overall survival was 80.0 against 74.3 percent, hazard ratio 0.74 (0.53 to 1.04), not statistically significant. Added to the summary and result
- 2026-11-17 · GETUG-AFU 17: The summary line said adjuvant radiotherapy doubles the rate of lasting bladder and erection problems. The trial's figures are 27 against 7 percent and 28 against 8 percent (Second accuracy pass; the abstract for DOI 10.1016/S1470-2045(20)30454-X read through PubMed) · The line now says more than triples
- 2026-11-17 · SPCG-4: The page said ProtecT found prostate cancer mortality under 3 percent in every arm at 15 years. The active monitoring arm was 3.1 percent (Second accuracy pass; the abstract for DOI 10.1056/NEJMoa2214122 read through PubMed) · Now reads 2.7 percent overall, between 2.2 and 3.1 percent by arm
- 2026-11-17 · ASCENDE-RT: The page said the brachytherapy boost had not been tested in a second randomised trial and was recommended on the strength of this trial alone (Second accuracy pass; the abstracts for DOI 10.1016/j.radonc.2020.09.047 and DOI 10.1016/j.ijrobp.2022.11.005 read through PubMed, and NICE NG131 recommendation 1.3.24 read as a raw page) · A randomised trial of a high-dose-rate boost in 216 men also found better relapse-free survival without a survival difference, and NICE cites both kinds of boost. The page now says so and adds the 10-year update: 85 against 67 percent free of biochemical progression, no difference in metastasis or survival
- 2026-11-17 · PIVOT: A note said the registry gives a planned target rather than the analysed number, and the page stopped at the 2017 report (Second accuracy pass; the ClinicalTrials.gov record NCT00007644 and the abstract for DOI 10.1016/j.eururo.2020.02.009 read through PubMed) · The registry gives 731 as actual enrolment. Added the 22-year all-cause mortality result: hazard ratio 0.84 (0.70 to 1.00, p=0.044), a mean of one year of life gained, not adjusted for multiple comparisons
- 2026-11-17 · EORTC 22863: The summary line said adding hormone therapy roughly halves the chance of dying of prostate cancer, and the radiotherapy was described as 50 Gy over 5 weeks, which the report does not say (Second accuracy pass; the abstract for DOI 10.1016/S1470-2045(10)70223-0 read through PubMed) · The line now gives the figures, about 30 percent against about 10 percent at ten years. The schedule now reads as the report has it: daily for 7 weeks, 50 Gy to the pelvis and a further 20 Gy to the prostate
- 2026-11-17 · RADICALS-HD: The summary line described the gain as cancer kept from spreading, when the endpoint also counts death from any cause, and the page did not mention that the same trial found no gain from six months of hormone therapy over none (Second accuracy pass; the abstracts for DOI 10.1016/S0140-6736(24)00549-X and DOI 10.1016/S0140-6736(24)00548-8 read through PubMed) · The line now says alive and free of cancer spread at ten years. The none against short-course comparison is added: hazard ratio 0.886 (0.688 to 1.140) in 1,480 men
- 2026-11-17 · PROFIT, RTOG 0415 and SWOG 8794: PROFIT's sponsor line named the Canadian Cancer Trials Group, which the registry does not. SWOG 8794 was said to have no registration. RTOG 0415 lacked its long-term report (Second accuracy pass; the ClinicalTrials.gov records NCT00304759 and NCT00394511, and the abstract for DOI 10.1200/JCO.23.02445 read through PubMed) · PROFIT's sponsor line now follows the registry and its hazard ratio is marked as having a 90 percent interval. SWOG 8794 now names NCT00394511. RTOG 0415 gains the 12.8-year result: non-inferiority confirmed, no significant difference in late grade 3 or worse events
- 2026-11-17 · TRITON3: rucaparib or physician's choice in metastatic castration-resistant prostate cancer: The record gave no overall survival result and its plain summary said the drug beats the alternatives. The trial's final overall survival did not differ significantly: in the BRCA subgroup a median of 23.2 against 21.2 months, hazard ratio 0.91 (0.68 to 1.20), and in all randomised men hazard ratio 0.99 (0.78 to 1.26) (Second accuracy pass; the results section of the ClinicalTrials.gov record NCT02975934 read through the registry API, and the current US label read through openFDA) · The plain summary, a new finding and the first caveat now give the final survival figures and say they were not significant, with the label's figure for the ATM subgroup
- 2026-11-17 · TRITON2: rucaparib in men with metastatic castration-resistant prostate cancer harbouring a BRCA1 or BRCA2 alteration: Rucaparib was called the second PARP inhibitor licensed in prostate cancer; in the United States it was approved on 15 May 2020, four days before olaparib. The record also stopped at the 2020 accelerated approval after a taxane, although the US label was revised in December 2025 and no longer requires a prior taxane (Second accuracy pass; the FDA announcements of 15 and 19 May 2020 read as Internet Archive copies, and the Rubraca label and approval history read through openFDA) · The order of approval is corrected, the approval date is given, the December 2025 label change is described and the FDA announcement is linked
- 2026-11-17 · KEYNOTE-199: pembrolizumab for treatment-refractory metastatic castration-resistant prostate cancer: The summary said the field keeps returning to immunotherapy with combinations and did not say how those ended. Both phase 3 pembrolizumab combinations were negative: KEYNOTE-921 with docetaxel (hazard ratio for death 0.92, 0.78 to 1.09) and KEYNOTE-641 with enzalutamide (1.04, 0.88 to 1.22, stopped for futility) (Second accuracy pass; the abstracts for DOI 10.1200/jco-24-01283 and DOI 10.1016/j.annonc.2025.05.007 read through Europe PMC) · The summary and a new caveat give both results and both papers are linked
- 2026-11-17 · TROPIC: prednisone plus cabazitaxel or mitoxantrone for metastatic castration-resistant prostate cancer progressing after docetaxel: Severe neutropenia was given as 303 of 377 cabazitaxel patients. 377 is the number randomised to mitoxantrone; 378 were randomised to cabazitaxel and 371 in each arm were treated and counted for safety (Second accuracy pass; the abstract for DOI 10.1016/s0140-6736(10)61389-x read through PubMed and the adverse event tables of the ClinicalTrials.gov record NCT00417079) · The wrong denominator is removed and the safety population is stated
- 2026-11-17 · The mutational landscape of lethal castration-resistant prostate cancer: The plain summary said fifty men had their tumours sequenced within hours of death. The paper sequenced 50 tumours, three of them from one patient, obtained at rapid autopsy, and gives no time (Second accuracy pass; the abstract for DOI 10.1038/nature11125 read through PubMed) · The summary and a caveat now say fifty tumours and rapid autopsy
- 2026-11-17 · Integrative genomic profiling of human prostate cancer: The record said the paper's copy-number finding became Decipher and the other commercial classifiers. Decipher is a 22-marker RNA expression classifier developed in a Mayo Clinic cohort, and nothing read derives it from this paper (Second accuracy pass; the abstracts for DOI 10.1016/j.ccr.2010.05.026 and DOI 10.1371/journal.pone.0066855 read through Europe PMC) · The summary and the interpretation now say the commercial classifiers were built separately on RNA expression, and the Decipher paper is linked
- 2026-11-17 · STOpCaP: docetaxel or bisphosphonates added to standard of care in hormone-sensitive prostate cancer, a systematic review and meta-analysis: The plain summary said docetaxel improved four-year survival by about nine percent. The paper reports an absolute improvement of 9 percent, which is nine percentage points (Second accuracy pass; the abstract for DOI 10.1016/s1470-2045(15)00489-1 read through PubMed) · The summary now says nine percentage points
- 2026-11-17 · Huggins and Hodges 1941: the effect of castration, of oestrogen and of androgen injection on serum phosphatases in metastatic carcinoma of the prostate: The plain summary said this was the first time any cancer had been made to regress by a drug or a hormone, which no source read states, and a caveat said prostate-specific antigen was not described until 1987, when the 1987 paper compared an existing marker with prostatic acid phosphatase (Second accuracy pass; the Nobel Prize facts page read as raw HTML and the abstract of the 1987 paper (PubMed id 2442609) read through Europe PMC) · The priority claim is replaced by the Nobel citation and the caveat now says what the 1987 paper showed
- 2026-11-17 · Four key-paper records with unsourced claims of being first or largest: AR-V7 was called the first predictive resistance biomarker in prostate cancer, the TMPRSS2 fusion the first recurrent rearrangement found in a common carcinoma, TRANSFORMER the first randomised evidence that treatment order matters, and the Foundation Medicine series the largest routine-practice picture. None of the four abstracts says so (Second accuracy pass; the abstracts for DOI 10.1056/nejmoa1315815, DOI 10.1126/science.1117679, DOI 10.1200/jco.20.02759 and DOI 10.1200/po.18.00283 read through PubMed) · Each sentence is reworded to what its abstract supports
- 2026-11-17 · GETUG-AFU 15: androgen deprivation alone or with docetaxel in non-castrate metastatic prostate cancer: The record carried only the 2013 analysis at 50 months and did not say whether longer follow-up changed it. It did not: at 83.9 months the hazard ratio for death was 0.88 (0.68 to 1.14) (Second accuracy pass; the abstract for DOI 10.1016/j.eururo.2015.11.005 read through Europe PMC) · A caveat gives the long-term result and the analysis by disease volume, and the 2016 report is linked
- 2026-11-17 · ERSPC: screening and prostate cancer mortality in a randomised European study: The record gave only the 2009 figures (a 20 percent reduction, 1,410 screened and 48 treated per death prevented) and named the 16-year report as the latest follow-up. The 23-year follow-up, published in October 2025, reports a rate ratio of 0.87 (0.80 to 0.95), an absolute reduction of 0.22 percent, and 456 men invited and 12 diagnosed per death prevented (Second accuracy pass; the abstract for DOI 10.1056/nejmoa2503223 read through PubMed) · The 23-year figures are now in the summary line, the summary, a caveat and the links; the 2009 figures are dated to nine years
- 2026-11-17 · PLCO: mortality results from a randomised prostate cancer screening trial: The record stopped at the trial's 7 to 10 year data, with no later report (Second accuracy pass; the abstract for DOI 10.1002/cncr.30474 read through PubMed) · A caveat now gives the 15-year result: 255 against 244 prostate cancer deaths, rate ratio 1.04 (0.87 to 1.24), with 86 percent of the control arm tested at some point. The paper is linked
- 2026-11-17 · Prostate cancer diagnosis and treatment after the introduction of prostate-specific antigen screening, 1986 to 2005: The summary line said the benefit reached one man in twenty of those treated. The paper says more than 20 men had to be diagnosed for each man who benefited (Second accuracy pass; the abstract for DOI 10.1093/jnci/djp278 read through PubMed) · Now reads fewer than one in twenty of the men diagnosed
- 2026-11-17 · Lead time and overdiagnosis in prostate-specific antigen screening: The summary presented the ranges of 5.4 to 6.9 years and 23 to 42 percent as disagreement between definitions of lead time. They are the spread across the three models for a single definition (Second accuracy pass; the abstract for DOI 10.1093/jnci/djp001 read through PubMed) · The sentence now says which definition the ranges belong to
- 2026-11-17 · USPSTF 2012: screening for prostate cancer, recommendation statement (grade D): The summary said testing and biopsy rates and the proportion of cancers found at a localised stage fell afterwards, with no source; the cited statement does not say so (Second accuracy pass; the abstracts for DOI 10.1001/jama.2015.14905 and DOI 10.1002/cncr.31549 read through PubMed) · Reworded to what those two papers report: screening rates down 18 percent between 2010 and 2013, early-stage incidence down, distant-stage incidence up from 2010 to 2014. Both are linked
- 2026-11-17 · USPSTF 2018: screening for prostate cancer, recommendation statement: The record called the figures a three-to-one ratio of metastatic cases prevented to deaths prevented. The statement gives about 3 and about 1.3 per 1,000, which is a little over two to one. It also did not say the recommendation is being updated (Second accuracy pass; the abstract for DOI 10.1001/jama.2018.3710 read through PubMed, and the task force's prostate page read as raw HTML) · The ratio is now stated as the two figures, and the record says an update was listed as in progress on 17 November 2026
- 2026-11-17 · Trans-ancestry genome-wide association meta-analysis of prostate cancer: The record called this the largest genetic study of prostate cancer and gave 269 as the number of known risk variants. A 2023 study pooled 156,319 cases and 788,443 controls and raised the total to 451 (Second accuracy pass; the abstract for DOI 10.1038/s41588-023-01534-4 read through PubMed) · The claim is now dated to 2021, and a caveat and a link give the 2023 study
- 2026-11-17 · Prostate cancer roadmap: Seven stale or wrong statements. Transdermal oestradiol was described as still being tested, when the trial reported in April 2026. ERSPC appeared only at its 2009 figures. RESTORE was given as 52 percent of men rechallenged with enzalutamide responding, when 15 of the 21 rechallenged did (71 percent) and 52 percent counts the 29 who completed testosterone. Overdiagnosis estimates of 1.7 to 67 percent were called a factor of thirty. The roadmap said there are no head-to-head sequencing data, when a 202-man randomised crossover trial exists. Prostate radiotherapy in STAMPEDE was given as improving survival without saying this was a subgroup. PARP inhibitors were said to have been licensed after a taxane (Second accuracy pass; abstracts for DOI 10.1056/NEJMoa2511781, 10.1056/nejmoa2503223, 10.1016/j.eururo.2013.12.062 and 10.1016/S1470-2045(19)30688-6 read through PubMed, the RESTORE and STAMPEDE full texts (PMC5875180, PMC6269599), and the US olaparib and rucaparib labels through openFDA) · Each statement corrected in place: the oestradiol result (87.1 against 85.9 percent metastasis-free at three years, non-inferior), the 23-year ERSPC figures, 15 of 21 with both percentages explained, almost forty, almost no sequencing data, the subgroup and whole-trial hazard ratios (0.68 and 0.92), and first licensed after an androgen receptor drug
- 2026-11-17 · Prostate cancer roadmap: Six entries in the watch list were wrong or out of date. STAMPEDE2 was said to test niraparib with abiraterone; the registry lists two comparisons, stereotactic radiotherapy and lutetium-177 PSMA-617, and a start in June 2024. PSMAddition's completion date was given as 11 February 2027; the registry gives 1 August 2028 and now posts an overall survival hazard ratio of 0.84 (0.63 to 1.13), not significant. PSMA-DC was said to test a radioligand in place of metastasis-directed treatment; both arms have radiotherapy to every metastasis. CAPItello-281 was called the first PI3K-pathway result in a biomarker-selected population, which IPATential150 was in 2021. ProBio and PROTEUS carried unsourced first and largest claims and no mention of their published results (Second accuracy pass; ClinicalTrials.gov records NCT06320067, NCT04720157, NCT05939414, NCT03903835 and NCT03767244 read through the API, and abstracts for DOI 10.1016/S0140-6736(21)00580-8, 10.1038/s41591-024-03204-2 and 10.1056/NEJMoa2603878 read through PubMed) · Each entry corrected from the registry, with the ProBio and PROTEUS results added
- 2026-11-17 · Take high-dose testosterone to phase 3, with progression-free survival through the second line as the primary endpoint: The record said 15 of 21 (52 percent) responded to enzalutamide rechallenge and that half respond again; 15 of 21 is 71 percent and the 52 percent counts 29 men. It also said the approach had never been taken to a definitive trial, when a phase 2/3 trial was registered on 1 October 2026 (Second accuracy pass; the RESTORE full text (PMC5875180) and ClinicalTrials.gov record NCT07852195 read through the API) · The response is now given as 15 of 21 with both percentages explained, and the registered ECOG-ACRIN trial (432 men, not yet recruiting) is described
- 2026-11-17 · Randomise the order of treatment, not only the drugs: The record said no trial has ever compared the orders in which treatments can be given. A Canadian randomised crossover trial of 202 men compared abiraterone then enzalutamide with the reverse, and TRANSFORMER is cited in the record itself (Second accuracy pass; the abstract for DOI 10.1016/S1470-2045(19)30688-6 read through PubMed) · Now says only a few small trials have done so, and names them
- 2026-11-17 · Judge a prostate screening programme on metastatic presentation, not on incidence or mortality: Three fields said preventing a metastatic presentation happens three times as often as preventing a death. The task force figures are about 3 and about 1.3 per 1,000, a little over twice (Second accuracy pass; the abstract for DOI 10.1001/jama.2018.3710 read through PubMed) · Now reads more than twice, as the summary already did
- 2026-11-17 · Watch for the cancer changing cell type before the biopsy says neuroendocrine, and act on it: The rationale said treatment-emergent small-cell disease has a median survival measured in months. The cited study reports a hazard ratio for death, not a median (Second accuracy pass; the abstract for DOI 10.1200/jco.2017.77.6880 read through PubMed) · Now gives the hazard ratio of 2.02
- 2026-11-17 · Test every man for DNA repair faults on the day his prostate cancer is found to have spread: The summary said PARP inhibitors were licensed after an androgen receptor drug and a taxane; the US olaparib and rucaparib labels require an androgen receptor drug only. It also did not say that labels now cover hormone-sensitive disease (Second accuracy pass; the US labels for Lynparza, Rubraca, Akeega and Talzenna read through openFDA) · Reworded, with the two hormone-sensitive indications added
- 2026-11-17 · Prostate cancer: surgery and radiotherapy side effect figures: The ProtecT trial's leaking and erection figures were attached to the robotic surgery record, and its bowel figures to the 20-fraction radiotherapy record, without saying the trial used neither. Of the operations in ProtecT, 324 were open, 23 laparoscopic and 25 robot-assisted, and the radiotherapy was 74 Gy in 37 fractions with hormone therapy beforehand (Second accuracy pass; the full text for DOI 10.1056/NEJMoa1606221 read through NCBI PubMed Central) · Both notes now say how the surgery and the radiotherapy were given in the trial. The response rate is now given as above 80 percent for most measures, as the 12-year paper states
- 2026-11-17 · Prostate cancer: radium-223 when cancer has reached the bone: Radium-223 was described as an option for symptomatic bone-only disease with no conditions, and as given with bone protection alongside. NICE TA412 recommends it only for people who have already had docetaxel or for whom docetaxel is contraindicated or not suitable, and the bone protection statement had no source (Second accuracy pass; the NICE TA412 recommendations chapter and the European Medicines Agency overview of Xofigo read as raw HTML) · The sentence now gives the TA412 conditions, links the appraisal, and replaces the bone protection clause with the agency's statement that radium-223 can increase the risk of fractures
- 2026-11-17 · Prostate cancer: questions to ask: The Prostate Cancer UK nurse line was described as there for questions at nine in the evening, but it is open Monday to Friday from 9am to 5pm, from 10am on Wednesdays. A carers' question credited the charity with saying a partner is often the first to notice a change, which its pages do not say. Hormone therapy was said to raise insulin where the charity says it changes insulin (Second accuracy pass; Prostate Cancer UK's specialist nurses, sex and relationships, erection treatments and hormone therapy pages read as raw HTML) · The opening hours are written in, the unsupported clause is removed and the hormone therapy sentence follows the charity's wording
- 2026-11-17 · Prostate cancer: NICE NG131 recommendation numbers: Four statements about NICE NG131 did not match the guideline. It was said to write one recommendation for each of the five Cambridge Prognostic Groups, but groups 4 and 5 share recommendations 1.3.11 and 1.3.12. Explaining the limits of nomograms was credited to 1.1.7 and is 1.1.8. Recommendation 1.3.5 was described as an assessment by both a surgeon and a clinical oncologist where it gives the opportunity to discuss options with both. Weekly tamoxifen was credited to 1.4.16 and is 1.4.17. The page also did not say that NICE has announced an update (Second accuracy pass; the NICE NG131 recommendations and overview pages read as raw HTML) · The numbers and wording are corrected, and a note now says NICE reviewed the guideline in May and August 2025 and will update the localised and locally advanced section
- 2026-11-17 · Metastatic spinal cord compression (MSCC): NICE NG234 was said to ask for dexamethasone and lying flat while the diagnosis is made, without qualification. The guideline offers dexamethasone to people with neurological symptoms or signs, says not to offer corticosteroids routinely without them, and asks for immobilisation where the spine may be unstable (Second accuracy pass; the NICE NG234 recommendations page read as raw HTML) · The sentence now gives both conditions and the full MRI wording, as soon as possible and always within 24 hours
- 2026-11-17 · Priapism: The record linked the NHS priapism page but gave only the four-hour threshold from Prostate Cancer UK. The NHS says to call 999 or go to accident and emergency for an erection lasting more than 3 to 4 hours (Second accuracy pass; the NHS priapism page read as raw HTML) · The NHS wording is added beside the charity's
- 2026-11-17 · Acute urinary retention: Chronic retention was said to show as dribbling, a weak flow and repeated infections. Prostate Cancer UK lists leaking urine at night, a weak flow and a feeling of not emptying the bladder fully, and says urine infections are more likely (Second accuracy pass; Prostate Cancer UK's urinary problems page read as raw HTML) · The signs now follow the charity's list
- 2026-11-17 · Prostate cancer: note on plain words: A note said surveys consistently find men are not asked about leaking or erections. No survey was cited or linked (Second accuracy pass; every source linked from the living-with layer read as raw HTML) · Replaced with what Prostate Cancer UK says: many men avoid asking for help, and a health professional might not be used to talking about sex
- 2026-11-17 · Prostate cancer: talazoparib eligibility: The talazoparib with enzalutamide row gave the setting as castration-resistant disease only and said the label states that no FDA-approved test is available. Both are stale: a supplement approved on 26 October 2026 added HRR gene-mutated castration-sensitive disease on TALAPRO-3, and the label now words the indication as mutations detected by an FDA-authorized test. The row also said only six of the twelve genes are on the olaparib list when seven are (Second accuracy pass; the Talzenna prescribing information effective 28 October 2026 read as raw text from DailyMed, the openFDA drugsfda record for the supplement date, and the Lynparza label gene list) · Setting, note and source label rewritten for both disease states and the current test wording; six corrected to seven
- 2026-11-17 · Prostate cancer: AR-V7 and the taxane choice: The Scher 2018 survival figures (14.3 against 7.3 months, 19.8 against 12.8 months) were given as the result in all 142 men, with one hazard ratio and no interval. They are from the 70 men classed as high risk, and the AR-V7-positive comparison was not statistically significant (hazard ratio 0.62, 0.28 to 1.39) (Second accuracy pass; the abstract for DOI 10.1001/jamaoncol.2018.1621 read through Europe PMC and NCBI PubMed Central) · The sentence now names the high-risk subgroup, gives both hazard ratios with intervals and says the AR-V7-positive difference was not significant
- 2026-11-17 · Prostate cancer: PTEN loss: The CAPItello-281 cut-off was written as no cytoplasmic staining in a specimen with at least 90% viable malignant cells; the trial defined PTEN deficiency as at least 90% of viable malignant cells with no specific cytoplasmic staining, and the 25.3% is of 6,003 men with a valid test result. The row did not say that capivasertib with abiraterone was approved in the United States on 12 June 2026 for PTEN-deficient hormone-sensitive disease, and the PSMA row called itself the only row an approval depends on (Second accuracy pass; the abstract for DOI 10.1016/j.annonc.2025.10.004 read through Europe PMC, the Truqap label and the drugsfda record read on openFDA) · Cut-off and denominator corrected, the approval and its progression-free and overall survival results added with two sources, and the PSMA row now says one of the few rows
- 2026-11-17 · Prostate cancer: PARP inhibitor gene lists: ATM was said to be in three of the four prostate PARP inhibitor eligibility lists; it is in two of five (olaparib alone and talazoparib with enzalutamide). The olaparib row said the remaining genes had one or two evaluable men each including CHEK2; the label table shows BRCA1 6 men with no response, CHEK2 4, BRIP1 2, PALB2 2 and one each for four others (Second accuracy pass; the Lynparza, Talzenna, Rubraca and Akeega prescribing information read as raw text from DailyMed) · Both sentences corrected to the label's counts
- 2026-11-17 · Prostate cancer: molecular landscape figures: Seven smaller faults. The ERG meta-analysis was said to cover 48 further studies (47). RB1 copy loss in 54% of 44 aggressive-variant samples did not say 8 of the 44 were xenografts. The IDH1 cohort was called 99 men (99 samples from 96 men). TMPRSS2 was called the partner in 194 of 203 TCGA events, mixing records and samples (189 of 203 samples). HOXB13 was given as 4.5% of positives, a denominator the paper does not state. ZFHX3 was said to separate Asian from White men most clearly, and mismatch repair deficiency was called the only setting with durable checkpoint responses, neither of which the sources say. The Aurora kinase trial results were called disappointing with no source (Second accuracy pass; abstracts for DOIs 10.1158/1055-9965.EPI-12-0042, 10.1158/1078-0432.CCR-15-1259, 10.1001/jamaoncol.2018.6760, 10.1158/1078-0432.CCR-21-2577, 10.1001/jamaoncol.2018.5801 and 10.1158/1078-0432.CCR-18-1912 read through Europe PMC and PubMed, and the public cBioPortal API recounted on the named studies) · Each sentence corrected or reworded to the source; the alisertib phase 2 result (primary endpoint not met, six-month progression-free survival 13.4%) is now stated and cited
- 2026-11-17 · Prostate cancer: PSMA PET: The proPSMA management-change figures were given as 41 of 150 and 23 of 152, denominators the report does not give and that do not produce its 28%. The radioligand eligibility thresholds were described as one rule for the registration trials, when VISION measured organs, nodes and bone in short axis and PSMAfore measured organs and bone in longest diameter (Second accuracy pass; the abstract for DOI 10.1016/S0140-6736(20)30314-7 read through PubMed and the Pluvicto prescribing information read as raw text from DailyMed) · Denominators removed; the size rule now distinguishes the two trials
- 2026-11-17 · IPATential150: ipatasertib plus abiraterone and prednisolone in metastatic castration-resistant prostate cancer: The record said overall survival was not demonstrated and that the PTEN-loss population benefits, without the final analysis published in 2025, which found no overall survival gain (PTEN-loss hazard ratio 0.94, 95% CI 0.76 to 1.17; intention-to-treat 0.91, 0.79 to 1.07). (Second accuracy pass; the abstract of the final analysis (PMID 39884884, DOI 10.1016/j.eururo.2024.12.015) read through PubMed and the DOI resolved through Crossref.) · The summary line, the meaning line and the caveat now state the final survival result, and the 2025 paper is linked.
- 2026-11-17 · Reciprocal feedback regulation of PI3K and androgen receptor signalling in PTEN-deficient prostate cancer: The caveat described the ipatasertib trial as the human confirmation and stopped there. That trial showed no survival gain at its final analysis, and capivasertib with abiraterone has since entered the US label for PTEN-deficient metastatic disease. (Second accuracy pass; the IPATential150 final analysis abstract read through PubMed and the Truqap label (effective 16 September 2026) read through openFDA.) · The caveat now gives both later developments.
- 2026-11-17 · Nuclear-localised androgen receptor splice variant 7 in circulating tumour cells as a predictive biomarker in castration-resistant prostate cancer: The summary line said men with AR-V7 lived roughly twice as long on chemotherapy, and the prostate resistance table repeated the medians, without saying the comparison was not statistically significant (hazard ratio 0.62, 95% CI 0.28 to 1.39, P = .25), came from the 70 high-risk men only and was not randomised. A finding said the effect was confined to the high-risk group, which the abstract does not state. (Second accuracy pass; the abstract for DOI 10.1001/jamaoncol.2018.1621 read through Europe PMC. The full text could not be retrieved.) · The summary line, summary, findings and the resistance table row now give the subgroup, the confidence intervals and the lack of significance.
- 2026-11-17 · Prostate cancer: The resistance table said androgen receptor ligand-binding-domain mutations cluster on four residues while naming F877L among them. A recount found the commonest records at L702H, T878A, H875Y and W742, with F877L in only 3 of 72 records. The same table said only one or two men with each rarer repair gene were evaluable for response to olaparib, when four with CHEK2 were, and gave reasons for every TheraP screening exclusion that the abstract does not give. (Second accuracy pass; cBioPortal public API recounted for prad_su2c_2019 and prostate_msk_2024, the Lynparza label read through openFDA, and the TheraP abstract read through Europe PMC.) · The residues, the evaluable counts and the TheraP sentence now match the sources.
- 2026-11-17 · Genomic correlates of clinical outcome in advanced prostate cancer: The summary line presented the RB1 survival result as coming from all 429 men. The outcome analysis was in the 128 men who received abiraterone or enzalutamide as first treatment. (Second accuracy pass; the abstract and full text (PMC6561293) for DOI 10.1073/pnas.1902651116 read through Europe PMC.) · The summary line now names the 128-man subset and the 18 alterations tested.
- 2026-11-17 · Prevalence of microsatellite instability in prostate cancer and response to immune checkpoint blockade: The summary line said that in men with microsatellite instability immunotherapy worked and kept working. Six of 11 treated men had a PSA fall of more than half and 5 were still on treatment. (Second accuracy pass; the abstract for DOI 10.1001/jamaoncol.2018.5801 read through Europe PMC.) · The summary line now gives the counts.
- 2026-11-17 · The TMPRSS2-ERG rearrangement, ERG expression and prostate cancer outcomes: a cohort study and meta-analysis: The summary line said 48 other studies were pooled. The meta-analysis added 47 studies to the cohort. (Second accuracy pass; the abstract for DOI 10.1158/1055-9965.EPI-12-0042 read through Europe PMC.) · Corrected to 47 other studies.
- 2026-11-17 · Mutation of the androgen-receptor gene in metastatic androgen-independent prostate cancer: The summary line implied that in all five men with a mutation the receptor could be switched on by other hormones. Only two mutant receptors were tested for function. (Second accuracy pass; the abstract for DOI 10.1056/NEJM199505253322101 read through Europe PMC.) · The summary line now says two mutant receptors were tested.
- 2026-11-17 · Prostate-specific membrane antigen expression in normal and malignant human tissues: The meaning line said the 1997 survey predicted salivary and lacrimal gland uptake. The abstract reports staining in prostate, duodenum, proximal renal tubules and colonic neuroendocrine cells and none in other normal tissues. (Second accuracy pass; the abstract for PMID 9815541 read through Europe PMC.) · The line now refers to the renal tubules and falling expression only.
- 2026-11-17 · Exome sequencing identifies recurrent SPOP, FOXA1 and MED12 mutations in prostate cancer: Summary lines made claims the abstracts do not: that nobody had linked SPOP to prostate cancer (this paper), that gain of 8q carries the worst prognosis (Stopsack 2022, where it is called an adverse factor), and that the HOXB13 figure of 4.5% was a share of positive results (Nicolosi 2019). A pathway note called AR-V7 the only routinely measured marker needing intact circulating cells. (Second accuracy pass; the abstracts for DOIs 10.1038/ng.2279, 10.1158/1078-0432.CCR-21-2577, 10.1001/jamaoncol.2018.6760 and 10.1200/JCO.18.01731 read through Europe PMC.) · Each sentence reworded to what its source states.
- 2026-11-17 · Homologous recombination repair gene mutation in prostate cancer: The talazoparib row quoted a superseded US label: an indication for castration-resistant disease only, and a statement that no FDA-approved test was available. The summary also said six of talazoparib's twelve genes are missing from olaparib's list; it is five (Second accuracy pass; the current TALZENNA label (effective 28 October 2026) read as raw text from DailyMed and compared gene by gene with the LYNPARZA label) · Both quotations replaced with the current label's words, a second row added for hormone-sensitive metastatic disease (added October 2026 on TALAPRO-3, radiographic progression-free survival hazard ratio 0.48, 95% CI 0.36 to 0.65), the test statement removed, and six corrected to five
- 2026-11-17 · PTEN alteration: The prostate note said PTEN loss selects a treatment on the strength of IPATential150 alone, without its final analysis, and gave the 70% against 39% adverse event rates as if they were for the PTEN-loss group (Second accuracy pass; the final analysis (DOI 10.1016/j.eururo.2024.12.015) and the CAPItello-281 report read through PubMed, and the current TRUQAP label read through openFDA) · Note now says ipatasertib gave no overall survival gain in the PTEN-loss group (hazard ratio 0.94, 95% CI 0.76 to 1.17), that the adverse event rates are for all treated men, and that capivasertib with abiraterone is the licensed PTEN-selected treatment (US label since June 2026; hazard ratio 0.81, 95% CI 0.66 to 0.98)
- 2026-11-17 · AR-V7 splice variant: The prostate note gave survival of 14.3 against 7.3 months on a taxane as the result in 142 men. It is the result in the 70 men classed as high risk, and it was not statistically significant (Second accuracy pass; the abstract for DOI 10.1001/jamaoncol.2018.1621 read through Europe PMC) · Note now names the 70-man high-risk group, says the series was not randomised, and gives both hazard ratios with their intervals (0.62, 0.28 to 1.39; 1.67, 1.00 to 2.81)
- 2026-11-17 · SPOP mutation: The summary line said tumours with an SPOP mutation never have the ERG fusion, and the scoring rule called the two mutually exclusive (Second accuracy pass; recounted on the public cBioPortal API, where 5 of 318 SPOP-mutant samples in prostate_msk_2024 and 1 of 55 in the TCGA PanCancer deposit also carry an ETS structural variant) · Changed to almost never, with the counts stated
- 2026-11-17 · AR amplification (gene and upstream enhancer): The summary called AR amplification the first molecular explanation of endocrine treatment failure in a solid tumour, and said that in plasma it is the commonest finding and adds nothing to standard markers. Neither abstract says so (Second accuracy pass; the abstracts for DOI 10.1038/ng0495-401 and DOI 10.1158/2159-8290.CD-17-0937 read through Europe PMC) · Reworded to an early explanation in prostate cancer, and to the paper's own finding that detecting AR amplification did not outperform standard prognostic biomarkers
- 2026-11-17 · Prostate cancer: The molecular history said the 1995 receptor mutations let the antiandrogen itself switch the receptor on, that the 2004 overexpression finding produced abiraterone, that BRCA2 reversion was identified in 2017 as the mechanism of PARP inhibitor resistance, and that carriers' localised disease is treated rather than watched because of the 2013 cohort. It also called mismatch repair deficiency the only setting with durable checkpoint responses (Second accuracy pass; each cited abstract read through Europe PMC or PubMed, with PMID 19359544 for enzalutamide) · Each line reworded to what its source states: two mutant receptors activated by progesterone and oestrogen; enzalutamide only; reversion already known in ovarian and breast cancer and found in prostate cancer; cause-specific survival and the authors' own conclusion; only removed
- 2026-11-17 · Molecular characterisation of neuroendocrine prostate cancer and identification of new drug targets: The paper was called the first evidence that neuroendocrine prostate cancer is a distinct molecular disease, which its abstract does not claim, and the later trial result was left unstated (Second accuracy pass; the abstract for DOI 10.1158/2159-8290.CD-11-0130 and the alisertib phase 2 report (DOI 10.1158/1078-0432.CCR-18-1912) read through Europe PMC and PubMed) · First removed; the alisertib trial in 60 men is now stated, with its missed primary endpoint and six-month radiographic progression-free survival of 13.4%
- 2026-11-17 · Concordance of circulating tumour DNA and matched metastatic tissue biopsy in prostate cancer: The summary line, and the tumour fraction readout, said blood and biopsy were taken on the same day. The abstract says plasma was collected at the time of the biopsy (Second accuracy pass; the abstract for DOI 10.1093/jnci/djx118 read through Europe PMC) · Reworded to at the same time in both places
- 2026-11-17 · Prostate cancer in the UK: three source links that resolved to other papers: Three PubMed ids were wrong. The id given for Delon and colleagues on cancer incidence by ethnic group opened a paper on a brain ryanodine receptor, the id for the 2024 UK consensus on PSA testing opened a paper on asthma inhalers, and the id for the long-term STAMPEDE prostate radiotherapy results opened a biosensor paper. The PACE-B patient-reported outcomes paper was also cited under a title it does not have. (Second accuracy pass; every PubMed id in the file fetched from PubMed and its title compared with the citation.) · The links now point to PubMed ids 35233092, 39038964 and 35671327, each read and confirmed, and the PACE-B paper carries its published title and volume (European Urology 2026;90(4):321-327).
- 2026-11-17 · Prostate cancer in the UK: waiting times: Suspected urological cancer was called the worst-performing tumour group in England on the 28-day standard, and the introduction said prostate waiting times were the worst of any cancer. In July 2026 three groups were below urology (suspected acute leukaemia, other blood cancers and lower gastrointestinal cancer), and at 62 days five tumour groups were below prostate. The page also stopped at July 2026 for England and at March 2026 for Scotland and Northern Ireland, and said referrals were growing when they fell between July and September. (Second accuracy pass; NHS England monthly combined waiting-times files for April to September 2026, the Public Health Scotland compliance workbooks for the quarters to March and June 2026 and the Northern Ireland workbooks, all read as raw data.) · The ranking claims are corrected, and the figures are now September 2026 for England (65.6% at 28 days, 70.1% at 62 days, 92.6% at 31 days), the quarter to June 2026 for Scotland (42.5% at 62 days on 1,003 referrals) and April to June 2026 for Northern Ireland, with the earlier figures kept where they show the trend. The Welsh figures could not be re-read because the Welsh Government page refused the request, and the page says so.
- 2026-11-17 · Prostate cancer in the UK: the National Prostate Cancer Audit: The audit figures were those of the 2025 report and a source label said the 2026 patient report had no document attached. The 2026 State of the Nation report was published in October 2026. One sentence credited a proposal to "the evidence agent", which is not a source, and the high-risk treatment indicator was given a twelve-month window the report text does not state. (Second accuracy pass; the 2025 and 2026 State of the Nation reports read from the audit's PDFs.) · The 2026 figures are now given with their diagnosis years (12% metastatic at diagnosis in England for 2023, 6,312 of 51,389; 7% of low-risk men having radical treatment, 418 of 6,000; 69% of high-risk men having it, 13,983 of 20,296; 10,218 prostatectomies in 2025), the 2025 ranges are kept and dated, and the unsourced sentence is removed.
- 2026-11-17 · Prostate cancer in the UK: introduction and first steps: The introduction called prostate cancer the second most common cancer in men, when it is the most common in men and the second most common overall, and gave ten-year survival of 78.9% without saying it is a prediction for men diagnosed in 2018. The PSA preparation advice named cycling, which the NHS page does not; the PSA referral table was said to have replaced a single fixed threshold, which NICE does not say; blood in the semen after biopsy was said to last more than two weeks where NICE says at least two; and template biopsy was given a core count that the NICE guidance does not contain. (Second accuracy pass; Cancer Research UK statistics pages, the NHS PSA test page and NICE NG12, NG131 and HTG237 read as raw text.) · Each sentence now follows its source, with the years added to the incidence, mortality and survival figures.
- 2026-11-17 · Prostate cancer in the UK: the screening recommendation: The page said BRCA1 carriers were dropped from the recommendation because geneticists advised the risk was not large enough. The committee's coversheet says the model was rerun for BRCA1 and BRCA2 separately, with the extra risk assumed for BRCA1 carriers removed, and was cost effective only for BRCA2. (Second accuracy pass; the UK National Screening Committee coversheet of 26 March 2026 read from its PDF.) · Reworded to the coversheet's account.
- 2026-11-17 · Prostate cancer in the UK: stereotactic radiotherapy and the Cancer Drugs Fund list: Stereotactic radiotherapy was described as five fractions of 36.25 Gy, which reads as five times the dose; the policy gives 36.25 Gy in five fractions in total. The Cancer Drugs Fund list was cited as version 1.408 at an address that now answers 404. (Second accuracy pass; NHS England commissioning policy 2106 and the Cancer Drugs Fund list version 1.410 of 12 November 2026 read from their PDFs.) · The dose is restated, and the list is cited at its current version and address; section A still holds no prostate indication.
- 2026-11-17 · TRANSFORM: The sponsor was given as Imperial College Healthcare NHS Trust and University College London Hospitals. The registry names Imperial College London as sponsor and the two trusts as trial centres. The summary also said the screening committee's model tested PSA alone, when it modelled PSA followed by MRI and biopsy. (Second accuracy pass; the ISRCTN13801649 record read through the registry's API and the UK National Screening Committee coversheet.) · Sponsor and modelling sentence corrected.
- 2026-11-17 · Prostate Cancer UK: The record said the charity is paying for half of TRANSFORM, is its larger funder and its co-sponsor, and that it called the screening recommendation "a positive first step". The charity's page puts the trial at about 60 million pounds with government commitments of 16 million and then up to 18 million, the sponsor is Imperial College London, and the quoted phrase was not found. (Second accuracy pass; the charity's TRANSFORM page and health professional briefing and the ISRCTN record read as raw text.) · The record now says the charity funds the trial jointly with the National Institute for Health and Care Research and gives the sums as the charity states them.
- 2026-11-17 · Caroline Moore: PRECISION's comparator was called a standard twelve-core biopsy; the paper says 10 to 12 cores. (Second accuracy pass; the abstract for PubMed id 29552975 read from PubMed.) · Corrected.
- 2026-11-17 · Nicholas van As: The record said PACE-B's five-year patient-reported outcomes found no difference in pad use, leakage, sexual or bowel problems. The paper calls the outcomes comparable and reports 64% against 69% of men leak-free. (Second accuracy pass; the abstract for PubMed id 42476886 read from PubMed.) · Reworded with the pad-free and leakage figures from the paper.
- 2026-11-17 · Rakesh Heer: The record named him research lead on TRANSFORM and joint chief investigator of ELIPSE with a named colleague. The registry confirms him as principal investigator of ELIPSE but names neither the colleague nor his title on TRANSFORM. (Second accuracy pass; the ISRCTN records for ELIPSE and TRANSFORM read through the registry's API.) · The record now says what the registry confirms and what could not be re-read.
- 2026-11-17 · Prostate cancer in the UK: centre volumes and rankings: Velindre Cancer Centre was credited with 629 courses of radical prostate radiotherapy, which is the figure for all of Wales; the audit workbook gives Velindre 205. Addenbrooke's was given the second lowest adjusted readmission rate in England when its 4.8 percent is the lowest. Morriston was given the lowest readmission rate in Wales when Hywel Dda had none after 11 operations. The Christie's 2.6 percent genitourinary complication rate was called the second lowest of any high-volume centre, a rank the workbook does not give (Second accuracy pass; the National Prostate Cancer Audit 2025 data tables workbook and the State of the Nation PDF read directly) · Velindre now reads 205 on the centre card and in the funding table, of 629 across Wales; the three rankings are restated from the workbook
- 2026-11-17 · Prostate cancer in the UK: who sponsors which trial: TRANSFORM was said to be sponsored by Imperial College Healthcare NHS Trust and UCLH, and UCLH was called co-sponsor of TRANSFORM and STAMPEDE2. The registry names Imperial College London as TRANSFORM's sponsor, with the two trusts as trial centres, and University College London as STAMPEDE2's. Leeds Teaching Hospitals was called sponsor of POINTER-PC, which the University of Leeds sponsors. STAR-TRAP was said to recruit across the UK when the registry lists two centres, both at The Royal Marsden (Second accuracy pass; the ISRCTN records read through the registry's query interface) · Sponsors and centres now follow the registry
- 2026-11-17 · Prostate cancer in the UK: surgical volume and readmission: A GIRFT study was said to show 30-day emergency readmission falling from 10.6 to 7.0 percent with trust volume, in more than 35,000 prostatectomies. The paper reports 90-day emergency readmission, for robot-assisted operations, among 35,629 prostatectomies, and it was not among the sources cited (Second accuracy pass; the abstract for PMID 34133832 read through PubMed) · The sentence now says 90 days and robot-assisted, gives 35,629, and cites the paper
- 2026-11-17 · Prostate cancer in the UK: Cancer Drugs Fund list and the abiraterone forms: The Cancer Drugs Fund list was cited as version 1.408 of 24 September 2026, at an address that no longer answers; the current list is version 1.410 of 12 November 2026. Both abiraterone forms were said to carry an exception for men who had been in STAMPEDE. Only form ABI4, for metastatic disease, does; form ABI5, for non-metastatic disease, does not (Second accuracy pass; version 1.410 of the list found through NHS England's document interface and read as text) · The funding note, the STAMPEDE history and the four-nations row now cite version 1.410, which still has no prostate drug in the managed access section, and place the exception on form ABI4 only
- 2026-11-17 · Prostate cancer in the UK: waiting times for suspected urological cancer: Suspected urological cancer was called the worst-performing group on England's 28-day Faster Diagnosis Standard in July 2026. Suspected lower gastrointestinal cancer, at 67.7 percent, and both suspected blood cancer groups were lower. The figures were also two months behind the latest release (Second accuracy pass; NHS England's monthly combined data for April to September 2026 downloaded and computed) · The claim now reads among the lowest, and the September 2026 figures are added beside July's: 65.6 percent at 28 days, 70.1 percent at 62 days and 92.6 percent at 31 days for prostate cancer. Scotland's quarter to June 2026 is added too, 42.5 percent at 62 days
- 2026-11-17 · Prostate cancer in the UK: Northern Ireland waiting times: Northern Ireland was said to publish no 62-day split by tumour site, and its revision of September 2026 was said to have lowered every quarter from December 2024. The release of 1 October 2026 does give the split, and the revision reached back to the quarter ending March 2024 (Second accuracy pass; the Department of Health's revised 62-day publication and the April to June 2026 workbook read directly) · The page now gives 28.4 percent for the quarter to June 2026 and 84 of 448 people with urological cancer treated within 62 days, and dates the revision from March 2024
- 2026-11-17 · Prostate cancer in the UK: what TRANSFORM compares: A gap note said the four screening strategies in TRANSFORM are not listed in the registry record and appear only in charity and press material. The record's plain English summary lists all four. A second note said no update had been posted to the ELIPSE record, which was edited on 12 August 2026 without a change to its recruitment end date (Second accuracy pass; the ISRCTN records read through the registry's query interface) · The note now sets out the four prostate health checks from the registry, and the ELIPSE note says what the record shows
- 2026-11-17 · Prostate cancer in the UK: smaller corrections: Blood in the semen after biopsy was said to last more than two weeks in 60 of 100 men where NICE says at least two weeks. PACE-B's five-year patient-reported outcomes were said to show no difference where the paper calls them comparable and reports no test. The inherited disease test directory was cited at version 9 and its criteria at 9.1 when both are now 9.2 of 9 November 2026. The Scottish position on abiraterone for high-risk non-metastatic disease omitted the national off-label guidance that the Welsh review records (Second accuracy pass; NICE NG131 box 1, the abstract for PMID 42476886, NHS England's eligibility criteria version 9.2 and the All Wales OW20 page read directly) · Each sentence now follows its source
- 2026-11-17 · Prostate cancer in the UK: waiting-time figures: The figures table led with July 2026 for England and the quarter to 31 March 2026 for Scotland and Northern Ireland, while the pathway text above it had moved to September 2026 and the quarter to 30 June 2026, so the same page gave two different current figures for each standard (Second accuracy pass; the NHS England September 2026 monthly and April to September cumulative CSV files, the Public Health Scotland table 1 workbook for the quarter to 30 June 2026 and the Northern Ireland April to June 2026 workbook, each recomputed from the raw file) · Five figure rows now lead with the latest period: England 28-day 65.6 percent for suspected urological cancer, prostate 62-day 70.1 percent and 31-day 92.6 percent in September 2026; Scotland 42.5 percent at 62 days and 89.8 percent at 31 days; Northern Ireland 28.4 percent for all cancers. The earlier figures stay, dated, and the rows cite the new source entries
- 2026-11-17 · Prostate cancer in the UK: National Prostate Cancer Audit figures: The figures table, the complications step and the four-nations row gave the 2025 audit report as current although the 2026 report was published in October 2026. The share of men with newly diagnosed metastatic disease given treatment intensification (47 percent overall, 66 percent under 75 and 29 percent at 75 and over) was given with no warning (Second accuracy pass; the full State of the Nation Report 2026 PDF read as text (table 2 and sections 1.1 and 4)) · Seven figure rows, the complications step and the nations row now use the 2026 report: 12 percent metastatic at diagnosis (6,312 of 51,389), 7 percent of CPG 1 men treated radically (418 of 6,000), 69 percent of high-risk men treated radically (13,983 of 20,296), 12 percent readmitted (1,137 of 9,725), 10,218 prostatectomies in 2025. The intensification figures are now dated to the 2025 report with the reason the 2026 report dropped the indicator: the audit found systemic underreporting of abiraterone, apalutamide and enzalutamide use, so the figures probably understate treatment
- 2026-11-17 · Prostate cancer in the UK: genomic test directory: Two source entries gave the rare and inherited disease test directory as version 9 of 8 April 2026 and its eligibility criteria as version 9.1 of 20 May 2026, while the four-nations row said version 9.2 (Second accuracy pass; the NHS England publication page record and the version 9.2 eligibility criteria PDF read as text) · Both entries now point at the version 9.2 files of 9 November 2026; the R430 testing criteria are unchanged
- 2026-11-17 · PSA and MRI-first prostate cancer screening: The summary gave the ERSPC result only at 16 years (rate ratio 0.80, 570 men invited and 18 diagnosed per death averted) although the 23-year analysis was published in 2025 (Second accuracy pass; the abstract for PubMed 41160819 read through NCBI) · The 23-year figures are added (rate ratio 0.87, 456 men invited and 12 diagnosed per death averted) with a link to the paper
- 2026-11-17 · Prostate cancer: standard-of-care rows on hormone therapy with radiotherapy: The locally advanced row said docetaxel improved 4-year overall survival from 89 to 93 percent in RTOG 0521 and that DART 01/05 confirmed the benefit of long-course hormone therapy, and the intermediate-risk row gave the RTOG 94-08 10-year survival gain, all without the later analyses in which those survival differences were no longer significant. The after-surgery row cited RADICALS-RT without its primary outcome (Second accuracy pass; the abstracts for PubMed 37179241, 35427469, 25702876, 34481017, 28463149, 14581419 and 38583574 read through NCBI) · RTOG 0521 is now reported as no significant survival difference at a median 10.4 years (hazard ratio 0.89, 90 percent confidence interval 0.70 to 1.14); DART 01/05 as not significant at 10 years; RTOG 94-08 with survival converging at 18 years while prostate cancer deaths stayed lower; the RTOG 92-02 high-grade result as a subset outside the original design; RADICALS-RT with 10-year freedom from distant metastases of 93 against 90 percent, not significant, and 39 percent of the observation group irradiated
- 2026-11-17 · The cost of androgen deprivation: DART 01/05 was said to have improved 5-year overall survival from 86 to 95 percent with no mention that the 10-year final results did not confirm it, and the RTOG 92-02 high-grade survival result was given without saying the subgroup was not part of the trial design (Second accuracy pass; the abstracts for PubMed 35427469, 28463149 and 14581419 read through NCBI) · The 10-year DART result is added (78.4 against 73.3 percent, hazard ratio 0.84, 0.55 to 1.27, not significant), with the RTOG 92-02 caveat and its whole-trial survival gain at a median 19.6 years
- 2026-11-17 · Prostate cancer programmes that failed: PEACE-3 was said to have made bone-protecting agents mandatory for everyone and the text concluded that the harm in ERA 223 was in the missing supportive drug, not in the radium. Bone protection became mandatory only from March 2018, part way through recruitment, and the paper does not make that causal statement (Second accuracy pass; the abstract for PubMed 41763609 read through NCBI) · The sentence now gives the March 2018 date and the authors own conclusion that a bone-protecting agent is required to reduce skeletal complications
- 2026-11-17 · STAMPEDE: STAMPEDE2 was said to have opened in 2023 with three comparisons including niraparib with abiraterone (Second accuracy pass; the ClinicalTrials.gov record NCT06320067 read through the registry interface) · It opened to recruitment in June 2024 with two comparisons, stereotactic radiotherapy to metastases and lutetium-177 PSMA-617
- 2026-11-17 · Prostate cancer research roadmap: what to watch: TALAPRO-3 and CAPItello-281 were listed as awaited, with completion dates only, although both have reported and changed a US label (Second accuracy pass; the Talzenna and Truqap prescribing information read on DailyMed and the approval dates on openFDA) · Both items now say the trial has reported: talazoparib with enzalutamide was added to the US label on 26 October 2026 (hazard ratio for radiographic progression 0.48, 0.36 to 0.65, in 599 men) and capivasertib with abiraterone was approved on 12 June 2026
- 2026-11-17 · Prostate cancer: molecular sections: Androgen receptor mutations were said to cluster on four residues including F877L, which is uncommon; ETS fusions were called mutually exclusive with SPOP mutation; androgen receptor amplification was called the first demonstration of its kind in a solid tumour and the commonest plasma finding that adds nothing; and overdiagnosis was called unquantified to within a factor of thirty for a range of 1.7 to 67 percent (Second accuracy pass; mutation and structural variant records recounted through the cBioPortal public interface, and the abstracts for DOI 10.1038/ng0495-401, DOI 10.1158/2159-8290.CD-17-0937 and PubMed 24439788 read through Europe PMC and NCBI) · The residues are now L702H, T878A, H875Y and W742, with F877L at 3 of 72 records; the exclusivity is almost always, with 5 of 318 SPOP-mutant samples also carrying an ETS variant; the amplification claims are cut back to what the abstracts say; the factor is almost forty
- 2026-11-17 · Aparicio 2016, aggressive variant prostate cancer: RB1 copy loss was given as 54 percent of 44 samples without saying that the 44 were 36 tumours and 8 patient-derived xenografts (Second accuracy pass; the abstract for PubMed 26546618 read through NCBI) · The findings line and the history entry on the prostate page now give the split
- 2026-11-17 · Pettersson 2012, TMPRSS2-ERG and outcome: The findings line and a source label described a meta-analysis of 48 studies; the paper pooled its own cohort with 47 additional studies (Second accuracy pass; the abstract for PubMed 22736790 read through NCBI) · Both now say a meta-analysis including 47 additional studies
- 2026-11-17 · Abida 2019, microsatellite instability in prostate cancer: The record said the finding opens the only durable immunotherapy route in prostate cancer, which the paper does not claim (Second accuracy pass; the abstract for DOI 10.1001/jamaoncol.2018.5801 read through Europe PMC) · Reworded to a durable immunotherapy route
- 2026-11-17 · Tumour BRCA1/2 mutation and HRR gene alterations: Talazoparib with enzalutamide was described for castration-resistant prostate cancer only, with a quotation from a superseded label (Second accuracy pass; the Talzenna prescribing information effective 28 October 2026 read on DailyMed, and the supplement approval date on openFDA) · The summary now includes castration-sensitive disease, approved 26 October 2026, and the FDA-authorized test; the threshold row quotes the current label
- 2026-11-17 · PTEN loss: The term said ipatasertib improved progression-free survival in IPATential150 without saying the final analysis found no survival gain, and that capivasertib had reported a benefit in CAPItello-281 without its approval (Second accuracy pass; the abstract for PubMed 39884884 read through NCBI and the Truqap prescribing information on DailyMed) · It now gives the final overall survival result (hazard ratio 0.94, 0.76 to 1.17) and the US approval of capivasertib with abiraterone and prednisone on 12 June 2026
- 2026-11-17 · Sipuleucel-T: The summary line, its plain-language version and eight translations called it the first therapeutic cancer vaccine approved, with no regulator named (Second accuracy pass; the abstract for DOI 10.1158/1078-0432.CCR-10-3126 read through Europe PMC) · All now say the first approved by the US Food and Drug Administration, or in the United States
- 2026-11-17 · Atrasentan: The record is marked withdrawn, which a reader could take to mean the drug is off the market; it was approved in the United States on 2 April 2025 for a kidney disease (Second accuracy pass; the openFDA approval record for NDA 219208 and the Vanrafia label) · The summary now says the withdrawn status describes the prostate cancer programme and that the kidney indication is an accelerated approval
- 2026-11-17 · PROFIT: The trial was linked to the Canadian Cancer Trials Group as its sponsor (Second accuracy pass; the ClinicalTrials.gov record NCT00304759, which names the Ontario Clinical Oncology Group) · The link now goes to the Ontario Clinical Oncology Group
- 2026-11-17 · Hashim Ahmed: He was said to lead the PART and CHRONOS studies. CHRONOS completed in March 2023, and PART is sponsored by the University of Oxford with him as a co-author of its protocol (Second accuracy pass; the ClinicalTrials.gov record NCT04049747, the ISRCTN17249875 registry record and the Europe PMC record for DOI 10.1111/bju.70448) · The sentence now gives his role in each trial and the completion of CHRONOS with 100 participants
- 2026-11-17 · United Kingdom: the Cancer Drugs Fund: The link to the national Cancer Drugs Fund list went to version 1.408, an address that now answers not found; the same dead link was on the skin cancer UK page (Second accuracy pass; both addresses requested directly, and the version 1.410 list read as text) · Both links now go to version 1.410 of 12 November 2026
- 2026-11-17 · Lung cancer (all types): NELSON result in women: The summary and the low-dose CT screening note gave the NELSON death rate ratio in women, 0.67 (0.38 to 1.14), beside the men's result without saying it was not statistically significant, and the 2020 history entry gave 0.67 with no interval at all. Men were the primary analysis and women a subgroup of 2,594 (Second accuracy pass; the abstract for DOI 10.1056/nejmoa1911793 read through Europe PMC) · All three passages now say the men's result was the primary analysis and that the result in the smaller group of women was not statistically significant, with both intervals shown
- 2026-11-17 · Lung cancer (all types) and non-small-cell lung cancer: survival years: UK survival figures were given without the years they apply to: ten-year survival of 11.1 percent and the figures by age are for 2018, five-year survival of 18.1 percent in England and 43.5 percent after surviving one year are for 2015 to 2019, and the deprivation figures are for 2018 to 2022. The outlook answers also gave the United States five-year figure without its diagnosis years (Second accuracy pass; Cancer Research UK's lung cancer survival page and the SEER Cancer Stat Facts page read as raw HTML) · Each figure now carries its years on both records, including 2016 to 2022 for the United States figures
- 2026-11-17 · Sublobar resection (segmentectomy and wedge): CALGB 140503 was said to have randomised 697 American patients. The trial recruited from 83 institutions in the United States, Canada and Australia (Second accuracy pass; the full text for DOI 10.1056/nejmoa2212083 read through NCBI PubMed Central) · The sentence now names the three countries
- 2026-11-17 · Lung cancer in never-smokers: The OLIVE cohort was said to have been set up in 2026; its protocol was published in 2026 and gives a study start date in 2024. An estimate of adenocarcinomas attributable to particulate pollution was called the IARC's own, which the paper does not say. On the non-small-cell record the Million Women Study was called the largest UK prospective study and never-smoker disease the fastest-growing part of lung cancer, neither with a source (Second accuracy pass; the full text of the OLIVE protocol (DOI 10.1136/bmjresp-2025-003966) and the abstracts for DOI 10.1016/s2213-2600(24)00428-4 and DOI 10.1002/ijc.30084 read through Europe PMC) · The OLIVE dates are corrected, the estimate is credited to the Lancet Respiratory Medicine analysis, the study is described as large, and the unsourced growth claim is removed
- 2026-11-17 · Radon: The summary line called radon the second commonest cause of lung cancer after smoking and the largest natural source of radiation most people meet, and three passages called indoor radon the largest part of the 5 percent of UK cases caused by ionising radiation. The cited page says none of these, and its UK shares put workplace exposures (13 percent) and outdoor air pollution (8 percent) above ionising radiation (5 percent) (Second accuracy pass; Cancer Research UK's lung cancer risk page and the UKradon page read as raw HTML) · The summary line now gives the sourced figure, an estimated 9 percent of lung cancer deaths in Europe linked to indoor radon, and the three passages say the ionising radiation share includes indoor radon
- 2026-11-17 · Lung-RADS and nodule reporting: NELSON was said to have given 9.2 percent of screened men one extra scan; the paper says at least one additional scan. A remark that current nodule protocols should lower false positives was credited to the US Preventive Services Task Force; it is from the evidence report written for it, which says they might (Second accuracy pass; the abstracts for DOI 10.1056/nejmoa1911793 and DOI 10.1001/jama.2021.0377 read through Europe PMC) · The wording now follows both abstracts, here and in the low-dose CT screening note and the screening harms answer
- 2026-11-17 · Non-small-cell lung cancer: why United States mortality is falling faster: The summary said United States mortality is falling faster than incidence because more cancers are found early and because targeted drugs and immunotherapy work. The cited SEER page gives the two trends and no cause (Second accuracy pass; the SEER Cancer Stat Facts page for lung and bronchus cancer read as raw HTML) · The causal clause is removed; the two rates of fall stay
- 2026-11-17 · Lurbinectedin: The lung pages said the 2020 United States accelerated approval for relapsed small-cell lung cancer had been converted to a full approval on the IMforte data, and that lurbinectedin had no NICE appraisal and no United Kingdom marketing authorisation. This appeared on the drug record, the small-cell rows of the standard of care, the failed programmes term and the England access term (Second accuracy pass; the United States Zepzelca label effective 30 September 2026 and the approval record for NDA 213702 read through openFDA, the EMA Zepzelca page and NICE TA1198 read as raw pages) · The accelerated indication for relapsed disease was removed from the United States label in September 2026. Maintenance with atezolizumab was approved there on 2 October 2025 and in the EU on 29 May 2026. NICE ended its appraisal on 13 October 2026 without a recommendation because the company withdrew its submission. All four places now say so, and the United Kingdom authorisation claim, which could not be checked, is removed
- 2026-11-17 · Lung cancer trials open today (registry snapshot): The term gave 323 open trials, 146 of them phase 3, and listed a trial the registry now records as withdrawn. It also said lung cancer has more open phase 2 trials than any other cancer and that most will never report outside a local journal, with no source (Second accuracy pass; the registry recheck of 17 November 2026 and the ClinicalTrials.gov record for NCT07428044 read through the registry API) · The list now leaves out withdrawn and completed trials and holds 322, of which 145 are phase 3 and 27 phase 2/3. The withdrawn trial is linked from the failed programmes term. The two unsourced statements are removed
- 2026-11-17 · Lung cancer: the failed and stopped programmes, and why: The term said mobocertinib was withdrawn across 2023 and 2024, that necitumumab added about a month of median survival and is almost never used, that MERMAID-1 and MERMAID-2 were both stopped, and that the 50 percent PD-L1 cut-off is the direct consequence of CheckMate 026 (Second accuracy pass; the Federal Register notice 2024-15371 read as raw text, the abstract for PMID 26045340 read through Europe PMC, NICE TA411 read as a raw page, and the ClinicalTrials.gov records for NCT04385368 and NCT04642469 read through the registry API) · The United States approval of mobocertinib dates from 15 September 2021 and was withdrawn on 15 July 2024. Necitumumab added 1.6 months (11.5 against 9.9) and NICE does not recommend it (TA411). The registry lists both MERMAID trials as completed with 89 and 30 participants. The sentence about the cut-off now states only what happened
- 2026-11-17 · Lung cancer: KRAS G12C treatment: The standard of care said sotorasib and adagrasib beat docetaxel without an overall survival gain and that adagrasib has no recommendation. Neither trial abstract reports overall survival, and NICE had ended its adagrasib appraisal (Second accuracy pass; the abstracts for PMID 36764316 and PMID 40783289 read through Europe PMC, the United States Lumakras and Krazati labels read through openFDA and NICE TA1076 read as a raw page) · The row gives the progression-free figures (5.6 against 4.5 months, hazard ratio 0.66; 5.5 against 3.8 months, hazard ratio 0.58), says neither drug has a confirmed survival gain and that both United States lung indications are still accelerated approvals, and cites TA1076 (2 July 2025)
- 2026-11-17 · Lung cancer: antibody-drug conjugates, bispecific antibodies and later lines: Datopotamab deruxtecan, telisotuzumab vedotin and zongertinib were described as approved without saying the approvals are accelerated. Patritumab deruxtecan was listed as in phase 3 in one row although its application had been withdrawn. Tarlatamab was called the first T-cell engager to improve survival in a common solid tumour with no source, and the HARMONi-3 interim result was stated as fact (Second accuracy pass; the FDA approval notice for datopotamab deruxtecan read as raw HTML, the Datroway, Emrelis and Hernexeos labels and approval records read through openFDA, the Merck press release of 29 May 2025 read as raw HTML, the abstracts for PMID 40454646 and PMID 40057343 read through Europe PMC and the registry record for NCT05899608) · Each approval is now called accelerated, with its date. The patritumab deruxtecan withdrawal of 29 May 2025 is stated in both rows. The first claim is replaced by the DeLLphi-304 figures (13.6 against 8.3 months, hazard ratio 0.60). The HARMONi-3 sentence is labelled as a sponsor report that could not be re-read
- 2026-11-17 · Lung cancer: first-line and perioperative immunotherapy: The IMpower110 survival result was given from the first analysis only, and Neotorch was given as an interim result in stage III disease (Second accuracy pass; the abstracts for DOI 10.1016/j.jtho.2021.06.019 and DOI 10.1016/j.annonc.2026.10.094 read through Europe PMC) · The IMpower110 update is added (20.2 against 14.7 months, hazard ratio 0.76, no longer statistically significant), and the final Neotorch analysis of stage II and III disease (hazard ratio 0.52) is given beside the interim figure
- 2026-11-17 · Lung cancer: spinal cord compression and palliation: The palliation row credited metastatic spinal cord compression to NICE NG59, which is the low back pain guideline, and said MRI is done the same day. The row on extensive-stage small-cell disease said thoracic radiotherapy tripled two-year survival in CREST (Second accuracy pass; NICE NG59 and NG234 read as raw pages and the abstract for DOI 10.1016/S0140-6736(14)61085-0 read through Europe PMC) · The guideline is NG234. Its wording is used: MRI as soon as possible and always within 24 hours, radiotherapy within 24 hours of the decision to treat. The CREST figures (13 against 3 percent) stand and are now described as a secondary analysis
- 2026-11-17 · Lung cancer drugs in England: what NICE has recommended: The term left out adjuvant alectinib, which NICE recommends, and the ended appraisals of adagrasib and lurbinectedin, and gave the cemiplimab recommendation without its PD-L1 and ROS1 conditions. The exon 20 row said mobocertinib was withdrawn worldwide in 2023 (Second accuracy pass; NICE TA1014, TA1076, TA1198 and TA1165 read as raw pages, every NICE number in the file reopened and its date and recommendation read, and the Federal Register notice 2024-15371 read as raw text) · TA1014 (13 November 2024), TA1076 (2 July 2025) and TA1198 (13 October 2026) are added. The cemiplimab conditions are stated. The mobocertinib sentence gives the withdrawal of United States approval on 15 July 2024
- 2026-11-17 · Lung trial sponsor links: Three trials were linked to the wrong company because the sponsor name shared a word: a trial run by EXACT Therapeutics was linked to Exact Sciences, one by United Immunity to United Therapeutics, and one by Heron Neutron Medical to Heron Therapeutics. A fourth, now sponsored by Gilead Sciences, was linked only to Tubulis (Second accuracy pass; the sponsor and collaborator fields of 155 registry records read through the ClinicalTrials.gov API and compared with each linked company) · The three wrong links are removed and Gilead is added to the fourth
- 2026-11-17 · Lurbinectedin basket trial, small-cell lung cancer cohort: The page described the 2020 United States accelerated approval for relapsed small-cell lung cancer as standing, and said lurbinectedin had no United Kingdom marketing authorisation and no NICE appraisal (Second accuracy pass; the United States Zepzelca label effective 30 September 2026 read through openFDA and DailyMed, the EMA Zepzelca page and NICE TA1198 read as raw pages) · The page now says the accelerated approval indication was removed from the United States label in September 2026, that the European Union authorised lurbinectedin on 29 May 2026 for maintenance with atezolizumab only, and that NICE ended its appraisal on 13 October 2026 without a recommendation. The United Kingdom authorisation claim, which could not be checked, is removed
- 2026-11-17 · TIME2: The page said the talc pleurodesis group needed fewer further pleural procedures. The trial found the reverse (Second accuracy pass; the full abstract for DOI 10.1001/jama.2012.5535 read through PubMed) · 22 percent of the talc group needed a further pleural procedure against 6 percent with the catheter. The page now says so, and adds that adverse events were commoner with the catheter (21 of 52 against 7 of 54)
- 2026-11-17 · SQUIRE: The page said necitumumab has no NICE appraisal, gave the survival gain as about six weeks, did not say the European authorisation had ended, and said the companion trial INSPIRE was stopped for excess thromboembolic death (Second accuracy pass; NICE TA411 and the EMA Portrazza page read as raw pages, and the abstracts for DOI 10.1016/S1470-2045(15)00021-2 and 10.1016/S1470-2045(15)70046-X read through PubMed) · NICE TA411 (28 September 2016) does not recommend it. Median survival was 11.5 against 9.9 months (hazard ratio 0.84). The European authorisation expired on 18 February 2021. INSPIRE stopped enrolling on its monitoring committee's advice and found no survival benefit. A claim about blood clots in SQUIRE that the abstract does not make is removed
- 2026-11-17 · IMpower110: The page gave only the first analysis (median survival 20.2 against 13.1 months, hazard ratio 0.59) and described NICE TA705 without its PD-L1 condition (Second accuracy pass; the abstract for DOI 10.1016/j.jtho.2021.06.019 read through Europe PMC and the NICE TA705 recommendation read as a raw page) · The page now adds the update 17 months later, in which the difference was no longer statistically significant (20.2 against 14.7 months, hazard ratio 0.76, 95 percent confidence interval 0.54 to 1.09) and the lower PD-L1 groups showed no significant benefit, and gives the NICE threshold
- 2026-11-17 · Neotorch: The page gave the interim analysis of stage III patients as the result and said the treatment more than doubles the time before the cancer returns, which the paper does not state (Second accuracy pass; the abstracts for DOI 10.1001/jama.2023.24735 and 10.1016/j.annonc.2026.10.094 read through PubMed) · The final analysis of all 501 patients, published in October 2026, is added: event-free survival hazard ratio 0.52 (0.39 to 0.68) and overall survival hazard ratio 0.66 (0.49 to 0.89) at a median 48.2 months. The summary line now says the risk is roughly halved
- 2026-11-17 · GEMSTONE-301: The page called the trial the only phase 3 evidence for consolidation immunotherapy after sequential chemoradiotherapy and said sugemalimab had no stage III authorisation in the United Kingdom (Second accuracy pass; the abstract for PACIFIC-5 (DOI 10.1186/s13045-025-01768-1) read through PubMed, the EMA Cejemly page and NICE TA798 read as raw pages) · PACIFIC-5 (December 2025) is added. The European Union authorisation for unresectable stage III disease is stated in place of the unchecked United Kingdom claim, and the conditions of NICE TA798 are given. The hazard ratio (0.64, 0.48 to 0.85) is added
- 2026-11-17 · STARS and ROSEL pooled analysis: The page listed SABRTooth among larger trials still running, and said the registry enrolment figures were planned targets (Second accuracy pass; the abstract for DOI 10.1183/13993003.00118-2020 read through Europe PMC and ClinicalTrials.gov API v2 for NCT00840749, NCT00687986, NCT02468024 and NCT02984761) · SABRTooth was a feasibility study of 24 patients that concluded in 2020 a phase 3 trial was not feasible in the NHS. The registry figures (36 and 22) are actual enrolment. STABLE-MATES and VALOR are still recruiting
- 2026-11-17 · JCOG0802 / WJOG4607L: The page said CALGB 140503 recruited in North America, and gave no figures for the main result (Second accuracy pass; the full text of DOI 10.1056/NEJMoa2212083 read through NCBI PMC and the abstract for DOI 10.1016/S0140-6736(21)02333-3 read through Europe PMC) · CALGB 140503 recruited in the United States, Canada and Australia. The primary result is added: five-year overall survival 94.3 against 91.1 percent (hazard ratio 0.663, 0.474 to 0.927)
- 2026-11-17 · Dutch Bone Metastasis Study: The page said 1,157 patients were randomised (Second accuracy pass; the abstract of the 1999 primary report (DOI 10.1016/S0167-8140(99)00110-3) read through Europe PMC) · 1,171 patients were randomised, 585 to one fraction and 586 to six. The primary report is now linked
- 2026-11-17 · J-ALEX: The page said J-ALEX was the reason Japan approved alectinib at 300 mg twice daily (Second accuracy pass; the abstracts for DOI 10.1016/S0140-6736(17)30565-2 and 10.1007/s40265-014-0329-y read through Europe PMC) · Japan had already approved alectinib at that dose on the earlier single-arm AF-001JP study. The sentence is corrected and the hazard ratio (0.34) added
- 2026-11-17 · CREST: The page said thoracic radiotherapy tripled two-year survival and that the commonest serious side effects were in single figures (Second accuracy pass; the abstract for DOI 10.1016/S0140-6736(14)61085-0 read through Europe PMC and ClinicalTrials.gov API v2 for NCT04402788) · Two-year survival was 13 against 3 percent. Fatigue affected 11 against 9 patients. The follow-on RAPTOR trial is named, with its registry status of suspended for scheduled interim monitoring
- 2026-11-17 · IALT: The page gave the benefit as four to five percent and said weaker later results were part of why the LACE analysis was done (Second accuracy pass; the abstracts for DOI 10.1056/NEJMoa031644 and 10.1200/JCO.2009.23.2272 read through PubMed and Europe PMC) · The benefit is 4.1 points at five years. The long-term result (hazard ratio 0.91, 0.81 to 1.02, p=0.10 at 7.5 years) is given, and the unsupported link to LACE, which was published first, is removed
- 2026-11-17 · JBR.10, BRF113928 cohort C, Lung ART: Side-effect figures were given at the wrong grade or organ: JBR.10 neutropenia of 88 percent was called grade 3 to 4, BRF113928 grade 3 or 4 events were called grade 3 or worse, and Lung ART was said to damage the heart (Second accuracy pass; the abstracts for DOI 10.1056/NEJMoa043623, 10.1016/S1470-2045(17)30679-4 and 10.1016/S1470-2045(21)00606-9 read through PubMed and Europe PMC) · Each now matches its abstract: neutropenia of any grade 88 percent, grade 3 or 4 events 69 percent, and late grade 3 to 4 cardiopulmonary toxicity 11 against 5 percent. The main result figures for JBR.10 and Lung ART are added
- 2026-11-17 · CheckMate 057, IMpower150, OAK, CheckMate 026: NICE TA713 was described without its PD-L1-positive condition. Three summary lines made claims no source supports: OAK as the largest trial of its kind, IMpower150 as the one regimen with evidence in EGFR or ALK disease, and CheckMate 026 as the reason for the 50 percent PD-L1 cut-off (Second accuracy pass; NICE TA713 and TA584 recommendations read as raw pages and the trial abstracts read through Europe PMC and PubMed) · The NICE conditions are stated and the three lines are reworded to what the papers say. Overall survival figures for IMpower150 (19.2 against 14.7 months, hazard ratio 0.78) and the confidence interval for OAK are added
- 2026-11-17 · JROSG 99-1, AMPLE: Summary lines overstated two results: whole-brain radiotherapy was said to halve new brain tumours, and the home drain to save about three hospital days (Second accuracy pass; the abstracts for DOI 10.1001/jama.295.21.2483 and 10.1001/jama.2017.17426 read through PubMed) · Twelve-month brain recurrence was 47 against 76 percent, and the hospital-day medians were 10 against 12 with an estimated difference of 2.92. Both lines now give those figures
- 2026-11-17 · Non-small-cell lung cancer: The molecular table described the cBioPortal study nsclc_ctdx_msk_2022 as 2,621 MSK-IMPACT samples. It is 2,621 samples from 1,127 patients, of which 1,667 are plasma cell-free DNA and 954 are tumour tissue, which is why its rates for several genes read lower. (Second accuracy pass; the study record and its sample attributes read from the public cBioPortal API.) · The cohort label now gives the patient count and the plasma and tissue split, and the STK11 row notes the plasma share.
- 2026-11-17 · Non-small-cell lung cancer: Six counts in the molecular table did not match the deposited data: EGFR L861Q was given as 30 and 15 samples (29 and 13); the three uncommon EGFR mutations were summed to 117 samples when 88 carry at least one; the HER2 allele share was 42 of 110 records (112); HER2 S310F was 7 to 12 samples per cohort (3 to 12); EZR was the ROS1 partner in 4 events (5); and KIF5B and CCDC6 were the RET partners in 36 and 4 events (33 and 7 samples). (Second accuracy pass; every count in the table recounted from the public cBioPortal API on 17 November 2026, 161 counts matching and these six not.) · Each figure replaced with the recount.
- 2026-11-17 · Non-small-cell lung cancer: The KRAS row said no non-G12C allele has an approved inhibitor and that pan-RAS inhibitors are in trials. Daraxonrasib, a pan-RAS inhibitor, has since been approved in the United States, for metastatic pancreatic adenocarcinoma only. (Second accuracy pass; the indications section of the US label for Rasonque, effective 27 August 2026, read on openFDA.) · The row now says no inhibitor is approved for lung cancer, names the pancreatic approval and links the label.
- 2026-11-17 · Non-small-cell lung cancer: The tumour mutational burden section gave the US tumour-agnostic pembrolizumab threshold as 10 or more mutations per megabase and said the basket trial behind it had no lung cohort. The label now reads 13 or more, and 25 of the 220 high-burden patients in KEYNOTE-158 had small-cell lung cancer. (Second accuracy pass; the indications and section 14.18 of the US pembrolizumab label effective 16 October 2026 read on openFDA, and the archived FDA notice of 16 June 2020.) · Both passages now give the 2020 value and the current one, and say the trial had no non-small-cell lung cohort.
- 2026-11-17 · Non-small-cell lung cancer: The same section said CheckMate 227 enrolled by tumour mutational burden and that its burden endpoint was later dropped, citing the 2018 paper, and the PD-L1 table said CheckMate 026 randomised on PD-L1 of 5% or more. CheckMate 227 randomised by PD-L1 level and added a co-primary endpoint in high-burden patients by amendment; the 2018 paper says nothing about the endpoint being dropped; CheckMate 026 enrolled at 1% or more and analysed the 423 patients at 5% or more. (Second accuracy pass; the full text of DOI 10.1056/NEJMoa1801946 read through NCBI PMC and the abstracts for DOIs 10.1056/NEJMoa1910231 and 10.1056/NEJMoa1613493 through Europe PMC.) · Designs corrected; the unsupported sentence replaced with the 2019 overall survival result by PD-L1 level (17.1 against 14.9 months at 1% or more, 17.2 against 12.2 months below 1%) and that paper linked; the CheckMate 026 result marked not statistically significant.
- 2026-11-17 · Non-small-cell lung cancer: The resistance table said tumours with EGFR, RB1 and TP53 mutations were 5% of EGFR-mutant lung cancers but supplied 18% of small-cell transformations. In the source, 18% of those patients transformed (7 of 39) and no transformation occurred without both alterations. (Second accuracy pass; the abstract for DOI 10.1016/j.jtho.2019.06.002 read through PubMed.) · Sentence rewritten with 43 of 863, 7 of 39 and the absence of transformations in the other groups.
- 2026-11-17 · Non-small-cell lung cancer: The NRG1 row said lung cancer had the highest incidence of NRG1 fusions of any tumour type in the Jonna series. Lung had the most cases (25 of 9,592, 0.3%); gallbladder, pancreatic and kidney cancers had a higher rate, 0.5% each. (Second accuracy pass; the full text of DOI 10.1158/1078-0432.CCR-19-0160 read through NCBI PMC.) · Row now gives the denominator and says which tumour types had the higher rate.
- 2026-11-17 · Non-small-cell lung cancer: Nine statements in the molecular tables went beyond their sources: a single crizotinib response written as a rule that amplified MET exon 14 tumours respond most deeply; a two-centre cohort that was four centres; KEAP1 loss in adenocarcinoma and NOTCH1 loss in squamous cancer both called truncation when most KEAP1 records are missense and half the NOTCH1 records are; CDKN2A exon 2 skipping where the paper says exon 1 beta; superlatives for the piano subtype; tumour mutational burden said not to correlate with PD-L1 in any dataset; an uncited 30% and 60% for targetable drivers; and the HER2 indication called exon 20 when the US labels say tyrosine kinase domain mutations. (Second accuracy pass; abstracts read through PubMed and Europe PMC, the full text of DOI 10.1038/nature11404 through Europe PMC, the cBioPortal recount, and the US labels for zongertinib and sevabertinib on openFDA.) · Each sentence reworded to what its source states; the uncited percentages replaced with the EGFR rates of the cited cohorts.
- 2026-11-17 · Small-cell lung cancer: The subtype table said no trial has randomised on small-cell subtype. PRISM (SWOG S2409, NCT06769126), a randomised phase 2 trial recruiting since November 2025, assigns maintenance treatment by subtype A, N, P or I and SLFN11 status. (Second accuracy pass; the ClinicalTrials.gov record read through the registry API on 17 November 2026.) · The row now names the trial, says it has not reported and links the registry record. The PARP finding for the POU2F3 subtype is now attributed to cell-line models, as the paper has it.
- 2026-11-17 · The genomic landscape of SMARCA4 alterations and associations with outcomes in patients with lung cancer: The plain summary and a finding said both classes of SMARCA4 alteration do better than average on immunotherapy. The paper found no difference in progression-free or overall survival on checkpoint inhibitors by SMARCA4 status. What it found is that, among patients with SMARCA4 alterations, those given a checkpoint inhibitor lived longer than those who were not (hazard ratio 0.67), and that class 1 had the highest response rate (Second accuracy pass; the abstract for PubMed id 32709715 and the open full text PMC7641983 read through NCBI efetch) · The summary, the finding and the closing sentence now state the comparison the paper made and say that survival on these drugs did not differ by SMARCA4 status
- 2026-11-17 · Association of high tumor mutation burden in non-small cell lung cancers with improved outcomes of PD-L1 blockade: The plain summary said the study's dividing line was about twice the number the licensing authorities use, and a finding gave that number as 10. The current US pembrolizumab label defines a high tumour mutational burden as 13 or more mutations per megabase (Second accuracy pass; the openFDA label for Keytruda effective 16 October 2026 read as raw JSON, and the abstract and full text for PubMed id 35708671) · The summary now gives the study's threshold of more than 19 against the 13 in the current US label, and the finding says 10 was the cutoff earlier studies had explored
- 2026-11-17 · PD-L1 immunohistochemistry comparability study in real-life clinical samples: Blueprint phase 2: The plain summary said the slides were read by pathologists from thirteen countries. The paper says 24 pathologists from 15 countries (Second accuracy pass; the open full text PMC8386299 for PubMed id 29800747 read through NCBI efetch) · The summary now says 24 pathologists from 15 countries
- 2026-11-17 · Concurrent RB1 and TP53 alterations define a subset of EGFR-mutant lung cancers at risk for histologic transformation: The plain summary said these cancers stop responding three times sooner than the rest. The paper gives a median time to stopping the EGFR inhibitor of 9.5 months, against 12.3 months with a TP53 alteration alone and 36.6 months with neither RB1 nor TP53 altered, so neither comparison is threefold (Second accuracy pass; the abstract for PubMed id 31228622 and the open full text PMC6764905 read through NCBI efetch) · The summary and the finding now give the three medians and name the comparison groups
- 2026-11-17 · Effects of co-occurring genomic alterations on outcomes in patients with KRAS-mutant non-small cell lung cancer: The plain summary said that with a KEAP1 co-mutation immunotherapy barely worked at all. The paper reports shorter overall survival from the start of immune therapy (hazard ratio 3.54) and no difference in how long immune therapy lasted (Second accuracy pass; the abstract for PubMed id 29089357 and the open full text PMC5771996 read through NCBI efetch) · The summary now says the first chemotherapy stopped sooner and survival after starting immunotherapy was shorter
- 2026-11-17 · Mutational signatures associated with tobacco smoking in human cancer: The record and two pathway notes said the tobacco signature is confined to directly exposed tissues and credited the paper with explaining why KRAS G12C dominates in lung cancer and G12D in bowel and pancreatic cancer. The paper says the signature is found mainly in those tissues and does not mention KRAS. A third note said the signature is absent from never-smoker genomes, where the cited study says no strong tobacco signature was detected (Second accuracy pass; the abstract for PubMed id 27811275 and the open full text PMC6141049 read through NCBI efetch, the abstract for DOI 10.1038/s41588-021-00920-0 read through Europe PMC, and KRAS mutations in the 230 TCGA adenocarcinomas recounted through the cBioPortal API) · The wording is now found mainly in, the KRAS link is described as an inference from the signature being mainly C to A changes, and the never-smoker sentence uses the study's own wording
- 2026-11-17 · Comprehensive molecular profiling of lung adenocarcinoma: The plain summary said the study found three new driver genes and that a tenth of tumours with no obvious oncogene were driven by newly counted alterations. The paper names RIT1 and MGA, and reports NF1, MET, ERBB2 and RIT1 aberrations in 13% of cases, enriched in tumours lacking an activated oncogene (Second accuracy pass; the abstract for PubMed id 25079552 and the open full text PMC4231481 read through NCBI efetch) · The summary now names the two genes and gives the 13% figure as the paper states it
- 2026-11-17 · ALK resistance mutations and efficacy of lorlatinib: The record was typed as a randomised controlled trial. It is a biomarker analysis of a single-arm registrational phase 2 study (Second accuracy pass; the abstract for PubMed id 30892989 read through NCBI efetch) · The paper type is now translational
- 2026-11-17 · EGFR T790M: the two 2005 reports: The Pao report was described as published within weeks of the single-patient report, and the single-patient report called T790M the most studied resistance mutation in cancer, with no source. The PLoS Medicine paper appeared online on 22 February 2005 and the New England Journal of Medicine paper on 24 February 2005 (Second accuracy pass; the PubMed records for ids 15737014 and 15728811 read as raw XML through NCBI efetch) · Both summaries now say the two reports appeared in the same week, and the superlative is removed
- 2026-11-17 · Blood-based tumor mutational burden as a predictor of clinical benefit with atezolizumab: A caveat said the prospective randomised test did not confirm a benefit, without naming the trial, its result or a source (Second accuracy pass; the abstract for PubMed id 35995953 (DOI 10.1038/s41591-022-01933-w) read through Europe PMC) · The caveat now names BFAST cohort C and its primary result, a progression-free survival hazard ratio of 0.77 with a 95% confidence interval of 0.59 to 1.00, and the paper is linked
- 2026-11-17 · Lung pathway notes: PD-1 checkpoint, MYC and the resistance map: Three notes made claims their sources do not: that PD-L1 is the only protein biomarker changing first-line treatment, that the EGFR and ALK resistance map is the best characterised in oncology, and that MYC paralogues are mutually exclusive in small-cell disease according to Rudin 2019, which says only that MYCL is predominant in the ASCL1 state and MYC in the NEUROD1 state (Second accuracy pass; the open full text PMC6538259 read through NCBI efetch, the testing guideline abstract for PubMed id 29396253, and the abstracts of the cited resistance papers read through Europe PMC) · The two superlatives are removed and the MYC sentence now says what the review says
- 2026-11-17 · Using multiplexed assays of oncogenic drivers in lung cancers to select targeted drugs: The plain summary said a thousand tumours were tested for ten genes at once with a driver in two thirds. Of 1,007 tumours tested for at least one gene, 733 were tested for all ten, and the 64% is of those 733 (Second accuracy pass; the abstract for PubMed id 24846037 read through NCBI efetch) · The summary now says up to ten genes and that the two thirds applies to those tested for all ten
- 2026-11-17 · Molecular subtypes of small cell lung cancer (Rudin 2019): A caveat said no trial has randomised on subtype, and the lung cancer page listed the subtypes as having no prospective trial. PRISM (SWOG S2409, NCT06769126), a randomised phase 2 trial that assigns patients by subtype A, N, P or I and SLFN11 status, has been recruiting since November 2025 (Second accuracy pass; the ClinicalTrials.gov record for NCT06769126 read through the registry API on 17 November 2026) · The caveat and the open problem on the non-small-cell lung cancer page now name the trial and say it has not reported
- 2026-11-17 · SMARCA4 in lung cancer: The lung note said both classes of SMARCA4 alteration did better than wild-type tumours on checkpoint inhibitors. The paper reports a higher response rate for class 1 only, and no difference in progression-free or overall survival on these drugs by SMARCA4 status; the benefit it reports is against SMARCA4-altered patients who were not given them (Second accuracy pass; the abstract for PMID 32709715 read through PubMed and the open full text PMC7641983 read through NCBI) · The note now states the comparison the paper made and that survival on these drugs did not differ from wild-type tumours
- 2026-11-17 · Histologic transformation: The lung note said the EGFR, RB1 and TP53 triple-mutant group supplies 18% of transformations. In the source, 18% (7 of 39) is the share of triple-mutant patients who transformed, and no transformation was seen in any other group. The note also gave 17.8 months as a time after transformation; it is the median time from diagnosis to transformation (Second accuracy pass; the abstracts for DOI 10.1016/j.jtho.2019.06.002 and DOI 10.1200/JCO.18.01585 read through Europe PMC, and the open full text PMC6764905) · Both sentences corrected from the abstracts
- 2026-11-17 · PD-L1 tumour-cell score: The lung note said CheckMate 026 randomised at a PD-L1 level of 5% or more. The trial randomised patients at 1% or more and drew its primary analysis in those at 5% or more, where the hazard ratio of 1.15 was not significant (Second accuracy pass; the abstract and open full text (PMC6487310) for DOI 10.1056/NEJMoa1613493) · The note now gives both thresholds, the hazard ratio and that it was not significant
- 2026-11-17 · EGFR exon 20 insertion: The lung note gave 3.4% in never smokers. That is the ERBB2 exon 20 insertion rate in the same cohort; the EGFR figure is 4 of 232, 1.7%, as the landscape table in the same file states (Second accuracy pass; a recount of study lung_nci_2022 from the cBioPortal public API) · Figure corrected to 1.7% (4 of 232)
- 2026-11-17 · HER2 in lung cancer: The target note said immunohistochemistry has no selective role in lung cancer, and the HER2 mutation readout said kinase-domain missense variants are not covered by the exon 20 indications. The US label for trastuzumab deruxtecan carries an accelerated approval for previously treated solid tumours of any origin that stain 3+, and the labels for zongertinib and sevabertinib are worded as tyrosine kinase domain activating mutations, not exon 20 insertions (Second accuracy pass; the current Enhertu, Hernexeos and Hyrnuo labels read through openFDA) · Both notes now give the label wording
- 2026-11-17 · KRAS in lung cancer: Three statements. Non-G12C alleles were said to have no approved inhibitor with no qualification, although daraxonrasib was approved in the United States on 26 August 2026 for pancreatic cancer only. KEAP1 loss was said to make a tumour resistant to everything, which the cited paper does not say. On the KRAS G12C readout, STK11 and KEAP1 loss were each put at about a fifth of G12C tumours, which is the rate in all adenocarcinomas in the cited cohort; among G12C tumours they are about a third and about a quarter (Second accuracy pass; openFDA labels and approval records, the abstract for DOI 10.1158/1078-0432.CCR-17-1841, and a recount of five cBioPortal lung adenocarcinoma cohorts) · The notes, the KRAS G12D readout and the open problem on the lung cancer page now say no inhibitor is approved for lung cancer and name the pancreatic approval; the KEAP1 sentence gives what the paper found; the co-mutation rates are the recounted ones
- 2026-11-17 · NRG1 in lung cancer: The lung note called 0.3% the highest incidence of any tumour type in the 21,858-sample series. Lung had the most cases, but gallbladder, pancreatic and kidney cancers had higher rates at 0.5%. The RNA sequencing note said the fusion had a different partner almost every time, where CD74 was the partner in 12 of the 25 lung cases. Invasive mucinous adenocarcinoma was called the histology with no other driver, which the discovery paper does not say (Second accuracy pass; the open full text PMC7470623 for DOI 10.1158/1078-0432.CCR-19-0160 and the abstract for DOI 10.1158/2159-8290.CD-13-0633) · The target, biomarker, technology and glossary notes now give the counts from the paper
- 2026-11-17 · STK11 or KEAP1 loss in KRAS-mutant lung adenocarcinoma: The quotation given for the scoring rule was a paraphrase of the abstract with its hazard ratios removed and abbreviations expanded. The summary credited a survival hazard ratio of 1.96 to KEAP1 loss, where the paper reports it for co-mutation of KEAP1 or NFE2L2 (Second accuracy pass; the abstracts for PMID 34740862 and PMID 29089357 read through PubMed and Europe PMC) · The quotation is now the abstract's own conclusion sentence, and the summary names KEAP1 or NFE2L2
- 2026-11-17 · Non-small-cell lung cancer: squamous disease: An open problem said squamous cell lung cancer still has no targeted therapy; afatinib is licensed in the United States for metastatic squamous disease after platinum chemotherapy. A note said guidelines recommend full molecular testing in squamous disease only for never smokers, small biopsies and mixed histology, with no source; the European guideline's exceptions are patients under 50, never smokers, former light smokers and long-time ex-smokers (Second accuracy pass; the afatinib label read through openFDA and the Pan-Asian adapted ESMO guideline (DOI 10.1016/j.esmoop.2024.103996) read as open full text) · The open problem now says no therapy matched to one of its own alterations and names afatinib; the note gives the guideline's wording and links it
- 2026-11-17 · Tumour mutational burden: The lung note called 10 mutations per megabase the licensed threshold. The US indication was converted to traditional approval on 16 October 2026 at 13 or more (Second accuracy pass; the FDA announcement read as raw HTML and the current Keytruda label read through openFDA) · The note now calls 10 the historical threshold, since raised to 13
- 2026-11-17 · EGFR C797S: The statement that a first plus third generation combination can cover C797S in trans and nothing can in cis was not in either paper cited (Second accuracy pass; the abstracts for the two cited papers and for PMID 25964297 read through Europe PMC) · The sentence now says the finding comes from cell models and cites Niederst 2015, added as a link
- 2026-11-17 · Unsourced superlatives in the lung target notes: Four superlatives had no source: EGFR as the widest ancestry gap of any biomarker in solid tumours, the PI3K pathway as having the longest record of failed trials in squamous disease, the immune-cell PD-L1 score as the least reproducible measurement in lung pathology, and MET amplification of a mutated allele as predicting the deepest responses, which rests on one patient (Second accuracy pass; the cited abstracts read through Europe PMC and PubMed) · Each is replaced by what the source states, or removed
- 2026-11-17 · Lung cancer in the UK: Scottish cardiothoracic surgery activity: The source label for Public Health Scotland's inpatient activity gave 18,846 cardiothoracic episodes at the Golden Jubilee in 2024/25, 16,539 at the Royal Infirmary of Edinburgh and 4,900 at Aberdeen Royal Infirmary. Those totals add together every admission-type row in the file (all inpatients, all inpatients and day cases, day case, elective, emergency and transfer), which counts each episode about three times (Second accuracy pass; the open data file for activity by location and specialty queried through the opendata.nhs.scot datastore for specialty C4, with hospital codes resolved against the registry's own lookup) · Label now gives the all inpatients and day cases rows: 6,652, 5,549 and 1,745 episodes, read from the file as revised on 29 September 2026. Glasgow Royal Infirmary recording none is confirmed
- 2026-11-17 · Lung cancer in the UK: cemiplimab with chemotherapy in Scotland: The Scottish Medicines Consortium entry for cemiplimab with platinum chemotherapy showed only the 2024 non-submission refusal and set it against NICE TA1165 as a medicine England funds and Scotland does not. SMC2906, published 9 November 2026, accepted it for restricted use (Second accuracy pass; the Scottish Medicines Consortium advice pages for SMC2724 and SMC2906 read as raw HTML) · Label now records that SMC2906 superseded the refusal, with its stopping rule of 36 three-weekly cycles. Pending submissions for serplulimab (SMC3006) and tarlatamab (SMC2993) are noted on their entries
- 2026-11-17 · Lung cancer in the UK: Cancer Drugs Fund list and NICE counts: The Cancer Drugs Fund list was cited as version 1.408 of 24 September 2026 at an address that now answers 404; the list is version 1.410 of 12 November 2026. The NICE counts were 181 products and 16 terminated appraisals; NICE now lists 183 and 17, after TA1198 (lurbinectedin with atezolizumab as maintenance in extensive-stage small-cell lung cancer) was terminated on 13 October 2026 (Second accuracy pass; the NHS England publication record read through the site's REST interface and the list itself read as text; the NICE lung cancer topic page and products list read as raw HTML) · Link, version and date updated, with the lung rows still in the Cancer Drugs Fund section named; both NICE labels carry the new counts and the date they were read
- 2026-11-17 · Lung cancer in the UK: waiting-time sources: The Public Health Scotland waiting-times link was described as the January to March 2026 release, but the address now serves April to June 2026, published 29 September 2026. The English figures were for July 2026 only, with two later monthly releases out (Second accuracy pass; the Public Health Scotland publication page and compliance workbook, and NHS England's September 2026 statistical release and combined monthly file, read directly) · Scottish label corrected and the new quarter added (lung 84.0 percent at 62 days, 99.0 percent at 31 days); the September 2026 English figures added beside July's (all cancers 76.2, 91.5 and 70.7 percent; lung 77.7 percent at 28 days, 60.9 percent at 62 days and 33.5 percent from screening)
- 2026-11-17 · Lung cancer in the UK: trial results quoted without their significance: Three source labels overstated or under-described a result. The UK Lung Cancer Screening trial's relative rate of 0.65 was given without saying it was not statistically significant on its own. CONVERT's hazard ratio of 1.18 was given without its interval. QUARTZ was said to show whole-brain radiotherapy added no quality-adjusted survival, where the paper reports a difference of 4.7 quality-adjusted life days (Second accuracy pass; the abstracts for PubMed ids 34806061, 28642008 and 27604504 read through PubMed) · Labels now give 0.65 (0.41 to 1.02, not significant alone) and the pooled 0.84 (0.76 to 0.92); 1.18 (0.95 to 1.45, not significant); and the 4.7 day difference with no difference in overall survival
- 2026-11-17 · Lung cancer in the UK: the Scottish screening pilot: The Scottish Screening Committee minutes of January 2023 were cited for one lung screening pilot in Lothian and Fife. The minutes say only that Scotland has one pilot, about halfway through, and name no place (Second accuracy pass; the minutes page on gov.scot read as raw HTML and searched for both place names) · The place is removed from the source label, which now says the minutes do not name it
- 2026-11-17 · Lung cancer in the UK: registry records: MEDLEY was listed with five centres and a target of 900; the registry now lists seven centres and 842 people enrolled, with recruitment closed on 30 September 2026. QuicDNA was said to run across all six Welsh health boards, where the record lists six boards without saying they are all of them. QUARTZ LUNG gave no enrolment (Second accuracy pass; the ISRCTN records 87248310, 22734699 and 52137148 read through the registry's interface) · MEDLEY label updated; the word all removed for QuicDNA; QUARTZ LUNG now carries the 52 enrolled and the recruitment end date of 14 September 2026 the record gives
- 2026-11-17 · Philip Crosbie: The profile gave a current visiting senior lectureship at the University of Leeds, which his university page dates from March 2018 to March 2023. It called him scientific lead on PROMINENT, a study his page does not mention. The paper on screening location was presented as his without saying another author leads it (Second accuracy pass; the University of Manchester profile read as raw HTML, the abstract for PubMed id 31481631 read through Europe PMC, and the ISRCTN record for the Yorkshire Lung Screening Trial) · The lectureship is given with its dates, the PROMINENT clause is replaced by the biomarker sub-study role the page does list, and the location paper carries its first author and its figure of 23 percent of 938 attendees
- 2026-11-17 · Roy Castle Lung Cancer Foundation: Information days were said to include dietitians; the charity's page says nutritionists. The Maggie's source beside it gave a review date of 12 November 2024, which belongs to an NHS link on that page, and 29 UK centres where the page lists 30 (Second accuracy pass; the charity's face-to-face support page and the Maggie's lung cancer page read as raw HTML) · Wording corrected to the charity's own; the Maggie's label now says last reviewed November 2024 and 30 UK centres
- 2026-11-17 · Matthew Callister: The Yorkshire Lung Screening Trial eligibility paper was listed with a DOI that belongs to a different paper (LIBELULE, a French liquid biopsy trial) and with a title the paper does not have (Second accuracy pass; both DOIs resolved through Crossref and the abstract for PubMed 39709114 read through NCBI efetch) · DOI corrected to 10.1016/j.jtho.2024.12.016 and the title replaced with the published one
- 2026-11-17 · Gary Middleton: The vistusertib paper from the National Lung Matrix Trial was listed with a DOI that resolves to a melanoma paper (Second accuracy pass; both DOIs resolved through Crossref and PubMed 40069402 read through NCBI efetch) · DOI corrected to 10.1038/s41698-025-00838-4
- 2026-11-17 · Sanjay Popat: The chief medical officer appointment and the appointment years could not be re-read: the Royal Marsden page refused the request and the only archived copy is from August 2022 (Second accuracy pass; the Royal Marsden consultant page requested with curl (403) and its Internet Archive copy read) · A sentence added to the summary saying what could and could not be re-checked
- 2026-11-17 · SUMMIT (lung health check study): The sponsor was given as University College London Hospitals when the registry names University College London as lead sponsor. The enrolment note said a blood-test arm recruited separately, which no source read says. The summary line called it the biggest single UK lung screening cohort with no source (Second accuracy pass; the ClinicalTrials.gov record for NCT03934866 read through the v2 API and the abstracts for PubMed 40154514 and 36709053 read through NCBI efetch) · Sponsor corrected, the enrolment note now gives the 12,773 analysed and the 13,035 the registry records, and the summary line states the cohort size instead
- 2026-11-17 · CHART: The mature analysis was said to show a 24 percent reduction in the risk of metastasis across the trial; that figure is for squamous cell cancers only (Second accuracy pass; the abstract for PubMed 10577699 read through NCBI efetch) · The sentence now says the reduction in metastasis was in the squamous cell cancers
- 2026-11-17 · Big Lung Trial: The trial was described as having four primary-treatment settings; its papers describe three. The summary line said chemotherapy was worth nothing at all after surgery or radiotherapy, where both papers say no benefit was shown but the results remain consistent with a small one. The radical radiotherapy paper was quoted but not linked (Second accuracy pass; the abstracts for PubMed 15454647, 15200998, 10817793 and 16094738 read through NCBI efetch and Europe PMC) · Settings corrected to three, the summary line and summary reworded to no survival benefit shown, the p value added and the radiotherapy paper linked
- 2026-11-17 · VIOLET: The summary line said the trial settled the question with the same survival; the paper reports no difference at 52 weeks and says long-term outcomes need more follow-up. The summary also said three quarters of English lung resections are done by keyhole or robot, which the audit it cited does not state (Second accuracy pass; the abstract for PubMed 38319202 read through NCBI efetch and the National Lung Cancer Audit 2026 report and data tables searched as raw text) · Summary line reworded to no difference in survival at one year and the unsupported clause removed
- 2026-11-17 · TRACERx: The enrolment note said the cohort recruited 842 in total; the registry records 814 enrolled (Second accuracy pass; the ClinicalTrials.gov record for NCT01888601 read through the v2 API) · Figure corrected to 814
- 2026-11-17 · Lung cancer on the NHS: waiting times: The 31-day step said Scotland is the only nation that publishes a lung-specific split, though the same paragraph quotes lung figures for England and Northern Ireland. It gave a 16-day median from decision to treat to systemic therapy as an all-lung figure; in the audit it is the small-cell figure, measured from diagnosis. The Scottish range was given as if for all boards when three island boards were at 100 percent on one or two patients (Second accuracy pass; the NHS England monthly data file, the Public Health Scotland and Northern Ireland workbooks and the National Lung Cancer Audit 2026 report read as raw files) · The claim removed, the 16 days attributed to small-cell lung cancer from diagnosis, and the range limited to mainland boards. The Welsh figure is labelled as not re-checked because the Welsh Government page refused the request
- 2026-11-17 · Lung cancer on the NHS: small-cell lung cancer and stage 3: The small-cell step listed durvalumab and atezolizumab as the first-line options in NICE NG122 and left out serplulimab, which recommendation 1.11.3 now also lists (TA1167, 18 June 2026). It called oral topotecan the only recommended drug in relapse, where recommendation 1.12.3 also offers an anthracycline regimen or further platinum. The stage 3 step said NG122 offers chemoradiotherapy with surgery where the guideline says consider (Second accuracy pass; the NG122 management chapter, last reviewed 9 November 2026, and TA1167 read from nice.org.uk with curl) · Serplulimab added, the relapse sentence reworded to the guideline's own terms, and offer changed to consider
- 2026-11-17 · Lung cancer on the NHS: where it is treated: Norfolk and Norwich was said to have the lowest unadjusted 90-day mortality of any high-volume unit when Royal Papworth, a larger unit, had no deaths. Guy's was called the largest thoracic service in the UK on an audit that covers England and Wales. The Royal Marsden was called sponsor of HALT, which the Institute of Cancer Research sponsors, and the Leeds trials unit was called sponsor of CONCORDE and MEDLEY, which the University of Leeds sponsors. Southampton was listed as a MEDLEY and REFINE-Lung site, which neither registry record shows. The service specification was said to require three full-time surgeons; it says three (Second accuracy pass; the National Lung Cancer Audit 2026 data tables read as a workbook, sixteen ISRCTN records read through the registry API and the NHS England thoracic surgery specification read as text) · Each statement corrected to what the source records
- 2026-11-17 · Lung cancer on the NHS: what is funded: The EGFR row said TA653 covers disease after a first-generation inhibitor; NICE's wording is after first-line treatment with an EGFR tyrosine kinase inhibitor. The exon 20 row cited version 1.408 of the Cancer Drugs Fund list, which has been replaced. The introduction gave annual cases and deaths without their years and called the deprivation gap the widest of any cancer with no source (Second accuracy pass; TA653 read from nice.org.uk, version 1.410 of the national Cancer Drugs Fund list (12 November 2026) read as text and the Cancer Research UK statistics pages read as raw HTML) · TA653 wording corrected, the note now cites version 1.410 and the entries read there, the years added and the unsourced superlative removed
- 2026-11-17 · Lung cancer in the UK: cemiplimab in Scotland: The funding table, the funding note and two rows of the four-nations table said Scotland does not fund cemiplimab with chemotherapy, that a monotherapy submission was due on 9 November 2026, that Scotland refuses six lung medicines England funds and funds none of nine listed regimens (Second accuracy pass; the Scottish Medicines Consortium pages for SMC2906 and SMC2905 read as raw HTML, both published 9 November 2026) · Cemiplimab with chemotherapy is shown as accepted for restricted use (SMC2906) and monotherapy as accepted for locally advanced disease (SMC2905); the counts now read five refused and one of nine funded. The serplulimab resubmission (SMC3006) and a second tarlatamab submission (SMC2993) are noted as in progress
- 2026-11-17 · Lung cancer in the UK: terminated NICE appraisals: The funding note said sixteen lung appraisals were terminated, fifteen because a company gave no evidence submission, and that atezolizumab was terminated three times. Three were terminated for other reasons, atezolizumab was terminated twice, and a seventeenth, lurbinectedin with atezolizumab (TA1198), was terminated on 13 October 2026. The topic counts of 181 products and 32 appraisals in development were also out of date (Second accuracy pass; the NICE page for each terminated appraisal and the NICE lung cancer products list read as raw HTML on 17 November 2026) · The paragraph now gives seventeen terminations with the stated reason for each group, and the current counts of 183 products, 30 in development and 33 awaiting development beside the earlier ones
- 2026-11-17 · Lung cancer in the UK: Cancer Drugs Fund list version: Statements about the national Cancer Drugs Fund list rested on version 1.408, whose address now returns an error, and one described the list as 320 pages (Second accuracy pass; version 1.410 of 12 November 2026 downloaded from NHS England and read as extracted text) · The four lung entries in the Cancer Drugs Fund section and the funding dates quoted were confirmed in version 1.410, which is now named; tarlatamab is still absent from it
- 2026-11-17 · Lung cancer in the UK: trial statuses: MEDLEY was shown as recruiting after it closed on 30 September 2026 with 842 of 900 people enrolled. HALT follow-up was given as ending on 31 December 2026 when the record says 31 December 2027. QUARTZ LUNG was described as running when its record shows recruitment ended on 14 September 2026 with 52 of 448 enrolled (Second accuracy pass; the ISRCTN records for ISRCTN87248310, ISRCTN53398136 and ISRCTN52137148 read through the registry public API) · All three statuses corrected from the registry records
- 2026-11-17 · Lung cancer in the UK: CONCORDE and SEARCH credits: CONCORDE was said to have Corinne Faivre-Finn and Alastair Greystoke as joint chief investigators and SEARCH to have Roy Castle Lung Cancer Foundation support; neither registry record says so (Second accuracy pass; ISRCTN10142971 and ISRCTN11668189 read through the registry API and the CONCORDE protocol paper (PubMed 33072895) read in full text through Europe PMC) · CONCORDE now says the two are joint senior authors of the protocol paper and that the registry names no chief investigator; SEARCH now names its two registered funders
- 2026-11-17 · Lung cancer in the UK: UKLS deaths and the Big Lung Trial: The history said UKLS had 30 lung cancer deaths against 46 without saying the difference was not statistically significant. It said the Big Lung Trial tested four settings and found chemotherapy worth nothing after surgery or radical radiotherapy; there were three settings and the authors called both results consistent with a small benefit. It said QUARTZ found no quality-adjusted survival gain where the paper reports 4.7 days (Second accuracy pass; the abstracts for PubMed 34806061, 15454647, 15200998, 16094738, 16157935 and 27604504 read through NCBI efetch) · The UKLS sentence carries the relative rate 0.65 with its confidence interval 0.41 to 1.02 and the pooled 0.84; the Big Lung Trial sentence gives three settings with both hazard ratios; QUARTZ gives 46.4 against 41.7 quality-adjusted days
- 2026-11-17 · Lung cancer in the UK: claims with no source: Five statements had no source that says them: that three quarters of English lung resections are minimally invasive, that the National Lung Matrix Trial was the largest umbrella trial ever run in lung cancer, that the Manchester pilot parked scanners in supermarket car parks, that lung has the steepest deprivation gradient of any common cancer, and that England has the fastest stage shift of any common cancer (Second accuracy pass; the cited abstracts (PubMed 38319202, 32669708, 29440588), the Cancer Research UK mortality page and the National Lung Cancer Audit 2026 report read as raw text) · Three were removed and two reworded to what the paper says: the largest umbrella trial in non-small-cell lung cancer according to its 2020 report, and scanners next to local shopping centres
- 2026-11-17 · Lung cancer in the UK: Scottish deprivation figures: The incidence note called lung the steepest deprivation gradient of any cancer site in the Public Health Scotland workbook; cancer of the larynx is steeper at 4.42 against 3.16. The earlier mortality ratio was given as 3.09 and is 3.08 (Second accuracy pass; Public Health Scotland deprivation tables 3 and 5 downloaded and every cancer site ratio computed) · Lung is now described as second after larynx, and the ratio reads 3.08
- 2026-11-17 · Lung cancer in the UK: UK cases and Welsh operations: New cases a year were given as 50,218, a precision the Cancer Research UK page does not give; it says around 50,200. The Welsh resection row repeated 1,910 operations from the audit infographic, which cannot be 17.1 percent of the 2,135 people diagnosed in Wales (Second accuracy pass; the Cancer Research UK incidence page read as raw HTML, and the National Lung Cancer Audit 2026 report and data tables read as extracted text) · The value now reads around 50,200; the Welsh count was removed and the reason stated
- 2026-11-17 · Lung cancer in the UK: waiting times: Scottish lung waiting times were for the quarter to March 2026 although the quarter to June 2026 was published on 29 September 2026. The Welsh figure of 60.1 percent could not be re-read because the Welsh Government page refused a script (Second accuracy pass; Public Health Scotland cancer waiting times table 1 of 29 September 2026 downloaded and read; gov.wales requested with curl (HTTP 403)) · Scotland now shows 99.0 percent at 31 days and 84.0 percent at 62 days for lung, against 94.8 and 73.6 percent for all cancers, with the board range; the Welsh figure is labelled as not re-read
- 2026-11-17 · Lung cancer in the UK: Northern Ireland survival: The Northern Ireland figures cited to the registry lung cancer report could not be re-read: the report answered HTTP 403 and has no archived copy (Second accuracy pass; the report requested with curl and looked up on the Internet Archive on 17 November 2026) · A sentence on the survival by stage row says the figures were not re-read; the routes to diagnosis figures were confirmed from an archived copy
- 2026-11-17 · Lung cancer: NICE NG122 recommendation numbers around surgery and chemoradiotherapy: The decision rows, questions and notes cited NG122 recommendation numbers that NICE changed in November 2026: perioperative durvalumab and pembrolizumab were given as 1.6.9 and 1.6.10, postoperative systemic therapy as 1.6.11 to 1.6.13, adjuvant osimertinib and alectinib as 1.6.14 and 1.6.15, and chemoradiotherapy as 1.6.17. The text also said only durvalumab was restricted to disease without EGFR mutations or ALK rearrangements, and omitted perioperative nivolumab (TA1127) (Second accuracy pass; the NG122 management chapter and its update information read as raw HTML from nice.org.uk) · Numbers corrected to 1.6.9 (nivolumab, durvalumab and pembrolizumab, all for disease without EGFR mutations or ALK rearrangements), 1.6.10 to 1.6.12, 1.6.13 and 1.6.14, and 1.6.16 wherever they appeared
- 2026-11-17 · Lung cancer: osimertinib after chemoradiotherapy: The locally advanced decision row and its question described osimertinib after chemoradiotherapy (the LAURA setting) as something to ask about funding for, with no NICE recommendation. NICE recommended it in TA1156 on 21 May 2026 and NG122 now carries it as recommendation 1.6.17 (Second accuracy pass; the NG122 management chapter and the TA1156 overview page read as raw HTML from nice.org.uk) · Both places now state the NICE recommendation, its mutations and its condition
- 2026-11-17 · Lung cancer: patritumab deruxtecan after a targeted drug stops working: The resistance decision row, its question and the draft term offered patritumab deruxtecan "in the HERTHENA-Lung02 setting" as a later option. The US licence application was withdrawn on 29 May 2025 because overall survival in HERTHENA-Lung02 did not reach statistical significance (Second accuracy pass; the manufacturers' press release of 29 May 2025 read as raw HTML, and ClinicalTrials.gov record NCT05338970) · All three places now say it is not a licensed option and why
- 2026-11-17 · Small-cell lung cancer: first treatment for extensive-stage disease: The small-cell decision row, a question and the checkpoint inhibitor note said NG122 1.11.3 points to durvalumab and atezolizumab. It now also lists serplulimab with carboplatin and etoposide (TA1167, 18 June 2026), for people with a performance status of 0 or 1 (Second accuracy pass; the NG122 management chapter and the TA1167 overview page read as raw HTML from nice.org.uk) · Serplulimab added in three places
- 2026-11-17 · Lung cancer: brain imaging and advance care planning credited to NICE: A newly diagnosed question said brain imaging was only an open NICE research question, when NG122 1.2.24 to 1.2.26 give recommendations by stage. A palliative care question credited NICE's lung guideline with advance care planning wording it no longer contains (Second accuracy pass; the NG122 diagnosis chapter, its update information and the NG142 recommendations read as raw HTML from nice.org.uk) · The brain imaging answer now gives the three recommendations; advance care planning is credited to NG142 1.6.1 and 1.6.2
- 2026-11-17 · Lung cancer: wording credited to the NHS and Marie Curie: The smoking term said NHS Better Health states that 28 days smoke free makes you 5 times more likely to quit for good; the page says "much more likely" and gives no multiple. Two places said the NHS explains the difference between genetic and genomic testing; its page treats them as one. Three places said Marie Curie describes palliative care as available at any point after a diagnosis; the page says after a terminal diagnosis (Second accuracy pass; the NHS Better Health quit smoking page, the NHS genetic and genomic testing page and the Marie Curie palliative care page read as raw HTML) · Each statement reworded to what the page says
- 2026-11-17 · Osimertinib with or without chemotherapy in EGFR-mutated advanced NSCLC: The record said overall survival was immature and that whether the combination extends life was still awaited, and said the abstract was truncated so no further figures could be quoted. The final overall survival analysis was published in 2026. (Second accuracy pass; the abstracts for PMID 37937763 and PMID 41104938 (DOI 10.1056/NEJMoa2510308) read through PubMed.) · Added the final result: median survival 47.5 months against 37.6, hazard ratio 0.77 (0.61 to 0.96), with grade 3 or higher adverse events in 70 percent against 34 percent. Added the 24-month progression-free figures and rewrote the summary line, interpretation and caveat.
- 2026-11-17 · Adjuvant atezolizumab after adjuvant chemotherapy in resected stage IB-IIIA non-small-cell lung cancer (IMpower010): The summary said the intention-to-treat population cleared the significance bar, and the record still described overall survival as immature. At the final analysis the intention-to-treat result was not statistically significant, and five-year survival has been reported. (Second accuracy pass; the abstracts for PMID 34555333 and PMID 40446184 (DOI 10.1200/JCO-24-01681) read through PubMed, and the US atezolizumab label read through openFDA.) · Stated the final intention-to-treat disease-free survival hazard ratio of 0.85 (0.71 to 1.01, P 0.07) and the five-year overall survival hazard ratios, none significant. Removed an unsourced claim that this was the first such trial.
- 2026-11-17 · Durvalumab plus platinum-etoposide versus platinum-etoposide in first-line treatment of extensive-stage small-cell lung cancer (CASPIAN): The summary said almost nobody with this disease is alive at three years, and described the trial as the second positive one in thirty years with no source. The three-year update reports 17.6 percent alive with durvalumab against 5.8 percent. (Second accuracy pass; the abstracts for PMID 31590988, PMID 33285097, PMID 35279527 and PMID 30280641 read through PubMed.) · Replaced the wording with the three-year figures, gave the later tremelimumab result (hazard ratio 0.82, 0.68 to 1.00, not significant), removed the thirty-year claims and linked both updates.
- 2026-11-17 · Chemotherapy in non-small cell lung cancer: a meta-analysis using updated data on individual patients from 52 randomised clinical trials: The 5 percent five-year benefit of chemotherapy after surgery was given as an established result, and a caveat called it real. The abstract says the cisplatin results reached conventional significance with radical radiotherapy and with supportive care, not with surgery. The lung cancer timeline repeated the figure the same way. (Second accuracy pass; the abstract for PMID 7580546 read through PubMed.) · Said in the finding, the caveat and the 1995 timeline entry that the surgical comparison did not reach conventional significance, and that the figures are for cisplatin-based regimens.
- 2026-11-17 · Paclitaxel-carboplatin alone or with bevacizumab for non-small-cell lung cancer: The record called this the first time a targeted agent extended survival in lung cancer, which erlotinib had done in 2005, and said the abstract does not quantify treatment-related deaths when it reports 15 in the bevacizumab group, 5 from pulmonary haemorrhage. Two other first claims had no source. (Second accuracy pass; the abstracts for PMID 17167137 and PMID 16014882 read through PubMed.) · Removed the first claims, cited the erlotinib trial, and added the treatment-related death and bleeding figures.
- 2026-11-17 · First-line lorlatinib or crizotinib in advanced ALK-positive lung cancer: The summary said lorlatinib's main side effect is cognitive. The abstract lists hyperlipidaemia, oedema, weight gain, peripheral neuropathy and cognitive effects as the most common, with severe events mainly altered lipid levels. Long-term follow-up was described without figures. (Second accuracy pass; the abstracts for PMID 33207094, PMID 38819031 and PMID 42217582 read through PubMed.) · Corrected the side-effect sentence, added the five-year (60 percent against 8 percent) and seven-year (55 percent against 3 percent) progression-free figures, and said that no overall survival analysis has been reported.
- 2026-11-17 · Alectinib versus crizotinib in untreated ALK-positive non-small-cell lung cancer: A caveat said alectinib had been superseded in first line by lorlatinib, which no head-to-head trial supports, and the record lacked the final overall survival analysis. (Second accuracy pass; the abstracts for PMID 28586279 and PMID 41110693 (DOI 10.1016/j.annonc.2025.09.018) read through PubMed.) · Reworded the caveat and added the final overall survival result: median 81.1 months against 54.2, hazard ratio 0.78 (0.56 to 1.08), not statistically significant.
- 2026-11-17 · Osimertinib after chemoradiotherapy in stage III EGFR-mutated NSCLC: The record called the result the strongest hazard ratio in the lung cancer literature with no source, and gave no overall survival figures although two interim analyses exist. (Second accuracy pass; the abstracts for PMID 38828946 and PMID 42714840 (DOI 10.1158/1078-0432.CCR-25-4523) read through PubMed.) · Removed the superlative and added the interim overall survival hazard ratios, 0.81 (0.42 to 1.56) and 0.67 (0.40 to 1.14), neither statistically significant.
- 2026-11-17 · Anaplastic lymphoma kinase inhibition in non-small-cell lung cancer: A caveat gave 6.4 months as the length of follow-up. The abstract gives it as the mean duration of treatment. (Second accuracy pass; the abstract for PMID 20979469 read through PubMed.) · Corrected to mean treatment duration and noted that most patients had been treated before.
- 2026-11-17 · Identification of the transforming EML4-ALK fusion gene in non-small-cell lung cancer: The summary called the four years to crizotinib's approval the shortest interval from driver to drug in solid tumour oncology, with no source, and a caveat gave a later prevalence of 3 to 5 percent that the cited papers do not state. (Second accuracy pass; the abstracts for PMID 17625570 and PMID 20979469 read through PubMed, and the crizotinib approval record read through openFDA.) · Removed the superlative, gave the US approval date of 26 August 2011, and quoted the 2 to 7 percent range from the 2010 crizotinib report.
- 2026-11-17 · Comprehensive genomic profiles of small cell lung cancer: The record said the NOTCH finding led to DLL3 as a drug target. The paper that introduced DLL3 as a target found it through increased expression in patient-derived tumours. (Second accuracy pass; the abstracts for PMID 26168399 and PMID 26311731 read through PubMed, and the full text PMC4861069.) · Reworded both sentences and linked the DLL3 paper.
- 2026-11-17 · Molecular subtypes of small cell lung cancer: a synthesis of human and mouse model data: The summary named four authors and eighteen colleagues. The paper has eighteen authors in all. (Second accuracy pass; the author list for PMID 30926931 read through PubMed.) · Changed to fourteen colleagues.
- 2026-11-17 · Lung cancer: The timeline listed IMpower010 as due to complete on 31 August 2027 when the registry shows it completed on 26 August 2026, and an open problem said resistance arrives within one to three years for every targeted drug, which the seven-year CROWN follow-up contradicts. (Second accuracy pass; the ClinicalTrials.gov records for NCT02486718 and NCT03703297 read through the registry API, and the abstract for PMID 42217582 read through PubMed.) · Corrected the completion entry, noted that ADRIATIC's listed date has passed without a registry update, reworded the resistance line and named the endpoints behind the 2023 and 2024 timeline figures.
- 2026-11-17 · Lung cancer roadmap: MARIPOSA side effects and survival: The 2020 to 2026 step said amivantamab with lazertinib came at ten times the treatment discontinuation of osimertinib. The trial reported 10 percent against 3 percent stopping all treatment for treatment-related adverse events, about three times. The step also gave the 23.7 against 16.6 months and the CodeBreaK 200 gain of 1.1 months without saying both are progression-free survival, and stopped before the final survival results of FLAURA2 and MARIPOSA (Second accuracy pass; the abstracts for DOI 10.1056/NEJMoa2403614, 10.1016/S0140-6736(23)00221-0, 10.1056/NEJMoa2510308 and 10.1056/NEJMoa2503001 read through Europe PMC and PubMed) · Discontinuation corrected to about three times (10 percent against 3), the endpoints named, and the final survival results added: 47.5 against 37.6 months in FLAURA2 (hazard ratio 0.77) and a hazard ratio of 0.75 in MARIPOSA
- 2026-11-17 · Lung cancer roadmap: registry watch: Four watch rows were stale. FLAURA2 and MARIPOSA were listed as awaiting overall survival although both final analyses were published in 2025. IMpower010 was given an estimated completion of 31 August 2027 although the registry records it as completed on 26 August 2026. ADRIATIC was given an estimated completion of 23 October 2026, a date that has passed with the tremelimumab arm still unreported (Second accuracy pass; ClinicalTrials.gov records NCT04035486, NCT04487080, NCT02486718 and NCT03703297 read through the registry API on 17 November 2026, and PubMed abstracts 41104938, 40923797, 40446184 and 42971506) · Each row now states what has reported and what the registry says today; interim survival results added to the LAURA, AEGEAN and DeLLphi-304 rows
- 2026-11-17 · Lung cancer roadmap: small-cell lung cancer: The small-cell step said nothing happened for a quarter of a century after 1999 and the summary called the disease static, although the step itself cites IMpower133 and CASPIAN, which showed significant survival gains in 2018 and 2019. The closing step gave a 62 percent grade 3 toxicity rate and a two-month gain; CASPIAN reported grade 3 or 4 adverse events in 62 percent of both arms and gains were 2.0 and 2.7 months. The summary called tarlatamab the first drug built for the disease with no source, and said resistance arrives within three years for every targeted drug although lorlatinib has 60 percent of patients progression-free at five years (Second accuracy pass; abstracts for DOI 10.1056/NEJMoa1809064, 10.1016/S0140-6736(19)32222-6 and 10.1200/JCO.24.00581 read through PubMed, and the US tarlatamab label read through openFDA) · Reworded to what the trials report; the unsourced first claim removed; the lorlatinib exception named
- 2026-11-17 · Lung cancer roadmap: never-smoker disease: The roadmap and the never-smoker idea said about one lung cancer in five occurs in never-smokers and that the proportion is rising; the cited Sherlock-Lung paper gives 10 to 25 percent and no trend. The idea also said never-smokers are outside every screening programme in the world and that nobody screens for the disease, with no source (Second accuracy pass; the full text of DOI 10.1038/s41588-021-00920-0 read through NCBI PMC and the TALENT abstract (PubMed 38042167)) · Range given as the source states it, the rising clause and the universal claim removed, and the TALENT study of 12,011 never-smokers in Taiwan described
- 2026-11-17 · Run small-cell lung cancer as one platform with shared controls and subtype stratification: The idea did not say that a subtype-assigned randomised trial now exists, credited the subtype proposal to Rudin and eighteen colleagues when the paper has eighteen authors in all, and gave median survival in extensive-stage disease as under a year when the cited trials report 12.3 and 13.0 months with current treatment (Second accuracy pass; ClinicalTrials.gov NCT06769126 read through the registry API, the Europe PMC author list for PubMed 30926931, and the IMpower133 and CASPIAN abstracts) · PRISM (SWOG S2409), recruiting since November 2025, added; seventeen colleagues; about a year
- 2026-11-17 · Smoking, smoking cessation, and lung cancer in the UK since 1950: The short summary called the 2, 3, 6 and 16 percent figures lifetime risks of dying of lung cancer. The paper gives them as cumulative risks by age 75 (Second accuracy pass; the abstract for DOI 10.1136/bmj.321.7257.323 read through PubMed) · Now says by age 75, and that stopping at 50 avoids more than half the risk
- 2026-11-17 · Screening by chest radiograph and lung cancer mortality: PLCO: The summary said NLST could use a chest X-ray control because PLCO had already shown X-ray screening does not work. NLST was designed while PLCO was still testing it, and PLCO reported afterwards (Second accuracy pass; the NLST full text (DOI 10.1056/NEJMoa1102873) read through NCBI PMC) · Order of events corrected; the short summary now gives the two death counts rather than calling them the same
- 2026-11-17 · Effects of a combination of beta carotene and vitamin A on lung cancer: CARET: The summary said CARET found harm at the same time as the Finnish ATBC trial. ATBC reported in April 1994, two years earlier (Second accuracy pass; the ATBC abstract (PubMed 8127329) read through PubMed) · Now says two years after ATBC, with its figures and a link
- 2026-11-17 · Evaluation of USPSTF Lung Cancer Screening Guidelines Among African American Adult Smokers: The summary said Black smokers carry a higher lung cancer risk at any given exposure, which the abstract does not state (Second accuracy pass; the abstract for DOI 10.1001/jamaoncol.2019.1402 read through PubMed) · Replaced with the paper's own figures: median 25.8 against 48.0 pack-years among people who developed lung cancer, and 32 against 56 percent of them eligible
- 2026-11-17 · Socioeconomic inequalities in lung cancer treatment: The record and the deprivation idea said poorer patients were less likely to be offered treatment at the same stage. The review measured receipt of treatment, and reports the stage-adjusted finding for surgery only (Second accuracy pass; the abstract for DOI 10.1371/journal.pmed.1001376 read through PubMed) · Reworded to receipt, with the stage finding limited to surgery and the null result for radiotherapy added
- 2026-11-17 · A Study of Trastuzumab Deruxtecan in People With Non-Small Cell Lung Cancer: The record said the HERCULES phase 2 trial of trastuzumab deruxtecan before surgery for HER2-altered non-small-cell lung cancer was now recruiting, with 14 participants planned. The registry entry, updated on 2026-10-20, lists it as withdrawn with an enrolment of 0 and the reason "sponsor closed study" (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT07428044 read as raw JSON and compared field by field by script, then read by hand) · Status set to withdrawn and enrolment to 0; the one-line summary and the summary now say the trial was withdrawn with no one enrolled and quote the registry's reason word for word; the dates are removed
- 2026-11-17 · D-BACE in Combination With Chemotherapy and Camrelizumab for Resectable II-IIIA or Potentially Resectable T3-4N2 Stage IIIB NSCLC: The record called this a phase 3 trial with 70 participants, named the drug "Carelizumab" in its title and gave pathological complete response as the primary outcome. The registry entry, revised on 2026-10-26, gives phase 2, a single-arm single-centre design in its official title, 46 participants, camrelizumab in the title and major pathological response as the primary outcome; the quoted eligibility line had also been reworded (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT07016126 read as raw JSON and compared field by field by script, then read by hand) · Phase set to 2 and enrolment to 46; the name, official title, one-line summary, primary outcome and quoted eligibility line now carry the registry's current words
- 2026-11-17 · Adaptive ctDNA-Based Personalization of Treatment During Consolidation Therapy in NSCLC: The record carried the title "Personalized Escalation of Consolidation Treatment Following Chemoradiotherapy and Immunotherapy in Stage III NSCLC in Stage III NSCLC", with the last words repeated, named Maximilian Diehn as sponsor and gave a primary completion date of 2026-08. The registry entry, updated on 2026-09-30, is titled "Adaptive ctDNA-Based Personalization of Treatment During Consolidation Therapy in NSCLC" (ADAPT-C), gives Stanford University as lead sponsor and 2026-12 as the primary completion date (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT04585490 read as raw JSON and compared field by field by script, then read by hand) · Name, official title, sponsor, one-line summary and summary now follow the registry; the summary says which title and sponsor the record carried before
- 2026-11-17 · Study of TUB-040 in Participants With Unresectable or Metastatic Nonsquamous Non-small Cell Lung Cancer: The record named Tubulis GmbH as the sponsor of this phase 2 trial of TUB-040. The registry entry gives Gilead Sciences as the lead sponsor and lists no collaborator. The summary also said the trial "started 2026-11", a date the registry marks as anticipated (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT07806773 read as raw JSON and compared field by field by script, then read by hand) · Sponsor, one-line summary and summary now name Gilead Sciences, and the summary says which company the record named before; "started" is now "planned to start"
- 2026-11-17 · MOnaliZumab in Combination With durvAlumab (MEDI4736) for tRreatmenT of Small Cell Lung Cancer: The record named Hirva Mamdani as the sponsor of the MOZART trial. The registry entry, updated on 2026-11-05, gives Khalil Choucair as the lead sponsor and sponsor-investigator (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT05903092 read as raw JSON and compared field by field by script, then read by hand) · Sponsor, one-line summary and summary now name the registry's lead sponsor, and the summary says which name the record carried before
- 2026-11-17 · High Dose Dexamethasone to Reduce Postoperative Pain After Video-Assisted Thoracoscopic Surgery Lobectomy /Segmentectomy: The record gave the sponsor of the DEXTER trial as "Rigshospitalet, Denmark". The registry now writes the lead sponsor as "Copenhagen University Hospital, Rigshospitalet" (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT07213375 read as raw JSON and compared field by field by script, then read by hand) · Sponsor, one-line summary and summary now use the registry's name for the hospital, and the summary says which name the record carried before
- 2026-11-17 · Lung cancer: registry trials whose recruitment status changed: 26 records carried a recruitment status the registry has since changed: NCT07162883 (was planned, now active and not recruiting, entry updated 2026-10-22), NCT04875585 (was recruiting, now active and not recruiting, entry updated 2026-11-06), NCT05429463 (was recruiting, now active and not recruiting, entry updated 2026-11-04), NCT06545747 (was recruiting, now active and not recruiting, entry updated 2026-11-12), NCT07063329 (was planned, now recruiting, entry updated 2026-10-28), NCT07289646 (was planned, now recruiting, entry updated 2026-10-15), NCT07739199 (was planned, now recruiting, entry updated 2026-11-12), NCT07718737 (was planned, now recruiting, entry updated 2026-10-26), NCT07798947 (was planned, now recruiting, entry updated 2026-10-19), NCT07416058 (was planned, now recruiting, entry updated 2026-10-23), NCT07135882 (was planned, now recruiting, entry updated 2026-10-23), NCT07723482 (was planned, now recruiting, entry updated 2026-10-05), NCT07739966 (was planned, now recruiting, entry updated 2026-11-10), NCT07551635 (was planned, now recruiting, entry updated 2026-10-26), NCT07739186 (was planned, now recruiting, entry updated 2026-11-12), NCT07829549 (was planned, now recruiting, entry updated 2026-11-12), NCT07680790 (was planned, now recruiting, entry updated 2026-11-10), NCT07802821 (was planned, now recruiting, entry updated 2026-10-30), NCT07707895 (was planned, now recruiting, entry updated 2026-11-12), NCT07754123 (was planned, now recruiting, entry updated 2026-11-02), NCT07746388 (was planned, now recruiting, entry updated 2026-11-04), NCT07756645 (was planned, now recruiting, entry updated 2026-10-20), NCT07688577 (was planned, now recruiting, entry updated 2026-11-04), NCT07672223 (was planned, now recruiting, entry updated 2026-11-12), NCT07160686 (was planned, now recruiting, entry updated 2026-09-28), NCT07824882 (was planned, now recruiting, entry updated 2026-10-23) (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · Status set to active for the 4 that have closed to recruitment and to recruiting for the 22 that have opened; each one-line summary now says "active but no longer recruiting" or "now recruiting" to match
- 2026-11-17 · Lung cancer: registry trials with an unknown status: 4 records said the trial was now recruiting or not yet recruiting where the registry now gives the overall status as unknown, because the entry has not been verified for two years or more: NCT06469879 (last known status not yet recruiting, last verified 2024-06), NCT06028633 (last known status not yet recruiting, last verified 2023-09), NCT06048315 (last known status not yet recruiting, last verified 2023-09), NCT06177925 (last known status recruiting, last verified 2023-12) (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · Each one-line summary and summary now says what the trial was last listed as, that the registry gives its status as unknown, and when the entry was last verified; the status field is left as it was because the schema has no value for unknown
- 2026-11-17 · Lung cancer: registry trials enrolling by invitation: 5 records said "now recruiting" where the registry status is enrolling by invitation, which a reader cannot simply join: NCT06281964, NCT06370416, NCT06759857, NCT04762459, NCT06419179 (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · Each one-line summary now says "now enrolling by invitation only"
- 2026-11-17 · Lung cancer: registry trial dates moved by their sponsors: 34 records carried a start or primary completion date the registry has since changed: NCT07063329 (start 2025-08-08 now 2025-08-22); NCT07289646 (start 2026-01-15 now 2026-09-04, primary completion 2029-01-31 now 2030-01); NCT07739199 (start 2026-08 now 2026-10-10); NCT07718737 (start 2026-08-28 now 2026-09-30); NCT07798947 (start 2026-09 now 2026-10-07); NCT07416058 (start 2026-01-31 now 2026-04-30, primary completion 2027-12-31 now 2027-12-30); NCT07135882 (start 2026-08-25 now 2026-10-30); NCT07723482 (start 2026-08 now 2026-09-25, primary completion 2027-08 now 2027-10); NCT07739966 (start 2026-10-01 now 2026-10-13); NCT07162883 (start 2025-09-01 now 2025-11-21); NCT07551635 (start 2026-09-04 now 2026-04-28); NCT07739186 (start 2026-08 now 2026-10-21); NCT07829549 (start 2026-09-30 now 2026-10-10); NCT07680790 (start 2026-09-30 now 2026-09-08); NCT07802821 (start 2026-10-15 now 2026-10-27); NCT07707895 (start 2026-09-29 now 2026-11-03); NCT07754123 (start 2026-09-17 now 2026-10-23); NCT07746388 (start 2026-09-01 now 2026-09-19); NCT07688577 (start 2026-08 now 2026-10-21); NCT07821684 (primary completion 2031-04 now 2030-03); NCT07361484 (start 2026-06-01 now 2026-12-01); NCT07672223 (start 2026-06-25 now 2026-07-30); NCT07160686 (start 2026-03-02 now 2026-04-28, primary completion 2029-09-30 now 2029-10-30); NCT07824882 (start 2026-09-10 now 2026-10-08); NCT07153055 (start 2026-08 now 2027-01); NCT07149649 (start 2026-08 now 2026-09-29); NCT04585490 (primary completion 2026-08 now 2026-12); NCT05522660 (primary completion 2026-09 now 2027-06); NCT04875585 (primary completion 2023-12 now 2025-11-17); NCT04267848 (primary completion 2027-01-31 now 2030-01-31); NCT05429463 (start 2022-11-11 now 2022-12-08, primary completion 2025-11-11 now 2026-11-30); NCT06449313 (start 2026-08 now 2026-11); NCT07322341 (start 2026-10-01 now 2026-10-06); NCT04181060 (primary completion 2026-12-31 now 2029-06-30) (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · The dates in each summary now match the registry, with "started" kept only where the registry marks the start date as actual
- 2026-11-17 · Lung cancer: registry trials past their primary completion: 8 records said the trial was "due to reach its primary completion" on a date the registry records as an actual, past date: NCT06031558, NCT03811002 (results posted), NCT04804644, NCT04536675, NCT04214262 (results posted), NCT04155034, NCT03977194, NCT04875585 (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and the date type compared by script) · Each summary now says the trial "reached its primary completion" on that date. Two of them have results posted on the registry; no outcome figures were added to these registry-only records
- 2026-11-17 · Lung cancer: registry trials that have not started: 129 records said the trial "started" on a date the registry gives only as an anticipated start, some of them in the future: NCT07710612, NCT07355231, NCT07796100, NCT07477808, NCT07234877, NCT07169903, NCT05496166, NCT05186506, NCT07637513, NCT07285148, NCT07742956, NCT06970782, NCT07360132, NCT05859217, NCT07808138, NCT07501169, NCT05764928, NCT07704645, NCT05767892, NCT04153734, NCT06526624, NCT06908772, NCT06456138, NCT06847334, NCT06045767, NCT07560358, NCT07076095, NCT07256509, NCT06955325, NCT07135882, NCT07402070, NCT07545213, NCT07318883, NCT06581380, NCT07728955, NCT07242274, NCT06672068, NCT07453394, NCT06683703, NCT07633873, NCT07816380, NCT06704620, NCT06469879, NCT07343479, NCT06961188, NCT06319313, NCT07383116, NCT07158840, NCT07146230, NCT07551414, NCT07819591, NCT07636226, NCT07101627, NCT07819578, NCT06323148, NCT04846634, NCT06780839, NCT07802834, NCT07603791, NCT06028633, NCT07450183, NCT01394679, NCT07608159, NCT06858410, NCT05745740, NCT04465968, NCT06198907, NCT07283198, NCT07756645, NCT07154368, NCT05393466, NCT06961201, NCT07821684, NCT06102161, NCT06775483, NCT05503888, NCT07043907, NCT06380348, NCT07827261, NCT06416852, NCT06951399, NCT05990127, NCT07617389, NCT07464327, NCT07294261, NCT07414927, NCT07472647, NCT07106528, NCT06719336, NCT07197853, NCT07831863, NCT07442565, NCT06982924, NCT07806773, NCT06048315, NCT07609251, NCT07361484, NCT06254599, NCT06885840, NCT07833202, NCT06525350, NCT07538271, NCT07376382, NCT07229339, NCT04772287, NCT06335355, NCT07554859, NCT07715396, NCT07822542, NCT07150416, NCT06828757, NCT07662811, NCT07153055, NCT06177925, NCT07251582, NCT06523673, NCT06475209, NCT07485452, NCT07388524, NCT07809256, NCT06951646, NCT07008716, NCT06449313, NCT07738172, NCT07412262, NCT04829708, NCT07655622, NCT06672133, NCT06441344 (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and the date type compared by script) · Each summary now says "planned to start" with the registry date
- 2026-11-17 · Lung cancer: registry trial enrolment figures: 7 records carried an enrolment figure the registry has since changed: NCT04214262 (415 now 417, actual), NCT07739199 (200 now 326, estimated), NCT07162883 (122 now 146, estimated), NCT07739186 (120 now 166, estimated), NCT07688577 (140 now 180, estimated), NCT07672223 (570 now 586, estimated), NCT07016126 (70 now 46, estimated) (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · The enrolment field and the summary now give the registry figure
- 2026-11-17 · Lung cancer: registry trials with an incomplete list of primary outcomes: 16 records gave only the first three of the primary outcome measures the registry lists: NCT07710612 (5 measures), NCT07285148 (4 measures), NCT07063329 (6 measures), NCT07798947 (4 measures), NCT07256509 (9 measures), NCT07402070 (9 measures), NCT07723482 (7 measures), NCT07318883 (4 measures), NCT07688577 (5 measures), NCT06525350 (4 measures), NCT07559929 (6 measures), NCT06478108 (4 measures), NCT04871412 (6 measures), NCT07602855 (11 measures), NCT07655622 (4 measures), NCT06851663 (10 measures). In 1 more the sentence stopped in the middle of a phrase: NCT07718243 (3 measures). NCT07739199 and NCT07739186 listed three measures where the registry now lists four, the first a safety measure for a new part A of the phase 2 stage, and for NCT07739186 overall survival in the phase 3 stage (Second accuracy pass; the primary outcome measures in the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared with the sentence by script) · Where the registry lists up to four measures all are now given in full; where it lists more, the first three are given followed by the number of further measures
- 2026-11-17 · Lung cancer: registry trial condition lists cut short: 5 records ended their list of registered conditions in the middle of a word, because the generator cut the list at 250 characters: NCT01993810 (6 conditions), NCT05096663 (9 conditions), NCT03637816 (9 conditions), NCT06500455 (10 conditions), NCT06814496 (11 conditions) (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · Each summary now gives the registry's full list of conditions
- 2026-11-17 · Lung cancer: registry trial sites and one quotation: 10 records had no site countries where the registry now lists them: NCT07416058 (China); NCT07739966 (Argentina, Austria, Belgium, Brazil, China, France, Italy, Japan, Spain, Taiwan, Tunisia, Turkey (Türkiye), United States); NCT07162883 (Bosnia and Herzegovina, China, Georgia, Greece, Hungary, India, Poland, Romania, Spain, Thailand, Turkey (Türkiye), Ukraine, Vietnam); NCT07802821 (China); NCT07754123 (China); NCT07746388 (United States); NCT07756645 (Puerto Rico, United States); NCT07688577 (United States); NCT07672223 (China); NCT07824882 (China). NCT07160686 listed one United Kingdom site, the University of Birmingham; the registry now lists eight hospitals and the university is the sponsor only. NCT06284317 listed five United Kingdom sites; the registry now lists seven, adding Velindre University NHS Trust in Cardiff and the Beatson West of Scotland Cancer Centre in Glasgow. NCT04875585 quoted an eligibility line giving stages IA3 to IIIA; the registry entry reads "Histologically confirmed primary diagnosis of resectable NSCLC, stages IA2-IIIA (max. single station N2)." (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · A Sites sentence is added to each of the 10 records; the two United Kingdom site notes and counts now match the registry; the quotation is replaced with the registry's exact sentence
- 2026-11-17 · Lurbinectedin: The plain-language line, the small-cell lung cancer summary, the 2020 approval row and the 2020 history entry still gave relapsed small-cell lung cancer as a current United States approval. The US label effective 30 September 2026 lists only maintenance with atezolizumab and records the accelerated approval indication as removed in September 2026. The UK coverage row said the NICE position had not been researched (Second accuracy pass; the Zepzelca label and the submission history for NDA 213702 read through openFDA, NICE TA1198 read as raw HTML from nice.org.uk and the EMA Zepzelca page read as raw HTML) · Each place now says the relapsed indication was removed. The UK coverage row gives the EU authorisation of 29 May 2026 and the NICE appraisal ended on 13 October 2026 without a recommendation because the company withdrew its submission (TA1198). Eight translations updated
- 2026-11-17 · ATLANTIS: The summary said the drug stayed on the market after the trial missed its survival goal and that the 2020 accelerated approval was maintained, with the FDA accepting IMforte and LAGOON as confirmatory routes. The accelerated approval indication was removed from the US label in September 2026 (Second accuracy pass; the Zepzelca label effective 30 September 2026 read through openFDA) · The tldr, summary and plain-language line now say the approval the trial was meant to confirm was removed. Eight translations updated
- 2026-11-17 · DeLLphi-304: Called the first trial in which a T-cell engager improved survival in a common solid tumour. The paper makes no such claim and no source was cited (Second accuracy pass; the abstract for PubMed 40454646 read through NCBI efetch) · The tldr, summary and plain-language line now state the result without the word first: tarlatamab improved survival over chemotherapy (13.6 against 8.3 months, hazard ratio 0.60). Eight translations updated
- 2026-11-17 · Zongertinib: The tldr said approved in 2025 and moved to first line in 2026 without saying the indication is still under accelerated approval (Second accuracy pass; the Hernexeos label effective 13 August 2026 and the submission history for NDA 219042 read through openFDA) · The tldr now says accelerated approval in the United States in 2025, extended to first-line use in 2026 and still under accelerated approval. Eight translations updated
- 2026-11-17 · Lung cancer: SMARCA4 and immunotherapy: The SMARCA4 row said SMARCA4-altered tumours responded better than wild-type ones to checkpoint blockade, class 1 best of all, and the pathway note said the two classes differ in immunotherapy benefit. The paper found a higher response rate in class 1 only (P=0.027) and no difference in progression-free or overall survival on checkpoint inhibitors by SMARCA4 status (Second accuracy pass; the full text of Schoenfeld 2020 (PMC7641983) read through NCBI efetch) · Both passages rewritten to the paper: longer survival with a checkpoint inhibitor among SMARCA4-altered patients (hazard ratio 0.67), highest response rate in class 1, no survival difference by SMARCA4 status
- 2026-11-17 · Lung cancer: molecular notes without a source: Four statements went beyond their sources: that T790M became the most studied resistance mutation in cancer; that NRG1 fusions turn up in the histology with no other driver because they are a ligand event; that NFE2L2 hotspots are a cleaner signal than KEAP1 truncation (81 of 103 KEAP1 mutations in the TCGA adenocarcinoma set are missense); and that CheckMate 227 found its survival benefit did not depend on mutational burden, which the abstract does not state. The NRG1 paper tldr said the fusion was found most often in lung cancer, where the rate was 0.3 percent (Second accuracy pass; abstracts for PubMed 15737014, 15728811, 24469108 and 31562796 read through NCBI efetch, the full text of PMC7470623, and cBioPortal mutation records recounted) · The T790M sentence is removed and the other four are reworded to what the sources say
- 2026-11-17 · Lung cancer in the UK: waiting times: The 28-day, 31-day and 62-day paragraphs quoted July 2026 for England and the quarter to March 2026 for Scotland, while another row on the page already used the quarter to June 2026. Two later English releases exist (Second accuracy pass; the NHS England statistical release for September 2026 and the April to September monthly data file (lung rows recomputed), and Public Health Scotland table 1 of 29 September 2026 read from the spreadsheet) · England now reads September 2026: 76.2, 91.5 and 70.7 percent for all cancers; lung 77.7 percent at 28 days (60.2 on the screening route), 93.2 at 31 days and 60.9 at 62 days (33.5 on the screening route). Scotland now reads the quarter to June 2026: 99.0 percent at 31 days (1,095 of 1,106) and 84.0 at 62 days, boards from 61.5 to 100 percent. Three source entries added
- 2026-11-17 · Lung cancer in the UK: Scottish cardiothoracic surgery activity: Two centre cards and the thoracic surgery text still gave 18,846, 16,539 and 4,900 cardiothoracic episodes for 2024/25, totals that add every admission-type row together (Second accuracy pass; the Public Health Scotland open data records for specialty C4 recounted from the API response) · Now 6,652 at the Golden Jubilee, 5,549 at the Royal Infirmary of Edinburgh and 1,745 at Aberdeen Royal Infirmary, all inpatients and day cases
- 2026-11-17 · Lung cancer in the UK: statements the sources do not make: The page said a Scottish screening pilot had run in Lothian and Fife (the minutes name no place), that Scotland is the only nation publishing a lung-specific waiting-time split (England publishes one monthly), that QuicDNA ran in all six Welsh health boards (the record lists six without saying all), that lurbinectedin had no published NICE appraisal (TA1198 was terminated on 13 October 2026), and that the live Cancer Drugs Fund file is version 1.408 (it is 1.410). MEDLEY listed five of its seven centres (Second accuracy pass; the Scottish Screening Committee minutes of January 2023, NICE TA1198, the ISRCTN records for MEDLEY and QuicDNA and the NHS England file addresses read as raw text) · Each statement reworded to its source; the two missing MEDLEY centres added; source entries added for TA1198, SMC2906 and SMC2905
- 2026-11-17 · Philip Crosbie: The record linked him to The Christie, and the tldr and the Wythenshawe card said the Manchester pilot put CT scanners in supermarket car parks. His university page places him at the University of Manchester and Wythenshawe Hospital, and the pilot paper says the lung health checks were next to local shopping centres (Second accuracy pass; the University of Manchester profile read as raw HTML and the abstract for PubMed 29440588 read through NCBI efetch) · Institution changed to Wythenshawe Hospital and the link from The Christie removed; both descriptions now say next to local shopping centres
- 2026-11-17 · Alastair Greystoke: Described as joint chief investigator of CONCORDE. The registry record names no chief investigator; the protocol paper marks Faivre-Finn and Greystoke as joint senior authors (Second accuracy pass; ISRCTN10142971 read through the registry API and the full text of the protocol paper (PMC7548952)) · Now joint senior author, with Corinne Faivre-Finn, of the protocol paper
- 2026-11-17 · Gary Middleton: The tldr stated as present fact that the National Lung Matrix Trial is the largest umbrella trial in lung cancer. The claim is the 2020 paper own description, of non-small-cell lung cancer (Second accuracy pass; the abstract for PubMed 32669708 read through NCBI efetch) · The tldr now attributes the description to the 2020 report
- 2026-11-17 · Lung cancer: first 60 days checklist: The stop-smoking item said NHS Better Health says 28 days smoke free makes you 5 times more likely to quit for good; the page gives no multiple. The palliative care item in the small-cell checklist said Marie Curie says palliative care can be offered at any point after a diagnosis; the page says after a terminal diagnosis (Second accuracy pass; the NHS Better Health quit smoking page and the Marie Curie palliative care page read as raw HTML) · Both items now use the pages own words
- 2026-11-17 · Lung cancer: nodule management in the precision matrix: Said 9.2 per cent of screened men in NELSON had an additional scan. The abstract says that on average 9.2 percent of screened participants had at least one additional scan for an initially indeterminate result (Second accuracy pass; the abstract for PubMed 31995683 read through NCBI efetch) · Reworded to the abstract
- 2026-11-17 · Colorectal cancer: NordICC and deaths from colorectal cancer: The screening paragraph, the screening question and the 2022 history row quoted NordICC's fall in cancer incidence as evidence that screening saves lives and did not say that its difference in deaths from colorectal cancer was not statistically significant (0.28 against 0.31 percent at 10 years, risk ratio 0.90, 95 percent confidence interval 0.64 to 1.16). They also stopped at the 10-year report although a 13-year report was published in May 2026 (Second accuracy pass; the abstract for DOI 10.1056/nejmoa2208375 read through Europe PMC and the abstract for PMID 42102826 (DOI 10.1016/S0140-6736(26)00508-8) read through PubMed) · All three places now give the mortality result and say it was not significant; the summary adds the 13-year figures (incidence 1.46 against 1.80 percent, deaths 0.41 against 0.47 percent, not significant) and the paper is linked
- 2026-11-17 · Colorectal cancer: liver metastasis trials: The answer on liver spread said perioperative chemotherapy in EORTC 40983 added about 9 points to three-year progression-free survival, which is the figure for resected patients only; across all randomised patients the gain was 7.3 points and not statistically significant (hazard ratio 0.79, p=0.058). It also said cetuximab chemotherapy in CELIM raised the proportion that became operable from 32 to 60 percent; those figures come from a retrospective blinded review of scans in 68 patients, and 36 of 106 patients had a complete resection (Second accuracy pass; the abstracts for PMID 18358928 and PMID 19942479 read through PubMed) · The answer now gives the result for all randomised patients first and says it was not significant, then the resected subgroup, and describes the CELIM figures as what they are
- 2026-11-17 · Colorectal cancer: screening threshold: The summary and the open problems gave the English screening threshold of 120 micrograms per gram with no mention that the National Cancer Plan for England commits to lowering it to 80 micrograms per gram nationally by 2028 (Second accuracy pass; the accessible version of the National Cancer Plan for England read as raw HTML from GOV.UK, first published 4 February 2026) · Both places now state the commitment
- 2026-11-17 · Colorectal cancer: survival figures without their years, and guideline credits: Survival by age and by deprivation, and the figures in the outlook answer, were given without the diagnosis years they apply to. The diagnosis row credited NICE NG151 with imaging and a baseline carcinoembryonic antigen test it does not contain, the follow-up row credited NG151 with a colonoscopy schedule that comes from the BSG guideline, and one symptom line added "lasting more than a few weeks" to the NHS wording (Second accuracy pass; the Cancer Research UK survival pages, the SEER page, the NICE NG151 recommendations and the NHS symptoms page read as raw HTML, and the full text of DOI 10.1136/gutjnl-2019-319858 read through Europe PMC) · The years are stated (2018 for the UK age figures, 2018 to 2022 for deprivation, 2016 to 2022 for the United States, 2016 to 2020 and 2013 to 2017 for England); the rows say which steps NG151 covers; the symptom line uses the NHS wording
- 2026-11-17 · Micropapillary adenocarcinoma of the colon and rectum: The summary cited the published summary of the 2019 WHO classification (Nagtegaal 2020) for listing this subtype. That paper does not mention micropapillary adenocarcinoma (Second accuracy pass; the full text of DOI 10.1111/his.13975 read through Europe PMC) · The sentence now rests on the paper that describes the variant and says the full classification could not be re-read on 17 November 2026
- 2026-11-17 · Adenoma-like adenocarcinoma of the colon and rectum: The statement that the fifth edition of the WHO classification added this subtype was cited to Nagtegaal 2020, which does not mention it (Second accuracy pass; the full text of DOI 10.1111/his.13975 and the abstract for DOI 10.1016/j.humpath.2020.09.008 read through Europe PMC) · The citation now points to the 2021 paper, whose abstract states it
- 2026-11-17 · Lynch syndrome-associated colorectal cancer: The summary cited Nagtegaal 2020 for Lynch syndrome being the commonest inherited cause of bowel cancer, which that paper does not say, and said gynaecological risk-reducing surgery follows the BSG guideline, which does not cover it; a care row credited the same surgery to NICE NG151 (Second accuracy pass; the full texts of DOI 10.1111/his.13975 and DOI 10.1136/gutjnl-2019-319915 read through Europe PMC, and the NICE NG151 recommendations read as raw HTML) · The comparison now uses Cancer Research UK's figures for Lynch syndrome and polyposis; the summary and the row say neither guideline covers gynaecological surgery, and the BSG surveillance start ages (25 and 35 by gene) are given
- 2026-11-17 · Familial adenomatous polyposis-associated colorectal cancer: The summary called desmoid tumours a leading cause of death after colectomy, listed thyroid cancer, hepatoblastoma and medulloblastoma, and called duodenal and ampullary cancer the leading gastrointestinal risk after colectomy. None of this is in the sources cited; the BSG guideline says desmoids cause significant complications and pose a low risk of death. The surveillance start age was given only from an American review (10 to 12) on a page that says UK practice follows the BSG guideline, which recommends 12 to 14 (Second accuracy pass; the full texts of DOI 10.1111/his.13975 and DOI 10.1136/gutjnl-2019-319915 and the abstracts for DOI 10.1016/j.giec.2021.08.007 and DOI 10.1016/j.bpg.2022.101793 read through Europe PMC) · The desmoid and duodenal sentences now follow the BSG guideline and the review abstract, the unsourced list of other tumours is removed, and the BSG start ages for colonic (12 to 14) and upper gastrointestinal (25) surveillance are stated
- 2026-11-17 · Anti-EGFR rechallenge: The page said no randomised comparison against standard later-line treatment had reported. The VELO phase 2 trial randomised 62 patients in 2023, and CAVE-2 GOIM randomised 156 in 2026 (Second accuracy pass; the PubMed abstracts for PMID 37200022 (DOI 10.1001/jamaoncol.2023.0655) and PMID 41443411 (DOI 10.1016/j.annonc.2025.12.014) read through NCBI efetch) · Gives the VELO result (median progression-free survival 4.0 against 2.5 months, hazard ratio 0.48) and the CAVE-2 GOIM result (avelumab added nothing to cetuximab rechallenge, overall survival hazard ratio 1.00), says the randomised evidence is phase 2 only, and links both papers
- 2026-11-17 · Liver-limited metastatic bowel cancer: The EORTC 40983 gain in progression-free survival was quoted without saying it missed statistical significance in all randomised patients, and without the later finding of no difference in overall survival. The English five-year survival figure had no cohort years (Second accuracy pass; the abstracts for DOI 10.1016/s0140-6736(08)60455-9 and DOI 10.1002/bjs.7032 read through Europe PMC, and the long-term report (PMID 24120480) read through NCBI efetch) · States hazard ratio 0.79, p=0.058 in all randomised patients; adds five-year overall survival of 51.2 against 47.8 percent at a median of 8.5 years (hazard ratio 0.88, p=0.34) with its link; gives the English figure as 44.2 percent after liver resection in people operated on between 1998 and 2004
- 2026-11-17 · Screening uptake: A 35 to 41 percent reduction among people actually screened was credited to the 2022 NordICC paper, and NordICC was quoted with no mortality result. The achievable uptake level was given as 73 to 76 percent. Survival by deprivation had no diagnosis years. The 2019 change of test was said to have improved uptake with no source (Second accuracy pass; the abstracts for DOI 10.1056/nejmoa2208375, DOI 10.1007/s10654-025-01209-w and DOI 10.1136/gutjnl-2015-310691 read through Europe PMC, PMID 42102826 read through NCBI efetch, and the GOV.UK programme standards and Cancer Research UK survival pages read as raw HTML) · Credits the 35 to 41 percent estimate to the 2025 instrumental variable analysis; adds that deaths from colorectal cancer were not significantly lower at 10 years (0.28 against 0.31 percent) or 13 years (0.41 against 0.47 percent); gives the uptake standard as 62 and 73 percent with coverage at 62 and 76; dates the survival figures to 2018 to 2022; replaces the uptake claim with the 2014 pilot result of 66.4 against 59.3 percent
- 2026-11-17 · NHS bowel cancer screening programme: The page gave the screening threshold of 120 micrograms per gram with no mention of the planned change, gave modelled figures per 100,000 screened without the 15-year horizon, and gave the achievable uptake level as 73 to 76 percent (Second accuracy pass; the National Cancer Plan for England and the GOV.UK programme standards read as raw HTML, and the abstract for DOI 10.1038/s41416-022-01919-y read through Europe PMC) · Adds the commitment to 80 micrograms per gram rolled out nationally by 2028, with a link; adds over 15 years; gives the uptake standard as 62 and 73 percent and the coverage standard as 62 and 76
- 2026-11-17 · Colorectal cancer screening (colonoscopy, FIT, stool DNA, blood): NordICC was listed among the results screening rests on with its fall in incidence and no mortality result (Second accuracy pass; the abstract for DOI 10.1056/nejmoa2208375 read through Europe PMC and PMID 42102826 read through NCBI efetch) · Adds that there was no statistically significant fall in deaths from colorectal cancer at 10 or at 13 years, with the 2026 paper linked. The same sentence is added to the colonoscopy note
- 2026-11-17 · Adenoma-carcinoma sequence: The 53 percent fall in mortality in the National Polyp Study was written as if measured in the same 1,418 patients as the fall in incidence, and the summary line called this the first human cancer to be mapped this way with no source (Second accuracy pass; the abstracts for DOI 10.1056/nejm199312303292701 and DOI 10.1056/nejmoa1100370 read through Europe PMC) · Gives the mortality result as 53 percent lower than the general population would predict among 2,602 patients, and removes the unsourced claim
- 2026-11-17 · Watch and wait after chemoradiotherapy for rectal cancer: The summary line said about a quarter of regrowths happen, almost all within two years. The source reports regrowth in 25.2 percent of patients, with 88 percent of regrowths in the first two years (Second accuracy pass; the abstract for DOI 10.1016/s0140-6736(18)31078-x read through Europe PMC) · Says the tumour regrows in about a quarter of people, mostly within two years and nearly always in the bowel wall
- 2026-11-17 · Sidedness: NICE NG151 recommendation 1.4.1 was cited for recording tumour side on every metastatic case. The recommendation covers RAS and BRAF V600E testing only (Second accuracy pass; the NICE NG151 recommendations page read as raw HTML) · Attributes only the RAS and BRAF V600E testing to 1.4.1
- 2026-11-17 · Faecal occult blood test (guaiac): The page said three randomised trials made the case and described two, and said England switched every invitation to the newer test in June 2019 (Second accuracy pass; the abstracts for DOI 10.1056/nejm199305133281901 and DOI 10.1016/s0140-6736(96)03386-7 read through Europe PMC and the GOV.UK programme overview read as raw HTML) · Drops the count of three, and says the newer test was introduced in June 2019 and is now sent with every invitation
- 2026-11-17 · Total mesorectal excision (TME): The note said the first 50 curative operations had no recurrence at two years. The paper reports 50 operations classed as curative or conceivably curative, out of more than 100 consecutive resections, followed for over two years (Second accuracy pass; the abstract for DOI 10.1002/bjs.1800691019 read through Europe PMC) · Reworded to the paper's description
- 2026-11-17 · HERACLES criteria (HER2 in bowel cancer): The summary line gave the 5 percent prevalence for RAS wild-type tumours. The validation cohorts were KRAS wild-type (Second accuracy pass; the abstract for DOI 10.1038/modpathol.2015.98 read through Europe PMC) · Says KRAS wild-type
- 2026-11-17 · Cytoreductive surgery: PRODIGE 7 was placed twenty years after the 2003 Dutch trial. It was published in 2021, eighteen years later (Second accuracy pass; publication years for DOI 10.1200/jco.2003.04.187 and DOI 10.1016/s1470-2045(20)30599-4 read through Europe PMC) · Says eighteen years
- 2026-11-17 · AZUR-1: The record said the results were pending a congress presentation and that the US filing was under priority review with a decision date of February 2027, and credited the 2022 Memorial Sloan Kettering paper with more than 40 complete responses (Second accuracy pass; the abstract for pmid 42927858 read through PubMed, the Jemperli label and approval history read from openFDA, and the abstract for pmid 35660797) · The record now gives the first results (154 enrolled; 68 percent of 144 patients with a clinical complete response sustained for 12 months, interim analysis) and the accelerated approval of 19 October 2026 for stage II and III mismatch repair deficient rectal cancer. The 2022 paper is described as 12 of 12. The colorectal summary and history rows say the same
- 2026-11-17 · CheckMate 8HW: The record said overall survival was not yet mature, gave the 54.1 against 5.9 month result with arm sizes of 202 and 101 when it comes from the 171 and 84 patients with centrally confirmed tumours, called a median an average, and cited a news page (Second accuracy pass; the abstract for pmid 42926106 read through PubMed and the abstract for DOI 10.1200/JCO.2025.43.16_suppl.3501 read through Crossref) · Final overall survival added (hazard ratio 0.61 against nivolumab across all lines and 0.37 against chemotherapy first line; 0.66 against nivolumab first line, which did not meet its threshold). Arm sizes, confidence intervals and the source corrected, and the tldr now says median
- 2026-11-17 · KEYNOTE-177: The overall survival hazard ratio of 0.73 was quoted without saying that the final analysis did not reach statistical significance (Second accuracy pass; the abstracts for pmid 35427471 and pmid 39631622 read through PubMed) · The summary now gives the final analysis (hazard ratio 0.74, 95 percent confidence interval 0.53 to 1.03, superiority not demonstrated) beside the 5-year figures, with the comparator's 36.7 months and 44.2 percent. The link now uses the paper's DOI
- 2026-11-17 · ATOMIC: The confidence interval was given as 0.34 to 0.72 where the abstract says 0.35 to 0.72, the record carried only interim conference figures, did not say that the United States had approved the treatment, and claimed three firsts with no source (Second accuracy pass; the abstract for DOI 10.1200/JCO.2025.43.17_suppl.LBA1 read through Crossref, the abstract for pmid 41880612 read through PubMed, and the Tecentriq label read from openFDA) · Published figures entered (3-year disease-free survival 86.3 against 76.2 percent, hazard ratio 0.50, 0.35 to 0.73), the October 2026 label change added, and the unsourced firsts removed
- 2026-11-17 · Cetuximab: The record listed the 2024 United States approval with adagrasib for KRAS G12C colorectal cancer as current. The colorectal summary also offered adagrasib with an EGFR antibody as a standard option (Second accuracy pass; the Krazati label effective 31 August 2026 and its approval history read from openFDA) · The approval row is marked withdrawn in 2026 with the dates, the tldr and summary no longer present the combination as current, and the colorectal summary names sotorasib with panitumumab only
- 2026-11-17 · CAVE mCRC: The tldr said adding a PD-L1 antibody to cetuximab rechallenge helped. CAVE had one arm and could not show that, and the randomised trial that followed found no benefit from avelumab (Second accuracy pass; the abstract for pmid 41443411 (CAVE-2 GOIM) read through PubMed and the abstract for DOI 10.1016/j.esmogo.2026.100398 read through Europe PMC) · The tldr, summary and replication line now report CAVE-2 GOIM: median overall survival 14.8 against 12.9 months, hazard ratio 1.00 (0.65 to 1.52). The pooled comparison is described as unrandomised
- 2026-11-17 · CHRONOS: The record said anti-EGFR rechallenge had no randomised evidence (Second accuracy pass; the abstracts for pmid 37200022 (VELO) and pmid 41443411 (CAVE-2 GOIM) read through PubMed) · The summary now gives VELO (62 patients, median progression-free survival 4.0 against 2.5 months, hazard ratio 0.48) and CAVE-2 GOIM, and says both were phase 2
- 2026-11-17 · CIRCULATE-Japan: GALAXY was given as more than 2,000 patients and credited to the 2023 paper, which analysed 1,039. ALTAIR was described as adjuvant escalation and dated 2024 with no figures, and the DYNAMIC-III result was missing (Second accuracy pass; the abstracts for DOI 10.1038/s41591-022-02115-4, 10.1038/s41591-026-04428-0 and 10.1038/s41591-025-04030-w read through Europe PMC and pmid 39284954 through PubMed) · Patient counts corrected for each report. ALTAIR is described as treatment after standard therapy with its result (243 randomised, hazard ratio 0.79, 0.60 to 1.05). DYNAMIC-III added here, in the idea record and in the state of the art line, which had called de-escalation proven without limiting it to stage II
- 2026-11-17 · CRYSTAL and FIRE-3: The overall survival gain of 23.5 against 20.0 months was credited to the 2009 paper, which found no significant survival difference; it comes from the 2011 update. FIRE-3 was reported without saying its primary endpoint was not met (Second accuracy pass; the abstracts for pmid 19339720, pmid 21502544 and DOI 10.1016/S1470-2045(14)70330-4 read through PubMed and Europe PMC) · The figure is credited to the 2011 analysis with its hazard ratio, and the summary says FIRE-3 missed its response endpoint and that survival was a secondary endpoint
- 2026-11-17 · BREAKWATER: The record said the approval was the first under Project FrontRunner and was being converted toward full approval, and gave hazard ratios without intervals or the size of the FOLFIRI cohort (Second accuracy pass; the abstracts for DOI 10.1200/JCO.2025.43.17_suppl.LBA3500 and LBA3503 read through Crossref, pmid 40444708 and pmid 42219860 through PubMed, and the Braftovi label of February 2026 from openFDA) · The unsourced first is removed, the February 2026 label wording is stated, the survival result is marked interim, and intervals and arm sizes are added
- 2026-11-17 · Bevacizumab: Two side effects were said to carry boxed warnings, and a 2023 United States approval with trifluridine and tipiracil was listed as a change to the bevacizumab label (Second accuracy pass; the Avastin label and the Avastin and Lonsurf approval histories read from openFDA) · The notes say the current label has no boxed warning, and the 2023 rows say the approval was a change to the Lonsurf label on 2 August 2023
- 2026-11-17 · Trifluridine and tipiracil: Takeda was listed as a company behind the drug (Second accuracy pass; the Lonsurf label and approval history read from openFDA, which name Taiho) · The company link now points to Taiho Oncology
- 2026-11-17 · Fruquintinib: The drug and its trial record said it is better tolerated and has fewer off-target effects than regorafenib, while also saying the two have never been compared (Second accuracy pass; the abstract for DOI 10.1016/S0140-6736(23)00772-9 read through Europe PMC) · The comparative wording is removed, and the FRESCO-2 result is now sourced to the paper and not a company news page
- 2026-11-17 · NCIC CO.17: The KRAS analysis was said to show a trend to harm in KRAS-mutant tumours and to be the first negative predictive biomarker in solid-tumour oncology (Second accuracy pass; the abstract for pmid 18946061 read through PubMed) · The summary gives the paper's hazard ratios (0.55 in wild-type, 0.98 in mutant tumours) and drops the priority claim. The paper is linked
- 2026-11-17 · National Polyp Study: The replication line said NordICC found a smaller effect and called it the first randomised trial, without saying its fall in deaths was not statistically significant (Second accuracy pass; the abstracts for pmid 36214590 and pmid 42102826 read through PubMed) · The line gives the incidence and mortality risk ratios at 10 and 13 years and says the mortality difference was not significant
- 2026-11-17 · IDEA collaboration: The summary said neuropathy roughly halves with three months and that every guideline carries the two-tier recommendation. The CAPOX record repeated the halving (Second accuracy pass; the full text PMC6426127 read through NCBI and NICE NG151 recommendation 1.3.18 read as raw HTML) · Neurotoxicity figures entered (47.7 to 16.6 percent with FOLFOX, 44.9 to 14.2 percent with CAPOX), and the guideline sentence now states what NG151 offers
- 2026-11-17 · TRIUMPH: The tldr called it the first trial to let a blood test alone enrol people into a targeted treatment, and a 2026 analysis of 66 patients was described as TRIUMPH patients (Second accuracy pass; the full text PMC8604726 and the abstract for DOI 10.1016/j.esmoop.2026.107734 read through Europe PMC) · The tldr carries the authors' narrower claim, and the 2026 analysis is described as TRIUMPH and MyPathway combined
- 2026-11-17 · CheckMate 142: The tldr said this trial first showed checkpoint immunotherapy works in mismatch repair deficient bowel cancer and that a second antibody makes it work better (Second accuracy pass; the abstracts for DOI 10.1200/JCO.2017.76.9901 and pmid 26028255 read through Europe PMC and PubMed) · The tldr says the trial established nivolumab alone and with ipilimumab, and that the comparison between cohorts was not randomised
- 2026-11-17 · Colorectal cancer: The summary and burden gave about 1.9 million cases and 900,000 deaths, and a 2 percent yearly rise under 50 across high-income countries. The history said MSH2 and MLH1 were both cloned in 1993, and the deprivation survival figures had no diagnosis years (Second accuracy pass; the GLOBOCAN 2024 fact sheet text, the abstract for DOI 10.3322/caac.70067, the abstracts for pmid 8252616, 8128251 and 8145827, and the Cancer Research UK survival page) · The first paragraph now matches the corrected core summary (about 2 million cases and 920,000 deaths in 2024), the trend is the American Cancer Society figure of 3 percent a year at ages 20 to 49, MLH1 is dated 1994, and the survival figures carry 2018 to 2022
- 2026-11-17 · Rectal cancer trials: Dutch TME, Swedish trial, watch and wait database: Three tldr lines understated or overstated their sources: radiotherapy was said to halve local recurrence where the falls were from 8.2 to 2.4 and 27 to 11 percent, and almost all regrowth was said to come within two years where the source says 88 percent. The database was also credited with a surveillance schedule the abstract does not give (Second accuracy pass; the abstracts for DOI 10.1056/NEJMoa010580, pmid 9091798 and DOI 10.1016/S0140-6736(18)31078-X) · Each line now follows the source's figures, the schedule sentence is replaced by what the abstract says about surveillance, and the Swedish record gives its survival figures
- 2026-11-17 · NICHE-2: Three-year disease-free survival was sourced to a 2024 conference link that could not be resolved, and the summary said no patient needed chemotherapy, which neither abstract states (Second accuracy pass; the abstract for pmid 42906564 read through PubMed) · The endpoint is sourced to the 2026 Lancet report, five-year overall survival added, and the chemotherapy sentence removed
- 2026-11-17 · PARADIGM: The summary and the left-sided pairing called the result the longest survival seen in first-line trials, with no source (Second accuracy pass; the abstracts for DOI 10.1200/JCO.2022.40.17_suppl.LBA1 and pmid 37071094) · The claim is removed in both places; the summary adds that progression-free survival did not differ
- 2026-11-17 · Adagrasib: The colorectal approval row, the KRAS G12C standard-of-care row, the England drug access summary and an open problem all described adagrasib with cetuximab as a current US accelerated approval awaiting its confirmatory trial. The indication was removed from the US label in August 2026 and the FDA lists the accelerated approval as withdrawn on 1 September 2026 (Second accuracy pass; the US Krazati label (effective 31 August 2026) and application history read through openFDA, the FDA page of withdrawn cancer accelerated approvals read as raw HTML, and the abstract for PubMed id 42611092 read through NCBI efetch) · The approval row is marked withdrawn with both sources linked, the standard-of-care row and open problem are rewritten, and the pages say a 2026 review reports that KRYSTAL-10 did not improve progression-free or overall survival against chemotherapy
- 2026-11-17 · KRYSTAL-10: The note called it the pending confirmatory trial for the adagrasib and cetuximab approval, and the colorectal page listed it in the pipeline. The registry gives primary completion on 16 January 2026 with no results posted, a 2026 review reports that it did not improve progression-free or overall survival, and the approval it was to confirm has been withdrawn (Second accuracy pass; ClinicalTrials.gov record NCT04793958 read through the registry API, the abstract for PubMed id 42611092 read through NCBI efetch, and the openFDA Krazati label) · The note is rewritten with the registry position, the reported outcome and the withdrawal, and the trial is removed from the colorectal pipeline list
- 2026-11-17 · Botensilimab and balstilimab randomised phase 2: The 123-patient result (response rate 21%, median overall survival 21.2 months) was credited to this randomised phase 2 trial and the record was given a reporting year of 2026. The paper reports a different trial, the single-arm phase 1b study NCT03860272. The phase 2 has 234 participants, is closed to recruitment and has posted no results, and the phase 3 BATTMAN trial was terminated after 21 patients. The standard-of-care row also called the approach the most promising (Second accuracy pass; the full text of PMC13725327 (DOI 10.1158/1078-0432.CCR-26-1610) read through Europe PMC, and ClinicalTrials.gov records NCT05608044 and NCT07152821 read through the registry API) · The note and the standard-of-care row now credit the figures to the phase 1b trial, give the status of the phase 2 and the termination of the phase 3, and the reporting year is removed
- 2026-11-17 · EORTC 40983: The record said the earlier report showed better progression-free survival and that the benefit stands, and it flagged overall survival as the primary endpoint. In all randomised patients the three-year progression-free survival gain (35.4% against 28.1%, hazard ratio 0.79) was not statistically significant (p=0.058); it reached significance only in eligible and resected patients. Progression-free survival was the primary endpoint (Second accuracy pass; the abstracts for DOI 10.1016/S0140-6736(08)60455-9 and DOI 10.1016/S1470-2045(13)70447-9 read through Europe PMC) · The one-line summary, summary, result and the liver standard-of-care row give both results with their p values, a progression-free survival outcome row carries the primary flag, and the 2008 paper is linked
- 2026-11-17 · Colorectal cancer: anti-EGFR rechallenge: Two standard-of-care rows said rechallenge rested on single-arm trials only and was unproven. VELO randomised 62 patients in 2023 (median progression-free survival 4.0 against 2.5 months, hazard ratio 0.48) and CAVE-2 GOIM randomised 156 in 2026 (avelumab added nothing to cetuximab, overall survival hazard ratio 1.00) (Second accuracy pass; the abstracts for PubMed ids 37200022 and 41443411 read through NCBI efetch, and those for CHRONOS and CAVE read through Europe PMC) · Both randomised phase 2 results are given, and the rows say no phase 3 trial has compared rechallenge with standard later-line treatment
- 2026-11-17 · Colorectal cancer: stage III, mismatch-repair deficient: The row gave ATOMIC three-year disease-free survival as 86 against 77 percent and called it the first adjuvant immunotherapy success in colorectal cancer. The paper reports 86.3% against 76.2% (hazard ratio 0.50) and makes no such claim (Second accuracy pass; the abstract for DOI 10.1056/NEJMoa2507874 read through Europe PMC) · Figures corrected to 86.3 against 76.2 percent with the hazard ratio and the number randomised; the priority claim is removed
- 2026-11-17 · DYNAMIC-III: The registry link opened an unrelated trial of atropine for clozapine-induced drooling. The record linked the trial to the Signatera assay, but the registry says the ctDNA analysis was done at Johns Hopkins University. The prognostic comparison was flagged as the primary endpoint, and the primary results were not given (Second accuracy pass; the ANZCTR record ACTRN12617001566325 read as raw HTML, and the abstract for DOI 10.1038/s41591-025-04030-w read through Europe PMC) · The link now opens the right registration, the Signatera link is removed, and the two primary endpoints are given (85.3% against 88.1% at three years in ctDNA-negative patients, 51% against 61% at two years in ctDNA-positive patients)
- 2026-11-17 · MOUNTAINEER-03: The record and a note on the MOUNTAINEER page said the trial was recruiting. The registry lists it as active and no longer recruiting (Second accuracy pass; ClinicalTrials.gov record NCT05253651 (updated 16 November 2026) read through the registry API) · Status changed to active and both sentences reworded
- 2026-11-17 · ALTAIR: The one-line summary said the drug delayed relapse by about four months. The difference in median disease-free survival (9.30 against 5.55 months, hazard ratio 0.79, p=0.107) was not statistically significant (Second accuracy pass; the abstract for DOI 10.1038/s41591-026-04428-0 read through Europe PMC) · The one-line summary now says the gap was not statistically significant
- 2026-11-17 · PROPHYLOCHIP-PRODIGE 15: The one-line summary said one patient in four had a serious complication. The paper reports grade 3 to 4 complications in 29 of 71 patients who had second-look surgery, 41% (Second accuracy pass; the abstract for DOI 10.1016/S1470-2045(20)30322-3 read through Europe PMC) · The one-line summary now says four in ten
- 2026-11-17 · New EPOC: Overall survival was flagged as the primary endpoint. The paper says progression-free survival was the primary endpoint and overall survival a secondary one. The same fault was found on PRIME (primary endpoint progression-free survival), OPUS (best overall response) and CLOCC (overall survival at 30 months) (Second accuracy pass; the abstracts for DOI 10.1016/S1470-2045(19)30798-3 and PubMed id 28376151 read through Europe PMC and NCBI efetch, and the primary outcomes of NCT00364013, NCT00125034 and NCT00043004 read through the ClinicalTrials.gov API) · The primary flag is moved to the right endpoint on New EPOC and PRIME and removed on OPUS and CLOCC, and each summary names the primary endpoint
- 2026-11-17 · CLOCC: The record and the liver standard-of-care row called CLOCC the only randomised evidence that ablation lengthens life. The authors call it the first randomised study to show it. AVF2107g carried two unsupported "first" claims and OPUS one (Second accuracy pass; the abstracts for PubMed id 28376151, DOI 10.1056/NEJMoa032691 and DOI 10.1093/annonc/mdq632 read through NCBI efetch and Europe PMC) · "Only" is replaced by the authors' own wording, and the unsupported "first" claims on AVF2107g and OPUS are removed
- 2026-11-17 · Colorectal cancer: open problems: An open problem said about one in ten circulating tumour DNA-negative stage II patients still recurs, and another said the KRAS G12C combinations were approved on response rate alone. In DYNAMIC 15 of 246 ctDNA-negative patients (6%) recurred or died, and sotorasib with panitumumab was approved on the randomised phase 3 CodeBreaK 300 trial (Second accuracy pass; the full text of PMC9701133 read through NCBI efetch and the abstract for DOI 10.1056/NEJMoa2308795 read through Europe PMC) · Both statements corrected
- 2026-11-17 · CAO/ARO/AIO-94: The replication note gave local recurrence of 7.1% against 10.1% "at eleven years". These are ten-year cumulative incidences reported at a median follow-up of 11 years (Second accuracy pass; the abstract for DOI 10.1200/JCO.2011.40.1836 read through Europe PMC) · Reworded to ten-year cumulative local relapse at a median follow-up of eleven years
- 2026-11-17 · Encorafenib: The European approval row dated the colorectal use with cetuximab to 2018. The date of 19 September 2018 is the medicine's first authorisation, and the EMA page does not date the colorectal extension; it now also lists first-line use with cetuximab and FOLFOX (Second accuracy pass; the EMA Braftovi medicine page read as raw HTML) · The note says what the date refers to and lists both colorectal uses
- 2026-11-17 · Colorectal cancer: ctDNA-guided treatment trials: The circulating tumour DNA table called DYNAMIC "the only randomised evidence" that a ctDNA result can change treatment, said escalation was "the question still open" and that randomised escalation trials "will answer it". DYNAMIC-III and ALTAIR have both reported. In DYNAMIC-III de-escalation in ctDNA-negative stage III colon cancer missed non-inferiority (three-year recurrence-free survival 85.3% against 88.1%) and escalation in ctDNA-positive patients did not help (two-year recurrence-free survival 51% against 61%). In ALTAIR trifluridine-tipiracil against placebo gave median disease-free survival 9.30 against 5.55 months, hazard ratio 0.79 (0.60 to 1.05), and the primary endpoint was not met. The table also described the DYNAMIC margin as applying "with two years of follow-up" when two years is the endpoint and median follow-up was 37 months (Second accuracy pass; the abstracts for PubMed ids 41115959 (DYNAMIC-III, Nat Med 2025), 42260101 (ALTAIR, Nat Med 2026) and 40055522 (DYNAMIC five-year results) read through PubMed efetch, and the DYNAMIC abstract for DOI 10.1056/NEJMoa2200075 read through Europe PMC) · The three rows now give the DYNAMIC five-year figures, both DYNAMIC-III results and the ALTAIR result, each with its source, and the margin sentence names the two-year endpoint
- 2026-11-17 · Colorectal cancer: botensilimab plus balstilimab: The immune table called the 17% response rate "the highest response rate reported with immunotherapy alone" in microsatellite-stable disease, which the paper does not say, and said "the randomised confirmation has not reported" without the registry position. The randomised phase 2 (NCT05608044) is closed to recruitment with no results posted, and the phase 3 BATTMAN trial (NCT07152821) was terminated after 21 patients because the pharmaceutical sponsor stopped funding it (Second accuracy pass; the abstract for DOI 10.1038/s41591-024-03083-7 read through Europe PMC, and ClinicalTrials.gov records NCT05608044 and NCT07152821 read through the registry API) · The superlative is removed and the sentence now gives the status of both randomised trials with registry links
- 2026-11-17 · Colorectal cancer: where the left-right line falls: Two rows said a left-right split "at the splenic flexure still maximised the prognostic separation" (Loree 2018) and that mutation rates varied within each side "as much as" between sides. The paper says the division that maximised prognostic differences fell before the transverse colon, and only that substantial variation remained within each side. The KRAS row also gave the paper's 70% to 43% gradient as a KRAS rate when the paper reports it for RAS mutations (Second accuracy pass; the abstract for DOI 10.1158/1078-0432.CCR-17-2484 read through PubMed efetch and the full text read through NCBI PMC (PMC5844818)) · Both rows now say the best split was before the transverse colon, the "as much as" wording is replaced, and the gradient is labelled RAS
- 2026-11-17 · Colorectal cancer: BRAF V600E prognosis and citations: The BRAF V600E row gave the overall survival hazard ratio of 1.91 as a figure for microsatellite-stable disease. In the pooled CAIRO, CAIRO2, COIN and FOCUS analysis it is BRAF-mutant against BRAF wild-type across all patients. The row also counted 266 of 770 MSI-high samples as carrying "any BRAF mutation" when 266 is V600E alone and any BRAF mutation is 323, said BEACON CRC made V600E "the first colorectal genotype with its own targeted regimen", which the paper does not claim, and credited the BREAKWATER paper to Kopetz when its first author is Elez (Second accuracy pass; the abstract for DOI 10.1158/1078-0432.CCR-14-0332 read through PubMed efetch, the abstracts for DOIs 10.1056/NEJMoa1908075 and 10.1056/NEJMoa2501912 read through Europe PMC, and the study crc_msk_2026 recounted through the cBioPortal public API) · The hazard ratio is given for the pooled population with the mismatch repair proficient finding beside it, the counts are 323 for any BRAF mutation and 266 against 308 for V600E, the "first" claim is removed, and the paper is cited as Elez 2025
- 2026-11-17 · Colorectal cancer: Lynch syndrome testing figures: The Lynch syndrome route said universal tumour testing found every carrier "at the cost of testing about a third more tumours". The JAMA paper says a selective strategy (testing those diagnosed at 70 or younger and older patients meeting the Bethesda guidelines) needed 34.8% fewer tumour tests than universal testing and missed 4.9% of carriers. The sensitivity figures of 100%, 87.8% and 85.4% were also given without saying they come from the population-based cohorts of 3,671 probands, not all 10,206 (Second accuracy pass; the abstract for DOI 10.1001/jama.2012.13088 read through PubMed efetch) · Both rows now name the population-based cohorts, and the route gives the selective strategy with its 4.9% and 34.8% figures
- 2026-11-17 · Colorectal cancer: tumour mutational burden, RET and KRAS G12D approvals: The landscape table said the tumour-agnostic tumour mutational burden approval "has no colorectal-specific evidence behind it", that the colorectal evidence for selpercatinib is "a handful of patients", and that KRAS G12D has "no approved inhibitor". The current Keytruda label defines the indication at 13 or more mutations per megabase and lists 10 colorectal cancers with 1 response. The Retevmo label lists colorectal cancer as the commonest tumour in its 75-patient table, 22 patients with 10 responses. A RAS inhibitor, daraxonrasib, now has a US label, for pancreatic adenocarcinoma only (Second accuracy pass; the openFDA drug labels for Keytruda (effective 16 October 2026), Retevmo (11 September 2026) and Rasonque (27 August 2026) read as raw JSON, and an openFDA search for labels naming G12D, which returned none) · Each sentence now gives the label's own figures with a link, and the G12D sentence is narrowed to colorectal cancer
- 2026-11-17 · Colorectal cancer: genomic landscape counts in the 2026 MSK cohort: Fourteen comparisons set MSI-high samples against "6,447 microsatellite-stable samples" in the study crc_msk_2026. Only 6,114 are called stable; the 6,447 also holds 281 indeterminate and 52 unreported samples. Deep deletion counts no longer matched the study: SMAD4 205 (now 226), SMAD2 74 (79), SMAD3 49 (63), APC 40 (48) and PTEN 70 (95). TRIM24-BRAF and MKRN1-BRAF were given as 9 and 4 samples, which are record counts for 7 and 3 samples. The sidedness note said its 3,483 samples were those "with a named subsite" although 320 rectosigmoid samples are left out. A parallel line called 196 TCGA samples microsatellite-stable when they include the MSI-low group (Second accuracy pass; every count in the landscape and sidedness tables recounted from the cBioPortal public API on eight studies (118 plain counts and about 60 derived ones matched)) · The comparison group is relabelled non-MSI-high and defined once, the deletion and fusion counts are corrected, and the sidedness note says which subsites are counted
- 2026-11-17 · Colorectal cancer: smaller source faults in the molecular tables: Pancreatic cancer was said to run "a similar G12D share" to colorectal cancer, where a TCGA pancreatic cohort gives 41.2% of KRAS records against about 28%. The 1990 Fearon and Vogelstein model was dated to "the following year" after the 1988 paper. MLH1 silencing was said to be reversed by 5-azacytidine when the paper used 5-aza-2'-deoxycytidine in cell lines. The chromosomal instability rate was given per cell division, not per chromosome per generation. The consensus subtypes were said to rest on 4,151 samples, which the paper gives as patients. Three statements went beyond their sources: that HER2 amplification was "first proposed" as a cause of cetuximab resistance by Bertotti 2011, that a 2017 guideline says the G13D claim "has never been confirmed prospectively and no label allows it", and that the Le 2017 paper "led to the first tissue-agnostic approval", which the FDA granted on 23 May 2017 before that paper appeared. NICHE-2 treatment was described as "four weeks", which its abstract does not state (Second accuracy pass; all 64 DOIs in the source table resolved through Europe PMC and their abstracts read, with full texts for the TCGA, Guinney, Kotani and Loree papers and the open copy of the biomarker guideline (PMC7263311), the FDA news release of 23 May 2017 read from its Internet Archive copy, and cBioPortal study paad_tcga_pan_can_atlas_2018 recounted) · Each statement now matches its source, with the pancreatic count and the FDA release added as sources
- 2026-11-17 · KRAS G12C metastatic colorectal cancer: specific features of a new emerging target population: A finding credited the paper with KRAS G12C being "about 3% of all colorectal cancers". The paper states that the mutation occurs in about 4% of colorectal cancers; the 3% is this site's own count from cBioPortal (Second accuracy pass; the abstract for PubMed id 32605718 read through PubMed) · The finding now gives the paper's figure of about 4%
- 2026-11-17 · Non-V600 BRAF mutations define a clinically distinct molecular subtype of metastatic colorectal cancer: A caveat said the cohort came from "three commercial sequencing databases". The paper pooled three reference laboratories, and the authors' affiliations are Mayo Clinic, MD Anderson and Foundation Medicine, so only one is a commercial laboratory (Second accuracy pass; the abstract for PubMed id 28486044 and the article front matter for PMC5549454 read through PubMed and PubMed Central) · The caveat now names the three laboratories
- 2026-11-17 · Clinical sequencing defines the genomic landscape of metastatic colorectal cancer: A caveat said ligand expression "was measured in a subset" of the 1,134 sequenced cancers. The paper analysed ligand expression in TCGA tumours, not in its own sequenced cohort (Second accuracy pass; the full text of PMC5765991 read through PubMed Central) · The caveat now says the ligand data came from TCGA tumours
- 2026-11-17 · Circulating tumor DNA-guided treatment with pertuzumab plus trastuzumab for HER2-amplified metastatic colorectal cancer: a phase 2 trial (TRIUMPH): The summary line called it "the first trial to let a blood test decide who joins", which is wider than the authors' own claim, and a caveat said concurrent RAS alterations in plasma predicted poor response. The trial enrolled RAS wild-type patients, and the post hoc analysis found alterations in receptor tyrosine kinase, RAS and PI3K pathway genes enriched in non-responders (Second accuracy pass; the full text of PMC8604726 read through Europe PMC) · "The first" is removed and the caveat now describes the post hoc finding as the paper does
- 2026-11-17 · HER2 overexpression and amplification as a potential therapeutic target in colorectal cancer: analysis of 3256 patients enrolled in the QUASAR, FOCUS and PICCOLO colorectal cancer trials: The summary line said HER2 overexpression rose to 5% "in the tumours with no RAS or BRAF mutation". The study tested KRAS and BRAF only, and 5.2% is the stage IV figure; at stage II-III it was 2.1% (Second accuracy pass; the abstract for PubMed id 26690310 and the full text of PMC4785607 read through PubMed and PubMed Central) · The line now says stage IV tumours with no KRAS or BRAF mutation
- 2026-11-17 · Assessment of a HER2 scoring system for colorectal cancer: results from a validation study: The record called these criteria the definition "the trials have used ever since" and "the threshold behind the colorectal HER2 approvals". The paper supports only that the criteria selected patients for the HERACLES trial (Second accuracy pass; the abstract for PubMed id 26449765 read through PubMed) · Both sentences now say the criteria selected patients for HERACLES
- 2026-11-17 · Somatic POLE proofreading domain mutation, immune response, and prognosis in colorectal cancer: The summary said the cancers came from three trials "and eight cohorts". The paper lists seven cohorts: Leiden 1 and 2, Oslo 1 and 2, Bern, AMC-AJCC-II and Epicolon-1 (Second accuracy pass; the abstract for PubMed id 28404093 read through PubMed) · The summary now says seven cohorts
- 2026-11-17 · The landscape of somatic mutation in normal colorectal epithelial cells: A caveat said the 42 individuals were "mostly from surgical specimens". The parts of the paper that could be read name transplant organ donors and an autopsy cohort among the tissue sources and do not support that description (Second accuracy pass; the abstract for PubMed id 31645730 and the open parts of the article page at nature.com read as raw text; the main text is not open) · The caveat now gives only the number of individuals
- 2026-11-17 · Colorectal key papers: priority and uniqueness claims: Four summary lines made claims their papers do not make: the cetuximab K-ras analysis as "the first randomised proof", CALGB/SWOG 80405 as "the one randomised trial" in which the consensus subtypes were measured, Giannakis 2016 as finding "four" new genes when the paper lists those four as examples, and responders to cetuximab stopping "within a year" (Misale 2012) (Second accuracy pass; the abstracts for PubMed ids 18946061, 31042420, 27149842 and 22722830 read through PubMed) · Each line is reworded to what its abstract states
- 2026-11-17 · Terminology, molecular features, epidemiology, and management of serrated colorectal neoplasia: A finding said WNT disruption in the serrated pathway "arrives through methylation and RNF43 rather than APC". The abstract says only that the pathway has RAS and RAF mutations, disrupted WNT signalling and widespread CpG island methylation, and the full text is not open (Second accuracy pass; the abstract for PubMed id 31323292 read through PubMed) · The finding now states what the abstract states
- 2026-11-17 · Adagrasib: The colorectal notes on adagrasib and on the KRAS G12C result presented adagrasib with cetuximab as a current option. The US accelerated approval for colorectal cancer was removed from the label in August 2026, and a 2026 review reports that the phase 3 trial KRYSTAL-10 did not improve progression-free or overall survival over chemotherapy. (Second accuracy pass; the Krazati label effective 31 August 2026 read through openFDA, the abstract for PubMed id 42611092 read through PubMed, and registry record NCT04793958 read through ClinicalTrials.gov.) · Both notes now state the removal, the reported trial result and that the registry held no posted results on 17 November 2026.
- 2026-11-17 · Colorectal cancer: The residual disease note called DYNAMIC the only randomised evidence in any solid tumour that a ctDNA result can safely change treatment and said escalation evidence was missing. IMvigor011 in bladder cancer and DYNAMIC-III in stage III colon cancer have both reported; in DYNAMIC-III escalated chemotherapy did not improve two-year recurrence-free survival (51% against 61%) and de-escalation missed its non-inferiority margin (85.3% against 88.1%). The timeline also called DYNAMIC the first such trial and KRAS the first negative predictive biomarker in solid tumours, which neither paper claims. (Second accuracy pass; abstracts for PubMed ids 41115959, 41124204, 35657320 and 18946061 read through PubMed.) · Note 6 of 7 and the ctDNA result note rewritten with the DYNAMIC-III figures and its DOI; the two priority claims removed from the timeline.
- 2026-11-17 · Colorectal cancer: The open problems said KRAS G12D has no approved inhibitor, that one analysis is the only randomised test of the consensus subtypes, that the 2026 botensilimab cohort was selected for the absence of liver metastases, and that the FDA diagnostics list carries only the old codon 12 and 13 definition for cetuximab. A RAS inhibitor is now on a US label for pancreatic cancer only; a 2026 pooled analysis of five trials also tested the subtypes; the cohort had no active liver metastases; and a May 2026 entry on the FDA list reads extended KRAS and NRAS for both antibodies. (Second accuracy pass; the Rasonque label read through openFDA, abstracts for PubMed ids 41252656 and 42663536 read through PubMed, and the FDA list of companion diagnostic devices fetched on 17 November 2026.) · Four open problem lines reworded; the same corrections made to the KRAS G12D result note and the consensus subtype term note.
- 2026-11-17 · Colorectal cancer: Mutation shares in MSI-high tumours were set against a group labelled microsatellite-stable, but the figures (RNF43 3.7%, ARID1A 6.3%, B2M 1.5%, FBXW7 12.6%) are for all 6,447 samples not called MSI-high. SMAD4 and TP53 loss were called rare in MSI-high disease on figures of 11.8% and 30.5%. Immunoscore's hazard ratio of 0.20 was credited to 2,681 patients when it comes from the 700-patient training set. Neoadjuvant nivolumab plus ipilimumab was described as four weeks of treatment, which neither the paper's abstract nor the registry says. (Second accuracy pass; sample-level recount of crc_msk_2026 from the cBioPortal public API, the abstracts for PubMed ids 29754777 and 38838311 read through PubMed, and registry record NCT03026140.) · Comparison group relabelled non-MSI-high in five target notes; rare changed to less common with the figures; the Immunoscore denominator clarified; four weeks changed to a short course in four places.
- 2026-11-17 · Sidedness: The note said its right against left comparison was measured in 7,237 samples with a named subsite. Only 3,483 of the 7,237 have one (908 right, 2,575 left). (Second accuracy pass; sample-level recount of crc_msk_2026 from the cBioPortal public API.) · Denominator corrected to the 3,483 of 7,237 samples.
- 2026-11-17 · ctDNA MRD positivity (molecular residual disease after curative treatment): The note said 16 of 20 patients who were ctDNA-positive after adjuvant chemotherapy relapsed. The 16 of 20 are patients positive after surgery who then received chemotherapy. (Second accuracy pass; the abstract for PubMed id 34625408 read through PubMed.) · Sentence corrected and attributed.
- 2026-11-17 · dMMR (mismatch repair deficiency by IHC): The note said universal testing cost about a third more tests than the Bethesda guidelines, and credited the guideline with MSI testing where a stain is equivocal. The paper's 34.8% is the saving in tumour tests from a selective strategy that missed 4.9% of carriers, and the guideline accepts immunohistochemistry or DNA-based MSI testing without that condition. The Lynch syndrome term note also applied the 100% sensitivity to all 312 carriers when it was measured in the population-based cohorts. (Second accuracy pass; the abstract for PubMed id 23073952 read through PubMed and the guideline's open co-publication PMC7263311 read through Europe PMC.) · Both sentences rewritten from the sources; the term note and note 7 of 7 aligned.
- 2026-11-17 · BRAF: The colorectal note attached the odds ratio of 203 to BRAF and MLH1 methylation and called BRAF V600E the one genotype with its own regimen. The odds ratio is for BRAF mutation and the CIMP phenotype, and KRAS G12C and HER2-positive disease also have labelled regimens. The mismatch repair note said sporadic cases almost always carry BRAF V600E; the guideline says about three-fourths. (Second accuracy pass; the abstract for PubMed id 16804544 read through PubMed, the guideline co-publication PMC7263311, and the Lumakras, Tukysa and Braftovi labels read through openFDA.) · BRAF, MLH1 and mismatch repair target notes reworded.
- 2026-11-17 · HER2 ISH amplified (ERBB2 gene amplification): Four notes said the HERACLES criteria select about 5% of RAS wild-type patients. The 5% (48 of 914) is of KRAS exon 2 wild-type patients. The HER2 target note also said amplification was first proposed as a cause of cetuximab resistance in the 2011 xenograft paper, which claims no priority. (Second accuracy pass; abstracts for PubMed ids 26449765, 27108243, 26690310 and 22586653 read through PubMed.) · KRAS wild-type substituted in the result, trastuzumab and tucatinib notes and note 7 of 7; the priority claim removed.
- 2026-11-17 · NRAS: The note said an exon 2-only assay misses nearly all NRAS mutations, against its own count of 137 records at codons 59 and 61 and 123 at codons 12 and 13. (Second accuracy pass; mutation records recounted from the cBioPortal public API (291 missense records, 168 outside codons 12 and 13).) · Now reads more than half, with the count.
- 2026-11-17 · Analysis of plasma cell-free DNA by ultradeep sequencing in patients with stages I to III colorectal cancer: A caveat gave 130 patients and 16 relapses. The full text reports 125 analysable patients and 24 recurrences; 16 is the number among patients with serial samples after treatment. (Second accuracy pass; the full text PMC6512280 read through Europe PMC.) · Caveat corrected to 125 patients and 24 relapses.
- 2026-11-17 · International validation of the consensus Immunoscore for the classification of colon cancer: This record and the 2016 Immunoscore paper were linked to Institut Curie. The authors' affiliations are INSERM, the Centre de Recherche des Cordeliers and Universite Pierre et Marie Curie, a different institution. (Second accuracy pass; author affiliations for PubMed ids 29754777 and 26982367 read through PubMed.) · Institution link removed from both paper records.
- 2026-11-17 · MSI-high (microsatellite instability by PCR or sequencing): The note called MSI-high status the entry criterion for every colorectal immunotherapy trial. Trials in microsatellite-stable disease exist and are described on the same page. The balstilimab note also gave response rates as if of 148 patients when they are of 101 evaluable. (Second accuracy pass; abstracts for PubMed ids 31003911 and 38871975 read through PubMed and the Keytruda and Opdivo labels read through openFDA.) · Now reads the approved checkpoint inhibitor indications; denominator added to the balstilimab note.
- 2026-11-17 · Colorectal cancer in the UK: Northern Ireland screening age: The page said Northern Ireland sends the bowel screening kit from age 60 and "still starts at 60". The programme now invites people aged 59 to 74, an age extension was launched on 2 November 2026, and everyone aged 50 to 74 is planned to be invited by 2031 (Second accuracy pass; the nidirect and Public Health Agency bowel screening pages and Bowel Cancer UK's screening page read as raw HTML on 17 November 2026) · The introduction, the screening route and the screening step of the timeline now give 59 to 74 with the extension and its 2031 goal, and the two source labels carry the new wording and date
- 2026-11-17 · Colorectal cancer in the UK: the 28-day standard: The Faster Diagnosis Standard target was given as 75 percent, with 80 percent as a 2025/26 planning figure. NHS England's planning framework for 2026/27 to 2028/29 sets 80 percent as the new threshold to be maintained (Second accuracy pass; the Medium Term Planning Framework long-read read through NHS England's own REST API, and the National Cancer Plan read as raw HTML) · The target now reads 80 percent from 2026/27, formerly 75 percent, with the National Cancer Plan commitment for March 2029, and the framework is cited
- 2026-11-17 · Colorectal cancer in the UK: stage at diagnosis in Scotland: The introduction said more than half of all cases in Scotland are found at stage 1 or 2. Cancer Research UK gives 49 percent of cases with a known stage in 2023; the 63.4 percent figure applies only to cancers found through screening. The UK incidence figure also carried no years (Second accuracy pass; Cancer Research UK's bowel cancer statistics and incidence pages read as raw HTML, and the Public Health Scotland screening report read as PDF text) · The sentence now gives both figures with what each covers, and the incidence figure carries its years, 2019 and 2021 to 2022
- 2026-11-17 · STAR-TREC: The record gave 344 people randomised at 34 centres with recruitment closing on 31 December 2023, called the trial phase 3, called its report the first randomised comparison of the two radiotherapy schedules, and gave no result. The paper reports a phase 2/3 trial that enrolled 503 people at 37 sites between 14 June 2017 and 8 April 2024, analysed 409, and makes no such first claim (Second accuracy pass; the abstract for pmid 42633754 read through PubMed and the ISRCTN14240288 record read through the registry's API) · Enrolment, sites, dates and phase corrected with the registry's figures attributed; the 12-month result added (78.5 percent against 60.6 percent free of total mesorectal excision, hazard ratio 1.90, 1.29 to 2.81) and marked as early, before the 30-month main measure; the first claim removed here and on David Sebag-Montefiore's record
- 2026-11-17 · QUASAR: The record linked Rachel Kerr and Oxford to QUASAR. Its authors are Gray, Barnwell, McConkey, Hills, Williams and D J Kerr of the Birmingham Clinical Trials Unit; Rachel Kerr led the later QUASAR 2 (Second accuracy pass; the abstract and author list for pmid 18083404 read through PubMed and the ISRCTN82375386 record read through the registry's API) · The link between Rachel Kerr and QUASAR removed in both directions and the institution changed to Birmingham
- 2026-11-17 · MRC COIN: The summary said the companion COIN-B trial supported treatment breaks, which its report does not say, and the enrolment note misdescribed which arms the 1,630 patients were in. Tim Maughan's summary line said RAS wild-type where the trial's primary analysis was in KRAS wild-type tumours (Second accuracy pass; the abstracts for pmid 21641636, pmid 21641867 and pmid 24703531 read through PubMed and Europe PMC) · The COIN-B claim removed and the intermittent arm's own result stated (non-inferiority not shown); the note now names the arms, 815 each; the line on Tim Maughan's record says KRAS
- 2026-11-17 · UK Flexible Sigmoidoscopy Screening Trial: The summary line said one test can prevent a bowel cancer for life; the 17-year report says protection lasts at least 17 years. Recruitment was dated from July 1995, where the paper gives 14 November 1994 to 30 March 1999, and the end of bowel scope screening was credited to an NHS England statement that the cited page does not contain (Second accuracy pass; the abstracts for pmid 20430429 and pmid 28236467 read through PubMed, the ISRCTN28352761 record, and the update log of the GOV.UK programme overview) · The line now says at least 17 years, the dates follow the paper with the registry date attributed, and the sentence cites the GOV.UK update of 14 January 2021
- 2026-11-17 · Colorectal cancer in the UK: source labels: The NG-350A rectal cancer trial was labelled as recruiting to 30 November 2026; the registry now gives recruitment ended 19 October 2026 with 33 people enrolled. Two other trials whose registry end dates have passed were labelled as recruiting. The FOCUS4-N paper was dated 10 November 2021 and a FOxTROT paper title said RAS where it says KRAS. TA242 was described as replaced by TA1136 where NICE says partially replaced (Second accuracy pass; ISRCTN records read through the registry's API, PubMed records for pmid 34516759 and pmid 39805350, and the NICE overview pages for TA242 and TA1136 read as raw HTML) · Each label corrected to the source's wording and dates
- 2026-11-17 · Colorectal cancer in the UK: The funding table and note described the Cancer Drugs Fund from list version 1.365 of June 2025 as the latest: nivolumab with ipilimumab first line as the one colorectal medicine inside the fund, nine medicines routinely funded, and fruquintinib absent from the list (Second accuracy pass; the current NHS England Cancer Drugs Fund list, version 1.410 dated 12 November 2026, found through the england.nhs.uk document API and read as text) · Rewritten from version 1.410: no colorectal-specific medicine is in the fund section, nivolumab with ipilimumab first line has been routinely funded since 27 August 2025, fruquintinib (form FRU1) since 21 October 2025, bevacizumab has forms BEV11 and BEV12, and fourteen colorectal forms are routinely funded. The best value framework sentence is now dated to the version that carried it
- 2026-11-17 · Colorectal cancer in the UK: Northern Ireland's bowel screening was given as ages 60 to 74, unchanged since 2010, and the gaps list said the Public Health Agency page still carried stale text about ages 60 to 71 and the older test (Second accuracy pass; the nidirect and Public Health Agency bowel screening pages read as raw text on 17 November 2026) · Now 59 to 74, with the planned extension to everyone aged 50 to 74 by 2031; the gaps entry says what the page now shows
- 2026-11-17 · Colorectal cancer in the UK: The page said there was no 2026 National Bowel Cancer Audit report and that the 2025 report was the latest (Second accuracy pass; the NATCAN reports page and the State of the Nation Report 2026 PDF, published October 2026, read as text) · The gaps entry and the four-nations table now name the 2026 report and its headline 2024 figures (37,250 diagnosed, 90-day mortality 2.6 percent, 77 percent minimally invasive) and say the page's audit figures are still those of the 2025 report
- 2026-11-17 · Colorectal cancer in the UK: Three trials were shown as recruiting after their registry recruitment dates had passed: the NG-350A rectal trial (closed 19 October 2026 with 33 enrolled), OrigAMI-3 and PACES. STAR-TREC's closing date ignored the published enrolment window, and COLO-PREVENT was called phase 2/3 (Second accuracy pass; each ISRCTN record read through the registry API, and the STAR-TREC report (PubMed 42633754) read as abstract) · Statuses corrected to the registry dates; STAR-TREC now gives both the registry date and the published enrolment to 8 April 2024 (503 people at 37 sites); COLO-PREVENT is phase 3 on the registry
- 2026-11-17 · Colorectal cancer in the UK: The STAR-TREC 12-month report was called the first randomised comparison of the two radiotherapy schedules for organ preservation, which the paper does not say (Second accuracy pass; the abstract for PubMed 42633754 read through PubMed) · Replaced with the reported result: 78.5 percent alive without total mesorectal excision at 12 months after chemoradiotherapy against 60.6 percent after short-course radiotherapy, among people who chose organ preservation
- 2026-11-17 · Colorectal cancer in the UK: NICE NG151 was said to send locally advanced rectal cancer and peritoneal disease to specialist centres; thirteen appraisals were said to recommend a medicine; regorafenib was said to have been refused once; BRAF V600E was put at 8 to 10 percent with a further 2 to 4 percent non-V600, which no cited source states; a survey figure was credited to a page that does not carry it; an audit count was given as 28 of 128 units (Second accuracy pass; the NG151 recommendations, the NICE colorectal products list and terminated appraisal TA334, the GeNotes colorectal page, the Bowel Cancer UK campaign page and the audit report read as raw text) · NG151 1.3.12 and 1.5.21 now read consider and discuss referral; fourteen appraisals; TA334 described as terminated for want of a submission; BRAF given as GeNotes states it, around 10 to 15 percent; the survey figure removed; the audit count given as the report states it, 28 units (22 percent)
- 2026-11-17 · When to seek urgent help with bowel cancer (NHS 111 and 999): The summary line said to call 999 for sudden severe tummy pain with green vomit and to ring the hospital line for a stoma that has stopped working. The NHS colostomy page puts severe tummy pain, green vomit and no stoma output under an urgent GP appointment or NHS 111. The adult sepsis list was also quoted with lips or tongue, which the NHS gives only for children (Second accuracy pass; the NHS pages on colostomy complications, bowel cancer symptoms and sepsis read as raw text) · The 999 line now carries only sepsis and heavy bleeding, the stoma signs moved to 111 or an urgent GP appointment, and the sepsis wording matches the adult list
- 2026-11-17 · Colorectal cancer: the temperature that means ring the hospital: Five passages said Macmillan tells people on chemotherapy to ring for a temperature outside the range their team gave them. Macmillan's pages give the numbers: above 37.5 C or below 36 C (Second accuracy pass; Macmillan's FOLFOX, capecitabine, irinotecan, oxaliplatin and sepsis pages read as raw text) · The two numbers are now given in the oncologist question, the neutropenia note and the chemotherapy and urgent help terms
- 2026-11-17 · Colorectal cancer: DPD testing and the immunotherapy trials: The chemotherapy after surgery row said NICE NG151 puts DPD testing in writing; the guideline does not mention DPD. The immunotherapy row credited all four NICE recommendations to KEYNOTE-177 and CheckMate 8HW; the recommendation for nivolumab with ipilimumab after chemotherapy rests on CheckMate 142 (Second accuracy pass; the NG151 recommendations chapter and the committee discussions of TA709, TA1065, TA716 and TA914 read on nice.org.uk) · DPD testing is now credited to Macmillan with a note that NG151 is silent on it, and the trial sentence is limited to the first-treatment recommendations
- 2026-11-17 · Work and money with bowel cancer (UK): The record said the employer must consider reasonable adjustments, gave working from home as one of Bowel Cancer UK's examples and listed Carer's Allowance among the benefits the page names. The page says the employer must make reasonable adjustments, does not give working from home as an example and does not name Carer's Allowance (Second accuracy pass; Bowel Cancer UK's work, money and travel page read as raw text) · Wording changed to must make, the example removed, and the benefits listed as the page lists them, in the term and in the colorectal note on financial toxicity
- 2026-11-17 · Lynch syndrome testing after bowel cancer, and what it means for the family: The order of tests left out a step and universal testing had no source. NICE says to offer testing to everyone with colorectal cancer at diagnosis and, where MLH1 is lost, to test for BRAF V600E first and then MLH1 promoter methylation (Second accuracy pass; the recommendations of NICE HTG430, formerly DG27, read on nice.org.uk) · The BRAF step added in the term and the decision row, and the NICE guidance linked
- 2026-11-17 · Colorectal cancer: late effects, nerve damage and fatigue notes: Notes on nerve damage, hernia and fatigue quoted Bowel Cancer UK for things its pages do not say today: nerve symptoms for up to two weeks after each cycle, no heavy lifting for three months, and anaemia as a cause of fatigue. The late effects page was rewritten in October 2026; the fatigue page never mentioned anaemia. A ranking of treatments credited to the ESMO fatigue guideline could not be checked (Second accuracy pass; the two Bowel Cancer UK pages read as raw text, the Internet Archive copy of the late effects page from 1 May 2026, and Crossref and PubMed for the guideline, whose publisher refused the full text) · Notes rewritten from the pages as they stand, the hernia advice kept with its archived source, and the unchecked ranking removed with the gap stated
- 2026-11-17 · Carers: what you can do and UK carer support (bowel cancer): The record credited Bowel Cancer UK with a peer support line and with advice its carer pages do not give, described a Colostomy UK page as covering bag changes and skin care when it offers a booklet and volunteers, and said the NHS carer's assessment is for people aged 18 or over looking after another adult, where the NHS says anyone over 18 can ask for one (Second accuracy pass; the Bowel Cancer UK, Colostomy UK and NHS carer pages read as raw text) · Reworded to what each page says, in the term and in the three questions for carers
- 2026-11-17 · Colorectal cancer: smaller wording corrections on the questions and notes: Bowel Cancer UK's question list was said to start with stage and grade (it starts with where the cancer is); Cancer Research UK's prehabilitation page was said to cover smoking (it does not); prehabilitation was said to be shown by evidence to reduce anxiety (the page says it can); NICE was said to require referral for exenterative surgery (it says consider); and the title of the 2012 paper on the low anterior resection syndrome score was wrong (Second accuracy pass; each cited page read as raw text, NG151 recommendation 1.3.12, and Europe PMC for DOI 10.1097/SLA.0b013e31824f1c21) · Each statement reworded to its source and the paper's title corrected
- 2026-11-17 · CheckMate 8HW: nivolumab plus ipilimumab in MSI-high metastatic colorectal cancer: A caveat said overall survival was not mature. The final results appeared in October 2026 and show longer survival with the combination. The summary line called the result the largest effect size of any first-line trial in colorectal cancer, and the comparison with nivolumab alone the first phase 3 to prove the point, neither with a source (Second accuracy pass; the abstracts for PubMed 39602630, 39874977 and 42926106 read through PubMed, and the US Opdivo label read through openFDA) · The caveat now gives the final overall survival results: hazard ratio 0.37 against chemotherapy in the first line, 0.61 against nivolumab alone across all lines, and a first-line comparison with nivolumab alone (0.66) that did not meet its threshold. Both superlatives removed and two links added
- 2026-11-17 · Botensilimab plus balstilimab in microsatellite stable metastatic colorectal cancer: A caveat said a randomised phase 3 had not yet reported. The phase 3 trial was terminated in July 2026 after 21 patients when the sponsor withdrew funding. The summary line called this the first immunotherapy signal in the 95 percent of bowel cancers that ignore it, which the paper does not say, and the record was typed as a randomised controlled trial although it is a single-arm phase 1 (Second accuracy pass; the full text (PMC11405281) read through Europe PMC and the registry records NCT07152821 and NCT05608044 read through the ClinicalTrials.gov API) · Caveat rewritten with both registry records; summary line reworded and now says none of the 24 patients with active liver metastases responded against 17 of 77 without; study type changed
- 2026-11-17 · Nonoperative management of mismatch repair-deficient tumors: The record named AZUR-1 as the multicentre trial to come and said one antibody now replaces chemotherapy, radiotherapy and surgery. AZUR-1 gave first results in October 2026 with a lower complete response rate than this single-arm study, which was also typed as a randomised controlled trial (Second accuracy pass; the abstracts for PubMed 40293177 and 42927858 read through PubMed and the registry record NCT05723562) · Caveat now gives the AZUR-1 result (68 percent of 144 patients met the 12-month clinical complete response endpoint), the interpretation says the drug can replace the other treatments for most patients, and the study type is corrected
- 2026-11-17 · Circulating tumor DNA analysis detects minimal residual disease in stage II colon cancer: The interpretation said the trials that followed turned the measurement into a treatment decision, without the two trials in which acting on a positive result did not help (Second accuracy pass; the abstracts for PubMed 35657320, 41115959 (DYNAMIC-III) and 42260101 (ALTAIR) read through PubMed) · Rewritten: a negative result can spare chemotherapy in stage II disease, while escalated chemotherapy in DYNAMIC-III (two-year recurrence-free survival 51 against 61 percent) and trifluridine-tipiracil in ALTAIR (hazard ratio 0.79, 0.60 to 1.05) failed to improve outcomes. Links added
- 2026-11-17 · Colorectal cancer: evidence timeline and open problems: The forward-looking timeline entry treated the ctDNA strategy trials and the NordICC mortality answer as still to come, although DYNAMIC-III, ALTAIR and the 13-year NordICC report have been published. An open problem said organ preservation had never been randomised against surgery, although STAR-TREC randomised 120 patients against total mesorectal excision in its first phase. The HER2 entry said HERACLES enrolled RAS wild-type patients and cited the wrong paper. The polypectomy entry gave the result as at 23 years, and the regorafenib entry called it the first drug approved for its setting (Second accuracy pass; the abstracts for PubMed 41115959, 42260101, 42102826, 42633754, 27108243, 22356322 and 23177514 read through PubMed and Europe PMC, and three registry records read through the ClinicalTrials.gov API) · Timeline entry now reports the two ctDNA trials and the non-significant NordICC mortality result (risk ratio 0.88, 0.68 to 1.08). The open problem states what STAR-TREC randomised and that function was not the primary endpoint of OPRA or STAR-TREC. HERACLES is described as KRAS exon 2 wild-type and linked. The polypectomy figure is given at a median 15.8 years (12 deaths against 25.4 expected). Regorafenib is the first drug to lengthen survival in that setting. Years added to the early-onset incidence figures
- 2026-11-17 · PRODIGE 23: neoadjuvant FOLFIRINOX in locally advanced rectal cancer: The record stopped at three-year disease-free survival and noted that overall survival was not the result. The seven-year report found an overall survival gain (Second accuracy pass; the abstract for PubMed 38986769 read through PubMed) · Caveat now gives seven-year overall survival of 81.9 against 76.1 percent (p=0.033), with a link
- 2026-11-17 · OPRA: organ preservation after total neoadjuvant therapy: A caveat said regrowth continues after three years and durability is not settled. The five-year update found 99 percent of regrowths within three years (Second accuracy pass; the abstract for PubMed 37883738 read through PubMed) · Caveat rewritten with the five-year figures: survival free of total mesorectal excision 39 and 54 percent, disease-free survival 71 and 69 percent
- 2026-11-17 · Operative versus nonoperative treatment for stage 0 distal rectal cancer: The summary line gave ten-year survival of 98 percent for the patients who were watched, and a finding gave 97.7 percent as across the series. The abstract does not say which patients the ten-year figure covers (Second accuracy pass; the abstract for DOI 10.1097/01.sla.0000141194.27992.32 read through Europe PMC; the full text is not available through PubMed Central) · Summary line now uses the figure the abstract ties to the watched group, five-year survival of 100 percent in 71 patients. The ten-year finding is labelled as not re-read in full
- 2026-11-17 · DESTINY-CRC01: trastuzumab deruxtecan in HER2-expressing colorectal cancer: A caveat said the lower HER2 cohorts showed little activity, where the final report found no responses, and gave a motive for the DESTINY-CRC02 dose that its report does not state. The record was typed as a randomised controlled trial although it is single arm, as was HERACLES (Second accuracy pass; the abstracts for PubMed 33961795, 37286557 and 39116902 read through PubMed) · Caveats corrected to the final report (no responses in cohorts B and C, lung inflammation in 8 of 86) and to what DESTINY-CRC02 compared; study type changed on both records
- 2026-11-17 · Levamisole and fluorouracil for adjuvant therapy of resected colon carcinoma: The summary line said this 1990 trial first showed that chemotherapy after surgery prevents recurrence. A 401-patient trial reported that in 1989 and describes this one as its confirmation (Second accuracy pass; the abstracts for PubMed 2300087 and 2778478 read through PubMed) · Summary line now says confirmed; a caveat and a link name the 1989 trial
- 2026-11-17 · MOSAIC, IDEA, RAPIDO and BEACON CRC: later reports: Four records described later follow-up loosely or not at all. MOSAIC credited later analyses with a high-risk stage II benefit that the final report does not state. IDEA omitted its final overall survival result and was called the largest de-escalation exercise in adjuvant oncology with no source. RAPIDO and BEACON CRC mentioned later reports without their figures (Second accuracy pass; the abstracts for PubMed 19451431, 33271092, 36661037, 33503393 and 40444708 read through PubMed) · MOSAIC: six-year overall survival 78.5 against 76.0 percent, no difference in stage II. IDEA: five-year overall survival 82.4 against 82.8 percent, non-inferiority not confirmed. RAPIDO: locoregional recurrence 10 against 6 percent at five years. BEACON CRC: updated median 9.3 months for both regimens against 5.9, and the BREAKWATER interim survival of 30.3 against 15.1 months. Links added
- 2026-11-17 · SUNLIGHT and CALGB/SWOG 80405: wording: SUNLIGHT was called the largest gain in its setting with no source, and CALGB/SWOG 80405 was said to have enrolled over seven years when the abstract gives November 2005 to March 2012 (Second accuracy pass; the abstracts for PubMed 37133585 and 28632865 read through PubMed and Europe PMC) · SUNLIGHT now states the 3.3-month difference; the CALGB caveat gives the enrolment dates
- 2026-11-17 · Colorectal cancer roadmap: what the registry says is coming: Eight forward-looking statements were stale. AZUR-1 was shown as awaited after its first results appeared; COBRA was shown as completing in June 2026 although its posted results say it never went on to phase 3; MOUNTAINEER-03 was shown as recruiting; BREAKWATER's completion date was 28 December 2027; JANUS's primary completion was 30 September 2026; the botensilimab trial was called the first randomised test with no mention that the phase 3 trial BATTMAN was terminated; and NordICC's 15-year answer was dated to 2036 (Second accuracy pass; ClinicalTrials.gov records NCT05723562, NCT04068103, NCT05253651, NCT04607421, NCT05610163, NCT05608044, NCT07152821 and NCT00883792 read through the registry API on 17 November 2026, and the abstract for PubMed 42927858 read through PubMed) · The step and the watch list now give AZUR-1's first result (68 percent of 144 with a clinical complete response lasting 12 months), COBRA's posted outcome, MOUNTAINEER-03 as active and not recruiting, BREAKWATER at 29 July 2027, JANUS at 1 July 2030, the BATTMAN termination after 21 participants, and NordICC's registry terms without the inferred date
- 2026-11-17 · Colorectal cancer roadmap: screening colonoscopy: NordICC's 18 percent fall in bowel cancer cases was given three times with no mortality result, and the National Polyp Study was described as halving deaths two decades later after a 23-year follow-up (Second accuracy pass; the abstracts for PubMed 36214590, 42102826 and 22356322 read through PubMed) · Each passage now says that deaths from colorectal cancer did not differ significantly at ten years (risk ratio 0.90, 0.64 to 1.16) or at 13 years (0.88, 0.68 to 1.08), and the National Polyp Study is given its median follow-up of 15.8 years, with 23 years as the maximum
- 2026-11-17 · Colorectal cancer roadmap: blood tests after surgery: The roadmap said no trial had shown whether adding treatment after a positive ctDNA result helps and that the CIRCULATE trials were built to answer it. Two trials have since reported (Second accuracy pass; the abstracts for PubMed 41115959 (DYNAMIC-III) and 42260101 (ALTAIR) read through PubMed) · The step now gives both results: in DYNAMIC-III escalated chemotherapy did not improve two-year recurrence-free survival (51 against 61 percent) and reduced treatment missed its non-inferiority margin (85.3 against 88.1 percent); in ALTAIR the hazard ratio was 0.79 (0.60 to 1.05), not significant
- 2026-11-17 · Colorectal cancer roadmap: trial figures: Several figures were credited to the wrong analysis or left unqualified. The IDEA figure of 83.1 against 83.3 percent was given as a CAPOX result when it is an exploratory analysis of both regimens, and the overall result (non-inferiority not confirmed) was missing. NICHE-2 was described as four weeks of nivolumab. ATOMIC was dated 2025 and PARADIGM 2022. Cercek's 82 of 103 patients and 92 percent were given as rectal results although they include non-rectal tumours. Heald's operation was said to cut recurrence from a quarter to a few percent, SUNLIGHT was called the largest gain in the refractory setting, CRYSTAL the first negative predictive biomarker, and United States incidence under 50 was said to rise about 3 percent a year, none of which the sources state (Second accuracy pass; the abstracts for PubMed 29590544, 38838311, 42906564, 41880612, 37071094, 40293177, 6751457, 37133585, 19339720 and 36856579 read through PubMed, and the NICHE arm description on ClinicalTrials.gov NCT03026140) · Each sentence now follows its source: the IDEA result by regimen and risk group, two doses of nivolumab and one of ipilimumab, ATOMIC published 2026 with 86.3 against 76.2 percent, PARADIGM published 2023, the Cercek cohorts named, Heald's 50 operations with no pelvic recurrence, and a rise of 1 to 2 percent a year under 55. The unsourced superlatives are removed
- 2026-11-17 · Treat screening uptake, not test sensitivity, as the thing to optimise: NordICC was called the only randomised trial of screening colonoscopy and given without its mortality result, and the United States move to screening from 45 was credited to one incidence paper (Second accuracy pass; the abstracts for PubMed 40158525 (COLONPREV), 41721101 (SCREESCO), 42102826 and 34003218 read through PubMed) · The record now names COLONPREV and SCREESCO, gives NordICC's non-significant mortality result at 10 and 13 years, and describes the 2021 recommendation as resting on an evidence review and modelling
- 2026-11-17 · Take ctDNA-guided de-escalation beyond stage II: The record treated stage III de-escalation as untested and listed DYNAMIC-III as a readout still to come. DYNAMIC-III reported in 2025 and reduced treatment did not meet its non-inferiority margin. It also said 60 to 70 percent of stage III patients are cured by surgery alone, with no source (Second accuracy pass; the abstracts for PubMed 41115959 and 42260101 read through PubMed) · The summary, rationale and test now carry the DYNAMIC-III and ALTAIR results with their figures, and the unsourced cure figure is removed
- 2026-11-17 · Randomise organ preservation against surgery in mismatch repair-proficient rectal cancer: The record said tumour regrowth continues beyond three years and that nobody has ever randomised organ preservation against surgery. OPRA's five-year update found 99 percent of regrowths within three years, and STAR-TREC randomised 120 people among two organ-preserving schedules and immediate surgery in its first phase (Second accuracy pass; the abstracts for PubMed 37883738, 42633754 and 42927858 read through PubMed) · The regrowth sentence is corrected, STAR-TREC's 12-month result (78.5 against 60.6 percent free of total mesorectal excision) and AZUR-1's 68 percent are added, and the summary says the evidence in locally advanced disease rests on historical comparison
- 2026-11-17 · Select microsatellite stable patients for immunotherapy by a measured immune biomarker: The botensilimab signal was described with no word on what followed it (Second accuracy pass; ClinicalTrials.gov records NCT05608044 and NCT07152821 read through the registry API, and the full text of PubMed 38871975 read through Europe PMC) · The record now says the randomised phase 2 trial has closed with no results posted and the phase 3 trial BATTMAN was terminated after 21 participants, and gives the response without active liver metastases as 17 of 77
- 2026-11-17 · Find peritoneal spread while it is still small: The Verwaal trial was said to have used heated intraperitoneal oxaliplatin and to share PRODIGE 7's entry limits of age under 70 and a Peritoneal Cancer Index of 25 or less. Neither is in the Verwaal reports; both describe PRODIGE 7 (Second accuracy pass; the abstracts for PubMed 14551293, 18521686 and 33476595 read through PubMed) · The drug name is removed from the Verwaal sentence and the entry limits are credited to PRODIGE 7 alone
- 2026-11-17 · Colonoscopic polypectomy and long-term prevention of colorectal-cancer deaths (National Polyp Study): The one-line summary said the result was measured twenty-three years after polyp removal. Twenty-three years was the longest follow-up; the median was 15.8 years (Second accuracy pass; the abstract for DOI 10.1056/NEJMoa1100370 read through PubMed) · The summary now gives both the maximum and the median follow-up
- 2026-11-17 · Colorectal cancer incidence patterns in the United States, 1974-2013: The record said the United States Preventive Services Task Force accepted this paper's argument in 2021 and called the paper the direct evidence behind screening from 45. The task force's statement rests on a commissioned evidence review and modelling (Second accuracy pass; the abstract of the 2021 recommendation statement, PubMed 34003218, read through PubMed) · Both sentences now state the 2021 recommendation and its basis without crediting it to this paper
- 2026-11-17 · Increasing incidence of colorectal cancer in young adults in Europe over the last 25 years: The falls in mortality for ages 30 to 39 and 40 to 49 were given without their periods, which differ (Second accuracy pass; the abstract for DOI 10.1136/gutjnl-2018-317592 read through PubMed) · The finding now gives 1990 to 2016 for the younger band and 1990 to 2009 for the older
- 2026-11-17 · Once-only flexible sigmoidoscopy, Nottingham, stool DNA, colibactin and adenoma detection papers: Five statements went beyond their sources: that England's bowel scope programme was replaced by wider stool testing, that Nottingham showed no all-cause effect as in every trial of its size, that the stool DNA test was the first of its kind to be approved, that colibactin is the strongest lead yet, and that the adenoma detection gradient is shallower today (Second accuracy pass; the GOV.UK programme overview and National Cancer Plan as read by the UK checker, the openFDA premarket approval record P130017, and the abstracts for PubMed 8942775, 40267983 and 20463339) · Each is reworded to what the source states: bowel scope no longer offered from January 2021, no all-cause figure in the Nottingham abstract, approval on 11 August 2014, the authors' own suggestion about early-life exposure, and the limits of a 2010 report
- 2026-11-17 · ESMO guidelines for metastatic colorectal, localised colon and rectal cancer: The three guideline records describe recommendations that could not be checked, and two dated the ATOMIC trial to 2025 (Second accuracy pass; PubMed records 36307056, 32702383 and 28881920 carry no abstract and the journal pages answered a script with a refusal; the ATOMIC abstract, PubMed 41880612, read through PubMed) · Each record now carries a caveat saying the guideline text was not re-read on 17 November 2026; ATOMIC is dated to its 2026 publication and the stage III DYNAMIC-III result is added to the localised colon caveat
- 2026-11-17 · Preoperative FOLFOX Versus Postoperative Risk-adapted Chemotherapy in Patients With Locally Advanced Rectal Cancer: The record said this phase 3 rectal cancer trial was now recruiting, with 550 participants and a primary completion date of 2030-08. The registry entry, updated on 2026-10-02, lists it as terminated with 215 participants enrolled, an actual primary completion date of 2026-07-31 and the reason "poor accrual" (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT04495088 read as raw JSON and compared field by field by script, then read by hand) · Status set to withdrawn, the value the corpus uses for a terminated trial, and enrolment to 215; the one-line summary and the summary now say the trial stopped early and quote the registry's reason word for word
- 2026-11-17 · KN 046 Plus Regorafenib in MSS Metastatic Colorectal Cancer: The record said this phase 2 trial of KN046 with regorafenib was now recruiting, with 98 participants and a primary completion date of 2026-12-31. The registry entry, updated on 2026-11-03, lists it as terminated with 20 participants enrolled, an actual primary completion date of 2026-09-30 and the reason "The study was stopped early due to lack of efficacy." No results are posted on the registry (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT05985109 read as raw JSON and compared field by field by script, then read by hand) · Status set to withdrawn, the value the corpus uses for a terminated trial, and enrolment to 20; the one-line summary and the summary now say the trial stopped early and quote the registry's reason word for word. No outcome figures were added
- 2026-11-17 · Nimotuzumab Combined With Trifluridine/Tipiracil in the Treatment of Refractory Metastatic Colorectal Cancer: The record said this phase 3 trial of nimotuzumab with trifluridine and tipiracil was not yet recruiting, with 420 participants planned. The registry entry, updated on 2026-10-05, lists it as withdrawn with an enrolment of 0 and the reason "Study withdrawn before enrollment of any participant due to a corporate strategic decision. Not related to safety or efficacy concerns." (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT06343116 read as raw JSON and compared field by field by script, then read by hand) · Status set to withdrawn and enrolment to 0; the one-line summary and the summary now say the trial was withdrawn with no one enrolled and quote the registry's reason word for word; the planned dates are removed
- 2026-11-17 · Aspirin in Colorectal Cancer Liver Metastases: The record said the ASAC trial of aspirin after resection of colorectal liver metastases was active but no longer recruiting. The registry entry, updated on 2026-10-16, lists it as completed, with a completion date of 2026-06-01. No results are posted on the registry (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT03326791 read as raw JSON and compared field by field by script, then read by hand) · Status set to completed; the one-line summary and the summary now say the registry lists the trial as completed and give the completion date. No outcome figures were added
- 2026-11-17 · Effect of Perioperative Lidocaine or High-Dose Dexamethasone on Immune Response in Colon Cancer Surgery (PILDI Study): The record described NCT07432399 as a recruiting trial of lidocaine or high-dose dexamethasone, and its one-line summary called it a trial of dexamethasone. The registry no longer serves that registration: its API returns the NCT06923787 entry for both numbers and lists NCT07432399 as an alias. That entry lists lidocaine and placebo only, the single condition colon cancer and the primary outcome "Difference in opioid use in 24 hours after surgery between arms" (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT07432399 read as raw JSON and compared field by field by script, then read by hand) · The one-line summary no longer names dexamethasone and says the registry files this registration under NCT06923787; the summary says its details are those read on 2026-09-24 and could not be read again on 17 November 2026; the link to dexamethasone is removed. The two records still need merging, as the existing note says
- 2026-11-17 · XS-03 in Combination With FOLFOX or FOLFIRI and Bevacizumab for Treatment of Metastatic Colorectal Cancer Patients With RAS Mutation: The record named NovaOnco Therapeutics Co., Ltd. as the sponsor and said the trial was now recruiting. The registry entry, updated on 2026-09-25, gives Shanghai Fosun Pharmaceutical Industrial Development Co. Ltd. as the lead sponsor with no collaborator, and a status of enrolling by invitation. The primary outcome sentence also stopped in the middle of the third measure (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT06936527 read as raw JSON and compared field by field by script, then read by hand) · Sponsor, one-line summary and summary now name the registry's lead sponsor, and the summary says which company the record named before; the one-line summary says "now enrolling by invitation only"; the third primary outcome measure is given in full
- 2026-11-17 · Colorectal cancer: registry trials whose recruitment status changed: 16 records carried a recruitment status the registry has since changed: NCT05462613 (was recruiting, now active and not recruiting, entry updated 2026-11-04), NCT06680739 (was recruiting, now active and not recruiting, entry updated 2026-11-13), NCT05343013 (was recruiting, now active and not recruiting, entry updated 2026-10-01), NCT02919644 (was recruiting, now active and not recruiting, entry updated 2026-10-28), NCT07178717 (was planned, now recruiting, entry updated 2026-11-13), NCT07808151 (was planned, now recruiting, entry updated 2026-11-12), NCT07595276 (was planned, now recruiting, entry updated 2026-10-05), NCT07815834 (was planned, now recruiting, entry updated 2026-10-13), NCT07791758 (was planned, now recruiting, entry updated 2026-10-16), NCT07271355 (was planned, now recruiting, entry updated 2026-11-05), NCT07198321 (was planned, now recruiting, entry updated 2026-11-16), NCT06788171 (was planned, now recruiting, entry updated 2026-10-01), NCT07622017 (was planned, now recruiting, entry updated 2026-11-16), NCT07469306 (was planned, now recruiting, entry updated 2026-10-01), NCT07645118 (was planned, now recruiting, entry updated 2026-10-28), NCT07472868 (was planned, now recruiting, entry updated 2026-10-27) (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · Status set to active for the 4 that have closed to recruitment and to recruiting for the 12 that have opened; each one-line summary now says "active but no longer recruiting" or "now recruiting" to match
- 2026-11-17 · Colorectal cancer: registry trials with an unknown status: 13 records said the trial was now recruiting, not yet recruiting or active where the registry now gives the overall status as unknown, because the entry has not been verified for two years or more: NCT03635021 (last known status active and not recruiting, last verified 2024-10), NCT06192680 (last known status not yet recruiting, last verified 2024-01), NCT06594692 (last known status not yet recruiting, last verified 2024-09), NCT06373133 (last known status not yet recruiting, last verified 2024-04), NCT05168839 (last known status recruiting, last verified 2024-09), NCT05462236 (last known status recruiting, last verified 2024-07), NCT04516681 (last known status recruiting, last verified 2024-10), NCT06202001 (last known status recruiting, last verified 2024-04), NCT06640166 (last known status recruiting, last verified 2024-10), NCT06593548 (last known status not yet recruiting, last verified 2024-09), NCT05239650 (last known status not yet recruiting, last verified 2022-02), NCT05628038 (last known status recruiting, last verified 2024-09), NCT05963490 (last known status recruiting, last verified 2024-09) (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · Each one-line summary and summary now says what the trial was last listed as, that the registry gives its status as unknown, and when the entry was last verified; the status field is left as it was because the schema has no value for unknown
- 2026-11-17 · Colorectal cancer: registry trials enrolling by invitation: 6 records said "now recruiting" where the registry status is enrolling by invitation, which a reader cannot simply join: NCT06936527, NCT07542262, NCT07620574, NCT07441369, NCT07683949, NCT06959589 (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · Each one-line summary now says "now enrolling by invitation only"
- 2026-11-17 · Colorectal cancer: registry trial dates moved by their sponsors: 37 records carried a start or primary completion date the registry has since changed: NCT07178717 (start 2025-10 now 2025-11-03); NCT07808151 (start 2026-09-30 now 2026-11-30); NCT07224022 (start 2026-10-01 now 2027-03-01, primary completion 2028-12-31 now 2029-12-31); NCT07595276 (start 2026-06-01 now 2026-05-27); NCT07815834 (start 2026-09-01 now 2026-10-01); NCT07612007 (start 2026-09-01 now 2026-12-01, primary completion 2028-08-01 now 2028-11-01); NCT07691489 (start 2026-08 now 2026-10); NCT07662031 (start 2026-08-31 now 2026-12-31, primary completion 2030-08-31 now 2030-12-31); NCT07791758 (start 2026-09-04 now 2026-09-28); NCT07621497 (start 2026-06-01 now 2026-11-01, primary completion 2029-06-01 now 2029-11-01); NCT07271355 (start 2026-10-01 now 2027-01-04); NCT06258525 (start 2026-07 now 2027-01, primary completion 2028-08 now 2030-01); NCT07198321 (start 2025-09-22 now 2025-09-18, primary completion 2027-09-22 now 2027-07-30); NCT06265285 (primary completion 2026-12-31 now 2027-05-01); NCT07018869 (primary completion 2029-05-23 now 2031-10-28); NCT07595874 (start 2026-12-27 now 2026-07-01, primary completion 2027-05-27 now 2028-12-16); NCT06680739 (start 2022-03-01 now 2022-12-09); NCT06307548 (primary completion 2027-06-01 now 2029-06-01); NCT07750158 (start 2026-08-25 now 2026-12-18); NCT07140679 (primary completion 2029-09-30 now 2029-07-31); NCT05411718 (primary completion 2026-11-30 now 2029-11-30); NCT05610163 (primary completion 2026-09-30 now 2030-07-01); NCT07740720 (start 2026-09-17 now 2026-12-17, primary completion 2029-12-12 now 2030-03-13); NCT06788171 (start 2025-03-01 now 2026-05-09, primary completion 2026-12-01 now 2027-09-01); NCT07580339 (start 2026-10 now 2026-11, primary completion 2030-10 now 2030-11); NCT06057831 (start 2026-05 now 2027-05, primary completion 2029-05 now 2030-05); NCT07622017 (start 2026-07 now 2026-10-05); NCT07516210 (start 2026-05 now 2026-11); NCT07735624 (start 2026-09 now 2026-08-10, primary completion 2027-01-01 now 2027-06-01); NCT05254327 (primary completion 2027-05 now 2030-05); NCT06841159 (start 2024-03-01 now 2024-09-30, primary completion 2026-03-01 now 2029-09-30); NCT06958432 (start 2025-05-01 now 2025-05-06); NCT07469306 (start 2026-08-10 now 2026-08-15); NCT07645118 (start 2026-07 now 2026-10); NCT07472868 (start 2026-03-03 now 2026-03-16); NCT05663775 (primary completion 2026-08 now 2027-08); NCT07005570 (start 2025-05-01 now 2025-05-23, primary completion 2027-07-01 now 2031-12-31) (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · The dates in each summary now match the registry, with "started" kept only where the registry marks the start date as actual
- 2026-11-17 · Colorectal cancer: registry trials past their primary completion: 25 records said the trial was "due to reach its primary completion" on a date the registry records as an actual, past date: NCT04776655, NCT03869892, NCT02572141, NCT02758951, NCT03828227, NCT04068103 (results posted), NCT01460589, NCT03326791, NCT03829462 (results posted), NCT03874026, NCT06778382, NCT00722228, NCT06349590, NCT02505750, NCT04094688 (results posted), NCT06507371, NCT04928807, NCT04922853, NCT00569127 (results posted), NCT02843191, NCT06484153, NCT05624138, NCT05972655, NCT05555888, NCT07448077 (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and the date type compared by script) · Each summary now says the trial "reached its primary completion" on that date. Four of them have results posted on the registry; no outcome figures were added to these registry-only records
- 2026-11-17 · Colorectal cancer: registry trials that have not started: 238 records said the trial "started" on a date the registry gives only as an anticipated start, some of them in the future: NCT05220800, NCT02448173, NCT07649473, NCT06192680, NCT07585279, NCT07683403, NCT07467590, NCT07405736, NCT07337551, NCT05231850, NCT05794750, NCT07156682, NCT07515820, NCT07021950, NCT07823322, NCT07025850, NCT05275374, NCT06899477, NCT07025239, NCT06423937, NCT06959693, NCT07134205, NCT07260058, NCT07656766, NCT07150403, NCT07134101, NCT06513221, NCT06446557, NCT07808151, NCT07059338, NCT07549906, NCT07677579, NCT07714447, NCT07359456, NCT07004413, NCT07807800, NCT06391905, NCT07268677, NCT07224022, NCT06652412, NCT07381309, NCT07535632, NCT05699746, NCT06092242, NCT07284394, NCT06400160, NCT07391566, NCT07395063, NCT07815834, NCT07612007, NCT07515963, NCT07691489, NCT07243938, NCT05776381, NCT07409844, NCT06099951, NCT07662031, NCT07071844, NCT06845306, NCT07622550, NCT06255379, NCT07621497, NCT06835179, NCT06646445, NCT07271355, NCT07051785, NCT07551778, NCT06594692, NCT06790212, NCT06839092, NCT07095309, NCT06195670, NCT07007728, NCT07729605, NCT07435584, NCT06860685, NCT07570914, NCT07132008, NCT06665087, NCT07730489, NCT07124884, NCT06301386, NCT06944548, NCT07123441, NCT07229872, NCT07261709, NCT07229846, NCT07403877, NCT06258525, NCT07814417, NCT06681038, NCT07286695, NCT06746545, NCT07591597, NCT05427669, NCT05747716, NCT07509398, NCT07321847, NCT06218888, NCT06890624, NCT06047015, NCT06858969, NCT06373133, NCT07358585, NCT07576725, NCT07353268, NCT07502014, NCT07750041, NCT06616259, NCT05909423, NCT07510932, NCT07362836, NCT06999252, NCT07333651, NCT07009145, NCT06176885, NCT07003022, NCT07537998, NCT07468630, NCT07085169, NCT07277322, NCT04516681, NCT05870800, NCT07060443, NCT05914389, NCT06558773, NCT05742425, NCT07750158, NCT07398326, NCT06722183, NCT03125980, NCT07551596, NCT07441369, NCT07683949, NCT07357220, NCT07134218, NCT07700719, NCT07007767, NCT07504198, NCT07116577, NCT07740720, NCT06666166, NCT06872606, NCT07796763, NCT07302126, NCT07248020, NCT07774754, NCT03975049, NCT07755124, NCT07815288, NCT06838546, NCT07297030, NCT07580339, NCT04889742, NCT07402486, NCT06057831, NCT07817888, NCT07061210, NCT04443543, NCT06751394, NCT06596343, NCT07737600, NCT07483684, NCT05969847, NCT05841134, NCT06014944, NCT07691632, NCT06855849, NCT07503639, NCT07381400, NCT07490119, NCT07548736, NCT07819604, NCT07000305, NCT06097416, NCT07543848, NCT06832917, NCT06599827, NCT06973343, NCT06972498, NCT06593548, NCT07621159, NCT06919510, NCT06450574, NCT07161115, NCT06304545, NCT07394192, NCT07797244, NCT05364489, NCT07209384, NCT07509424, NCT07527520, NCT07175636, NCT07779694, NCT05359393, NCT05807984, NCT05484024, NCT07209215, NCT07070713, NCT05239650, NCT07384377, NCT06390982, NCT07397442, NCT07516210, NCT06443671, NCT06679673, NCT07810114, NCT07528209, NCT05794971, NCT06930118, NCT06254521, NCT06829355, NCT07113275, NCT07292298, NCT07594925, NCT07686796, NCT07314528, NCT07551479, NCT07645118, NCT07118800, NCT07430696, NCT07277842, NCT07596290, NCT07662278, NCT07383285, NCT07381777, NCT05998122, NCT07811076, NCT07074353, NCT07059949, NCT05479240, NCT07837362, NCT07521605, NCT06760520, NCT07581626, NCT07214142, NCT07651007, NCT07150949 (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and the date type compared by script) · Each summary now says "planned to start" with the registry date
- 2026-11-17 · Colorectal cancer: registry trial enrolment figures: 4 records carried an enrolment figure the registry has since changed: NCT06778382 (22 now 35, estimated), NCT06680739 (60 now 35, actual), NCT05343013 (25 now 30, actual), NCT07740720 (90 now 100, estimated) (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · The enrolment field and the summary now give the registry figure
- 2026-11-17 · Colorectal cancer: registry trials with an incomplete list of primary outcomes: 19 records gave only the first three of the primary outcome measures the registry lists: NCT04428905 (18 measures), NCT07807800 (11 measures), NCT07391566 (4 measures), NCT06860685 (4 measures), NCT07198321 (5 measures), NCT07693751 (4 measures), NCT06134388 (6 measures), NCT05868226 (6 measures), NCT06758830 (8 measures), NCT06307548 (4 measures), NCT03965234 (4 measures), NCT07503756 (6 measures), NCT07227636 (4 measures), NCT07737600 (5 measures), NCT07621159 (4 measures), NCT06885385 (4 measures), NCT05254327 (4 measures), NCT06881537 (5 measures), NCT07314528 (7 measures). In 4 more the sentence stopped in the middle of a phrase: NCT06936527 (3 measures), NCT06400160 (3 measures), NCT06907342 (3 measures), NCT07272772 (3 measures). NCT07005570 listed two measures where the registry now lists one, "Complete Response Rate" (Second accuracy pass; the primary outcome measures in the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared with the sentence by script) · Where the registry lists up to four measures all are now given in full; where it lists more, the first three are given followed by the number of further measures
- 2026-11-17 · Colorectal cancer: registry text changed in seven trial records: NCT06788171 and NCT07239466 quoted an eligibility line that is not in the registry entry today. NCT06958432 gave the control arm as conventional radiotherapy plus chemotherapy and PD-1 inhibitor; the registry now gives it without the PD-1 inhibitor. NCT07808151 listed three conditions where the registry now lists colon cancer alone. NCT07693751 lacked Czechia among its site countries. NCT07740720 and NCT07814417 had no site countries; the registry now lists Japan for the first and China, Spain and the United States for the second (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · Each record now carries the registry's current words, and the two quotations are replaced with the lines the registry gives
- 2026-11-17 · Adagrasib: The record said adagrasib is approved in lung and colorectal cancer. The colorectal use, with cetuximab, was an accelerated approval granted on 21 June 2024 and it was removed from the US label in August 2026. The dosing line still listed the cetuximab combination, and the regulatory events listed a full approval in lung cancer in June 2025 that the label and Drugs@FDA do not show (Second accuracy pass; the US Krazati label effective 31 August 2026 and the Drugs@FDA submission history read through openFDA, with the FDA approval notice and the FDA list of withdrawn cancer accelerated approvals read by curl) · The summary, plain-language line, approvals row, dosing line and events now say the colorectal indication was an accelerated approval that was removed in August 2026; the unsupported 2025 event was taken out; the eight translations were updated
- 2026-11-17 · KRAS G12C colorectal cancer: The page said sotorasib with panitumumab and adagrasib with cetuximab are both approved, counted two approved combinations, and described KRYSTAL-10 as a trial that will decide first-line use. Adagrasib's colorectal approval was removed in August 2026, KRYSTAL-10 was a second-line trial, and a review published in August 2026 reports that it did not improve progression-free or overall survival against chemotherapy (Second accuracy pass; the US Krazati label read through openFDA, the registry record NCT04793958 read through the ClinicalTrials.gov API, and the abstract for PubMed 42611092 read through PubMed) · The summary, subtype list, standard of care row, state of the art, history and open problems were corrected, and KRYSTAL-10 was taken out of the pipeline list
- 2026-11-17 · KRYSTAL-10: The trial was shown as active with no outcome. A review published in August 2026 reports that adagrasib with cetuximab did not improve progression-free or overall survival against chemotherapy, and the accelerated approval the trial was to confirm was withdrawn in 2026. The registry has no results posted (Second accuracy pass; the abstract for DOI 10.1007/s12029-026-01547-2 read through PubMed, the registry record read through the ClinicalTrials.gov API, and the FDA list of withdrawn cancer accelerated approvals read by curl) · The status is now negative and the summary and plain-language line say what the review reports and that the trial's own report was not found
- 2026-11-17 · KRAS G12C: The record quoted a colorectal indication for adagrasib with cetuximab from the Krazati label, which no longer carries it, and said both KRAS G12C drugs are given with an EGFR antibody in bowel cancer. The same claim stood in the report-field text, the pathway and technology summaries, the pairing record, the KRYSTAL-1 paper record, the guideline history and Rona Yaeger's record (Second accuracy pass; the US Krazati label effective 31 August 2026 read through openFDA) · The threshold is kept and marked withdrawn with a note, the quotation was removed, and each of the other records now gives the 21 June 2024 accelerated approval and its removal in August 2026
- 2026-11-17 · Colorectal cancer in the UK: National Bowel Cancer Audit figures: Ten audit figures were those of the 2025 report, for 2023. The 2026 report, for people diagnosed or operated on in 2024, was published in October 2026: 46.8 percent diagnosed at stage 1 or 2 among those staged, 7.6 percent under 50, 90-day mortality 2.6 percent against a tightened target of under 5 percent, 77 percent of resections minimally invasive, 83 percent of units doing 20 or more rectal resections, 67 percent adjuvant chemotherapy in stage 3 colon cancer, 27 percent neoadjuvant radiotherapy, 40 percent with an unclosed ileostomy at 18 months, two-year survival 85.3 percent (Second accuracy pass; the National Bowel Cancer Audit State of the Nation Report 2026 read as PDF text) · The ten figures now carry the 2026 report's values, periods and source; the narrative rows that quote 2023 figures stay dated to 2023
- 2026-11-17 · United Kingdom: The page said bowel screening starts at about 60 in Northern Ireland and that Northern Ireland invites people aged 60 to 74. It invites people aged 59 to 74 and plans to invite everyone aged 50 to 74 by 2031 (Second accuracy pass; the nidirect and Public Health Agency bowel screening pages read by curl) · The text, the detail and the source label now say 59 to 74
- 2026-11-17 · Getting urgent help with bowel cancer: The short summary sent people to 999 for sudden severe tummy pain with green vomit and to the hospital line for a stoma that has stopped working, and gave a temperature outside your team's range. The NHS puts those stoma signs under an urgent GP appointment or 111, and Macmillan gives above 37.5 C or below 36 C. Two colorectal red cards repeated the temperature wording (Second accuracy pass; the NHS colostomy complications, bowel cancer symptoms and sepsis pages and Macmillan's FOLFOX page read by curl) · The short summary and both red cards now follow the NHS and Macmillan wording
- 2026-11-17 · Work and money with bowel cancer in the UK: The short summary said an employer must consider reasonable adjustments and listed a phased return among them. Bowel Cancer UK says the employer must make reasonable adjustments and that you may want to ask for a phased return. The first-60-days checklist told people to ask Macmillan's welfare rights advisers, a phrase the sources read do not use (Second accuracy pass; Bowel Cancer UK's work, money and travel page (updated July 2025) read by curl) · The summary was reworded, and the checklist now says to call Macmillan on 0808 808 0000
- 2026-11-17 · Colorectal cancer trials open today: The list of open trials included two trials the registry now lists as terminated, one withdrawn and one completed (Second accuracy pass; the registry records read through the ClinicalTrials.gov API by the registry checker and the statuses now on the records) · Trials whose status is withdrawn or completed are left off the list, and the summary says so
- 2026-11-17 · Circulating tumour DNA after bowel cancer surgery: the randomised trials: The replication note said randomised escalation trials are ongoing, and a technology note said the randomised evidence is for de-escalation, not escalation. DYNAMIC-III (2025) found that de-escalation for ctDNA-negative stage III patients did not meet non-inferiority (85.3 against 88.1 percent) and that escalation for ctDNA-positive patients did not help (51 against 61 percent); ALTAIR (2026) did not meet its primary endpoint (hazard ratio 0.79, 0.60 to 1.05) (Second accuracy pass; the abstracts for DOI 10.1038/s41591-025-04030-w and DOI 10.1038/s41591-026-04428-0 read through Europe PMC) · Both notes now give the two results
- 2026-11-17 · CIRCULATE-Japan: Jeanne Tie, who leads DYNAMIC and DYNAMIC-III, was listed among the people of CIRCULATE-Japan. She is not an author of the design paper, the GALAXY reports or the ALTAIR report (Second accuracy pass; the author lists for PubMed 33931919, 36646802, 39284954, 39293512 and 42260101 read through PubMed) · The link was removed
- 2026-11-17 · Henriksen 2022: serial ctDNA in stage III colorectal cancer: The source label and a sentence called this cohort IMPROVE-IT. The paper names IMPROVE-IT as a separate, ongoing randomised trial; its own cohort is 168 patients treated at Danish and Spanish hospitals (Second accuracy pass; the abstract for PubMed 34625408 and the full text PMC9401484 read through PubMed and Europe PMC) · The name was removed from the label and the sentence. The record id keeps the old name because ids may not change
- 2026-11-17 · BRAF in colorectal cancer: The note said BRAF V600E carries the worst prognosis of any common colorectal genotype, hazard ratio 1.91. The source compares BRAF-mutant with BRAF wild-type tumours in a pooled analysis of four trials (1.91, 1.66 to 2.19) and ranks nothing. Notes on RET said the colorectal evidence is a handful of patients with no trial; the selpercatinib label reports 22 patients with colorectal cancer, 10 of whom responded (Second accuracy pass; the abstract for PubMed 25139339 read through PubMed and the US Retevmo label effective 11 September 2026 read through openFDA) · The BRAF notes state the comparison; the RET notes give the label figures
- 2026-11-17 · Colonoscopy: The summary said removing adenomas cuts colorectal cancer incidence and mortality by half or more, citing the National Polyp Study and NordICC. The National Polyp Study found 76 to 90 percent fewer cancers than expected; NordICC found 18 percent fewer cancers at 10 years on an invitation basis and no statistically significant fall in deaths. The 2026 NordICC paper record had no findings (Second accuracy pass; the abstracts for PubMed 8247072, 36214590 and 42102826 read through Europe PMC and PubMed) · The summary gives each trial's own result, and the 13-year findings were added to the paper record
- 2026-11-17 · Heald 1982: total mesorectal excision: Three places said none of Heald's first 50 curative operations recurred in the pelvis within two years. The paper reports 50 operations classed as curative or conceivably curative, out of more than 100 consecutive resections, followed for more than two years with no pelvic or staple-line recurrence (Second accuracy pass; the abstract for DOI 10.1002/bjs.1800691019 read through Europe PMC) · The plain summary, the rectal cancer page and the colorectal history row were reworded
- 2026-11-17 · Habr-Gama 2004: watch and wait: The roadmap and an idea summary gave ten-year overall survival of 97.7 percent for the 71 of 265 patients who were watched. The abstract ties that figure to the series and gives five-year overall survival of 100 percent for the 71 (Second accuracy pass; the abstract for PubMed 15383798 read through PubMed) · Both now give the five-year figure for the watched group
- 2026-11-17 · NICHE-2: Four places said four weeks or a month of immunotherapy before surgery, and one credited the three-year result to the 2024 paper. The schedule was ipilimumab on day 1 and nivolumab on days 1 and 15; the 2024 paper had a median follow-up of 26 months and the three-year result is in the 2026 report (Second accuracy pass; the abstracts for PubMed 38838311 and 42906564 read through PubMed and registry record NCT03026140) · The texts now say a short course, the three-year result is credited to 2026, and eight translations were updated
- 2026-11-17 · AZUR-1: A roadmap, a standard of care row and a plain-language line described AZUR-1 as the registrational trial with no result, or as having met its main goal. Its first results, online on 25 October 2026, are an interim analysis: 68 percent of 144 patients followed for at least 12 months had a clinical complete response sustained for 12 months on dostarlimab alone (Second accuracy pass; the abstract for PubMed 42927858 read through PubMed and the US Jemperli label effective 19 October 2026 read through openFDA) · The three texts now give the result, and eight translations were updated
- 2026-11-17 · Anti-EGFR antibody and chemotherapy in left-sided disease: The plain-language line said the combination gives the longest survival in first-line trials, with no source (Second accuracy pass; the abstract for PubMed 37071094 (PARADIGM) read through PubMed) · The line now gives the PARADIGM result, 37.9 against 34.3 months, as about 38 months and longer than with bevacizumab; eight translations were updated
- 2026-11-17 · Bevacizumab interactions: The interaction text called bleeding and gastrointestinal perforation boxed warnings. The current US label has no boxed warning; both are listed under warnings and precautions (Second accuracy pass; the US Avastin label effective 6 January 2025 read through openFDA) · Reworded
- 2026-11-17 · Tumour mutational burden cut-off: The assay record and the KEYNOTE-158 paper record gave the pembrolizumab cut-off as 10 mutations per megabase. The US label now defines TMB-high as 13 or more; the FDA companion diagnostic list still gives 10 for FoundationOne CDx (Second accuracy pass; the US Keytruda label effective 16 October 2026 read through openFDA and the FDA companion diagnostics list read by curl) · Both records now state the label's 13 alongside the list's 10
- 2026-11-17 · Colibactin: The plain summary called colibactin the strongest current lead on why bowel cancer is rising in the under-50s. The 2025 genome study says only that early-life exposure may contribute. Two paper records claimed KRAS was the first negative predictive biomarker, which no source read says; HERACLES was described as enrolling RAS wild-type patients when eligibility was KRAS exon 2 wild-type (Second accuracy pass; the abstracts for PubMed 40267983, 19339720 and 27108243 read through PubMed) · Each was reworded to what its source says
- 2026-11-17 · Pancreatic ductal adenocarcinoma: daraxonrasib in Europe: The summary said no European or UK decision on daraxonrasib had been made by 24 September 2026. The European Medicines Agency's human medicines committee adopted a positive opinion on 15 October 2026, for adults with metastatic pancreatic adenocarcinoma after at least one prior systemic therapy (Second accuracy pass; the European Medicines Agency medicines list and its page for Rasonque read as raw text on 17 November 2026, and the United States approval date and indication confirmed against the openFDA label and the abstract for pmid 42802048) · The summary now gives the committee's recommendation and its date, and says the European Commission's decision was still to come on 17 November 2026. The UK half of the sentence keeps its September date because the UK regulator's record was not read
- 2026-11-17 · Pancreatic ductal adenocarcinoma: zoldonrasib and the biomarker-directed drugs: The summary said zoldonrasib combinations produced response rates never before seen in this disease, with no source, and called olaparib and zenocutuzumab the only biomarker-directed drugs, also with no source (Second accuracy pass; the abstracts for pmid 42645357 and pmid 42663603 read through Europe PMC, which say early zoldonrasib combination studies still require confirmation) · The sentence now says early studies of zoldonrasib combinations have extended the progress to KRAS G12D disease, although confirmation is required. The word only is removed
- 2026-11-17 · Pancreatic ductal adenocarcinoma: UK figures and survival years: The share of cases found at stage I or II was given for Northern Ireland (30 percent) and Wales (19 percent) without years; they are yearly averages for 2015 to 2019 and 2017 to 2019. The outlook answer gave United States and UK survival without the diagnosis years (Second accuracy pass; Cancer Research UK's pancreatic cancer statistics and survival pages and the SEER Cancer Stat Facts page read as raw HTML) · Each figure now carries its years: 2016 to 2022 for the United States figure and England and Wales diagnoses up to 2018 for the UK one
- 2026-11-17 · Pancreatic ductal adenocarcinoma: what NICE does and does not say: Three statements went beyond the guidance cited. The symptoms list said NG12 1.2.5 calls for an urgent CT, where the recommendation is to consider one. The diagnosis row credited a baseline CA 19-9 and germline testing for every patient to NG85, which recommends neither. The relatives answer said NG85 offers yearly surveillance; NG85 gives no interval (Second accuracy pass; the recommendations of NICE NG85 and NG12 read as raw HTML from nice.org.uk) · The symptoms line says the finding is a reason to consider an urgent CT, the row says the two tests are not NG85 recommendations and come from the NCCN guideline, and the word yearly is removed from the NG85 sentence
- 2026-11-17 · Pancreatic ductal adenocarcinoma: early-onset risk and the WHO variants: The summary said the usual risk factors apply before age 60 with alcohol most strongly; the pooled study found alcohol at 26 g or more a day associated with early-onset disease (odds ratio 1.49), most strongly before 45, and did not rank it above the other factors. The list of ductal variants was credited to a 2020 overview of the WHO classification that does not list them (Second accuracy pass; the abstract for DOI 10.1097/mpa.0000000000000392 read through Europe PMC and the full text of DOI 10.1111/his.13975 (PMC7003895)) · The alcohol sentence now gives the study's own finding. The variants sentence says the WHO volume could not be re-read on 17 November 2026 and that the overview does not list them. The 2009 timeline note on the 1 mm margin rule now says what happened that year
- 2026-11-17 · Undifferentiated carcinoma of the pancreas with osteoclast-like giant cells: The page gave the frequency as 1.4 percent in the largest series, 38 of about 2,700 cases. The paper gives no such denominator: 10 of its 38 cases came from 2,346 systematically reviewed pancreatectomies at one centre, which is 0.4 percent of them and 1.4 percent of the 735 invasive pancreatic carcinomas. The Wikipedia link pointed to giant-cell tumour of bone, a different disease (Second accuracy pass; the full text of DOI 10.1097/pas.0000000000000689 (PMC4987218) read through NCBI, and the Wikipedia title checked through its API) · The burden line and the summary now give the paper's own counts, the unsourced word largest is removed, and the link points to the pancreatic cancer article
- 2026-11-17 · Adenosquamous carcinoma of the pancreas: The 30 percent squamous threshold was credited to the 2019 WHO classification through a 2020 overview that does not give it (Second accuracy pass; the full texts of DOI 10.1111/his.13975 (PMC7003895) and DOI 10.1016/j.jss.2011.06.015 (PMC3210865) read) · The threshold is now credited to the Armed Forces Institute of Pathology definition quoted in the SEER study, with a note that the WHO volume could not be re-read on 17 November 2026
- 2026-11-17 · Mucinous cystic neoplasm of the pancreas with associated invasive carcinoma: The page said the DIPLOMA trial showed the oncological outcome of minimally invasive distal pancreatectomy is not inferior. The trial's primary endpoint was complete (R0) resection, 73 against 69 percent; survival was a secondary outcome (Second accuracy pass; the abstract for DOI 10.1016/j.lanepe.2023.100673 read through Europe PMC) · The sentence now says the rate of complete (R0) resection is not inferior
- 2026-11-17 · Neoadjuvant therapy versus surgery first for resectable and borderline resectable pancreatic cancer: The summary line said that for clearly removable tumours "the largest trial found no gain". The trial it meant, NORPACT-1, was a phase 2 trial of 140 patients. The Japanese Prep-02/JSAP05 trial randomised 364 patients with resectable disease and found longer survival with neoadjuvant gemcitabine plus S-1 (median 37.0 against 26.6 months, hazard ratio 0.73, 95% CI 0.56 to 0.95), and the page did not mention it. The page also gave no figures for NORPACT-1, where survival by intention to treat ran the other way (25.1 against 38.5 months, hazard ratio 1.52) (Second accuracy pass; the abstracts for pmid 40235447 (Annals of Surgery, DOI 10.1097/sla.0000000000006730) and DOI 10.1016/s2468-1253(23)00405-3 read through Europe PMC) · The summary line now says the trials disagree and names both by size; the text gives both results with their hazard ratios, and the Prep-02/JSAP05 paper is linked
- 2026-11-17 · Vascular resection in pancreatic cancer surgery: The page said a meta-analysis found vein resection associated with slightly higher postoperative mortality, without saying the difference was not statistically significant (risk difference 0.01, 95% CI 0.00 to 0.03, P = 0.2) (Second accuracy pass; the abstract for DOI 10.1002/bjs.9969 read through PubMed) · The sentence now gives the 1-point difference and says it was not statistically significant
- 2026-11-17 · New-onset diabetes as a signal of pancreatic cancer: The page credited a 1.8 to 1.9 times risk to long-standing type 2 diabetes. Those are the pooled figures for diabetes of any duration (odds ratio 1.82, relative risk 1.94); for diabetes of five years or more the same meta-analysis gives 1.5 (Second accuracy pass; the abstracts for DOI 10.1038/sj.bjc.6602619 and DOI 10.1016/j.ejca.2011.03.003 read through PubMed) · The sentence now says diabetes of any duration and gives both pooled figures
- 2026-11-17 · Familial pancreatic cancer and inherited risk: The page said that in the UK germline testing is offered to every patient with pancreatic cancer. None of the page's sources says so: NICE NG85 asks only that every patient is asked whether a first-degree relative has had the disease. The share of surveillance-detected cancers at stage I in CAPS5 (77.8 percent) was given without saying it is seven of nine (Second accuracy pass; NICE NG85 recommendations read as raw HTML and the abstract for DOI 10.1200/jco.22.00298 read through Europe PMC. The NHS genomic test directory page returned no content) · The sentence now gives the NG85 recommendation (1.1.14) and says which patients the NHS tests could not be confirmed on 17 November 2026; the CAPS5 figure carries its denominator
- 2026-11-17 · Pancreatic intraepithelial neoplasia (PanIN): The page said 71 percent of high-grade lesions were accompanied by cystic change and lobular fibrosis. The autopsy study says 71 percent of pancreata with PanIN-3 showed cystic change, and separately that PanIN-3 came with higher-grade extralobular fibrosis (Second accuracy pass; the abstract for DOI 10.1097/mpa.0000000000000786 read through Europe PMC) · Reworded to the study's two findings
- 2026-11-17 · RASolute 302: The record gave no European status for daraxonrasib, and the pancreatic summary and second-line row said no European decision had been made by 24 September 2026. The European Medicines Agency's human medicines committee adopted a positive opinion on 15 October 2026. The summary also called the trial the biggest result in the disease's history with no source, and gave 59 sites where the registry lists 60 (Second accuracy pass; the European Medicines Agency page for Rasonque read as raw text, the openFDA Rasonque label and FDA approval notice read, the abstract for DOI 10.1056/NEJMoa2605555 read through PubMed, and ClinicalTrials.gov NCT06625320 read through its API) · The trial record, the pancreatic summary, the second-line rows and the open problem now give the committee's recommendation and its date and say the European Commission's decision had not been published on 17 November 2026. The superlative is replaced by the figures, with the confidence interval and progression-free survival added
- 2026-11-17 · Pancreatic ductal adenocarcinoma: neoadjuvant chemotherapy for resectable disease: The resectable row, the list of failed programmes and the open problem named NORPACT-1, PREOPANC-2, SWOG S1505 and NEONAX and left out Prep-02/JSAP05, a randomised trial of 364 patients in which neoadjuvant gemcitabine plus S-1 gave median survival of 37.0 months against 26.6 months with surgery first (hazard ratio 0.73, 95% CI 0.56 to 0.95) (Second accuracy pass; the abstract for DOI 10.1097/SLA.0000000000006730 (pmid 40235447) read through Europe PMC, and the NORPACT-1 and PREOPANC-2 abstracts read through PubMed) · The trial and its figures are added in each place with a link, and the text now says the negative results apply to FOLFIRINOX
- 2026-11-17 · PANOVA-3: The page said the result was presented at the ASCO 2025 plenary and called the device the first new approval in locally advanced disease in decades. The ASCO abstract is LBA4005, which is not a plenary abstract, and no source supports the claim of a first approval (Second accuracy pass; the abstract for DOI 10.1200/JCO-25-00746 read through PubMed, the ASCO abstract found through Crossref, and the openFDA device record P250034 read) · The page now says the annual meeting and the abstract number, gives the approval date of 11 February 2026, drops the unsourced claim, and says progression-free survival and response were not improved
- 2026-11-17 · MRTX1133: The page credited the xenograft results to Nature 2023 and said development had slowed. The paper is in Nature Medicine 2022, and the phase 1 trial was terminated in March 2025 after 63 participants, the registry giving formulation challenges as the reason (Second accuracy pass; the abstract for DOI 10.1038/s41591-022-02007-7 read through Europe PMC and ClinicalTrials.gov NCT05737706 read through its API) · The citation, the figure (regression in eight of 11 pancreatic models) and the termination are corrected, the paper is linked, the pancreatic timeline entry is reworded and the drug is removed from the pancreatic pipeline list
- 2026-11-17 · Zoldonrasib: The record gave the status as phase 2 and called zoldonrasib the first drug aimed specifically at KRAS G12D. Three phase 3 trials are recruiting, and an earlier KRAS G12D inhibitor had already entered the clinic in 2023 (Second accuracy pass; ClinicalTrials.gov NCT07621718, NCT07805954, NCT07777822 and NCT05737706 read through the API, and the response figures checked against the OncLive report read as raw HTML) · The status is phase 3, the word first is removed and the summary line says the 50 percent response is from 30 second-line patients in a company-reported cohort
- 2026-11-17 · High-risk pancreatic surveillance (CAPS / PRECEDE): The page said that in CAPS5 77 percent of screen-detected cancers were stage I against about 15 percent in the general population, with five-year survival of about 73 percent. In CAPS5 it was seven of nine cancers; the 73.3 percent survival is for all the CAPS studies together, and the comparison figure is not in the paper (Second accuracy pass; the abstract for DOI 10.1200/JCO.22.00298 read through Europe PMC) · The sentence now gives seven of nine for CAPS5, 57.9 percent of 19 across all CAPS studies, the survival figure with its cohort, and the paper's own comparison with cancers found outside surveillance. The paper is linked
- 2026-11-17 · POLO: The page said roughly 5 to 7 percent of patients have inherited BRCA mutations. The trial report says 4 to 7 percent (Second accuracy pass; the full text of DOI 10.1056/NEJMoa1903387 (PMC6810605) read through NCBI) · Both sentences now say 4 to 7 percent, and the summary adds that 7.5 percent of the 3,315 people screened for the trial carried one
- 2026-11-17 · NAPOLI 3 and NALIRIFOX: The result line gave an overall survival hazard ratio of 0.84 beside a progression-free figure from the Lancet paper, whose own survival hazard ratio is 0.83. The NALIRIFOX record dated the European first-line indication to 2025 (Second accuracy pass; the Lancet abstract and full text (PMC11664154), the openFDA Onivyde label and the European Medicines Agency page for Onivyde read as raw text) · The result now gives 0.83 (95% CI 0.70 to 0.99) from the paper and 0.84 from the US label, and the European entry says the first-line indication came through a 2024 variation
- 2026-11-17 · Pancreatic ductal adenocarcinoma: timeline: The timeline dated gemcitabine's approval to 1997 and gave daraxonrasib's phase 1/2 survival as about 14.5 months. Gemcitabine was approved in the United States on 15 May 1996, and the published phase 1/2 report gives 13.1 months in 26 second-line patients with RAS G12 mutations (Second accuracy pass; the openFDA approval record for NDA020509 and the abstract for DOI 10.1056/NEJMoa2505783 read through Europe PMC) · The gemcitabine entry is moved to 1996 with the 1997 publication noted, and the 2025 entry carries the published figure
- 2026-11-17 · Pancreatic cancer: the failed and stopped programmes and why: The erlotinib sentence gave RTOG 0848 as 29.9 against 28.1 months and LAP07 as 13.6 against 11.9 months without saying that the longer figure in each pair was the arm without erlotinib. An open problem also said stereotactic radiotherapy and electroporation were equivalent in CROSSFIRE, a 68-patient trial stopped for futility (Second accuracy pass; the abstracts for DOI 10.1097/COC.0000000000000633, DOI 10.1001/jama.2016.4324 and DOI 10.1016/S2468-1253(24)00017-7 read through PubMed) · Each pair now names its arms, and the open problem says survival did not differ in a small trial stopped early
- 2026-11-17 · Pancreatic cancer trials open today: The registry counts and the six UK hospitals named for RASolute 303 were those of 24 September 2026. The registry now lists 11 UK sites for RASolute 303, and one of the 284 listed trials has been withdrawn (Second accuracy pass; ClinicalTrials.gov read through its API for NCT07491445 and NCT07097064, and the 284 registry records cached by the registry checker on the same day) · The UK site list is updated, and a dated sentence gives the 17 November 2026 statuses beside the September snapshot
- 2026-11-17 · SCALOP-2: The record said the trial finished in 2021 and that its main results had not appeared in an indexed journal by September 2026, and the SCALOP record repeated this. The results were published in the European Journal of Cancer in July 2024 and were negative: high-dose chemoradiotherapy did not improve overall survival (16.9 against 15.6 months, p 0.68) and nelfinavir did not improve progression-free survival (10.0 against 11.1 months, p 0.98) (Second accuracy pass; a Europe PMC search for the trial name found PMID 39059185 (DOI 10.1016/j.ejca.2024.114236), and the abstract was read through PubMed) · Status is now negative with 2024 as the year reported; the tldr, summary and result give the published figures with their 60 percent confidence intervals, the paper is linked, the open-trials tag is removed, and the SCALOP record's closing sentence states the result
- 2026-11-17 · SCALOP: The nine-month progression-free survival figures (62.9 against 51.4 percent) were given as the result without saying that the difference in this primary endpoint was not statistically significant (Second accuracy pass; the abstract for DOI 10.1016/S1470-2045(13)70021-4 read through PubMed, where the authors call the difference non-significant and ask for caution) · The tldr, summary and result now say the primary endpoint difference was not significant
- 2026-11-17 · PRIMUS 001: The record gave 500 participants, read as if recruitment were continuing, and said the 30 hospitals ran from Aberdeen to Plymouth. The ISRCTN record gives recruitment ending on 31 December 2024 with a total final enrolment of 411 against a target of 460, and no Plymouth site is listed (Second accuracy pass; ISRCTN75002153 read through the ISRCTN API (record updated 10 July 2026) and ClinicalTrials.gov NCT04151277 read through the registry API) · Enrolment is now 411 with a note explaining the registry gap, the summary gives the ISRCTN dates, and the tldr says recruitment closed at the end of 2024 with no results published and names Southampton
- 2026-11-17 · STARPAC2: The setting was given as pancreatic cancer in general and the record was linked to metastatic disease; the trial is restricted to locally advanced, non-metastatic disease. The record also said recruitment had closed at a single site, where the ISRCTN record lists eight UK centres and a recruitment end date of 31 December 2026 (Second accuracy pass; the eligibility criteria of NCT04241276 read through the registry API, the protocol abstract (DOI 10.1186/s12885-024-13333-z) read through PubMed, and ISRCTN11503604 read through the ISRCTN API) · Setting, summary and tldr now say locally advanced disease, the metastatic link is removed, both registries' accounts of recruitment and sites are given, and the protocol paper is linked
- 2026-11-17 · SWOG S1505: The record said neither arm exceeded the 40 percent two-year survival threshold, although the estimates were 47 and 48 percent, and that a third of enrolled patients were not resectable on central review (Second accuracy pass; the abstract for DOI 10.1001/jamaoncol.2020.7328 read through Europe PMC, which says neither estimate was significantly higher than 40 percent and that 43 of 147 patients (29 percent) were ineligible) · The tldr, summary and result now say not significantly higher than the threshold, and the tldr gives 29 percent
- 2026-11-17 · APACT: The record said the trial recruited at six UK sites and that there were two treatment-related deaths per arm. The registry lists three UK hospitals, each entered twice, and the paper reports two deaths per arm from treatment-emergent adverse events (Second accuracy pass; the locations of NCT01964430 read through the registry API and the abstract for DOI 10.1200/JCO.22.01134 read through Europe PMC) · The summary and the UK note now give three hospitals in six registry entries, and the deaths are described in the paper's words
- 2026-11-17 · PANFIRE-2: The record said CROSSFIRE found electroporation equivalent to stereotactic radiotherapy, and gave major complications as 21 events across 29 of 50 participants. CROSSFIRE was a superiority trial stopped for futility with no significant difference, and the 29 participants had 14 minor and 21 major complications between them (Second accuracy pass; the abstracts for DOI 10.1016/S2468-1253(24)00017-7 and DOI 10.1148/radiol.2019191109 read through PubMed) · The summary now says no significant survival difference and early stopping for futility, and the result gives both complication counts
- 2026-11-17 · Pancreatic trials: first, only and largest claims: Six trial records carried claims no source supported: CanStem111P as the largest first-line pancreatic cancer trial ever run, CONKO-007 as the largest test of radiotherapy after chemotherapy, CROSSFIRE as the only randomised comparison of the two local treatments, NAPOLI-1 as the first licensed second-line treatment, CONKO-003 as the first randomised trial to show second-line benefit and as superseded by daraxonrasib, and PRODIGE 4 as the first change in treatment in fourteen years (Second accuracy pass; each trial's abstract read through PubMed. The CanStem111P authors claim only the largest cohort given nab-paclitaxel with gemcitabine in a trial) · Each sentence now states what the paper reports: the numbers randomised, the authors' own claim for CanStem111P, and daraxonrasib as the alternative where it is available
- 2026-11-17 · Daraxonrasib: The pancreatic note said no European authorisation was listed on 24 September 2026 and that the first-line trial RASolute 303 recruits at six UK hospitals. The EMA's medicines committee adopted a positive opinion on 15 October 2026, and the registry now lists 11 UK sites (Second accuracy pass; the EMA page for Rasonque read as raw HTML on 17 November 2026, and NCT07491445 (updated 16 November 2026) read through the registry API) · The note now gives the committee opinion, its indication and the absence of a Commission decision so far, links the EMA page, says NICE's search could not be read by script, and gives 11 UK sites
- 2026-11-17 · RASolute 303: The registry note gave primary completion as June 2028 and six UK sites; the RASolute 305 note said United States sites only. The registry now gives April 2028 and 11 UK sites for RASolute 303, and RASolute 305 lists sites in Austria, Italy, the Netherlands and three London hospitals (Second accuracy pass; NCT07491445 and NCT07621718 read through the ClinicalTrials.gov API on 17 November 2026) · Both notes are re-dated and carry the current completion date, countries and UK sites
- 2026-11-17 · Adagrasib: The pancreatic note said the Krazati label covers KRAS G12C lung and colorectal cancer. The label effective 31 August 2026 lists lung cancer only and records the accelerated colorectal indication as removed in August 2026 (Second accuracy pass; the Krazati label read through the openFDA label API, including its recent major changes section) · The note now says lung cancer only and that the colorectal indication was removed; the label dates quoted for sotorasib and dostarlimab are also brought up to the current labels
- 2026-11-17 · NALIRIFOX: The European approval row for first-line NALIRIFOX carried the year 2025. The extension of indication received its committee opinion on 21 March 2024 and its European Commission decision on 25 April 2024 (Second accuracy pass; the EMA procedural steps document for Onivyde pegylated liposomal (variation II/0034) read as text from the PDF) · The row now carries 2024 and both dates
- 2026-11-17 · Erlotinib: The pancreatic note said the drug was licensed on a trial whose survival gain was under two weeks, gave the US licence as metastatic disease, and quoted RTOG 0848 survival as 29.9 against 28.1 months in a sentence that read as favouring neither arm; the same order fault put the ESPAC4 subgroup figures the wrong way round in the PRODIGE 24 note (Second accuracy pass; the erlotinib label read through openFDA (median survival 6.5 against 6.0 months, hazard ratio 0.81, locally advanced, unresectable or metastatic disease), and the abstracts for DOI 10.1097/COC.0000000000000633 and DOI 10.1200/JCO.24.01118 read through PubMed) · The note gives the label's medians and indication, RTOG 0848 reads 28.1 months with erlotinib against 29.9 without, and the ESPAC4 subgroup reads 25.9 months with the combination against 20.7 with gemcitabine alone
- 2026-11-17 · MRTX1133 phase 1/2: The note called MRTX1133 the first KRAS G12D inhibitor in the clinic, with no source, and did not give the reason the trial stopped (Second accuracy pass; NCT05737706 read through the ClinicalTrials.gov API, which gives formulation challenges as the reason for termination) · The unsourced claim is removed and the registry's reason is given
- 2026-11-17 · Pancreatic cancer: The molecular table said mutant KRAS copy gains were recorded in 296 of 853 dosage-annotated samples in the MSK 2024 study. 853 is the number of annotated samples without a gain, not the number annotated. (Second accuracy pass; the study's clinical attributes recounted from the public cBioPortal API (KRAS_MUT_GAINED above 0 in 296 of 1,157 samples, KRAS_MUT_DOSAGE 2+ in 201 and 3+ in 95 of 1,149).) · Now 296 of 1,157, with the attribute values quoted as deposited.
- 2026-11-17 · Pancreatic cancer: The KRAS allele rows described their denominators as KRAS mutation records when four of the six were counts of KRAS-mutant samples (2,188 samples against 2,207 records in the MSK 2024 study). Percentages were within one point. (Second accuracy pass; mutation records and mutated samples recounted for all seven studies from the public cBioPortal API.) · The G12D and G12C rows and the file's method note now name each denominator correctly and give the record totals.
- 2026-11-17 · Pancreatic cancer: The microenvironment section and the mismatch repair pathway note called tumour-agnostic pembrolizumab the only approved immunotherapy route. Dostarlimab is also approved in the United States for mismatch repair deficient solid tumours, and pembrolizumab also for tumours with a high mutational burden. (Second accuracy pass; the current US labels for Jemperli and Keytruda read through openFDA (effective 19 and 16 October 2026).) · Reworded to name both drugs and the three tumour-agnostic indications; both labels linked.
- 2026-11-17 · Pancreatic cancer: The ENDPAC new-onset diabetes score was listed as awaiting independent validation, with only the developers' 2018 figures. A retrospective external validation in 70,050 English primary care patients, published in November 2026, found lower accuracy (sensitivity 62.6%, specificity 72.5%). (Second accuracy pass; PubMed abstract for DOI 10.3399/BJGP.2024.0253 and the ClinicalTrials.gov record NCT04662879 read on 17 November 2026.) · Status changed, the UK validation and the status of the prospective Early Detection Initiative added with their sources.
- 2026-11-17 · Pancreatic cancer: Four statements went beyond their cited sources: that KRAS G12R is almost unique to pancreatic cancer (credited to Bournet 2016), that no approval or guideline uses ctDNA (credited to Lee 2019), that Collisson 2011 was the first published claim that subtypes should stratify treatment, and that daraxonrasib halved the death hazard when the hazard ratio was 0.40. (Second accuracy pass; the abstracts for PubMed ids 27010960, 31250894, 21460848 and 42223072 read through Europe PMC and PubMed.) · The G12R phrase is removed; the other three are reworded to what each paper states.
- 2026-11-17 · Real-time targeted genome profile analysis of pancreatic ductal adenocarcinomas identifies genetic alterations that might be targeted with existing drugs or used as biomarkers: The record called the cohort the largest such series, which the paper does not claim, and said the tumour-agnostic pembrolizumab label defines a high mutational burden as 10 mutations per megabase. The current US label says 13 or more. (Second accuracy pass; the abstract for PubMed id 30836094 and the Keytruda label read through openFDA (effective 16 October 2026).) · The largest claim is removed and the caveat gives the label's current threshold.
- 2026-11-17 · Core signaling pathways in human pancreatic cancers revealed by global genomic analyses: The summary line called the study the first complete reading of the coding genes and said the same dozen pathways were hit in almost every case. The paper claims no priority and reports each pathway altered in 67 to 100% of tumours. (Second accuracy pass; the abstract for PubMed id 18772397 read through PubMed.) · Summary line reworded with the paper's range.
- 2026-11-17 · Pancreatic cancer genomes reveal aberrations in axon guidance pathway genes: The record said ATM's appearance in this paper is the origin of its place on germline testing panels. The paper reports ATM as a somatically mutated gene and makes no such claim. (Second accuracy pass; the abstract for PubMed id 23103869 read through Europe PMC.) · The sentence now says only what the paper found.
- 2026-11-17 · Whole-exome sequencing of pancreatic cancer defines genetic diversity and therapeutic targets: The record said the study first tied MYC amplification to the squamous, chemotherapy-resistant end of the disease. The paper ties it to poor outcome and the adenosquamous subtype and claims neither priority nor chemotherapy resistance. (Second accuracy pass; the abstract for PubMed id 25855536 read through Europe PMC.) · Reworded to the paper's finding.
- 2026-11-17 · Subtypes of pancreatic ductal adenocarcinoma and their differing responses to therapy: The summary line called these the first gene-expression subtypes of pancreatic cancer, a priority the paper does not claim. (Second accuracy pass; the abstract for PubMed id 21460848 read through Europe PMC.) · The word first is removed.
- 2026-11-17 · Prospective evaluation of germline alterations in patients with exocrine pancreatic neoplasms: The summary line described the 615 patients as consecutive. The paper describes them as unselected. (Second accuracy pass; the abstract for PubMed id 29506128 read through Europe PMC.) · Consecutive changed to unselected.
- 2026-11-17 · PRECEDE consortium recommendations (Gonda 2021): A caveat said outcomes from the consortium cohort were not yet published. The consortium published baseline imaging findings in 2024 and five-year imaging findings in November 2026 (Second accuracy pass; a Europe PMC title search for PRECEDE, and the abstracts for PubMed ids 38626807 and 42976359 read through PubMed) · The caveat now gives the published results: 1,400 high-risk people imaged at baseline, one with stage IIB cancer at entry and 35.1 percent with pancreatic cysts, and five-year imaging in 2,085 carriers of BRCA1, BRCA2, ATM or PALB2 variants. It also says the recommendations paper has no indexed abstract and its full text could not be re-read
- 2026-11-17 · Circulating tumour DNA as a prognostic marker (Pietrasz 2017): The summary line gave survival of 6.5 against 19 months as a result among all 135 patients. It is a result among the 104 with advanced disease. The allele fraction after surgery was called a mean, which the abstract does not say (Second accuracy pass; the abstract for DOI 10.1158/1078-0432.CCR-16-0806 read through PubMed) · The summary line now says 104 patients with advanced pancreatic cancer and calls the figures medians; the word mean is removed
- 2026-11-17 · Hedgehog inhibition and chemotherapy delivery in mice (Olive 2009): A caveat said hedgehog inhibitors were harmful or futile in patients and cited Rhim 2014 and Ozdemir 2014, which are mouse studies (Second accuracy pass; the abstracts for PubMed ids 24856585, 26527777 and 26390428 read through PubMed) · The caveat now cites the patient trials: vismodegib with gemcitabine did not improve response, progression-free or overall survival in a randomised phase 2 trial, and a phase 2 trial of IPI-926 with gemcitabine indicated detrimental effects. The two mouse studies are named as the explanation
- 2026-11-17 · Recurrent GNAS mutations in pancreatic cysts (Wu 2011): The summary line said two-thirds of the mucinous kind of cyst carry a GNAS mutation. The 66 percent is of IPMNs only, and the paper found no GNAS mutations in other cystic neoplasms (Second accuracy pass; the abstract for DOI 10.1126/scitranslmed.3002543 read through PubMed) · The summary line now says two-thirds of one precancerous kind, IPMN
- 2026-11-17 · KRAS G12D as a prognostic factor (Bournet 2016): The record said the 33 percent wild-type rate showed the assay over-called wild-type because of low cellularity. The paper reports 72 of 219 samples as wild-type and says nothing about missed mutations or cellularity (Second accuracy pass; the full text PMC4822095 read through Europe PMC) · Both sentences now state the count and say the paper does not report how many were missed mutations
- 2026-11-17 · Pancreatic molecular papers: unsupported wording: Six records carried wording their abstracts do not support: a Smad4 stain said to support choosing radiotherapy when the authors asked for validation (Crane 2011); cyst classification called almost perfect (Springer 2015); neoadjuvant chemotherapy said to lower ctDNA from 69 to 21 percent when two patient groups were compared (Groot 2019); and three first claims with no source (Royal 2010, Brahmer 2012, Sausen 2015) (Second accuracy pass; each abstract read through PubMed) · Each sentence now follows its abstract: the radiotherapy idea is called still unvalidated, the sensitivity range is stated, the two groups are named, and first becomes early
- 2026-11-17 · Daraxonrasib: The pancreatic note described the practical threshold as a RAS-mutant or RAS-unknown tumour and did not say the drug is approved. Daraxonrasib was approved in the United States on 26 August 2026 for adults with metastatic pancreatic adenocarcinoma after at least one prior systemic therapy or who are not candidates for multiagent systemic therapy, and the indication names no RAS mutation. The KRAS note also said the drug halved the death hazard when the hazard ratio was 0.40. (Second accuracy pass; the RASONQUE label and the Drugs@FDA record NDA220910 read through openFDA, and the abstract for PubMed id 42223072 read through Europe PMC.) · The label wording and approval date are added to the daraxonrasib, KRAS and KRAS inhibitor notes, and the hazard ratio is given as 0.40, a 60% reduction.
- 2026-11-17 · KRAS G12D: The note said G12D-selective inhibitors such as zoldonrasib and MRTX1133 are in trials. The only MRTX1133 trial is terminated. It also described Bournet's comparison group as other alleles when it was all other patients, wild-type included. (Second accuracy pass; ClinicalTrials.gov records NCT05737706 (terminated, formulation challenges), NCT07621718 and NCT07805954 read on 17 November 2026, and the abstract for PubMed id 27010960.) · Now says zoldonrasib is in phase 3 first-line pancreatic trials and the MRTX1133 trial was terminated; the comparison group is named correctly.
- 2026-11-17 · KRAS: The pancreatic note gave mutant allele copy gain as 296 of 853 dosage-annotated MSK samples. 853 is the number of annotated samples without a gain. (Second accuracy pass; the KRAS_MUT_GAINED attribute of the pdac_msk_2024 study recounted from the public cBioPortal API (296 of 1,157).) · Now 296 of 1,157, matching the molecular table.
- 2026-11-17 · MSI-high: This note and the mismatch repair target note called pembrolizumab the only approved immunotherapy route in pancreatic cancer. Dostarlimab is also approved in the United States for mismatch repair deficient recurrent or advanced solid tumours. (Second accuracy pass; the current US labels for Jemperli and Keytruda read through openFDA (effective 19 and 16 October 2026).) · Both notes now name pembrolizumab and dostarlimab as the tumour-agnostic routes.
- 2026-11-17 · NRG1: The note and the 2018 history entry on the pancreatic cancer page said NRG1 fusions were found in every KRAS wild-type tumour of the Heining series, and the history entry said Philip 2022 put NRG1 fusions at about 1% of all pancreatic cancers. Heining found targetable gene fusions in all four wild-type tumours of 17, with NRG1 recurrent among them, and Philip reports 1.3% of KRAS wild-type tumours. (Second accuracy pass; the abstracts for Heining 2018 and Jones 2019 read through Europe PMC and for PubMed id 35302596 through PubMed.) · Both passages reworded to the papers' figures; the US accelerated approval of zenocutuzumab for NRG1 fusion pancreatic adenocarcinoma is added from its label.
- 2026-11-17 · ATM: The note called ATM the second commonest pancreatic risk gene after BRCA2, citing Hu 2018, where ATM (2.3%) is commoner than BRCA2 (1.9%), and gave ATM mutations in 3 to 5% of tumours when the seven cBioPortal studies range from 1.0 to 4.5%. (Second accuracy pass; the abstracts for PubMed ids 29922827 and 28767289 and the cBioPortal recount of 17 November 2026.) · Reworded to each paper's ranking and to 1 to 5% by cohort, 3.0% in the largest.
- 2026-11-17 · ctDNA MRD positive: The note and the monitoring note on the pancreatic cancer page said ctDNA is found in 37 to 49% of patients after surgery. Lee 2019 reports 37% after surgery, and 49% is the preoperative rate in Groot 2019. The Signatera note said every ctDNA-positive patient recurred in both series, which only Lee reports. (Second accuracy pass; the abstracts for PubMed ids 31250894 and 31142500 read through PubMed.) · Now 37% after surgery in one series, and the recurrence statement is credited to Lee alone.
- 2026-11-17 · DETECT-A: a blood test plus PET-CT to screen 10,006 women with no symptoms: The summary said 15 of the 26 cancers found by the blood test were confined to the organ or regional and that standard screening found 24 cancers by mammography or colonoscopy. The paper reports 17 of 26 localised or regional, with 15 the number who had PET-CT, and the 24 were 20 breast, three lung and one colorectal cancer. A clause saying 99% were not led to an unnecessary procedure is not in the paper. (Second accuracy pass; the full text of PubMed id 32345712 read through NCBI PMC (PMC7509949).) · Figures corrected and the unsupported clause removed.
- 2026-11-17 · CCGA: the Circulating Cell-free Genome Atlas behind the Galleri methylation test: The summary said pancreatic cancer was one of the twelve pre-specified cancers and that its sensitivity was above the all-cancer average at each stage. The abstract gives 67.6% against 40.7% for stages I to III in the twelve cancers as a group and neither lists them nor gives a pancreatic figure; the full text could not be read. (Second accuracy pass; the abstract for PubMed id 34176681 read through PubMed, and the publisher page answered 403.) · Reworded to the abstract's figures with a note that the full text was not re-read.
- 2026-11-17 · KRAS allelic imbalance and mutant KRAS dosage in pancreatic cancer: The summary credited Chan-Seng-Yue 2020 with the finding that squamous histology is the tissue form of the basal-like signature, which is Hayashi 2020. Its imbalance categories and the 12% hybrid figure, repeated on the GATA6 term, are not in the abstract and the full text is not open. (Second accuracy pass; the abstracts for PubMed id 31932696 and for Hayashi 2020 read through PubMed and Europe PMC.) · Attribution corrected, and both terms now say which figures could not be re-read.
- 2026-11-17 · Pancreatic cancer: The state-of-the-art line and the stroma note said every checkpoint trial outside mismatch repair deficiency has failed, on the strength of three studies; the precursor note gave about three years between high-grade dysplasia and cancer where Noe 2020 says an average of more than three years. (Second accuracy pass; the abstracts for PubMed ids 20842054, 22658128 and 31318392 read through PubMed and for Noe 2020 through Europe PMC.) · Reworded to checkpoint inhibitors given alone or as a pair, and to more than three years on average.
- 2026-11-17 · FoundationOne CDx: The pancreatic note said the 3,594-tumour Singhi series was run on this panel. The abstract describes capture panels of as many as 315 genes and does not name the test; the KDM6A target note also said Waddell 2015 first proposed the gene as a driver, a priority the paper does not claim. (Second accuracy pass; the abstracts and author affiliations for PubMed ids 30836094 and 25719666 read through Europe PMC.) · Reworded to what each abstract states.
- 2026-11-17 · Pancreatic cancer in the UK: waiting-time targets: The pathway and the four-nations table gave the Faster Diagnosis Standard as "75%, rising to 80% by March 2026" and a 62-day planning target of 75% by March 2026. Both dates have passed: the standard is now 80%, and the 2026/27 planning target is every trust at 94% for 31 days and 80% for 62 days by March 2027 (Second accuracy pass; the NHS England Medium Term Planning Framework 2026/27 to 2028/29 and the 2023 standards long-read read as raw text through NHS England's content API) · The three passages now give the current standards and planning targets and cite the framework
- 2026-11-17 · Pancreatic cancer in the UK: medicines NICE is working on: The funding note said zolbetuximab (ID6444) was in development at NICE and that trastuzumab deruxtecan (ID6511) was expected on 13 May 2027. NICE suspended the zolbetuximab appraisal on 17 February 2026 after the company said the trial missed its primary endpoint and that it would not seek a licence; ID6511 is awaiting development with no date. Pancreas SABR was described as at topic prioritisation when the decision, on 29 June 2022, was not selected (Second accuracy pass; the three NICE project pages read with plain curl) · The note now gives each status as NICE states it
- 2026-11-17 · EUROPAC: The record, the tests table and the trial list presented EUROPAC as recruiting and as the route into surveillance for families, and said the charity's page gave no year for a planned closure to new people. The page says EUROPAC will not be accepting new people from 1 January 2026. The protocol paper was also listed under a title it does not have (Second accuracy pass; the Pancreatic Cancer UK family history page read as raw HTML, the ISRCTN record read through the registry API, and PubMed record 40180412) · Status changed from recruiting to active, five passages now say new entry closed on 1 January 2026 and point to a genetics clinic referral, and the protocol title is the published one
- 2026-11-17 · Pancreatic cancer in the UK: trial statuses and sites: STARPAC2 was listed as recruiting to 31 December 2026; ClinicalTrials.gov (NCT04241276, updated 7 January 2026) shows it active and no longer recruiting, while the ISRCTN record was last edited in January 2025. Oxford was listed as a CRISTAL-APC site, and the registry lists Cambridge, The Christie and Barts only (Second accuracy pass; ISRCTN11503604 and ISRCTN16463547 read through the ISRCTN API and NCT04241276 through the ClinicalTrials.gov API) · STARPAC2 status now gives both registers, and Oxford is removed from CRISTAL-APC in the trial list and the centre list
- 2026-11-17 · Pancreatic cancer in the UK: NICE and SMC details: TA914 was said to cover endometrial, gastric, small intestine and biliary cancers only; it also covers colorectal cancer. TA644 was said to have been withdrawn; it was replaced by TA1118, terminated on 7 January 2026. The Scottish column said larotrectinib was accepted, but the SMC search returns no advice on it. Creon micro was described as accepted when the SMC accepted it for restricted use in young infants. An erlotinib non-submission was given as 141/04 in 2004; it is 382/07, 9 July 2007. The pathway said NG85 tells teams to offer a coeliac plexus block where it says consider (Second accuracy pass; the NICE TA914, TA644, TA1118 and NG85 pages and the SMC advice pages and keyword search read as raw HTML) · Each statement now matches the page, with entrectinib SMC2295 named as the accepted NTRK medicine in Scotland
- 2026-11-17 · Pancreatic cancer in the UK: ESPAC-4 hospitals: The legacy section said the 2017 Lancet report of ESPAC-4 covered 732 patients at 106 hospitals in the UK, France, Germany and Sweden. The paper says 92 hospitals in England, Scotland, Wales, Germany, France and Sweden; 106 is the registry's figure at the close of recruitment (Second accuracy pass; the abstract for PubMed record 28129987 read through PubMed, and ISRCTN96397434 through the registry API) · The sentence now gives 92 hospitals and the countries as the paper lists them
- 2026-11-17 · Andrew Biankin: The record said he co-authored the STARPAC2 protocol and listed it among his papers. He is not a named author or collaborator: the paper lists the Precision-Panc team as a group author and thanks him in the acknowledgements for protocol development (Second accuracy pass; the full text of PMC11748870 read through Europe PMC) · The sentence now says what the paper says and the paper is removed from his list
- 2026-11-17 · Paula Ghaneh and Daniel Palmer: Paula Ghaneh was called chief investigator of ESPAC5 and Daniel Palmer its co-lead. The paper shows her as first and corresponding author and him as second author; neither title is in the abstract or the registry record, and the full text could not be read (Second accuracy pass; the abstract for PubMed record 36521500 and ISRCTN89500674 read through the registry API) · The records now say first author and second author
- 2026-11-17 · Pancreatic cancer in the UK: mismatch repair figure: The tests table gave mismatch repair deficiency as about 1 percent with no source (Second accuracy pass; the abstract for PubMed record 29367431 read through PubMed) · The row now gives 7 of 833 tumours, 0.8 percent, and cites the paper
- 2026-11-17 · Pancreatic cancer: daraxonrasib in the chemotherapy and trial decisions: Three passages described daraxonrasib only as a trial drug given after first-line chemotherapy in RASolute 302. The trial has reported and the drug is approved in the United States (Second accuracy pass; the abstract for DOI 10.1056/NEJMoa2605555 read through Europe PMC, and the openFDA drugsfda record for NDA220910 and its label read through the API) · The text now gives the RASolute 302 result (500 previously treated patients, median survival 13.2 months against 6.7 with chemotherapy), the United States approval of 26 August 2026 and its indication, and tells the reader to ask about access where they are treated
- 2026-11-17 · Pancreatic cancer: NALIRIFOX funding: The chemotherapy decision and one question said only that NHS funding for NALIRIFOX "follows NICE appraisals". NICE terminated its appraisal on 2 April 2025 without a recommendation because the company made no evidence submission (Second accuracy pass; the NICE TA1052 page read as raw text with curl) · Both passages now state the terminated appraisal and its number, and the guideline line names TA1052
- 2026-11-17 · Pancreatic cancer: surgery first or chemotherapy first: The row said the question was unresolved for resectable disease "after PREOPANC-2, NORPACT-1 and Alliance A021806", as if A021806 had reported. It has not. The row also said borderline disease gets chemotherapy "for two to four months, usually modified FOLFIRINOX", which none of the cited pages states (Second accuracy pass; ClinicalTrials.gov records NCT04340141 and NCT04927780 read through the registry API, a Europe PMC search for A021806 reports in 2025 and 2026, and the Cancer Research UK and Pancreatic Cancer UK pages read as raw text) · The row now says the question is unresolved after PREOPANC-2 and NORPACT-1 and that PREOPANC-3 and Alliance A021806 had posted no results on 17 November 2026. The unsourced duration and regimen are replaced by what the patient pages say
- 2026-11-17 · Pancreatic cancer: NICE NG85 wording in the jaundice and chemotherapy decisions: The jaundice row credited NICE with offering a metal stent to people who will have chemotherapy first and with a fully covered stent as a firm recommendation. NG85 1.7.3 covers only people not yet fit for surgery and 1.7.4 is a "consider". One question said NG85 "offers" gemcitabine combination therapy, which 1.9.5 says to consider (Second accuracy pass; the NICE NG85 recommendations page read as raw text with curl) · The stent sentence follows 1.7.3 and 1.7.4 and credits the chemotherapy-first use to Pancreatic Cancer UK. The question now reads offer, consider, offer
- 2026-11-17 · Pancreatic enzyme replacement therapy (PERT): how to take it: The term, a nutrition note and one question said the UK consensus guideline (Phillips 2021) "sets the same starting doses" as Pancreatic Cancer UK's 50,000 to 75,000 units a meal and 25,000 to 50,000 a snack. The guideline says at least 50,000 units with meals and 25,000 with snacks (Second accuracy pass; the full text of DOI 10.1136/bmjgast-2021-000643 read through Europe PMC (PMC8212181), statement 4.2) · All three now give the guideline's own figures
- 2026-11-17 · Eating after a Whipple operation: slow stomach emptying, dumping and bowel changes: The summary explained dumping after a classic Whipple operation and gave dietary advice for it in a sentence credited to Macmillan. None of the pages cited on the record mentions dumping (Second accuracy pass; every cited page read as raw text and searched, then Macmillan's dumping syndrome page read the same way) · The sentence now reports what Macmillan's dumping syndrome page says, notes that the page is written about stomach and oesophagus operations, and links it
- 2026-11-17 · Coeliac plexus block for pancreatic cancer pain: The summary said Macmillan describes an injection of "local anaesthetic and a steroid (or a nerve-destroying agent)". Macmillan's page names only local anaesthetic and a steroid (Second accuracy pass; the Macmillan page on managing symptoms of pancreatic cancer read as raw text) · The added words are removed
- 2026-11-17 · Pancreatic cancer: cachexia guideline, palliative care trial and three attributions: Three passages said the ASCO cachexia guideline "recommends dietary counselling and says no drug is established". The guideline says counselling may be offered, that no drug can be recommended as the standard of care, and that a progesterone analog or a short course of corticosteroid may be used. One question cited a trial of early palliative care without saying it was in metastatic lung cancer. Three sentences credited a statement to the wrong body: home care services to Macmillan (Cancer Research UK), a nasojejunal tube to Pancreatic Cancer UK (Macmillan), and the GP to the NHS page on palliative teams (Second accuracy pass; the abstracts for DOI 10.1200/JCO.20.00611 and DOI 10.1056/NEJMoa1000678 read through Europe PMC, and the NHS, Macmillan, Cancer Research UK and Pancreatic Cancer UK pages read as raw text) · The guideline is reported in its own terms, the trial is named with its population, and each statement is credited to the page that makes it
- 2026-11-17 · Pancreatic cancer roadmap: The roadmap said the question of chemotherapy before surgery for resectable tumours had been left open by PREOPANC-2 and NORPACT-1 and did not mention Prep-02/JSAP05, a randomised trial of 364 patients in which neoadjuvant gemcitabine plus S-1 gave a median survival of 37.0 months against 26.6 months with surgery first (hazard ratio 0.73, 95% CI 0.56 to 0.95). The idea record on the same question said two trials had failed to show that chemotherapy first beats operating first, when PREOPANC-2 had no surgery-first arm, and an open problem on the pancreatic cancer page said the case for neoadjuvant treatment rested on secondary endpoints (Second accuracy pass; the abstracts for PubMed ids 40235447, 38237621 and 40945523 read through PubMed) · The roadmap summary and step, the idea's summary line and text, the PREOPANC-2 paper record and the open problem now give the Prep-02/JSAP05 result with a link, and say the negative results apply to FOLFIRINOX
- 2026-11-17 · Pancreatic cancer roadmap: Several statuses were those of 24 September 2026. RASolute 303 was given a primary completion date of June 2028 where the registry now lists April 2028. IMCODE003 was described as enrolling when it is active and no longer recruiting. The roadmap said cachexia had no phase 3 trial, when a phase 2b/3 trial of ponsegromab with first-line chemotherapy in metastatic pancreatic cancer (982 planned participants) is recruiting. The PRECEDE cohort was described with no results, when it has published baseline imaging in 1,400 people and five-year imaging in 2,085 gene carriers (Second accuracy pass; ClinicalTrials.gov records NCT07491445, NCT05968326, NCT06989437 and NCT04970056 read through the API on 17 November 2026, and the abstracts for PubMed ids 38626807 and 42976359 read through PubMed) · Dates and statuses corrected in the steps, the watch list and the ideas, the two PRECEDE reports and the ponsegromab trial added with links, and the note on registry dates now says when they were re-read
- 2026-11-17 · Pancreatic cancer roadmap: The roadmap called daraxonrasib the first RAS inhibitor approved for pancreatic cancer and olaparib the first biomarker-directed approval in the disease, with no source for either; the 2018 ASCO guideline had already recommended pembrolizumab for mismatch repair deficient tumours. It gave no European status for daraxonrasib. It also said ESPAC-3 showed fluorouracil and gemcitabine equivalent, when the trial found no survival difference, and that KRYSTAL-1 did the same for adagrasib as CodeBreaK 100 did for sotorasib, when KRYSTAL-1 reports responses in 7 of 21 pancreatic patients and no pancreatic survival figure (Second accuracy pass; the Drugs@FDA record NDA220910 and the Rasonque label read through openFDA, the FDA approval notice and the European Medicines Agency page for Rasonque read as raw text, and the abstracts for PubMed ids 29791286, 20823433 and 37099736 read through PubMed) · The two unsourced firsts are removed, the US approval date of 26 August 2026, the label indication and the European committee opinion of 15 October 2026 are added, and the ESPAC-3 and KRYSTAL-1 sentences give what each paper reports
- 2026-11-17 · Run the ENDPAC score on every new diabetes diagnosis after 50 and scan the high scorers: The idea and the ENDPAC paper record gave only the developers' 2018 figures (pancreatic cancer in 3.6% of high scorers) and named the Hungarian NODES cohort and a US cohort as the studies attempting validation. A retrospective validation in 70,050 people in UK primary care, published in November 2026, found lower accuracy: area under the curve 0.734, sensitivity 62.6%, specificity 72.5% and a positive predictive value of 0.6%. The NODES registry record has not been updated since April 2023 (Second accuracy pass; the abstract for DOI 10.3399/BJGP.2024.0253 read through PubMed, and ClinicalTrials.gov records NCT04662879 and NCT04164602 read through the API) · The UK validation and its figures are added to the idea, the paper record, the roadmap and the open problem, the summary line no longer presents 1 in 30 as settled, and the prospective Early Detection Initiative (8,869 participants, primary completion estimated January 2028) is named with its status
- 2026-11-17 · Pancreatic enzyme replacement by default: The idea, the roadmap, the Roberts 2019 paper record and an open problem gave the UK prescribing rate for enzyme replacement as about one patient in five, the figure from a 2019 study, as though it were current. The National Pancreatic Cancer Audit's 2026 report found a primary care prescription in 55% of people diagnosed in England in 2022 and 2023, and in 67% of those who lived at least 90 days. The idea also said NICE NG85 recommends pancreatin after resection, where the guideline says to consider it before and after resection (Second accuracy pass; the text of the audit's State of the Nation Report 2026 and the recommendations of NICE NG85 read as raw text, and the abstract for PubMed id 30385188 read through PubMed) · Each passage now gives the 2019 figure with its source and the audit's 2026 figure beside it, the NG85 wording is corrected, and the idea no longer says the trial has never been done
- 2026-11-17 · Probability of pancreatic cancer following diabetes (Chari 2005): The record said a 1% rate of pancreatic cancer after new diabetes is a hundred times the general population's. The paper's own comparison is an observed-to-expected ratio of 7.94 (Second accuracy pass; the abstract for PubMed id 16083707 read through PubMed) · Now says about eight times the expected rate, with the ratio
- 2026-11-17 · Ponsegromab for the Treatment of Cancer Cachexia: The summary line said patients gained about two to three kilograms. The trial reports differences from placebo of 1.22, 1.92 and 2.81 kg at the three doses. The record also called ponsegromab the first drug to reverse cachexia mechanistically, with no source, and said a phase 3 programme remained to come (Second accuracy pass; the abstract for PubMed id 39282907 read through PubMed and ClinicalTrials.gov NCT06989437 read through the API) · Now about one to three kilograms more than placebo depending on dose; the unsourced first is removed and the recruiting phase 2b/3 trial in metastatic pancreatic cancer is named and linked. The cachexia idea record carries the same trial
- 2026-11-17 · Estimated Projection of US Cancer Incidence and Death to 2040: The record said the 2014 projection made pancreatic cancer the second cause of cancer death in the UK's peer countries. That projection covers the United States only (Second accuracy pass; the abstract for PubMed id 24840647 read through PubMed) · Now says the United States
- 2026-11-17 · Relationship of carbohydrate antigen 19-9 and Lewis antigens in pancreatic cancer: The summary line said about one patient in ten cannot make CA 19-9 and that a normal test then says nothing. The 1987 paper studied 20 patients and gives no population frequency; a later series of 1,482 patients found 8.4% Lewis-negative, with CA 19-9 raised in 27.4% of them. The roadmap and an open problem repeated the tenth (Second accuracy pass; the abstracts for PubMed ids 3308077 and 30131287 read through PubMed) · The summary line and the repeats now give the later series' figure, and a caveat with a link says the marker is not always absent
- 2026-11-17 · PANOVA-3 paper and PRODIGE 24 five-year paper: The PANOVA-3 record called the Optune Pax approval the first new approval in locally advanced disease in decades, and the PRODIGE 24 five-year record called its result the best survival ever recorded in a pancreatic cancer trial. Neither claim has a source (Second accuracy pass; the openFDA device record P250034 and the abstracts for PubMed ids 40448572 and 36048453 read) · Both claims are removed; the approval date of 11 February 2026 is given with a link, and the PRODIGE 24 record gives the median of 53.5 months
- 2026-11-17 · RAS inhibitor combinations and sequence: The summary line called daraxonrasib the first drug against the KRAS protein. Sotorasib and adagrasib were tested in pancreatic cancer before it. The idea record on neoadjuvant chemotherapy said a third of patients never receive adjuvant chemotherapy after a Whipple operation, with no source (Second accuracy pass; the abstracts for PubMed ids 36546651 and 37099736 read through PubMed, and the text of the National Pancreatic Cancer Audit's State of the Nation Report 2026) · The summary line now says a drug that blocks the RAS proteins; the neoadjuvant idea gives the audit's figure, 62% in England receiving chemotherapy or chemoradiotherapy within 14 weeks of a Whipple operation, as 38% who did not
- 2026-11-17 · Evaluation of the Stereotactic MR-guided Adaptive Radiotherapy for Locally Advanced Pancreatic Cancers: The record said this phase 2 radiotherapy trial was not yet recruiting, with 160 participants planned. The registry entry, updated on 2026-11-12, lists it as withdrawn with an enrolment of 0 and the reason "The study was withdrawn prior to enrollment due to lack of funding" (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT07097064 read as raw JSON and compared field by field by script, then read by hand) · Status set to withdrawn and enrolment to 0; the one-line summary and the summary now say the trial was withdrawn before anyone was enrolled and quote the registry's reason word for word
- 2026-11-17 · Gemcitabine, Cisplatin and Nab-Paclitaxel as Neoadjuvant Treatment for Patients With Resectable or Borderline Resectable Pancreatic Cancer: The record named Emory University as the sponsor. The registry entry, updated on 2026-09-30, gives Georgetown University as the lead sponsor with Emory University as a collaborator, a primary completion date of 2027-08-31 where the record had 2026-12-31, and a reworded first eligibility line (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT06423326 read as raw JSON and compared field by field by script, then read by hand) · Sponsor, one-line summary and summary now name Georgetown University and note Emory as a collaborator; the date follows the registry; the quoted eligibility line gives the registry's current words
- 2026-11-17 · GnP Combined With SHR-1701 and Apatinib as First-Line Treatment for Metastatic PDAC: The record described a trial for locally advanced or metastatic pancreatic cancer that was not yet recruiting. The registry entry, updated on 2026-11-09, now covers metastatic disease only in its titles, conditions and eligibility text, and the trial opened on 2026-09-21 (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT07802184 read as raw JSON and compared field by field by script, then read by hand) · Name, official title, conditions and the quoted eligibility line now carry the registry's current words; the link to the locally advanced page is removed; status set to recruiting with the actual start date
- 2026-11-17 · Liquid Biopsy (ctDNA) Guided Treatment in Localized Pancreatic Cancer: Neoadjuvant CTX vs. Upfront Surgery: The record called this a phase 3 trial that was recruiting and did not say it was randomised. The registry entry, updated on 2026-11-03, gives phase 2, randomised allocation, a status of active and not recruiting, and an actual primary completion date of 2026-06-01 where the record had 2025-12-31 as a date still to come (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT06391892 read as raw JSON and compared field by field by script, then read by hand) · Phase set to 2 in the field, the one-line summary and the summary, which now says randomised; status set to active; the summary says the trial reached its primary completion on 2026-06-01
- 2026-11-17 · Tissue Adhesive Glue Modified Cyanoacrylate (Glubran® 2) in Soft Pancreas: The record called this a phase 3 trial. The registry entry, updated on 2026-10-01, gives the phase as not applicable (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT07155525 read as raw JSON and compared field by field by script, then read by hand) · Phase set to not applicable; the one-line summary and the summary now say the registry lists no phase
- 2026-11-17 · Pancreatic cancer: registry trials whose recruitment status changed: 17 records carried a recruitment status the registry has since changed: NCT04172532 (was recruiting, now active and not recruiting, entry updated 2026-11-12), NCT06391892 (was recruiting, now active and not recruiting, entry updated 2026-11-03), NCT06736717 (was recruiting, now active and not recruiting, entry updated 2026-09-25), NCT05093231 (was recruiting, now active and not recruiting, entry updated 2026-10-30), NCT05065801 (was recruiting, now active and not recruiting, entry updated 2026-11-09), NCT05053971 (was recruiting, now active and not recruiting, entry updated 2026-11-16), NCT05927142 (was recruiting, now active and not recruiting, entry updated 2026-09-29), NCT04605913 (was recruiting, now active and not recruiting, entry updated 2026-09-29), NCT04858009 (was recruiting, now active and not recruiting, entry updated 2026-10-15), NCT07705919 (was planned, now recruiting, entry updated 2026-10-05), NCT07670312 (was planned, now recruiting, entry updated 2026-09-25), NCT07678593 (was planned, now recruiting, entry updated 2026-10-05), NCT07802184 (was planned, now recruiting, entry updated 2026-11-09), NCT07748611 (was planned, now recruiting, entry updated 2026-10-29), NCT07683221 (was planned, now recruiting, entry updated 2026-10-13), NCT07806058 (was planned, now recruiting, entry updated 2026-10-26), NCT07784374 (was planned, now recruiting, entry updated 2026-10-02) (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · Status set to active for the 9 that have closed to recruitment and to recruiting for the 8 that have opened; each one-line summary now says "active but no longer recruiting" or "now recruiting" to match
- 2026-11-17 · Pancreatic cancer: registry trials with an unknown status: Six records said the trial was now recruiting or not yet recruiting where the registry now gives the overall status as unknown, because the entry has not been verified for two years or more: NCT05988372 (last known status not yet recruiting, last verified 2023-10), NCT06329947 (last known status not yet recruiting, last verified 2024-05), NCT05679583 (last known status recruiting, last verified 2022-12), NCT06229340 (last known status recruiting, last verified 2024-01), NCT06389760 (last known status recruiting, last verified 2024-10), NCT06151262 (last known status recruiting, last verified 2023-10) (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · Each one-line summary and summary now says what the trial was last listed as, that the registry gives its status as unknown, and when the entry was last verified; the status field is left as it was because the schema has no value for unknown
- 2026-11-17 · Pancreatic cancer: two registry trials enrolling by invitation: Two records said "now recruiting" where the registry status is enrolling by invitation, which a reader cannot simply join: NCT07386704, NCT07344220 (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · Each one-line summary now says "now enrolling by invitation only"
- 2026-11-17 · Pancreatic cancer: registry trial dates moved by their sponsors: 24 records carried a start or primary completion date the registry has since changed: NCT06115499 (primary completion 2029-01-15 now 2027-12-01); NCT07705919 (start 2026-09-01 now 2026-10-01); NCT07670312 (start 2026-07-15 now 2026-09-17, primary completion 2029-07-15 now 2029-09-15); NCT07802184 (start 2026-09-10 now 2026-09-21); NCT07748611 (start 2027-02-03 now 2026-10-26); NCT07683221 (start 2026-07-15 now 2026-08-05); NCT07806058 (start 2026-09 now 2026-10-15); NCT07784374 (start 2026-08-31 now 2026-08-20); NCT06423326 (primary completion 2026-12-31 now 2027-08-31); NCT06391892 (primary completion 2025-12-31 now 2026-06-01); NCT06736717 (start 2025-02 now 2025-02-17, primary completion 2026-07 now 2026-07-16); NCT07285044 (primary completion 2026-12-18 now 2027-05-15); NCT05944237 (start 2023-07-13 now 2023-08-02, primary completion 2027-06 now 2028-12-31); NCT05969860 (start 2023-08-23 now 2023-09-21, primary completion 2027-01-01 now 2027-12-01); NCT05093231 (primary completion 2027-07-31 now 2026-11-30); NCT05065801 (primary completion 2026-12-30 now 2027-04-30); NCT03492671 (primary completion 2026-09 now 2027-10); NCT06605430 (primary completion 2026-09 now 2027-12); NCT06807437 (primary completion 2027-11-01 now 2029-04-01); NCT05927142 (primary completion 2026-04 now 2026-12); NCT04605913 (start 2022-04-01 now 2022-04-25, primary completion 2027-10-01 now 2026-08-17); NCT03190941 (primary completion 2027-06-29 now 2029-06-29); NCT07226856 (primary completion 2028-12-01 now 2027-02-28); NCT04858009 (primary completion 2029-07 now 2030-09-30) (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · The dates in each summary now match the registry, with "started" kept only where the registry marks the start date as actual
- 2026-11-17 · Pancreatic cancer: registry trials past their primary completion: 7 records said the trial was "due to reach its primary completion" on a date the registry records as an actual, past date: NCT04084496, NCT06391892, NCT06736717, NCT05218889, NCT06435260, NCT04605913, NCT06050317 (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and the date type compared by script) · Each summary now says the trial "reached its primary completion" on that date. None of them has results posted on the registry
- 2026-11-17 · Pancreatic cancer: registry trials that have not started: 100 records said the trial "started" on a date the registry gives only as an anticipated start, some of them in the future: NCT06866977, NCT07469956, NCT07138846, NCT07153289, NCT07415525, NCT06991491, NCT07745790, NCT04901741, NCT07814066, NCT07705074, NCT07028424, NCT06383078, NCT07277452, NCT07477418, NCT06759090, NCT06513455, NCT05810792, NCT07765836, NCT07230301, NCT07773363, NCT06573398, NCT07691593, NCT06233877, NCT06750861, NCT07689539, NCT06370754, NCT07578337, NCT06621095, NCT06009029, NCT03301805, NCT06166589, NCT06255912, NCT07705919, NCT06361030, NCT07711067, NCT07336953, NCT07678593, NCT07333287, NCT07824960, NCT07175389, NCT07342725, NCT05637567, NCT06888648, NCT06571461, NCT06662669, NCT05562297, NCT06999512, NCT05988372, NCT07827053, NCT07689955, NCT07698613, NCT07561463, NCT07504263, NCT07631871, NCT06329947, NCT07155629, NCT06259058, NCT06172036, NCT06492941, NCT04381130, NCT06916975, NCT06217042, NCT06210360, NCT06669078, NCT06234072, NCT06177522, NCT06700603, NCT07814859, NCT07233850, NCT07825753, NCT07752875, NCT06938503, NCT07446270, NCT07108504, NCT06888674, NCT07208539, NCT06409429, NCT07736612, NCT07802418, NCT07803783, NCT07665684, NCT07828756, NCT07383922, NCT07429643, NCT07831330, NCT07656376, NCT07301229, NCT07833241, NCT07272109, NCT06196671, NCT07235930, NCT07632118, NCT07306286, NCT07595172, NCT07504471, NCT07306273, NCT07619521, NCT06427460, NCT07240766, NCT06491472 (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and the date type compared by script) · Each summary now says "planned to start" with the registry date
- 2026-11-17 · Pancreatic cancer: registry trial enrolment figures: Two records carried an enrolment figure the registry has since changed: NCT06736717 (21 now 16, actual), NCT05132504 (30 now 39, estimated) (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · The enrolment field and the summary now give the registry figure
- 2026-11-17 · Pancreatic cancer: registry trials with an incomplete list of primary outcomes: 17 records gave only the first three of the primary outcome measures the registry lists, and three of those sentences stopped in the middle of a phrase. In NCT05438459 the measure left out was overall survival in pancreatic cancer. The records: NCT07670312 (4 measures), NCT07678593 (10 measures), NCT07752875 (4 measures), NCT07665684 (6 measures), NCT07828756 (4 measures), NCT07656376 (4 measures), NCT06483555 (4 measures), NCT05438459 (4 measures), NCT06782412 (3 measures), NCT05187338 (4 measures), NCT06844422 (4 measures), NCT07098598 (3 measures), NCT05944237 (4 measures), NCT05518903 (6 measures), NCT03608631 (4 measures), NCT06145074 (18 measures), NCT05720338 (4 measures) (Second accuracy pass; the primary outcome measures in the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared with the sentence by script) · Where the registry lists four measures all four are now given; where it lists more, the first three are given in full followed by the number of further measures
- 2026-11-17 · Pancreatic cancer: registry text changed in five trial records: NCT04605913 quoted an eligibility line with "adeno-squamous" where the registry reads "adenosquamous". NCT07769502 listed sites in Israel, Japan and Taiwan; the registry now also lists the United States. NCT07670312 lacked Gastric Cancer, which the registry now lists among the conditions. NCT05944237 lacked the tablet form now named in the second intervention and the sixth UK site, Ninewells Hospital in Dundee. NCT07784374 carried a brief title the sponsor has since shortened (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · Each record now carries the registry's current words
- 2026-11-17 · Pancreatic cancer trials open today: The term linked three trials that are not open: SCALOP-2, which reported in 2024 that neither high-dose chemoradiotherapy nor nelfinavir improved survival; Precision Promise, which the registry lists as completed; and NCT07097064, withdrawn before enrolling anyone for lack of funding. It also called PRIMUS 001 open, though recruitment closed on 31 December 2024 (Second accuracy pass; the ClinicalTrials.gov records read through the v2 API, the SCALOP-2 abstract (PubMed id 39059185) read through PubMed, and the ISRCTN75002153 record read through the ISRCTN API) · The three trials are no longer linked from the term, the text says PRIMUS 001 has closed to recruitment, and the withdrawn trial is recorded under the failed and stopped programmes with the registry's reason
- 2026-11-17 · Pancreatic cancer: the failed and stopped programmes and why: The summary called MRTX1133 "the first KRAS G12D inhibitor in the clinic", which no source read supports, and gave no reason for the trial stopping (Second accuracy pass; ClinicalTrials.gov record NCT05737706 read through the v2 API) · "first" is removed and the registry's reason, formulation challenges, is given
- 2026-11-17 · Pancreatic cancer drugs in England: NICE appraisals: The summary said the NICE in-development list carried zolbetuximab for claudin 18.2-positive disease, without saying the appraisal is suspended (Second accuracy pass; the NICE page for GID-TA11560 read as raw text with curl) · The text now says NICE suspended the appraisal on 17 February 2026 after the company announced that the trial missed its primary endpoint.
- 2026-11-17 · POLO: The record called olaparib maintenance "the first biomarker-directed drug approval in pancreatic cancer". No source read says so, and the 2018 ASCO guideline already recommended pembrolizumab for mismatch repair deficient tumours (Second accuracy pass; the abstract for PubMed id 29791286 read through PubMed and the Lynparza label read through the openFDA API) · "first" is removed from the summary and the short description
- 2026-11-17 · Resectable pancreatic ductal adenocarcinoma: The page set out the evidence on chemotherapy before surgery without Prep-02/JSAP05, the one randomised trial that found a survival gain over surgery first in resectable disease. It also said every patient "should be offered" germline testing, and the borderline resectable page said testing "is offered to all"; the guidance read says testing may be discussed with everyone diagnosed (Second accuracy pass; the abstracts for DOI 10.1097/SLA.0000000000006730 and DOI 10.1200/JCO.18.01489 read through PubMed and Europe PMC) · Prep-02/JSAP05 is added with its figures (364 patients, median survival 37.0 against 26.6 months, hazard ratio 0.73, 95% CI 0.56 to 0.95). The germline sentences on both pages now follow the ASCO provisional clinical opinion, and both papers are linked
- 2026-11-17 · Pancreatic cancer with a BRCA or PALB2 variant: The row for treatment after a PARP inhibitor named daraxonrasib "after first-line chemotherapy" without saying it is approved or what the trial found (Second accuracy pass; the Rasonque label and the Drugs@FDA record for NDA220910 read through the openFDA API, and the RASolute 302 abstract (PubMed id 42223072) read through PubMed) · The row gives the United States approval of 26 August 2026 and the result: median survival 13.2 against 6.7 months with chemotherapy, hazard ratio 0.40
- 2026-11-17 · Pancreatic cancer: The base record said 2026 brought "the first new approval in locally advanced disease in decades" and "the first RAS inhibitor", and that daraxonrasib "delivered the biggest result in the disease's history". None of the three has a source. A guideline paper's caveat read as if the Alliance A021806 trial had reported, the treatment decision tool said NHS funding for NALIRIFOX "follows NICE appraisals", and an open problem said daraxonrasib is approved in the United States "only after first-line chemotherapy", which leaves out half of the label (Second accuracy pass; the openFDA device record P250034, the Rasonque label, the RASolute 302 abstract, ClinicalTrials.gov record NCT04340141 and the NICE TA1052 page, each read as raw text) · The record gives the two approvals with their dates (Optune Pax 11 February 2026, daraxonrasib 26 August 2026) and the trial medians in place of the superlatives. The caveat says Alliance A021806 has not reported. The decision tool says NICE could not make a recommendation on NALIRIFOX because the company made no submission (TA1052, terminated April 2025). The open problem gives the label wording: after at least one prior systemic therapy, or for people who are not candidates for multi-agent treatment
- 2026-11-17 · Daraxonrasib: The summary gave the phase 1/2 result as "median OS ~14.5 months in second-line pancreatic cancer versus historical ~6 months", a figure the published report does not contain, and did not give the European status. The RASolute 302 record repeated the 14.5 months, and the plain-language line said the drug was "in a decisive trial" (Second accuracy pass; the abstract for DOI 10.1056/NEJMoa2505783 read through PubMed, and the EMA page for Rasonque read as raw text with curl) · The summary gives the published medians (13.1 months in 26 second-line patients with RAS G12 mutations at 300 mg; 15.6 months among 38 patients with RAS G12, G13 or Q61 mutations), the RASolute 302 result and the EMA committee's positive opinion of 15 October 2026, with no European Commission decision published on 17 November 2026. The plain-language line says the drug is approved in the United States
- 2026-11-17 · RASolute 302: The plain-language line called the trial "the biggest result in the disease's history", which no source read supports (Second accuracy pass; the abstract for PubMed id 42223072 read through PubMed and the Rasonque label read through the openFDA API) · The line now gives the result: 13.2 months against 6.7 with chemotherapy
- 2026-11-17 · Zoldonrasib: The plain-language line called it "the first drug aimed at KRAS G12D". MRTX1133, another KRAS G12D inhibitor, entered a phase 1 trial in March 2023 (Second accuracy pass; ClinicalTrials.gov records NCT05737706, NCT07621718 and NCT07805954 read through the v2 API) · "first" is removed and the line says phase 3 trials are recruiting
- 2026-11-17 · KRAS inhibitors: The technology, the KRAS target and the KRAS roadmap listed MRTX1133 as a current KRAS G12D inhibitor and described daraxonrasib as still in phase 3; the roadmap gave "~14.5-month OS in second-line pancreatic cancer". The only MRTX1133 trial was terminated in March 2025 and daraxonrasib was approved in the United States on 26 August 2026 (Second accuracy pass; ClinicalTrials.gov record NCT05737706 read through the v2 API, the Rasonque label read through the openFDA API, and the RASolute 302 abstract read through PubMed) · All three records now say the MRTX1133 trial was terminated, give the approval, and quote the phase 3 result (13.2 against 6.7 months, hazard ratio 0.40)
- 2026-11-17 · Study of MRTX1133 in Patients With Advanced Solid Tumors Harboring a KRAS G12D Mutation: The short description said the trial was in non-small-cell lung cancer and rectal cancer and the summary said it was "due to reach its primary completion". The registry describes a trial in advanced solid tumours with a KRAS G12D mutation that was terminated (Second accuracy pass; ClinicalTrials.gov record NCT05737706 read through the v2 API) · The record now says the trial was terminated after 63 participants and completed on 10 March 2025, with formulation challenges as the registry's reason
- 2026-11-17 · NALIRIFOX: The European approval note said "NALIRIFOX label 2025". The first-line indication was added to Onivyde by a variation whose committee opinion was published on 22 March 2024 (Second accuracy pass; the EMA page for Onivyde pegylated liposomal read as raw text with curl) · The note now reads "first-line NALIRIFOX indication added by a 2024 variation"
- 2026-11-17 · Study of Daraxonrasib and Daraxonrasib + GnP as First-line Treatment in Patients With Metastatic Pancreatic Adenocarcinoma: The summary gave the primary completion date as 2028-06. The registry now lists April 2028 (Second accuracy pass; ClinicalTrials.gov record NCT07491445, last updated 16 November 2026, read through the v2 API) · Changed to 2028-04
- 2026-11-17 · RASolute 302: The eight translations of the plain-language line still called the result the biggest in the history of the disease and said it was presented at the ASCO 2026 plenary session, wording removed from the English line because no source read supports it (Second accuracy pass; the abstract of the trial report (PubMed 42223072) read in raw text) · Each translation now gives the survival figures the abstract reports, 13.2 months against 6.7
- 2026-11-17 · Triple-negative breast cancer: OlympiA population: The summary said OlympiA added adjuvant olaparib for germline BRCA carriers "with residual disease". The trial enrolled people with high-risk, HER2-negative early breast cancer after neoadjuvant or adjuvant chemotherapy, not only those with residual disease (Second accuracy pass; the abstract for DOI 10.1056/nejmoa2105215 read through Europe PMC) · The sentence now gives the trial population and the one-year course
- 2026-11-17 · Triple-negative breast cancer: iza-bren phase 3: The summary said iza-bren posted "the first positive phase 3 for a bispecific ADC". The report of the company announcement of 26 February 2026 describes interim topline results and calls it the third phase 3 study in which the drug reached its primary end points (Second accuracy pass; the OncLive report of the news release read as raw HTML, and ClinicalTrials.gov record NCT06382142 read through the registry API) · The sentence now says both primary endpoints were met at an interim analysis, that the result came by news release, that it was the drug's third such phase 3, and that no full report was found in Europe PMC
- 2026-11-17 · Triple-negative breast cancer: age and grade figures from Howlader 2014: The summary gave "10 to 30 percent less likely to be diagnosed at older ages" and "6.4 to 20 times more likely to present with high-grade disease" as figures for triple-negative patients. The paper gives those ranges for three subtypes together, triple-negative and the two HER2-positive subtypes, compared with hormone receptor-positive, HER2-negative patients (Second accuracy pass; the abstract for DOI 10.1093/jnci/dju055 read through Europe PMC) · The sentence now names the three subtypes and the comparison group
- 2026-11-17 · Triple-negative breast cancer: ER-low cohorts and the NG101 citation: The summary credited an Italian series (Dieci 2021) with matching germline BRCA prevalence in ER-low tumours; that paper reports response and survival only and does not mention BRCA. It also cited NG101 1.3.1 to 1.3.4 for recording receptor status at multidisciplinary meetings, which is recommendation 1.3.5 (Second accuracy pass; the full text of DOI 10.1038/s41523-021-00308-7 read through Europe PMC, and the NICE NG101 recommendations read from the Internet Archive copy of 15 November 2026) · Dieci is now cited for response and survival only, and the citation reads 1.3.1 to 1.3.5
- 2026-11-17 · Triple-negative breast cancer: history of claudin-low: The 2010 history entry said claudin-low was "described" as a subtype that year. The 2010 paper says it characterises a "recently identified" subtype (Second accuracy pass; the abstract for DOI 10.1186/bcr2635 read through Europe PMC) · The entry title now says the subtype was characterised in detail in 2010
- 2026-11-17 · Basal-like 1 triple-negative breast cancer (BL1): The biomarker list called an HRD score of 42 or more or a tumour BRCA mutation "the response marker for platinum in GeparSixto". The paper found that this marker predicted response to chemotherapy overall and concluded that "HRD does not predict for carboplatin benefit" (Second accuracy pass; the abstract for DOI 10.1093/annonc/mdy460 read through Europe PMC) · The line now says the marker predicted response overall but not extra benefit from carboplatin
- 2026-11-17 · Luminal androgen receptor triple-negative breast cancer (LAR): The treatment row here and on the apocrine carcinoma page gave "clinical benefit rates of 19 to 33 percent" for bicalutamide and enzalutamide. The 33 percent is the enzalutamide trial's evaluable subgroup with androgen receptor staining of 10 percent or more; the rate in all enrolled patients was 25 percent at 16 weeks, and 19 percent is the six-month rate with bicalutamide (Second accuracy pass; the abstracts for DOI 10.1158/1078-0432.ccr-12-3327 and DOI 10.1200/jco.2016.71.3495 read through Europe PMC) · Both rows now give 19 percent for bicalutamide and, for enzalutamide, 25 percent of all enrolled patients and 33 percent of the evaluable subgroup
- 2026-11-17 · Secretory carcinoma of the breast: The burden line said "under 0.15 percent of breast cancers in most series" and credited Thomas 2023, which gives no such figure. The European working group review cited on the same page says less than 0.02 percent of all breast cancers (Second accuracy pass; the full texts of DOI 10.1038/s41523-023-00554-x and DOI 10.3390/cancers13225694 read through Europe PMC) · The burden now reads less than 0.02 percent of all breast cancers, credited to Cserni 2021
- 2026-11-17 · PD-L1 assay discordance: The page said only the 22C3 combined positive score of 10 has an approved and NICE-recommended drug attached in triple-negative disease. NICE TA639 recommends atezolizumab with nab-paclitaxel when PD-L1 is 1 percent or more, and TA801 itself requires both a score of 10 or more and immune-cell staining below 1 percent (Second accuracy pass; the recommendations of NICE TA801 and TA639 read in Internet Archive copies, and the current US atezolizumab label read through openFDA) · Separates the US position, where the atezolizumab indication was withdrawn, from NICE guidance, which uses both scores
- 2026-11-17 · PD-L1 combined positive score 10 in triple-negative breast cancer: The NICE recommendation was described only as covering the appraised population, and atezolizumab approval was said to be withdrawn without saying where. The early-disease response figure was given without its analysis (Second accuracy pass; NICE TA801 recommendation 1.1 read in an Internet Archive copy, the IMpassion130 and KEYNOTE-522 abstracts read through Europe PMC, and the US atezolizumab label read through openFDA) · States the TA801 conditions, says the withdrawal was the US approval, adds the whole-population survival hazard ratio of 0.84 with its interval, and notes that the response figure is from the first interim analysis of 602 patients
- 2026-11-17 · Health disparities in cancer outcomes: Basal-like tumours were given as 16 percent of all other women in the Carolina study, and every ethnic minority group in England was said to have less favourable tumours (Second accuracy pass; the abstracts for DOI 10.1001/jama.295.21.2492 and DOI 10.1038/s41416-021-01409-7 read through Europe PMC) · Gives 14 percent in postmenopausal African American women and 16 percent in non-African American women, and says the England finding held for every group aged 53 to 70 and every group except Indian women aged 30 to 46
- 2026-11-17 · Ki-67 in triple-negative breast cancer: High Ki-67 was called an independent predictor of both relapse-free and overall survival in the Korean series. The multivariate result with the hazard ratio of 7.82 is for relapse-free survival only (Second accuracy pass; the abstract for DOI 10.1186/bcr2834 read through Europe PMC) · Says high Ki-67 was associated with worse relapse-free and overall survival and was an independent factor for relapse-free survival
- 2026-11-17 · Risk-reducing surgery for BRCA carriers: The oophorectomy cohort was called the largest prospective study. Its authors call it a large prospective study (Second accuracy pass; the abstract for DOI 10.1093/jnci/djw177 read through Europe PMC) · Now reads a large prospective study
- 2026-11-17 · Basal-like breast cancer: The TP53, mitotic index and grade comparison from the Carolina study was given without its comparator (Second accuracy pass; the abstract for DOI 10.1001/jama.295.21.2492 read through Europe PMC) · States that the comparison is with luminal A tumours
- 2026-11-17 · Datopotamab deruxtecan: The triple-negative supplement said there was no triple-negative indication in the EU and that the US indication had no EU counterpart yet. The CHMP adopted a positive opinion on first-line use in triple-negative disease in June 2026 and the EMA overview of Datroway now describes that use. The same stale line sat on the TROPION-Breast02 trial note. (Second accuracy pass; the EMA Datroway product page (last updated 30 September 2026), the CHMP post-authorisation summary of opinion PDF and the EMA variation page fetched as raw text with curl.) · Added an EU 2026 row for the triple-negative indication, corrected the note, the link label and the TROPION-Breast02 note, and said that the Commission decision date could not be read.
- 2026-11-17 · Sacituzumab govitecan: The EU status of sacituzumab govitecan with pembrolizumab (ASCENT-04) was missing, and the ASCENT-04 note carried a status dated 24 September 2026. The EMA records the extension of indication as accepted, with a Commission decision on 21 August 2026. (Second accuracy pass; the EMA procedural-steps PDF for Trodelvy read as text, the EMA product page, and a search of the NICE site on 17 November 2026.) · Added the EU row with opinion and decision dates and the caveat that the product page indication list is not yet updated, added the 22 June 2026 decision date to the ASCENT-03 row, and updated the ASCENT-04 trial note.
- 2026-11-17 · Residual disease after neoadjuvant therapy in triple-negative breast cancer: the decision point: TROPION-Breast03 was described as an open trial. The registry lists it as active and not recruiting, with 1,174 participants enrolled. (Second accuracy pass; ClinicalTrials.gov v2 API records for NCT05629585 and NCT05633654.) · Reworded: ASCENT-05 still recruiting, TROPION-Breast03 fully enrolled and closed to recruitment, neither with results posted.
- 2026-11-17 · Palliation in advanced triple-negative breast cancer: brain, bone, pleura, skin and end of life: TUXEDO-4 was described as recruiting. The registry lists it as active and not recruiting, with 28 participants and no posted results. (Second accuracy pass; ClinicalTrials.gov v2 API record for NCT06048718.) · Status corrected and the registry number added.
- 2026-11-17 · Trastuzumab deruxtecan: NICE TA1197 was called the rapid review of TA992, in the drug note and in the DESTINY-Breast04 note. The NICE overview, recommendation and history pages do not use that phrase; they say TA1197 updates and replaces TA992. (Second accuracy pass; the NICE TA1197 pages fetched as raw text with curl and the TA992 page from its Internet Archive copy.) · Reworded both notes to say that TA1197 updates and replaces TA992.
- 2026-11-17 · Atezolizumab: The March 2019 US accelerated approval was sourced to the 2018 IMpassion130 paper, which cannot support it, and the withdrawal gave only August 2021. (Second accuracy pass; the FDA approval notice, the FDA list of withdrawn cancer accelerated approvals and the Roche media release of 27 August 2021 fetched as raw text.) · Sources replaced with the FDA pages; the approval day (8 March 2019), the announcement day (27 August 2021) and the FDA withdrawal date (6 October 2021) are now stated.
- 2026-11-17 · NeoTRIP (NeoTRIPaPDL1): The record said the main endpoint, event-free survival, was still awaited and that no result was indexed. The event-free survival analysis was published as congress abstract LBA19 in Annals of Oncology in October 2023 (Second accuracy pass; the Crossref record for DOI 10.1016/j.annonc.2023.10.009 and the full text of a 2026 meta-analysis (DOI 10.3389/fimmu.2026.1958007) read through Europe PMC; the abstract page itself refused a script) · The record now says event-free survival was not improved, gives the hazard ratio of 1.08 (0.76 to 1.54) as cited by the meta-analysis, says that figure is second-hand, and links both sources
- 2026-11-17 · A-BRAVE: The co-primary endpoints were given as disease-free survival overall and in PD-L1-positive patients, and the record said the primary paper's figures were not indexed. The co-primary endpoints were disease-free survival in all randomised patients and in stratum B, and the primary paper was published in Annals of Oncology in 2025 (Second accuracy pass; the abstract for DOI 10.1016/j.annonc.2025.08.005 (PMID 40885530) read through PubMed) · Endpoints corrected; added disease-free survival hazard ratio 0.81 (0.61 to 1.09, p 0.172), three-year rates 68.3 versus 63.2 percent, the stratum B result and the descriptive overall survival hazard ratio 0.66 (0.45 to 0.97), with a link
- 2026-11-17 · BARBICAN: The record said the trial was due to report in 2026 and gave no result. Its pathological complete response result was published as congress abstract 123MO in Annals of Oncology in September 2021 (Second accuracy pass; the Crossref record for DOI 10.1016/j.annonc.2021.08.404 and the full text of two reviews citing it (DOI 10.3390/jpm13030526 and PMC10320476) read through Europe PMC; ClinicalTrials.gov read through its API) · The record now reports no gain from ipatasertib (49.3 versus 48.5 percent, p 0.729, as cited by the 2023 review), says the figures are second-hand and that the registry still shows no posted results
- 2026-11-17 · GeparDouze / NSABP B-59: The setting named the United States, Canada and Germany. The paper says 353 sites in four countries: the United States, Germany, Canada and Spain (Second accuracy pass; the full text of DOI 10.1038/s41591-026-04565-6 read through Europe PMC) · Setting corrected to four countries and 353 sites; link label now leads with the first author
- 2026-11-17 · SWOG S1418 / NRG BR006: The record said the trial was due to report from 2026 and dated its check to September 2026. The registry now records primary completion as reached on 3 August 2026 (Second accuracy pass; ClinicalTrials.gov NCT02954874 read through its API and PubMed and Europe PMC searched on 17 November 2026) · Status redated: primary completion 3 August 2026, no results posted or published on 17 November 2026
- 2026-11-17 · c-TRAK TN: The result line said 72 percent of ctDNA-positive patients had metastases on staging. The paper gives 72 percent of the 32 patients allocated to intervention (23 of 32) (Second accuracy pass; the abstract for DOI 10.1016/j.annonc.2022.11.005 read through PubMed) · Result line and outcome note now give the denominator
- 2026-11-17 · KEYNOTE-119: Neutropenia was given as below 1 versus 10 percent. Those figures are for decreased neutrophil count; neutropenia was 0 versus 13 percent (Second accuracy pass; the abstract for DOI 10.1016/S1470-2045(20)30754-3 read through PubMed) · Both figures now stated under their own names
- 2026-11-17 · SYSUCC-001: The short summary said fewer women relapsed at five years (83 versus 73 percent). Those are the shares alive and free of relapse, not the shares who relapsed (Second accuracy pass; the abstract for DOI 10.1001/jama.2020.23370 read through PubMed) · Reworded to say more women were alive and free of relapse
- 2026-11-17 · BROCADE3: The record called BROCADE3 the only phase 3 of a PARP inhibitor given with chemotherapy. BrighTNess, in the same file, is a phase 3 of veliparib with chemotherapy before surgery (Second accuracy pass; the abstracts for DOI 10.1016/S1470-2045(20)30447-2 and 10.1016/S1470-2045(18)30111-6 read through PubMed) · Reworded to a phase 3 trial in advanced disease, naming BrighTNess
- 2026-11-17 · PAKT: CAPItello-290 was described as using the same regimen and reporting in 2026. It gave paclitaxel at 80 mg/m2 against 90 in PAKT and was published online in December 2025 (Second accuracy pass; the abstracts for DOI 10.1200/JCO.19.00368 and 10.1016/j.annonc.2025.12.012 read through PubMed) · Reworded with the dose and the publication date
- 2026-11-17 · PARTNER: The treatment note and the germline BRCA standard-of-care row presented PARTNER as recruiting and on offer. The registry entry is unchanged since November 2022 and both cohorts have reported (Second accuracy pass; the abstracts for DOI 10.1038/s41586-024-07384-2 and 10.1038/s41467-025-59151-0 read through PubMed, and ClinicalTrials.gov read through its API) · Note now says the status is out of date and links both papers; the row gives the germline BRCA cohort result (pathological complete response 64.1 versus 69.8 percent, three-year event-free survival 96.4 versus 80.1 percent)
- 2026-11-17 · Triple-negative breast cancer: standard of care rows: Four statements were stale or loose: TROPION-Breast03 was offered although it has finished recruiting; TUXEDO-4 was called recruiting; the atezolizumab withdrawal was dated August 2021 only, where the FDA dates it 6 October 2021; and an 88 percent intracranial control rate was credited to a 26-patient HER2-low series when it applies to the 16 patients given trastuzumab deruxtecan in a 26-patient series of metastatic breast cancer (Second accuracy pass; ClinicalTrials.gov NCT05629585 and NCT06048718 read through the API, the FDA list of withdrawn accelerated approvals, and the abstract for DOI 10.1016/j.esmoop.2024.102995 read through PubMed) · Each row reworded to the source
- 2026-11-17 · Triple-negative breast cancer: open problems: PIK3CA, AKT1 and PTEN alterations were said to be present in a fifth to a quarter of tumours. CAPItello-290 found them in 30.7 percent and PAKT in 28 of 140 patients (Second accuracy pass; the abstracts for DOI 10.1016/j.annonc.2025.12.012 and 10.1200/JCO.19.00368 read through PubMed) · Replaced with the two trial figures
- 2026-11-17 · Triple-negative breast cancer: The molecular table said LOTUS randomised 124 patients with PTEN-low tumours. The trial randomised 124 patients in all, 48 of them PTEN-low, and the result in that group was not statistically significant. (Second accuracy pass; the abstract for DOI 10.1016/S1470-2045(17)30450-3 read through Europe PMC.) · Row now gives 124 randomised, 48 PTEN-low, hazard ratio 0.59 with 95% CI 0.26 to 1.32, not significant.
- 2026-11-17 · Triple-negative breast cancer: Three figures from the Swedish whole-genome study were misread: PALB2 variants were given as two biallelic of five when the paper reports four; 29 tumours with biallelic BRCA1 or BRCA2 loss were set against all 237 genomes when the paper counts them among the 139 HRDetect-high tumours; and 93% was given as the success rate at 30-fold depth when it is the overall rate (237 of 254). (Second accuracy pass; the full text of DOI 10.1038/s41591-019-0582-4 (PMC6859071) read through Europe PMC.) · PALB2 row, BRCA1 row and the Sweden row corrected to the paper's counts and denominators.
- 2026-11-17 · Triple-negative breast cancer: The ctDNA table gave relapse hazard ratios of 25.1 and 25.2 for detection during follow-up. In the 2015 study 25.1 came from a single sample after surgery and serial follow-up gave 12.0. (Second accuracy pass; the abstracts for DOI 10.1126/scitranslmed.aab0021 and 10.1001/jamaoncol.2019.1838 read through Europe PMC.) · Row now gives 25.1 (single post-surgical sample) and 12.0 (serial samples) for the first study and 25.2 for the second.
- 2026-11-17 · Triple-negative breast cancer: The Burstein subtype row gave log-rank P 0.042 and 0.041 as the disease-free and disease-specific survival values. Both are disease-free survival values (for the worst and best subtypes); the disease-specific values are 0.039 and 0.029. (Second accuracy pass; the abstract for DOI 10.1158/1078-0432.CCR-14-0432 read through Europe PMC.) · All four P values now stated against the right endpoint.
- 2026-11-17 · Triple-negative breast cancer: The NOTCH pathway note said gamma-secretase inhibitors were active in NOTCH1-rearranged models only, which the second cited paper contradicts. The androgen receptor row described 78 enzalutamide patients as all those with 10% or more nuclear staining when they were the evaluable subgroup. The tumour mutational burden row called a 62-patient cohort heavily pretreated, which the paper does not say. The MYC row gave a denominator of 550 the paper does not state. (Second accuracy pass; the abstracts for DOI 10.1158/2159-8290.CD-13-0830, 10.1158/1078-0432.CCR-14-1348 and 10.1200/JCO.2016.71.3495 read through Europe PMC, and the full texts of 10.1158/1078-0432.CCR-19-3507 and 10.1093/annonc/mdy024.) · Each statement reworded to what its source says.
- 2026-11-17 · Triple-negative breast cancer: The Ashkenazi row credited a 1997 paper with a statement about triple-negative tumours and testing order that it does not make. The basal-like 2 statement was credited to the 2011 Lehmann paper when the 2016 paper makes it. A United States mortality comparison had no link to its source, and the lower and upper ends of the two PD-L1 ranges (41% and 38%) were not written out or sourced. (Second accuracy pass; abstracts for DOI 10.1056/NEJM199705153362001 and 10.3322/caac.21583 read through Europe PMC, full texts of 10.1371/journal.pone.0157368 and 10.1093/jnci/djab108, and the ClinicalTrials.gov results for NCT02819518.) · Unsupported clause removed, citation moved, source links added, and the 40.9% and 323 of 847 figures stated.
- 2026-11-17 · The clonal and mutational evolution spectrum of primary triple-negative breast cancers: A caveat said the 104 tumours came from one Canadian centre. The paper acknowledges tumour banks in British Columbia, Alberta and Cambridge. (Second accuracy pass; the author manuscript PMC3863681 read through NCBI.) · Caveat rewritten.
- 2026-11-17 · Inherited breast cancer in Nigerian women: The page said West Africa carries the highest germline burden reported for an unselected breast cancer population. The paper does not claim this, and a Bahamian study on this site reports a higher share. (Second accuracy pass; the abstract and full text of DOI 10.1200/JCO.2018.78.3977 (PMC6161833) read through Europe PMC.) · Sentence replaced with the paper's own conclusion.
- 2026-11-17 · Genomic and transcriptomic landscape of triple-negative breast cancers: subtypes and treatment strategies: The summary line called this the largest Chinese triple-negative cohort, which the paper does not say. (Second accuracy pass; the abstract for DOI 10.1016/j.ccell.2019.02.001 read through Europe PMC.) · The word largest removed.
- 2026-11-17 · Exploring the spectrum of HER2 in non-metastatic triple negative breast cancer: from HER2-null to HER2-low, including HER2-ultralow status: The record said three-quarters of chemotherapy-naive triple-negative cancers would be HER2-low or ultralow. The paper gives 37.6% ultralow and 24.0% low, 61.6% together. One finding also ran two different comparisons into one: ultralow tumours were smaller and lower grade than HER2-null tumours, and had more BRCA1 promoter methylation than HER2-low tumours. (Second accuracy pass; the abstract for DOI 10.1007/s00428-026-04425-1 read through PubMed.) · Now says about six in ten (61.6%), and the finding names the comparator for each difference.
- 2026-11-17 · MYC pathway activation in triple-negative breast cancer is synthetic lethal with CDK inhibition: A caveat said the CDK inhibitor used was not a clinical agent. The paper tested two: purvalanol A, an experimental compound, and dinaciclib, which it describes as in phase 2 clinical trials at the time. (Second accuracy pass; the full text for DOI 10.1084/jem.20111512 read through Europe PMC (PMC3328367).) · Caveat rewritten to name both inhibitors and the clinical status of dinaciclib.
- 2026-11-17 · Capivasertib plus paclitaxel versus placebo plus paclitaxel as first-line therapy for metastatic triple-negative breast cancer: the PAKT trial: The short summary gave the phase 2 survival gain (19.1 against 12.6 months) without saying the phase 3 trial did not confirm it, and the caveat said only that CAPItello-290 was negative, with no figures. CAPItello-290 randomised 812 patients: median overall survival 17.7 against 18.0 months, hazard ratio 0.92 (95% CI 0.78 to 1.08), not significant, and 20.4 months in both arms in PIK3CA, AKT1 or PTEN-altered tumours. (Second accuracy pass; the CAPItello-290 abstract (Annals of Oncology 2026, PubMed 41422862) found by PubMed search and read as raw text.) · The short summary now says the larger phase 3 trial found no survival gain, and the caveat carries the phase 3 figures.
- 2026-11-17 · Ipatasertib plus paclitaxel versus placebo plus paclitaxel as first-line therapy for metastatic triple-negative breast cancer (LOTUS): The caveat said the phase 3 IPATunity130 trial did not confirm the benefit, without figures. In its triple-negative cohort of 255 patients with PIK3CA, AKT1 or PTEN-altered tumours, median progression-free survival was 7.4 months with ipatasertib and 6.1 with placebo, hazard ratio 1.02 (95% CI 0.71 to 1.45). (Second accuracy pass; the posted results for NCT03337724 read through the ClinicalTrials.gov API, and the cohort A abstract (PubMed 39058425).) · Caveat now gives the phase 3 result.
- 2026-11-17 · Assessment of molecular relapse detection in early-stage breast cancer: The record gave the 10.7-month median lead time of the blood test over clinical relapse as a result in the main cohort of 101 women. The abstract reports it for the combined cohort of 144, which adds 43 patients from an earlier study. (Second accuracy pass; the abstract for DOI 10.1001/jamaoncol.2019.1838 read through PubMed.) · The short summary and the summary now attribute the lead time to the combined cohort.
- 2026-11-17 · The impact of expanded access to germline high penetrance genetic testing for women with a new diagnosis of invasive breast cancer or high-grade DCIS: The summary gave risk-reducing surgery uptake of 100% with a result before surgery against 71.4% after, without saying the difference was not statistically significant (P = .24). (Second accuracy pass; the abstract for DOI 10.1016/j.clbc.2026.07.031 read through PubMed.) · The summary now says the difference was not statistically significant.
- 2026-11-17 · Triple-negative breast cancer: Four statements in the molecular notes. (1) Blood tumour DNA was called a stratification factor in ASCENT-05 and ZEST: the ASCENT-05 registry record does not mention it, and in ZEST detectable tumour DNA was an entry requirement, with 40 participants enrolled. (2) The c-TRAK TN line read as if 72% of all women with a positive blood test already had metastases: it was 23 of the 32 allocated to intervention and staged, and none of the five who started pembrolizumab had sustained clearance. (3) High mutation burden was given as 18% of pretreated metastatic disease: 40% of that 62-patient cohort had immunotherapy as first-line treatment. (4) A combined positive score of 10 or more was given as 38% of KEYNOTE-355 screening: it is 38% of the 847 patients randomised. (Second accuracy pass; the ClinicalTrials.gov records NCT05633654 and NCT04915755, the c-TRAK TN abstract (DOI 10.1016/j.annonc.2022.11.005), the full text of Barroso-Sousa 2020 (PMC7269810) and the current US prescribing information for pembrolizumab from openFDA, each read as raw text.) · The state-of-the-art lines and the notes on the ctDNA, tumour mutational burden and pembrolizumab records were reworded to what the sources say, with denominators.
- 2026-11-17 · Triple-negative breast cancer in the UK: PARTNER trial story: The page said the Nature 2024 paper reported 100 percent three-year survival in 39 germline BRCA patients given olaparib after a 48-hour gap, and did not say the main trial was negative. The Nature 2024 paper is the cohort of 559 patients without a germline BRCA mutation, in which olaparib did not improve response or survival. The BRCA figures come from a separate small cohort published in Nature Communications in 2025 and a University of Cambridge release dated 13 May 2025, which the page dated 2024 (Second accuracy pass; the abstracts for DOI 10.1038/s41586-024-07384-2 and DOI 10.1038/s41467-025-59151-0 read through Europe PMC, and the University of Cambridge release read as raw HTML) · The story now gives the negative main result first (response 51 against 52 percent), then the germline BRCA cohort with its unimproved response rate, its survival figures and the authors' statement that the result needs confirmation in larger trials. The source date is 13 May 2025 and the Nature Communications paper is cited
- 2026-11-17 · Jean Abraham: The summary credited the germline BRCA survival result to the Nature 2024 paper and left out that olaparib made no difference in the 559 patients without a germline BRCA mutation (Second accuracy pass; the two PARTNER abstracts read through Europe PMC and the University of Cambridge release read as raw HTML) · The summary now states the negative result of the main cohort, attributes the germline BRCA result to the Nature Communications 2025 paper and the release of 13 May 2025, and lists that paper
- 2026-11-17 · Triple-negative breast cancer in the UK: OlympiA story: The page said the NEJM 2021 paper showed olaparib improves overall survival, named Breast Cancer Now among the trial's funders and gave Andrew Tutt's affiliation as the ICR and The Royal Marsden. In the 2021 paper the survival difference was not statistically significant at the interim analysis, the funders named are the National Cancer Institute and AstraZeneca, and the affiliation is the ICR and King's College London (Second accuracy pass; the abstracts and affiliations for DOI 10.1056/NEJMoa2105215 and DOI 10.1016/j.annonc.2022.09.159 read through Europe PMC) · The story now says the 2021 paper showed better invasive disease-free survival (hazard ratio 0.58) and that overall survival became significant at the second interim analysis (hazard ratio 0.68, Annals of Oncology 2022), with the funders and affiliation as the paper gives them
- 2026-11-17 · Triple-negative breast cancer in the UK: NICE appraisals awaiting development: The page said no NICE appraisal of first-line sacituzumab govitecan could be found and that two appraisals for triple-negative disease were awaiting development. NICE lists four: the two for datopotamab deruxtecan, sacituzumab govitecan for untreated PD-L1-negative disease (ID6433, selected in December 2023) and adjuvant pembrolizumab after neoadjuvant chemotherapy (ID6546) (Second accuracy pass; the NICE site search and the project pages GID-TA11489, GID-TA11724, GID-TA11535 and GID-TA11835 read as raw HTML) · The funding note, the sacituzumab row and the gaps list now name all four topics, each awaiting development with no expected publication date, and say that no topic for sacituzumab govitecan with pembrolizumab was found
- 2026-11-17 · Triple-negative breast cancer in the UK: Cancer Drugs Fund status: The page said the Cancer Drugs Fund list could not be read and treated every recommended medicine as routine commissioning. The list is readable, and it shows trastuzumab deruxtecan for HER2-low disease with interim Cancer Drugs Fund funding from 17 September 2026 and baseline commissioning expected on 5 January 2027. The timeline also still said a HER2-low result opens a medicine only in Scotland (Second accuracy pass; the National Cancer Drugs Fund list version 1.41 (12 November 2026), located through the NHS England publication API and read as PDF text, and NICE TA1197 read as raw HTML) · The funding note and the HER2-low row give the interim funding and its dates and the list's limit of two prior lines of chemotherapy for metastatic disease. The other six appraisals are confirmed in the list's baseline-funded section. The timeline names TA1197 for England, Wales and Northern Ireland
- 2026-11-17 · Triple-negative breast cancer in the UK: waiting-time targets in England: The Faster Diagnosis Standard was given as 75 percent rising to 80 percent by March 2026, and the 62-day planning target as 75 percent by March 2026. Both dates have passed. The planning framework for 2026/27 holds the Faster Diagnosis Standard at a new threshold of 80 percent and asks every trust to reach 94 percent for 31 days and 80 percent for 62 days by March 2027 (Second accuracy pass; NHS England's standards letter, 2025/26 planning guidance and Medium Term Planning Framework 2026/27 to 2028/29 read through the NHS England content API) · The timeline targets and the four-nations row carry the 2026/27 figures and the framework is cited
- 2026-11-17 · Triple-negative breast cancer in the UK: genomic test criteria: The R208 criteria listed bilateral breast cancer with both cancers under 50, and R216 listed triple-positive breast cancer at 35 or under. The Test Directory says bilateral breast cancer under 60 and HER2-positive breast cancer at 35 or under. The page also said the directory could not be read (Second accuracy pass; the rare and inherited disease eligibility criteria version 9.2 (9 November 2026) read as PDF text and the cancer test directory version 16.0 read from its spreadsheet, both located through the NHS England publication API) · Both criteria follow the directory, with a note that the GeNotes summary of February 2024 gives the older wording. The M234 entry quotes the directory's own condition, and the notes give the directory versions read
- 2026-11-17 · Triple-negative breast cancer in the UK: platinum chemotherapy row: The row for platinum-containing chemotherapy before surgery carried the reference TA851, so the page showed a NICE TA851 label on a recommendation that comes from guideline NG101 (Second accuracy pass; NICE NG101 recommendation 1.8.1 and TA851 read as raw HTML) · The reference is NG101, and the note says NICE's survival figures are for a minimum follow-up of three years
- 2026-11-17 · Triple-negative breast cancer in the UK: trial sites: IZABRIGHT-Breast01 was shown as recruiting in the UK, but no UK site is listed as recruiting. TROPION-Breast05 listed Liverpool, Oxford and Surrey as open and named hospitals the registry does not name. The Liverpool site is terminated, Oxford and Surrey are suspended and one London site is withdrawn. TroFuse-012 lacked its Cambridge site, and TROPION-Breast02 was said to have finished recruiting in 2024, a date the registry does not give (Second accuracy pass; the ClinicalTrials.gov records NCT06926868, NCT06103864, NCT06393374, NCT05374512, NCT03740893 and NCT05633654 read through the registry API) · Statuses are dated 17 November 2026, the site lists and seven centre rows follow the registry, and sites not yet recruiting are noted for PHOENIX and ASCENT-05
- 2026-11-17 · Ellen Copson: The summary said she returned to academic medicine in 2012 and had been a consultant since 2009. Her hospital page gives neither statement in those words (Second accuracy pass; the University Hospital Southampton consultant page read as raw HTML) · The summary now says she began registrar training in Southampton in 2001 and has worked at the trust since June 2009, and the 2012 clause is removed
- 2026-11-17 · Triple-negative breast cancer in the UK: Northern Ireland figures: The Northern Ireland case counts, stage shares and five-year net survival of 83.2 percent could not be read again, and the survival figure does not state the diagnosis years it applies to (Second accuracy pass; the Northern Ireland Cancer Registry page and data tables answered 403 and the Internet Archive held no copy) · The two figures and the gaps list now say the tables could not be re-read on 17 November 2026 and that the survival cohort is not recorded
- 2026-11-17 · Triple-negative breast cancer: chemotherapy before surgery: The decision row and its question said NICE quotes 85 rather than 81 out of 100 people alive at 3 years when the chemotherapy before surgery contains a platinum. NICE gives those figures for studies with a minimum follow-up of 3 years, so they are not a 3-year survival rate (Second accuracy pass; NICE NG101 recommendation 1.8.1 and its table 3 read as raw HTML from nice.org.uk) · Both places now say 85 rather than 81 out of 100 surviving in studies with a minimum follow-up of 3 years
- 2026-11-17 · Triple-negative breast cancer: fertility preservation: The guideline line cited NICE CG156 as the fertility guideline. CG156 was replaced by NICE NG257, published 31 March 2026 (Second accuracy pass; the NICE pages for CG156 and NG257 read as raw HTML) · The line now cites NICE NG257 and says it replaced CG156
- 2026-11-17 · Triple-negative breast cancer: first treatment for metastatic disease: A question said NICE TA801 funds pembrolizumab with chemotherapy only where the combined positive score is 10 or more. TA801 sets two tumour conditions: a combined positive score of 10 or more and immune cell staining of less than 1 percent, with paclitaxel or nab-paclitaxel (Second accuracy pass; NICE TA801 recommendation 1.1 read as raw HTML from nice.org.uk) · The question now gives both conditions and names the chemotherapy
- 2026-11-17 · Triple-negative breast cancer: radiotherapy schedule: A question said NICE says offer 40 Gy in 15 fractions after implant reconstruction. NICE says offer it when the lymph nodes are treated and consider it after implant-based reconstruction (Second accuracy pass; NICE NG101 recommendations 1.13.13, 1.13.14 and 1.13.16 read as raw HTML) · The question now separates offer from consider
- 2026-11-17 · Triple-negative breast cancer: adding pembrolizumab before surgery: The decision row and a question said the KEYNOTE-522 gains were seen whatever the PD-L1 result. Neither published abstract gives a result by PD-L1 and the full text could not be read. What could be read is that the trial enrolled people whatever their PD-L1 result and that neither the NICE recommendation nor the licence depends on a score (Second accuracy pass; the abstracts for pmid 32101663 and pmid 39282906 read through PubMed and Europe PMC, NICE TA851, and the current FDA label for pembrolizumab read through openFDA) · Both places now say the trial enrolled people whatever their PD-L1 result and that the gains were reported for the whole trial population
- 2026-11-17 · Triple-negative breast cancer: questions when newly diagnosed: A question credited the NHS with saying you can ask your GP or consultant to refer you for a second opinion. None of the NHS pages the record cites says this (Second accuracy pass; the cited NHS pages read as raw HTML) · The NHS clause was removed; the Triple Negative Breast Cancer Foundation advice stays
- 2026-11-17 · Adjuvant Capecitabine for Breast Cancer after Preoperative Chemotherapy: The registry link for CREATE-X opened a different study, a phase 2 trial of irinotecan and S-1 registered as UMIN000000855. The summary also gave the treatment as six months of capecitabine, a length neither the abstract nor the registry record states (Second accuracy pass; the UMIN-CTR pages for both receipt numbers and the WHO ICTRP record for UMIN000000843 read as raw text, and the abstract for PubMed 28564564 read through PubMed) · Link now opens the CREATE-X record (receipt number R000000989). The treatment length is removed here and in the roadmap step
- 2026-11-17 · Triple-negative breast cancer roadmap: from a remainder defined by three negative tests to immunotherapy, antibody-drug conjugates and the residual disease problem: The Carolina Breast Cancer Study figure was given as 16 percent of others. The platinum step said survival followed pathological complete response without saying that the CALGB 40603 long-term report found no significant gain from carboplatin. The c-TRAK TN figure of 72 percent with metastases was applied to all ctDNA-positive women. The 2000 gene expression study was said to cover 65 tumours. TROPION-Breast03 primary completion was 20 September 2027. No trial alternating or sequencing the conjugates and no later ctDNA-guided trial was named. The izalontamab brengitecan phase 3 was given as positive without saying it is unpublished (Second accuracy pass; the abstracts for PubMed 16757721, 35044810, 42992319, 36423745 and 10963602 read through PubMed, and ClinicalTrials.gov records NCT05629585, NCT07151586, NCT06992336, NCT04915755 and NCT06382142 read through the registry API) · Gives 14 percent in postmenopausal African American women and 16 percent in non-African American women. Adds the CALGB 40603 long-term result and the 2026 pooled analysis (hazard ratio 0.71). States 72 percent of the 32 women allocated to staging. Says 65 specimens from 42 people. Date changed to 1 September 2027. Names the ALTER trial, the Chinese ctDNA trial and ZEST. Labels the izalontamab result as a sponsor announcement with no registry results or indexed report
- 2026-11-17 · Triple-negative breast cancer: The history gave the Carolina figure as 16 percent in other women and called the group young rather than premenopausal. It applied the c-TRAK TN metastasis figure to all ctDNA-positive women, and an open problem said ctDNA misses brain-only relapse, which the cited abstract does not say (Second accuracy pass; the abstracts for PubMed 16757721 and 36423745 read through PubMed) · History now gives 14 percent in postmenopausal African American women and 16 percent in non-African American women, and 72 percent of those sent for staging scans (23 of 32). The open problem now says seven women relapsed with no ctDNA detected first
- 2026-11-17 · Long-term efficacy and safety of addition of carboplatin with or without veliparib to standard neoadjuvant chemotherapy in triple-negative breast cancer: 4-year follow-up data from BrighTNess, a randomized phase III trial: Called the third trial to show carboplatin benefit lasting beyond pathological complete response. CALGB 40603, the third trial, found no significant long-term gain (Second accuracy pass; the abstracts for PubMed 35044810 and 42992319 read through PubMed) · Says it is one of two trials with GeparSixto, states the CALGB 40603 result and the pooled hazard ratio of 0.71, and links both papers. The CALGB 40603 paper record gains the same caveat
- 2026-11-17 · Survival analysis of carboplatin added to an anthracycline/taxane-based neoadjuvant chemotherapy and HRD score as predictor of response-final results from GeparSixto: Said disease-free survival improved with carboplatin in both repair-deficient and non-deficient tumours, quoting hazard ratios of 0.49 and 0.44 without saying neither was statistically significant (Second accuracy pass; the abstract for DOI 10.1093/annonc/mdy460 read through PubMed) · Adds the confidence intervals (0.23 to 1.04 and 0.17 to 1.17), the p values and that neither is significant on its own
- 2026-11-17 · Tumor-Infiltrating Lymphocytes in Triple-Negative Breast Cancer: Said a prospective trial of chemotherapy omission was the missing step, and the de-escalation idea said none had started. Two are recruiting (Second accuracy pass; ClinicalTrials.gov records NCT06078384 (ETNA) and NCT06476119 (OPTImaL) read through the registry API) · Both the paper record and the idea now describe ETNA and OPTImaL with their sizes and completion dates and link the registry records
- 2026-11-17 · PD-L1 Immunohistochemistry Assay Comparison in Atezolizumab Plus nab-Paclitaxel-Treated Advanced Triple-Negative Breast Cancer: Said the benefit of atezolizumab was concentrated in tumours all three tests called positive. The paper reports double-positive cases, positive on SP142 and one other assay (Second accuracy pass; the abstract for DOI 10.1093/jnci/djab108 read through PubMed) · Reworded to tumours that SP142 also called positive, here and in the PD-L1 harmonisation idea
- 2026-11-17 · Association of Circulating Tumor DNA and Circulating Tumor Cells After Neoadjuvant Chemotherapy With Disease Recurrence in Patients With Triple-Negative Breast Cancer: Preplanned Secondary Analysis of the BRE12-158 Randomized Clinical Trial: Said no completed trial had acted on the result, although c-TRAK TN had done so and reported in 2023. The summary line gave the result as if all 196 women were analysed for ctDNA (Second accuracy pass; the abstracts for PubMed 32644110 and 36423745 read through PubMed and ClinicalTrials.gov NCT04915755 read through the registry API) · Names c-TRAK TN and the ZEST trial, and says 142 of the 196 were analysed. The c-TRAK TN paper record now says nearly three quarters of those scanned, not of all who tested positive, had metastases
- 2026-11-17 · ctDNA-guided adjuvant decisions after residual disease: escalate the positive, spare the negative: Credited the ten-year relapse-free survival of 81 percent for the lowest residual cancer burden class to the 2022 pooled analysis. It is the triple-negative figure from the 2017 single-institution series (Second accuracy pass; the abstracts for PubMed 28135148 and 34902335 read through PubMed) · Attribution corrected to the 2017 MD Anderson series
- 2026-11-17 · A UK audit of trial access and germline testing uptake in triple-negative breast cancer: Said a triple-negative diagnosis predicts a BRCA1 variant in 57 percent of carriers, which reverses the finding (Second accuracy pass; the abstract for PubMed 18779615 read through PubMed) · Now says 57 percent of BRCA1 carriers had triple-negative cancers against 14 percent of non-carriers
- 2026-11-17 · Ethnicity-stratified outcome reporting for triple-negative breast cancer in NHS cancer statistics: Said Black ethnicity was an independent risk factor for distant relapse in the POSH cohort. The multivariate result was in oestrogen receptor-positive patients (Second accuracy pass; the abstract for DOI 10.1038/bjc.2013.650 read through PubMed) · States the subgroup and the hazard ratio of 1.60
- 2026-11-17 · Race, breast cancer subtypes, and survival in the Carolina Breast Cancer Study: The short summary gave the basal-like subtype as 39 percent in premenopausal African American women against 16 percent in other women. The paper gives 14 percent in postmenopausal African American women and 16 percent in non-African American women of any age (Second accuracy pass; the abstract for DOI 10.1001/jama.295.21.2492 read through PubMed) · The short summary now gives all three figures
- 2026-11-17 · Triple-negative breast cancer: clinical features and patterns of recurrence: The record said women who reach five years without relapse are largely safe. The paper says only that the higher risk compared with other breast cancers was seen within five years of diagnosis and not thereafter. The diagnosis years were missing from the short summary (Second accuracy pass; the abstract for DOI 10.1158/1078-0432.CCR-06-3045 read through PubMed) · Says that after five years the risk was no higher than for other breast cancers, and adds the diagnosis years 1987 to 1997
- 2026-11-17 · Neoadjuvant Chemotherapy, Endocrine Therapy, and Targeted Therapy for Breast Cancer: ASCO Guideline: The record said the systematic review closed before the KEYNOTE-522 event-free survival data and that the immunotherapy line was reversed within a year. The review included the second interim analysis of KEYNOTE-522 with preliminary event-free survival, and the two documents were published online 15 months apart (Second accuracy pass; the full text read on the PubMed Central page and the publication dates read from the PubMed records) · States what the review included and gives 15 months, here and in the rapid update record
- 2026-11-17 · Biomarkers for Systemic Therapy in Metastatic Breast Cancer: ASCO Guideline Update: The short summary said there is no routine test for tumour DNA in blood, and the record called PD-L1 the one selection assay. The guideline recommends tumour DNA in blood for PIK3CA testing and advises only against its routine use to monitor response, and it recommends germline BRCA testing to select PARP inhibitors (Second accuracy pass; the abstract for DOI 10.1200/JCO.22.01063 read through PubMed) · Both statements reworded to what the guideline says
- 2026-11-17 · Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications: The record said the basal-like group did worst. The paper reports the shortest survival in both the basal-like and the ERBB2-overexpressing groups (Second accuracy pass; the full text for DOI 10.1073/pnas.191367098 read on the PubMed Central page) · Names both groups in the short summary, summary and findings
- 2026-11-17 · Molecular portraits of human breast tumours: The record counted 65 breast tumours. The paper studied 65 specimens from 42 people, with 20 tumours sampled twice. A caveat gave an unsourced figure for the overlap between basal-like and triple-negative tumours (Second accuracy pass; the abstract for DOI 10.1038/35021093 read through PubMed) · Now reads 65 tumour samples from 42 people, and the unsourced figure is removed
- 2026-11-17 · Repeated observation of breast tumor subtypes in independent gene expression data sets: A finding said the subgroups were reproduced in two independent data sets, where the paper says some of the same subtypes were found. The record also said every triple-negative patient under 60 is now offered germline testing without saying this is the UK criterion (Second accuracy pass; the abstract and full text for DOI 10.1073/pnas.0932692100 and the abstract for DOI 10.1016/j.gimo.2023.100849 read through PubMed and PubMed Central) · Finding reworded, the 18 BRCA1 carriers counted, and the testing rule attributed to UK mainstream criteria with its source linked
- 2026-11-17 · Germline BRCA1 mutations and a basal epithelial phenotype in breast cancer: The short summary said BRCA1 cancers are nine times more likely to have the basal pattern. The figure is an odds ratio and applies only among tumours negative for the oestrogen receptor and HER2 (Second accuracy pass; the abstract for DOI 10.1093/jnci/djg050 read through PubMed) · Now reads nine times the odds among those tumours
- 2026-11-17 · US incidence of breast cancer subtypes defined by joint hormone receptor and HER2 status: The short summary called this the first national count and gave 12.2 percent of cases as triple-negative. The registries cover 28 percent of the US population and the 12.2 percent is of cases with known receptor status (Second accuracy pass; the abstract for DOI 10.1093/jnci/dju055 read through PubMed) · Short summary reworded with the coverage and the denominator
- 2026-11-17 · Update on triple-negative breast cancer disparities for the United States: A population-based study from the United States Cancer Statistics database, 2010 through 2014: The record presented odds of a triple-negative diagnosis among breast cancer cases as incidence, gave no comparator groups, and explained the 8.4 percent share by unknown-status coding, which the authors do not say (Second accuracy pass; the abstract and full text for DOI 10.1002/cncr.32207 read through PubMed and NCBI PubMed Central) · Names the comparators, says the figures are unadjusted odds, gives the authors' own explanation and adds that 39 states were covered
- 2026-11-17 · Sites of distant recurrence and clinical outcomes in patients with metastatic triple-negative breast cancer: high incidence of central nervous system metastases: The record said first-line trials now report median survival near two years. In KEYNOTE-355 that holds only for tumours with a PD-L1 combined positive score of 10 or more (23.0 months); in all patients the median was 17.2 months (Second accuracy pass; the abstract for DOI 10.1056/NEJMoa2202809 read through PubMed) · Gives both figures and links the paper
- 2026-11-17 · Refinement of Triple-Negative Breast Cancer Molecular Subtypes: Implications for Neoadjuvant Chemotherapy Selection: The record called this the first evidence that molecular subtype predicts chemotherapy response. The paper itself credits an earlier report by Masuda and colleagues. The short summary called complete response simply response (Second accuracy pass; the abstract and full text for DOI 10.1371/journal.pone.0157368 read through PubMed and NCBI PubMed Central) · Reworded as pooled evidence from 306 patients building on the earlier report, and the short summary says complete response
- 2026-11-17 · Comprehensive genomic analysis identifies novel subtypes and targets of triple-negative breast cancer: The record said the study arrived at the same partition as the refined Lehmann scheme, and that B7-H4 antibody-drug conjugates are only now in early trials. The paper says its clustering did not separate the two Lehmann basal-like groups, and a phase 3 trial of a B7-H4 conjugate has been recruiting since 2025 (Second accuracy pass; the full text for DOI 10.1158/1078-0432.CCR-14-0432 read through NCBI PubMed Central and NCT07044336 read on ClinicalTrials.gov) · Says which groups the two schemes share, and gives the phase 3 trial, which is in endometrial cancer
- 2026-11-17 · Management of Hereditary Breast Cancer: American Society of Clinical Oncology, American Society for Radiation Oncology, and Society of Surgical Oncology Guideline: The record gave the share of triple-negative patients with a germline BRCA variant as 11 to 17 percent, a figure the guideline abstract does not contain and for which no source was cited (Second accuracy pass; the abstract for DOI 10.1200/JCO.20.00299 read through PubMed) · Figure removed
- 2026-11-17 · Breast Cancer, Version 4.2026, NCCN Clinical Practice Guidelines In Oncology: The record called NCCN the first major guideline to place an antibody-drug conjugate first line, which the abstract does not say, and quoted category grades from the full guideline, which needs registration and could not be re-read (Second accuracy pass; the abstract for DOI 10.6004/jnccn.2026.0033 read through PubMed) · The priority claim is removed and the grades are labelled as not re-read. The three ESMO guideline records carry the same label
- 2026-11-17 · Circulating Tumor DNA Guided Boost Therapy in Early Triple Negative Breast Patients With Residual Disease After Neoadjuvant Therapy: The record said the trial tests anlotinib, benmelstobart and capecitabine in 411 participants. The registry entry, updated on 2026-10-23, now records the interventions as "Apatinib, Camrelizumab" and "Capecitabine", an anticipated enrolment of 430, and a new official title that no longer names anlotinib (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT06992336 read as raw JSON and compared field by field by script, then the interventions and arms read by hand) · The one-line summary, the interventions sentence and the linked drugs now name apatinib, camrelizumab and capecitabine, the official title is the registry's current one, and enrolment is set to 430
- 2026-11-17 · Triple-negative breast cancer: registry trials whose recruitment status changed: Eight records carried a recruitment status the registry has since changed: NCT04159818 (was recruiting, now active and not recruiting, entry updated 2026-11-12), NCT05866432 (was recruiting, now active and not recruiting, entry updated 2026-10-30), NCT06787339 (was planned, now recruiting, entry updated 2026-09-28), NCT07015853 (was planned, now recruiting, entry updated 2026-11-16), NCT07299409 (was planned, now recruiting, entry updated 2026-10-06), NCT07372079 (was planned, now recruiting, entry updated 2026-09-28), NCT07698106 (was planned, now recruiting, entry updated 2026-10-29), NCT07729956 (was planned, now recruiting, entry updated 2026-09-30) (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · Status set to recruiting for the six that have opened and to active for the two that have closed to recruitment; each one-line summary now says "now recruiting" or "active but no longer recruiting" to match
- 2026-11-17 · Triple-negative breast cancer: registry trials with an unknown status: Five records said the trial was now recruiting or not yet recruiting where the registry now gives the overall status as unknown, because the entry has not been verified for two years or more: NCT04481763 (last known status recruiting, last verified 2020-07), NCT05019690 (last known status recruiting, last verified 2024-09), NCT06125080 (last known status recruiting, last verified 2023-11), NCT06268327 (last known status not yet recruiting, last verified 2024-01), NCT06586866 (last known status not yet recruiting, last verified 2024-09) (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · Each one-line summary and summary now says what the trial was last listed as, that the registry gives its status as unknown, and when the entry was last verified; the status field is left as it was because the schema has no value for unknown
- 2026-11-17 · Triple-negative breast cancer: three registry trials enrolling by invitation: Three records said "now recruiting" where the registry status is enrolling by invitation, which a reader cannot simply join: NCT06910072, NCT07359404, NCT07376629 (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · Each one-line summary now says "now enrolling by invitation only"
- 2026-11-17 · Triple-negative breast cancer: registry trial dates moved by their sponsors: Eighteen records carried a start or primary completion date the registry has since changed: NCT03168880 (primary completion 2025-11-30 now 2026-11-30); NCT04159818 (primary completion 2022-12-15 now 2030-12-15); NCT05451784 (primary completion 2025-11-30 now 2028-09-30); NCT05528133 (primary completion 2028-05 now 2029-09); NCT05866432 (primary completion 2026-05-23 now 2026-07-01); NCT05999149 (primary completion 2025-08-30 now 2026-05-30); NCT06787339 (start 2025-01-15 now 2025-01-20); NCT06793332 (start 2024-12-17 now 2025-02-05, primary completion 2026-01-31 now 2026-12-31); NCT06802757 (start 2025-05-01 now 2025-04-30, primary completion 2026-06-30 now 2026-12-31); NCT07015853 (start 2026-06 now 2026-12, primary completion 2033-07 now 2033-12); NCT07299409 (start 2025-12 now 2026-01-20); NCT07372079 (start 2026-04-05 now 2026-04-14); NCT07419880 (start 2026-01-30 now 2027-01-01, primary completion 2027-01-30 now 2027-12-31); NCT07698106 (start 2026-08 now 2026-10-20); NCT07721259 (start 2026-09-30 now 2026-11-01); NCT07723469 (start 2026-10 now 2026-12); NCT07729956 (start 2026-08 now 2026-09-02); NCT07743190 (start 2026-10-01 now 2027-01-01, primary completion 2029-03-01 now 2029-09-22) (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · The dates in each summary now match the registry, with "started" kept only where the registry marks the start date as actual
- 2026-11-17 · Triple-negative breast cancer: registry trials past their primary completion: Five records said the trial was "due to reach its primary completion" on a date the registry records as an actual, past date: NCT02488967, NCT02641847, NCT04296175, NCT05866432, NCT05999149 (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and the date type compared by script) · Each summary now says the trial "reached its primary completion" on that date. None of the five has results posted on the registry
- 2026-11-17 · Triple-negative breast cancer: registry trials that have not started: Seventy-four records said the trial "started" on a date the registry gives only as an anticipated start, some of them in the future: NCT02802423, NCT04137653, NCT04301739, NCT05067530, NCT05128734, NCT05522491, NCT05806385, NCT05821686, NCT05862610, NCT06125080, NCT06140576, NCT06202313, NCT06229067, NCT06268327, NCT06313463, NCT06371807, NCT06401005, NCT06404736, NCT06407310, NCT06586866, NCT06632405, NCT06691594, NCT06724263, NCT06731153, NCT06746870, NCT06795503, NCT06819319, NCT06849492, NCT06851299, NCT06977542, NCT07005583, NCT07007780, NCT07011823, NCT07015853, NCT07021261, NCT07040644, NCT07054242, NCT07101614, NCT07103447, NCT07104266, NCT07107217, NCT07139470, NCT07151586, NCT07153965, NCT07161791, NCT07244874, NCT07255612, NCT07256964, NCT07271992, NCT07292142, NCT07407517, NCT07413601, NCT07419880, NCT07441512, NCT07458113, NCT07458204, NCT07486687, NCT07487519, NCT07509515, NCT07528768, NCT07552181, NCT07556848, NCT07581314, NCT07610525, NCT07681297, NCT07721259, NCT07723469, NCT07730125, NCT07743190, NCT07743632, NCT07743723, NCT07752550, NCT07762703, NCT07786428 (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and the date type compared by script) · Each summary now says "planned to start" with the registry date
- 2026-11-17 · Neoadjuvant Therapy for Early Triple-Negative Breast Cancer: A Response-Guided Approach Using Iparomlimab and Tuvonralimab Injection in Combination With Chemotherapy: The summary quoted an eligibility line that is no longer in the registry entry word for word; the sponsor has reworded the criterion (Second accuracy pass; the eligibility criteria in the ClinicalTrials.gov v2 API record for NCT07372079 read as raw JSON and searched for the quoted text by script) · The quotation now gives the registry's current words: "* According to the latest ASCO/CAP guideline definitions, histopathologically confirmed early-stage triple-negative invasive breast carcinoma meeting all of the following criteria"
- 2026-11-17 · BL-B01D1-307: The page, the cancer summary, the ASCO 2026 digest and the antibody-drug conjugate roadmap called this the first phase 3 win for a bispecific antibody-drug conjugate. The report they cite calls it the third phase 3 trial in which the drug reached its primary endpoints. The digest also called an interim analysis full data, and the trial was described as sponsored by SystImmune and Bristol Myers Squibb when the registry names Sichuan Baili Pharmaceutical. (Second accuracy pass; the OncLive report of the 26 February 2026 announcement read as raw HTML, ClinicalTrials.gov NCT06382142, and the publisher abstract for DOI 10.1200/JCO.2026.44.17_suppl.LBA1003 read through Crossref.) · The first-win claim removed from all four places. The trial page now gives the interim figures from the congress abstract (median progression-free survival 8.5 against 3.1 months, overall survival 15.9 against 12.5 months), links it, names the registry sponsor, and says no full paper or registry results existed on 17 November 2026.
- 2026-11-17 · Triple-negative breast cancer: The summary said OlympiA added olaparib for BRCA carriers with residual disease, and the residual-disease row gave it NCCN Category 1. OlympiA enrolled high-risk early breast cancer after neoadjuvant or adjuvant chemotherapy, and the review the site cites tabulates the NCCN rating as Category 2A, preferred. (Second accuracy pass; the abstract for DOI 10.1056/NEJMoa2105215 read through PubMed, and the full text of PMC13114725 read through NCBI.) · Summary reworded to high-risk early disease after chemotherapy; the row now says Category 2A, preferred, and that the NCCN text itself was not re-read.
- 2026-11-17 · PD-L1: The share of metastatic triple-negative tumours with a combined positive score of 10 or more was given as 35 to 40% from KEYNOTE-355 screening, sourced to Wikipedia. The US prescribing information gives 38% of the 847 randomised patients. (Second accuracy pass; the pembrolizumab label (effective 16 October 2026) read through the openFDA label API.) · Figure changed to 38%, described as a share of randomised patients, and sourced to the label.
- 2026-11-17 · TROPION-Breast03: The trial page and the calendar gave the estimated primary completion as 20 September 2027. The registry gives 1 September 2027. (Second accuracy pass; ClinicalTrials.gov NCT05629585 read through the v2 API.) · Date corrected in the summary, the notes and the calendar entry.
- 2026-11-17 · Datopotamab deruxtecan: The UK coverage row and the regional approvals table described the EU authorisation as covering hormone receptor-positive breast cancer only. A first-line triple-negative indication was added in 2026. (Second accuracy pass; the EMA Datroway overview and the opinion on variation page read as raw HTML.) · Both now record the positive opinion of 23 June 2026 for first-line triple-negative breast cancer in people who are not candidates for PD-1 or PD-L1 inhibitors.
- 2026-11-17 · Sacituzumab govitecan: The regional approvals table listed only the 2021 and 2023 EU indications. (Second accuracy pass; the EMA procedural-steps document for Trodelvy read as text.) · The two first-line triple-negative uses added, with Commission decisions of 22 June 2026 (alone) and 21 August 2026 (with pembrolizumab).
- 2026-11-17 · Atezolizumab: The IMpassion130 note and the biomarker table said the US triple-negative approval was withdrawn in August 2021. That is the month the company announced it; the FDA dates the withdrawal 6 October 2021. (Second accuracy pass; the FDA list of withdrawn cancer accelerated approvals read as raw HTML.) · Both now give the announcement month and the FDA date.
- 2026-11-17 · Ipatasertib: The note said the BARBICAN trial tests ipatasertib before surgery and gave no result, although a result was reported in 2021. (Second accuracy pass; the full text of DOI 10.3390/jpm13030526 (PMC10055616) read through NCBI, and ClinicalTrials.gov NCT05498896.) · Note now gives the reported rates, 49.3% with ipatasertib against 48.5% without, as second-hand from a review, and says no full paper has followed.
- 2026-11-17 · The risk of cancer associated with specific mutations of BRCA1 and BRCA2 among Ashkenazi Jews: The paper was linked to Mary-Claire King, who is not one of its authors. (Second accuracy pass; the author list for PubMed 9145676 read through NCBI.) · Link removed.
- 2026-11-17 · How NHS cancer care works: waiting-time standards: The coverage text gave the Faster Diagnosis Standard target as 75%, rising to 80% by March 2026. (Second accuracy pass; NHS England's Medium Term Planning Framework 2026/27 to 2028/29 read through the site's JSON feed.) · Now gives 80%, the threshold set from 2026/27.
- 2026-11-17 · BL-B01D1-307: The plain-language line and its eight translations called the result the first phase 3 win for a bispecific antibody-drug conjugate. The report cited for it calls it the third phase 3 study in which the drug reached its primary end points (Second accuracy pass; the OncLive report of the 26 February 2026 news release read in raw text) · The word first is removed from the plain-language line and from each translation
- 2026-11-17 · Gallbladder cancer: adjuvant treatment: The standard-of-care row said ACTICCA-1 "is awaited". The trial was published on 24 October 2026: gemcitabine and cisplatin did not significantly improve disease-free survival over capecitabine (median 25.7 against 17.9 months, log-rank P = 0.097) in 591 patients, 16 percent of them with gallbladder cancer. The 2019 history entry also called BILCAP the adjuvant standard without saying it missed its primary endpoint (Second accuracy pass; the abstract for pmid 42926109 (Annals of Oncology 2026) and the BILCAP abstract for DOI 10.1016/S1470-2045(18)30915-X read through Europe PMC, and the registry record NCT02170090) · The row now gives the ACTICCA-1 result and says capecitabine stays the standard; the history entry gives the BILCAP intention-to-treat hazard ratio 0.81 (0.63 to 1.04) and says it was not statistically significant
- 2026-11-17 · Gallbladder cancer: survival figures: The one-year survival figure for England was dated to people diagnosed 2016 to 2020; it is for people diagnosed 2018 to 2022. The US figures by stage were given for people diagnosed 2013 to 2019 as "almost 30 percent" regional and "almost 5 percent" distant; Cancer Research UK now quotes people diagnosed 2016 to 2022, with 30 percent and around 5 percent (Second accuracy pass; Cancer Research UK's survival page for gallbladder cancer read as raw HTML) · Each figure now carries its own diagnosis years and the current SEER values; the state-of-the-art line says around 5 percent rather than under 5 percent
- 2026-11-17 · Gallbladder cancer: T2 tumour location in the timeline: The 2015 history entry gave five-year survival as "42.6 versus 64.7 percent" for peritoneal-side against hepatic-side tumours, which reverses the two. In the source 42.6 percent is the hepatic side and 64.7 percent the peritoneal side (Second accuracy pass; the abstract for DOI 10.1097/sla.0000000000000728 read through Europe PMC) · The entry now reads 64.7 percent on the peritoneal side against 42.6 percent on the hepatic side
- 2026-11-17 · T2a versus T2b gallbladder cancer: The page said a Korean validation study found T1b and T2a, and T2b and T3, hard to separate. The study found the opposite: those pairs differed significantly, and it was T2a and T2b that showed no significant survival difference (Second accuracy pass; the abstract for DOI 10.4143/crt.2019.271 and the full text PMC7176947 read through Europe PMC) · Rewritten to say there was no significant difference between T2a and T2b, although T1b differed from T2a and T2b from T3
- 2026-11-17 · Gallbladder cancer: first-line treatment and NICE: The page said pembrolizumab with chemotherapy was "not appraised by NICE". NICE opened an appraisal, TA966, and terminated it without a recommendation because the company made no submission (Second accuracy pass; the NICE search results page read as raw HTML) · The row now names TA966 and says why NICE made no recommendation
- 2026-11-17 · Radical (extended) cholecystectomy: The page said patients with "N2 (four or more nodes or aortocaval) involvement" do not benefit, crediting the 2015 AHPBA consensus. The consensus met in January 2014 and speaks of N2 nodal stations, a location in the staging then in use, not the later count of four or more nodes (Second accuracy pass; the abstract for DOI 10.1111/hpb.12444 read through Europe PMC, with the 7th edition node classes checked in the full text of DOI 10.1186/1477-7819-10-87) · Reworded to N2 nodal stations with the date of the consensus; the staging list on the gallbladder cancer page was aligned
- 2026-11-17 · Adenosquamous and squamous carcinoma of the gallbladder: A SEER study was said to show surgery, node dissection, radiotherapy and chemotherapy each linked to better cancer-specific survival. That was the result for advanced-stage disease; in early-stage disease only radical node dissection helped. A molecular cohort was also called European where its authors say Caucasian (Second accuracy pass; the abstracts for DOI 10.21037/jgo-22-1292 and DOI 10.1111/his.70179 read through Europe PMC) · The sentence now separates early from advanced disease, and the cohort is described in its authors' word
- 2026-11-17 · Neuroendocrine carcinoma of the gallbladder: A population study was said to show surgery and adjuvant chemotherapy each linked to better survival; it reports better five-year outcomes for surgery, radiation and chemotherapy combined, and better early survival with surgery alone. Cancer Research UK was credited with saying these tumours are not necessarily treated like adenocarcinoma, which it says of lymphoma and melanoma. The summary also called its sources the largest series (Second accuracy pass; the abstract and full text for DOI 10.3390/jpm13061009 read through Europe PMC, and Cancer Research UK's types page read as raw HTML) · Treatment sentences reworded to the paper's finding, the Cancer Research UK clause removed, and "largest series" replaced by "the two series cited here"
- 2026-11-17 · Cystic duct margin: The page said frozen section of the bile duct margin "generally agrees with the final result". In the series cited the final diagnosis differed in 5 of 20 patients, and its authors conclude the two may be inconsistent in 25 percent (Second accuracy pass; the abstract for DOI 10.1080/13651820510028873 read through Europe PMC) · The sentence now gives 5 of 20
- 2026-11-17 · Incidental gallbladder cancer: The Danish incidence of 0.08 percent was given for patients "under 60"; the study's group is 60 or under. An open problem said two analyses disagree on disease-specific benefit of re-resection for T1b, but one of them reports overall survival only (Second accuracy pass; the abstracts for DOI 10.61409/a11240787, DOI 10.14701/ahbps.26-091 and DOI 10.1016/j.hpb.2024.01.012 read through Europe PMC) · Age group corrected here and on the gallbladder cancer page; the open problem now states what each analysis found
- 2026-11-17 · Gallbladder cancer: open problems: The list said two meta-analyses suggest peritoneal-side T2 tumours may not need liver resection, which neither cited abstract says; called a 237-patient series the largest; said UK survival is not reported outside Northern Ireland, when England's is; and said incidence is falling in the United States, where it fell and then levelled off after 2002 (Second accuracy pass; the abstracts for DOI 10.1016/j.hpb.2021.12.019, DOI 10.3389/fonc.2023.1186378, DOI 10.1002/jhbp.593 and DOI 10.1097/coc.0000000000000918 read through Europe PMC, and Cancer Research UK's statistics pages read as raw HTML) · Each item reworded to what its source states; the T1b series now carries its DOI, and the complication excess with segmental resection is described as borderline (confidence interval 1.00 to 3.60)
- 2026-11-17 · Port-site metastasis: The fall in port-site metastasis was credited to retrieval bags, avoiding perforation and fewer extractions through the umbilicus. The abstract gives no reasons and the full text is not open (Second accuracy pass; the abstract for DOI 10.1016/j.surg.2016.08.007 read through Europe PMC) · Replaced with the abstract's own finding that the extraction site is at significantly higher risk than the other port sites
- 2026-11-17 · Carcinoma in situ and dysplasia of the gallbladder: The summary line said Cancer Research UK notes some doctors do not regard stage 0 as a true cancer. Its page says stage 0 is not seen as an invasive cancer. The primary sclerosing cholangitis figures were also given per 100 explants where the study examined 72 gallbladders (Second accuracy pass; Cancer Research UK's stages and grades page read as raw HTML and the abstract for DOI 10.1097/01.pas.0000213435.99492.8a read through Europe PMC) · Both reworded to the source
- 2026-11-17 · Gallbladder cancer: urgent symptoms: Two lines told readers with jaundice to seek help "the same day", crediting the NHS. The NHS page says to ask for an urgent GP appointment or get help from 111 and gives no time (Second accuracy pass; the NHS symptoms page for gallbladder cancer read as raw HTML) · Both lines now use the NHS wording
- 2026-11-17 · ACTICCA-1 and the gallbladder cancer adjuvant row: The trial was described as active with results awaited. Its results were published in Annals of Oncology on 24 October 2026: 591 patients randomised, median disease-free survival 25.7 versus 17.9 months (log-rank p 0.097), so the primary endpoint was not met, and overall survival was not significantly improved (Second accuracy pass; a Europe PMC search for the trial found PMID 42926109 (DOI 10.1016/j.annonc.2026.10.1278) and its abstract was read; the registry record NCT02170090 was read through the ClinicalTrials.gov API) · Status is now negative with the reported figures, the restricted mean survival analysis and the authors' conclusion; the standard-of-care row says capecitabine remains the standard; the trial is removed from the pipeline list. The gallbladder cohort is not reported separately in the abstract and the page says the full text could not be read
- 2026-11-17 · NIFE, liposomal irinotecan and Gallbladder cancer in biliary tract cancer trials: NIFE was described as a first treatment for bile duct and gallbladder cancer with no results paper. Its protocol excludes gallbladder cancer, and the results were published in the Journal of Clinical Oncology in 2024 (Second accuracy pass; the registry record NCT03044587 listed PMID 38843469, whose abstract was read through Europe PMC (DOI 10.1200/JCO.23.01566); the protocol full text (PMC6813114) was read for the eligibility criteria) · The record now says gallbladder cancer was excluded and gives the results: 91 randomised, 51 percent free of progression at four months with liposomal irinotecan (primary endpoint met), median survival 15.9 versus 13.6 months with no significant difference. NIFE is removed from the list of trials open to gallbladder cancer
- 2026-11-17 · HERIZON-BTC-01, HERIZON-BTC-302 and the gallbladder cancer HER2 row: The pages said 53 percent of HERIZON-BTC-01 patients had gallbladder cancer. The Ziihera label gives that share for the 62 patients with IHC 3+ tumours, not for the 87 enrolled. The record also called this the largest gallbladder share of any approval study, which no source states (Second accuracy pass; section 14.2 of the Ziihera label read through openFDA, and the final results (PMC12635922) read in full text) · Every mention now says 53 percent of the 62 patients with IHC 3+ tumours, and adds their 52 percent response rate and the final median survival of 15.5 months (18.1 months with IHC 3+). The NICE recommendation chapter could not be re-read and the record says so
- 2026-11-17 · EORTC 1607: The record said no results had been published. They appeared in Translational Oncology on 31 August 2026 (Second accuracy pass; Europe PMC search found PMID 42679517 (DOI 10.1016/j.tranon.2026.103003) and the abstract was read) · Adds the six-month progression-free survival rate of 61.2 percent, median progression-free survival of 8.3 months, response rate of 40.8 percent and median survival of 13.4 months
- 2026-11-17 · EA2251 ivonescimab trial: The trial was named EA2197, which is the number of the OPT-IN gallbladder trial. The registry identifies NCT07765433 as EA2251 (Second accuracy pass; both registry records read through the ClinicalTrials.gov API) · The name and summary now say EA2251. The page address still carries the old number
- 2026-11-17 · ComboMATCH binimetinib with FOLFOX: The record gave 66 patients planned and primary completion in October 2026. The registry now records 18 patients enrolled and primary completion on 8 August 2026 (Second accuracy pass; NCT05564403 read through the ClinicalTrials.gov API (record updated 12 November 2026)) · Enrolment and dates now follow the registry
- 2026-11-17 · ABC-12: The record gave a target of 70 patients and an end date of September 2026. The registry gives a target of 50 and an overall end date of 28 February 2027 (Second accuracy pass; ISRCTN11210442 read through the ISRCTN query API) · Target, recruitment dates and end date now follow the registry
- 2026-11-17 · BILCAP: The gallbladder note said the long-term analysis kept the intention-to-treat hazard ratio at 0.84. The primary figure was 0.81 (0.63 to 1.04, p 0.097), and neither is statistically significant (Second accuracy pass; the abstracts for DOI 10.1016/S1470-2045(18)30915-X and 10.1200/JCO.21.02568 read through Europe PMC) · The note now says the primary endpoint was not met, gives both hazard ratios and says the case for capecitabine rests on the sensitivity and per-protocol analyses
- 2026-11-17 · ASCOT (JCOG1202) and the gallbladder cancer adjuvant row: The record said S-1 is licensed in Europe only for gastric cancer. The European licence also covers some patients with metastatic colorectal cancer (Second accuracy pass; the therapeutic indication on the EMA page for Teysuno read in raw text) · The record gives both licensed uses and says neither is biliary cancer; the EMA page is linked
- 2026-11-17 · Gallbladder cancer first-line and adjuvant rows: The first-line row gave an ESMO benefit score of 3 for durvalumab; the published score is 4. The adjuvant row gave a benefit score of A for capecitabine, where the guideline gives a grade of recommendation and no benefit score (Second accuracy pass; the Pan-Asian adapted ESMO biliary guideline (PMC11410730), which reproduces the ESMO scores, read in full text) · The durvalumab score is now 4 and the unsupported capecitabine score is removed
- 2026-11-17 · ROAR and KEYNOTE-158: ROAR's ninth cohort was given as adenocarcinoma of unknown primary; the registry lists germ cell tumours. KEYNOTE-158 responders were said to stay in response for years, where the paper gives a range of 6.2 to more than 26.6 months (Second accuracy pass; NCT02034110 read through the ClinicalTrials.gov API and the abstract for DOI 10.1002/ijc.33013 read through Europe PMC) · Both statements now follow their sources
- 2026-11-17 · OPT-IN, POLCAGB, RUGB and NEOGB: OPT-IN was called the first randomised test of chemotherapy before re-resection although GAIN started in 2019; POLCAGB was called the largest gallbladder-only randomised trial and RUGB the only phase 2/3 radiotherapy trial, which the registry records for the other trials contradict (Second accuracy pass; the five registry records read through the ClinicalTrials.gov API and compared) · The first, largest, only and third claims are removed or reworded to what the registry shows
- 2026-11-17 · NIFTY: The record said the authors attributed the change between the two readings to reader variation, which the abstract does not say, and described the pooled analysis with NALIRICC without figures (Second accuracy pass; abstracts for DOI 10.1001/jamaoncol.2023.0016, PMID 38870977 and PMID 40147791 read through Europe PMC and PubMed) · The record now gives the discordance rates the abstract reports, the NALIRICC result (hazard ratio 0.87, 0.56 to 1.35) and the pooled result (hazard ratio 0.65 for progression, 0.77 for survival with p 0.051), with both papers linked
- 2026-11-17 · Surgery for gallbladder cancer, Duodenal stenting and gastrojejunostomy and TOPAZ-1: Three statements went further than their sources: liver resection extent in peritoneal-side T2 tumours, an endoscopic bypass comparison credited to the wrong journal and without its lower technical success, and benefit across all subgroups in the TOPAZ-1 three-year update (Second accuracy pass; abstracts for DOI 10.1016/j.surg.2017.05.004, 10.1055/s-0043-101695, 10.14701/ahbps.26-091 and 10.1016/j.jhep.2025.05.003 read through Europe PMC and PubMed) · Each now gives the source's own figures and wording, including the non-significant disease-specific survival result in the T1b meta-analysis
- 2026-11-17 · Regional differences in gallbladder cancer pathogenesis: insights from a multi-institutional comparison of tumor mutations: The record, the gallbladder regional table and the state-of-the-art line said Japanese patients had fewer gallstone-associated cancers and a higher mutation count, without saying the gallstone difference was not statistically significant (36% against 67% and 69%, P = 0.13) and although the median mutation count was 6 in Japan against 7 in Chile. The SMAD4 survival figures (10 against 25 months) were given without saying they are medians from a univariate analysis (Second accuracy pass; the abstract for DOI 10.1002/cncr.31850 read in raw text from Europe PMC) · The gallstone difference is now stated as not significant and removed from the one-line summary; the three median mutation counts are given as the abstract gives them; the SMAD4 figures are labelled as medians on univariate analysis
- 2026-11-17 · The mutational landscape and actionable targets of gallbladder cancer: an ancestry-informed and comparative analysis of a Chilean population: The panel was described as covering 143 genes and as not covering ELF3, SMAD4 or ARID1A comprehensively. The paper says the panel covers 161 genes and does not detect alterations in ARID1A; SMAD4 was detected in 5 of 56 tumours. The record also said Chile has the world's highest gallbladder cancer mortality, which the paper does not say (Second accuracy pass; the open full text (PMC12531073) for DOI 10.3389/fonc.2025.1658528 read through Europe PMC) · Panel size corrected to 161 genes in the paper record and the regional table; the caveat now gives the authors' own limitation; the mortality claim replaced with the paper's wording, one of the highest reported incidence rates
- 2026-11-17 · Gallbladder cancer: The immune microenvironment line said higher PD-L1 marks tumours with shorter survival and cited both the Western and the Indian series. The Indian series found no association between PD-L1 and overall survival (P = 0.546); only the Western series found shorter survival, at a tumour proportion score above 10% (Second accuracy pass; the abstracts for DOI 10.1111/his.13669 and DOI 10.3390/cancers13071682 read in raw text from Europe PMC) · The sentence now gives each series its own result
- 2026-11-17 · Integrated genomic analysis reveals mutated ELF3 as a potential gallbladder cancer vaccine candidate: The record and the gallbladder history line called the cohort three-continent. The patients came from Korea, India and Chile: three countries on two continents (Second accuracy pass; the abstract for DOI 10.1038/s41467-020-17880-4 read in raw text from Europe PMC) · Changed to three-country in both places
- 2026-11-17 · Genomic characterization of co-existing neoplasia and carcinoma lesions reveals distinct evolutionary paths of gallbladder cancer: The record, the precancer term, the pathway note and the gallbladder page said CTNNB1 mutation is an early event that drives or starts the adenoma route, and that ageing is the main source of mutations. The paper says CTNNB1 mutations have a critical role and were found in 5 of 11 patients, and that ageing is a major factor (Second accuracy pass; the abstract and open full text (PMC8346570) for DOI 10.1038/s41467-021-25012-9 read through Europe PMC) · Reworded to the paper's own claims in all five places; the caveat now gives the 11 patients and 33 tumour samples from the full text
- 2026-11-17 · Zanidatamab zovodotin: The payload was described as auristatin-class, which neither cited review says, and the note said the programme did not advance, which the registry does not state (Second accuracy pass; the ADC table in the open full text for DOI 10.3390/cancers16040800, the open full text of Niquille and colleagues in mAbs 2024 (PMC10936611), and the ClinicalTrials.gov record for NCT03821233 read as raw text) · Payload now reads cytotoxic tubulin inhibitor, the wording of the mAbs review; the note now gives the registry facts: phase 1 completed with 112 patients enrolled and no later-phase study of the conjugate registered
- 2026-11-17 · Details of human epidermal growth factor receptor 2 status in 454 cases of biliary tract cancer: The 83% heterogeneity figure was written in six places as applying to HER2-positive tumours. The abstract gives it for biliary tract cancer cases without stating the denominator, and the full text is not open. The record also called the 31.3% gallbladder rate the highest in the literature, which the source does not say (Second accuracy pass; the abstract for DOI 10.1016/j.humpath.2020.08.006 read in raw text from Europe PMC) · The figure is now given as 83% of cases with the abstract's wording in the paper record, a caveat names the missing denominator, and the rate is called the highest among the series on this site
- 2026-11-17 · Genomic spectra of biliary tract cancer: The record and the gallbladder history line said the paper shows gallbladder cancer carries HER2 and ELF3 and that IDH belongs to the intrahepatic ducts. The abstract ties only the APOBEC signature to gallbladder and extrahepatic tumours and FGFR2 fusions to intrahepatic tumours; ELF3 is one of 32 significantly altered genes across all sites (Second accuracy pass; the abstract for DOI 10.1038/ng.3375 read in raw text from Europe PMC) · Both lines reworded to what the abstract states
- 2026-11-17 · Pertuzumab and trastuzumab for HER2-positive, metastatic biliary tract cancer (MyPathway): The record called the study the first prospective evidence that HER2 blockade works in biliary cancer and said it led to HERIZON-BTC-01 and prompted the later randomised trials. The abstract says only that the response rate is promising for starting randomised trials of the two antibodies (Second accuracy pass; the abstract for DOI 10.1016/s1470-2045(21)00336-3 read in raw text from Europe PMC) · The priority and causal claims are removed; the one-line summary now reports the authors' conclusion
- 2026-11-17 · Whole-exome and targeted gene sequencing of gallbladder carcinoma identifies recurrent mutations in the ErbB pathway: Five paper records carried a first or largest claim that their sources do not make: this one (the first exome study), the Indian 376-patient series, the 244-sample Memorial Sloan Kettering series, the 1,502-tumour profiling series and the 174-case PD-L1 series (the largest of their kind) (Second accuracy pass; each abstract read in raw text from Europe PMC by DOI (10.1038/ng.3030, 10.1200/go-25-00332, 10.1158/1078-0432.ccr-22-1954, 10.21037/jgo.2018.08.18, 10.1111/his.13669)) · Each claim replaced with the cohort size or a plain description; the gallbladder history title and the HER3 note changed to match
- 2026-11-17 · Intracholecystic papillary-tubular neoplasm (ICPN): A sentence on CTNNB1 mutation in incipient ICPNs had no source; neither cited paper says it (Second accuracy pass; the two cited abstracts read from Europe PMC, and a Europe PMC search found the 2025 letter that reports it (DOI 10.5858/arpa.2025-0261-le), which has no abstract) · The letter is added as a link and the sentence says it was read by its title only
- 2026-11-17 · Molecular and clinical determinants of targeted therapy treatment in biliary tract cancer: The cBioPortal link was labelled Hepatobiliary Cancer, MSK 2026, and the record said SMAD4 co-mutation predicts a worse response. The study is named Biliary Tract Cancers (MSK, Clin Cancer Res 2026) and the abstract says SMAD4 alterations were associated with inferior outcomes (Second accuracy pass; the cBioPortal study record read through its public API and the abstract for DOI 10.1158/1078-0432.ccr-26-0428 read from Europe PMC) · Link label and wording corrected
- 2026-11-17 · Gallbladder cancer in the UK: NTRK fusion funding row: The row said NICE recommends entrectinib through the Cancer Drugs Fund under TA644. NICE replaced TA644 on 7 January 2026 with TA1118, a terminated appraisal in which it could make no recommendation because Roche did not provide a complete evidence submission. The Scottish cell had the two medicines the wrong way round: it said larotrectinib was accepted and entrectinib should be checked (Second accuracy pass; the NICE pages for TA644, TA1118 and TA630 read as raw HTML, the Cancer Drugs Fund list version 1.410 read as PDF text, and the Scottish Medicines Consortium advice SMC2295 and its search read as raw HTML) · The row, the NTRK test row and the funding note now say larotrectinib remains in the Fund under TA630, entrectinib is no longer recommended in England, and in Scotland entrectinib is accepted (SMC2295, 8 March 2021) with no larotrectinib advice found
- 2026-11-17 · Gallbladder cancer in the UK: Cancer Drugs Fund details: The page said zanidatamab was not yet on the Cancer Drugs Fund list and that the current list could not be read. It gave pembrolizumab two biliary forms, PEMB27 and PEMB28, and dated futibatinib funding from 12 December 2024 (Second accuracy pass; the National Cancer Drugs Fund list version 1.410 (12 November 2026), located through the NHS England publication index and read as PDF text) · Zanidatamab is now shown in Section B with form ZANI1 and funding from 5 August 2026. The pembrolizumab biliary form is PEMB28 only, since PEMB27 is the small intestine form. Futibatinib funding is dated 10 December 2024. The gap notes were rewritten
- 2026-11-17 · Gallbladder cancer in the UK: BILCAP and adjuvant capecitabine: The trial story said median survival was longer with capecitabine in the pre-specified analysis without saying the primary intention-to-treat comparison was not statistically significant. The funding row called capecitabine the only adjuvant treatment with a randomised trial behind it (Second accuracy pass; the abstracts for DOI 10.1016/S1470-2045(18)30915-X and DOI 10.1016/S0140-6736(22)02038-4 read through Europe PMC) · The story now gives the intention-to-treat result (51.1 against 36.4 months, hazard ratio 0.81, 95% CI 0.63 to 1.04, p=0.097) beside the per-protocol result, and names the Japanese ASCOT trial of adjuvant S-1. The word only was removed
- 2026-11-17 · Gallbladder cancer in the UK: waiting-time targets: The page still gave targets that expired in March 2026: a Faster Diagnosis Standard of 75% rising to 80% by March 2026, and a 62-day planning target of 75% by March 2026 (Second accuracy pass; the NHS England Medium Term Planning Framework for 2026/27 to 2028/29 read through the site's WordPress API) · The Faster Diagnosis Standard is now given as 80%, and the 62-day planning target as every trust at 80% by March 2027, in the timeline and the four-nations table
- 2026-11-17 · Gallbladder cancer in the UK: UK trials: The zanidatamab early access study (ISRCTN13555554) was shown as recruiting to 30 September 2026, a date that has passed, and the microbiome study (ISRCTN11210442) carried a note to check its status (Second accuracy pass; both records read through the ISRCTN registry API) · Each status now gives the recruitment end date on the register, says it has passed, and gives the listed study end date
- 2026-11-17 · Gallbladder cancer in the UK: specialist MDT and centre list: The page credited NHS England specification 2260 with saying the specialist MDT decides on incidental cancers found at cholecystectomy, which the specification does not mention. The Glasgow Royal Infirmary row cited the AMMF centre list, which lists the Beatson instead. The IDH1 test was described in words GeNotes does not use (Second accuracy pass; specification 2260 read as PDF text and the AMMF and GeNotes pages read as raw HTML) · The text now says the MDT takes every decision for the new and recurrent cases referred to it. The AMMF source was removed from the Glasgow row and the IDH1 test wording follows GeNotes
- 2026-11-17 · John Bridgewater: The record called him chief investigator of ABC-06 and leader of the National Cancer Research Institute biliary tract subgroup. The ABC-06 paper names Juan Valle as chief investigator, and his UCLH page describes the upper gastrointestinal study group (Second accuracy pass; the ABC-06 full text (PMC8082275) read through Europe PMC, ClinicalTrials.gov NCT01926236, and the Internet Archive copy of his UCLH consultant page) · He is now described as a co-author and trial management group member of ABC-06, and the study group wording follows the UCLH page
- 2026-11-17 · Mairéad McNamara: The record said she co-authored ABC-06 and listed it as her paper. She is not among its named authors. It also named two international collaborations that neither of her profile pages mentions (Second accuracy pass; the ABC-06 author list and full text read through Europe PMC, and her Christie and University of Manchester pages read as raw HTML) · The ABC-06 credit and trial link were removed, the paper was replaced with one she first-authored (DOI 10.1016/j.ejso.2020.03.202, resolved through Europe PMC), and the two memberships were removed
- 2026-11-17 · NIFTY: liposomal irinotecan in second-line biliary tract cancer (Lancet Oncol 2021): The record gave the first analysis (median progression-free survival 7.1 against 1.4 months, hazard ratio 0.56) as the trial's result. An updated analysis with the scans read again superseded it. (Second accuracy pass; the abstract for DOI 10.1001/jamaoncol.2023.0016 (PMID 36951834) and the NALIRICC abstract (PMID 38870977) read through Europe PMC.) · The one-line description, summary, findings and caveats now give the updated 4.2 against 1.7 months (hazard ratio 0.61, 95 percent CI 0.44 to 0.86) beside the first figure, and NALIRICC's 2.6 against 2.3 months (hazard ratio 0.87, 0.56 to 1.35, not significant). Both papers are linked.
- 2026-11-17 · Management of incidental gallbladder cancer in the nationwide CAPBIL study: The record left out the abstract's adjuvant result, and the gallbladder cancer page said the CAPBIL cohort saw no benefit from chemotherapy after surgery. The incidental-cancer paper reports longer disease-free survival (35 against 15 months) and overall survival (47 against 26 months) in those who completed adjuvant chemotherapy, unmatched. (Second accuracy pass; the abstracts for DOI 10.1093/bjs/znag050 and DOI 10.1016/j.hpb.2026.06.009 read through Europe PMC.) · Added the overall survival figures (72 against 26 months with liver resection), the adjuvant figures and the multivariable hazard ratios. The open problem on the cancer page and the companion paper now state both the matched null result in 516 patients and the unmatched signal, and the companion paper is described as gallbladder-only, not mostly gallbladder.
- 2026-11-17 · Timing of revision surgery for incidental gallbladder cancer: individual patient data meta-analysis: The hazard ratio 1.29 was given without its comparison and read as no difference by timing in general. It compares revision at four weeks or sooner with revision after four weeks only. The one-line description also said individual records from more than 2,000 patients were pooled; 2,067 is the review total and the abstract does not give the pooled number. (Second accuracy pass; the abstract for DOI 10.1016/j.hpb.2025.12.017 read through Europe PMC as raw text.) · Named the comparison in the summary, finding and one-line description, in the Ethun 2017 record and in the 2017 history entry on the cancer page. The summary now says the pooled number is not given.
- 2026-11-17 · Optimal surgical treatment in patients with T1b gallbladder cancer: The one-line description said about 95 in 100 were alive and free of the disease at five years. The figure (94.6 percent) is disease-specific survival, the share who had not died of the cancer. (Second accuracy pass; the abstract for DOI 10.1002/jhbp.593 read through Europe PMC.) · Reworded to about 95 in 100 had not died of the cancer at five years.
- 2026-11-17 · Detecting Early Recurrence With Circulating Tumor DNA in Stage I-III Biliary Tract Cancer: A median relapse-free survival of 6.6 months was written as the cancer coming back within about seven months on average, and the record said conventional markers were not prognostic when the abstract tested CA 19-9 alone. (Second accuracy pass; the abstract for DOI 10.1200/po-24-00443 read through Europe PMC.) · The one-line description now says half had relapsed within about seven months. The marker sentence names CA 19-9 with its hazard ratio (1.17, 0.24 to 5.71), and the abstract's sensitivity (15 of 16 recurrences) and 3.7-month lead time are added.
- 2026-11-17 · Prophylactic (preventive) cholecystectomy: The record said gallbladder cancer almost always arises with gallstones, that Chile's guarantee covers symptomatic gallstones, that 284,139 notifications had been issued by 2024, and (in the cancer page history and the Samaniego paper record) that the programme is the only one of its kind in the world. The cited abstract supports none of these four points as written. (Second accuracy pass; the abstracts for DOI 10.4067/s0034-98872024001001028 and DOI 10.21037/cco.2019.08.03 read through Europe PMC. The journal's own site refused a scripted read.) · Reworded to what the abstracts say: stones are present in 80 percent of Indian patients, the guarantee covers gallstones at ages 35 to 49, and a 2024 evaluation counted 284,139 notifications since 2006. The only-in-the-world wording is removed and a caveat notes that the end year of the count is not given.
- 2026-11-17 · Second-line FOLFOX chemotherapy versus active symptom control (ABC-06): The one-line description called a difference between medians (6.2 against 5.3 months) about a month longer on average. (Second accuracy pass; the abstract for DOI 10.1016/S1470-2045(21)00027-9 read through Europe PMC.) · Reworded to a median of about a month longer.
- 2026-11-17 · Biliary Tract Cancers, Version 2.2025, NCCN Clinical Practice Guidelines In Oncology: The record said the article concentrates on treatment after surgery. The abstract says it also covers systemic treatment for advanced disease. (Second accuracy pass; the abstract for DOI 10.6004/jnccn.2025.0042 read through Europe PMC. The journal's full text refused a scripted read.) · Added the advanced-disease scope, and a caveat that the trials behind the adjuvant section could not be re-read.
- 2026-11-17 · HER2/HER3 pathway in biliary tract malignancies: systematic review and meta-analysis: The record said one in five advanced gallbladder cancers may qualify for a HER2 drug, and placed the review in Manchester alone. The review estimates up to 20 percent over-expression with close to 60 percent of those amplified, and its first author is in Milan. (Second accuracy pass; the abstract and affiliations for DOI 10.1007/s10555-016-9645-x and the HERIZON-BTC-01 abstract (PMID 37276871) read through Europe PMC.) · Reworded to the review's own two-step estimate and to Milan and Manchester.
- 2026-11-17 · Gallbladder cancer: Two notes attached to terms misdescribed their sources: the T2 survival comparison (42.6 against 64.7 percent) was credited to all 437 patients in the Shindoh series when it covers the 252 with T2 tumours, and polyps were said to be found on 4 to 6 percent of scans when the source says more than 4 percent, and 5.8 percent in its own cohort. (Second accuracy pass; the abstracts for DOI 10.1097/SLA.0000000000000728 and DOI 10.1001/jamanetworkopen.2020.5143 read through Europe PMC, with the open full text of the second.) · Both notes reworded to the source's figures.
- 2026-11-17 · Gallbladder cancer roadmap: The roadmap still treated ACTICCA-1 as pending: it "will say" whether gemcitabine and cisplatin beat capecitabine after surgery, with a watch item for a December 2025 completion. The trial was published on 24 October 2026 and did not meet its primary endpoint (Second accuracy pass; the abstract for PubMed id 42926109 (Annals of Oncology 2026, DOI 10.1016/j.annonc.2026.10.1278) read through Europe PMC, and the registry record NCT02170090 read through the ClinicalTrials.gov API) · The summary, two steps and the watch item now give the result: 591 randomised, 16 percent with gallbladder cancer, median disease-free survival 25.7 against 17.9 months, log-rank p=0.097, not statistically significant, so capecitabine remains the standard. The registry still lists 789 enrolled and no posted results, and the page says so
- 2026-11-17 · Gallbladder cancer roadmap: The roadmap said no trial acts on a positive residual disease blood test after biliary surgery and that none was found on ClinicalTrials.gov. Two registered studies do: a Chinese study of test-guided adjuvant chemotherapy (NCT06171321, 94 planned, status unknown) and a US pilot across gastrointestinal cancers (NCT05482516, 11 enrolled). The same idea record said 23 percent of 167 patients tested positive, when the source gives 19 of 82 tested in that window (Second accuracy pass; a ClinicalTrials.gov API search and the two registry records, and the abstract for PubMed id 42583118 read through Europe PMC) · The step, the watch item and the idea now say that no randomised trial was found and name the two small non-randomised studies; the detection rate is given as 19 of 82 in a cohort of 167
- 2026-11-17 · A national incidental gallbladder cancer pathway: The summary said a 2021 systematic review showed selective histology policies risk missing cancers. The review concludes the opposite: selective histology "can be preferred if a careful intraoperative macroscopic examination is done and patient risk factors are taken into consideration" (Second accuracy pass; the abstract for PubMed id 33773526 read through Europe PMC) · The sentence now says the question is contested and gives the review's own conclusion
- 2026-11-17 · Redesign Chile's prophylactic cholecystectomy programme: The summary said Mapuche ancestry raises risk independently of gallstones and body mass index. The cited analysis points to gallstones as a mediator of the ancestry effect and found ancestry lowered body mass index slightly (Second accuracy pass; the abstract for PubMed id 37627062 read through Europe PMC) · The sentence now says the risk runs partly through gallstone disease and not through body mass index
- 2026-11-17 · Gallbladder cancer ideas: separate reporting, T1b surgery, chemoradiation and neoadjuvant trials: Four smaller faults across the idea records: the CAPBIL surgical cohort was called "mostly gallbladder" when all 516 patients had gallbladder cancer; the T1b record said 95 in 100 were "alive without the disease" when the endpoint was disease-specific survival and called the series the largest; the chemoradiation record credited the share of local relapses in SWOG S0809 to margin-positive patients when the 23 of 47 is for all patients; and the neoadjuvant record said OPT-IN "reports in 2028" when July 2028 is an estimated primary completion date (Second accuracy pass; abstracts for PubMed ids 42399203, 30562839 and 25964250 read through Europe PMC and the registry record NCT04559139) · Each sentence now states what its source states
- 2026-11-17 · Gallbladder cancer: Pakistan: The Karachi case-control study was quoted as having 78 percent women among patients. That is the control group; 67 percent of the cancer patients were women (Second accuracy pass; the abstract for PubMed id 22151791 read through Europe PMC) · The figure is now 67 percent of the cancer patients
- 2026-11-17 · Gallbladder cancer: central and eastern Europe: The card said Poland, the Czech Republic and Slovakia have Europe's highest rates. In GLOBOCAN 2024 Slovenia is highest (1.83 per 100,000), then Slovakia (1.42), Bosnia and Herzegovina (1.28), Croatia (1.22), Poland (1.14) and Czechia (1.09) (Second accuracy pass; IARC's GLOBOCAN 2024 estimates in the repository file, checked against the Cancer Today API for eight populations) · The summary now lists the six countries in order with their rates
- 2026-11-17 · Gallbladder cancer in India: The card said India runs the world's only randomised trials of treatment before surgery, although OPT-IN in the United States and GAIN in Germany are randomised trials of the same kind. It gave patients' ages as their fifties and sixties where the review says the fifth and sixth decades, quoted a mustard oil odds ratio of 1.33 without saying it was not statistically significant, and another card said India has about a tenth of world cases when GLOBOCAN 2024 gives about a quarter (Second accuracy pass; registry records NCT04559139, NCT03673072, NCT02867865, NCT06712420 and NCT06214572, the abstracts for PubMed ids 31484488 and 32142159 read through Europe PMC, and the IARC Cancer Today API) · The trials sentence now names the three Indian trials and the two elsewhere, with planned enrolment marked as planned; the age, the odds ratio and the share of world cases follow their sources
- 2026-11-17 · Gallbladder cancer in Chile and the Andes: Four statements went beyond their sources: that registry data made Chile's incidence the highest "in the world" (the paper says Central and South America), that ancestry "tracks risk gene by gene" (the paper gives 3.7 percent higher mortality per 1 percent Mapuche ancestry and no association for Native American ancestry as a whole), that the 2024 American Journal of Epidemiology analysis is "the most rigorous test yet" (it has no abstract and could not be read), and the programme number 26 (the regulator's page now returns HTTP 404) (Second accuracy pass; abstracts for PubMed ids 27678326 and 40052976 and DOI 10.1371/journal.pgen.1006756 read through Europe PMC, the record for PubMed id 38576158, and the Internet Archive copy of the Superintendencia de Salud page) · The first two now carry the sources' own figures; the third is labelled as unread; the page says the number could not be re-checked
- 2026-11-17 · Gallbladder cancer: the Americas: The card gave the excess among Native American people in Alaska and the Southwest as five to eight times (the meta-analysis gives 6-fold in men and 8.5-fold in women), credited a US Hispanic rate to GLOBOCAN 2018 and to women (the abstract does neither), and said the rise among non-Hispanic Black Americans was mostly at distant stage without the figures (Second accuracy pass; abstracts for PubMed ids 42024224, 31922259 and 38963040 read through Europe PMC) · The three sentences now give the sources' figures, including the plateau after 2014 and the 2.72 percent a year rise in distant-stage disease
- 2026-11-17 · HER2 and genomic testing for biliary cancer on the NHS: The record said a HER2 antibody is funded on the NHS for HER2-positive disease and put that beside "about one in ten" tumours. NICE's recommendation defines HER2-positive as an immunohistochemistry score of 3, a narrower group than the one in ten, which also counts tumours scoring 2 with gene amplification (Second accuracy pass; the NICE TA1153 recommendation chapter and the GOV.UK MHRA announcement read as raw text, and the abstract for PubMed id 42595189) · The term, the standard-of-care row, the question sheet and the HER2 testing idea now say the recommendation covers tumours scoring 3
- 2026-11-17 · Emotional support and helplines (UK): Two quotations did not match their sources: Cancer Research UK writes "about the cancer", not "about your cancer", and Macmillan defines a carer as giving "any unpaid help and support", not "help or support". The NHS gallbladder cancer support page quoted for the helpline numbers has been retired and now redirects to the main page (Second accuracy pass; the Cancer Research UK, Macmillan and NHS pages fetched as raw text, and the Internet Archive copies of the NHS page from 18 April and 10 August 2026) · Both quotations now carry the sources' exact words (the second is on the carers record), and the NHS wording is dated to its archived copy
- 2026-11-17 · Evaluating Novel Therapies in ctDNA Positive GI Cancers: The record said the trial was now recruiting, with a registry enrolment figure of 20 and a primary completion due in 2026-12. The registry entry, updated on 2026-10-06, lists it as active and not recruiting, with an actual enrolment of 11 and an actual primary completion date of 2026-08-20 (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT05482516 read as raw JSON and compared field by field by script) · Status set to active, the one-line summary now says active and no longer recruiting, enrolment set to 11 and the summary now says the trial reached its primary completion in 2026-08-20
- 2026-11-17 · Durvalumab and Tremelimumab in Combination With Propranolol and Chemotherapy for Treatment of Advanced Hepatopancreabiliary Tumors (BLOCKED): The record said the trial was now recruiting, with a registry enrolment figure of 62 and a primary completion due in 2026-10-01. The registry entry, updated on 2026-11-04, lists it as active and not recruiting, with an actual enrolment of 15 and an anticipated primary completion of 2027-05-01 (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT05451043 read as raw JSON and compared field by field by script) · Status set to active, the one-line summary now says active and no longer recruiting, enrolment set to 15 and the primary completion date set to 2027-05-01
- 2026-11-17 · Agnostic Therapy in Rare Solid Tumors: The record said the trial was now recruiting, with a primary completion due in 2026-07. The registry entry, updated on 2026-10-13, lists it as active and not recruiting, with an anticipated primary completion of 2027-07 (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT06638931 read as raw JSON and compared field by field by script) · Status set to active, the one-line summary now says active and no longer recruiting, and the primary completion date set to 2027-07
- 2026-11-17 · Polymeric Micellar Paclitaxel-Based Combination Therapy for Advanced Biliary Tract Cancer: The record said the trial was active and no longer recruiting, with a registry enrolment figure of 61 and a primary completion due in 2026-08-12. The registry entry, updated on 2026-10-15, lists it as recruiting, with an anticipated enrolment of 25 and an anticipated primary completion of 2027-08-07 (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT07159204 read as raw JSON and compared field by field by script) · Status set to recruiting, the one-line summary now says now recruiting, enrolment set to 25 and the primary completion date set to 2027-08-07
- 2026-11-17 · Trifluridine/Tipiracil + Oxaliplatin in Participants With Advanced or Metastatic Biliary Tract Cancer: The record said the trial was now recruiting. The registry entry, updated on 2026-10-20, lists it as suspended and gives the reason as "Interim Analysis" (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT07146646 read as raw JSON and compared field by field by script) · Status set to active, and the one-line summary and the summary now say the registry lists the study as suspended and quote its stated reason
- 2026-11-17 · Target Therapy With GEMOX in Recectable Gallbladder Carcinoma Patients Monitored by ctDNA: The record said the trial was now recruiting. The registry now gives its overall status as unknown, with a last known status of recruiting, and the entry was last verified in 2024-10 (Second accuracy pass; the ClinicalTrials.gov v2 API record for NCT04183712 read as raw JSON and compared field by field by script) · The one-line summary and the summary now say the trial was last listed as recruiting and that the registry gives its status as unknown; the status field is left as recruiting because the schema has no value for unknown
- 2026-11-17 · Gallbladder cancer: three registry trials enrolling by invitation: Three records said "now recruiting" where the registry status is enrolling by invitation, which a reader cannot simply join: NCT06901622, NCT06963060, NCT07310069 (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · Each one-line summary now says "now enrolling by invitation only"
- 2026-11-17 · Gallbladder cancer: registry trial dates moved by their sponsors: Three records carried dates the registry has since changed. NCT06423170 gave a primary completion of 2025-12-31 where the registry now gives 2027-12-31. NCT07433673 gave a start of 2026-08 and a primary completion of 2028-01 where the registry now gives 2027-01 and 2032-01. NCT07723534 gave a start of 2026-08 where the registry now gives 2026-11 (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and compared field by field by script) · The dates in each summary now match the registry
- 2026-11-17 · Gallbladder cancer: registry trials past their primary completion: Eleven records said the trial was "due to reach its primary completion" on a date the registry records as an actual, past date: NCT03364530, NCT03482102, NCT04005339, NCT04042831, NCT04634058, NCT04781192, NCT04941287, NCT04969887, NCT05451290, NCT05640791, NCT05994001 (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and the date type compared by script) · Each summary now says the trial "reached its primary completion" on that date. Three of them (NCT04005339, NCT04042831 and NCT04941287) now have results posted on the registry; no outcome figures were added
- 2026-11-17 · Gallbladder cancer: registry trials that have not started: Forty-nine records said the trial "started" on a date the registry gives only as an anticipated start, some of them in the future: NCT04910386, NCT05985707, NCT06037655, NCT06389500, NCT06502080, NCT06530823, NCT06531291, NCT06654947, NCT06662877, NCT06903273, NCT06919848, NCT06993025, NCT07003815, NCT07099794, NCT07109167, NCT07129018, NCT07151872, NCT07159217, NCT07160283, NCT07263360, NCT07269158, NCT07296666, NCT07351591, NCT07433673, NCT07437287, NCT07471165, NCT07472933, NCT07514533, NCT07530445, NCT07561775, NCT07598318, NCT07599995, NCT07623642, NCT07631884, NCT07632235, NCT07636798, NCT07638501, NCT07649980, NCT07701122, NCT07723534, NCT07726446, NCT07729033, NCT07729917, NCT07733115, NCT07745296, NCT07767994, NCT07780838, NCT07793253, NCT07829874 (Second accuracy pass; the ClinicalTrials.gov v2 API record for each trial read as raw JSON and the date type compared by script) · Each summary now says "planned to start" with the registry date
- 2026-11-17 · Incidental gallbladder cancer decision tool: The quoted standard-of-care row and a suggested question still treated ACTICCA-1 as awaited and as a trial to join, and the source label gave BILCAP an ESMO-MCBS grade of A, which is a guideline grade of recommendation and not a benefit score (Second accuracy pass; the ACTICCA-1 abstract (DOI 10.1016/j.annonc.2026.10.1278) read through Europe PMC, and ClinicalTrials.gov NCT02170090 read as JSON) · The row now matches the corrected gallbladder row, the question no longer names a closed trial, and the benefit score is removed from the label
- 2026-11-17 · Zanidatamab: The US approval note said 53% of HERIZON-BTC-01 patients had gallbladder cancer. The label gives that share for the 62 patients with IHC 3+ tumours in its efficacy population (Second accuracy pass; the Ziihera label read through openFDA) · The note now says 53% of the 62 patients with IHC 3+ tumours
- 2026-11-17 · Entrectinib in UK coverage: The UK coverage row gave TA643 for NTRK fusion-positive solid tumours and TA644 for ROS1-positive lung cancer, which is the wrong way round (Second accuracy pass; NICE pages for TA643, TA644 and the terminated TA1118 read in raw text) · The numbers are swapped back, and the note says NICE states TA644 has been updated and replaced by TA1118, a terminated appraisal
- 2026-11-17 · India: gallbladder trials: The card said India runs the world's only randomised trials of treatment before surgery in gallbladder cancer. ClinicalTrials.gov lists OPT-IN in the United States and GAIN in Germany (Second accuracy pass; ClinicalTrials.gov records NCT04559139 and NCT03673072 read as JSON) · The card now says India runs three randomised trials in gallbladder cancer
- 2026-11-17 · Gallbladder cancer: Chile's cholecystectomy guarantee: Three passages called it the only such programme in the world and dated 284,139 notifications to 2024. The cited abstract counts them since 2006, gives no end year and makes no such comparison (Second accuracy pass; the abstract for DOI 10.4067/s0034-98872024001001028 read through Europe PMC) · The count is now credited to an evaluation published in 2024, and the only-in-the-world wording is removed
- 2026-11-16 · Hypercalcaemia of malignancy: best-supported version: The page said both sources start with assessment and intravenous saline. The Endocrine Society 2023 abstract, the only part of that guideline read, covers drug treatment and does not mention fluids. (Second accuracy pass; Endocrine Society 2023 abstract, read raw from the Europe PMC REST API) · Saline and assessment are now attributed to the Society for Endocrinology 2016 guidance alone.
- 2026-11-16 · Hypercalcaemia of malignancy: medicines review step: The step told the reader to stop medicines that raise calcium and named calcium and vitamin D supplements. The Society for Endocrinology guidance says only to take a history of drugs including supplements and over-the-counter preparations, and lists thiazide diuretics, lithium, hypervitaminosis D and A and theophylline toxicity among the causes. (Second accuracy pass; Society for Endocrinology 2016 full text XML from Europe PMC (PMC5314807)) · The step now says to ask about all medicines and gives the source's own list of causes. The missing instruction is named in the gaps.
- 2026-11-16 · Hypercalcaemia of malignancy: bisphosphonate step: The step and summary said to give the bisphosphonate at a lower dose in kidney impairment and presented it as the routine next step. The source says to consider intravenous bisphosphonates if further treatment is required after saline, and to give more slowly and consider dose reduction in renal impairment. (Second accuracy pass; Society for Endocrinology 2016 full text XML from Europe PMC (PMC5314807)) · Reworded to the source's conditions, with the pamidronate and ibandronic acid doses and the pamidronate rate of 20 mg an hour added and quoted.
- 2026-11-16 · Hypercalcaemia of malignancy: glucocorticoid step and its grade: The page said to add a bisphosphonate or denosumab if calcium stays high or for persistent severe hypercalcaemia. The Endocrine Society says where patients continue to have severe or symptomatic hypercalcaemia. The page also described steroids as for lymphoma and other calcitriol-driven causes, where the Society for Endocrinology lists them second line in lymphoma, other granulomatous diseases or vitamin D poisoning. (Second accuracy pass; Endocrine Society 2023 abstract and Society for Endocrinology 2016 full text, both read raw from Europe PMC) · Step, summary and grade claim reworded to the sources' conditions.
- 2026-11-16 · Hypercalcaemia of malignancy: certainty ratings: Low certainty was printed against the recommendations on tumours with high calcitriol and on parathyroid carcinoma. The abstract states low certainty in a sentence directly before the three recommendations on denosumab and calcitonin and does not rate the later two separately. (Second accuracy pass; Endocrine Society 2023 abstract, read raw from the Europe PMC REST API; the full text returned 403) · Those two are now given as conditional recommendations without a certainty rating, and the gap is stated.
- 2026-11-16 · Hypercalcaemia of malignancy: reasons given for steps: Several reasons were not in either source: that calcitonin works within hours, that loop diuretics worsen dehydration, that bisphosphonates damage the kidneys if given too fast, that low albumin hides a high calcium, and that a low PTH points to the cancer, where the source says malignancy or other less common causes. Tiredness and drowsiness were listed as symptoms and are not in the source's list. (Second accuracy pass; Society for Endocrinology 2016 full text XML and Endocrine Society 2023 abstract, both from Europe PMC) · Each reason now says only what the cited source says. The symptoms follow the source's list.
- 2026-11-16 · Metastatic spinal cord compression: who is immobilised: The page said to immobilise anyone with suspected cord compression and neurological signs or signs of spinal instability, lying flat. NICE says to immobilise people with suspected or confirmed cord compression whose neurological symptoms or signs suggest spinal instability, nursed supine, flat or with partial elevation. (Second accuracy pass; NICE NG234 recommendations 1.4.1 and 1.4.3, fetched and read as raw text) · The step, the summary and the best-in-class statement now give the NICE scope and position.
- 2026-11-16 · Metastatic spinal cord compression: the 24-hour coordinator: The summary said the coordinator service must be covered 24 hours a day. NICE says services should ensure the role is covered 24 hours a day, 7 days a week. (Second accuracy pass; NICE NG234 recommendation 1.1.7, read as raw text) · Changed must to should and added 7 days a week.
- 2026-11-16 · Metastatic spinal cord compression: who is given dexamethasone: The patient summary and the clinical summary read as if everyone with suspected compression is given dexamethasone, stopped if cord compression is ruled out. NICE offers 16 mg to people with neurological symptoms or signs, does not offer corticosteroids routinely without them, and says to stop when both spinal metastases and cord compression are ruled out. (Second accuracy pass; NICE NG234 recommendations 1.8.1, 1.8.2, 1.8.3 and 1.8.7, read as raw text) · Both summaries and the step now carry the scope and the stopping rule as NICE words them.
- 2026-11-16 · Metastatic spinal cord compression: neck pain and how fast to call: The page listed new back or neck pain as a warning sign, and called pain alone an emergency needing a same-day call. NICE's list names back pain only, treats neurological symptoms as the emergency, and asks for advice within 24 hours for pain alone. (Second accuracy pass; NICE NG234 box 1 and recommendations 1.3.2, 1.3.3 and 1.3.6, read as raw text) · Removed neck, separated pain (urgent call) from nerve symptoms (emergency), and named the neck omission in the gaps.
- 2026-11-16 · Metastatic spinal cord compression: the ASTRO radiotherapy schedules: The page gave ASTRO's four schedules as being for patients not having surgery. The abstract says they are for patients who are ineligible for surgery and receiving conventional radiotherapy. (Second accuracy pass; ASTRO 2024 guideline abstract from Europe PMC, read as raw text) · Summary and best-in-class statement reworded to not eligible for surgery.
- 2026-11-16 · Metastatic spinal cord compression: the team plan and the reasons given for steps: The team-planning step dropped NICE's scope (before an invasive intervention is offered). Twelve reasons said more than NICE does, for example that plain X-rays miss most spinal metastases, that metastases are often at more than one level, and that radiotherapy shrinks the tumour. NICE's rationale says plain X-rays with neurological assessment had very low sensitivity for cord compression and that radiotherapy is likely to prevent further neurological damage. (Second accuracy pass; NICE NG234 recommendation 1.11.1 and the rationale and impact page, read as raw text) · The step and the reasons were reworded to what the recommendations and rationale state.
- 2026-11-16 · Metastatic spinal cord compression: two quotations: Two quoted NICE recommendations carried a space before a closing bracket that is not in the source. (Second accuracy pass; NICE NG234 recommendations 1.2.2 and 1.5.2, checked by script against the page markup) · Both quotations now match the source character for character.
- 2026-11-16 · Tumour lysis syndrome: rasburicase duration: The page gave rasburicase "for 3 to 5 days" as a fixed range. The Italian consensus says the course continues "for at least 3-5 days". (Second accuracy pass; Italian consensus conference (Haematologica 2008), open PDF fetched raw and read as text) · The step, summary and best-in-class statement now say at least 3 to 5 days, and the step names the inpatient setting.
- 2026-11-16 · Tumour lysis syndrome: when hydration starts: The page said hydration starts 48 hours before treatment. The source says at least 48 hours before, when possible. (Second accuracy pass; Italian consensus conference (Haematologica 2008), open PDF fetched raw and read as text) · The step and summary now say at least 48 hours.
- 2026-11-16 · Tumour lysis syndrome: what the guidelines agree on: The page said hydration for everyone at risk and allopurinol at lower risk were common ground. The sources differ: the international panel recommends close monitoring alone for low risk and keeps allopurinol or rasburicase for intermediate risk, while the Italian consensus gives low-risk patients allopurinol, hydration and urine alkalinisation. (Second accuracy pass; Coiffier 2008 abstract from Europe PMC and the Italian consensus PDF, both read as raw text) · The best-in-class statement now sets out both schemes and who holds each.
- 2026-11-16 · Tumour lysis syndrome: newest guidelines: The page called the 2015 British guideline the most recent and said no newer guideline was found. The British Society for Haematology published an updated guideline in 2025, and a United States expert panel published one in 2023. (Second accuracy pass; Crossref and Europe PMC records for 10.1111/bjh.70092 and 10.1016/j.ctrv.2023.102603) · Both are now listed as sources. The 2025 guideline could not be read and the 2023 one was read as an abstract, and the page says so.
- 2026-11-16 · Tumour lysis syndrome: name of the 2008 international guideline: The page called the 2008 Journal of Clinical Oncology guideline "Cairo 2008". Its first author is Coiffier, and Cairo is the last author. (Second accuracy pass; Europe PMC and Crossref records for 10.1200/JCO.2007.15.0177) · Renamed to Coiffier 2008 throughout.
- 2026-11-16 · Tumour lysis syndrome: when rasburicase cannot be given: The step said to use allopurinol and hydration instead, and said rasburicase can cause severe haemolysis and methaemoglobinaemia in these patients. The source recommends oral allopurinol, hydration and urine alkalinisation, and says only that these are disorders that can cause haemolytic anaemia. (Second accuracy pass; Italian consensus conference (Haematologica 2008), open PDF fetched raw and read as text) · The action now includes urine alkalinisation and the reason is reworded to the source.
- 2026-11-16 · Tumour lysis syndrome: reasons given for four steps: Four reasons said more than their sources: that the cheaper drug is enough at lower risk, that urine flow clears potassium, that monitoring is timed to when the syndrome usually develops, and that treatment protects the kidneys. The sources give none of these. (Second accuracy pass; Coiffier 2008 abstract and the Italian consensus PDF read as raw text) · Each reason now says only what its source says. The hydration reason names uric acid and phosphate, and the monitoring reason says the schedule rests on good clinical practice.
- 2026-11-16 · Tumour lysis syndrome: rasburicase dosing in use and the venetoclax variant: The page said lower and single fixed doses of rasburicase are in use, which no source read states, and it wrote one clinician's handling of a single high-risk patient on venetoclax as a general rule. (Second accuracy pass; Italian consensus PDF and Gribben 2020 full text from Europe PMC) · The dosing sentence now says only that lower doses were under evaluation in 2008, and the venetoclax variant is attributed to the case described.
- 2026-11-16 · Suspected febrile neutropenia (neutropenic sepsis): the first hour: grade for the evidence on antibiotic timing: The sentence quoted to show that NICE rated the timing evidence very low was a real sentence from the wrong section: NICE wrote it about training for staff. The timing section says the same grade in different words. (Second accuracy pass; NICE CG151 full guideline PDF, read as raw text) · Quote replaced with the sentence from the timing section; the grade itself (very low) was right and stays.
- 2026-11-16 · Suspected febrile neutropenia (neutropenic sepsis): the first hour: the Cardiff and Vale variant: The page said the Cardiff policy suspects neutropenic sepsis after treatment in the last 6 weeks with a temperature over 38 °C or under 36 °C. The policy says unwell, or signs or symptoms of infection, or such a temperature, so a temperature is not required. (Second accuracy pass; Cardiff and Vale UHB 529 policy PDF, read as raw text) · Variant reworded to the three alternatives the policy gives.
- 2026-11-16 · Suspected febrile neutropenia (neutropenic sepsis): the first hour: cultures and central lines: The page said to take a peripheral culture as well as a line culture and that this shows whether the line is the source. NICE asks for an additional peripheral blood culture if clinically feasible, to help find the cause, and does not mention a line culture. (Second accuracy pass; NICE CG151 recommendations page and full guideline, read as raw text) · Step and summary reworded; "if clinically feasible" added.
- 2026-11-16 · Suspected febrile neutropenia (neutropenic sepsis): the first hour: aminoglycosides and glycopeptides: The page said NICE advises against a routine glycopeptide in general and gave line infection and resistant organisms as the exceptions. NICE's glycopeptide recommendation is about people with a central venous access device, and it gives no examples of the exceptions. (Second accuracy pass; NICE CG151 recommendations page, read as raw text) · Step and summary reworded to NICE's scope; the examples removed.
- 2026-11-16 · Suspected febrile neutropenia (neutropenic sepsis): the first hour: escalation after one hour: The step left out one of the three actions NICE lists when a high-risk patient does not respond within 1 hour of any intervention (informing the responsible consultant) and described the trigger as the first hour of antibiotics and fluids. (Second accuracy pass; NICE NG253 guideline PDF (last updated 23 September 2026), read as raw text) · All three actions now given, with NICE's own trigger.
- 2026-11-16 · Suspected febrile neutropenia (neutropenic sepsis): the first hour: reasons given for steps: Several reasons said more than the cited source: that steroids or paracetamol can hide a fever, that most episodes start at home, that cultures after the first dose are less likely to grow, that the antibiotic is the one step that changes the outcome, and that the start time should be recorded. None is in the source cited for its step. (Second accuracy pass; NICE CG151, NICE NG253 and SEOM 2018, read as raw text) · Each reason reworded to what its source says; the instruction to record the start time removed; "intravenous" removed from the NICE step, which says only empiric antibiotic therapy.
- 2026-11-16 · Suspected febrile neutropenia (neutropenic sepsis): the first hour: MASCC score and CISNE figures: The page said a MASCC score under 21 means the standard inpatient pathway and 21 or more means going home on oral antibiotics; the MASCC statement says standard neutropenic sepsis management and likely to be suitable for outpatient or ambulatory emergency care. It also gave the CISNE accuracy figures as if for all 1,133 patients, when they are for the validation subset. (Second accuracy pass; MASCC 2026 statement (Europe PMC full text) and the CISNE 2015 abstract) · Wording changed to the sources' own terms.
- 2026-11-16 · Fever and neutropenia in children with cancer or after stem cell transplant: waiting for the neutrophil count: The page said some centres give the first antibiotic only once the count confirms neutropenia and others give it before. The guideline does not describe what centres do. It reports that some panel members preferred to give antibiotics while the count is awaited and others preferred to wait, because antibiotics carry resistance, cost and toxicity. (Second accuracy pass; International Pediatric Fever and Neutropenia Guideline 2023, full text from Europe PMC read as raw text) · The variant now attributes the two positions to panel members, as the guideline does.
- 2026-11-16 · Fever and neutropenia in children with cancer or after stem cell transplant: knowledge gaps: The page said the panel named early stopping of antibiotics in high-risk children with a respiratory virus as a knowledge gap. The panel names early stopping in high-risk fever and neutropenia in general. The respiratory virus trial is the one it decided not to base a recommendation on. (Second accuracy pass; International Pediatric Fever and Neutropenia Guideline 2023, full text from Europe PMC read as raw text) · The gap now says what the panel says.
- 2026-11-16 · Fever and neutropenia in children with cancer or after stem cell transplant: best in class: The page called the guideline the only current guideline built on paediatric trials alone. The guideline says it is founded on direct paediatric cancer or transplant data. It does not claim to be the only one, and its earlier editions drew on observational studies as well as trials. (Second accuracy pass; International Pediatric Fever and Neutropenia Guideline 2023, full text from Europe PMC read as raw text) · The statement now says the guideline is founded on data from children with cancer or a transplant.
- 2026-11-16 · Fever and neutropenia in children with cancer or after stem cell transplant: prolonged fever steps: The CT and antifungal steps were triggered by fever lasting 96 hours or more. The guideline's recommendations are for fever with neutropenia lasting 96 hours or more in children at high risk of invasive fungal disease. The step on fungal blood tests also showed one grade for two tests that the guideline grades differently. (Second accuracy pass; International Pediatric Fever and Neutropenia Guideline 2023, full text from Europe PMC, and the Children's Oncology Group endorsement PDF, both read as raw text) · The steps and summary now say fever and neutropenia, and the step gives both grades: moderate-quality evidence for PCR and low-quality for beta-D-glucan.
- 2026-11-16 · Fever and neutropenia in children with cancer or after stem cell transplant: reasons given for six steps: Six reasons said more than the sources: that most fevers start at home, that a child can get worse within hours, that germs sit in one channel of a line and not another, that extra antibiotics do not help the stable child, that fever alone often settles, and that trials showed early stopping was as safe as continuing. None of these sentences is in the guideline or in NICE CG151. (Second accuracy pass; International Pediatric Fever and Neutropenia Guideline 2023 from Europe PMC and the NICE CG151 recommendations page, both read as raw text) · Each reason now gives what the source says, and where the 2023 paper gives no reasoning the page says so.
- 2026-11-16 · Management of chemotherapy extravasation by drug class: surgery after 10 days: The step on not debriding an ulcer also said to refer for surgery where necrosis or pain persists beyond 10 days, and cited the Japanese guideline for it. The Japanese guideline says nothing about 10 days. The figure is from the West Midlands network guideline. (Second accuracy pass; Japanese guideline full text read as Europe PMC XML and West Midlands guideline PDF read as text) · The 10 day referral is now its own step, cited to the West Midlands guideline with its sentence quoted.
- 2026-11-16 · Management of chemotherapy extravasation by drug class: strength of the ONS/ASCO recommendations: The page said the full ONS/ASCO guideline could not be read and gave its recommendations without their strength. The full text now reads on PubMed Central. It makes one strong recommendation (use an antidote where one exists) and four conditional ones (hyaluronidase for taxanes, a compress, a compress for more than 24 hours, surgical referral after a central line leak), all on very low certainty evidence. (Second accuracy pass; ONS/ASCO guideline full text read as HTML on PubMed Central) · The source is marked as read in full, the five graded recommendations are quoted, and the summary and best in class statement give each strength.
- 2026-11-16 · Management of chemotherapy extravasation by drug class: surgical referral: The step said to refer promptly after any central line extravasation of a vesicant and to consider referral for DNA-binding vesicants, large volumes and grade 2 symptoms. The guideline suggests referral or escalation of care, as a conditional recommendation, for central line leaks of vesicants or irritants with vesicant properties, and names those groups as the likeliest to benefit. For peripheral lines it leaves referral to clinical judgement. (Second accuracy pass; ONS/ASCO guideline full text read as HTML on PubMed Central) · Step reworded to the recommendation, with its grade and the guideline's own sentence.
- 2026-11-16 · Management of chemotherapy extravasation by drug class: sodium thiosulfate for bendamustine: The page said to give sodium thiosulfate for bendamustine extravasation. The full guideline says to consider it for bendamustine, and recommends it outright only for more than 20 ml of concentrated cisplatin. (Second accuracy pass; ONS/ASCO guideline Table 4 read as HTML on PubMed Central) · Step, summary and best in class statement now say consider for bendamustine.
- 2026-11-16 · Management of chemotherapy extravasation by drug class: hyaluronidase timing: The page gave the ONS/ASCO window for hyaluronidase as within 2 to 4 hours. That is the algorithm's figure. The antidote table in the full guideline says within 3 to 4 hours. (Second accuracy pass; ONS/ASCO algorithm PDF read as an image and full text read on PubMed Central) · Both figures are now stated in the UK variant and in the gaps.
- 2026-11-16 · Management of chemotherapy extravasation by drug class: dexrazoxane dose cap and infusion site: The step said no single dexrazoxane dose above 2000 mg and to infuse into a large vein of another limb. The label caps a single dose at 2000 mg for patients with a body surface area over 2 m2, and says a large vein of an extremity or area other than the affected one. (Second accuracy pass; Savene Summary of Product Characteristics, EMA PDF read as text) · Both steps reworded to the label.
- 2026-11-16 · Management of chemotherapy extravasation by drug class: who the sources cover: The page said children are outside every source read. The ONS/ASCO guideline is for adults and the West Midlands guideline for patients over 16 in adult services, but the Japanese guideline states no age range. (Second accuracy pass; all three guidelines read in full) · Population reworded to what each source says.
- 2026-11-16 · Management of chemotherapy extravasation by drug class: compress for unlisted drugs: The step applied the ONS advice on choosing a compress to irritants and non-vesicants. The passage is about agents the guideline does not name for cold or warm, within its scope of vesicants and irritants with vesicant properties. (Second accuracy pass; ONS Voice summary read as HTML and full text on PubMed Central) · Step reworded, and the West Midlands rule of a dry cold compress for non-vesicants noted.
- 2026-11-16 · Management of chemotherapy extravasation by drug class: UK flush-out referral: The UK variant said vesicant and irritant extravasations are referred for a saline flush-out. The flowchart shows that referral as one branch for a peripheral cannula leak, and the text says to consult the team where relevant. (Second accuracy pass; West Midlands guideline PDF read as text and the flowchart page as an image) · Variant reworded.
- 2026-11-16 · Superior vena cava obstruction in cancer: urgent stenting step and summary: The page told teams to treat the vein urgently with a stent for airway compromise, laryngeal oedema or reduced consciousness, with thrombolysis if there is clot, and said the review reserves stenting for those cases after complete staging. The review says direct opening by stenting and angioplasty with thrombolysis should be considered, gives acute central airway obstruction, severe laryngeal oedema and coma from cerebral oedema as examples, and puts it before the diagnostic work-up, not after staging. (Second accuracy pass; Annals of Palliative Medicine 2024 review, full text read as raw text from the journal's open page) · Step, summary and best-in-class statement reworded to the review's own terms and order.
- 2026-11-16 · Superior vena cava obstruction in cancer: steroids before a biopsy: The page said no source read addressed whether steroids before a biopsy can obscure a diagnosis. The Annals of Palliative Medicine review, read in full for the first time, says steroids should be given after pathological confirmation because they can reduce the diagnostic sensitivity of a biopsy, especially in blood cancers. (Second accuracy pass; Annals of Palliative Medicine 2024 review, full text read as raw text from the journal's open page) · A step with the review's sentence was added, and the gap now states the difference from the Scottish guideline, which considers dexamethasone immediately.
- 2026-11-16 · Superior vena cava obstruction in cancer: graded guidance: The page presented NICE's two ungraded 2005 recommendations as the only national guidance and had not read SIGN 137. SIGN 137 (February 2014) gives a grade B recommendation for chemotherapy or radiotherapy first in small-cell lung cancer with a stent for relapse or persistent obstruction, and a good practice point that a stent may be considered as primary treatment in non-small-cell lung cancer. (Second accuracy pass; SIGN 137 PDF from sign.ac.uk, section 10.3 read with pdftotext and the page image for the grade symbols) · SIGN 137 added as a source, a variant and two grade rows, with the note that SIGN says some of its recommendations may be out of date.
- 2026-11-16 · Superior vena cava obstruction in cancer: when the Scottish guideline stents: The page said the Scottish palliative guideline considers a stent for severe symptoms or failure of other treatment. It says only to consider referral for stenting, thrombectomy, thrombolysis and anticoagulation if the patient's condition permits. (Second accuracy pass; Scottish Palliative Care Guidelines page, fetched raw and read as text) · Best-in-class statement corrected.
- 2026-11-16 · Superior vena cava obstruction in cancer: time limits: The page said assessment is needed the same day, told readers to ring 999, and said a stent relieves symptoms within days. No source read gives a time limit or a number of days: the Scottish guideline says to refer urgently, and the reviews say a stent may give relief more rapidly or immediately. (Second accuracy pass; all five original sources re-read as raw text) · Reworded to urgent assessment and quick relief, and the missing time limit is named as a gap.
- 2026-11-16 · Superior vena cava obstruction in cancer: signs, imaging and reasons: The summary called stridor a late sign, where the Scottish guideline marks only papilloedema as late. The imaging step cited the vascular review for a chest X-ray it does not mention. Several reasons went beyond the sources: that steroids are widely given and have side effects, that supportive drugs are for breathlessness and distress, and that steroids are unsupported in every source when two do not mention them. (Second accuracy pass; Scottish Palliative Care Guidelines page and Sen 2026 full text (Europe PMC), read as raw text) · Each statement reworded to what its source says.
- 2026-11-16 · Infusion and hypersensitivity reactions during anticancer treatment: positioning the patient: The step said to keep the patient lying flat "or sitting if breathing is easier". Resuscitation Council UK says semi-recumbent, not sitting, and warns that sitting up suddenly can be fatal. The page also left out that ESMO advises differently: Trendelenburg for low blood pressure and sitting up for respiratory distress, graded [V, C]. (Second accuracy pass; Resuscitation Council UK May 2021 PDF read with pdftotext, and the ESMO 2017 guideline read from an Internet Archive capture of the journal page) · Step reworded to semi-recumbent with no standing, walking or sudden sitting up, and the ESMO and RCUK positions set side by side in the summary and best-in-class statement.
- 2026-11-16 · Infusion and hypersensitivity reactions during anticancer treatment: ESMO on antihistamines and corticosteroids: The page said ESMO "recommends" H1 plus H2 antihistamines [I, B] and "allows" corticosteroids [V, D]. ESMO grades [I, B] the finding that the combination is superior to either alone and says the drugs "may be given" [V, C]. Grade D in its scheme means generally not recommended, and the sentence it grades [V, D] says corticosteroids prevent biphasic reactions but are not critical. (Second accuracy pass; ESMO 2017 guideline and its Table 8, read from an Internet Archive capture of the journal page) · Summary, best-in-class statement and variant reworded to the source's own terms, with grade D explained and the conditional doses stated.
- 2026-11-16 · Infusion and hypersensitivity reactions during anticancer treatment: grades in the best-in-class statement: The statement attached ESMO's [IV, B] to repeating adrenaline after 5 minutes and wrote "no rechallenge except by desensitisation [V, B]". ESMO's [IV, B] is for the intramuscular dose, its repeat interval is every 5 to 15 minutes at [V, C], and its [V, B] statement against rechallenge carries no exception. ESMO describes desensitisation for patients whose reactions recur despite premedication. (Second accuracy pass; ESMO 2017 guideline read from an Internet Archive capture of the journal page) · Each grade moved to the statement it belongs to, the exception removed, and desensitisation described in ESMO's scope in the summary and variant.
- 2026-11-16 · Infusion and hypersensitivity reactions during anticancer treatment: reasons given beside five steps: Five reasons said more than the sources. Adrenaline was called "the only drug" and delay "the main avoidable cause of death", where RCUK says the most important drug and that delay is associated with fatal outcomes. The page also said the next dose after anaphylaxis "could be fatal", that cytokine-release reactions become milder, that a missed steroid dose raises the chance of a reaction, and that tryptase and the allergy clinic decide on desensitisation. None of those is in the cited text. (Second accuracy pass; Resuscitation Council UK May 2021 PDF and ESMO 2017 guideline, both read as raw text) · Each reason rewritten to what the cited passage says.
- 2026-11-16 · Infusion and hypersensitivity reactions during anticancer treatment: ESMO sources and ranitidine: The page did not mention that a correction notice to the ESMO guideline was published in 2018 (Annals of Oncology 29, supplement 4, page iv260). It also said ranitidine "has since been withdrawn in several countries" with no source. One ESMO quotation had a stray space and the ESMO adrenaline site was given as anterolateral thigh where ESMO prints lateral thigh. (Second accuracy pass; Europe PMC and Crossref records for the correction notice, which itself could not be read, and the European Medicines Agency ranitidine referral page read raw) · Correction notice added as a source that could not be read, with a gap saying its content is unknown. Ranitidine statement now cites the EMA suspension in the EU in 2020. Quotation and injection site corrected.
- 2026-11-17 · Infusion and hypersensitivity reactions during anticancer treatment: what ASCO-ONS standard 3.12.7 covers: The summary cited ASCO-ONS standard 3.12.7 for keeping resuscitation equipment at hand. That standard asks for a hypersensitivity management policy and for order sets and medicines to be accessible in time; it does not mention equipment. (Second accuracy pass; ASCO-ONS 2024 standards, full text read from the Oncology Nursing Society's open PDF) · The equipment advice is now cited to ESMO alone and the standard is described in its own sentence.
- 2026-11-16 · Raised intracranial pressure in cancer: dexamethasone step: The step said to start dexamethasone with a proton pump inhibitor and gave reasons (sleep, ulcers) the source does not give. The Scottish guideline says to consider a trial of corticosteroids with gastroprotection, usually a proton pump inhibitor, with the second dose before 2 pm if possible, and marks the use as off-label. (Second accuracy pass; Scottish Palliative Care Guidelines page re-fetched and read as raw text) · Reworded to a trial to be considered, with the route, "if possible" and the off-label marking; the unsourced reasons removed.
- 2026-11-16 · Raised intracranial pressure in cancer: steroids without symptoms: The page said the American guideline found insufficient evidence for people without symptoms or mass effect. Its statement covers only people with no symptoms and no mass effect. The page also said NICE offers dexamethasone only to symptomatic people; NICE does not say only. (Second accuracy pass; CNS 2019 abstract from Europe PMC and NICE NG122 chapter, read as raw text) · Reworded in the summary, the best-in-class statement and the variant; the European guideline's own statement added.
- 2026-11-16 · Raised intracranial pressure in cancer: European guideline: The page said the EANO-ESMO 2021 guideline could not be read, that no graded guideline addressed raised pressure, and that no source addressed surgery. The guideline says to use the lowest dose of steroids for the shortest time, to consider steroids only in symptomatic patients, to image anyone with symptoms or signs of raised pressure and to consider surgery for acute symptoms. (Second accuracy pass; typeset article read in full from the Sapienza University of Rome repository with pdftotext) · Source marked as read; three steps, three graded statements and its scope added; best-in-class statement and gaps rewritten.
- 2026-11-16 · Raised intracranial pressure in cancer: when to call: The patient summary said a same-day call was needed and that a scan is usually done and dexamethasone started. No source gives a time for contact; the Scottish guideline says a scan may be necessary and a steroid trial should be considered. New confusion was listed where the source lists personality change. (Second accuracy pass; Scottish Palliative Care Guidelines page read as raw text) · Reworded to urgent contact, a scan that may be needed and a steroid that is often tried; the missing timing named in gaps.
- 2026-11-16 · Raised intracranial pressure in cancer: reasons given for steps: Seven reasons said more than their source: pressure rising when lying flat, a scan only where it would change management, blood thinners worsening bleeding, a list of steroid side effects, sitting up lowering pressure, named treatments after referral, and a dose comparison credited to the wrong guideline. (Second accuracy pass; each cited passage re-read in the raw source text) · Each reason reworded to what its source says; the side-effect list removed.
- 2026-11-16 · Raised intracranial pressure in cancer: cause and the dose trial: The summary said raised pressure is usually from brain metastases or a primary tumour with oedema; the source says it may be caused by the cancer or a bleed into or around the tumour. The variant said 4 or 8 mg a day matched 16 mg; the trial's authors drew that conclusion for 4 mg only. (Second accuracy pass; Scottish guideline page and Vecht 1994 abstract from Europe PMC) · Summary reworded to the source's sentence; the trial's day 7 figures given as printed.
- 2026-11-16 · Neutropenic enterocolitis (typhlitis): imaging, cultures, antibiotics and conservative care: antibiotic step and its grade: The step gave grade BIII beside the sentence on how long to continue antibiotics. The guideline prints BIII for the choice of antibiotic and prints no grade for the duration sentence. (Second accuracy pass; AGIHO/DGHO 2017 guideline, full-text XML from Europe PMC (PMC5748412) read as raw text) · Split into two steps: the choice of antibiotic with its own quotation and grade BIII, and the duration with its quotation and no grade. The summary now says the duration is stated without a grade.
- 2026-11-16 · Neutropenic enterocolitis (typhlitis): imaging, cultures, antibiotics and conservative care: patient summary: The patient summary said the bowel is rested and that antibiotics are given through a drip. The guideline's conservative therapy is a bland diet, hydration and pain treatment, and it does not state a route for the antibiotics. (Second accuracy pass; AGIHO/DGHO 2017 guideline, full-text XML from Europe PMC (PMC5748412) read as raw text) · Changed to antibiotics, fluids, pain relief and a bland diet. Bowel rest and parenteral nutrition now appear as a variant, attributed to the 2025 cohort paper that describes them.
- 2026-11-16 · Neutropenic enterocolitis (typhlitis): imaging, cultures, antibiotics and conservative care: reasons given for four steps: Four reasons said more than the guideline does: that bacteria often enter the blood, that most episodes settle as the blood count recovers, that fungal infection is easily missed, and that surgery is risky and kept for a failing bowel. The guideline says none of these. (Second accuracy pass; AGIHO/DGHO 2017 guideline, full-text XML from Europe PMC (PMC5748412) read as raw text) · Each reason reworded to the guideline's own passage, including its pooled 6.2% frequency of fungal infection.
- 2026-11-16 · Neutropenic enterocolitis (typhlitis): imaging, cultures, antibiotics and conservative care: best in class statement: It said every recommendation is expert opinion because there are no trials, and named a carbapenem until the neutropenia recovers. The guideline defines level III more widely, does not say there are no trials, names imipenem-cilastatin or meropenem, and says until clinical signs and neutropenia resolve. (Second accuracy pass; AGIHO/DGHO 2017 guideline, full-text XML from Europe PMC (PMC5748412) read as raw text) · Reworded to the guideline's level III definition, the named drugs and both conditions for stopping; the unsupported reason removed.
- 2026-11-16 · Neutropenic enterocolitis (typhlitis): imaging, cultures, antibiotics and conservative care: summary figures: The summary called high-dose cytarabine with anthracyclines the main risk factor and dropped part of the population behind the 5.3% incidence. The guideline says a major risk factor, and high-dose chemotherapy for solid tumours. (Second accuracy pass; AGIHO/DGHO 2017 guideline, full-text XML from Europe PMC (PMC5748412) read as raw text) · Changed to a major risk factor and added for solid tumours.
- 2026-11-16 · Neutropenic enterocolitis (typhlitis): imaging, cultures, antibiotics and conservative care: growth factor variant: The variant was named for G-CSF and said it is given to speed neutrophil recovery. The guideline says only that haematopoietic growth factors might be considered and does not name G-CSF or a purpose. (Second accuracy pass; AGIHO/DGHO 2017 full text and the Clinical Infectious Diseases cohort full text (PMC13017227), both from Europe PMC as raw text) · Renamed to haematopoietic growth factors, with G-CSF attributed to the 2025 cohort paper, which describes it as part of usual medical treatment.
- 2026-11-16 · Neutropenic enterocolitis (typhlitis): imaging, cultures, antibiotics and conservative care: gaps: A gap said no randomised trial has tested any treatment, the duration of antibiotics or the timing of surgery. Neither source says this. The guideline reports no studies on added metronidazole or vancomycin and no evidence for growth factors, and the cohort paper says management relies mainly on expert opinion and that surgery is controversial. (Second accuracy pass; AGIHO/DGHO 2017 full text and the Clinical Infectious Diseases cohort full text (PMC13017227), both from Europe PMC as raw text) · Gap reworded to what each source states.
- 2026-11-16 · Hyperleukocytosis and leukostasis in acute leukaemia: the ASFA apheresis guideline: The page said the ASFA tenth edition categorises leukocytapheresis for hyperleukocytosis but that its category could not be read, and listed no graded statement at all. The full text prints Category III, Grade 2B, says no randomised trial has been done and says arranging apheresis should not delay chemotherapy. The ELN 2019 recommendations say leukapheresis should be avoided in acute promyelocytic leukaemia. (Second accuracy pass; ASFA tenth edition and ELN 2019 APL recommendations read in full as HTML from PubMed Central) · The category and grade are now stated as printed, and two steps were added, one from each guideline.
- 2026-11-16 · Hyperleukocytosis and leukostasis in acute leukaemia: uric acid and hydroxyurea steps: Two steps were firmer than the source. The article says high uric acid from tumour lysis can be addressed with allopurinol or rasburicase depending on the uric acid level and institution guidelines, and that hydroxyurea can be used before induction chemotherapy in eligible patients. The page told the reader to give both as a rule. (Second accuracy pass; ATS Scholar 2025 full text re-read as XML from Europe PMC) · Both steps and the summary now carry the source's conditions.
- 2026-11-16 · Hyperleukocytosis and leukostasis in acute leukaemia: symptoms and reasons: The trigger and the short description listed breathlessness, confusion, headache and visual change, and several reasons gave mechanisms (potassium and phosphate release, kidney protection, DIC raising the risk of bleeding in the brain) that the text read does not state. The article names the lungs, kidneys and brain and gives no symptom list in its text. (Second accuracy pass; ATS Scholar 2025 full text re-read as XML from Europe PMC) · The trigger, short description and reasons now say only what the article says.
- 2026-11-16 · Hyperleukocytosis and leukostasis in acute leukaemia: leukapheresis and red cell exchange: The page said the Journal of Hospital Medicine review calls leukapheresis optional and that the sources agree on correcting DIC and on ATRA. The abstract calls the role of leukapheresis controversial and does not mention DIC or ATRA. The red cell exchange entry gave no numbers. (Second accuracy pass; two abstracts re-read raw from Europe PMC and the Blood 2024 letter read in full as HTML from PubMed Central) · The best-in-class statement was rewritten source by source, and the red cell exchange variant now carries the study's size, design and the authors' own caution.
- 2026-11-16 · Hyperleukocytosis and leukostasis in acute leukaemia: why AML more than ALL: The summary said leukostasis is commoner in AML because myeloid blasts are larger. The article says it is in part because of the size of myeloid blasts. (Second accuracy pass; ATS Scholar 2025 full text re-read as XML from Europe PMC) · Reworded to the source.
- 2026-11-16 · Hyponatraemia and SIADH in cancer: re-lowering after over-correction: The step and summary said the Society for Endocrinology advises hypotonic fluid with or without desmopressin. Its text says "with or without concurrent anti-diuresis" and does not name a drug; it is the Korean 2022 guideline that names 5% dextrose and desmopressin. The step also called osmotic demyelination permanent, which the cited source does not say. (Second accuracy pass; Society for Endocrinology 2016 and Korean Society of Nephrology 2022, full text from Europe PMC read as raw text) · Step reworded to an antidiuretic and a serious injury to the brain; the summary now attributes desmopressin to the Korean guideline.
- 2026-11-16 · Hyponatraemia and SIADH in cancer: European bolus regimen and the SALSA figures: The summary said the European bolus of 150 mL over 20 minutes is repeated two to three times, which reads as three or four doses; the Korean guideline's table says two to three doses as needed. The SALSA overcorrection rates were given without the trial's definition (a rise above 12 mmol/L in 24 hours or 18 mmol/L in 48 hours), which is looser than the 10 mmol/L limit on the page, and the Korean recommendation was said to rest on SALSA alone when it also cites a cohort study. (Second accuracy pass; Korean Society of Nephrology 2022 from Europe PMC and SALSA 2021 full text from PubMed Central, read as raw text) · Regimen reworded to two to three times as needed; the trial's definition and the cohort study added; a gap now sets out the three different definitions of over-correction.
- 2026-11-16 · Hyponatraemia and SIADH in cancer: bolus step scope: The step told readers to give boluses rather than a continuous infusion without limit. The Korean recommendation is for symptomatic severe hyponatraemia; for moderate symptoms it allows either method. (Second accuracy pass; Korean Society of Nephrology 2022, full text from Europe PMC read as raw text) · Step now begins "For severe symptoms".
- 2026-11-16 · Hyponatraemia and SIADH in cancer: SIADH treatment step: The step added "a vaptan such as tolvaptan, with close sodium checks" under the Korean guideline, which does not name tolvaptan for SIAD or give that monitoring instruction. (Second accuracy pass; Korean Society of Nephrology 2022, full text from Europe PMC read as raw text) · Step reduced to what the guideline says: oral salt with a loop diuretic, or a vaptan, as second-line options.
- 2026-11-16 · Hyponatraemia and SIADH in cancer: brain disease step: The step said to avoid fluid restriction in people with a brain tumour or brain metastases. The Korean guideline lists brain tumour among cerebral diseases, does not name metastases, and says clinicians should avoid volume depletion through water restriction. (Second accuracy pass; Korean Society of Nephrology 2022, full text from Europe PMC read as raw text) · Step reworded to a brain tumour or other brain disease, and to avoiding drying the person out with fluid restriction.
- 2026-11-16 · Hyponatraemia and SIADH in cancer: finding the cause: Three explanations went beyond the cited passages: that a moderate sodium can cause fits, that the drug history is reviewed (cited to a source that does not say it), and that SIADH and fluid loss need opposite treatments. (Second accuracy pass; Society for Endocrinology 2016 from Europe PMC and The Oncologist 2012 full text from PubMed Central, read as raw text) · Reworded to serious brain symptoms; the medicines review moved to the step citing The Oncologist, which states it; "opposite" replaced by "treated differently", with the review's point that the wrong treatment can worsen the sodium.
- 2026-11-16 · Hyponatraemia and SIADH in cancer: best in class and vaptan variant: The best-in-class statement said the guidelines agree on limits of 10 mmol/L in 24 hours and 18 mmol/L in 48 hours; the Korean guideline aims for 4 to 6 mmol/L at first and says not to correct by 10 mmol/L or more in a day, without an 18 mmol/L limit. The vaptan variant said "a response in all" where the review says an initial response, and said vaptans are used "often during chemotherapy", which no text then read stated. (Second accuracy pass; Korean Society of Nephrology 2022 from Europe PMC, the 2026 scoping review abstract, and The Oncologist 2012 full text from PubMed Central) · Each source's own figures given; "initial response" restored; the chemotherapy point reworded to what The Oncologist review says about cisplatin hydration.
- 2026-11-16 · Malignant pericardial effusion: warning signs in the short summary and the trigger: The page listed a fast heartbeat, feeling faint and low blood pressure as presenting signs. Neither readable source states them. The 2024 review lists breathlessness (90%), chest pain, swelling and raised neck veins, and describes tamponade as haemodynamic compromise. (Second accuracy pass; CJC Open 2024 systematic review, full text XML from Europe PMC read raw) · The short summary, the one-line version and the trigger now name only the signs the review states.
- 2026-11-16 · Malignant pericardial effusion: reasons given for the steps: Four reasons said more than the sources. "Draining even some fluid relieves it", "only treating the cancer stops the fluid being made", cytology "guides treatment", and planning for "time at home" are in neither source. The step on tamponade at presentation gave a figure from 41 case reports as a general fact. (Second accuracy pass; CJC Open 2024 review and the PMAP trial, full text XML from Europe PMC read raw) · Each reason now says what its source says. The tamponade figure is scoped to published case reports.
- 2026-11-16 · Malignant pericardial effusion: what the PMAP trial is first at: The page called PMAP the first randomised comparison of procedures. Its authors claim less: the first randomised trial to show balloon pericardiotomy superior to pericardiocentesis for recurrence. The page also left out that median survival was 88 days after balloon pericardiotomy against 180 days after pericardiocentesis (P = .238), that procedural success was 92% against 100%, and that 45 of 130 people screened were excluded for needing urgent drainage out of hours. (Second accuracy pass; PMAP trial, full text XML from Europe PMC read raw) · The claim is reworded to the authors' own, and the survival, success and exclusion figures are added.
- 2026-11-16 · Malignant pericardial effusion: cytology and open questions: The page gave cytology a sensitivity of 92% without saying that this is the review quoting the 2015 ESC guideline, or that in the review's own 41 cases cytology found malignant cells in 39%. Two open questions were stated as conclusions no source draws: that a larger trial is needed before balloon pericardiotomy is standard, and that drain duration is not settled. (Second accuracy pass; CJC Open 2024 review and the PMAP trial, full text XML from Europe PMC read raw) · The 92% figure is attributed and the 39% figure added. The open questions now state the authors' own limitations and that neither source gives a drain duration.
- 2026-11-16 · Malignant bowel obstruction in advanced cancer: the stent step: The page said a stent has lower early death rates than emergency surgery. MASCC 2022 reports two pooled analyses: one found more deaths within 30 days after surgery, the other, of randomised trials only, found no difference in deaths and a shorter hospital stay with a stent. (Second accuracy pass; MASCC 2022 full text from Europe PMC, read as raw text) · The reason now gives both findings and says the blockage returns more often after a stent.
- 2026-11-16 · Malignant bowel obstruction in advanced cancer: who it covers: The page limited the guideline to mechanical obstruction. MASCC 2022 covers functional obstruction as well, where the bowel stops moving because cancer has infiltrated its nerves or mesentery. (Second accuracy pass; MASCC 2022 full text from Europe PMC, read as raw text) · Population rewritten to the source's scope, with its exclusions.
- 2026-11-16 · Malignant bowel obstruction in advanced cancer: fluids in the last days of life: The reason said mouth care helps more than drip fluids. MASCC 2022 does not mention mouth care. (Second accuracy pass; MASCC 2022 full text from Europe PMC, read as raw text) · The unsupported clause was removed.
- 2026-11-16 · Malignant bowel obstruction in advanced cancer: hyoscine butylbromide: The page called hyoscine butylbromide cheaper, said it is probably inferior to octreotide, and told the reader to use it for colic. MASCC 2022 says nothing about cost, says it may be inferior for vomiting, and suggests only that it may reduce abdominal pain (level III, grade D); the trials found no significant difference from octreotide in colicky pain. (Second accuracy pass; MASCC 2022 full text from Europe PMC, read as raw text) · Variant, summary and pain step reworded to the source, and the grade for pain added.
- 2026-11-16 · Malignant bowel obstruction in advanced cancer: strength of several steps: Four steps were firmer than the guideline: ask for a surgical opinion, give a somatostatin analogue, prefer the least invasive operation, and avoid prokinetics in complete obstruction. MASCC 2022 says a surgical opinion should be considered, a somatostatin analogue may reduce vomiting, less invasive operations should be considered, and prokinetics likely should be avoided. (Second accuracy pass; MASCC 2022 full text from Europe PMC, read as raw text) · Each reworded to the strength the source uses.
- 2026-11-16 · Malignant bowel obstruction in advanced cancer: reasons not in the source: Several reasons said things MASCC 2022 does not: that food and tablets make vomiting worse, that a gastrostomy avoids a nasal tube for weeks, that a surgeon should be asked before the person becomes too unwell, and that home parenteral nutrition is for someone well enough to live for months. (Second accuracy pass; MASCC 2022 full text from Europe PMC, read as raw text) · Reworded to what the guideline says, with its own list of signs that home parenteral nutrition may help.
- 2026-11-16 · Malignant bowel obstruction in advanced cancer: octreotide in the 2021 update: The page said MASCC 2021 calls octreotide the drug of choice in non-operable obstruction. The abstract says octreotide remains the drug of choice in managing malignant bowel obstruction, and separately that the panel recommends it in non-operable obstruction. (Second accuracy pass; MASCC 2021 abstract from Europe PMC) · Both statements now given as the abstract words them.
- 2026-11-16 · Malignant bowel obstruction in advanced cancer: a guideline that had not been read: The WSES-PSOGI-SICO 2026 guidelines were listed as found but not read. They read in full: five weak statements, on low or very low certainty evidence, for obstruction from peritoneal surface malignancy. (Second accuracy pass; WSES-PSOGI-SICO 2026 full text from Europe PMC, read as raw text) · Added as a source, a variant and two quoted, graded statements.
- 2026-11-16 · Cytokine release syndrome after CAR T-cell therapy: admitting an outpatient: The step told teams to admit an outpatient at the first sign of fever, low blood pressure or confusion. SITC says admission should be strongly considered. (Second accuracy pass; SITC 2020 guideline, full text from Europe PMC read as raw text) · Step reworded to "strongly consider admission".
- 2026-11-16 · Cytokine release syndrome after CAR T-cell therapy: heart assessment and growth factors: The step and summary said to assess the heart with troponin and an echocardiogram after 72 hours of severe CRS, and to hold G-CSF. SITC says a cardiac assessment should be considered, which may include troponin or an echocardiogram, and says to consider holding G-CSF. (Second accuracy pass; SITC 2020 guideline, full text from Europe PMC read as raw text) · Step and summary reworded to the strength the source gives.
- 2026-11-16 · Cytokine release syndrome after CAR T-cell therapy: when to call the doctor: The list of warning signs left out several that SITC names: a change from baseline in systolic blood pressure, arrhythmia, rising creatinine or liver tests, jerky movements, and values outside the normal range for age in children. (Second accuracy pass; SITC 2020 guideline, full text from Europe PMC read as raw text) · Missing signs added to the step.
- 2026-11-16 · Cytokine release syndrome after CAR T-cell therapy: tocilizumab availability: The step said tocilizumab must be on hand for every patient with staff trained to give it. SITC's sentence is in its section on the COVID-19 pandemic and says adequate staff and supportive care are a prerequisite and that it is important to ensure tocilizumab is available. (Second accuracy pass; SITC 2020 guideline, full text from Europe PMC read as raw text) · Step reworded and its scope stated.
- 2026-11-16 · Cytokine release syndrome after CAR T-cell therapy: tocilizumab in children: The summary read as if tocilizumab "may be considered" at grade 3 in children. SITC says it should be given at grade 3 (LE: 3) and may be given for prolonged grade 2 or intolerance to fever (LE: 4). (Second accuracy pass; SITC 2020 guideline, full text from Europe PMC read as raw text) · Summary reworded and the LE: 4 statement added to the evidence table.
- 2026-11-16 · Cytokine release syndrome after CAR T-cell therapy: reasons given for steps: Several reasons were not in the sources, among them that further tocilizumab doses add risk without benefit (EBMT gives no reason for its four-dose limit) and that third-line drugs act on other parts of the cytokine cascade (siltuximab acts on IL-6 itself). (Second accuracy pass; SITC 2020 and EBMT/JACIE 2019, full texts from Europe PMC read as raw text) · Reasons rewritten to what each source says.
- 2026-11-16 · Cytokine release syndrome after CAR T-cell therapy: early tocilizumab variant: The variant said tocilizumab was given before CRS develops and that the approach is not recommended routinely. SITC describes a trial giving it at the time of fever to children with more than 40% marrow blasts, which reduced grade 4 CRS by one third, and makes no recommendation either way. (Second accuracy pass; SITC 2020 guideline, full text from Europe PMC read as raw text) · Variant rewritten with the source's detail.
- 2026-11-16 · Cytokine release syndrome after CAR T-cell therapy: FDA variant and bispecific scope: The page said the clinical need for on-site tocilizumab had not changed, and that CRS after bispecific antibodies is managed under step-up dosing protocols. Neither statement is in a cited source: the FDA notice covers only the end of the REMS, and ASTCT says only that its definition applies to any immune effector cell engaging therapy. (Second accuracy pass; FDA safety communication of 26 June 2025 and ASTCT 2019 author manuscript, read as raw text) · Both statements replaced with what the sources say.
- 2026-11-16 · Suspected central line infection in a person with cancer: edition status of the IDSA guideline: The page called the 2009 IDSA guideline the current edition. IDSA's own page marks it as archived and says it is currently being updated. No replacement has been published. (Second accuracy pass; IDSA guideline page for the 2009 intravascular catheter guideline, fetched raw and read as text) · The summary, best-in-class statement and gaps now say the guideline is archived and being updated, and that it is quoted as the latest IDSA text.
- 2026-11-16 · Suspected central line infection in a person with cancer: removal for tunnel infection or port abscess: The page said ESMO adds tunnel infection and port abscess to IDSA's reasons for removing a line. IDSA itself says patients with tunnel infection or port abscess require removal of the catheter, with drainage if indicated, graded A-II. (Second accuracy pass; IDSA 2009 full text re-read from PubMed Central as raw text) · The step now cites and quotes IDSA with its A-II grade, and the summary attributes the point to both guidelines.
- 2026-11-16 · Suspected central line infection in a person with cancer: candidaemia: The summary said to suspect candidaemia in septic patients with certain risk factors. IDSA recommends giving empirical treatment for candidaemia to those patients (B-II). (Second accuracy pass; IDSA 2009 full text re-read from PubMed Central as raw text) · Reworded to empirical treatment, with the risk factors as IDSA lists them.
- 2026-11-16 · Suspected central line infection in a person with cancer: keeping the line: The step said to consider keeping the line in all other cases. IDSA says treatment should be attempted without removal only in uncomplicated infection, and that lines infected with Bacillus, Micrococcus or Propionibacteria should generally be removed. The reason given was not IDSA's, which is that many patients have limited access sites. (Second accuracy pass; IDSA 2009 full text re-read from PubMed Central as raw text) · Step reworded and re-quoted from IDSA's salvage recommendation; the summary now names the extra organisms; an unprovable statement that no randomised trial was cited was removed.
- 2026-11-16 · Suspected central line infection in a person with cancer: first antibiotics: The step said to start antibiotics at once. The ESMO sentence it quotes says only that antibiotics should not be delayed until blood culture results are available. Combination cover was also scoped to neutropenic or septic patients, where IDSA adds patients known to be colonised with resistant organisms. (Second accuracy pass; IDSA 2009 re-read as raw text; the ESMO guideline could not be re-read on 16 November 2026 and its quote stands as read on 11 November) · Step reworded to starting without waiting for culture results; colonised patients added; the gaps name what rests on ESMO alone.
- 2026-11-16 · Suspected central line infection in a person with cancer: reasons given for steps: Four explanations said more than IDSA does: that two lumen samples still allow a comparison (IDSA calls their interpretation unresolved), that an exit-site swab identifies the germ before blood cultures grow, that bowel organisms are the concern in neutropenia, and that all the listed infections rarely clear and spread to heart valves or bones (IDSA says this of S. aureus). The grading scheme was also described as the Canadian Task Force scheme when IDSA prints it as the IDSA and US Public Health Service system adapted from it. (Second accuracy pass; IDSA 2009 full text re-read from PubMed Central as raw text) · Each explanation reworded to what the guideline says; grading scheme description corrected.
- 2026-11-16 · Hyperviscosity syndrome in Waldenström macroglobulinaemia and myeloma: the emergency step: The step cited to the BSH guideline said to act the same day. BSH calls hyperviscosity a haematological emergency and sets no time limit, and no source read gives a clock. (Second accuracy pass; BSH 2022 guideline, published PDF read with pdftotext) · Step now says to treat hyperviscosity as a haematological emergency and states that no source sets a limit in hours.
- 2026-11-16 · Hyperviscosity syndrome in Waldenström macroglobulinaemia and myeloma: red cell transfusion: The page said two reviews advise holding back red cell transfusion in stable patients. Weaver 2020 says to take caution to avoid excessive red cell or platelet transfusion, and Gertz 2018 says only that a theoretical argument could be made to withhold it, with no scientific proof. (Second accuracy pass; Weaver 2020 (Europe PMC full-text XML) and Gertz 2018 (PubMed Central page), read as raw text) · Step, summary, best-in-class statement and gaps reworded to avoiding excessive red cell or platelet transfusion, with Gertz's caveat kept.
- 2026-11-16 · Hyperviscosity syndrome in Waldenström macroglobulinaemia and myeloma: the ASFA rating: The summary and best-in-class statement said both guidelines give plasma exchange their top rating. ASFA category I is its first-line category, but grade 1B (strong recommendation, moderate-quality evidence) is not its highest grade, which is 1A. (Second accuracy pass; ASFA Tenth Special Issue 2026, PubMed Central page read as raw text) · Reworded to category I with grade 1B, described as ASFA defines them.
- 2026-11-16 · Hyperviscosity syndrome in Waldenström macroglobulinaemia and myeloma: reasons given for four steps: Four explanations went beyond the cited passage: that the retina shows strain first, that IgG moves back into the blood and rebounds (cited to ASFA, which gives only the intravascular share and fluid shifts), that the IgM flare lasts a few weeks (the ECOG study found it still raised at 2 months in 13 of 22 patients), and that a flare cannot be predicted (cited to BSH, which does not say so; the ECOG study does). (Second accuracy pass; ASFA 2026, BSH 2022 and Gertz 2018 full text, Ghobrial 2004 abstract) · Each explanation cut back to what its source says, and the point about prediction attributed to Ghobrial 2004.
- 2026-11-16 · Hyperviscosity syndrome in Waldenström macroglobulinaemia and myeloma: grade table and the Mayo guideline: A Weaver 2020 quote about when to perform fundoscopy was attached to the claim that retinal signs can be present without visual symptoms and are a reason to treat, which that sentence does not state. The page also said the Mayo mSMART guideline could not be read; it reads today on PubMed Central. (Second accuracy pass; Weaver 2020 full-text XML; mSMART 2016 author manuscript on PubMed Central (PMC5556979), read as raw text) · Quote replaced with the Weaver sentence that states the claim. mSMART added as a source with its two graded recommendations (plasma exchange promptly before cytoreductive therapy, level 3 grade B; rituximab alone best avoided at a very high IgM without plasmapheresis, level 3 grade A).
- 2026-11-16 · ICANS after CAR T-cell therapy: ICE score, ASTCT grading, steroids and seizure management: steroid doses in the summary: The summary gave dexamethasone 0.5 mg/kg (at most 10 mg a dose) and methylprednisolone 2 mg/kg/day as the steroids for ICANS. SITC 2020 prints those doses in its section on cytokine release syndrome. For neurotoxicity it reports the axicabtagene ciloleucel package insert: dexamethasone 10 mg intravenously every 6 hours up to grade 3 and methylprednisolone 1000 mg daily for 3 days for grade 4. (Second accuracy pass; SITC 2020 guideline, full-text XML from Europe PMC read as raw text) · The summary now gives the label doses SITC reports for neurotoxicity, says the mg/kg doses come from the CRS section, and a gap says no product label was read.
- 2026-11-16 · ICANS after CAR T-cell therapy: ICE score, ASTCT grading, steroids and seizure management: how common seizures are: The short summary for patients said seizures happen rarely. SITC 2020 reports non-convulsive status epilepticus in roughly 10% of patients at some centres and generalised seizures in roughly 30% in one study. (Second accuracy pass; SITC 2020 guideline, full-text XML from Europe PMC read as raw text) · Changed "rarely" to "sometimes".
- 2026-11-16 · ICANS after CAR T-cell therapy: ICE score, ASTCT grading, steroids and seizure management: preventive levetiracetam: The step and summary said to give preventive levetiracetam to high-risk patients. SITC 2020 says only that it may be considered for them, and that the evidence is insufficient to recommend it for everyone. (Second accuracy pass; SITC 2020 guideline, full-text XML from Europe PMC read as raw text) · Reworded to "consider" and "may be considered", and added the EBMT/JACIE 2019 exception for a history of seizures or central nervous system disease.
- 2026-11-16 · ICANS after CAR T-cell therapy: ICE score, ASTCT grading, steroids and seizure management: reasons given for four steps: Four reasons said more than the sources: that brain swelling can be fatal within hours, that scans are normal early on, that CD28 products cause more neurotoxicity, and that tocilizumab does not cross into the brain well. The sources say cerebral oedema is rare and can have a fulminant onset, that most patients have unremarkable scans, that neurotoxicity is reported more often and earlier with CD28 products, and that tocilizumab may not penetrate the central nervous system. The monitoring step also said "at least twice daily" where SITC says twice daily. (Second accuracy pass; SITC 2020 and ASTCT 2019, full text read as raw text) · Each reason reworded to what the source says, and "at least" removed.
- 2026-11-16 · IEC-HS after CAR T-cell therapy: recognition and stepwise treatment: first treatment: The step and summary said to start anakinra, with or without corticosteroids, and the patient summary said anakinra and steroids. The ASTCT text says therapy should be initiated with corticosteroids with or without anakinra, and its figure says anakinra with or without corticosteroids, so either drug or both is first line. (Second accuracy pass; ASTCT 2023 (Hines et al.), PMC author manuscript read as raw XML through NCBI efetch) · Reworded to anakinra, corticosteroids or both, and the 48-hour wait now carries the source's where possible.
- 2026-11-16 · IEC-HS after CAR T-cell therapy: recognition and stepwise treatment: tocilizumab: The step said do not give tocilizumab for IEC-HS without CRS features. The ASTCT panel says such use was specifically discouraged and that individual patient considerations may make it appropriate. (Second accuracy pass; ASTCT 2023 (Hines et al.), PMC author manuscript read as raw XML through NCBI efetch) · Action changed to avoid tocilizumab.
- 2026-11-16 · IEC-HS after CAR T-cell therapy: recognition and stepwise treatment: fibrinogen replacement: The reason given said disproportionately low fibrinogen is a hallmark and causes delayed, severe bleeding, citing SITC. SITC says it may possibly be one hallmark and needs close follow-up and cryoprecipitate; the bleeding sentence is not in SITC. (Second accuracy pass; SITC 2020, Europe PMC full-text XML) · Reason reworded to what SITC says.
- 2026-11-16 · IEC-HS after CAR T-cell therapy: recognition and stepwise treatment: SITC advice on anakinra and steroids: The summary and best-in-class statement listed anakinra and steroids for late tocilizumab-refractory toxicity as a SITC recommendation. SITC says these agents may be considered for late-onset pathology which may be tocilizumab-refractory. (Second accuracy pass; SITC 2020, Europe PMC full-text XML) · Reworded to may be considered in both places.
- 2026-11-16 · IEC-HS after CAR T-cell therapy: recognition and stepwise treatment: how common it is: The patient summary and plain description called IEC-HS rare. No cited source gives a frequency; ASTCT speaks of a variable incidence across products and patients. (Second accuracy pass; ASTCT 2023 and SITC 2020, read as raw text) · Rarely changed to sometimes, and rare removed.
- 2026-11-16 · IEC-HS after CAR T-cell therapy: recognition and stepwise treatment: who it applies to: The population said the ASTCT panel notes similar toxicity after bispecific T-cell engagers. The paper only lists engagers among triggers of secondary HLH and names differences between engagers and CAR T cells as future research. (Second accuracy pass; ASTCT 2023 (Hines et al.), PMC author manuscript read as raw XML through NCBI efetch) · Population reworded to those two statements.
- 2026-11-16 · IEC-HS after CAR T-cell therapy: recognition and stepwise treatment: best in class and etoposide: The statement called ASTCT 2023 the only document that defines, grades and treats IEC-HS, which no source says, and the summary gave the etoposide dose without saying the panel reached no clear consensus on its use. Monitoring of kidney and liver tests was given as daily where the source says frequent, for example daily. (Second accuracy pass; ASTCT 2023 (Hines et al.), PMC author manuscript read as raw XML through NCBI efetch; EBMT 2025 and ASTCT 2026 abstracts from Europe PMC) · Claim limited to the sources read in full, the lack of consensus stated, monitoring reworded, and newer EBMT guidance named in the gaps.
- 2026-11-16 · IEC-HS after CAR T-cell therapy: recognition and stepwise treatment: safety switch and tapering: The safety-switch variant said commercial products in the sources do not describe such a switch, and the gaps said the panel reached no consensus on how fast to taper. Neither is in the source, which calls the best approach to tapering unclear. (Second accuracy pass; ASTCT 2023 (Hines et al.), PMC author manuscript read as raw XML through NCBI efetch) · Clause removed and the gap reworded.
- 2026-11-16 · Acute oncology telephone triage: who may triage: The page said a trained nurse triages the call. The UKONS toolkit is written for all staff of any discipline who are trained and assessed as competent, and the UKAOS review found 41 providers using non-clinical staff to answer and triage first in core hours. (Second accuracy pass; UKONS 24-Hour Triage Toolkit, 3rd edition (December 2024), PDF from ukons.org read in full as text, and the UKAOS June 2025 review PDF) · Reworded to trained staff, and added the toolkit's rule that anyone without the training and competency assessment should not triage calls.
- 2026-11-16 · Acute oncology telephone triage: sources: The page said the UKONS toolkit manual was not public and that ukons.org did not list the tool, and it dated version 3 to 2025. UKONS lists both the manual and the poster on its public resources page, and the manual is dated December 2024. (Second accuracy pass; ukons.org public resources page and the toolkit PDF, fetched raw and read with pdftotext) · Added the manual as a source read in full, corrected the year to 2024, and rewrote the gap.
- 2026-11-16 · Acute oncology telephone triage: grading step: The step said each symptom is graded using the current NCI-CTCAE criteria. The tool's sentence about the latest NCI-CTCAE applies to problems the tool does not list; listed symptoms are graded on the tool's own descriptors, which the manual says are based on NCI-CTCAE. (Second accuracy pass; UKONS Triage Assessment Tool version 3 poster and toolkit manual, read as text) · Reworded the step to grade against the tool and to use NCI-CTCAE and further clinical advice for a problem the tool does not list.
- 2026-11-16 · Acute oncology telephone triage: fever step: The step named paracetamol and ibuprofen as medicines that hide a fever, and the short summary said the temperature threshold always applies. The tool says analgesia or steroids, and scopes the threshold to people within 6 to 8 weeks of treatment or immunocompromised. (Second accuracy pass; UKONS Triage Assessment Tool version 3 poster, read as text and as a rendered image) · Reworded to painkillers or steroids, added the tool's advice to arrange a blood count if in doubt, and scoped the threshold.
- 2026-11-16 · Acute oncology telephone triage: who it covers: The page said people who have had cell therapy are covered at any time and that the tool prompts for ICANS and CRS after cell therapy. The tool makes the any-time statement only for immune checkpoint inhibitors, and its ICANS and CRS prompts sit under altered consciousness and headache without naming cell therapy. The manual's scope also includes radiotherapy, bone marrow transplant and teenagers and young adults in adult services. (Second accuracy pass; UKONS toolkit manual and version 3 poster, read as text) · Rewrote the population from the manual's own scope and reworded the prompts.
- 2026-11-16 · Acute oncology telephone triage: chest pain step: The reason given for the chest pain step listed a heart problem, a clot in the lung or a drug reaction. The tool gives no causes; it marks chest pain red at every grade above none. (Second accuracy pass; UKONS Triage Assessment Tool version 3 poster, read as text and as a rendered image) · Replaced the reason with what the tool states.
- 2026-11-16 · Acute oncology telephone triage: digital check-ins variant: The page said the Leicester evaluation found no safety signal. The paper reports two patients who came to hospital on the same day as a green check-in, judged not attributable to the tool, and no other safety signal. (Second accuracy pass; JMIR Cancer 2026 full text XML from Europe PMC) · Reworded to state the two cases and the authors' judgement.
- 2026-11-16 · Immune-related myocarditis from checkpoint inhibitors: second-line treatment: The step and the summary told the reader to add antithymocyte globulin, mycophenolate, abatacept or alemtuzumab if steroids had not worked in 24 hours. SITC 2021 says these "should be considered", at its level of evidence 4. (Second accuracy pass; SITC 2021, full text from Europe PMC read as raw text) · Reworded to "consider adding" in the step and "should be considered" in the summary.
- 2026-11-16 · Immune-related myocarditis from checkpoint inhibitors: infliximab: The page said to avoid infliximab. SITC 2021 and ESC 2022 both say caution is advised against its use in steroid-refractory myocarditis, which is weaker than a prohibition. (Second accuracy pass; SITC 2021 full text from Europe PMC and the 2022 ESC cardio-oncology guideline PDF from a university repository, both read as raw text) · Reworded to "be cautious about using infliximab" and "caution is advised against".
- 2026-11-16 · Immune-related myocarditis from checkpoint inhibitors: best in class and the ESC guideline: The page said SITC 2021 was the only readable guideline with a myocarditis protocol and that the 2022 ESC cardio-oncology guideline could not be read. An open copy of the ESC guideline was found and it differs: methylprednisolone 500 to 1000 mg rather than 1000 mg, steroid resistance judged after 3 days rather than 24 hours, no preferred second-line agent, the checkpoint inhibitor interrupted in every suspected case (Class I, Level C), and ECG, natriuretic peptide and troponin recommended for all patients before treatment (Class I, Level B). (Second accuracy pass; 2022 ESC cardio-oncology guideline, published PDF from the San Raffaele University repository read with pdftotext) · Best in class rewritten to state both guidelines and their grades; two ESC steps, three variants and six ESC grade rows added; the source is now marked as read in full.
- 2026-11-16 · Immune-related myocarditis from checkpoint inhibitors: tests: The step asked for cardiac MRI with right heart catheterisation and biopsy "where needed". SITC 2021 says cardiac MRI "if available (with or without right heart catheterization and myocardial biopsy)". (Second accuracy pass; SITC 2021, full text from Europe PMC read as raw text) · Step and summary now say MRI if available, with or without catheterisation and biopsy.
- 2026-11-16 · Immune-related myocarditis from checkpoint inhibitors: reasons given for steps: Three reasons said more than any source read: that myocarditis "can deteriorate within hours", that each of the three overlapping conditions "raises the danger of the others", and that restarting "risks a recurrence". The patient summary also said "very high-dose steroids straight away" where the source says high-dose steroids as soon as the diagnosis is considered likely. (Second accuracy pass; SITC 2021 full text from Europe PMC read as raw text) · Reasons reworded to what SITC states (a high death rate, a troponin rise without symptoms, the three conditions seen together); patient summary reworded.
- 2026-11-16 · T-cell engaging bispecific antibodies: transfer to a community centre: The page said care moves to a community centre only after the full step-up phase. The European Delphi consensus (statement 4, 90% agreement) adds an exception: patients who had no CRS or ICANS with the first two doses may move earlier. (Second accuracy pass; European Delphi consensus (eJHaem 2026), full text from Europe PMC) · Exception added to the step and the summary, and the statement quoted with its agreement figure.
- 2026-11-16 · T-cell engaging bispecific antibodies: when CRS starts: A step said CRS and ICANS can begin within hours. The consensus gives the typical onset as within 1 to 3 days of a dose, with a median of 2 days for teclistamab and elranatamab. (Second accuracy pass; European Delphi consensus (eJHaem 2026), full text from Europe PMC) · Reworded to 1 to 3 days.
- 2026-11-16 · T-cell engaging bispecific antibodies: tocilizumab on site: The step asked for two doses of tocilizumab wherever step-up doses are given. The consensus statement is about centres giving step-up doses to outpatients. (Second accuracy pass; European Delphi consensus (eJHaem 2026), full text from Europe PMC) · Scoped to outpatient step-up dosing.
- 2026-11-16 · T-cell engaging bispecific antibodies: doses and schedules: The page gave no dose or schedule and said the product labels had not been read. The current US and EU labels for teclistamab, elranatamab and linvoseltamab are readable and give the step-up doses, days, premedication and observation periods. (Second accuracy pass; US prescribing information from DailyMed and EU product information from the EMA, read as raw text) · Three label steps added with doses, days, premedication and hospital or monitoring periods, and the difference between the US labels (hospitalise) and the EU SmPCs (monitor near a healthcare facility) set out.
- 2026-11-16 · T-cell engaging bispecific antibodies: timing in the plain summary: The plain summary said doses are stepped up over the first week or two and that every patient needs a carer. Linvoseltamab reaches its full dose on day 15, and a carer is a condition of outpatient dosing and of the EU linvoseltamab label, not of the US labels, which say to stay in hospital. (Second accuracy pass; US and EU labels read as raw text) · Reworded to one to three weeks, and to a hospital stay or staying near the hospital, usually with a carer.
- 2026-11-16 · T-cell engaging bispecific antibodies: who the guidance covers: The page said other T-cell engagers (blinatumomab, tarlatamab, talquetamab, tebentafusp) follow the same principles, and applied the European consensus to linvoseltamab. No source read says the first, and the consensus says it was written for the two drugs given under the skin and that intravenous and GPRC5D-directed agents need more data. (Second accuracy pass; European Delphi consensus (eJHaem 2026), full text from Europe PMC) · Unsupported sentence removed and the consensus's own limit stated.
- 2026-11-16 · T-cell engaging bispecific antibodies: what the guidelines share: The page credited the IMWG 2024 guideline with step-up dosing advice and said the IgG threshold of 400 mg/dL is shared across the myeloma sources. The IMWG abstract, the only part read, says neither, and the threshold appears only in the European Delphi consensus. The labels are also less firm on infection prophylaxis than the consensus. (Second accuracy pass; Europe PMC abstracts, the Delphi full text and the labels) · Best-in-class statement rewritten to say which source says what.
- 2026-11-16 · T-cell engaging bispecific antibodies: reasons given for steps: Four reasons went beyond the sources: that infections are the main cause of serious harm, that the drugs deplete antibody-making cells, that frailty raises the risk of severe CRS, and that CAR T-cell treatment algorithms can be shared. A neutrophil count of 1000/mm3, which the consensus says is generally recommended, was written as a firm limit. (Second accuracy pass; European Delphi consensus and ASTCT 2019 full texts) · Each reason reworded to what the source says.
- 2026-11-16 · Contrast scans after a previous contrast reaction: the American manual: The page dated the ACR Manual on Contrast Media 2024, said it could not be read, and gave no detail of North American premedication. The edition on the ACR site is 2026 and it reads in full: it says premedication may be considered after a prior reaction to the same class, prints oral regimens (prednisone 50 mg at 13, 7 and 1 hours before, or methylprednisolone 32 mg at 12 and 2 hours before) and a 4 to 5 hour intravenous regimen, and says shorter regimens have not been shown to be effective. (Second accuracy pass; ACR Manual on Contrast Media 2026, complete PDF from acr.org read as text) · Year corrected to 2026, the regimens, indications and the manual's own doubts added, and the disagreement with the ESUR 30-minute emergency regimen and on skin testing stated. A 2025 ACR and AAAAI consensus statement, readable only as an abstract, is now named as a gap.
- 2026-11-16 · Contrast scans after a previous contrast reaction: danger signs after a delayed skin reaction: The list of danger signs included eye involvement, which the source does not name, and left out bleeding or eroded patches, abnormal liver or kidney tests and swollen glands, which it does. (Second accuracy pass; ESUR Contrast Media Safety Committee Guidelines 2026, PDF read as text) · The list now follows the ESUR footnote.
- 2026-11-16 · Contrast scans after a previous contrast reaction: observation and switching steps: The page said most immediate reactions start within 30 minutes, which neither ESUR document says (they define an immediate reaction as one within an hour). It also told teams to switch agent without the source's condition that the agent that caused the reaction is known, and described tryptase as showing whether a reaction was allergic when the ESUR paper says a normal result does not exclude one. (Second accuracy pass; ESUR Guidelines 2026 PDF and the 2025 ESUR paper (Europe PMC full text) read as text) · The three explanations were reworded to what the sources say.
- 2026-11-16 · Rapid drug desensitisation after an allergic reaction to a platinum, taxane or antibody: what ESMO means by rechallenge: The page said ESMO's advice against rechallenge after a severe reaction was compatible with desensitisation if rechallenge meant giving the drug again at the usual rate. ESMO describes rechallenge as restarting at a reduced rate with additional premedication, and its drug table separately lists "Consider desensitisation" [IV, B] for anthracyclines, carboplatin, oxaliplatin, docetaxel, paclitaxel and etoposide. The step also gave a reason (a second full-speed exposure can be fatal) that ESMO does not state. (Second accuracy pass; ESMO 2017 infusion reaction guideline, full text read as raw text from an Internet Archive capture) · The step, summary and best-in-class statement now give ESMO's own description, quote the table entry with its grade, and say that ESMO does not state outright whether its advice covers desensitisation.
- 2026-11-16 · Rapid drug desensitisation after an allergic reaction to a platinum, taxane or antibody: same-day desensitisation: The page said breakthrough reactions occurred in 15% with same-day desensitisation and 24% with planned desensitisation. The 15% is the rate across all 164 desensitisations in the 44 patients whose first was same-day, most of which were later ordinary desensitisations; only three patients reacted during the same-day procedure itself. The same-day group also had milder first reactions (severe in 5% against 22%). (Second accuracy pass; Torres Górriz 2026, full text read as Europe PMC XML) · The variant now states what each percentage counts, the three same-day reactions, and the difference between the groups.
- 2026-11-16 · Rapid drug desensitisation after an allergic reaction to a platinum, taxane or antibody: survival evidence: The summary read as if 241 women had been desensitised. The cohort had 241 patients in all, of whom 25 were desensitised (11 after a platinum reaction and 14 after a taxane reaction). (Second accuracy pass; Zwimpfer 2024, full text read as Europe PMC XML) · The summary and the gap now give the 25 desensitised patients and the 4 breakthrough reactions among the platinum group.
- 2026-11-16 · Rapid drug desensitisation after an allergic reaction to a platinum, taxane or antibody: grades and strength of the allergy guidance: The page gave the grade of three practice parameter statements as "Suggestion" and told the reader to offer desensitisation. The practice parameter prints each as strength conditional, certainty of evidence low, and says desensitisation may be performed. Its grading was also described as GRADE wording without the certainty scale or the authors' statement that it is not a GRADE document. (Second accuracy pass; AAAAI/ACAAI 2022 practice parameter, accepted manuscript read as raw PDF text) · Statements are now quoted from the consensus-based statements themselves with the printed strength and certainty, the action reads consider, and the grading scheme is described as the source describes it.
- 2026-11-16 · Rapid drug desensitisation after an allergic reaction to a platinum, taxane or antibody: where and by whom: The step on doing desensitisation in an experienced centre with anaphylaxis treatment at hand cited an ESMO sentence that only describes past use in experienced centres for reactions recurring despite premedication, and gave a reason that came from a different paper. (Second accuracy pass; ESMO 2017 full text and AAAAI/ACAAI 2022 accepted manuscript, both read as raw text) · The step now quotes the practice parameter: protocols can last several hours with dose doubling every 15 to 20 minutes, usually in inpatient units or infusion centres with trained staff. The reaction pattern is attributed to Castells 2008.
- 2026-11-16 · Rapid drug desensitisation after an allergic reaction to a platinum, taxane or antibody: the 12-step protocol and who it covers: The page called the 12-step protocol the best-known, said its three bags were of increasing concentration and that later desensitisations were outpatient, scoped the protocol to adults and said children are less studied. The abstract read says most later infusions were outpatient and does not describe the bags; no source read makes the statements about adults or children. (Second accuracy pass; Castells 2008 abstract from Europe PMC and AAAAI/ACAAI 2022 accepted manuscript) · Reworded to the sources: the practice parameter's phrase "most experienced published approach", three-bag attributed to it, most later infusions outpatient, and the population no longer limited to adults.
- 2026-11-16 · Immune-related pneumonitis from checkpoint inhibitors, by grade: re-challenge after severe pneumonitis: The page said SITC allows re-challenge only after grade 2 pneumonitis. SITC's lung recommendation covers grade 2, and its general rule says the decision to re-challenge after a grade 3 or 4 side effect should be risk-adjusted (LE: 3); it does not rule it out. (Second accuracy pass; SITC 2021 full text, Europe PMC XML read as raw text) · The variant now states both positions, and the general rule is in the grades table with its sentence quoted.
- 2026-11-16 · Immune-related pneumonitis from checkpoint inhibitors, by grade: drugs after steroids: The step told the team to add a second immunosuppressant with a lung specialist, said steroid-refractory pneumonitis has a high death rate, and left cyclophosphamide out of the grade line. SITC lists five drugs as options in no particular order, names a pulmonary specialist only for tapering mycophenolate, gives cyclophosphamide LE: 3, and states no death rate. (Second accuracy pass; SITC 2021 full text, Europe PMC XML read as raw text) · The step now offers the list as options with the doses SITC prints, the grade line covers all five drugs, and the death-rate sentence is gone.
- 2026-11-16 · Immune-related pneumonitis from checkpoint inhibitors, by grade: holding and stopping the checkpoint inhibitor: The page gave ASCO's rules as flat orders: hold for grade 2, suspend for grade 3, stop for good after grade 4. The ASCO abstract says treatment may be suspended for most grade 2 toxicities, grade 3 generally warrants suspension, and permanent discontinuation is in general recommended for grade 4; it is a general statement and not specific to the lung. (Second accuracy pass; ASCO 2021 abstract from Europe PMC, read as raw text) · The step, summary and variant now carry the abstract's own hedges and say the lung table could not be read.
- 2026-11-16 · Immune-related pneumonitis from checkpoint inhibitors, by grade: grading by CT extent: The page said the Chinese 2022 consensus adds CT extent to the grade. The consensus says it is reproducing the grading of the National Comprehensive Cancer Network guidelines. It also wrote that the consensus limits Pneumocystis prophylaxis, where the source's word is treatment. (Second accuracy pass; Chinese 2022 consensus full text, Europe PMC XML fetched and read as raw text) · The variant now attributes the grading as the source does, gives all four grades, and uses the source's wording on Pneumocystis.
- 2026-11-16 · Immune-related pneumonitis from checkpoint inhibitors, by grade: patient summary and reasons: The patient summary said to report symptoms the same day, that a CT confirms pneumonitis and that severe cases need hospital; step reasons said pneumonitis can worsen quickly and that a slower taper reduces relapse; the summary named prior chest radiotherapy as a risk and the population said adults. None of these is stated in a source that was read. SITC says prior curative-intent radiotherapy and sets no reporting clock or admission rule. (Second accuracy pass; SITC 2021 and Chinese 2022 consensus full texts, read as raw text) · The unsupported statements were removed or reworded to the sources, and the missing admission rule and grade definitions are named in the gaps.
- 2026-11-16 · Immune-related diarrhoea and colitis from checkpoint inhibitors, by grade: restarting immunotherapy: The page said many people can safely restart immunotherapy, and gave the restart rule as grade 1 or less on under 10 mg prednisone a day for everyone. SITC 2021 gives no restart rate, only that 34% of re-challenged patients in one retrospective study had colitis again, and its rule is written for people withheld after grade 3 or more and reads grade 1 or less or baseline. (Second accuracy pass; SITC 2021 full text from Europe PMC, read as raw XML) · The claim that many can restart was removed, the 34% figure stated, and the rule reworded with its scope and the words or baseline.
- 2026-11-16 · Immune-related diarrhoea and colitis from checkpoint inhibitors, by grade: reasons given for steps: Ten reasons said things SITC 2021 does not say, among them that a biologic works faster than more steroid, that stopping steroids too fast risks a flare, that the sickest patients may not absorb tablets, that complications need a surgeon, and that a new infection is a common reason for steroid failure. (Second accuracy pass; SITC 2021 full text from Europe PMC, read as raw XML) · Each reason was rewritten to what the guideline states, for example that no improvement in 3 to 5 days is treated as steroid-refractory and that infliximab is given to reduce the risk of recurrence.
- 2026-11-16 · Immune-related diarrhoea and colitis from checkpoint inhibitors, by grade: what a grade means: The page managed colitis by grade without saying what the grades are. SITC 2021 grades by CTCAE version 5.0, which counts stools a day over baseline: fewer than 4 more is grade 1, 4 to 6 is grade 2 and 7 or more is grade 3. (Second accuracy pass; CTCAE version 5.0 spreadsheet from the US National Cancer Institute, fetched and unpacked by script) · The definitions were added to the summary and to a new grading step, with CTCAE version 5.0 as a source and a note on version 6.0.
- 2026-11-16 · Immune-related diarrhoea and colitis from checkpoint inhibitors, by grade: ASCO general rules: The page said ASCO holds treatment for grade 2. The ASCO 2021 abstract says treatment may be suspended for most grade 2 toxicities, and these are general rules for all organs, not its colitis table. (Second accuracy pass; ASCO 2021 abstract from the Europe PMC record) · Reworded with the abstract's own hedges and a statement that ASCO's colitis table was not read.
- 2026-11-16 · Immune-related diarrhoea and colitis from checkpoint inhibitors, by grade: early biologic and patient advice: The page said SITC does not recommend an early biologic as first line and called the evidence observational; SITC says neither, only that evidence is emerging. The patient summary told readers to ring a 24-hour line, which no cited source describes, and told staff not to wait for every test, where SITC says treatment should start as is deemed clinically appropriate. (Second accuracy pass; SITC 2021 full text from Europe PMC, read as raw XML) · Reworded to say where SITC's recommendations place the biologic, to ask readers to tell their cancer team, and to use SITC's wording on starting treatment.
- 2026-11-17 · Immune-mediated hepatitis from checkpoint inhibitors, by grade: what to report and when: The patient summary and the trigger listed dark urine, pale stools and right upper abdominal pain, and told readers to call the same day. Neither guideline lists those symptoms or gives a time; AASLD mentions jaundice and says in general that patients should be taught to report symptoms. (Second accuracy pass; SITC 2021 and AASLD 2023 full text from Europe PMC, read as raw text) · Kept jaundice only, removed the unsourced symptoms and the same-day clock, and named the absence in the gaps.
- 2026-11-17 · Immune-mediated hepatitis from checkpoint inhibitors, by grade: patient summary: The summary said that if the tests rise, immunotherapy is paused and steroids are given. AASLD says that for grade 1 treatment continues with more frequent monitoring; SITC's general rule also allows treatment to continue at grade 1. (Second accuracy pass; AASLD 2023 and SITC 2021 full text from Europe PMC, read as raw text) · Reworded by grade: a mild rise is watched, a larger rise means a pause and usually steroids.
- 2026-11-17 · Immune-mediated hepatitis from checkpoint inhibitors, by grade: mycophenolate step: The step said to add mycophenolate mofetil if hepatitis returns on the taper. SITC says it may be given (LE: 4) if ALT or AST is not grade 1 or less within 10 to 14 days of starting steroids, or if toxicity recurs after the steroid taper. (Second accuracy pass; SITC 2021 full text from Europe PMC, read as raw text) · Changed the order to "consider", "on the taper" to "after the steroid taper", and the reason to match.
- 2026-11-17 · Immune-mediated hepatitis from checkpoint inhibitors, by grade: clinical summary: The summary wrote three "consider" recommendations as settled practice: SITC's abdominal imaging at grade 3 or more, and AASLD's intravenous steroids and hospital admission for jaundice. Both sources say these should be considered. (Second accuracy pass; SITC 2021 and AASLD 2023 full text from Europe PMC, read as raw text) · Restored "should be considered" in each place, and added AASLD's grade thresholds, which the page did not give.
- 2026-11-17 · Immune-mediated hepatitis from checkpoint inhibitors, by grade: reasons given for steps: Six reasons said things no source states: that mild rises often settle, that severe hepatitis progresses within days, that a fast taper risks rebound, that biopsy rarely changes early management, how steroids work, and what the workup tests separate. (Second accuracy pass; SITC 2021 and AASLD 2023 full text from Europe PMC, read as raw text) · Replaced each with what the sources say, such as AASLD's grade thresholds and its note on rebound hepatitis after tapering.
- 2026-11-17 · Immune-mediated hepatitis from checkpoint inhibitors, by grade: comparison of guidelines: The comparison said SITC and AASLD agree on excluding metastases, infection and viral hepatitis, gave AASLD's ungraded guidance the grade "expert opinion", and said SITC's other agents rest on use in autoimmune hepatitis. AASLD's immunotherapy text does not name viral hepatitis and prints no grade, and SITC ties only tacrolimus and mycophenolate to autoimmune hepatitis, not antithymocyte globulin. (Second accuracy pass; SITC 2021 and AASLD 2023 full text from Europe PMC, read as raw text) · Stated what each source says, marked AASLD as not graded, and added SITC's general rule that re-challenge after grade 3 or 4 is a risk-adjusted decision (LE: 3).
- 2026-11-17 · Immune-mediated hepatitis from checkpoint inhibitors, by grade: who it applies to: The population said adults, and that people with liver metastases or chronic liver disease need other causes excluded with particular care. Neither source scopes its hepatitis advice to adults; SITC says only that abdominal imaging should be considered at grade 3 or more in those groups. (Second accuracy pass; SITC 2021 full text from Europe PMC, read as raw text) · Reworded to SITC's sentence and to "people".
- 2026-11-17 · Neurological side effects of checkpoint inhibitors: myasthenia mortality figure: The summary said myasthenia gravis carries about 20% mortality. Neither cited source prints a mortality figure; SITC 2021 says only that it has a high potential for patient fatality. (Second accuracy pass; SITC 2021 full text (Europe PMC XML) and the 2021 consensus definitions (Europe PMC XML), searched by script) · Figure removed, SITC's wording used, and the missing figure named in gaps.
- 2026-11-17 · Neurological side effects of checkpoint inhibitors: refractory encephalitis: The step ordered IVIG, plasma exchange or rituximab, and the summary said rituximab comes next. SITC 2021 says patients may receive any of the three if pulse steroids fail. (Second accuracy pass; SITC 2021 full text (Europe PMC XML)) · Step reworded to consider, and the summary now lists the three options with the schedules SITC prints.
- 2026-11-17 · Neurological side effects of checkpoint inhibitors: how common they are: The page said immunotherapy rarely affects the nerves or brain, and placed the first 3 months timing on severe events. SITC 2021 gives 4 to 12% overall with under 1% at grade 3 or more, and the timing is for all neurological events. (Second accuracy pass; SITC 2021 full text (Europe PMC XML)) · Reworded: serious cases are uncommon, and the timing stands as its own sentence.
- 2026-11-17 · Neurological side effects of checkpoint inhibitors: reasons given for steps: Several reasons said more than the source: that myasthenia often comes with heart inflammation, that prompt treatment reduces respiratory failure, that IVIG or plasma exchange is the standard treatment for Guillain-Barré syndrome, and that encephalitis and Guillain-Barré are the most dangerous. SITC 2021 states none of these. (Second accuracy pass; SITC 2021 full text (Europe PMC XML)) · Each reason reworded to what SITC states.
- 2026-11-17 · Neurological side effects of checkpoint inhibitors: evidence behind the LE: 1 grades: The page said there are no trials in checkpoint inhibitor patients. No source read says that; SITC cites systematic reviews of myasthenia gravis, autoimmune encephalitis and Guillain-Barré syndrome outside immunotherapy. (Second accuracy pass; SITC 2021 reference list (Europe PMC XML)) · Reworded to name the reviews SITC cites.
- 2026-11-17 · Neurological side effects of checkpoint inhibitors: ASCO exception and who is at risk: The page said ASCO makes neurological toxicity an exception at grade 1, and that encephalitis goes with PD-(L)1 blockade. The ASCO abstract says some neurologic toxicities; SITC says PD-(L)1 monotherapy or combinations. A general statement on flares of earlier neurological disease went beyond the definitions paper. (Second accuracy pass; ASCO 2021 abstract (Europe PMC record), SITC 2021 and the consensus definitions (Europe PMC XML)) · Each reworded to the source, with the ASCO sentence quoted.
- 2026-11-17 · Neurological side effects of checkpoint inhibitors: restarting treatment: A gap said restarting is not addressed beyond permanent discontinuation. SITC 2021 has general rechallenge rules (LE: 3) that apply where the neurological section is silent. (Second accuracy pass; SITC 2021 full text (Europe PMC XML)) · Gap rewritten to state those rules and where SITC is silent.
- 2026-11-17 · Checkpoint inhibitor hypophysitis and adrenal insufficiency: how often cortisol is tested: The page said the guidelines it read set an interval for thyroid tests but none for cortisol. The ESE 2022 guideline does: it suggests a morning cortisol with TSH, free T4, glucose and electrolytes at baseline, repeated every 4 to 6 weeks, preferably before each treatment cycle. (Second accuracy pass; ESE 2022, full text read as raw XML from Europe PMC) · The gap now gives the ESE interval and says only that SITC sets none for cortisol.
- 2026-11-17 · Checkpoint inhibitor hypophysitis and adrenal insufficiency: sick day rules: The page told readers to double the dose when ill and said most adrenal crises can be prevented that way, but none of its cited sources said "double" or "most". The rule is in the Society for Endocrinology 2016 adrenal crisis guidance, which had not been read: double the daily oral dose during illness with fever that needs bed rest or antibiotics, and use an injection during prolonged vomiting or diarrhoea. (Second accuracy pass; Society for Endocrinology 2016 adrenal crisis guidance, full text read as raw XML from Europe PMC) · The 2016 guidance was added as a source with a new step quoting Sick Day Rule 1, and the unsupported "most" was removed.
- 2026-11-17 · Checkpoint inhibitor hypophysitis and adrenal insufficiency: emergency hydrocortisone dose: The 100 mg then 50 mg every 6 hours dose was attributed to the Society for Endocrinology 2018 guidance as well as ESE 2022. The readable text of the 2018 guidance states no dose; its algorithms are images. The dose itself is right and is stated by ESE 2022 and by the Society's 2016 adrenal crisis guidance. (Second accuracy pass; Society for Endocrinology 2018 and 2016 guidance and ESE 2022, full texts read as raw XML from Europe PMC) · The dose is now attributed to ESE 2022 and the 2016 guidance, and a gap says the 2018 algorithm figures were not read.
- 2026-11-17 · Checkpoint inhibitor hypophysitis and adrenal insufficiency: continuing immunotherapy: The step said to continue or restart immunotherapy once stable, and the page presented SITC 2021 as disagreeing. ESE 2022 adds an exception the page left out: it suggests oncology and endocrinology discuss interruption when hypophysitis compresses the optic chiasm or optic nerve. SITC's general endocrine rule also says immunotherapy should generally be continued as soon as patients are stable, and the ASCO abstract's rule is "in general". (Second accuracy pass; ESE 2022 full text and SITC 2021 full text from Europe PMC, ASCO 2021 abstract) · The step and summary now carry the optic compression exception, SITC's general rule is stated beside its hypophysitis advice, and the ASCO hedge is restored.
- 2026-11-17 · Checkpoint inhibitor hypophysitis and adrenal insufficiency: stopping emergency steroids: The step said to stop emergency steroids only if the pre-treatment cortisol was above 450 nmol/L. The Society for Endocrinology says treatment can be discontinued at that point if cortisol deficiency was the only indication, and that there are exceptions listed in an algorithm figure. (Second accuracy pass; Society for Endocrinology 2018 guidance, full text read as raw XML from Europe PMC) · Reworded to "consider stopping", with the exceptions and the unread figure named.
- 2026-11-17 · Checkpoint inhibitor hypophysitis and adrenal insufficiency: reasons given for steps: Several reasons said more than the sources: that thyroid hormone "speeds up cortisol use", that aldosterone "keeps salt and blood pressure up", that giving steroids while waiting "is safe", and that people already on high-dose steroids "need interpretation by an endocrinologist". The sources say only that thyroxine can trigger an adrenal crisis, that adrenal destruction causes a lack of aldosterone, that waiting for results can be fatal, and that steroid use can make cortisol results inaccurate. (Second accuracy pass; ESE 2022 and Society for Endocrinology 2018 full texts from Europe PMC, SITC 2021 full text) · Each reason was reworded to what its source states.
- 2026-11-17 · Checkpoint inhibitor hypophysitis and adrenal insufficiency: replacement dose and tapering: The 15 to 25 mg hydrocortisone step did not say that ESE scopes it to hypophysitis with confirmed ACTH deficiency, and the page said the Society for Endocrinology tapers "after 48 to 72 hours of recovery" when the source says tapering can start after clinical recovery, which usually takes 48 to 72 hours. The reason for not stopping steroids abruptly was also not the one ESE gives. (Second accuracy pass; ESE 2022 and Society for Endocrinology 2018 full texts read as raw XML from Europe PMC) · The step is scoped as ESE scopes it, the taper timing is corrected, and the abrupt-stop step now gives ESE's reason (prednisolone 7.5 mg equivalent or more for over 3 weeks can cause secondary adrenal insufficiency).
- 2026-11-17 · Immune-related type 1 diabetes and diabetic ketoacidosis from checkpoint inhibitors: who wrote the 2025 definitions: The page called the 2025 consensus disease definitions for endocrine side effects a SITC document. The paper says it is part of the irAE Program in partnership with Project Data Sphere. It is published in the journal of SITC but is not a SITC guideline. (Second accuracy pass; the 2025 definitions, full text read as XML from Europe PMC (PMC12530408)) · Every mention now says "2025 consensus definitions" and the source entry states who produced them.
- 2026-11-17 · Immune-related type 1 diabetes and diabetic ketoacidosis from checkpoint inhibitors: starting insulin: The step said to start insulin urgently once immune diabetes is suspected. The sentence it quotes says the patient should be considered for urgent insulin initiation. (Second accuracy pass; the 2025 definitions, full text read as XML from Europe PMC (PMC12530408)) · The step and summary now say to consider starting insulin urgently, and note that the same paper suggests a low threshold for insulin while tests are under way.
- 2026-11-17 · Immune-related type 1 diabetes and diabetic ketoacidosis from checkpoint inhibitors: autoantibody figures: The summary said autoantibodies are found in 40 to 53%. No source prints that range. SITC 2021 says autoantibodies are found in up to 53% of cases, most of them GAD65. The 2025 definitions say GAD65 antibodies are positive in 40 to 43%. (Second accuracy pass; SITC 2021 and the 2025 definitions, full text read as XML from Europe PMC) · Both figures are now given separately, each with its source.
- 2026-11-17 · Immune-related type 1 diabetes and diabetic ketoacidosis from checkpoint inhibitors: restarting treatment: A gap said no source says whether or when to restart the checkpoint inhibitor after ketoacidosis. SITC 2021 says it should be held until ketoacidosis is resolved and that immunotherapy should generally be continued as soon as patients are stable. The variant also restated a step instead of the real difference, which is that ESE sees no need to interrupt treatment for endocrine problems other than severe thyroid eye disease or optic compression (low quality evidence). (Second accuracy pass; SITC 2021 and ESE 2022, full text read as XML from Europe PMC) · The gap now says what the sources do state and what they leave open, and the variant gives the ESE position with its grade.
- 2026-11-17 · Immune-related type 1 diabetes and diabetic ketoacidosis from checkpoint inhibitors: figures and reasons that said more than the sources: The page said blood sugar can rise within days, that insulin production falls over the first weeks, that onset can follow the first dose, that 76% of cases follow PD-1 inhibitors, that steroids do not help, that a normal C-peptide does not exclude the diagnosis, and that insulin was withdrawn in one study. The sources say it develops rapidly, that C-peptide can decline rapidly, that onset can be shortly after treatment begins, up to 76%, that there is no evidence steroids rescue the insulin-making cells, a detectable C-peptide, and one single patient. (Second accuracy pass; SITC 2021, ESE 2022 and the 2025 definitions, full text read as XML from Europe PMC) · Each statement was reworded to what its source says.
- 2026-11-17 · Kidney inflammation (immune-related nephritis) from checkpoint inhibitors: how common it is: The summary said only about 2 to 5% of kidney injury on checkpoint inhibitors is caused by the drug. The ADQI report says up to 20% of treated patients develop acute kidney injury and that the incidence directly attributable to the drug is likely only 2 to 5%, which is a share of patients, not of kidney injuries. (Second accuracy pass; ADQI 2026 consensus report, full text read as raw text from PubMed Central) · The summary now gives 2 to 5% as the incidence of kidney injury caused by the drug.
- 2026-11-17 · Kidney inflammation (immune-related nephritis) from checkpoint inhibitors: when steroids start: A step told teams to aim to start steroids within 3 days of diagnosis, and another said under SITC's grade that recovery is better when they start within 3 days. ADQI reports an association between treatment within 3 days and kidney recovery and suggests early empirical steroids even while a biopsy is awaited. It sets no 3 day target, and SITC says nothing about timing. (Second accuracy pass; ADQI 2026 full text from PubMed Central and SITC 2021 full text from Europe PMC) · The step now says to consider starting steroids early, even while a biopsy is awaited, gives the 3 day finding as an association, and carries ADQI's printed rating (97% consensus, moderate certainty).
- 2026-11-17 · Kidney inflammation (immune-related nephritis) from checkpoint inhibitors: what could be read: The page said the ADQI report could be read only as its abstract, that it prints no grades, and that it is the only nephrology-led statement. The full text is readable on PubMed Central, it prints a strength of consensus and a certainty of evidence for each of its 16 statements, and it cites an earlier position statement from the American Society of Onco-Nephrology. (Second accuracy pass; ADQI 2026 full text read as raw text from PubMed Central) · The report was read in full. Its statements on referral, biopsy, steroid dose and duration, second-line drugs, rechallenge and transplant recipients were added with their printed ratings, and the claim that it is the only nephrology-led statement was removed.
- 2026-11-17 · Kidney inflammation (immune-related nephritis) from checkpoint inhibitors: patient summary: The short summary said the problem usually causes no symptoms. Neither source read says so. SITC says it typically shows as reduced kidney function, indicated by a rising creatinine. (Second accuracy pass; SITC 2021 full text from Europe PMC and ADQI 2026 full text from PubMed Central) · The sentence now says it is usually found when the creatinine on a blood test rises.
- 2026-11-17 · Kidney inflammation (immune-related nephritis) from checkpoint inhibitors: reasons given for steps: Several reasons went beyond the sources: that other causes of kidney injury can be fixed, that larger falls often need a biopsy, that a second immunosuppressant can rescue the kidneys, and that restarting waits until the kidneys have recovered. SITC grades second-line drugs LE: 4 and ADQI says the data for any specific drug are limited. ADQI says rechallenge may be considered once the injury has been treated, after weighing risk against benefit. (Second accuracy pass; SITC 2021 full text from Europe PMC and ADQI 2026 full text from PubMed Central) · Each reason now says only what its source says, and the rechallenge step uses the wording of ADQI consensus statement 13.
- 2026-11-17 · Kidney inflammation (immune-related nephritis) from checkpoint inhibitors: where SITC and ADQI differ: The page said ADQI agrees with SITC. The two differ in places: ADQI suggests nephrology referral for every suspected case where SITC asks for it at stage 2 or 3 and says to consider it for persistent stage 1, ADQI decides on biopsy by clinical suspicion and supports empirical steroids first when suspicion is high, and ADQI leaves the 0.3 mg/dL creatinine rise out of its definition. (Second accuracy pass; SITC 2021 full text from Europe PMC and ADQI 2026 full text from PubMed Central) · The differences are now stated as variants with each body's printed grade, and the unread ESMO guideline is no longer listed as support for the best-supported version.
- 2026-11-17 · Blood side effects of checkpoint inhibitors: aplastic anaemia grade: The page put the grade LE: 4 on ciclosporin for aplastic anaemia and wrote ciclosporin or ATG as equals. SITC says an immunosuppressant beyond steroids such as ciclosporin should be used and prints no grade for that sentence. The LE: 4 follows the next sentence, that ATG may also be considered, and cites one case report of ATG. (Second accuracy pass; SITC 2021 full text from Europe PMC (PMC8237720) read as raw text, with its reference list) · The step now quotes both sentences and gives LE: 4 to ATG only. The summary, best-in-class statement and grades table say the ciclosporin sentence carries no grade.
- 2026-11-17 · Blood side effects of checkpoint inhibitors: when to stop and restart the checkpoint inhibitor: The page said SITC does not address restarting beyond one small series, and had no step on withholding treatment. SITC's haematological section says it works within SITC's general rules. Those say to withhold temporarily at grade 2, allow re-challenge after grade 2 once controlled on 10 mg or less of prednisone a day (LE: 3), and make re-challenge after grade 3 or 4 a risk-adjusted decision (LE: 3). (Second accuracy pass; SITC 2021 full text from Europe PMC (PMC8237720) read as raw text, with its reference list) · Added a step to withhold at grade 2 with the SITC sentence quoted, two grade rows for the re-challenge rules, and rewrote the gap to say the haematological section has no rule of its own.
- 2026-11-17 · Blood side effects of checkpoint inhibitors: IVIG for haemolysis: The page said to add IVIG if haemolytic anaemia does not respond to steroids. SITC limits this to severe haemolytic anaemia that does not respond to steroids, at 0.5 g/kg/day for 5 days (LE: 3). (Second accuracy pass; SITC 2021 full text from Europe PMC (PMC8237720) read as raw text, with its reference list) · The step, summary and grades table now say severe and give the dose.
- 2026-11-17 · Blood side effects of checkpoint inhibitors: ASCO exception at grade 1: The page said ASCO makes haematological toxicity an exception to continuing treatment at grade 1. The abstract says that in general treatment continues at grade 1 except for some neurological, haematological and cardiac toxicities, and does not say which. (Second accuracy pass; ASCO 2021 abstract from the Europe PMC record read as raw text) · Restored in general and some, quoted the abstract sentence, and said in gaps that the ASCO haematological table was not read.
- 2026-11-17 · Blood side effects of checkpoint inhibitors: how common and how dangerous: The summary called aplastic anaemia rare and haemophagocytic lymphohistiocytosis the most dangerous, and the patient lines said rarely and very rarely. SITC calls haematological toxicities uncommon, says rarely only of pure red cell aplasia, and calls HLH life-threatening without ranking it against the others. (Second accuracy pass; SITC 2021 full text from Europe PMC (PMC8237720) read as raw text, with its reference list) · Reworded to uncommon and life-threatening, removed rare from aplastic anaemia, and added the medians SITC prints for pure red cell aplasia and for resolution.
- 2026-11-17 · Blood side effects of checkpoint inhibitors: who is involved and baseline platelet count: The patient summary said a haematologist is usually involved and the first step ordered a baseline platelet count. SITC says referral to a haematologist should be considered case by case, and that a baseline platelet count may be prudent. (Second accuracy pass; SITC 2021 full text from Europe PMC (PMC8237720) read as raw text, with its reference list) · Reworded to a blood specialist may be involved and to consider recording a baseline.
- 2026-11-17 · Blood side effects of checkpoint inhibitors: reasons given for steps: Several reasons stated things SITC does not: that steroids stop the immune system destroying platelets, that chemotherapy neutropenia does not need steroids, that marrow failure rarely responds to steroids alone, and that ADAMTS13 excludes a clotting-related cause. (Second accuracy pass; SITC 2021 full text from Europe PMC (PMC8237720) read as raw text, with its reference list) · Each reason now says only what SITC says, such as that treatment should follow haematological standard of care and that aplastic anaemia is frequently difficult to treat.
- 2026-11-17 · Blood side effects of checkpoint inhibitors: warning signs for patients: The patient lines listed symptoms to report with no source, and SITC gives none. Macmillan's pembrolizumab page says to contact the hospital straight away on the 24-hour number for unexplained bruising or bleeding, a temperature above 37.5 degrees, yellow skin or eyes and dark urine. Cancer Research UK lists breakdown of red blood cells as a rare side effect causing yellowing, dark urine and a high temperature. (Second accuracy pass; Macmillan and Cancer Research UK pembrolizumab pages fetched with curl on 17 November 2026 and read as tag-stripped text) · Added both pages as sources, added a patient step quoting Macmillan, matched the wording, and said in gaps that the advice comes from pages about one drug.
- 2026-11-17 · Blood side effects of checkpoint inhibitors: missing recommendations: The steps left out two SITC recommendations the page's own grades relied on or its scope named: IVIG 0.5 g/kg/day for 5 days with steroids for severe immune thrombocytopenia (LE: 1) with rituximab as an option (LE: 2), and steroids for pure red cell aplasia (LE: 3). (Second accuracy pass; SITC 2021 full text from Europe PMC (PMC8237720) read as raw text, with its reference list) · Added a step for each with the SITC sentence quoted, and said in gaps that SITC gives no platelet or haemoglobin threshold for severe.
- 2026-11-17 · Thyroid inflammation and underactive thyroid from checkpoint inhibitors: source of the 2025 definitions: The 2025 consensus definitions for endocrine side effects were called a SITC document. The paper says it comes from the irAE Program in partnership with Project Data Sphere and it is only published in SITC's journal. (Second accuracy pass; the paper's full text from Europe PMC, read as raw text) · Renamed the source and every mention of it to the 2025 consensus definitions and corrected the publisher.
- 2026-11-17 · Thyroid inflammation and underactive thyroid from checkpoint inhibitors: cortisol check: The page said SITC tells clinicians to check morning cortisol before starting levothyroxine and that both guidelines agree on this. SITC says to test morning cortisol in people with symptoms of hypothyroidism or a high TSH with low free T4 and gives no timing. The advice to exclude adrenal insufficiency before levothyroxine is in the hypophysitis section of the ESE guideline. (Second accuracy pass; SITC 2021 and ESE 2022 full texts from Europe PMC, read as raw text) · Reworded the step, summary and best in class statement so each claim is attributed to the source and section that makes it, and added a note under gaps.
- 2026-11-17 · Thyroid inflammation and underactive thyroid from checkpoint inhibitors: Graves' disease step: The step on a persistently low TSH with high free T4 cited the 2025 definitions and said assessment includes TSH receptor antibodies. That trigger is SITC 2021's recommendation, and the 2025 paper says only that the antibody test should be considered when Graves' disease is suspected. (Second accuracy pass; both full texts from Europe PMC, read as raw text) · Cited the step to SITC 2021 with its own sentence and wrote the antibody test as something to consider.
- 2026-11-17 · Thyroid inflammation and underactive thyroid from checkpoint inhibitors: grade for stopping levothyroxine: The quotation for tapering and stopping low-dose levothyroxine was recommendation R 2.6, which is about pituitary inflammation, not the thyroid. The thyroid section makes the same suggestion in its own sentence, also graded +000. (Second accuracy pass; ESE 2022 full text from Europe PMC, read as raw text) · Replaced the quotation with the sentence from the thyroid section.
- 2026-11-17 · Thyroid inflammation and underactive thyroid from checkpoint inhibitors: hedged advice written as orders: Continuing immunotherapy and watching mild cases without treatment were written as flat instructions. SITC says immunotherapy should generally be continued, ESE only suggests that mild cases can be monitored, and ESE suggests discussing a pause for severe thyroid eye disease, which the page did not mention. (Second accuracy pass; SITC 2021 and ESE 2022 full texts from Europe PMC, read as raw text) · Restored the hedges and added the ESE statement on severe thyroid eye disease with its grade.
- 2026-11-17 · Thyroid inflammation and underactive thyroid from checkpoint inhibitors: reasons and symptoms: Several reasons said more than the sources: that a mildly raised TSH often settles, that early dose changes overshoot, that problems start months after the last dose, and that ESE's thresholds are better supported. The patient summary said immunotherapy often inflames the thyroid and listed sweating, which no source read names. (Second accuracy pass; SITC 2021, ESE 2022 and the 2025 definitions as raw text) · Rewrote each reason to what the source states, gave the ESE grades as low and very low quality evidence, and took the symptoms from the 2025 definitions table.
- 2026-11-17 · Joint, muscle and other rheumatic side effects of checkpoint inhibitors: which guideline is the more recent: The page called the EULAR statement the more recent of the two guidelines. EULAR was first published online in April 2020 from a literature search to December 2018, and SITC was published in June 2021. (Second accuracy pass; EULAR and SITC full-text XML from Europe PMC, publication dates read from the records) · The comparison now gives both dates and says SITC is the later.
- 2026-11-17 · Joint, muscle and other rheumatic side effects of checkpoint inhibitors: steroid, DMARD and myositis steps: Three steps were written as orders (give steroids at set doses, add a conventional DMARD, give high-dose steroids). SITC says prednisone may start at 10 to 20 mg/day and that 40 to 60 mg/day may be required; EULAR says a conventional DMARD, and high-dose steroids with IVIG or plasma exchange, should be considered. (Second accuracy pass; SITC 2021 and EULAR full text from Europe PMC, read as raw XML) · The steps now say consider, may start and may be required, as the sources do.
- 2026-11-17 · Joint, muscle and other rheumatic side effects of checkpoint inhibitors: where the guidelines agree: The page said SITC and EULAR agree on withdrawing immunotherapy for life-threatening myositis and on early referral, and that both rest on case series and expert opinion. SITC holds the checkpoint inhibitor for grade 3 myositis and stops it permanently only with heart involvement, refers at grade 2 or higher where EULAR refers whatever the grade, and grades its arthritis doses LE: 1 from reviews of rheumatoid arthritis and polymyalgia. SITC's general rule of withholding at grade 2 was missing. (Second accuracy pass; SITC 2021 and EULAR full text from Europe PMC, read as raw XML) · Each guideline's position on referral, steroids and holding the checkpoint inhibitor is now stated separately, with SITC's general rules and their grades.
- 2026-11-17 · Joint, muscle and other rheumatic side effects of checkpoint inhibitors: reasons given for steps: Several reasons said more than the sources: that many mild cases settle without steroids (EULAR does not say so, and 76% of arthritis patients in its review had steroids), that biological drugs are used without clear harm to the cancer response (EULAR reports one study of TNF inhibitors and says further data are needed), and that most arthritis can be treated while immunotherapy carries on (EULAR reports it was stopped in 25% of arthritis cases). (Second accuracy pass; EULAR full text from Europe PMC, read as raw XML) · The reasons now give what EULAR states, with its figures and its caveat, and add its caution about sulfasalazine.
- 2026-11-17 · Joint, muscle and other rheumatic side effects of checkpoint inhibitors: how common each problem is: The summary said polymyalgia, sicca syndrome and vasculitis are less common than arthritis, citing SITC. SITC calls vasculitis and myositis rare but says the frequency of polymyalgia is hard to assess and gives no such ranking. (Second accuracy pass; SITC 2021 full text from Europe PMC, read as raw XML) · The summary now says only what SITC states.
- 2026-11-17 · Joint, muscle and other rheumatic side effects of checkpoint inhibitors: what to report and how fast: The patient summary called muscle weakness, trouble swallowing or breathlessness urgent and said immunotherapy often continues, with no cited source for what to report, and the second point is true of arthritis only. (Second accuracy pass; Macmillan Cancer Support pembrolizumab page, fetched with curl and read as text; EULAR full text) · Macmillan's advice is now cited: the 24-hour number for new aches, joint stiffness or swelling, or muscle weakness, and straight away for breathlessness or palpitations. A gap says no source read sets a time limit for trouble swallowing.
- 2026-11-17 · Immune-related skin reactions from checkpoint inhibitors: when treatment is paused: The page told readers to interrupt the checkpoint inhibitor for a rash in general. SITC 2021 prints that instruction only in its recommendation on rashes with blisters or mucosal involvement. Its general rules say treatment may continue with monitoring at grade 1 and is temporarily withheld at grade 2. (Second accuracy pass; SITC 2021 full-text XML from Europe PMC, read raw) · The step is now scoped to blistering or mucosal rashes, and the grade 1 and grade 2 general rules are separate quoted steps.
- 2026-11-17 · Immune-related skin reactions from checkpoint inhibitors: severe reactions: The page said no source read set out steroid doses or intensive care management for severe skin reactions, and gave none. The MASCC 2020 recommendations are open in PubMed Central and give them, including permanent discontinuation and immediate intensive care or burns unit admission for grade 4 Stevens-Johnson syndrome, DRESS and toxic epidermal necrolysis. (Second accuracy pass; MASCC 2020 author manuscript in PubMed Central, read as HTML with tags stripped) · MASCC 2020 added as a source with eight quoted steps covering grade 3 rash, grade 3 blistering, severe cutaneous reactions, grade 3 itch and re-challenge.
- 2026-11-17 · Immune-related skin reactions from checkpoint inhibitors: patient warning signs: The page told patients to contact the cancer team the same day for blisters, peeling or mouth sores, with no source for the symptoms or the time limit. (Second accuracy pass; Macmillan pembrolizumab page reviewed 1 August 2026, read raw) · Wording now follows Macmillan, which says to ring the 24-hour number straight away for a spreading rash or swollen, blistering or peeling skin. The page says no source read sets a limit in hours.
- 2026-11-17 · Immune-related skin reactions from checkpoint inhibitors: erosive mucocutaneous disease: The page said to treat erosive mucocutaneous disease as an emergency that usually needs systemic immune suppression, and that the 2024 consensus separates it from classic SJS/TEN. The consensus says systemic immunosuppression is often required, makes no such comparison, and states that treatment is outside its scope. (Second accuracy pass; D-irAE 2024 full-text XML from Europe PMC, read raw) · Reworded to what the consensus says.
- 2026-11-17 · Immune-related skin reactions from checkpoint inhibitors: steroid-sparing drugs: The page said to use a steroid-sparing drug if a rash returned on taper. SITC 2021 recommends a dermatological consultation or a steroid-sparing agent, because grade 2 or higher reactions may recur. (Second accuracy pass; SITC 2021 full-text XML from Europe PMC, read raw) · Reworded, with the dermatology option restored.
- 2026-11-17 · Immune-related skin reactions from checkpoint inhibitors: timing and frequency: The summary gave 3 to 4 weeks as the average onset of rash and called the severe forms rare. SITC 2021 gives 3 to 4 weeks for skin reactions as a class and 2 to 5 weeks for rash, and calls the severe forms uncommon. (Second accuracy pass; SITC 2021 full-text XML from Europe PMC, read raw) · Both onset figures are stated against the right subject and the severe forms are called uncommon.
- 2026-11-17 · Immune-related skin reactions from checkpoint inhibitors: ASCO general rules: The page gave the ASCO 2021 rules as flat orders (hold at grade 2, permanent stop at grade 4). The abstract hedges each one: in general, may be suspended for most grade 2, generally, and in general at grade 4. (Second accuracy pass; ASCO 2021 abstract from Europe PMC) · Hedges restored in the best-in-class statement and the evidence rows.
- 2026-11-17 · Immune-related skin reactions from checkpoint inhibitors: reasons given for steps: Several reasons said more than the sources: that itch often responds, that most rashes allow a restart, that most skin reactions do not rule out further immunotherapy, that steroid-treated skin becomes infected, and that antibiotics or chemotherapy are the usual other causes. (Second accuracy pass; SITC 2021, D-irAE 2024 and MASCC 2020, read raw) · Each reason reworded to what the cited source states.
- 2026-11-17 · Eye side effects of checkpoint inhibitors: posterior uveitis and panuveitis: The step told clinicians to step up to steroid injections in or around the eye and to hold the checkpoint inhibitor only if disease still progressed. The 2025 consensus says intraocular steroids may be considered for posterior uveitis and panuveitis (injection around the eye is offered for intermediate uveitis), and then to consider holding the checkpoint inhibitor, oral steroids or both, with no "only". (Second accuracy pass; 2025 Eye-irAE consensus definitions, full-text XML and supplemental tables from Europe PMC) · The step now says to consider a steroid injection into the eye, then to consider holding the checkpoint inhibitor, oral steroids or both, and the panel's disagreement about "last resort" is a variant.
- 2026-11-17 · Eye side effects of checkpoint inhibitors: optic neuritis: The step read as a flat order to withhold the checkpoint inhibitor and give IV methylprednisolone 1000 mg daily for 3 to 5 days. The source says most cases should be treated, gives the dose as 1000 mg daily or equivalent, and says it can be continued for 3 to 5 days; its supplemental table says to consider holding the checkpoint inhibitor. (Second accuracy pass; 2025 Eye-irAE consensus definitions, full-text XML and supplemental tables from Europe PMC) · The step and summary carry the source's hedges, and the difference between the main text and the table is stated.
- 2026-11-17 · Eye side effects of checkpoint inhibitors: when the checkpoint inhibitor is held: The page said SITC does not say when the checkpoint inhibitor may continue, and that it continues for anterior uveitis without exception. SITC's general rule withholds treatment at grade 2 unless otherwise specified, and the 2025 supplemental table says not to hold for anterior uveitis unless it cannot be adequately controlled. (Second accuracy pass; SITC 2021 full text and the 2025 supplemental tables from Europe PMC) · The SITC general rule, the exception and the unreconciled difference between the two sources are now stated.
- 2026-11-17 · Eye side effects of checkpoint inhibitors: steroids before eye review: The page said not to start steroids until an ophthalmologist has examined the eye. SITC says steroids for eye symptoms should be started under the guidance of an ophthalmologist. (Second accuracy pass; SITC 2021 full text from Europe PMC) · Reworded to "under the guidance of an ophthalmologist".
- 2026-11-17 · Eye side effects of checkpoint inhibitors: warning signs for patients: The page called sudden loss of vision an emergency and called optic nerve inflammation rare, with no source for either. No guideline read gives a time limit or a frequency. (Second accuracy pass; Macmillan pembrolizumab page read as raw HTML, SITC 2021 full text) · The patient advice now follows Macmillan (contact the hospital straight away on the 24-hour number about eye symptoms or changes to eyesight), and "rarely" is removed.
- 2026-11-17 · Eye side effects of checkpoint inhibitors: reasons and summary: Several reasons said more than the sources: permanent loss of vision, hypophysitis as a cause of double vision, local treatment often avoiding a stop to cancer treatment, a statement that there are no trials, and a list of eye conditions attributed to SITC that includes thyroid eye disease, which SITC does not name. The dry eye step also dropped one of the two conditions for a treatment break. (Second accuracy pass; SITC 2021 and 2025 Eye-irAE consensus full texts from Europe PMC) · Each is reworded to what its source states.
- 2026-11-17 · Steroid taper after immunotherapy side effects, with Pneumocystis prophylaxis and adrenal checks: adrenal check before stopping: The step said to check a morning cortisol before stopping steroids after 3 weeks or more at 7.5 mg prednisolone. The source says more than 3 weeks, says the test should be considered rather than done in every case, and does not mention adrenal crisis in that passage. (Second accuracy pass; ESE 2022 full text from Europe PMC, read as raw text) · The step now says more than 3 weeks and "consider", and the reason given is the one the source states.
- 2026-11-17 · Steroid taper after immunotherapy side effects, with Pneumocystis prophylaxis and adrenal checks: Pneumocystis prophylaxis: The step told clinicians to start prophylaxis for anyone on 20 mg of prednisone a day for a month. The ASCO and IDSA sentence recommends prophylaxis for chemotherapy regimens above a 3.5% risk and gives that steroid dose as an example; it was not written for immunotherapy side effects. The page also said the antibiotic is often given and that co-trimoxazole is the most effective preventive, which no source read states. (Second accuracy pass; ASCO/IDSA 2018 and ECIL-5 abstracts from Europe PMC, read as raw text) · The step now says "consider", states that the rule is applied by analogy, and describes co-trimoxazole as the drug of choice.
- 2026-11-17 · Steroid taper after immunotherapy side effects, with Pneumocystis prophylaxis and adrenal checks: second-line treatment: The step said to move to a second-line immunosuppressant rather than continuing high doses, because prolonged high-dose steroids add harm without benefit. SITC says only that second-line treatment should be considered when the first steroids do not work. (Second accuracy pass; SITC 2021 full text from Europe PMC, read as raw text) · The step now says "consider" and the unsupported reason is removed.
- 2026-11-17 · Steroid taper after immunotherapy side effects, with Pneumocystis prophylaxis and adrenal checks: taper length: The page said steroids are tapered because stopping too fast can bring the side effect back, that the taper starts once the side effect has settled, and that SITC agrees with 4 to 6 weeks for most organs. The ASCO abstract gives neither the reason nor the starting condition, and SITC states 4 to 6 weeks only for hepatitis, pneumonitis, myocarditis and sicca syndrome, at different levels of evidence. Osteoporosis was also listed under a level of evidence that belongs to a different sentence. (Second accuracy pass; ASCO 2021 abstract and SITC 2021 full text from Europe PMC, read as raw text) · The reason and condition are removed, the organs are named with the level of evidence SITC prints for each, and osteoporosis is listed separately.
- 2026-11-17 · Steroid taper after immunotherapy side effects, with Pneumocystis prophylaxis and adrenal checks: 2024 adrenal guideline: The page said the full text of the 2024 European Society of Endocrinology and Endocrine Society guideline was not in PubMed Central and gave only one sentence from its abstract. The guideline was co-published in an open access journal that PubMed Central holds. (Second accuracy pass; full text of the Journal of Clinical Endocrinology and Metabolism co-publication from Europe PMC, read as raw text) · The guideline is now read in full: a step on the morning cortisol, four graded recommendations and its disagreement with ESE 2022 on Synacthen testing were added.
- 2026-11-17 · Steroid taper after immunotherapy side effects, with Pneumocystis prophylaxis and adrenal checks: rheumatic side effects: The variant said EULAR tapers with a conventional DMARD above 10 mg a day of prednisone. EULAR says a conventional DMARD should be considered in that case. (Second accuracy pass; EULAR 2021 full text from Europe PMC, read as raw text) · Reworded to "should be considered", with the level of agreement EULAR prints for that statement.
- 2026-11-17 · Restarting a checkpoint inhibitor after an immune-related side effect: the grade on stopping after myocarditis: One step covered both Guillain-Barré syndrome and myocarditis and carried the grade LE: 1. SITC prints LE: 1 only after its Guillain-Barré sentence. Its myocarditis sentence, that permanent discontinuation should be seriously considered, carries no level of evidence. (Second accuracy pass; SITC 2021 full text from Europe PMC, read as raw text) · The step was split in three: Guillain-Barré syndrome (LE: 1), grade 3 or 4 myasthenia gravis and myocarditis, each with its own quoted sentence, the last two with no grade.
- 2026-11-17 · Restarting a checkpoint inhibitor after an immune-related side effect: statements written as orders: Three steps turned permissions into instructions: restart after a grade 2 side effect, restart only once grade 2 pneumonitis has resolved, and continue or restart when an endocrine side effect is controlled. SITC says such patients may be re-challenged, and the European Society of Endocrinology says a controlled endocrinopathy is not a contraindication. The summary also dropped the words in general from the ASCO rule on grade 4. (Second accuracy pass; SITC 2021 and ESE 2022 full texts from Europe PMC and the ASCO 2021 abstract, read as raw text) · The steps now say consider restarting, and that a controlled endocrine side effect is not a reason against continuing. The ASCO hedge is restored.
- 2026-11-17 · Restarting a checkpoint inhibitor after an immune-related side effect: monitoring after restart: The step on monitoring after restarting applied to every organ but quoted a SITC sentence that is about diarrhoea and colitis only, and it said to treat a recurrence as before, which SITC says only of colitis that returns after steroids are tapered. (Second accuracy pass; SITC 2021 full text and the VigiBase cohort abstract from Europe PMC, read as raw text) · The step now rests on the VigiBase authors' conclusion about appropriate monitoring and standard treatment algorithms, names SITC's colitis and pneumonitis advice as such, and the gaps say no source read sets a schedule.
- 2026-11-17 · Restarting a checkpoint inhibitor after an immune-related side effect: reasons given for steps: The reasons given for ten steps went beyond the sources, among them that restarting on high-dose steroids risks a flare, that severe side effects are more likely to return, that bowel inflammation can persist without symptoms, that a replaced hormone shortage does not get worse with more treatment, and that these side effects can be fatal if they return. SITC says recurrence was higher after grade 3 or higher events in two small series but not significantly so, and the other reasons are not in any source read. (Second accuracy pass; SITC 2021 and ESE 2022 full texts from Europe PMC, read as raw text) · Each reason now says what the cited source says, with its denominator where there is one.
- 2026-11-17 · Restarting a checkpoint inhibitor after an immune-related side effect: patient summary and recurrence figure: The short summary said very severe reactions usually mean stopping for good and called restarting a shared decision, and a step told teams to tell the patient that about 3 in 10 have the same side effect again. SITC makes restarting after a grade 3 or 4 side effect a risk-adjusted decision, no source read uses the words shared decision, and the 28.8% is 130 of 452 informative rechallenges in reports to a WHO safety database. (Second accuracy pass; SITC 2021 full text and the VigiBase cohort abstract from Europe PMC, read as raw text) · The summary names the conditions and the strength of each statement, the figure carries its denominator and its origin, and the step asks for the risk to be weighed.
- 2026-11-17 · Restarting a checkpoint inhibitor after an immune-related side effect: switching drug class: The page said the sources read do not address switching to a different checkpoint inhibitor. SITC reports a colitis study in which restarting with an anti-CTLA-4 drug was a risk factor for recurrence and says switching may be considered after an infusion reaction, and the VigiBase abstract names recurrence by regimen as an outcome without giving the result. (Second accuracy pass; SITC 2021 full text and the VigiBase cohort abstract from Europe PMC, read as raw text) · The gap now says that no source read recommends switching to lower the risk, and states what each source does say.
- 2026-11-17 · Restarting a checkpoint inhibitor after an immune-related side effect: the kidney consensus: The page said the ADQI report could be read only as its abstract and prints no grades. The full text is readable on PubMed Central. Its statement 13 says rechallenge may be considered, with 100% consensus and low certainty of evidence, and the figure of under 20% comes from the largest single study. (Second accuracy pass; ADQI 2026 full text read as raw text from PubMed Central) · The source is marked as read in full, the statement is quoted with its consensus strength and certainty, and the variant says where the figure comes from.
- 2026-11-17 · Baseline tests and monitoring before and during checkpoint inhibitor treatment: heart tests before and during treatment: The page said the 2022 ESC cardio-oncology guideline could not be read and gave only the SITC position (an ECG for people at higher risk, troponin optional). The ESC guideline recommends an ECG, natriuretic peptide and troponin in all patients before starting (Class I, Level B) and says an ECG and troponin should be considered before doses 2, 3 and 4 and then every three doses (Class IIa, Level B). (Second accuracy pass; 2022 ESC cardio-oncology guideline, published PDF from an open university repository read as text, Recommendation Table 15) · ESC steps, grades and a variant added; the disagreement with SITC is stated, including SITC's remark that scheduled troponin has not been shown to improve outcome in people without symptoms.
- 2026-11-17 · Baseline tests and monitoring before and during checkpoint inhibitor treatment: best-supported version: The page said SITC and ESE agree on a baseline blood count and metabolic panel, liver tests before each dose and thyroid tests every 6 to 12 months after treatment. ESE lists none of these: its baseline is TSH, free T4, cortisol, glucose and electrolytes, repeated every 4 to 6 weeks. (Second accuracy pass; ESE 2022 full text from Europe PMC) · Statement rewritten so each test is credited to the guideline that asks for it.
- 2026-11-17 · Baseline tests and monitoring before and during checkpoint inhibitor treatment: pre-existing autoimmune disease: One step applied SITC's recommendation for life-threatening autoimmune disease to everyone with autoimmune disease, told the team to arrange specialist monitoring where SITC says it should be considered, and said the survival benefit outweighs the risk for many cancers, which SITC does not say. (Second accuracy pass; SITC 2021 full text from Europe PMC) · Split into two steps, one for life-threatening and one for non-life-threatening autoimmune disease, each with SITC's own sentence and hedges.
- 2026-11-17 · Baseline tests and monitoring before and during checkpoint inhibitor treatment: card, lung referral and reasons: The education step read as if an alert card were part of SITC's recommendation, where SITC says a reference card may be given. The lung step dropped SITC's words "where possible" and "for the consideration of". Several reasons said more than the sources: myocarditis "rare but often fatal" (SITC says uncommon with a high death rate), hepatitis "usually" silent (SITC says often), and side effects in general found on blood tests before symptoms. (Second accuracy pass; SITC 2021 and ESE 2022 full text from Europe PMC) · Steps and reasons reworded to the sources. The patient card is now a separate step resting on the European pembrolizumab label, which requires one.
- 2026-11-17 · Baseline tests and monitoring before and during checkpoint inhibitor treatment: transplant recipients: The page said nothing about people with an organ or donor stem cell transplant. No guideline read gives advice, but the European pembrolizumab label asks that benefit be weighed against the risk of organ rejection and of graft-versus-host disease. (Second accuracy pass; Keytruda product information PDF from the European Medicines Agency read as text) · Step added, scoped to pembrolizumab, with a gap saying the other labels were not read.
- 2026-11-17 · Lymphodepleting chemotherapy before CAR T-cell infusion: omitting it when counts are low: The page gave only the tisagenlecleucel label's allowance to omit lymphodepletion at a low white count, and said the EBMT 2019 checklists could not be read. The checklist does read, and it recommends lymphodepletion for all patients regardless of white count or lymphocyte count. (Second accuracy pass; EBMT/JACIE 2019 Tables 5A and 5B, read as text from the open Haematologica PDF) · The disagreement is now stated in the best-in-class statement and the variant, and the checklist items (product on site, infection controlled, creatinine clearance above 30 mL/min, AST and ALT below 5 times the upper limit) were added.
- 2026-11-17 · Lymphodepleting chemotherapy before CAR T-cell infusion: infection before starting: The step said to treat and control any active infection and cited SITC 2020, whose quoted sentence covers suspected fungal infection only. Its reason, that lymphodepletion and the cytokine surge make infection much worse, is in neither source. EBMT 2019 states the general rule and gives the reason as a likely cytokine-driven worsening of inflammation. (Second accuracy pass; SITC 2020 and EBMT/JACIE 2019 full text from Europe PMC) · Split into two steps, the general rule quoted from EBMT 2019 and the fungal rule from SITC 2020, each with the reason its source gives.
- 2026-11-17 · Lymphodepleting chemotherapy before CAR T-cell infusion: choice of regimen: The step told teams to use fludarabine with cyclophosphamide and said it gave better persistence and disease control than older regimens. EBMT 2019 only describes it as the most widely used regimen, and the comparison comes from one trial of 30 patients with B-ALL. (Second accuracy pass; EBMT/JACIE 2019 full text from Europe PMC) · The step now says to use the regimen in the product's label, and the reason names the trial's size and comparators.
- 2026-11-17 · Lymphodepleting chemotherapy before CAR T-cell infusion: reasons given for three steps: Three reasons were not in the labels cited: that starting without a product leaves a person with low counts and no treatment, that live vaccines cause infection in a suppressed immune system, and that all four reasons to delay raise the risk of severe cytokine release syndrome. The labels give no reason for the first, say the safety of live vaccines has not been studied, and place the delay instruction in the cytokine release section without that claim. The vaccine step also turned not recommended into a check that none had been given. (Second accuracy pass; YESCARTA and KYMRIAH US prescribing information, PDFs from fda.gov read with pdftotext) · Each reason now says what the label says, and the vaccine step reads avoid live vaccines for at least 6 weeks before lymphodepletion.
- 2026-11-17 · Lymphodepleting chemotherapy before CAR T-cell infusion: who it covers and where sources agree: The population included myeloma although no myeloma product was read, and the best-in-class statement said EBMT and SITC agree on confirming the product and close monitoring, which SITC does not state. (Second accuracy pass; both US labels, EBMT/JACIE 2019 and SITC 2020 read as raw text) · The population is scoped to the two products read, and the statement now says which source holds each point.
- 2026-11-17 · Lymphodepleting chemotherapy before CAR T-cell infusion: timing, repeat and kidney rules: The page left out the days allowed between lymphodepletion and tisagenlecleucel infusion, gave only guideline advice on repeating lymphodepletion after a delay, and said the kidney dose rule had not been read. The US label sets 2 to 14 days for B-ALL, 2 to 11 for DLBCL and 2 to 6 for follicular lymphoma. The European product information says to re-treat after more than 4 weeks with a white count above 1,000 cells per microlitre (Kymriah) or after more than 2 weeks (Yescarta). A US fludarabine label reduces the dose by 20% at a creatinine clearance of 30 to 70 and says not to give it below 30. (Second accuracy pass; KYMRIAH US label, Kymriah and Yescarta European product information PDFs from ema.europa.eu, and a fludarabine label from DailyMed) · Added a timing step, variants for the European rules including the cytarabine and etoposide alternative for B-ALL, and the fludarabine rule with its limits named in gaps.
- 2026-11-17 · Prolonged low blood counts after CAR T-cell therapy (ICAHT): stem cell boost step: The step told teams to give an autologous stem cell boost for low counts that do not respond to G-CSF. The 2025 Haematologica review calls the boost an emerging, promising option and its table says its use should be considered. (Second accuracy pass; Haematologica 2025 review, full text from Europe PMC and the publisher PDF read as text) · The step now says consider a boost, and gives the review's scope: grade 3 or higher early ICAHT not responding to G-CSF beyond day +14.
- 2026-11-17 · Prolonged low blood counts after CAR T-cell therapy (ICAHT): evidence levels in the summary and GM-CSF step: The summary gave SITC's evidence level 4 to antibacterial and antifungal prophylaxis, a sentence SITC prints with no level, and the growth factor step showed level 4 over the GM-CSF advice, which SITC prints at level 5. (Second accuracy pass; SITC 2020 guideline, full text XML from Europe PMC) · Each SITC recommendation now carries the level printed after it, prophylaxis is marked as having none, and the GM-CSF advice is marked LE: 5.
- 2026-11-17 · Prolonged low blood counts after CAR T-cell therapy (ICAHT): bone marrow steps: The page said a marrow biopsy is usually not needed in the first 28 days and that a later biopsy looks for relapse, infection of the marrow or a new marrow disorder. SITC says a biopsy may not be indicated, and that the later one assesses response and cellularity, including viral pathogens. (Second accuracy pass; SITC 2020 guideline, full text XML from Europe PMC) · Both steps now use SITC's wording; relapse and a new myeloid cancer are attributed to the 2025 review, which does say them.
- 2026-11-17 · Prolonged low blood counts after CAR T-cell therapy (ICAHT): alternatives and their evidence: The prednisone entry called its evidence a series of 17 patients, where the review says six of 17 patients who had responded lost that response when the steroid stopped. The bendamustine entry said no comparative trial was cited, where the review cites two studies. The allogeneic transplant step gave a reason, serious risks of its own, that the review does not give. (Second accuracy pass; Haematologica 2025 review, full text from Europe PMC and the publisher PDF read as text) · Each entry now says what the review says and no more.
- 2026-11-17 · Prolonged low blood counts after CAR T-cell therapy (ICAHT): patient summary and grading: The patient summary said most low counts recover within weeks, which no source read states. The page also said the ICAHT grade thresholds could not be given, although two open papers reprint the EHA/EBMT table. (Second accuracy pass; Haematologica 2025 review (publisher PDF, table read as text and by eye) and Rejeski and colleagues, Blood Advances 2024 (full text XML from Europe PMC)) · The summary now says early low counts often recover on their own, and the grade thresholds for early and late ICAHT are stated, with the one difference in notation between the two papers.
- 2026-11-17 · Bridging therapy while CAR T cells are manufactured: when bridging can be omitted: The page said bridging could be left out when lymphoma is stable and low-burden. EBMT/JACIE 2019 says it can be omitted "in the presence of stable, low burden disease" in its general paragraph, not for lymphoma only. (Second accuracy pass; EBMT/JACIE 2019 full text from Europe PMC and the Haematologica PDF, read as raw text) · The step and the best-in-class statement now say stable, low-burden disease.
- 2026-11-17 · Bridging therapy while CAR T cells are manufactured: the two-week gap before lymphodepletion: The step gave a two-week gap between bridging and lymphodepletion as general advice, with a reason the source does not give. It comes only from an industry-supported expert panel on myeloma before anti-BCMA CAR T, which says "if feasible" and states no reason. (Second accuracy pass; Rasche et al. 2026 full text from Europe PMC, read as raw text) · The step is now scoped to myeloma, says the panel gives no reason, and the gaps say no other source read gives a wash-out before lymphodepletion.
- 2026-11-17 · Bridging therapy while CAR T cells are manufactured: acute lymphoblastic leukaemia: The page said bridging is used to lower the blast count, usually with reduced doses, because more leukaemia means a higher risk of severe cytokine release syndrome. The source says the risk of CRS correlates with blast burden, that bridging is especially important, and that doses are often reduced; it does not say severe. (Second accuracy pass; EBMT/JACIE 2019 full text, read as raw text) · Step and summary reworded to the source's words.
- 2026-11-17 · Bridging therapy while CAR T cells are manufactured: who is bridged in myeloma: The page gave only the industry-supported panel's view that nearly everyone should be bridged, and said the International Myeloma Working Group guidance had not been read. That guidance is open access and says to strongly consider bridging for patients with high disease burden or at risk of morbidity during manufacturing. (Second accuracy pass; Costa et al., Leukemia 2025, full text from Europe PMC, read as raw text) · Added as a source, a step, a variant and a grade, and the disagreement is stated in the best-in-class statement.
- 2026-11-17 · Bridging therapy while CAR T cells are manufactured: talquetamab and the CARTITUDE-4 figure: The summary said talquetamab for heavily pretreated patients, where the panel says to consider it in those who have exhausted other effective therapies. The evidence line called the CARTITUDE-4 analysis retrospective, which the paper does not, and did not say that no patient numbers are given. (Second accuracy pass; Rasche et al. 2026 full text from Europe PMC, read as raw text) · Both reworded; the page now says the paper gives no patient numbers.
- 2026-11-17 · Bridging therapy while CAR T cells are manufactured: reasons and restaging: Several reasons were not in the sources: that chemotherapy before collection damages T cells, that long half-life drugs are still in the body at infusion, that unneeded treatment only adds side effects, and that disease burden guides monitoring. The SITC step said restage, where SITC says disease burden should be assessed before CAR T-cell therapy. (Second accuracy pass; EBMT/JACIE 2019 and SITC 2020 full texts, read as raw text) · Each reason reworded to what the source states; restage changed to assess.
- 2026-11-17 · Infection screening, prophylaxis and immunoglobulin replacement around CAR T-cell therapy: hepatitis B: The page told teams to give antiviral prophylaxis to everyone with a history of hepatitis B and said ASTCT gives entecavir for "at-risk" hepatitis B. ASTCT 2024 gives entecavir 0.5 mg daily to people who are HBsAg positive, and for people who are HBsAg negative and anti-HBc positive it offers either prophylaxis or monthly HBV DNA and liver enzyme checks. (Second accuracy pass; ASTCT 2024 author manuscript, Europe PMC full-text XML read raw) · Split into two steps quoted from ASTCT, with EBMT/JACIE 2019's tenofovir advice kept as a variant.
- 2026-11-17 · Infection screening, prophylaxis and immunoglobulin replacement around CAR T-cell therapy: where the guidelines differ: The page said the sources differed only on the immunoglobulin threshold and the hepatitis B medicine. EBMT/JACIE 2019 also continues antiviral and Pneumocystis prophylaxis until 1 year and/or a CD4 count above 0.2 x 10^9/L rather than 6 months, and does not recommend routine antibacterial prophylaxis, while SITC 2020 says all patients with persistent neutropenia should have antibacterial and antifungal prophylaxis. (Second accuracy pass; EBMT/JACIE 2019 Table 10 read from the open Haematologica PDF with pdftotext, SITC 2020 Europe PMC full text) · Both disagreements added to the best-in-class statement and as variants; the unread 2022 European edition named in gaps.
- 2026-11-17 · Infection screening, prophylaxis and immunoglobulin replacement around CAR T-cell therapy: immunoglobulin monitoring: One step cited SITC 2020 and its grade LE: 4 for measuring IgG before treatment and in the first 3 months. That advice is ASTCT's, ungraded and scoped to B-cell cancers; SITC's graded sentence is monthly B-cell counts and immunoglobulins. (Second accuracy pass; ASTCT 2024 and SITC 2020 full text from Europe PMC) · Split into an ASTCT step and a SITC step, each with its own quote.
- 2026-11-17 · Infection screening, prophylaxis and immunoglobulin replacement around CAR T-cell therapy: reasons given: Several reasons said more than the sources: that immunoglobulin replacement prevents infections (ASTCT says trial evidence is lacking), that tocilizumab hides infection (the drug is not named), that infections are the commonest serious problem after CAR T cells, and that treatment is delayed for any symptomatic chest infection (ASTCT says severe). (Second accuracy pass; ASTCT 2024 author manuscript, Europe PMC full-text XML read raw) · Each sentence reworded to what ASTCT states.
- 2026-11-17 · Infection screening, prophylaxis and immunoglobulin replacement around CAR T-cell therapy: herpes virus quote: The quoted ASTCT sentence began with the word "Antiviral", which is the section heading above it and not part of the sentence. (Second accuracy pass; quote checked by script against the Europe PMC full-text XML) · Heading word removed so the quote is the sentence as printed.
- 2026-11-17 · Infection screening, prophylaxis and immunoglobulin replacement around CAR T-cell therapy: fever advice for patients: The advice to ring the team at once for any fever had no source. The guidelines cited are written for clinicians; the Yescarta medication guide says to call or get emergency help right away for a fever of 38°C (100.4°F) or higher. (Second accuracy pass; YESCARTA US prescribing information revised 07/2026, FDA PDF read with pdftotext) · Label added as a source, a step added with its words quoted, and the patient summary reworded to 38°C or higher.
- 2026-11-17 · Leukapheresis to collect T cells for CAR T-cell manufacture: lymphocyte count before collection: The page told teams to aim for a lymphocyte count above 0.5 x 10^9/L. EBMT/JACIE 2019 sets no such target: its checklist recommends a count above 0.2 x 10^9/L (a threshold it ties to recovery after corticosteroids, higher in small children) and its text says only that manufacture is likelier to succeed above 0.5. The quoted sentence also printed the figures as 0.1×109/L, a mangled superscript. (Second accuracy pass; EBMT/JACIE 2019 text and Table 3, open Haematologica PDF read as text) · Step rewritten with the checklist values for haemoglobin, haematocrit, neutrophils, platelets and lymphocytes; the mangled quote removed and the figures stated in prose.
- 2026-11-17 · Leukapheresis to collect T cells for CAR T-cell manufacture: washout periods: The page said the EBMT washout table could not be read and gave no interval for any treatment. The table reads in the open PDF: 4 weeks after donor lymphocyte infusion, 3 to 4 weeks after high-dose chemotherapy, 1 week after CNS-directed therapy, 3 days after short-acting cytotoxic drugs, ideally 7 days off systemic corticosteroids, and off immunosuppression and free of graft-versus-host disease after a donor transplant. (Second accuracy pass; EBMT/JACIE 2019 Table 4, open Haematologica PDF read as text) · Two steps added with the intervals and the table's comments; the gap rewritten.
- 2026-11-17 · Leukapheresis to collect T cells for CAR T-cell manufacture: infection tests: The page gave only testing on the day of collection and said regulations require it. EBMT/JACIE 2019 also says serology for hepatitis B, hepatitis C, HIV, syphilis and HTLV is to be done within 30 days of leukapheresis with results available at collection and shipment, and calls it mandatory in some countries. (Second accuracy pass; EBMT/JACIE 2019 text and Tables 2 and 3, open Haematologica PDF read as text) · A step added for the 30-day serology; the reason reworded to what the source says; both timings named in the gaps.
- 2026-11-17 · Leukapheresis to collect T cells for CAR T-cell manufacture: timing against chemotherapy: The page told teams to collect before salvage chemotherapy. EBMT/JACIE 2019 says it may be possible to arrange this, depending on disease burden. (Second accuracy pass; EBMT/JACIE 2019 full text read as text) · Reworded to consider arranging collection before salvage chemotherapy.
- 2026-11-17 · Leukapheresis to collect T cells for CAR T-cell manufacture: reasons given for steps: Three reasons went beyond the sources: that collected cells must go straight into a booked production run (EBMT says only that manufacturing capacity is limited, and describes cells frozen for up to 18 months), that earlier treatments do not stop CAR T cells working (SITC says only that such patients may be treated), and that any approved device may be used (EBMT says any commercially available device is suitable in principle). (Second accuracy pass; EBMT/JACIE 2019 and SITC 2020 full texts read as text) · Each reworded to the source.
- 2026-11-17 · Leukapheresis to collect T cells for CAR T-cell manufacture: how long it takes and how it is done: The patient summary said collection takes a few hours and described blood being returned, with no source; the page also said the EBMT supplement on technique had not been read, and said without a source that several manufacturers had changed their requirements since 2019. (Second accuracy pass; KYMRIAH US medication guide revised 6/2025 (fda.gov PDF) and the EBMT/JACIE 2019 online supplement (haematologica.org PDF), both read as text) · The duration now cites the tisagenlecleucel patient guide (3 to 6 hours, may need repeating); steps added from the supplement on venous access, identity and labelling, anticoagulant, collection target and sealing; the unsourced sentence about manufacturers removed.
- 2026-11-17 · Referral and eligibility assessment for CAR T-cell therapy: how long to stay near the hospital: The page told patients they would need to stay near the hospital for at least two weeks, as if that were general. Two weeks is the revised US labelling. The European Kymriah label says at least 4 weeks within 2 hours of travel and no driving for 8 weeks, the European Yescarta label says at least 2 weeks and no driving for at least 4 weeks, EBMT 2019 says at least 4 weeks with a caregiver always present, and the US idecabtagene vicleucel label says at least one week. (Second accuracy pass; the FDA communication, the seven US labels on DailyMed, the European Yescarta and Kymriah product information and the EBMT/JACIE 2019 PDF, read as raw text) · The summary for patients now says often two to four weeks depending on the product and the country; the step is scoped to the US, a European step and a variant give the longer periods, and the one-week label is named.
- 2026-11-17 · Referral and eligibility assessment for CAR T-cell therapy: grade on performance status: One step joined two SITC recommendations and printed LE: 4 for both. SITC prints LE: 4 only after the recommendation on intracranial haemorrhage; its recommendation on performance status carries no evidence level. (Second accuracy pass; SITC 2020 full-text XML from Europe PMC) · The step was split in two; the performance status step now carries SITC's own sentence and no grade.
- 2026-11-17 · Referral and eligibility assessment for CAR T-cell therapy: timing of referral and cell collection: The step said to refer early, before further lines of chemotherapy. EBMT/JACIE 2019 says something narrower: depending on the disease burden, it may be possible to arrange cell collection before salvage chemotherapy, because cumulative chemotherapy affects T-cell quality. (Second accuracy pass; EBMT/JACIE 2019 open PDF read through pdftotext) · The step now says to consider arranging collection before salvage chemotherapy where the disease burden allows.
- 2026-11-17 · Referral and eligibility assessment for CAR T-cell therapy: eligibility thresholds: The page said no organ-function thresholds could be given and that the EBMT eligibility tables could not be read. They read as text in the open PDF: ECOG performance status above 2 not recommended, no upper age limit, creatinine clearance above 30 mL/min, ejection fraction above 40%, neutrophils above 1.0 x 10^9/L, AST and ALT below 5 times the upper limit of normal. The page also said SITC and EBMT agree on the work-up, when SITC leaves the levels to each institution. (Second accuracy pass; EBMT/JACIE 2019 open PDF, Tables 1 and 2, read through pdftotext) · Two steps and the summary now give the EBMT thresholds, the comparison with SITC is stated as it is, and the page says the 2022 update could not be read in full.
- 2026-11-17 · Referral and eligibility assessment for CAR T-cell therapy: reasons given for steps: Several reasons went beyond the sources: that side effects usually start in the first two weeks, that a recent brain bleed makes neurotoxicity especially dangerous, that heart function must be adequate, that earlier treatments do not stop CAR T cells working, and neurological referral was limited to lymphoma where SITC says any current disease in the central nervous system. Myeloma was summarised as within the labels, where EBMT 2025 rates CAR T cells S/II after three prior lines. (Second accuracy pass; SITC 2020, EBMT 2025 and the FDA communication, read as raw text) · Each reason was reworded to what its source says, and the myeloma line now gives the EBMT table row with the current US label wording alongside.
- 2026-11-17 · Opioid switching and dose conversion in adults with cancer pain: conversion ratios: The page gave only the WHO annex ratios and no figure at all for fentanyl patches, buprenorphine patches or diamorphine, although its own sources differ. WHO divides oral morphine by 1.5 for oral oxycodone and the Scottish guidelines by 2. A 12 microgram an hour fentanyl patch is 30 mg of oral morphine a day in WHO, 45 mg in the Australian guideline and 30 to 60 mg in the Scottish guidelines. A 20 microgram an hour buprenorphine patch is 30 mg in the Australian guideline while the Scottish guidelines put 5 micrograms an hour at 12 mg. Only the Scottish guidelines print subcutaneous diamorphine, oral morphine divided by 3. (Second accuracy pass; WHO 2018 PDF through pdftotext, the Australian guideline from PubMed Central, and three Scottish Palliative Care Guidelines pages fetched with curl) · Each source's ratio is now on the page with its direction and who states it, and the Scottish guidelines were added as a source.
- 2026-11-17 · Opioid switching and dose conversion in adults with cancer pain: starting dose reduction: A variant described a fixed reduction of 25 to 50% at every switch. That figure is in no source read. WHO calls around 50% prudent in four named situations, the Scottish guidelines say to consider up to 30% when changing opioid, and the Australian guideline sets no figure. (Second accuracy pass; WHO 2018 annex 6, Australian guideline P18 and the Scottish conversion tables page, all as raw text) · The variant now states the three sourced positions and the unsourced figure is gone.
- 2026-11-17 · Opioid switching and dose conversion in adults with cancer pain: methadone: The page said conversion to or from methadone is left to specialists. The Australian guideline says methadone should be initiated and titrated only by specialists, and WHO recommends specialist guidance for conversions to regular methadone. Neither speaks of switching away from it. (Second accuracy pass; Australian guideline P4 and WHO 2018 table A6.2 footnote, raw text) · The step and summary now use each source's own scope, and the Scottish advice of inpatient admission was added.
- 2026-11-17 · Opioid switching and dose conversion in adults with cancer pain: monitoring after a switch: The page said to monitor closely in the first days for drowsiness, confusion and slow breathing. WHO asks for careful clinical monitoring during conversion to avoid under-dosing, excessive dosing and adverse effects, and gives no number of days and no symptom list. (Second accuracy pass; WHO 2018 annex 6, PDF through pdftotext) · The step and the patient summary now match WHO, and the gaps say no source read sets a time limit.
- 2026-11-17 · Opioid switching and dose conversion in adults with cancer pain: adding up the dose: The page said to add regular and rescue doses and that leaving out rescue doses underestimates need. The cited passage says only to look up the conversion for the total 24 hour opioid dose. (Second accuracy pass; Australian guideline P18 from PubMed Central, raw text) · The step now says total 24 hour dose and the unsupported reason was replaced.
- 2026-11-17 · Opioid switching and dose conversion in adults with cancer pain: grades on three steps: Three steps carried grades the source does not print. The side effect half of the switching recommendation comes from ASCO, not ESMO, NCCN and NICE, and two sentences of P18 name no source guideline at all. (Second accuracy pass; Australian guideline P18 from PubMed Central, raw text) · The grades now give the attribution printed for each sentence, or say none is printed.
- 2026-11-17 · Opioid switching and dose conversion in adults with cancer pain: before switching: The first step said to rule out other causes and to reduce the dose if pain is controlled for any side effect. NICE says to optimise laxatives and anti-emetics before considering a switch, and to consider a dose reduction for central nervous system effects when pain is controlled. (Second accuracy pass; NICE CG140 recommendations page, plain curl) · The step now follows NICE 1.1.20, 1.1.23 and 1.1.25.
- 2026-11-17 · Opioid switching and dose conversion in adults with cancer pain: kidney disease: The page said nothing on which opioid to choose in kidney disease, and the sources differ. The Australian guideline prefers buprenorphine, fentanyl, hydromorphone, methadone or oxycodone in severe kidney disease. The Scottish guidelines say to avoid hydromorphone and diamorphine in stage 4 and 5 chronic kidney disease. (Second accuracy pass; Australian guideline P6 and P7 and the Scottish choosing and changing opioids page, raw text) · A step and a variant now state both positions.
- 2026-11-17 · Opioid switching and dose conversion in adults with cancer pain: dose threshold and reasons: The 50% reduction step said around 1 g of oral morphine a day and explained that these are the situations most likely to cause overdose. WHO says morphine or equivalent doses of 1 g or more in 24 hours, without naming the route, and gives no such reason. (Second accuracy pass; WHO 2018 annex 6, PDF through pdftotext) · The step now uses WHO's wording and the invented reason was removed.
- 2026-11-17 · Strong opioids for cancer pain: summary and best in class: The page said NICE and WHO agree that oral morphine comes first. WHO says any opioid may be considered for maintenance and found no obviously best opioid; only NICE offers oral morphine first. (Second accuracy pass; WHO 2018 full text, PDF read with pdftotext, and NICE CG140 recommendations page read raw) · Summary and best-in-class statement rewritten to give each body's position.
- 2026-11-17 · Strong opioids for cancer pain: drowsiness and confusion step: The step and the summary told the prescriber to reduce the dose or switch opioid. NICE 1.1.25 says to consider dose reduction if pain is controlled or consider switching if it is not, for persistent or moderate-to-severe central nervous system side effects; it does not name confusion. (Second accuracy pass; NICE CG140 recommendations page read raw) · Both options worded as consider, in the source's own terms.
- 2026-11-17 · Strong opioids for cancer pain: choice of morphine form: The step gave the choice of sustained-release or immediate-release morphine by patient preference and cited WHO. WHO does not mention preference; NICE 1.1.4 does. (Second accuracy pass; WHO 2018 full text and NICE CG140 recommendations page read raw) · Step now cites and quotes NICE 1.1.4; the WHO sentence moved to the grades table.
- 2026-11-17 · Strong opioids for cancer pain: constipation after laxatives: The step said to consider a peripherally acting opioid antagonist where laxatives have failed and explained how the drugs work. NICE 1.1.21 only points to its appraisals of naldemedine and naloxegol, and no source read describes the mechanism. (Second accuracy pass; NICE CG140, TA651 and TA345 pages read raw) · Step reworded as a pointer, with each appraisal's own wording; the mechanism sentence removed.
- 2026-11-17 · Strong opioids for cancer pain: nausea and drowsiness: The page said nausea and mild drowsiness are common and usually pass. NICE says each may occur, that nausea is likely to be transient and that drowsiness is often transient. The Australian guideline's routine as-required anti-emetic from the start was not on the page. (Second accuracy pass; NICE CG140 and the Australian guideline (PMC article page) read raw) · Reworded; drowsiness and the driving warning given their own step; anti-emetic difference added as a variant.
- 2026-11-17 · Strong opioids for cancer pain: starting and rescue doses from NCCN: The page said NCCN could be read only as an abstract and gave none of its doses. The journal article for version 2.2025 now reads in full and prints 5 to 7.5 mg (elsewhere 5 to 15 mg) of oral morphine for opioid-naive adults and a rescue dose of usually 10 to 20% of the daily opioid; nccn.org lists version 2.2026. (Second accuracy pass; JNCCN article page read raw) · NCCN doses added beside NICE, WHO and ASCO; version 2.2026 named as unread.
- 2026-11-17 · Strong opioids for cancer pain: reasons given for steps: Several reasons were not in any source read: that fear stops many people taking enough, that constipation never wears off, that changing laxative usually works, that fentanyl is kept for when morphine fails, and a symptom list for kidney and liver failure. (Second accuracy pass; all cited sources read raw) · Each reason replaced with what the sources say, or removed.
- 2026-11-17 · Strong opioids for cancer pain: advice to patients: The simple summary called morphine safe and listed symptoms to report without a source. ASCO's wording was also given as lowest effective dose where the abstract says lowest possible dose. (Second accuracy pass; NHS morphine page and NICE information for the public read raw; ASCO abstract from Europe PMC) · Reworded to the NHS and NICE wording, with the NHS page added as a source; ASCO wording corrected.
- 2026-11-17 · Strong opioids for cancer pain: breakthrough and breathing steps: The rescue step left out that NICE 1.1.14 applies to people on maintenance oral morphine, and the naloxone step said only that breathing is slow where the Australian guideline gives a respiratory rate of 8 a minute or less. (Second accuracy pass; NICE CG140 and the Australian guideline read raw) · Scope and threshold added.
- 2026-11-17 · Syringe driver (syringe pump): setting up and checking a continuous subcutaneous infusion in palliative care: pump versions variant: The page said other pump models (T34 version 2, CME version 3) have different menus and alarms from the BD BodyGuard T. The Scottish guideline says the versions differ in labelling and that there are no differences in preparing and operating them. (Second accuracy pass; Scottish Palliative Care Guidelines version differences page, raw HTML read as text) · Variant rewritten to the guideline's statement, with the MHRA's 2021 summary of T34 problems and alert MDA/2020/007 added as sources.
- 2026-11-17 · Syringe driver (syringe pump): setting up and checking a continuous subcutaneous infusion in palliative care: share of incidents: The page gave about a quarter of palliative medication incidents as the finding of national incident data for 2016 to 2021. That figure is the paper's background statement citing earlier work. The study itself counted 1,317 infusion incidents among 7,506 palliative medication incidents. (Second accuracy pass; Brown et al., Palliative Medicine, full text from Europe PMC) · Summary now gives the study's own counts, and the grades row says the 25% figure is background.
- 2026-11-17 · Syringe driver (syringe pump): setting up and checking a continuous subcutaneous infusion in palliative care: two quotations: Two quotations from the Scottish guideline were not character for character: one had straight quote marks and added spaces around START INFUSION and RESUME, and one had a straight apostrophe where the source prints a typographic one. (Second accuracy pass; Scottish Palliative Care Guidelines monitoring page, checked by script) · Both quotations replaced with the source's exact characters.
- 2026-11-17 · Syringe driver (syringe pump): setting up and checking a continuous subcutaneous infusion in palliative care: reasons given for steps: Nine reasons said more than the sources do, among them that medicine is absorbed poorly from swollen skin, that incompatible mixtures block the line, that the pump takes hours to reach a steady level, that accidental button presses change the rate, and that many people link a syringe driver with dying. The guideline gives none of these. It says a syringe pump can sometimes be associated with end of life care. (Second accuracy pass; Scottish Palliative Care Guidelines and NICE NG31, raw text) · Each reason reworded to what the cited passage says.
- 2026-11-17 · Syringe driver (syringe pump): setting up and checking a continuous subcutaneous infusion in palliative care: when a pump is considered: The first step and the trigger joined the NICE threshold of more than 2 or 3 as-needed doses in 24 hours to the failing oral route as one NICE recommendation, and a variant said NICE reserves the pump for that threshold with separate doses as the default. NICE 1.5.6 is a consider recommendation for the last days of life, and the oral route reasons come from the Scottish guideline. (Second accuracy pass; NICE NG31 recommendations page and Scottish Palliative Care Guidelines, raw text) · Step, trigger and variant reworded, each reason attributed to its source. The lowest effective dose is now scoped to people who have not had symptom medicines before, as NICE 1.5.7 says.
- 2026-11-17 · Syringe driver (syringe pump): setting up and checking a continuous subcutaneous infusion in palliative care: what patients should report: The patient summary told people to call the nurses if the pump alarms, the skin is sore or symptoms are not controlled, with no source cited for the list. The guideline's own patient leaflet lists the alarm, the light not flashing, leakage, red, swollen or painful skin, a dislodged tube, and a change of colour or cloudiness in the liquid. (Second accuracy pass; Scottish Palliative Care Guidelines patient information leaflet, issue date 22 February 2024, PDF read with pdftotext) · Leaflet added as a source and the patient wording matched to it.
- 2026-11-17 · Syringe driver (syringe pump): setting up and checking a continuous subcutaneous infusion in palliative care: checks and mixing: The page said to check 15 to 30 minutes after set-up and to mix no more than three medicines without advice. The guideline says within the first 15 to 30 minutes after set up, reloading or resiting, and allows more than three where the health board has already agreed it. (Second accuracy pass; Scottish Palliative Care Guidelines, raw HTML read as text) · Wording corrected, and the guideline's limits on breakthrough bolus volume and its rule that cyclizine needs water for injection added.
- 2026-11-17 · Antiemetic prophylaxis for chemotherapy by emetic risk: breakthrough nausea and vomiting: The step said to give olanzapine 10 mg once daily for 3 days whenever sickness breaks through. MASCC/ESMO suggest olanzapine only if it was not already used for primary prophylaxis, and say some evidence supports 10 mg daily for 3 days. People on high-risk regimens already receive olanzapine. (Second accuracy pass; MASCC/ESMO 2023 multiple-day and breakthrough paper and the ESMO Open summary, full text from Europe PMC) · Step, summary and grade now carry the qualifier, and the quote is the full recommendation sentence.
- 2026-11-17 · Antiemetic prophylaxis for chemotherapy by emetic risk: oxaliplatin: The page said to add an NK1 receptor antagonist for women under 50 having oxaliplatin. The guideline recommendation is a suggestion for women aged 50 or under, graded III, B, with palonosetron preferred and no evidence for routine use in women over 50. (Second accuracy pass; ESMO Open summary of the MASCC/ESMO 2023 guideline, full text from Europe PMC) · Oxaliplatin now has its own step worded as a suggestion with the grade, and the difference from the abstract wording is named in the gaps.
- 2026-11-17 · Antiemetic prophylaxis for chemotherapy by emetic risk: days 2 to 4 after high-risk chemotherapy: The page said olanzapine alone continues after anthracycline with cyclophosphamide, and left aprepitant out of the delayed phase after cisplatin. The guideline says that if aprepitant 125 mg is used on day 1, aprepitant 80 mg should be given on days 2 and 3. (Second accuracy pass; MASCC/ESMO 2023 high emetic risk paper, Table 2, full text from Europe PMC) · Both steps, the summary and the grade row now include the aprepitant doses and no longer say alone.
- 2026-11-17 · Antiemetic prophylaxis for chemotherapy by emetic risk: low emetic risk: The page said to give one anti-sickness drug for low-risk chemotherapy. The guideline says a single agent, such as dexamethasone, a 5-HT3 receptor antagonist or metoclopramide, may be considered, graded II, B. (Second accuracy pass; ESMO Open summary of the MASCC/ESMO 2023 guideline, full text from Europe PMC) · Reworded to consider, with the agents and the grade.
- 2026-11-17 · Antiemetic prophylaxis for chemotherapy by emetic risk: high-dose and multiple-day chemotherapy: The high-dose step named aprepitant only, where the recommendation is for an NK1 receptor antagonist, and omitted that olanzapine is given at bedtime and continued for 2 to 3 days. The multiple-day cisplatin step carried the grade I/A on a sentence about the number of days, which the guideline grades II, B. (Second accuracy pass; MASCC/ESMO 2023 multiple-day and high-dose paper, full text from Europe PMC) · Both steps reworded; the multiple-day grade now gives both parts and the aprepitant doses of 125 mg then 80 mg.
- 2026-11-17 · Antiemetic prophylaxis for chemotherapy by emetic risk: olanzapine dose and ASCO: A variant called 5 mg of olanzapine a standard dose in ASCO 2020 and the page said the guidelines differ on dose. The ASCO abstract lists 5 mg as an option, as MASCC/ESMO do. The step also ordered 10 mg, where the guideline says 10 mg is the best investigated dose, once daily for 4 days. (Second accuracy pass; ASCO 2020 abstract from Europe PMC and the MASCC/ESMO 2023 high emetic risk paper) · Variant renamed, the claimed difference removed, and the step reworded with the 4 day schedule.
- 2026-11-17 · Antiemetic prophylaxis for chemotherapy by emetic risk: sources and gaps: The page said the ESMO Open summary of the guideline could not be read, that no guideline read covers oral anticancer drugs, and that the SPARED trial was published as an abstract only. The summary is open in PubMed Central, it says that in general only on-demand antiemetics are recommended for oral drugs, and SPARED was published in full in 2024. (Second accuracy pass; ESMO Open summary read in full from Europe PMC, SPARED abstract from Europe PMC) · Summary guideline now read and cited with grades for moderate, low and minimal risk; oral drug wording corrected; SPARED added as a source with its result.
- 2026-11-17 · Antiemetic prophylaxis for chemotherapy by emetic risk: reasons given for steps: Several reasons said more than the sources: that sickness is now largely preventable, that sickness after cisplatin peaks a few days later, a list of steroid side effects, and that nausea in advanced cancer is covered by separate guidance. None is in the texts read. (Second accuracy pass; all cited MASCC/ESMO texts read as raw text) · Removed or reworded to what the sources state.
- 2026-11-17 · Anticoagulation for a blood clot in a person with cancer: how long treatment lasts: The short summary for patients said treatment is usually for at least six months. That is the ITAC and ASCO position only. NICE NG158 offers 3 to 6 months and then a review. (Second accuracy pass; NICE NG158 recommendation 1.3.15, read as raw HTML) · The summary now says at least three to six months and often longer while the cancer is active.
- 2026-11-17 · Anticoagulation for a blood clot in a person with cancer: clots found by chance: The reason given was that clots found by chance carried the same risks as symptomatic ones in the trials. ITAC does not say that. It reports a meta-analysis of 774 people with incidental clots in which tablets and injections did not differ in new clots or major bleeding. (Second accuracy pass; ITAC 2022 full text from Europe PMC) · The reason now states what the meta-analysis compared.
- 2026-11-17 · Anticoagulation for a blood clot in a person with cancer: bleeding with the tablets: The page said the tablets caused more bleeding from tumours of the stomach, bowel and urinary tract, and that injections may be safer. ITAC says the data show more gastrointestinal bleeding with edoxaban and rivaroxaban and asks for caution in gastrointestinal cancers. ASCO asks for caution in gastrointestinal and genitourinary cancers. (Second accuracy pass; ITAC 2022 full text from Europe PMC and the ASCO 2023 update from an Internet Archive capture) · Reworded to the caution each guideline prints.
- 2026-11-17 · Anticoagulation for a blood clot in a person with cancer: a clot that recurs on treatment: The step told the reader to increase the heparin dose or switch drug. ITAC lists these as three options that can be considered, on very low quality evidence. ASH suggests a higher dose or continuing the same dose. (Second accuracy pass; ITAC 2022 from Europe PMC and ASH 2021 full text from PubMed Central) · The step now says consider, lists all three options, and the ASH position is on the page.
- 2026-11-17 · Anticoagulation for a blood clot in a person with cancer: vena cava filters: The page gave one position on filters. The guidelines differ. ITAC says a filter might be considered if anticoagulation is contraindicated or a pulmonary embolism recurs. ASH suggests not using a filter for recurrence despite anticoagulation. NICE allows one for recurrence only after its treatment failure steps. (Second accuracy pass; ASH 2021 full text from PubMed Central and NICE NG158 recommendations 1.7.1 to 1.7.3) · All three positions are now stated with their grades.
- 2026-11-17 · Anticoagulation for a blood clot in a person with cancer: reasons given for five steps: Reasons for the steps on drug interactions, review at six months, low platelets, kidney failure and filters stated mechanisms the cited guideline does not state, such as drugs building up in kidney failure. The six month step also listed bleeding as a factor where ITAC lists benefit and risk, tolerability, drug availability, preference and cancer activity. (Second accuracy pass; ITAC 2022, ASCO 2023 and NICE NG158 read as raw text) · Each reason now says only what its source says.
- 2026-11-17 · Anticoagulation for a blood clot in a person with cancer: edoxaban lead-in and the ASCO grading scale: The best in class statement said edoxaban follows 5 days of injections where ITAC says at least 5 days. The ASCO evidence scale was given as high to low where ASCO prints high, intermediate, low or insufficient. (Second accuracy pass; ITAC 2022 from Europe PMC and the ASCO 2023 update from an Internet Archive capture) · Both corrected.
- 2026-11-17 · Blood clot risk assessment and preventive anticoagulation during outpatient cancer treatment: who is offered a blood thinner: The step for a Khorana score of 2 or more said that under ASCO the choice was apixaban, rivaroxaban or low-molecular-weight heparin. The ASCO abstract, the only ASCO text that could be read, says only that these may be offered to selected high-risk outpatients and gives no score. The American Society of Hematology 2021 guideline, which takes a different position for intermediate and high risk, was missing from the page. (Second accuracy pass; ASCO 2020 abstract from Europe PMC, ASH 2021 full text from PubMed Central, ITAC 2022 full text from Europe PMC) · ASCO is no longer tied to a score of 2. ASH 2021 is added as a source with its recommendations for low, intermediate and high risk and their grades.
- 2026-11-17 · Blood clot risk assessment and preventive anticoagulation during outpatient cancer treatment: bleeding and doses: The page said prevention at a score of 2 or more came without a rise in major bleeding, and that no source read gave a dose. That is what one pooled analysis cited by ITAC found, but the AVERT trial itself reported major bleeding in 3.5% on apixaban against 1.8% on placebo. The trial abstracts do print the doses tested: apixaban 2.5 mg twice daily and rivaroxaban 10 mg daily. (Second accuracy pass; AVERT and CASSINI abstracts from Europe PMC, ITAC 2022 full text) · The AVERT bleeding figure is stated beside the pooled result, and the two trial doses and the event counts with their denominators are on the page, labelled as trial doses.
- 2026-11-17 · Blood clot risk assessment and preventive anticoagulation during outpatient cancer treatment: how long: The page said that wherever prevention is given it continues for as long as chemotherapy does. NICE says this only of its two groups, myeloma on thalidomide, pomalidomide or lenalidomide and pancreatic cancer. The other guidelines read set no length for an oral anticoagulant at a high score. (Second accuracy pass; NICE NG89 recommendations page, read as HTML) · The step is limited to the two NICE groups, and the 180 days used in the CASSINI and AVERT trials is stated.
- 2026-11-17 · Blood clot risk assessment and preventive anticoagulation during outpatient cancer treatment: warning signs: The page told patients to report signs of a clot the same day and to tell their team at once. No cited source gave a time limit or a list of signs. (Second accuracy pass; NHS page on deep vein thrombosis, read as HTML) · The NHS page is now cited and the wording follows it: an urgent GP appointment or NHS 111 for a suspected clot, and 999 or A&E if there is also shortness of breath or chest pain.
- 2026-11-17 · Blood clot risk assessment and preventive anticoagulation during outpatient cancer treatment: reasons given for steps: Three reasons were in no source read: that for most people the chance of a clot is lower than the chance of a bleed, that the raised risk lasts as long as the treatment, and that myeloma drugs with steroids raise the risk substantially. The page also said the guidelines agree that pancreatic cancer and myeloma are exceptions, when only ITAC and NICE name both. (Second accuracy pass; Khorana 2008 full text from PubMed Central, ITAC 2022 and ASH 2021 full texts) · Each reason is replaced with what the sources print, including the share of patients at low risk in the Khorana study and the aspirin against heparin comparison in myeloma, and the agreement is stated body by body.
- 2026-11-17 · Blood clot risk assessment and preventive anticoagulation during outpatient cancer treatment: the Khorana score: The page said the cancer sites that score 1 or 2 points could not be listed because only the abstract had been read, and its quotation printed platelet and white cell counts with mangled superscripts. (Second accuracy pass; Khorana 2008 full text from PubMed Central) · The sites are now listed (stomach and pancreas 2 points; lung, lymphoma, gynaecological, bladder and testicular 1 point), the paper's own inconsistency between its tables is noted, and the quotation is replaced with a sentence from the full text.
- 2026-11-17 · Blood clot prevention around cancer surgery: where the guidelines differ: The page gave one course as if agreed and did not say when the first dose is given. ITAC 2022 recommends starting heparin 2 to 12 hours before the operation (grade 1A) while ASH 2021 suggests starting after it (conditional, low certainty); ITAC does not recommend stockings or leg pumps alone unless drugs are contraindicated (grade 2A); and ITAC judged the evidence insufficient for tablets in place of LMWH (grade 2B) the year before ASCO added them. (Second accuracy pass; ITAC 2022 full text (Europe PMC XML) and ASH 2021 full text (PubMed Central), read raw) · ITAC and ASH added as sources, with a step on first-dose timing, a step on renal impairment, four variants stating who holds each position, and the grades as printed.
- 2026-11-17 · Blood clot prevention around cancer surgery: the 28-day course: The step and the best-in-class statement credited NICE with extended prophylaxis after surgery in the abdomen or pelvis. NICE NG89 recommendation 1.14.4 says only major cancer surgery in the abdomen; the pelvis is named by ITAC, ASH and the Cochrane review. (Second accuracy pass; NICE NG89 recommendations page, read raw) · The step now says abdomen for NICE, and each body's scope and grade is stated separately.
- 2026-11-17 · Blood clot prevention around cancer surgery: tablets for the extended course: The page said tablets replace four weeks of daily injections. ASCO 2023 offers rivaroxaban or apixaban after an initial period of LMWH or unfractionated heparin, and in the rivaroxaban trial tablets began about 7 days after surgery. The page also gave ASCO's grading scale with two levels missing. (Second accuracy pass; ASCO 2023 update full text from an Internet Archive capture of the publisher page, read raw) · Reworded to say the tablets follow an initial period of injections, with the trial figures ASCO prints and the evidence quality (low) beside the weak recommendation; grading scale corrected.
- 2026-11-17 · Blood clot prevention around cancer surgery: warning signs and how fast to act: The page told people to seek help at once for calf pain, breathlessness or chest pain and cited NICE NG89, which says only that people should be told the signs and the importance of seeking help and gives no symptom list or time limit. (Second accuracy pass; NICE NG89 read raw, and the NHS page on deep vein thrombosis (last reviewed 30 April 2026) read raw) · The NHS page is added as a source and the advice now follows it: an urgent GP appointment or NHS 111 for a suspected clot in the leg, and 999 or A&E if there is also breathlessness or chest pain.
- 2026-11-17 · Blood clot prevention around cancer surgery: plain-language summary and reasons: The summary for patients said injections are usually given and often continued for four weeks, and several reasons given for steps were not in the cited source (that stockings carry no bleeding risk, that they cause pressure sores, that injections beat stockings alone, that people on anticoagulants need a bridging plan). NICE says to add injections where clot risk outweighs bleeding risk and to consider the 28-day course. (Second accuracy pass; NICE NG89 recommendations page and the Cochrane review full text (PubMed Central), read raw) · Reworded to what NICE says; the reason for the longer course now cites the Cochrane review (risk raised for four to six weeks, up to a third of clots after discharge).
- 2026-11-17 · Blood clot prevention around cancer surgery: thoracic surgery and NICE's review: The page said thoracic surgery follows the same pattern with every step a 'consider', and that it was not known whether NICE would add oral drugs. NICE says to start mechanical prophylaxis on admission for thoracic surgery and to consider adding LMWH first, fondaparinux if LMWH is contraindicated; and NICE reviewed NG89 in September 2024 and decided not to update it. (Second accuracy pass; NICE NG89 recommendations and overview pages, read raw) · Both statements corrected.
- 2026-11-17 · Catheter-related thrombosis in people with cancer: warning signs and urgency: The page told patients to report a swollen face and swollen chest veins and to tell their team the same day. No cited source said either. Macmillan's page on PICC lines lists swelling, redness or tenderness in the arm, chest area or neck, a swollen hand, shortness of breath and tightness in the chest, says to contact the hospital doctor or nurse, and gives no time limit. (Second accuracy pass; Macmillan Cancer Support, PICC lines, read as raw HTML) · Signs reworded to Macmillan's list, the same-day wording removed, Macmillan added as a source, and a note added that no source read gives a time limit.
- 2026-11-17 · Catheter-related thrombosis in people with cancer: steps resting on ESMO 2015: Two steps gave a duration or threshold from the ESMO 2015 guideline alone (3 to 5 days of anticoagulation before removing a line, and removal if the clot reaches the superior vena cava), with a third figure (about 3 months after removal) in the grades. ESMO 2015 could not be re-read: the publisher returned 403 and the Internet Archive 429. (Second accuracy pass; ESMO 2015 not readable; ASH 2021 full text from PubMed Central and the ASCO 2013 abstract from Europe PMC read instead) · The three points were taken out of the steps and grades and named in the gaps as unconfirmed. The removal steps now rest on ASH 2021 (remove an infected, mispositioned, malfunctioning or unneeded line) and the ASCO 2013 abstract (remove if a confirmed clot worsens despite anticoagulation).
- 2026-11-17 · Catheter-related thrombosis in people with cancer: where the guidelines differ: The page gave only ITAC and ESMO and described routine prophylaxis as simply not recommended. ASH 2021 also covers line clots and differs in strength: it suggests keeping the line (conditional, very low certainty), suggests against prophylaxis only conditionally on low-certainty evidence, allows it to be considered for selected high-risk patients, and names neither a drug nor a duration. (Second accuracy pass; ASH 2021 recommendations 18, 19 and 29 with remarks, full text from PubMed Central) · ASH added as a source with its grades, and the differences from ITAC stated in the summary and the best-in-class statement. A variant on thrombolysis was added, saying no source read recommends it for line clots.
- 2026-11-17 · Catheter-related thrombosis in people with cancer: reasons given for steps: Several reasons said more than the sources: that ultrasound finds most line clots (the quoted sensitivity starts at 56%), that trials showed bleeding outweighing benefit for most people, that treatment stops the clot travelling to the lungs, and that a DOAC here means apixaban, rivaroxaban or edoxaban. ITAC states only bleeding risk as the consideration and names the drug class, not the drugs. (Second accuracy pass; ITAC 2022 full text from Europe PMC, ASH 2021 full text from PubMed Central) · Each reason reworded to what the cited text says.
- 2026-11-17 · Preventing chemotherapy sickness (nausea and vomiting) in children: 5-HT3 antagonists in the delayed phase: The page told readers not to use 5-HT3 antagonists such as ondansetron in the delayed phase after any chemotherapy, and said they work only for the first day's sickness. The guideline prints that recommendation for highly emetogenic chemotherapy only, and gives no such reason. (Second accuracy pass; Children's Oncology Group summary dated 15 January 2024, public PDF read as raw text) · The step, the summary and the grade row are now scoped to highly emetogenic chemotherapy, and the reason is reworded to what the guideline says.
- 2026-11-17 · Preventing chemotherapy sickness (nausea and vomiting) in children: grade of the emetogenicity classification: One grade, strong with very low to high quality evidence, was given for the whole classification and attached to an abstract that prints no grade. The summary prints very low to high for the high and moderate lists, very low to moderate for the low list and very low to low for the minimal list. (Second accuracy pass; Children's Oncology Group summary dated 15 January 2024, public PDF read as raw text) · The grade was taken off the step and three grade rows now give each list its printed grade.
- 2026-11-17 · Preventing chemotherapy sickness (nausea and vomiting) in children: reasons given for steps: Several reasons said more than the sources: that two medicines are enough for most children, that children on mild chemotherapy are rarely sick, that sickness can come a day or more later, that uncontrolled sickness makes the next cycle worse, that lorazepam helped adults, and that the medicines are given before the chemotherapy. None is in the texts read. The guideline says only that limited adult data support benzodiazepines and that earlier sickness is a risk factor for anticipatory sickness. (Second accuracy pass; Children's Oncology Group summary and the four POGO abstracts from Europe PMC, read as raw text) · Reasons and the patient summary reworded to what the sources state, and a gap says the summary gives no timings.
- 2026-11-17 · Preventing chemotherapy sickness (nausea and vomiting) in children: age range: The page said all the prevention and treatment guidelines apply up to age 18. The summary prints that limit for the anticipatory guideline and the breakthrough guideline, and prints no age range for the acute and delayed prevention guideline. (Second accuracy pass; Children's Oncology Group summary dated 15 January 2024, public PDF read as raw text) · Population reworded guideline by guideline.
- 2026-11-17 · Preventing chemotherapy sickness (nausea and vomiting) in children: ASCO update and later guidance: The page said the ASCO 2020 update could not be read and that no later paediatric guidance existed. The ASCO abstract is readable and says the option of fosaprepitant was added to its paediatric recommendations, and an international consensus on when dexamethasone is used as an antiemetic in children was published in 2026. (Second accuracy pass; Europe PMC records and abstracts) · ASCO source marked as abstract read with its statement added, the 2026 consensus added as a source and a variant, and the classification corrigendum named in the gaps.
- 2026-11-17 · Anticoagulants and antiplatelet drugs around elective endoscopy and endoscopic biopsy: continuing warfarin for low-risk procedures: The step telling people to continue warfarin carried the grade "strong recommendation, low quality evidence", which belongs to the neighbouring sentence about checking the INR. BSG/ESGE 2021 grades continuing warfarin as a weak recommendation on low quality evidence. (Second accuracy pass; BSG/ESGE 2021 full text, Europe PMC XML read raw) · The step now quotes the sentence it describes with its weak grade, and the INR check is listed separately with its strong grade.
- 2026-11-17 · Anticoagulants and antiplatelet drugs around elective endoscopy and endoscopic biopsy: bridging and recent clots: One step named recent clots among the reasons to bridge warfarin with heparin and printed the grade of a different recommendation (strong, high, which is for involving a cardiologist about antiplatelet drugs). BSG/ESGE grades bridging strong on low quality evidence, and says most people with a clot in the last 3 months take a direct oral anticoagulant, for whom bridging is not recommended and a high-risk procedure may be better deferred. (Second accuracy pass; BSG/ESGE 2021 full text and tables, Europe PMC XML read raw) · Bridging now has its own step with its own quote, grade and schedule, and the caveat about recent clots is stated.
- 2026-11-17 · Anticoagulants and antiplatelet drugs around elective endoscopy and endoscopic biopsy: the ACG-CAG position: The page said ACG-CAG suggests continuing warfarin for elective endoscopy generally and against bridging, as a flat disagreement with BSG/ESGE. ACG-CAG makes an exception for advanced procedures, where it says 5 days of interruption without bridging would be appropriate, says bridging may be appropriate for people with mechanical valves and some others, and describes a pause of 1 to 2 days for direct oral anticoagulants, which the page did not give. (Second accuracy pass; ACG-CAG 2022 full text, Europe PMC XML read raw) · The summary, best-in-class statement and variants now state these exceptions and both guidelines' intervals.
- 2026-11-17 · Anticoagulants and antiplatelet drugs around elective endoscopy and endoscopic biopsy: grades in the summary: The summary gave one grade to pairs of recommendations that BSG/ESGE grades separately: the ampullectomy exception for aspirin is a weak recommendation, and the INR check below 1.5 rests on low quality evidence, not high. (Second accuracy pass; BSG/ESGE 2021 full text, Europe PMC XML read raw) · Each recommendation now carries its own grade.
- 2026-11-17 · Anticoagulants and antiplatelet drugs around elective endoscopy and endoscopic biopsy: reasons and scope: Three reasons given to patients went beyond the source: that aspirin adds little bleeding risk, that all three antiplatelet drugs act for about a week (the source gives 5 to 7 days for clopidogrel only), and why a morning dose is missed. The population also included people on heparin injections and with cancer-associated clots, whom neither guideline addresses. (Second accuracy pass; BSG/ESGE 2021 and ACG-CAG 2022 full texts, Europe PMC XML read raw) · The reasons are reworded to what the guideline states and the population says what the guidelines do not cover.
- 2026-11-17 · Blood thinners around cancer surgery, biopsy and endoscopy: heparin bridging in venous thromboembolism: The page said ASH 2018 recommends strongly against heparin bridging for people on anticoagulation for venous thromboembolism in general. The recommendation covers only people at low to moderate risk of recurrence whose warfarin-type drug is interrupted (strong, moderate certainty); the guideline's table puts a clot within the past 3 months at high risk, and the text says the recommendation does not apply to people who also have a mechanical heart valve. (Second accuracy pass; ASH 2018 full text read raw from PubMed Central (recommendation 14, table 4)) · Summary, best-in-class statement and grade row now give the scope and certainty, and quote the recommendation itself.
- 2026-11-17 · Blood thinners around cancer surgery, biopsy and endoscopy: aspirin, warfarin and restart steps: Three steps read as if they applied to any procedure. The source recommends continuing aspirin for endoscopic procedures with ampullectomy as an exception, and stopping warfarin 5 days before a high-risk endoscopic procedure in people at low thrombotic risk. The restart step added that bleeding must first be controlled, which the recommendation does not say. (Second accuracy pass; BSG and ESGE 2021 full text read raw from Europe PMC) · Steps scoped to endoscopy and to low thrombotic risk, the ampullectomy exception named, and the restart step worded as the recommendation is.
- 2026-11-17 · Blood thinners around cancer surgery, biopsy and endoscopy: direct oral anticoagulant schedule: The step gave the one or two day gap as a flat order and the patient summary said one to three days. The figures are the schedule of the PAUSE cohort study, which covered apixaban, dabigatran and rivaroxaban but not edoxaban, used longer gaps for dabigatran with creatinine clearance below 50 mL/min, and left out people with creatinine clearance below 25 to 30 mL/min. No source prints one to three days. (Second accuracy pass; PAUSE full text read raw from PubMed Central) · Step attributed to PAUSE with its drugs and its dabigatran exception; patient summary now says one or two days; limits added to the population and gaps.
- 2026-11-17 · Blood thinners around cancer surgery, biopsy and endoscopy: what CHEST 2022 is said to recommend: The best-in-class statement said CHEST 2022 gives a pharmacokinetic schedule for stopping direct oral anticoagulants based on PAUSE and bases its bridging recommendation on BRIDGE. Only the abstract of CHEST 2022 can be read, and it names neither trial and prints no schedule. A 2023 corrigendum to the guideline was also not mentioned. (Second accuracy pass; CHEST 2022 abstract read raw from Europe PMC, journal site refused a script) · Statement now says which CHEST recommendations were not read; the corrigendum is named in the gaps.
- 2026-11-17 · Blood thinners around cancer surgery, biopsy and endoscopy: reasons given for steps and patient summary: Several reasons said more than the source: that timing mistakes are common and a written plan prevents them, that aspirin protects more than it harms around most procedures, and that all three P2Y12 drugs affect platelets for about a week (the source gives 5 to 7 days for clopidogrel only). The patient summary said most people do not need heparin injections, where the sources limit this to atrial fibrillation and lower-risk venous thromboembolism. The advice not to stop a blood thinner on your own had no source. (Second accuracy pass; BSG and ESGE 2021 full text, CHEST 2022 abstract and the NHS anticoagulants page, all read raw) · Reasons reworded to the source text; patient summary limited to atrial fibrillation; NHS page added as the source for the advice to patients.
- 2026-11-17 · Blood thinners around cancer surgery, biopsy and endoscopy: limits of the BRIDGE trial: The page applied the BRIDGE result to cancer surgery without saying that the trial excluded people with a mechanical heart valve or a stroke in the previous 12 weeks, and that its authors state major cancer surgery was not represented. (Second accuracy pass; BRIDGE full text read raw through the NCBI E-utilities API) · Exclusions added to the step, the summary and the gaps.
- 2026-11-17 · Pneumocystis pneumonia (PJP) prophylaxis with co-trimoxazole: dapsone: The grade A-IIt was shown beside the sentence about testing for G6PD deficiency. AGIHO prints that grade on the dapsone dose sentence and does not grade the testing sentence. The step also left out AGIHO's condition that dapsone can be used if blood toxicity is not a major concern, and gave haemolysis as a reason that AGIHO does not state. (Second accuracy pass; AGIHO 2020 full text from Europe PMC, read as raw text) · The step now quotes the graded dapsone sentence, keeps the condition, says the G6PD test is recommended without a grade, and attributes the haemolysis explanation to the children's guideline, which states it.
- 2026-11-17 · Pneumocystis pneumonia (PJP) prophylaxis with co-trimoxazole: when to stop: The page said no source gave a duration beyond the entire period at risk, and explained that period in words the ECIL abstract does not contain. The guideline for children with solid tumours, which has an open authors' copy, suggests continuing until 3 months after treatment ends or 6 months after a stem cell transplant once the lymphocyte count is normal. (Second accuracy pass; authors' version of the 2017 children's guideline from White Rose Research Online and the Zydelig product information from the EMA, read as raw text) · Two sourced steps on duration were added, for idelalisib and for children with solid tumours, and the page now says that no source read sets a CD4 count for stopping.
- 2026-11-17 · Pneumocystis pneumonia (PJP) prophylaxis with co-trimoxazole: where guidelines agree: The page said the guidelines, including ASCO and IDSA, agree on the core indications, and that ECIL adds other regimens case by case. The ASCO and IDSA abstract names no diseases, only a 3.5% risk threshold for chemotherapy regimens, and ECIL says additional indications are proposed depending on the treatment regimen. (Second accuracy pass; ASCO/IDSA 2018 and ECIL-5 abstracts from Europe PMC, read as raw text) · The agreement is now stated for AGIHO and ECIL-5 only, and each body's position is given in its own terms.
- 2026-11-17 · Pneumocystis pneumonia (PJP) prophylaxis with co-trimoxazole: plain summary: The short summary called the infection rare and said the tablet is often taken three days a week, and the page said prevention works very well and that many more people at intermediate risk would need treating. No source read gives a frequency or says which schedule is commonest, and AGIHO says only that the number needed to treat is likely to be higher. (Second accuracy pass; AGIHO 2020 full text from Europe PMC, read as raw text) · The unsupported words were removed and the intermediate-risk reason now says "probably".
- 2026-11-17 · Pneumocystis pneumonia (PJP) prophylaxis with co-trimoxazole: who the guidelines cover: The page said AGIHO addresses adults and described the treatments as suppressing cell-mediated immunity. Neither statement is in the AGIHO text, and the ECIL grading scheme was described with a range the abstract does not print. (Second accuracy pass; AGIHO 2020 full text and ECIL-5 abstract from Europe PMC, read as raw text) · Reworded to what the texts show.
- 2026-11-17 · Antibacterial (fluoroquinolone) prophylaxis during expected severe neutropenia: whether it saves lives: The page said no guideline read reports a mortality benefit and that the paediatric synthesis found none. NICE's own 2012 evidence review reports fewer deaths with quinolone prophylaxis (32 of 1,295 against 57 of 1,286, RR 0.615, 95% CI 0.4 to 0.946, moderate quality), and the paediatric synthesis found lower infection-related mortality with co-trimoxazole. (Second accuracy pass; NICE CG151 full guideline (PDF text, Table 5.3) and the paediatric guideline full text (Europe PMC XML, Table 2)) · The gap is rewritten as a stated disagreement between NICE, AGIHO, the paediatric panel, ECIL and the Australian and New Zealand consensus, with each figure and its source.
- 2026-11-17 · Antibacterial (fluoroquinolone) prophylaxis during expected severe neutropenia: children having a transplant: One step covered ALL induction and both kinds of stem cell transplant and showed the grade weak recommendation, low-quality evidence. That grade is printed for ALL induction only. The transplant recommendations are printed as weak recommendation, moderate-quality evidence. (Second accuracy pass; paediatric guideline full text (Europe PMC XML, recommendations 2, 4 and 5)) · The step is split in two and the transplant step carries the moderate-quality grade.
- 2026-11-17 · Antibacterial (fluoroquinolone) prophylaxis during expected severe neutropenia: what the 2026 paediatric update kept: The page said the 2026 update maintains the 2020 recommendations. The abstract, which is all that can be read, says two were maintained (consider prophylaxis in AML and relapsed ALL, and use levofloxacin) and is silent on the rest. (Second accuracy pass; 2026 update abstract (Europe PMC)) · The statement now names the two maintained recommendations, and the gaps say which steps rest on the 2020 text.
- 2026-11-17 · Antibacterial (fluoroquinolone) prophylaxis during expected severe neutropenia: where the guidelines agree: The summary said the guidelines agree on a fluoroquinolone given only during severe neutropenia, the ASCO/IDSA position was written as limited to high-risk adults only, and the Australian and New Zealand consensus was said to limit prophylaxis to the highest risk. Those three were read as abstracts, which name no drug or duration, say recommended for patients at high risk, and say harms outweighed benefit in most circumstances. (Second accuracy pass; ASCO/IDSA 2018 and Australian and New Zealand 2025 abstracts (Europe PMC)) · Each position is reworded to what its abstract says, and the AGIHO guideline (full text) and the ECIL appraisal (abstract) are added with their grades and figures.
- 2026-11-17 · Antibacterial (fluoroquinolone) prophylaxis during expected severe neutropenia: side effect warnings and stopping: The MHRA restrictions and the warning about disabling, long-lasting side effects such as tendon damage were attributed to the paediatric guideline, which mentions neither, and the rule to stop when the count recovers and its reason were attributed to a recommendation that states no stop rule. (Second accuracy pass; MHRA Drug Safety Update of 22 January 2024 (gov.uk, full text), AGIHO guideline (Europe PMC XML) and the paediatric guideline full text) · A new step quotes the MHRA advice, a new step gives the AGIHO start and stop rule with its grades (B-IIu and A-IIu), and the paediatric steps keep only what that guideline says.
- 2026-11-17 · Antibacterial (fluoroquinolone) prophylaxis during expected severe neutropenia: dose and licence: The page said the dose comes from the product licence, which was not read. The UK ciprofloxacin summary of product characteristics read today (500 mg tablets, Milpharm, revised 29 May 2026) lists the management of neutropenic patients with fever, not prophylaxis in neutropenia, and no guideline read prints a prophylaxis dose. (Second accuracy pass; electronic medicines compendium SmPC (HTML, full text) and the guidelines above) · The gap now says no source read prints a prophylaxis dose and that NICE's licence note dates from September 2024. No dose is given on the page.
- 2026-11-17 · Antibacterial (fluoroquinolone) prophylaxis during expected severe neutropenia: when to ring: The patient summary told readers to call at once with a fever and cited no source, and a sentence about people already on a fluoroquinolone had no source. NICE says to suspect neutropenic sepsis in anyone on anticancer treatment who becomes unwell and to refer immediately. (Second accuracy pass; NICE CG151 recommendations page (HTML, recommendations 1.1.1.1, 1.3.1.1 and 1.3.1.2)) · The wording is now feeling unwell, with or without a fever, the step cites and quotes NICE, and the unsourced sentence is replaced by the MHRA and AGIHO cautions.
- 2026-11-17 · Antifungal prophylaxis for people at high risk of invasive fungal infection: adults with acute lymphoblastic leukaemia: The page said the risk of fungal infection in adults with acute lymphoblastic leukaemia was too low for prophylaxis to help. ECIL-10 reports invasive fungal disease in 4 to 18% of these adults, similar to acute myeloid leukaemia. The obstacle is the interaction of voriconazole and posaconazole with vincristine. (Second accuracy pass; ECIL-10 (Leukemia 2025), full text from Europe PMC read as raw text) · Adult ALL was taken out of the low-risk step. The vincristine step now gives the 4 to 18% figure, the D-II grade against voriconazole and posaconazole, and the C-III options ECIL allows with caution.
- 2026-11-17 · Antifungal prophylaxis for people at high risk of invasive fungal infection: drug level targets: The page said therapeutic drug monitoring targets were in the ECIL-10 tables. ECIL-10 prints doses and recommends monitoring serum levels in steroid-treated graft-versus-host disease, but prints no target. (Second accuracy pass; ECIL-10 full text and the AGIHO 2022 update (J Antimicrob Chemother 2023), both from Europe PMC as raw text) · The gap was rewritten. A step now gives the targets AGIHO prints (for example posaconazole above 0.7 mg/L for prophylaxis, voriconazole above 1 mg/L and below 4.5 mg/L) with AGIHO's own caveat that they are not generally applicable to prophylaxis.
- 2026-11-17 · Antifungal prophylaxis for people at high risk of invasive fungal infection: when to give it in children: The step told teams to give prophylaxis during severe neutropenia and said no trial had compared start and stop rules. The 2020 paediatric guideline says to consider it (a weak recommendation on low quality evidence) and says only that data are limited. (Second accuracy pass; Lehrnbecher et al. (J Clin Oncol 2020), full text from Europe PMC read as raw text) · The action now begins with consider and the reason says data are limited.
- 2026-11-17 · Antifungal prophylaxis for people at high risk of invasive fungal infection: children with acute lymphoblastic leukaemia: The step said to consider prophylaxis for children aged 12 or older or with insufficient treatment response, and cited ECIL-10. ECIL-10 says these children are recommended to receive it. Consider is the strength of the separate 2020 paediatric guideline, which names no subgroup. The trial behind it was called European; the source calls it international. (Second accuracy pass; ECIL-10 full text and Lehrnbecher et al. 2020, from Europe PMC as raw text) · The step now states the ECIL recommendation as printed, says ECIL prints no grade for it, and gives the weaker 2020 paediatric position beside it.
- 2026-11-17 · Antifungal prophylaxis for people at high risk of invasive fungal infection: the 2026 paediatric update: The summary said the 2026 update maintains all the 2020 paediatric recommendations. Its abstract, the only part that could be read, says three strong recommendations were maintained (AML, allogeneic transplant, and an echinocandin or mould-active azole) and says nothing of the others. The publisher now lists the 2020 guideline as archived. (Second accuracy pass; abstract from Europe PMC and the POGO guideline page read raw; the journal page answered 403) · The summary names the three recommendations the abstract confirms, and the gaps say the 2020 text is archived and the 2026 full text was not read.
- 2026-11-17 · Antifungal prophylaxis for people at high risk of invasive fungal infection: who is covered: The page said most people with solid tumours or lymphoma are not covered by a prophylaxis recommendation. They are covered: the paediatric guideline makes a strong recommendation against routine prophylaxis for them and ECIL-10 grades routine prophylaxis in lymphoma D-II. (Second accuracy pass; Lehrnbecher et al. 2020 and ECIL-10, full texts from Europe PMC as raw text) · The population now says the guidelines recommend against routine prophylaxis in these groups.
- 2026-11-17 · Antifungal prophylaxis for people at high risk of invasive fungal infection: reasons given for four steps: Four reasons said more than the sources. Drug interactions were listed among what outweighs benefit at low risk, where the source says inconvenience. Vincristine nerve damage was called severe. Fungal infections after CAR-T were said to be higher with steroids, where ECIL speaks of infections in general and immunosuppressive drugs. The patient summary placed the infections in the lungs and said the medicine is usually posaconazole, which holds for adults only. (Second accuracy pass; ECIL-10 and Lehrnbecher et al. 2020, full texts from Europe PMC as raw text) · Each sentence was reworded to what the source prints.
- 2026-11-17 · Antifungal prophylaxis for people at high risk of invasive fungal infection: venetoclax and other targeted drugs: The page named the venetoclax interaction but gave no dose change, and had no step for adults after an allogeneic transplant. ECIL-9 says to reduce venetoclax by 75% when posaconazole is given (A-I). AGIHO says by at least 75% with posaconazole or voriconazole and by 50% with fluconazole or isavuconazole. The EHA summary says to 70 mg. (Second accuracy pass; ECIL-9 (Leukemia 2022), AGIHO 2022 update and the EHA guideline summary (HemaSphere 2022), full texts from Europe PMC as raw text) · Three sources and four steps were added (venetoclax, gilteritinib and QT monitoring, drug levels, adult allogeneic transplant), with each body's figure attributed to it.
- 2026-11-17 · Hepatitis B testing before cancer treatment and prevention of reactivation: UK guidance: The page said no current UK national guideline had been found. NICE CG165 (published 2013, last updated October 2017) has a section on prophylaxis during immunosuppressive therapy: it sets a minimum of 6 months after treatment for most groups and offers lamivudine to people who are HBsAg negative and anti-HBc positive starting rituximab. It does not say who should be tested. (Second accuracy pass; NICE CG165 recommendations 1.5.45 to 1.5.50, read as raw text from nice.org.uk) · NICE added as a source and as a variant, and the gap rewritten to say what NICE covers and what it does not.
- 2026-11-17 · Hepatitis B testing before cancer treatment and prevention of reactivation: the APASL variant: The page said APASL gives antivirals to the high-risk group and, in the other groups, only to people with advanced fibrosis or cirrhosis. APASL treats every HBsAg-positive patient in the moderate-risk group whatever the fibrosis; only anti-HBc-positive patients in that group, and everyone in the low-risk group, need advanced fibrosis or cirrhosis to be treated. (Second accuracy pass; APASL 2021 recommendations 4.1 and 4.2, full-text XML from Europe PMC) · The variant now states the rule for each of the three risk groups.
- 2026-11-17 · Hepatitis B testing before cancer treatment and prevention of reactivation: best in class: The page presented antivirals for everyone with chronic hepatitis B, and for past infection with any stem cell transplant, as common ground between ASCO and APASL. APASL monitors HBsAg-positive patients on low-risk therapy who have no advanced fibrosis, and for anti-HBc-positive patients it places autologous transplant in the moderate-risk group. The page also left out AASLD, which continues prophylaxis for at least 6 months, or at least 12 after anti-CD20 therapy. (Second accuracy pass; APASL 2021 Table 4 and recommendation 5.2 (Europe PMC XML), AASLD 2018 guidance section 6D (PubMed Central XML through E-utilities)) · The statement now attributes the rule to ASCO, states both APASL differences, and gives the AASLD and NICE durations beside ASCO's.
- 2026-11-17 · Hepatitis B testing before cancer treatment and prevention of reactivation: grading and scope wording: Several statements said more than the sources: that APASL used a modified GRADE scale with levels I, II-1, II-2, II-3 and III (the paper says GRADE and defines no levels), that ASCO's scope is adults and that it grades weak recommendations (it says neither), that anti-CD20 antibodies and transplant carry the highest risk (ASCO says an established high risk), that the risk with other drugs is probably low (ASCO says it is unclear and possibly low), and that solid-tumour cost-effectiveness was inconsistent across six analyses (four of the six concern solid tumours, two for and two against). (Second accuracy pass; ASCO 2020 author manuscript (PubMed Central XML through E-utilities) and APASL 2021 (Europe PMC XML)) · Each statement reworded to what the source prints.
- 2026-11-17 · Hepatitis B testing before cancer treatment and prevention of reactivation: newer editions: The page did not say that the AGA guideline it names through ASCO's table was replaced in 2025, that AASLD published a new guideline in November 2025 which leaves its 2018 reactivation guidance in force, or that APASL published a 2026 update of its general hepatitis B guideline. (Second accuracy pass; AGA 2025 abstract (Europe PMC), AASLD and IDSA 2025 full text (publisher page), APASL 2026 abstract (Europe PMC)) · AGA 2025 and AASLD 2018 and 2025 added as sources; the gaps name what could be read only as an abstract.
- 2026-11-17 · Aciclovir prophylaxis against herpes simplex and shingles during cancer treatment: corticosteroid step: The grade CIIu was attached to the sentence about continuing for 6 months after steroids stop, and the step told the reader to continue. AGIHO 2022 prints CIIu after the sentence saying it might be reasonable to consider prophylaxis, and says only that it might be continued for up to 6 months. (Second accuracy pass; AGIHO 2022 full text, Europe PMC XML read as raw text) · The step now quotes the sentence that carries the grade and says prophylaxis may be continued for up to 6 months.
- 2026-11-17 · Aciclovir prophylaxis against herpes simplex and shingles during cancer treatment: valaciclovir variant: The page said AGIHO grades valaciclovir BIIu with proteasome inhibitors. AGIHO 2022 prints BIIu for valaciclovir with bortezomib only. (Second accuracy pass; AGIHO 2022 full text and Table 7, Europe PMC XML read as raw text) · Changed to bortezomib.
- 2026-11-17 · Aciclovir prophylaxis against herpes simplex and shingles during cancer treatment: first step: The step said to start aciclovir in anyone who has had herpes simplex, and gave a reactivation rate as if it came from ASCO/IDSA. The ASCO/IDSA abstract says seropositive patients, which is a blood test result and not a history of cold sores, and the rate (37 to 68%) is from AGIHO 2022. (Second accuracy pass; ASCO/IDSA 2018 abstract from Europe PMC and AGIHO 2022 full text) · Reworded to seropositive, and the figure is attributed to AGIHO 2022 as printed.
- 2026-11-17 · Aciclovir prophylaxis against herpes simplex and shingles during cancer treatment: serology step: The step said not to wait for serology. AGIHO 2022 calls serology useful before intensive leukaemia treatment (BIII) and calls prophylaxis for everyone with an indication equally appropriate. (Second accuracy pass; AGIHO 2022 full text, Europe PMC XML read as raw text) · The step now gives both options with the grade.
- 2026-11-17 · Aciclovir prophylaxis against herpes simplex and shingles during cancer treatment: kidney function step: The reason given, that all three medicines are cleared by the kidneys and build up, is not in the cited guideline, which says only that dose adjustment is required. (Second accuracy pass; AGIHO 2022 full text, Europe PMC XML read as raw text) · Reason replaced with what the guideline says, and the missing adjusted doses are named as a gap.
- 2026-11-17 · Aciclovir prophylaxis against herpes simplex and shingles during cancer treatment: low-dose variant: The variant named ixazomib and gave the dose as 400 mg once daily or lower. AGIHO 2015 does not name ixazomib and prints doses as low as 200 to 400 mg of aciclovir daily or twice daily with proteasome inhibitors. (Second accuracy pass; AGIHO 2015 full text, Europe PMC XML read as raw text) · Reworded to the printed range and to proteasome inhibitor treatment.
- 2026-11-17 · Aciclovir prophylaxis against herpes simplex and shingles during cancer treatment: transplant, CAR T and daratumumab: The page said AGIHO and ASCO/IDSA agree on allogeneic transplant and left transplant guidelines unread. AGIHO 2022 excludes transplant and cell therapy. It also grades aciclovir with daratumumab only C IIt, for selected cases, while the US daratumumab label tells prescribers to start within 1 week and continue for 3 months after treatment. (Second accuracy pass; AGIHO 2020 autologous and AGIHO 2016 allogeneic transplant guidelines (Europe PMC XML), ECIL-2 and ECIL 11 slide sets (PDF text), ASTCT 2024 (XML) and the daratumumab label (DailyMed XML)) · Steps added for autologous transplant, allogeneic transplant, CAR T and daratumumab, with each source's own doses, durations and grades, and the disagreements stated as variants.
- 2026-11-17 · Influenza during cancer treatment: antiviral prophylaxis after exposure and early treatment: treating before a test result: The summary and the best in class statement said all the guidelines, AGIHO 2025 included, recommend starting an antiviral without waiting for a test result. AGIHO's graded sentence (AIIt) is to start as early as possible after symptom onset and diagnosis, and it says nothing about treating before a result. (Second accuracy pass; AGIHO 2025 full text from Europe PMC XML, IDSA 2018 from idsociety.org, UKHSA guidance of 23 October 2026 and the CDC clinician summary, all read raw) · The statement about not waiting for a result is now attributed to UKHSA and CDC, and IDSA's wording (documented or suspected influenza) is given as IDSA's. The step now reads start as early as possible after symptoms begin.
- 2026-11-17 · Influenza during cancer treatment: antiviral prophylaxis after exposure and early treatment: late start of prophylaxis: A step sourced to UKHSA said that after 48 hours the plan changes to watching closely and treating at the first symptom. UKHSA allows prophylaxis after 48 hours as off-label use on specialist advice. It is IDSA that advises against once-daily prophylaxis after 48 hours. IDSA's and AGIHO's limits (48 and 72 hours) were also written as cut-offs where both sources say ideally. (Second accuracy pass; UKHSA guidance and IDSA 2018 full text, read raw) · The step now says who holds each position, and ideally is restored wherever IDSA's 48 hours and AGIHO's 72 hours are quoted.
- 2026-11-17 · Influenza during cancer treatment: antiviral prophylaxis after exposure and early treatment: length of the preventive course: The page said UKHSA and NICE TA158 offer prophylaxis for 10 days. The NICE recommendations give no duration. The 10 days is in UKHSA and in the product information. (Second accuracy pass; NICE TA158 recommendations chapter, UKHSA guidance, the US label and the European product information, read raw) · The 10 days is now attributed to UKHSA and the labels, and CDC's 7 days is shown beside IDSA's.
- 2026-11-17 · Influenza during cancer treatment: antiviral prophylaxis after exposure and early treatment: length of treatment: The page said IDSA treats for 5 days. IDSA's 5 days is for uncomplicated influenza in otherwise healthy outpatients. The page also took the 10 day course for immunosuppressed patients only from UKHSA's citation of the manufacturer. (Second accuracy pass; IDSA 2018 full text, and the Tamiflu product information from the European Medicines Agency read with pdftotext) · IDSA's 5 days is now scoped as IDSA scopes it, and the European product information is cited directly for 10 days in immunocompromised patients.
- 2026-11-17 · Influenza during cancer treatment: antiviral prophylaxis after exposure and early treatment: steps written as orders: Two steps were written as orders where the source says consider: agreeing a plan in advance (IDSA, clinicians can consider, C-III) and repeating the test with resistance testing after 5 days without a response (UKHSA, consider). A third step said to use usually oral oseltamivir on IDSA's authority, where IDSA names three neuraminidase inhibitors without ranking them. (Second accuracy pass; IDSA 2018 and UKHSA guidance, read raw) · The two steps now begin with consider, and the preference for oseltamivir is attributed to UKHSA.
- 2026-11-17 · Influenza during cancer treatment: antiviral prophylaxis after exposure and early treatment: advice to patients: The patient summary told readers to tell their cancer team the same day and said flu can become serious quickly. No source read gives a same day limit or the word quickly. The NHS flu page tells people with flu symptoms and a weakened immune system, for example from chemotherapy, to ask for an urgent GP appointment or get help from NHS 111. (Second accuracy pass; NHS flu page (last reviewed 3 February 2026) and UKHSA guidance, read raw) · The patient summary now gives the NHS urgent advice and the 48 hour start, the NHS page is cited, and a gap says no source sets a time limit for reporting contact.
- 2026-11-17 · Influenza during cancer treatment: antiviral prophylaxis after exposure and early treatment: dose and kidney function: The page gave no treatment dose and nothing on kidney function. The sources also differ: UKHSA's guidance of 23 October 2026 keeps 75 mg at a creatinine clearance of 11 to 60 mL/min, where the US label, CDC and the European product information give 30 mg. (Second accuracy pass; UKHSA guidance, Tamiflu US label from DailyMed, CDC clinician summary from the Internet Archive capture of 13 November 2026, and the European product information, read raw) · A step now gives the licensed adult dose with the label's sentence, and a variant sets out each source's dose by kidney function with who says which.
- 2026-11-17 · Influenza during cancer treatment: antiviral prophylaxis after exposure and early treatment: reasons given for steps: Six reasons beside steps said more than the cited source: for example that a day spent waiting uses half the window, that the licensed dose is the one the trials tested, and that treating a household member shortens the time they are infectious. Two evidence lines stated that trials excluded immunocompromised people, which the labels do not say, and one called UKHSA's advice expert opinion although UKHSA does not grade. (Second accuracy pass; each cited source read raw) · Each reason now says what its source says and names the source, and the unsupported statements were removed.
- 2026-11-17 · Influenza during cancer treatment: antiviral prophylaxis after exposure and early treatment: baloxavir and CDC: The page said CDC guidance had not been read and so left out CDC's position that baloxavir is not recommended as monotherapy in severely immunosuppressed people. (Second accuracy pass; CDC clinician summary (updated 10 March 2026) read from the Internet Archive capture of 13 November 2026, because cdc.gov refuses a script) · CDC is now a source, and its positions on baloxavir, on not waiting for laboratory confirmation, on longer courses and on 7 days of prophylaxis are shown.
- 2026-11-17 · DPD deficiency testing before fluorouracil, capecitabine or tegafur: dosing by activity score: The page gave CPIC's dose reduction as 50% at activity score 1 and 25 to 50% at score 1.5, and said CPIC and the Dutch guideline disagreed at score 1.5. That is the table printed in 2017; CPIC revised it on its guideline page in November 2018 so that every intermediate metaboliser, score 1 or 1.5, starts at half the standard dose and is then titrated, and the two guidelines now agree. (Second accuracy pass; CPIC guideline page updates read as raw JSON from the CPIC and ClinPGx public APIs, the page itself serving no text without JavaScript) · Step, grades, summary, best in class and the Dutch variant rewritten to the November 2018 revision, with the 2017 printed figure kept and labelled as the published table
- 2026-11-17 · DPD deficiency testing before fluorouracil, capecitabine or tegafur: CPIC edition: A gap said a newer CPIC revision could not be ruled out because the CPIC website serves no text to a script. Three posted updates exist and read today: November 2018, January 2024 and July 2026, the last assigning HapB3 an allele value of 0.75 for the forthcoming guideline. (Second accuracy pass; CPIC and ClinPGx public APIs read as raw JSON) · Updates added as a source, and the gap replaced with what the updates say and what is still unpublished
- 2026-11-17 · DPD deficiency testing before fluorouracil, capecitabine or tegafur: what patients are told to report: The page told patients that severe diarrhoea, mouth sores, fever or infection need same-day contact with the team. No cited source gave that symptom list or that time limit. (Second accuracy pass; Xeloda EU product information read as raw text with pdftotext) · Reworded to the patient leaflet's own wording, a step added for the first-cycle symptoms the leaflet links to DPD deficiency, and the leaflet's advice to stop and contact the doctor cited
- 2026-11-17 · DPD deficiency testing before fluorouracil, capecitabine or tegafur: Xeloda product information year: The EU product information was dated 2026. The EMA product page gives its last update as 2 May 2025. (Second accuracy pass; EMA Xeloda EPAR page read as raw text) · Year corrected to 2025
- 2026-11-17 · DPD deficiency testing before fluorouracil, capecitabine or tegafur: turnaround time: A gap said no source read addressed the turnaround time of testing. The French evaluation states it: the result should reach the clinician ideally within seven days and at most ten. (Second accuracy pass; Haute Autorite de Sante evaluation of September 2023 read as raw text) · Gap replaced with the French figures, and the evaluation added as a source
- 2026-11-17 · Renal dose adjustment of anticancer drugs in reduced kidney function: date of the guideline: The page called ADDIKD the newest guidance of 2025. The guideline itself is version 1, published on eviQ on 26 October 2022; 2025 is the year of the three journal papers that describe it. (Second accuracy pass; eviQ guideline page and PDF read raw, Crossref for the papers) · The summary and short summary now give 2022 for the guideline and 2025 for the papers, and the guideline is cited as its own source.
- 2026-11-17 · Renal dose adjustment of anticancer drugs in reduced kidney function: cisplatin evidence grade: The page said ADDIKD's cisplatin recommendations had low to moderate certainty. The guideline grades all three cisplatin recommendations moderate, with strength strong, conditional and strong. (Second accuracy pass; ADDIKD guideline 2022 PDF, section 4.11, read as text) · The grade is corrected and the strength of each recommendation is stated.
- 2026-11-17 · Renal dose adjustment of anticancer drugs in reduced kidney function: cisplatin in reduced kidney function: The page gave only the eGFR 45 to 59 band for cisplatin and left out that ADDIKD advises avoiding cisplatin below an eGFR of 45, and the size of the reductions it suggests (25 to 50 per cent above 50 mg/m2, 25 per cent at or below it, or full dose under stated conditions). (Second accuracy pass; ADDIKD guideline 2022 PDF, Table 15, and the integration paper's supplementary appendix, read as text) · The cisplatin entry now gives each band as the guideline prints it, and a matching methotrexate entry was added.
- 2026-11-17 · Renal dose adjustment of anticancer drugs in reduced kidney function: measured GFR: The step told readers to use a measured GFR for the first dose of carboplatin, cisplatin and high-dose methotrexate. ADDIKD says measured GFR is preferred, not required. The step's quotation also printed a mangled superscript. (Second accuracy pass; ADDIKD Part 1, Europe PMC full text) · The step now says prefer, and the quotation was removed with the figure stated in prose.
- 2026-11-17 · Renal dose adjustment of anticancer drugs in reduced kidney function: carboplatin and body surface area: The summary said BSA-adjusted eGFR goes into the Calvert formula for carboplatin. ADDIKD prefers measured GFR and calls BSA-adjusted eGFR a suitable alternative, especially at eGFR 45 to 125, with non-curative intent and without unusual body composition. (Second accuracy pass; ADDIKD Part 1, Europe PMC full text) · The summary now states the preference and the conditions.
- 2026-11-17 · Renal dose adjustment of anticancer drugs in reduced kidney function: changing kidney function: The step said not to rely on a creatinine-based estimate when kidney function is changing. ADDIKD says to interpret such estimates cautiously, and KDIGO says to consider and adapt dosing when GFR is not in a steady state. (Second accuracy pass; ADDIKD Part 1 full text and KDIGO 2024 PDF) · The step and the summary are reworded to the sources.
- 2026-11-17 · Renal dose adjustment of anticancer drugs in reduced kidney function: scope: The page did not say that ADDIKD's dose recommendations apply to the first cycle only and exclude acute kidney injury, and it said dialysis patients need specialist pharmacy input where the guideline says nephrology, clinical pharmacology or both. (Second accuracy pass; ADDIKD guideline 2022 PDF, scope section) · The population is rewritten from the scope section and a first-cycle step was added.
- 2026-11-17 · Renal dose adjustment of anticancer drugs in reduced kidney function: Lancet Oncology reference: The 2023 Lancet Oncology update was credited to Krens and colleagues. Its first author is Giraud; Krens is the third author and led the 2019 paper. (Second accuracy pass; Crossref record for the DOI) · The source entry and the text now name Giraud et al. for 2023 and Krens et al. for 2019.
- 2026-11-17 · Renal dose adjustment of anticancer drugs in reduced kidney function: claims with no source: Several statements went beyond the sources: that no randomised trials exist, that ADDIKD is the only international guideline, that KDIGO names CKD-EPI for drug dosing, that older references mostly use Cockcroft-Gault, that uptake outside Australia is unmeasured, and three reasons given under steps. (Second accuracy pass; ADDIKD papers, KDIGO 2024 and BOPA 5.0 read raw) · Each is reworded to what the source says or attributed to its authors, and a difference between KDIGO and ADDIKD on eGFR not indexed to body size is now stated.
- 2026-11-17 · Hepatic dose adjustment of anticancer drugs in reduced liver function: the Lancet Oncology reference: The 2023 update of the Lancet Oncology dose recommendations was attributed to Krens and colleagues. Crossref and Europe PMC give Giraud as first author; Krens is first author of the 2019 original. The page also said UK pharmacists use it and that BOPA names it, where BOPA only lists "Lancet guidelines" as an example source. (Second accuracy pass; DOI resolved through Crossref and Europe PMC, BOPA version 5.0 PDF read with pdftotext) · Authorship corrected in the source list, the variant and the gaps, and the BOPA mention reworded to what its table says.
- 2026-11-17 · Hepatic dose adjustment of anticancer drugs in reduced liver function: which scale to use: A step told clinicians to use Child-Pugh in cirrhosis and liver cancers and gave as the reason that it captures liver function better than two blood tests. The FDA draft gives this advice to drug developers designing studies, makes the liver cancer advice conditional, and does not make that comparison. (Second accuracy pass; FDA September 2026 draft guidance, PDF read with pdftotext) · Step reworded as the FDA's advice to developers, with the reason in the draft's own terms.
- 2026-11-17 · Hepatic dose adjustment of anticancer drugs in reduced liver function: reasons given for steps: Three reasons said more than their sources: that liver function can change quickly and drugs build up dangerously (not in BOPA), that a raised result not from the liver should not cut a dose (not in the FDA draft), and that the label holds the only measured data (not in the label). (Second accuracy pass; BOPA version 5.0, FDA draft and DailyMed label read as raw text) · Each reason rewritten to what the cited text states.
- 2026-11-17 · Hepatic dose adjustment of anticancer drugs in reduced liver function: when the label is silent: The step on weighing treatment goals and monitoring closely when a label has no data was cited to Yang 2022, which advises drug developers and gives no such clinical advice. (Second accuracy pass; Yang 2022 full text from Europe PMC) · Step marked as practice with no citation, and the gap named.
- 2026-11-17 · Hepatic dose adjustment of anticancer drugs in reduced liver function: disagreement between the scales: The summary said the clinician should apply the label's scale and gave the disagreement figures as if for people with cancer, and said people with impairment are usually excluded from trials. The paper says prescribers should attempt to use the label's scale, its figures come from dedicated studies in people without cancer for five drugs from one company, and Yang says often. (Second accuracy pass; Elmeliegy 2021 and Yang 2022 full text from Europe PMC, percentages recomputed from Table 2) · Summary reworded and the limits of the figures added to the gaps.
- 2026-11-17 · Hepatic dose adjustment of anticancer drugs in reduced liver function: a quotation: One quoted FDA sentence is split in the PDF by a footnote and a page break, so it matched the source only after removing more than line breaks. (Second accuracy pass; FDA draft PDF, quotes checked by script) · Replaced with the unbroken sentence before it.
- 2026-11-17 · Carboplatin dose calculation by the Calvert formula and kidney function: the ceiling on kidney function: The page said the NCI letter that caps the kidney function entered into the formula could not be read, and gave no ceiling. The letter of 22 October 2010 survives in a public web archive and says the GFR used should not exceed 125 ml/min, with maximum doses of 900 mg at AUC 6, 750 mg at AUC 5 and 600 mg at AUC 4; NRG Oncology trial instructions print the same and add a minimum creatinine of 0.7 mg/dL. (Second accuracy pass; NCI CTEP letter (scanned copy in the University of North Texas web archive, read by eye and OCR) and NRG-GY027 protocol Appendix IV (PDF from ClinicalTrials.gov, pdftotext)) · Added the ceiling, the maximum doses and the creatinine floors as a step and a variant, each attributed to its body, with ADDIKD's different approach beside them.
- 2026-11-17 · Carboplatin dose calculation by the Calvert formula and kidney function: best in class: The page said the UK cohorts of Samani 2022 supported CKD-EPI over Cockcroft-Gault. Samani's main results do not report CKD-EPI; its best performers were CamGFR v2 and Cockcroft-Gault with adjusted body weight. (Second accuracy pass; Samani 2022, full text from Europe PMC) · Removed the claim and stated what Samani found.
- 2026-11-17 · Carboplatin dose calculation by the Calvert formula and kidney function: the US label: The page said the US label still uses creatinine clearance in the formula. The label prints the Calvert formula with GFR, notes that GFR was measured by 51Cr-EDTA clearance, and names no estimating equation; creatinine clearance appears only in its separate dose table in mg per square metre. (Second accuracy pass; Accord carboplatin label version 9, DailyMed SPL XML) · Corrected the sentence.
- 2026-11-17 · Carboplatin dose calculation by the Calvert formula and kidney function: measured GFR: The page said measured GFR is always used for curative treatment and attached the grade "clinical consensus; strong" to a sentence from a paper that prints no grade. ADDIKD calls measured GFR preferred and the estimate a suitable alternative, and prints the grade for its recommendation 2 in the Part 1 paper. (Second accuracy pass; ADDIKD Part 1 and integration papers, full text from Europe PMC) · Reworded to preferred, moved the step to the Part 1 sentence, and said which recommendation the grade belongs to.
- 2026-11-17 · Carboplatin dose calculation by the Calvert formula and kidney function: adjusted body weight: The page gave the threshold for adjusted body weight as above a BMI of 25. Samani 2022 and the NRG Oncology instructions both say a BMI of 25 or more. (Second accuracy pass; Samani 2022 (Europe PMC) and NRG-GY027 Appendix IV (pdftotext)) · Corrected the threshold.
- 2026-11-17 · Carboplatin dose calculation by the Calvert formula and kidney function: unsupported reasons: Four statements had no support in the cited text: that Cockcroft-Gault can overdose with modern assays (ADDIKD says over or under), that centres use AUC 6 because estimates run low, that lowering the AUC would cut the dose twice, and that carboplatin leaves the body almost entirely through the kidneys (the label says the major route). (Second accuracy pass; ADDIKD Part 1, Samani 2022 and the Accord label, read as raw text) · Reworded each to what its source says or removed it.
- 2026-11-17 · Body surface area and weight-based dose calculation: which formula: The summary named DuBois and Mosteller as the formulas with no source for either. No source read names Mosteller. ASCO 2012 recommends any of the standard formulae and finds no evidence for one over another, the two cited studies used DuBois, and BOPA names the Boyd formula for children. (Second accuracy pass; ASCO 2012 executive summary read in full from PubMed Central, Chatelut 2012 from Europe PMC XML, BOPA version 5.0 PDF as text) · The summary now gives ASCO's recommendation and the DuBois formula as Chatelut prints it, a step quotes the recommendation, and gaps says the Mosteller formula is not printed because no source read as text prints it.
- 2026-11-17 · Body surface area and weight-based dose calculation: actual weight and cure: The step telling teams to use actual weight, especially when treatment aims to cure, cited the 2021 update, whose abstract says neither. Both come from the 2012 guideline. The fixed-dose exceptions (carboplatin and bleomycin as examples, vincristine capped at 2.0 mg in CHOP and CVP) were missing from the page. (Second accuracy pass; ASCO 2012 executive summary read in full from PubMed Central, ASCO 2021 abstract from Europe PMC) · Actual body weight is now its own step quoting the 2012 recommendation, the cure wording is attributed to 2012, and a new step quotes the 2012 fixed-dose exceptions. Gaps says whether the 2021 update keeps them could not be read.
- 2026-11-17 · Body surface area and weight-based dose calculation: capping: The page said ASCO recommends against capping body surface area and that a cap is often set at 2.0 m2. No ASCO text that could be read uses those words: ASCO recommends actual body weight and full weight-based doses, and 2.0 m2 was one hospital's cap. (Second accuracy pass; ASCO 2012 and 2021 abstracts and the 2012 executive summary, Tübingen cohort full text from Europe PMC XML) · Reworded to what ASCO says, with the view that this amounts to a position against routine capping attributed to the cohort's authors. The patient summary no longer says capped.
- 2026-11-17 · Body surface area and weight-based dose calculation: capping cohort: The Tübingen cohort was described as 730 women. The paper says 730 patients and does not give their sex. (Second accuracy pass; full text from Europe PMC XML) · Changed to 730 patients in the summary and the variant.
- 2026-11-17 · Body surface area and weight-based dose calculation: re-weighing study: The page said changes came after a median of 84 days, that the alternative was weighing every cycle, that the data were Dutch, and gave the three and six month intervals as firm. The abstract says changes occurred after 84 days in tumour types with more than 100 patients, compares with monthly weighing only, does not name the country, and advises at least every 3 months for some groups and that 6 months should be considered for the rest. (Second accuracy pass; abstract from Europe PMC, full text not downloadable) · Reworded to the abstract, country removed, and the 10% change stated as the authors' assumption.
- 2026-11-17 · Body surface area and weight-based dose calculation: immunotherapy and targeted therapy: The reason given was that there is no evidence these drugs need reducing for body size, and the step said give. The 2021 abstract says only that fewer studies have addressed them, and that full approved doses be offered. (Second accuracy pass; ASCO 2021 abstract from Europe PMC) · The step now says offer and the reason states what the abstract says.
- 2026-11-17 · Body surface area and weight-based dose calculation: who verifies, and children: The page said a pharmacist checks each calculation; BOPA's verifier is a registered pharmacy professional, who may be a pharmacy technician. The page also said no source sets a re-weighing interval, although BOPA asks for a child's weight before every cycle. (Second accuracy pass; BOPA version 5.0 PDF as text) · Reworded to pharmacy professional, the no-interval statement scoped to adults, and a step added for children with BOPA's two weights and Boyd formula.
- 2026-11-17 · Body surface area and weight-based dose calculation: recording the calculation: The page said the dose calculation is written on the order, and its quotation ran four list items of the standard together. ASCO-ONS standard 3.9.8 says the calculation method, the variables, how often they are re-evaluated and what change prompts confirmation or recalculation must be in the patient's medical record. (Second accuracy pass; ASCO-ONS 2024 standards, full text read from the Oncology Nursing Society's open PDF) · The step, the summary and the best practice statement now say medical record, and the quotation is one list item copied whole.
- 2026-11-17 · Dose banding of anticancer drugs: the permitted difference from the calculated dose: The page said the NHS England tables and their limits could not be read and stated no limit. NHS England's guide to the tables (version 1, April 2016) sets two rules: no banded dose of traditional SACT more than 6% from the calculated dose, and none of a monoclonal antibody more than 10%, without specific prior agreement. (Second accuracy pass; NHS England guide and three national tables, PDFs fetched from england.nhs.uk and read with pdftotext) · Both rules added as steps with verbatim quotes, the summary, short summary and best-in-class statement rewritten, and the variance printed in the tisotumab vedotin, nivolumab and durvalumab tables reported, including two durvalumab bands above 10%.
- 2026-11-17 · Dose banding of anticancer drugs: short summaries: The short summaries said the difference is kept small, that studies show about the same amount of drug in the blood, and that banding does not make the treatment weaker. The one pharmacokinetic study cited compared the spread across patients and had no efficacy data, and its authors note one patient's exposure may change by as much as the dose. (Second accuracy pass; Chatelut 2012 full text, Europe PMC XML) · The unsupported sentences were removed and replaced by the NHS England limits.
- 2026-11-17 · Dose banding of anticancer drugs: clinical outcome evidence: The page said no clinical outcome study compares banded with individually calculated doses. A 2023 retrospective study of weekly paclitaxel in breast cancer did, and found no difference in efficacy or toxicity with doses fitted up to 10%. (Second accuracy pass; Puisset 2023 abstract from Europe PMC, DOI resolved) · The gap was rewritten and the study added as a source and a variant, marked as read in abstract only.
- 2026-11-17 · Dose banding of anticancer drugs: who it applies to: The page limited banding to parenteral drugs. The NHS England tables page also lists tables for capecitabine, oral temozolomide and trifluridine tipiracil. (Second accuracy pass; NHS England tables page, Internet Archive copy of 9 July 2026 read as text) · The word parenteral was removed.
- 2026-11-17 · Dose banding of anticancer drugs: children: The step said not to band doses for children unless a paediatric protocol says so, and gave a reason. BOPA says only that dose banding is not usually used in children and gives no reason or exception. (Second accuracy pass; BOPA standards version 5.0, PDF read with pdftotext) · The step was reworded to BOPA's words, and a variant added for the 2018 study that tested the NHS England tables on children's data.
- 2026-11-17 · Dose banding of anticancer drugs: recalculating after a change: The step applied to any drug when weight or kidney function changes. BOPA's passage is about drugs dosed by AUC on estimated kidney function, when creatinine or weight changes significantly, and its warning names carboplatin. (Second accuracy pass; BOPA standards version 5.0, PDF read with pdftotext) · The step was scoped to carboplatin dosed on estimated kidney function and a significant change.
- 2026-11-17 · Dose banding of anticancer drugs: the exposure step: The step said to use bands that keep each patient's exposure close to individual dosing. The study showed a similar spread across patients, and says one patient's exposure may change by as much as the dose. (Second accuracy pass; Chatelut 2012 full text, Europe PMC XML) · The step was reworded to the paper's conclusion, with the caveat for one patient in the reason.
- 2026-11-17 · G-CSF prophylaxis by febrile neutropenia risk: ASCO edition: The page followed the 2015 ASCO guideline and said no later ASCO update existed. ASCO published a new edition in February 2026. Its abstract keeps the threshold of about 20 per cent but says that below 20 per cent prophylaxis "may be warranted" for people at high risk, where the 2015 abstract said it "is recommended". (Second accuracy pass; ASCO 2026 abstract (PMID 41740078) read raw from Europe PMC, DOI resolved through Crossref) · Added the 2026 edition, re-quoted the two ASCO steps from it, marked 2015 as superseded and stated the change in wording.
- 2026-11-17 · G-CSF prophylaxis by febrile neutropenia risk: intermediate-risk step: The step said to give G-CSF when a person's risk factors raise the overall risk on a 10 to 20 per cent regimen. EORTC says particular attention should be given to those factors, ASCO 2026 says prophylaxis may be warranted, and SEOM says the decision must be individualised. None writes it as an order, and "other illnesses" was not in the source cited. (Second accuracy pass; EORTC 2010 and ASCO 2026 abstracts from Europe PMC, SEOM 2018 full text from Europe PMC) · Step reworded to "may be warranted" with the ASCO 2026 sentence quoted, and each body's wording given in the summary.
- 2026-11-17 · G-CSF prophylaxis by febrile neutropenia risk: NICE reasoning: The page said NICE does not offer routine G-CSF because for most regimens the risk is too low, and that NICE frames the same rule as ASCO and EORTC from the other side. NICE uses no risk threshold. Its guideline group decided against routine G-CSF because of the high cost per benefit in its economic model, and did not review G-CSF used as part of a regimen or to maintain dose intensity. (Second accuracy pass; NICE CG151 recommendations page and 2012 full guideline PDF, read as raw text) · Reason rewritten from the full guideline and NICE presented as a different position.
- 2026-11-17 · G-CSF prophylaxis by febrile neutropenia risk: best in class: The page said ASCO and EORTC agree that patient factors decide at 10 to 20 per cent and on secondary prophylaxis, that AGIHO 2026 is the most recent guideline, and that every other source read dates from 2010 to 2015. The ASCO abstract mentions neither 10 to 20 per cent nor secondary prophylaxis, ASCO 2026 is later than AGIHO 2026, and the NCCN abstract read is from 2022. (Second accuracy pass; abstracts and publication dates from Europe PMC) · Statement rewritten with each position attributed to the body that holds it, and SEOM 2018 grades added.
- 2026-11-17 · G-CSF prophylaxis by febrile neutropenia risk: reasons given for steps: Several reasons went beyond the EORTC and ASCO abstracts: that a frailer person is at higher risk, that a person who has had one episode is likely to have another, that pegfilgrastim avoids daily courses "stopped too early" (the source says some patients receive suboptimal daily G-CSF), and that dose-dense treatment is only worth its toxicity where it improves survival. (Second accuracy pass; EORTC 2010 and ASCO 2015 abstracts from Europe PMC) · Each reason reworded to what its source says. The advice not to stop filgrastim early is now sourced to the EU product information.
- 2026-11-17 · G-CSF prophylaxis by febrile neutropenia risk: doses and timing: The page gave no dose or timing and said none could be read. The product information states them: pegfilgrastim 6 mg once per cycle at least 24 hours after chemotherapy, with the US label adding not within 14 days before chemotherapy, and filgrastim 5 micrograms per kg a day from at least 24 hours after chemotherapy until the neutrophil count has recovered. SEOM 2018 grades its prophylaxis recommendations. (Second accuracy pass; EMA product information PDFs for Neulasta and Zarzio, DailyMed labels for Neulasta and Neupogen, SEOM 2018 full text) · Added two dosing steps, a patient step on what to report, three variants and five graded rows, each from the raw text.
- 2026-11-17 · Anaemia during chemotherapy: the clot warning: The page said these injections raise the risk of venous clots, a risk already raised in cancer. ASCO/ASH say thromboembolism, and report a 50% to 75% higher risk of thromboembolism and arterial events; they do not make the second statement. (Second accuracy pass; ASCO/ASH 2019 full text from PubMed Central, read as raw text) · Corrected to thromboembolism, the source's figure added, and the unsupported phrase removed.
- 2026-11-17 · Anaemia during chemotherapy: treatment aimed at cure: The patient summary said these injections are not used when the aim is cure, as if every body agreed. That is the ASCO/ASH and US label position; NICE recommends them within UK licences whose quoted indications do not mention treatment intent. (Second accuracy pass; ASCO/ASH 2019, NICE TA323 chapters 1 and 3, and the US labels on DailyMed, all read as raw text) · Each position is now attributed to the body that holds it, with SEOM 2020 added.
- 2026-11-17 · Anaemia during chemotherapy: reasons given for steps: Several reasons said more than the guideline: that a transfusion works faster, that continuing an ineffective injection adds cost, that iron improves the response, and that ASCO/ASH found no case for use outside chemotherapy. The guideline says iron may improve the response, gives the licence as the reason outside chemotherapy, and does not make the other two statements. Two steps were also worded as orders where the guideline says may. (Second accuracy pass; ASCO/ASH 2019 full text from PubMed Central, read as raw text) · Each reason reworded to what the guideline states, and the two steps reworded to may and consider.
- 2026-11-17 · Anaemia during chemotherapy: doses, ceiling and iron thresholds: The page said no guideline after 2019 existed and that the labels were not read, and gave no dose, no haemoglobin ceiling and no iron threshold. SEOM published a guideline in 2021, and the US labels and NICE's account of the UK licences can be read. (Second accuracy pass; Epogen and Aranesp labels from DailyMed, NICE TA323 chapter 3, and SEOM 2020 from Europe PMC, all read as raw text) · Starting doses and dose changes, the UK target of up to 120 g/litre, SEOM's 12 g/dL ceiling and the differing iron thresholds added, each attributed.
- 2026-11-17 · Neratinib: diarrhoea prevention with loperamide or dose escalation: stopping rules: The page gave the interrupt and stop rules as if they applied to everyone and said that with capecitabine only capecitabine has its own rules. The US label has a separate neratinib table for the combination, in which grade 4 diarrhoea means holding both drugs, not stopping neratinib for good. (Second accuracy pass; US prescribing information on DailyMed, SPL XML read as raw text) · The interrupt and stop steps now say they are for neratinib taken alone, and the population and gaps say the combination has its own table.
- 2026-11-17 · Neratinib: diarrhoea prevention with loperamide or dose escalation: CONTROL figures: The page gave only the trial paper's figures for loperamide alone (31 per cent grade 3, 20 per cent stopping). The US label's own table gives 32 and 18 per cent among 109 people, and the paper prints the four-week escalation's stopping rate as 6 per cent in its abstract and 7 per cent in its results. The page also gave no cohort sizes. (Second accuracy pass; US prescribing information, the CONTROL paper through Europe PMC and the ClinicalTrials.gov record, all read as raw text) · Both sets of figures are now on the page with who prints which, the cohort sizes are given from the registry, and the differences are named in the gaps.
- 2026-11-17 · Neratinib: diarrhoea prevention with loperamide or dose escalation: colestipol cohort: The page said colestipol was added to scheduled loperamide, implying the 56-day schedule. In that cohort scheduled loperamide ran for 28 days only. (Second accuracy pass; CONTROL final report, full text through Europe PMC) · The variant now gives the 28-day loperamide schedule.
- 2026-11-17 · Neratinib: diarrhoea prevention with loperamide or dose escalation: reasons given for steps: Several reasons said more than the label does: that either strategy makes severe diarrhoea much less likely, that escalation gave the least severe diarrhoea (grade 3 was lowest, but diarrhoea of any grade was more common, 98 against 78 per cent), that the bowel adapts, and what fewer or more bowel movements mean. The label was also said to make both strategies standard, where it says escalation may be considered. (Second accuracy pass; US prescribing information on DailyMed, SPL XML read as raw text) · The reasons and the best-in-class statement were reworded to what the label and the trial report say.
- 2026-11-17 · Neratinib: diarrhoea prevention with loperamide or dose escalation: fluid advice: The patient summary told readers to drink about two litres a day as standing advice. The label gives that figure only for managing diarrhoea once it has started. (Second accuracy pass; US prescribing information on DailyMed, SPL XML read as raw text) · The sentence was removed from the patient summary; the figure stays in the management step, and the gap is named.
- 2026-11-17 · Neratinib: diarrhoea prevention with loperamide or dose escalation: label date and EU label: The US label was dated July 2026, which is when DailyMed published the version; the label text is revised December 2025. The EU summary of product characteristics had not been read, so the page did not show that it has no escalation schedule, prints no loperamide doses and contraindicates neratinib in severe hepatic impairment, where the US label starts at 80 mg. (Second accuracy pass; DailyMed SPL XML and version history, and the European Medicines Agency product information PDF read with pdftotext) · The source year and wording were corrected, and the EU label was added as a source and a variant with each difference stated.
- 2026-11-17 · Intrathecal chemotherapy safety and vinca alkaloids given only by minibag: why a minibag is used: The page said a minibag cannot be connected to a spinal needle, so the wrong-route error becomes impossible. No source says that. ISMP says the bag holds a volume too large for intrathecal administration and that no intrathecal error has been reported with a minibag, and WHO says the bag is meant to deter. (Second accuracy pass; ISMP Targeted Medication Safety Best Practices for Hospitals 2026-2027 and WHO Alert No. 115, both read as PDF text) · The reason now gives the ISMP rationale and attributes the absence of reported cases to ISMP.
- 2026-11-17 · Intrathecal chemotherapy safety and vinca alkaloids given only by minibag: patient summary: The patient summary said vincristine and related drugs are fatal if injected into the spinal fluid. The US label says this usually results in death and ISMP writes of the few survivors. (Second accuracy pass; Hospira vincristine label revised December 2025 (DailyMed XML) and ISMP Best Practices 2026-2027 (PDF text)) · Reworded to say they usually cause paralysis and death.
- 2026-11-17 · Intrathecal chemotherapy safety and vinca alkaloids given only by minibag: count of past errors: The page attributed 55 incidents worldwide to HSC 2008/001, whose full text could not be re-read. The figure is the World Health Organization's, counted from 1968 to July 2007, and ISMP now reports more than 130 cases with vincristine. (Second accuracy pass; WHO Information Exchange System Alert No. 115 and ISMP Best Practices 2026-2027, both read as PDF text) · The count is attributed to WHO with its dates, and the ISMP figure and the Scottish count of at least 13 UK patients since 1985 are given beside it.
- 2026-11-17 · Intrathecal chemotherapy safety and vinca alkaloids given only by minibag: reasons given for the register and the room: Two reasons were stated as fact with no source: that most fatal errors happened when someone unfamiliar with the process gave the injection, and that a curtained bay next to an intravenous chemotherapy chair is where mix-ups happen. No source read says either. (Second accuracy pass; NHS Specialist Pharmacy Service articles of 13 June 2025 (HTML, tags stripped) and WHO Alert No. 115) · Both reasons were reworded to what the NHS text supports.
- 2026-11-17 · Intrathecal chemotherapy safety and vinca alkaloids given only by minibag: checks before the injection: The step cited the English circular for a time out, which is the American standards' term and is in no UK text read. The NHS requires recorded checks by the administering doctor and a second member of staff on the register, with the patient involved if they wish. (Second accuracy pass; NHS Specialist Pharmacy Service, Managing an intrathecal chemotherapy service, 13 June 2025 (HTML, tags stripped)) · The step now states the NHS requirement and quotes it; the separate step on recording the checks was merged into it.
- 2026-11-17 · Intrathecal chemotherapy safety and vinca alkaloids given only by minibag: children: The page said paediatric units rely on intrathecal separation and checks rather than minibags, which no source states, and the plain summary said vincristine is put up as a drip bag without exception. UK guidance supplies vinca alkaloids for children in a syringe of 10 mL or more, and Scottish guidance gives the reason that the practical difficulties of infusions for children outweigh the benefits. ISMP recommends minibags for children too, for example 25 mL. (Second accuracy pass; NHS Specialist Pharmacy Service, Preventing harm from incorrect vinca alkaloid administration, 13 June 2025; Scotland's CEL 22 (2009); ISMP Best Practices 2026-2027) · The variant now gives each position with its source and figures, and the plain summary says usually.
- 2026-11-17 · Intrathecal chemotherapy safety and vinca alkaloids given only by minibag: who it applies to: The population named intraventricular chemotherapy, acute lymphoblastic leukaemia and CNS lymphoma, none of which is in a source that could be read. The NHS text says leukaemias or lymphomas. (Second accuracy pass; NHS Specialist Pharmacy Service, Understanding risks with intrathecal administration, 13 June 2025 (HTML, tags stripped)) · Reworded to leukaemias and lymphomas.
- 2026-11-17 · Intrathecal chemotherapy safety and vinca alkaloids given only by minibag: sources behind the steps: Eight quotations rested on two documents that could not be re-read: the full text of HSC 2008/001 and the ASCO-ONS standards of 2024. The page also lacked the rule on who may prescribe and give the injection, the NRFit connector requirement, the minibag volumes and the label wording. (Second accuracy pass; NHS Specialist Pharmacy Service articles of 13 June 2025, Scotland's CEL 21 and CEL 22 (2009), ISMP Best Practices 2026-2027, WHO Alert No. 115 and the Hospira label, all read as raw text) · The steps were re-sourced to the NHS restatement of June 2025 and to ISMP, every quotation is now checked verbatim, the unread wording is named under gaps, and the missing rules were added.
- 2026-11-17 · Irinotecan: early cholinergic syndrome and late diarrhoea: admission and stool tests: The step on admitting for fever, dehydration or melaena and sending stool for infection and C. difficile was cited to the US label, which gives no admission criteria and no stool tests. Those come from the eviQ algorithm, which also says antibiotics as appropriate. (Second accuracy pass; Camptosar label (DailyMed XML) and eviQ ID 3238 PDF, read as raw text) · Split into three steps: eviQ's admission criteria, the UK product information's own admission triggers, and the US label's fluids and antibiotics advice.
- 2026-11-17 · Irinotecan: early cholinergic syndrome and late diarrhoea: loperamide at night: The summary said the US label and eviQ agree on 4 mg of loperamide every 4 hours at night. Only the US label prints that. eviQ gives no night dose and the UK product information says the 2-hourly regimen should not be modified. (Second accuracy pass; Camptosar label, eviQ PDF and the UK Campto summary of product characteristics (emc), read as raw text) · The night dose is now attributed to the US label alone, quoted in the step, and the disagreement is listed in the gaps.
- 2026-11-17 · Irinotecan: early cholinergic syndrome and late diarrhoea: atropine: The step told staff to give atropine and said giving it before the next dose usually prevents the symptoms. The US label says to consider atropine 0.25 to 1 mg for prevention or treatment and gives no success rate. The UK product information, not read before, says 0.25 mg under the skin should be given as treatment, with prevention recommended only after an acute and severe episode. (Second accuracy pass; Camptosar label (DailyMed XML) and UK Campto summary of product characteristics (emc), read as raw text) · Step reworded to consider, with the dose; a UK step and variant added; the patient summary now says atropine can prevent or ease the symptoms.
- 2026-11-17 · Irinotecan: early cholinergic syndrome and late diarrhoea: antibiotics: The page said the label and eviQ agree on antibiotics for fever, ileus or severe neutropenia. That list is the US label's. eviQ says antibiotics as appropriate, and the UK product information says a broad-spectrum antibiotic when diarrhoea comes with a neutrophil count below 500 per cubic millimetre. (Second accuracy pass; the three sources read as raw text) · Each source's position is now stated separately.
- 2026-11-17 · Irinotecan: early cholinergic syndrome and late diarrhoea: urgency wording: The page said the warning signs need assessment the same day, and the patient summary left out the label's first trigger, diarrhoea for the first time during treatment. No source read sets a same-day limit. The label says to contact the doctor, and Macmillan says to contact the hospital straight away on the 24-hour number. (Second accuracy pass; Camptosar label section 17 and Macmillan's irinotecan page, read as raw text) · Same day removed, Macmillan's wording added as its own step, the first-diarrhoea trigger added to the patient summary, and the missing time limit named in the gaps.
- 2026-11-17 · Irinotecan: early cholinergic syndrome and late diarrhoea: reasons given for steps: Several reasons said more than the sources: that UGT1A1 variants raise the risk of severe side effects in general (the US label says severe neutropenia), that loperamide at the first loose stool stops diarrhoea becoming dangerous, that the high dose is needed, and a nerve mechanism the label does not give. (Second accuracy pass; Camptosar label and UK Campto summary of product characteristics, read as raw text) · Each reason reworded to what the labels state, with the UK text's anticholinesterase explanation attributed to it.
- 2026-11-17 · Intrathecal chemotherapy in children: keeping vincristine away from the spinal injection: The page said no source it had read covered how vinca alkaloids are kept apart from intrathecal drugs in a children's unit, and gave no step. Sources that say so can be read. UK guidance gives a child's vinca alkaloid in a syringe of 10 mL or greater in a children's unit, and Scotland's guidance says intrathecal and intravenous bolus or vinca chemotherapy should never be given at the same time, with the usual order relaxed only for children under general anaesthesia or where a paediatric protocol requires it. ISMP in the United States says a minibag for every patient, with 25 mL as its example for children. (Second accuracy pass; NHS Specialist Pharmacy Service pages of 13 June 2025, Scottish circulars CEL 21 (2009), CEL 22 (2009) and DL(2025)14, and the ISMP best practices for 2026 and 2027, each read as raw text) · Two steps added with their sources, the UK and United States positions shown side by side as a variant, and the gap rewritten.
- 2026-11-17 · Intrathecal chemotherapy in children: doses: The page said doses had not been read from any source. The US label for methotrexate injection prints an intrathecal dose by age: 6 mg under 1 year, 8 mg from 1 to under 2 years, 10 mg from 2 to under 3 years, 12 mg from 3 to under 9 years, and 12 to 15 mg from 9 years, preservative-free only. US cytarabine labels print no age bands, and no label read gives a hydrocortisone dose. (Second accuracy pass; US prescribing information for Methotrexate Injection and Cytarabine Injection (Hospira), raw label text from the openFDA API) · Methotrexate label doses added with their source; the gap now says exactly which doses were and were not found.
- 2026-11-17 · Intrathecal chemotherapy in children: platelets before lumbar puncture: The step said to give platelets only if the count is below 20 and printed the grade as strong, moderate certainty against the JAMA abstract. The recommendation is that transfusion is recommended below 20, and the abstract prints the grade for its group of recommendations as strong with high or moderate certainty; strong and moderate is how the Children's Oncology Group endorsement prints it. The page also said NICE NG24 takes a different position without saying what it is or citing it. (Second accuracy pass; JAMA 2025 abstract through Europe PMC, the Children's Oncology Group endorsement PDF and the NICE NG24 platelet chapter, read as raw text) · Step reworded, grade given as each source prints it, and NICE's 2015 recommendation (consider raising the count above 50 for invasive procedures, lumbar puncture not named) stated and cited as a variant.
- 2026-11-17 · Intrathecal chemotherapy in children: lying flat afterwards: The page said children lie prone or flat for up to an hour to spread the drug, and that the practice rests on an experiment in four rhesus monkeys. Great Ormond Street Hospital says at least one hour, and the abstract of the experiment says four animals in a non-human primate model without naming the species. (Second accuracy pass; the Great Ormond Street Hospital page read as raw text and the 1995 abstract through Europe PMC) · Wording changed to a set time, at least an hour at Great Ormond Street, and to four non-human primates.
- 2026-11-17 · Intrathecal chemotherapy in children: needles and headache: Four statements said more than the nursing review does: the smallest needle (it says smaller, where clinically applicable), a pencil-point hole that closes more readily (it says a less traumatic hole), a blood patch for a headache that persists and a risk only when leukaemia is not in remission (it says for post-puncture headache, with remission status weighed), and reasons for telling the family that no source gave. (Second accuracy pass; full text of the author manuscript (PMC5685494) through NCBI E-utilities, and the NHS lumbar puncture page) · Each statement reworded to the source; the advice to report a headache that does not go away now cites the NHS page, and a gap says it is not written for children having chemotherapy.
- 2026-11-17 · Intrathecal chemotherapy in children: evidence figures in the variants: The page gave 92 lumbar punctures at counts under 20,000 in the Brazilian series, a sum the abstract does not print (it prints 25 and 67), and compared the Dutch cohort's 58 per cent survival after a blood-stained first sample with blasts against 72.6 per cent only, leaving out the paper's own comparison of 82 per cent after a blood-stained sample without blasts. (Second accuracy pass; abstracts through Europe PMC and PubMed) · The two printed counts are given in place of the sum, and all three Dutch figures are given.
- 2026-11-17 · Care in the last days of life: noisy breathing and guideline disagreement: The page said guidelines treat noisy breathing with a medicine once it is distressing and showed no dissent. The ESMO 2021 end of life guideline, which had not been read, says there is no evidence to support anticholinergic medicines for noisy breathing [V, C], that suction should not be used [V, D], that artificial hydration does not improve or prevent thirst [V, C], and recommends octreotide and haloperidol for vomiting in bowel obstruction [II, A], each differing from NICE NG31. (Second accuracy pass; ESMO Open 2021 full text read as Europe PMC XML, NICE NG31 recommendations read as raw text) · ESMO 2021 added as a source with its four graded statements quoted, and the three disagreements stated in the best in class section and the variants.
- 2026-11-17 · Care in the last days of life: medicines for agitation and delirium: The page said NICE advises an antipsychotic for delirium. NICE NG31 1.5.27 says to consider a trial of an antipsychotic to manage delirium or agitation. (Second accuracy pass; NICE NG31 recommendations read as raw text) · The step and the summary now say delirium or agitation, and that each medicine is a trial to consider.
- 2026-11-17 · Care in the last days of life: anticipatory medicines: The page listed five symptoms for anticipatory prescribing and said the usual set of medicines was local practice without a source. NICE NG31 1.6.2 lists agitation, anxiety, breathlessness, nausea and vomiting, noisy respiratory secretions and pain, and prints no doses. The Scottish Palliative Care Guidelines name four medicines and print their doses. (Second accuracy pass; NICE NG31 and the Scottish Palliative Care Guidelines pages on last days of life and anticipatory prescribing read as raw text) · The step lists NICE's symptoms in full, and a variant gives the Scottish medicines and doses as printed, with the differing Korean and ESMO doses beside them.
- 2026-11-17 · Care in the last days of life: assisted hydration: The page said to stop assisted hydration if there are signs of harm, said MASCC applies to anyone with a prognosis of a week or more, said MASCC usually gives fluids under the skin, and said NICE and MASCC both reject a blanket policy. NICE says reduce or stop for signs of possible harm or if the person no longer wants it. MASCC is about advanced cancer, says the intravenous route is usual in hospital and the subcutaneous route well established at home and in hospices, and is the only one of the two to reject a blanket policy in words. (Second accuracy pass; NICE NG31 read as raw text, MASCC 2024 full text read as Europe PMC XML) · Step, summary, best in class statement and variant reworded to each source.
- 2026-11-17 · Care in the last days of life: Korean guideline on delirium: The page said the Korean guideline asks for restraints, catheters and intravenous lines to be minimised in delirium. Recommendation 15 says physical restraints, indwelling catheters and intravenous injections. (Second accuracy pass; Korean guideline full text read as Europe PMC XML) · Wording corrected and the recommendation quoted verbatim.
- 2026-11-17 · Care in the last days of life: reasons given for steps: Eleven of the reasons printed beside steps said things no cited source says, for example that tests cause discomfort, that a ReSPECT form should shape care, that injections into muscle hurt, that oxygen does not ease breathlessness unless levels are low, and that symptoms often come at night. Three actions were firmer than NICE: use non-drug measures (NICE says consider), reposition for noisy breathing (NICE names no measure), and help to drink only if swallowing is safe (NICE says support the person to drink if they wish and are able, and check for swallowing problems). (Second accuracy pass; NICE NG31 recommendations read as raw text) · Each reason and action reworded to what NICE, the Korean guideline or ESMO states. Steps for pain and for nausea and vomiting, which NICE covers and the page lacked, were added with verbatim quotes.
- 2026-11-17 · Care in the last days of life: edition of NICE NG31: The page said NG31 has not been updated since December 2015. NICE lists one minor change, a link to its shared decision making guideline added in October 2021; the recommendations themselves are unchanged. (Second accuracy pass; NICE NG31 update information page read as raw text) · Gap and source note corrected, and the 2017 quality standard QS144 noted.
- 2026-11-17 · Palliative sedation for refractory suffering at the end of life: midazolam with risk factors: The page said to halve the doses of midazolam where there are risk factors. The EAPC medication table says to halve the initial dose and to wait longer, 6 to 8 hours, before raising the maintenance dose. The page also said about 1 mg an hour where the table prints 1 mg/h, and gave only the deep-sedation bolus although the framework says light sedation should generally be tried first. (Second accuracy pass; EAPC 2024 Table 5 and statements 22 and 27, read as text from the Europe PMC XML and Supplemental File 4) · Step and summary now give the initial-dose halving, the 6 to 8 hour interval, 1 mg/h, and both the light and deep bolus doses.
- 2026-11-17 · Palliative sedation for refractory suffering at the end of life: planned interruption of sedation: The page told teams to plan an interruption of sedation outside emergencies. The source says whenever possible, and statement 32 says that in the final stage of life lowering the dose is in most instances not recommended because the distress may return. (Second accuracy pass; EAPC 2024 article and Supplemental File 4, raw text) · Step now says whenever possible and that the dose is not routinely lowered in the final stage of life.
- 2026-11-17 · Palliative sedation for refractory suffering at the end of life: specialist consultation: The page said specialist advice comes before sedation, in the patient summary and in the first step. Statement 13 strongly recommends consultation only where there is uncertainty about the evaluation, and the authors say it was deliberately not made mandatory. (Second accuracy pass; EAPC 2024 Supplemental File 4, raw text) · Reworded with the condition; assessment by a physician with palliative care expertise (statement 12) added.
- 2026-11-17 · Palliative sedation for refractory suffering at the end of life: opioids and missing statements: The page did not say that opioids and haloperidol should not be used to sedate, did not give the distinction from euthanasia, the precautions for existential distress, informed consent, or the monitoring intervals. Its gaps said the 42 statements could not be read; they are text in Supplemental File 4. (Second accuracy pass; EAPC 2024 Supplemental File 4 from the open Europe PMC supplementary files) · Steps added or rewritten from statements 3, 5, 14, 18, 24, 28, 30 and 33; the gap is corrected.
- 2026-11-17 · Palliative sedation for refractory suffering at the end of life: monitoring: The monitoring step asked for a validated tool, a word the source does not use, gave no frequency, and its quote was cut off mid-sentence with spaces added inside brackets. Statement 30 says at least every 20 minutes until sedation is adequate, then at least three times a day. (Second accuracy pass; EAPC 2024 Supplemental File 4 and article XML) · Step rewritten from statement 30 and quoted verbatim.
- 2026-11-17 · Palliative sedation for refractory suffering at the end of life: propofol: The page said propofol is given under an anaesthetist's supervision. Statement 27 says by an anaesthetist or a person with sufficient experience in its use, and that midazolam and levomepromazine are changed for it rather than added to. (Second accuracy pass; EAPC 2024 statement 27 and Table 5) · Step and summary reworded.
- 2026-11-17 · Palliative sedation for refractory suffering at the end of life: how often it is needed: The patient summary began Very occasionally. No source says that; the EAPC article estimates that palliative sedation precedes 10 to 18 per cent of deaths of palliative care patients in Europe. It also said the family takes part without saying this needs the patient's consent. (Second accuracy pass; EAPC 2024 article XML and statement 16) · Reworded to Sometimes, and to those close to them if the person agrees.
- 2026-11-17 · Palliative sedation for refractory suffering at the end of life: life-expectancy limits: A variant said some guidelines allow continuous deep sedation only within hours to days or up to two weeks. No source read states those limits. The ESMO 2021 guideline says sedation where expected prognosis is more than 2 weeks remains controversial. (Second accuracy pass; ESMO 2021 full text from Europe PMC; EAPC 2024) · Variant reworded to what the two sources say; the gap names the national guidelines that were not read.
- 2026-11-17 · Palliative sedation for refractory suffering at the end of life: ESMO 2014 and grades: The page said the ESMO 2014 guideline could not be read. It was read from an Internet Archive capture of the publisher's free page; it rates all its assertions level V and its dose table differs from the EAPC one. The page called the Korean grade low evidence where the guideline defines D as expert opinion, and gave EAPC consensus levels as high or very high where Table 3 prints one level for each statement. (Second accuracy pass; ESMO 2014 archived full text; Korean supplementary table; EAPC 2024 Table 3) · ESMO 2014 and ESMO 2021 added as read sources with a variant on doses; grades set to the printed level; one grade quote corrected to the source's spacing.
- 2026-11-17 · Diarrhoea from chemotherapy, targeted therapy and radiotherapy: which expert panel said what: The page said the 2004 expert panel (Benson and colleagues) would use high-dose loperamide from the first day, then octreotide, and high-dose octreotide first for severe diarrhoea. NCI PDQ cites that sequence to an earlier panel (Kornblau and colleagues, 2000). What PDQ reports of the 2004 panel is intravenous fluids, antibiotics and octreotide at 100 to 150 micrograms three times a day for complicated diarrhoea. (Second accuracy pass; NCI PDQ Gastrointestinal Complications (updated 3 April 2025) read as raw text with its reference list, and both abstracts read through Europe PMC) · The variant and the best-in-class statement now name the 2000 panel, the 2004 panel's advice on radiotherapy diarrhoea is a separate variant, and the 2000 paper is listed as a source.
- 2026-11-17 · Diarrhoea from chemotherapy, targeted therapy and radiotherapy: when a patient should call: The short summary told patients to call if diarrhoea did not settle within a day or reached seven or more stools a day. No source read gives either as a reason to call; seven or more is a grading threshold, counted as an increase over the person's usual number. eviQ's patient sheet says to contact the team immediately for diarrhoea not controlled by medicine, blood, severe stomach pain and bloating, new vomiting, signs of dehydration or a temperature of 38°C or higher. (Second accuracy pass; eviQ patient information sheet ID 3097 version 4, PDF read with pdftotext) · The summary now gives eviQ's warning signs, a step for them was added with the sheet as its source, and a gap says no source sets an hour limit or stool count for ringing.
- 2026-11-17 · Diarrhoea from chemotherapy, targeted therapy and radiotherapy: octreotide step: The octreotide step carried a MASCC/ISOO level II grade, but its dose (100 to 150 micrograms three times a day, rising to 500) is eviQ's, which is ungraded. The MASCC/ISOO grade belongs to a different statement: at least 100 micrograms twice a day in chemotherapy associated with stem cell transplant. The step also said to give octreotide where eviQ says it may be considered, for chemotherapy and targeted therapy diarrhoea. (Second accuracy pass; eviQ ID 779 page and MASCC/ISOO 2014 full text (Europe PMC), read as raw text) · The grade was removed from the step and kept in the grades table beside its own statement; the step now says consider, and names the kinds of diarrhoea it applies to.
- 2026-11-17 · Diarrhoea from chemotherapy, targeted therapy and radiotherapy: loperamide daily maximum: The page gave loperamide doses with no daily maximum. eviQ gives a maximum of 16 mg a day and says higher doses may be used for a short period in chemotherapy diarrhoea; NCI PDQ describes a maximum of about 12 mg a day. (Second accuracy pass; eviQ ID 779 and NCI PDQ read as raw text) · Both figures are now in the summary, the loperamide step and the best-in-class statement, with who says which.
- 2026-11-17 · Diarrhoea from chemotherapy, targeted therapy and radiotherapy: admission, tests and stopping treatment: Three steps said more than the page they cited. Admission and interrupting treatment were written as orders where the eviQ page says people will often require admission and treatment may need to be interrupted; and the C. difficile test and blood count were cited to a page that names only stool culture and electrolytes. The firmer wording and the two tests are in eviQ's separate algorithm, which the page had not cited. (Second accuracy pass; eviQ algorithm ID 3237 version 4 (May 2026), PDF read with pdftotext, beside the eviQ ID 779 page) · The algorithm was added as a source with its own step for admission tests, and the other steps now give the page's wording and the algorithm's side by side.
- 2026-11-17 · Diarrhoea from chemotherapy, targeted therapy and radiotherapy: reasons given for steps: Several reasons said things the sources do not: that diarrhoea turns dangerous within a day or two, that immunotherapy diarrhoea needs steroids and that loperamide can hide it, that octreotide controls most diarrhoea loperamide does not, and that carrying on treatment makes diarrhoea worse. (Second accuracy pass; eviQ ID 779 page read as raw text) · Each reason was reworded to what eviQ states.
- 2026-11-17 · Diarrhoea from chemotherapy, targeted therapy and radiotherapy: grading scale: The page did not print the stool counts for each grade, and did not say that a newer grading scale exists. CTCAE version 5.0 counts an increase of fewer than 4, 4 to 6, and 7 or more stools a day over baseline for grades 1 to 3; version 6.0 (2025) redefines grade 1 as a change in consistency or frequency and adds a need for intravenous treatment to grade 3. (Second accuracy pass; eviQ grading table and the NCI CTCAE version 5.0 and 6.0 spreadsheets) · The version 5.0 counts are now in the summary and the grading step, and a gap describes what version 6.0 changes.
- 2026-11-17 · Anticipatory nausea and vomiting before chemotherapy: the Japanese guideline: The page said the Japanese antiemesis guideline 2023 prefers systematic desensitisation among behavioural treatments. The guideline says the opposite: it suggests not performing non-drug interventions for anticipatory nausea and vomiting (strength of recommendation 2, quality of evidence D), and says benzodiazepines are given. Only the systematic review behind it had been read, and only as an abstract. (Second accuracy pass; the JSCO 2023 guideline update summary (Int J Clin Oncol 2024), full text read from Europe PMC) · The variant, summary and best-in-class statement now give the guideline's recommendation and grade, and set it against the MASCC/ESMO recommendation for behavioural therapy.
- 2026-11-17 · Anticipatory nausea and vomiting before chemotherapy: children: The page said it covered children but cited no children's guideline. POGO (2021, endorsed by the Children's Oncology Group) makes a strong recommendation to optimise control of acute and delayed sickness, conditional recommendations to consider hypnosis, systematic desensitisation or relaxation techniques and to consider lorazepam for a child who has already had anticipatory sickness, and no recommendation on treating an episode. (Second accuracy pass; the Children's Oncology Group summary of endorsed guidelines (version date 15 January 2024), full PDF text, and the POGO abstract from Europe PMC) · A step for children and the three POGO grades were added. No dose for children is given because none of the sources that could be read prints one.
- 2026-11-17 · Anticipatory nausea and vomiting before chemotherapy: reasons given for steps: Four steps gave reasons their sources do not state: that the problem is easily missed unless asked about, that behavioural techniques unlearn the link, that a benzodiazepine reduces the learned nausea, and that relaxation prevents the experiences anticipatory nausea grows from. One step told staff to ask before each cycle, where the source only defines the problem, and one said to treat breakthrough nausea promptly, a timing the source does not give. (Second accuracy pass; eviQ page, MASCC/ESMO 2016 abstract and MASCC/ESMO 2023 full text read as raw text) · Each reason now says what the source says, and the two actions were reworded to match their sources.
- 2026-11-17 · Anticipatory nausea and vomiting before chemotherapy: benzodiazepine wording: The step said to offer a benzodiazepine and the summary for patients said lorazepam is taken on the morning of treatment. eviQ says benzodiazepines may be useful and gives lorazepam 0.5 to 1 mg the night before, repeated 1 to 2 hours before treatment, as an example. (Second accuracy pass; eviQ page (last reviewed 28 August 2024) read as raw text) · The step now says consider, the patient summary says again before treatment, and the dose is labelled as eviQ's example for adults.
- 2026-11-17 · Anticipatory nausea and vomiting before chemotherapy: the nurse-delivered trial: The page described a 20-minute session given by a nurse in place of a course from a psychologist. In the trial the nurse taught the exercise once before the first cycle and gave a recording, which the person used daily at home and at each session; the paper makes no comparison with a psychologist's course. (Second accuracy pass; Hunter et al., Cancer Medicine 2020, full text from Europe PMC) · The variant now describes the intervention as the methods section does.
- 2026-11-17 · Anticipatory nausea and vomiting before chemotherapy: the Japanese meta-analysis: The summary called the studies small and mostly at high risk of bias. The abstract gives no study sizes and says most were at high or unclear risk of bias; the pooled result comes from two studies. (Second accuracy pass; abstract from Europe PMC) · Reworded to two studies, with most included studies at high or unclear risk of bias.
- 2026-11-17 · Nausea and vomiting in advanced cancer not caused by cancer treatment: which guideline is current: The page called the 2016 MASCC/ESMO statement the latest and gave metoclopramide alone as the drug of choice. MASCC published an updated guideline in 2021 that names metoclopramide and haloperidol as first line, and methotrimeprazine (levomepromazine) and olanzapine as second line, each on level II evidence. (Second accuracy pass; MASCC 2021 updated guideline (DOI 10.1007/s00520-021-06437-w), abstract read raw from Europe PMC) · The 2021 guideline is now the cited source for the first-choice step, the summary and the best-supported version, with its grades; the 2016 statement is kept as the earlier position.
- 2026-11-17 · Nausea and vomiting in advanced cancer not caused by cancer treatment: bowel obstruction: The page said MASCC and the Scottish guideline agree on octreotide with an antiemetic for bowel obstruction that cannot be operated on. The Scottish bowel obstruction guideline calls octreotide a second-line antisecretory after hyoscine butylbromide, and NICE NG31, for the last days of life, also starts with hyoscine butylbromide. (Second accuracy pass; Scottish Palliative Care Guidelines bowel obstruction page and NICE NG31 recommendations, read raw; MASCC 2022 full text from Europe PMC) · The disagreement is now stated in the step, in the best-supported version and as a variant, with the MASCC 2022 grades.
- 2026-11-17 · Nausea and vomiting in advanced cancer not caused by cancer treatment: metoclopramide limits: The page recommended metoclopramide titrated to effect and did not say that regulators limit it. The MHRA and the EMA restrict metoclopramide to short-term use (up to 5 days) and an adult maximum of 30 mg in 24 hours because of neurological side effects, and the Scottish guideline marks its own higher and longer palliative doses as unlicensed. (Second accuracy pass; MHRA Drug Safety Update (August 2013) and the EMA referral page, read raw; Scottish nausea and vomiting guideline) · A step, a summary sentence and a gap now state the limit, the Scottish doses beside it, and that neither regulator's notice mentions palliative care. A step on avoiding dopamine-blocking antiemetics in Parkinson's disease was added from the Scottish guideline.
- 2026-11-17 · Nausea and vomiting in advanced cancer not caused by cancer treatment: Scottish cause-based choices: The page listed prochlorperazine among the first-line drugs for oral and oesophageal causes, and wrote several reasons more strongly than the guideline does: that prokinetics cause severe cramping, that QT risk rises at higher doses, and that an anticholinergic cancels a prokinetic. The guideline lists cyclizine alone for oral, pharyngeal or oesophageal irritation, says prokinetics may trigger colic, and says anticholinergics will diminish the prokinetic effect. (Second accuracy pass; Scottish Palliative Care Guidelines nausea and vomiting page (2025), read raw) · The variant and the reasons were reworded to the guideline's own statements.
- 2026-11-17 · EGFR inhibitor rash: pre-emptive skin care, antibiotics and dose changes: how long minocycline is given: The page gave minocycline 100 mg daily "for the first 6 weeks" on the MASCC 2011 guideline. Six weeks is the guideline's figure for doxycycline. For minocycline it says a study found it effective "during the first 8 weeks". (Second accuracy pass; MASCC 2011 full text read as XML from Europe PMC) · The step and summary now give 6 weeks for doxycycline and say minocycline was studied over the first 8 weeks, with the guideline's sentence quoted.
- 2026-11-17 · EGFR inhibitor rash: pre-emptive skin care, antibiotics and dose changes: the grade on the first step: The first step carried "Level II, grade A (systemic tetracycline)". The guideline's sentence has no grade, the grade in its table belongs to minocycline and doxycycline, and the same table lists tetracycline 500 mg twice daily as not recommended. (Second accuracy pass; MASCC 2011 Table 3 read as XML from Europe PMC) · The grade now says the sentence is ungraded and gives the table's grades with the drugs named as printed.
- 2026-11-17 · EGFR inhibitor rash: pre-emptive skin care, antibiotics and dose changes: when panitumumab is stopped: The panitumumab step said to stop at the fourth grade 3 reaction and left out the label's other stopping rule: stop after a grade 4 skin reaction, or a grade 3 reaction that does not recover after 1 or 2 doses are withheld. (Second accuracy pass; VECTIBIX US label read as DailyMed XML) · The step, summary and grades now carry the full rule, with the label's sentence quoted.
- 2026-11-17 · EGFR inhibitor rash: pre-emptive skin care, antibiotics and dose changes: reasons given for steps: Several reasons were not in any cited source: that early treatment prevents most severe cases, that the antibiotic works by an anti-inflammatory effect, that the chlorhexidine wash is to prevent nail infection, that drying acne creams make the rash worse, and that a lower afatinib dose usually lets treatment continue. (Second accuracy pass; MASCC 2011, the RYBREVANT and GILOTRIF labels and the trial abstracts read as raw text) · Each reason now says only what its source says, with trial percentages where a source prints them.
- 2026-11-17 · EGFR inhibitor rash: pre-emptive skin care, antibiotics and dose changes: tetracycline cautions: The page said tetracyclines are not suitable in pregnancy or for some people with allergies without a source. The NHS says doxycycline is not usually used in pregnancy or breastfeeding, though a doctor may recommend it if the benefits outweigh the risks. (Second accuracy pass; NHS doxycycline page read with curl) · The NHS page is now a cited source and the wording follows it. Minocycline's cautions are named as not read.
- 2026-11-17 · EGFR inhibitor rash: pre-emptive skin care, antibiotics and dose changes: the amivantamab label and COCOON: The page said the US amivantamab label recommends the COCOON regimen on the strength of the trial. The label does not name the trial, and it words two items differently: chlorhexidine as a wash for hands and feet, and clindamycin for the next 9 months. (Second accuracy pass; RYBREVANT label read as DailyMed XML and the COCOON abstract from Europe PMC) · The page now says the label's regimen matches the one tested in COCOON and gives both wordings.
- 2026-11-17 · EGFR inhibitor rash: pre-emptive skin care, antibiotics and dose changes: the most recent guidance: The page called the unread ESMO 2021 guideline the most recent international guideline. A 2025 European and US expert consensus on this rash is newer. It backs pre-emptive doxycycline at 100 or 200 mg a day and did not reach consensus on hydrocortisone 1 per cent, favouring a medium-potency steroid cream. (Second accuracy pass; J Eur Acad Dermatol Venereol 2025 position paper read as full text XML from Europe PMC) · The consensus paper is added as a source and as a variant showing where it differs from MASCC 2011.
- 2026-11-17 · EGFR inhibitor rash: pre-emptive skin care, antibiotics and dose changes: who the drugs are for: The page listed panitumumab for colorectal and head and neck cancer. Its label covers metastatic colorectal cancer only; head and neck cancer applies to cetuximab. (Second accuracy pass; VECTIBIX and ERBITUX US labels read as DailyMed XML) · The population line now gives each drug its own cancers.
- 2026-11-17 · Hand-foot syndrome and hand-foot skin reaction: capecitabine dose table: The capecitabine rule gave the restart doses but left out when to stop for good. The US label's table says to discontinue permanently at the fourth appearance of grade 2 and the third appearance of grade 3. (Second accuracy pass; XELODA US prescribing information, DailyMed text read raw) · Both stop rules added to the step and the summary, with the EU text's version (percentages of the starting dose) as a variant.
- 2026-11-17 · Hand-foot syndrome and hand-foot skin reaction: liposomal doxorubicin dose rule: The page said to reduce by 25 per cent after grade 3 and gave no stop rule. The label reduces by 25 per cent after grade 3 or 4, and after a lower grade in someone who has had grade 3 or 4 before, and says to discontinue if the reaction has not resolved after 2 weeks. (Second accuracy pass; DOXIL US prescribing information, DailyMed XML read raw) · Step and summary now give the full table, including the stop rule and that the dose is not raised again.
- 2026-11-17 · Hand-foot syndrome and hand-foot skin reaction: general skin care: The page told readers to avoid heat and hot water, wear well-fitting shoes, keep hands and feet cool and moisturise every day. The cited review says to avoid mechanical and chemical stress, limit water exposure, wear gloves that fit and use emollients regularly; heat, footwear and a daily schedule are in no source read. (Second accuracy pass; Japanese oncodermatology review 2025, full text from Europe PMC) · Reworded to what the review says, in the summary for patients, the plain summary and the step; the missing advice is named in the gaps.
- 2026-11-17 · Hand-foot syndrome and hand-foot skin reaction: pyridoxine: The page said pyridoxine does not work and must not be given. The review says most randomised trials were negative and efficacy is insufficient; the quote also carried a section heading and altered spacing. (Second accuracy pass; Japanese oncodermatology review 2025 and the Xeloda EU product information, read raw) · Reworded to do not rely on it, the exact sentence quoted, and the EU text's advice against vitamin B6 with cisplatin added.
- 2026-11-17 · Hand-foot syndrome and hand-foot skin reaction: reasons given for dose rules: Several reasons were not in any source: that stopping lets skin recover quickly, that a lower dose makes a repeat less likely, that stepping down lets most people continue on regorafenib or cabozantinib, and that diclofenac prevented most moderate cases. (Second accuracy pass; US labels from DailyMed and the D-TORCH abstract from Europe PMC) · Each reason replaced by what the label or trial prints, with figures and denominators.
- 2026-11-17 · Hand-foot syndrome and hand-foot skin reaction: grading: The summary said doses are modified by CTCAE grade. The liposomal doxorubicin and sorafenib labels print their own grade descriptions, and the page gave no grade definitions. (Second accuracy pass; CTCAE version 5.0 spreadsheet and the labels, read raw) · CTCAE definitions for grades 1 to 3 added, the two labels with their own scales named, and the EU capecitabine definitions given in a variant.
- 2026-11-17 · Hand-foot syndrome and hand-foot skin reaction: who should stop tablets: The plain summary told everyone to stop the tablets if hands or feet hurt or blister. Only the capecitabine label and leaflet tell patients to stop themselves; the other labels say to tell the healthcare provider. (Second accuracy pass; Xeloda US label and EU leaflet, Stivarga, Doxil and Cabometyx labels, read raw) · Stop rule limited to capecitabine; a step added on reporting symptoms promptly with the other medicines.
- 2026-11-17 · Hand-foot syndrome and hand-foot skin reaction: urea cream and S-1 wording: The blinded urea trial was described as 20 per cent cream that failed its primary measure, but its abstract prints neither the strength nor a primary measure. The S-1 step said to consider a switch once the reaction settled, where the panel recommends the switch at full dose, preferably once it is grade 1 or less. (Second accuracy pass; Lee 2020 abstract and the S-1 recommendations full text, Europe PMC) · Strength attributed to the review that reports it, result stated as cumulative incidence, and the S-1 step reworded to the panel's recommendation.
- 2026-11-17 · Red cell and platelet transfusion thresholds in cancer care: platelets before central lines and other procedures: The summary said AABB and ICTMG set lower thresholds than NICE for some procedures and gave central venous catheters as the example. NICE says not to offer platelets to adults having central venous cannulation at all, while AABB and ICTMG recommend them below 10 x 10^9/L for catheters in compressible sites. NICE's higher threshold of 50 to 75 x 10^9/L for people at high risk of bleeding was also missing. (Second accuracy pass; NICE NG24 platelet chapter read as raw text, the AABB and ICTMG 2025 abstract from Europe PMC and the British Society for Haematology summary PDF) · The summary, the best in class statement and the step now set the three bodies' positions side by side, a step for NICE's procedure thresholds was added, and the British Society for Haematology thresholds (20 for central lines, 40 for lumbar puncture) were added as a variant.
- 2026-11-17 · Red cell and platelet transfusion thresholds in cancer care: "consider" written as an order: The page told readers to give one unit of red cells at a time and to apply the 7 g/dL threshold to people with cancer. NICE says to consider single units, and AABB suggests considering transfusion below 7 g/dL on a conditional grade. (Second accuracy pass; NICE NG24 red blood cell chapter read as raw text and the AABB 2023 abstract from Europe PMC) · Both steps and the summary now say consider.
- 2026-11-17 · Red cell and platelet transfusion thresholds in cancer care: grades: Four grades were not as printed, and the AABB option of 8 g/dL with cardiovascular disease was given the strong grade that belongs to the 7 g/dL recommendation before it. The platelet abstract grades its recommendations in two groups, which the page showed as individual grades. (Second accuracy pass; AABB 2023 and AABB and ICTMG 2025 abstracts from Europe PMC) · Grades now carry the printed wording, the group grading is stated, and the 8 g/dL option is described as a choice with no separate grade in the abstract.
- 2026-11-17 · Red cell and platelet transfusion thresholds in cancer care: patient summary: The patient summary said that giving less blood at higher thresholds had proved just as safe, which had the direction backwards and was stronger than the source. AABB says that for most patient populations restrictive thresholds did not adversely affect outcomes that matter to patients. (Second accuracy pass; AABB 2023 abstract from Europe PMC) · Reworded to say that waiting for a lower level did not lead to worse outcomes for most groups of patients.
- 2026-11-17 · Red cell and platelet transfusion thresholds in cancer care: reasons given for steps: Seven reasons said more than the sources: that 70 g/L was used in most trials, that each extra unit adds risk, that higher platelet counts add risk without benefit, that brain or eye bleeding is dangerous at otherwise safe counts, and others. None is in the text read. (Second accuracy pass; NICE NG24 chapters read as raw text and the AABB abstracts from Europe PMC) · Each reason now says only what the cited passage says.
- 2026-11-17 · Red cell and platelet transfusion thresholds in cancer care: autologous transplant and scope: The page merged ASCO's statement (adults at experienced centres may be transfused at the first sign of bleeding) with AABB and ICTMG's (prophylaxis not recommended), wrote NICE's 10 x 10^9/L threshold as if it were scoped to chemotherapy and transplant, and changed NICE's autoimmune thrombocytopenia to immune thrombocytopenia. NICE NG24 is a general guideline with no cancer-specific recommendations and no autologous transplant exception. (Second accuracy pass; NICE NG24 read as raw text, ASCO 2018 and AABB and ICTMG 2025 abstracts from Europe PMC) · Each body's position is stated separately, NICE's scope is stated in the population and gaps, and the term was restored.
- 2026-11-17 · Premedication against infusion reactions before taxanes, antibodies and other reaction-prone anticancer drugs: daratumumab: The page gave ESMO's 2017 scheme as current: premedication one hour before every infusion and an oral corticosteroid on the two days after. The current US and EU product information say 1 to 3 hours before, use dexamethasone 20 mg rather than methylprednisolone 100 mg in combination regimens, and make the two days of oral corticosteroid afterwards firm only when daratumumab is given alone. (Second accuracy pass; Darzalex US prescribing information (DailyMed, revised June 2026) and EMA product information, read as raw text) · The step now rests on the label, with ESMO's 2017 wording and grade kept beside it and the monotherapy and combination doses set out separately.
- 2026-11-17 · Premedication against infusion reactions before taxanes, antibodies and other reaction-prone anticancer drugs: cetuximab: The page said cetuximab premedication can stop after the second infusion, on ESMO alone. The EU product information requires an antihistamine and a corticosteroid before the first infusion and recommends them before all later ones. The US label asks for an H1 antihistamine before the first dose and later doses as judged necessary. (Second accuracy pass; Erbitux EMA product information and US prescribing information (DailyMed), read as raw text) · A cetuximab step added that states all three positions, and the variant now says who holds each.
- 2026-11-17 · Premedication against infusion reactions before taxanes, antibodies and other reaction-prone anticancer drugs: docetaxel in prostate cancer: The page gave dexamethasone 8 mg at 12, 3 and 1 hours before docetaxel for prostate cancer without its scope. The US label gives it for metastatic castration-resistant prostate cancer because prednisone is given as well. The EU text also gives it in hormone-sensitive disease, and says the usual three-day regimen can be used unless contraindicated. (Second accuracy pass; docetaxel US prescribing information (DailyMed, revised July 2026) and Taxotere EMA product information, read as raw text) · Scope added to the step and the summary, with the label sentence quoted.
- 2026-11-17 · Premedication against infusion reactions before taxanes, antibodies and other reaction-prone anticancer drugs: reasons given beside the steps: Five reasons said more than the sources. Paclitaxel reactions were said to be mostly caused by histamine and mostly prevented, where ESMO says they are believed to be due to mast cell mediators and gives no such figure. Missed tablets were called a common reason for a reaction, rituximab reactions were said to come mainly from cytokine release and to be made milder by premedication, and the docetaxel steroid was said to lower the risk of an allergic reaction where the label says the severity. (Second accuracy pass; ESMO 2017 guideline read from an Internet Archive capture, and the US labels read from DailyMed) · Each reason rewritten to what the cited passage says, with the source's own figures.
- 2026-11-17 · Premedication against infusion reactions before taxanes, antibodies and other reaction-prone anticancer drugs: rituximab and scope wording: The page said glucocorticoids are considered in lymphoma and CLL. ESMO says non-Hodgkin lymphoma and CLL and grades that [IV, B], and the EU product information adds that it applies when the regimen does not already contain a glucocorticoid. The page also said no premedication before PD-L1 antibodies and platinums as classes, where ESMO's rows are for atezolizumab, carboplatin and oxaliplatin. (Second accuracy pass; ESMO 2017 guideline and MabThera EMA product information, read as raw text) · Wording narrowed to what the sources name.
- 2026-11-17 · Premedication against infusion reactions before taxanes, antibodies and other reaction-prone anticancer drugs: sources, ranitidine and stopping premedication: The page did not mention the 2018 correction notice to the ESMO guideline, said ranitidine had been withdrawn in several countries with no source, and said nothing about cabazitaxel or about the studies of stopping paclitaxel premedication after two uneventful doses. (Second accuracy pass; Europe PMC records, the EMA ranitidine referral page, the FDA announcement of 1 April 2020, the Jevtana label, and the abstracts of five studies) · Correction notice listed as unread with a gap. Ranitidine statement now cites the EU suspension and the US withdrawal request, and says no source read names a replacement. Cabazitaxel step added from its label. Stopping premedication and leaving out the H2 antagonist are now variants with their figures, marked as abstracts only and as supported by no guideline or label read.
- 2026-11-17 · GVHD prophylaxis after allogeneic stem cell transplant: adding antithymocyte globulin: The step said to consider adding rabbit antithymocyte globulin. The Brazilian review says its use with standard prophylaxis is recommended over standard prophylaxis alone. Its grade and the grade for mycophenolate were also paraphrased rather than printed. (Second accuracy pass; ABHH/SBTMO 2026 full text read as XML from Europe PMC) · The step now says to add it, names the source, and both grades use the printed wording.
- 2026-11-17 · GVHD prophylaxis after allogeneic stem cell transplant: Epstein-Barr virus: The step told teams to monitor for Epstein-Barr virus reactivation when cyclophosphamide is combined with antithymocyte globulin. The review reports the higher risk and gives no instruction to monitor. (Second accuracy pass; ABHH/SBTMO 2026 full text read as XML from Europe PMC) · The step now says to take the risk into account, and the gaps say no source read gives a monitoring schedule.
- 2026-11-17 · GVHD prophylaxis after allogeneic stem cell transplant: haploidentical donors: The step read as an instruction to give post-transplant cyclophosphamide to haploidentical donors. The review only compares two partners for it, a calcineurin inhibitor against sirolimus, and finds no significant difference in GVHD rather than the same control. (Second accuracy pass; ABHH/SBTMO 2026 full text read as XML from Europe PMC) · The step is now about the choice of partner drug and uses the source's wording on GVHD.
- 2026-11-17 · GVHD prophylaxis after allogeneic stem cell transplant: reduced-intensity conditioning: The summary listed a calcineurin inhibitor with mycophenolate after reduced-intensity conditioning without saying that the review found higher risks of acute and chronic GVHD with it, and recommends it for a lower risk of death. (Second accuracy pass; ABHH/SBTMO 2026 full text read as XML from Europe PMC) · The higher GVHD risks and the reason for the recommendation are now stated.
- 2026-11-17 · GVHD prophylaxis after allogeneic stem cell transplant: abatacept: The page said the abatacept trials were pooled by the review, which pooled only questions 3 to 8, and gave only the review's reading. The US label says the randomised matched cohort did not significantly improve survival free of grade III to IV acute GVHD and did improve overall survival at day 180. (Second accuracy pass; ABHH/SBTMO 2026 full text and the Orencia label read as XML from DailyMed) · Both readings are now given, with the licensed indication and dose from the label.
- 2026-11-17 · GVHD prophylaxis after allogeneic stem cell transplant: who it covers and why: The page said the European and Brazilian recommendations address adults, and that antithymocyte globulin and cyclophosphamide both reduce GVHD according to the European abstract. Neither source states an age range and the abstract makes no efficacy statement. The page also said less kidney and lung toxicity for any calcineurin inhibitor, which the review finds for tacrolimus. (Second accuracy pass; EBMT 2024 abstract from Europe PMC and ABHH/SBTMO 2026 full text) · Reworded to what each source states.
- 2026-11-17 · GVHD prophylaxis after allogeneic stem cell transplant: the BMT CTN 1703 trial: The page said the trial could be read only as its abstract and gave no regimen, toxicity or infection data. The author manuscript is open in PubMed Central. (Second accuracy pass; full text read as XML through NCBI E-utilities, and the two-year report's abstract from Europe PMC) · The trial's doses and days, taper timing, GVHD and infection rates, event counts and two-year result are now on the page as the trial prints them.
- 2026-11-17 · Veno-occlusive disease of the liver after stem cell transplant: how long defibrotide is given: The step said to continue defibrotide for at least 21 days and until the signs resolve, with no upper limit. The US label says to continue until resolution or up to a maximum of 60 days. The European label gives no maximum. (Second accuracy pass; Defitelio US prescribing information section 2.1 (DailyMed label XML) and the European product information section 4.2 (EMA PDF), both read as raw text) · The step now states the 60 day maximum from the US label and says the European label gives none.
- 2026-11-17 · Veno-occlusive disease of the liver after stem cell transplant: blood thinners with defibrotide: The page gave only the US position and cited the wrong label section for it. The US label contraindicates systemic anticoagulants and fibrinolytics (section 4) and says not to start defibrotide during active bleeding (section 5.1). The European label contraindicates only thrombolytic therapy and says other anticoagulants are not recommended and need careful monitoring. (Second accuracy pass; Defitelio US prescribing information sections 2.3, 4 and 5.1 and the European product information sections 4.3 to 4.5, read as raw text) · The step, summary and gaps now give both labels, the active bleeding warning was added, and the section reference was corrected.
- 2026-11-17 · Veno-occlusive disease of the liver after stem cell transplant: defibrotide for prevention: The page said the European Medicines Agency's text could not be read and that the EBMT papers had manufacturer support. The Agency's communication of 13 June 2022 reads today and says Defitelio is not to be used as prophylaxis. The 2026 EBMT paediatric paper declares no competing interests, and it says the HARMONY trial included lower-risk patients, not mainly lower-risk patients. (Second accuracy pass; EMA direct healthcare professional communication and product information (EMA PDFs), the 2026 EBMT paper full text (Europe PMC) and the 2012 paediatric trial abstract (Europe PMC)) · The Agency is now quoted directly and added to the grades, each document's funding declaration is stated separately, trial arm sizes were added, and the European indication (severe disease) is given.
- 2026-11-17 · Veno-occlusive disease of the liver after stem cell transplant: best in class statement: The statement said all sources agree defibrotide at 25 mg/kg a day is the only specific medicine, and that the EBMT documents were written with grants from its maker. The British abstract gives no treatment dose and does not say only, and the 2023 EBMT paper declares author honoraria, not a grant. (Second accuracy pass; BCSH/BSBMT 2013 abstract (Europe PMC) and the EBMT 2023 paper full text (publisher's free PDF, pdftotext)) · Reworded to say who gives the dose and who calls defibrotide the only agent with proven efficacy, and to describe the funding declarations accurately.
- 2026-11-17 · Veno-occlusive disease of the liver after stem cell transplant: ursodeoxycholic acid step: The step gave the EBMT schedule (start of conditioning to day 90) without saying that the EBMT statement is written for adults. (Second accuracy pass; EBMT 2020 position statement and EBMT Handbook 2024 chapter 49, full text) · The step now begins In adults and notes that the British guideline suggests the medicine (2C) without a schedule in its abstract.
- 2026-11-17 · Veno-occlusive disease of the liver after stem cell transplant: reasons given for steps: Several reasons said more than the sources: that prostaglandin E1 caused harm (the cited EBMT text does not name it), that a transjugular biopsy is safer when platelets are low, that too little fluid harms the kidneys, that some units add high-dose methylprednisolone particularly in children, that no trial supports the platelet threshold of 30, and that elastography is not available everywhere (the Handbook says that of pressure measurement). (Second accuracy pass; EBMT 2020 full text, BCSH/BSBMT 2013 abstract and EBMT Handbook 2024 chapter 49, read as raw text) · Each sentence was reworded to what its source states, or removed.
- 2026-11-17 · Platelet transfusion thresholds for children with cancer: UK thresholds said to be missing: The page said no guideline read gave a platelet threshold for central line insertion in children, that nothing read addressed raising the threshold for infection, and that the thresholds for bleeding came from NICE alone. The NHS Blood and Transplant summary of the British Society for Haematology children's guideline suggests 20 x 10^9/L with sepsis, severe mucositis and some other risks or before a non-tunnelled central line, 40 before lumbar puncture, 50 for moderate haemorrhage, surgery or a tunnelled line, and 75 to 100 for major haemorrhage or surgery at critical sites, with a typical dose of 10 to 20 mL/kg and normally no more than one pack. (Second accuracy pass; NHS Blood and Transplant bookmark BLC736.1P (updated November 2020), PDF read as raw text) · Three steps, a variant and a source were added, the lumbar puncture difference is now 20 against 40 with who says which, and the gaps were rewritten.
- 2026-11-17 · Platelet transfusion thresholds for children with cancer: a newer British guideline: The page listed the 2016 British Society for Haematology guideline with its 2020 addendum as the UK source. A new edition was published online on 12 November 2026 in Transfusion Medicine (doi 10.1111/tme.70117). (Second accuracy pass; Europe PMC search and Crossref record; the publisher refused a script and no abstract is held) · The new edition is listed as a source that could not be read, and the gaps say the UK figures come from the summary of the earlier edition and may have changed.
- 2026-11-17 · Platelet transfusion thresholds for children with cancer: grades on two steps: The steps for the thresholds of 10 and of 20 before lumbar puncture cited the journal abstract and carried the grade Strong / Moderate. The abstract prints only a group grade, strong recommendations with high/moderate-certainty evidence, and Strong / Moderate is printed in the Children's Oncology Group reprint. (Second accuracy pass; AABB and ICTMG 2025 abstract from Europe PMC and the Children's Oncology Group endorsement PDF read as raw text) · Each step now carries the grade its cited source prints, and the reprint's grade is named beside it and kept in the grades table.
- 2026-11-17 · Platelet transfusion thresholds for children with cancer: NICE and lumbar puncture: A variant said NICE treats a lumbar puncture as an invasive procedure needing a count above 50. NICE NG24 does not mention lumbar puncture, and its recommendation is to consider raising the count above 50 for invasive procedures or surgery in general. Its recommendation to consider 50 to 75 for people at high risk of bleeding was missing, and its scope was given as children over 1 year where NICE says aged 1 year and over. (Second accuracy pass; NICE NG24 platelet chapter, overview and inclusions chapter read as raw text) · The variant now rests on the British threshold of 40, NICE is described as naming no lumbar puncture threshold, the 50 to 75 recommendation was added and the scope corrected.
- 2026-11-17 · Platelet transfusion thresholds for children with cancer: reasons given for steps: Several reasons said more than the sources: that transfusing at higher counts did not reduce bleeding or deaths (the source says restrictive strategies probably did not increase them), that each transfusion carries a risk of reactions, that a single dose is usually enough, and that bleeding in the brain or eye does most harm. The patient summary made the first claim without the word probably. (Second accuracy pass; AABB and ICTMG 2025 abstract from Europe PMC and NICE NG24 read as raw text) · Each reason and the patient summary now say only what the cited passage says.
- 2026-11-17 · Stem cell transplant: conditioning, the low blood count phase and engraftment monitoring: engraftment syndrome step: The step said to stop G-CSF, take cultures and start antibiotics, "then consider a corticosteroid". The EBMT Handbook chapter does not hedge: after 48 hours, without stopping the antibiotics, it says to start methylprednisolone 1 mg/kg every 12 hours intravenously for 3 days and taper it over a week. (Second accuracy pass; EBMT Handbook 2024 chapter 42, chapter XML from NCBI's open-access archive read as text) · The step and the summary now give the timing and dose as the Handbook prints them, marked as the chapter authors' advice with no grade.
- 2026-11-17 · Stem cell transplant: conditioning, the low blood count phase and engraftment monitoring: protective rooms: The page said the 2009 global guidelines and the EBMT Handbook differ on protective environment rooms, and that the Handbook no longer endorses one. The Handbook repeats the 2009 room specification for allogeneic recipients. Its "no consensus" sentence is about protective (reverse) isolation for neutropenic patients. Both sources say the benefit after an autologous transplant is not well established. (Second accuracy pass; EBMT Handbook 2024 chapter 27 and the 2009 guidelines (PMC3103296), both read as raw text) · Best-in-class statement, the variant and the gap rewritten; the scoping to allogeneic and autologous recipients added.
- 2026-11-17 · Stem cell transplant: conditioning, the low blood count phase and engraftment monitoring: patient summary: The patient summary said conditioning lasts a few days and clears space in the bone marrow, that counts stay very low for about two to four weeks, and that blood is checked every day. No cited source gave those durations, the Handbook calls the "space" purpose controversial, and daily counts come from the busulfan label alone. (Second accuracy pass; EBMT Handbook 2024 chapter 13, the busulfan label and the NHS transplant pages (last reviewed 7 May 2026), read as raw text) · Reworded to what the NHS pages and the Handbook say (usually 1 to 10 days, some weeks, checked often); NHS added as a source; a gap says no source gives a general figure for how long counts stay low.
- 2026-11-17 · Stem cell transplant: conditioning, the low blood count phase and engraftment monitoring: growth factor after transplant: The page named the 2015 ASCO guideline on white cell growth factors and said no later graded recommendation was found. ASCO published an update on 25 February 2026. Its abstract, like the 2015 one, states no recommendation on growth factor after a transplant. (Second accuracy pass; ASCO 2026 abstract from Europe PMC (a sibling checker's copy fetched 17 November 2026); full text refuses a script) · The variant and the gap now name the 2026 update, added as an abstract-only source.
- 2026-11-17 · Stem cell transplant: conditioning, the low blood count phase and engraftment monitoring: daily blood counts: The step told teams to take a blood count every day until engraftment for any regimen, citing the busulfan label. The label speaks for busulfan with cyclophosphamide only, and the FACT-JACIE standards require a policy for monitoring counts without setting how often. (Second accuracy pass; Busulfex label version 33 (DailyMed XML) and FACT-JACIE 9.1 standard B7.7.2 (PDF read with pdftotext)) · The step is scoped to busulfan and its reason names B7.7.2.
- 2026-11-17 · Stem cell transplant: conditioning, the low blood count phase and engraftment monitoring: grades: Three grades were attached too widely. The hand hygiene step carried AII for its visitors clause, which the 2009 guidelines grade BIII. The room variant said every room feature other than the filter is BIII, but the 12 air changes an hour carry no printed grade. The order-set step cited B7.4.2 for content that is in B7.4.1. (Second accuracy pass; 2009 global guidelines (PMC3103296) and FACT-JACIE 9.1, read as raw text) · Each grade now names the part it belongs to.
- 2026-11-17 · Stem cell transplant: conditioning, the low blood count phase and engraftment monitoring: reasons given for steps: Several reasons said more than their sources: that a single room lowers infection (the 2009 guidelines say the effect of isolation precautions has not been evaluated), that hands carry most hospital infections, that irradiation stops white cells dividing, that platelets usually recover after neutrophils, that graft failure treatments work better before infection sets in, that checks happen at the bedside, and that the doses are among the highest in medicine. (Second accuracy pass; each cited source read as raw text) · Each reason reworded to what its source states, or the unsupported part removed.
- 2026-11-17 · Procedural pain and sedation for children with cancer: who gives deep sedation drugs: The page said an anaesthetist must be available for propofol, sevoflurane or opioids with ketamine. NICE CG112 recommendation 1.4.4 says a healthcare professional trained in delivering anaesthetic agents. (Second accuracy pass; NICE CG112 recommendations page read as raw HTML) · Reworded in the summary and the step to NICE's own words.
- 2026-11-17 · Procedural pain and sedation for children with cancer: bone marrow tests: The page said the Children's Oncology Group review could be read only as an abstract and gave no recommendation for bone marrow tests. The full text is open, and it makes a strong recommendation on moderate evidence for moderate sedation or general anaesthesia for bone marrow aspirates and biopsies, and no overall recommendation on the level of sedation. (Second accuracy pass; Children's Oncology Group review read in full as HTML from PubMed Central) · Added the recommendation as a step with its grade and quote, a variant setting it beside the international guideline, and rewrote the gap.
- 2026-11-17 · Procedural pain and sedation for children with cancer: what the international guideline covers: The page attributed to the 2020 international guideline a list of procedures (port access, venepuncture, cannulation, injections, bone marrow aspiration), hypnosis only where staff are trained, and reasons such as low cost and children needing to keep still. Only its abstract could be read, and the abstract says none of these. (Second accuracy pass; abstract from Europe PMC, full text still behind a browser check) · Reworded the population, three steps and their reasons to what the abstract says.
- 2026-11-17 · Procedural pain and sedation for children with cancer: reasons given for NICE steps: Eight steps citing NICE CG112 gave reasons NICE does not state (for example that fasting protects the airway, that parents make children calmer, that sedatives outlast the procedure), said capnography and blood pressure are monitored where possible when NICE's condition is that the monitoring does not wake the child, and said a local anaesthetic is used in all cases when NICE says to consider one. (Second accuracy pass; NICE CG112 recommendations page read as raw HTML) · Each reason now says what the recommendation says, the monitoring condition and the word consider are restored, and infants are added to those needing specialist advice.
- 2026-11-17 · Procedural pain and sedation for children with cancer: fasting times: The page gave only the 1 hour figure for clear fluids and dated NICE's fasting note to October 2018. NICE's note prints 2 hours for clear fluids, 4 for breast milk and 6 for solids with clear fluids reduced to 1 hour, and was added in February 2019. A 2022 European guideline recommends 3 hours for breast milk. (Second accuracy pass; NICE CG112 recommendations and update pages as raw HTML, ESAIC 2022 abstract from Europe PMC) · Printed the fasting hours as each body gives them, added the 2022 guideline and the 2020 sedation consensus as a variant, and corrected the date.
- 2026-11-17 · Clostridioides difficile infection in children with cancer or after stem cell transplant: probiotics as treatment: One step covered both bezlotoxumab and probiotics as treatment and gave them a single grade, conditional on moderate quality evidence. The guideline grades the monoclonal antibody recommendation moderate and the probiotics recommendation low. (Second accuracy pass; Patel et al. 2024, full text from Europe PMC read raw) · Split into two steps, each with its own quotation and grade.
- 2026-11-17 · Clostridioides difficile infection in children with cancer or after stem cell transplant: fidaxomicin evidence: The page said one trial that included children with cancer showed more cures and halved relapses with fidaxomicin. The guideline's figures are a pooled result from eight trials, one of them in children, and it prints a risk ratio rather than saying relapses were halved. (Second accuracy pass; Patel et al. 2024, full text from Europe PMC read raw) · Reworded to the pooled result of eight trials and removed the word halved.
- 2026-11-17 · Clostridioides difficile infection in children with cancer or after stem cell transplant: doses: The page said doses come from local formularies, which no source states, and gave none. The IDSA and SHEA 2017 update prints children's doses, for example vancomycin 10 mg/kg per dose four times a day for 10 days. (Second accuracy pass; IDSA and SHEA 2017 update, full text from PubMed Central, dose table read cell by cell) · Added the IDSA and SHEA doses, maxima and durations as a named variant, with the fidaxomicin dose from the US label, and corrected the gap.
- 2026-11-17 · Clostridioides difficile infection in children with cancer or after stem cell transplant: faecal transplant and probiotics wording: The faecal transplant step said not to use it routinely during treatment, a limit the recommendation does not carry, and the probiotics step said probiotics have caused infections where the guideline says three cases were attributed to a probiotic by the study authors. (Second accuracy pass; Patel et al. 2024, full text from Europe PMC read raw) · Both steps reworded to what the guideline says.
- 2026-11-17 · Clostridioides difficile infection in children with cancer or after stem cell transplant: endorsement document grades: The page said the Children's Oncology Group document reprints the guideline's grades. It prints low quality evidence for fidaxomicin in recurrence and for monoclonal antibodies, where the guideline prints moderate. (Second accuracy pass; COG endorsement PDF read with pdftotext beside the guideline text) · The difference is now stated in the source note and in the gaps; the page keeps the guideline's own grades.
- 2026-11-17 · Clostridioides difficile infection in children with cancer or after stem cell transplant: comparison with other guidelines: The page said adult IDSA and SHEA guidance could not be read and made no comparison. The 2021 update now reads in full, the 2017 update covers children, and NICE NG199 covers under-18s. They differ from the paediatric oncology guideline on metronidazole, fidaxomicin and faecal transplant. (Second accuracy pass; idsociety.org, PubMed Central and nice.org.uk read raw) · Added testing, single room and hand washing steps from IDSA and SHEA 2017, variants for IDSA and SHEA, NICE and adult guidance, and a note that bezlotoxumab's European authorisation was withdrawn on 20 October 2026.
- 2026-11-17 · Peripheral blood stem cell mobilisation and collection for transplant, in patients and donors: plerixafor on the US label: The page said the US label gives plerixafor to every patient with lymphoma or myeloma. The label is an indication for non-Hodgkin lymphoma and myeloma that sets no blood count threshold; it does not say every patient should have it. The page also left out the dose. (Second accuracy pass; Mozobil US prescribing information on DailyMed, label XML read as raw text) · Reworded in the summary, the best-in-class statement and the variant; the label dose (0.24 mg/kg or a fixed 20 mg at 83 kg or less, not above 40 mg a day, reduced by one third to 0.16 mg/kg when creatinine clearance is 50 mL/min or less) added with verbatim quotes.
- 2026-11-17 · Peripheral blood stem cell mobilisation and collection for transplant, in patients and donors: filgrastim in healthy donors: The page used the US filgrastim label for donor steps and said only that the plerixafor and motixafortide labels cover autologous collection. The filgrastim label states that mobilisation in healthy donors is not an approved indication. (Second accuracy pass; Neupogen US prescribing information (revised 07/2026) on DailyMed, label XML read as raw text) · Population, summary, the donor variant and a new grade row now say so, with the label sentence quoted.
- 2026-11-17 · Peripheral blood stem cell mobilisation and collection for transplant, in patients and donors: collecting after lenalidomide: The page gave no rule on when to collect in people with myeloma taking lenalidomide. The US lenalidomide label says mobilisation should occur within 4 cycles, and the IMWG consensus of 2009 recommends collection preferably within the first 4 cycles. (Second accuracy pass; Revlimid US prescribing information on DailyMed read as raw text, and the IMWG 2009 abstract from Europe PMC) · New step, new variant and a grade row added, each with a verbatim quote; the gap on daratumumab now says no source read gives a number of cycles for it.
- 2026-11-17 · Peripheral blood stem cell mobilisation and collection for transplant, in patients and donors: reasons given for steps: Nine reasons given beside steps said more than the sources: that splenic rupture and serious reactions are rare (the label gives no frequency), that a later collection often fails (the Handbook says increased risk), and several rationales for donor standards that the standards do not print. (Second accuracy pass; EBMT Handbook chapter 16 XML, WMDA 2024 Standards PDF and the Neupogen label, all read as raw text) · Each reason reworded to what the cited source says.
- 2026-11-17 · Peripheral blood stem cell mobilisation and collection for transplant, in patients and donors: choosing and monitoring: The page told teams to choose G-CSF alone where the Handbook says it should be considered, dropped the Handbook's words at the latest from the start of CD34 monitoring, and called the donor plerixafor reports case series where the Handbook says studies. (Second accuracy pass; EBMT Handbook 8th edition chapter 16, XML from the NCBI open-access archive read as raw text) · Wording brought into line with the Handbook.
- 2026-11-17 · Peripheral blood stem cell mobilisation and collection for transplant, in patients and donors: motixafortide: The motixafortide variant gave the timing but not the dose or the premedication the label requires before every dose because of anaphylactic shock (0.7% in clinical studies). (Second accuracy pass; Aphexda US prescribing information on DailyMed, label XML read as raw text) · Dose (1.25 mg/kg), premedication and the reaction figures added to the variant.
- 2026-11-17 · Preventing mouth sores (oral mucositis) in children having cancer treatment or a stem cell transplant: MASCC/ISOO statements: The page called the MASCC/ISOO 2020 guideline an adult guideline and labelled its palifermin, mouth care and chlorhexidine statements as adult recommendations. The guideline names adults only in its photobiomodulation statements, sets no age limit on the others, and includes one statement about children (chewing gum). (Second accuracy pass; MASCC/ISOO 2020 full text XML from Europe PMC, read raw) · The labels now say the guideline is not specific to children, and the chewing gum statement was added.
- 2026-11-17 · Preventing mouth sores (oral mucositis) in children having cancer treatment or a stem cell transplant: how cryotherapy is given: The cryotherapy step said to give ice chips or ice water and that it suits infusions of 30 minutes or less. No cited source described the method, ice water is in none of them, and the 2021 guideline says only that infusions in the trials lasted 30 minutes or less where a time was given. (Second accuracy pass; Children's Oncology Group endorsement PDF and the 2015 edition full text from Europe PMC, read raw) · The step now says ice held in the mouth, credits the description (ice cubes or chips, replenished through the infusion) to the 2015 edition, and reports the 30 minutes as what the trials did.
- 2026-11-17 · Preventing mouth sores (oral mucositis) in children having cancer treatment or a stem cell transplant: dental check and mouth care reasons: The reasons said bad teeth become a source of infection once the white count falls, and that mouth care routines lowered mucositis in trials mostly in adults. MASCC/ISOO says neither: it gives an expert opinion that a dental evaluation is desirable to reduce infection from the teeth while finding too little evidence that it prevents mucositis, and it suggests mouth care protocols on a small number of trials. (Second accuracy pass; MASCC/ISOO 2020 full text XML from Europe PMC, read raw) · Both reasons were rewritten to the guideline's own statements.
- 2026-11-17 · Preventing mouth sores (oral mucositis) in children having cancer treatment or a stem cell transplant: photobiomodulation in very young children: The page said photobiomodulation was workable in children of all ages. The panel's remarks add that giving it to very young children needs help from family members and may not always be successful. (Second accuracy pass; Children's Oncology Group endorsement PDF, read with pdftotext) · The caveat was added, and the patient summary no longer calls the treatment gentle or implies it is recommended for all head and neck radiotherapy.
- 2026-11-17 · Preventing mouth sores (oral mucositis) in children having cancer treatment or a stem cell transplant: MASCC/ISOO grading scheme: The grading scheme was described as levels of evidence I to V, with recommends for stronger and suggests for weaker evidence. The guideline text gives no such range or gloss: it assigns a level of evidence and classes each guideline as a recommendation, a suggestion or no guideline possible. (Second accuracy pass; MASCC/ISOO 2020 full text XML from Europe PMC, read raw) · The scheme description was rewritten from the guideline's methods section.
- 2026-11-17 · Oral care and oral mucositis prevention during cancer treatment: best in class: The page called MASCC/ISOO 2020 the only international guideline on mucositis prevention. ESMO also published a guideline on oral and gastrointestinal mucosal injury in 2015. (Second accuracy pass; Europe PMC and Crossref records read on 17 November 2026) · The claim was removed and the ESMO guideline is named as existing, older and not readable.
- 2026-11-17 · Oral care and oral mucositis prevention during cancer treatment: oral cryotherapy: The summary gave 30 minutes of cryotherapy for both bolus 5-fluorouracil and high-dose melphalan. MASCC/ISOO print 30 minutes only for bolus 5-fluorouracil and give no duration for melphalan. The reasons given (drug level in the blood, melphalan clearing quickly) are not in the guideline. (Second accuracy pass; MASCC/ISOO 2020 full text from Europe PMC, read as raw XML) · The two statements are now separate, the melphalan one says no duration is printed, and the reasons follow the guideline's own wording.
- 2026-11-17 · Oral care and oral mucositis prevention during cancer treatment: photobiomodulation settings: The page said the table of laser settings was an image and could not be read. It is text in the source. (Second accuracy pass; MASCC/ISOO 2020 full text from Europe PMC, Table 2 parsed cell by cell) · The five protocols are now given as a variant, with the guideline's note on a possible heating effect.
- 2026-11-17 · Oral care and oral mucositis prevention during cancer treatment: benzydamine: The step told readers to use benzydamine during head and neck radiotherapy without its limit. MASCC/ISOO recommend it for a moderate dose, printed as below 50 Gy, and only suggest it with chemoradiotherapy. (Second accuracy pass; MASCC/ISOO 2020 full text from Europe PMC) · The dose limit and the weaker grade with chemoradiotherapy are now in the step.
- 2026-11-17 · Oral care and oral mucositis prevention during cancer treatment: brushing and flossing in low counts: The steps said to brush at least twice a day, not to replace the brush daily, and not to start flossing during low counts. The 2024 statement says twice daily, that a brush does not need daily replacement, and that low counts are not the best time to start. (Second accuracy pass; MASCC/ISOO 2024 practice statement full text from Europe PMC) · The three steps now carry the statement's own strength of wording.
- 2026-11-17 · Oral care and oral mucositis prevention during cancer treatment: when to call: The page told patients to call if they cannot eat or drink, citing the NCI booklet, which says only to call if the mouth hurts a lot. (Second accuracy pass; NCI booklet PDF and NCI web page on mouth and throat problems, both read as raw text) · The step now cites the NCI web page, which says to call when pain makes it difficult to eat, drink or sleep or with a fever of 38 °C or higher.
- 2026-11-17 · Oral care and oral mucositis prevention during cancer treatment: palifermin and chlorhexidine variants: The palifermin variant said adults where the guideline says patients with blood cancers, and gave no dose or availability. The chlorhexidine summary read as a prohibition where the guideline makes a suggestion that leaves its use for infection open. (Second accuracy pass; MASCC/ISOO 2020 full text, the Kepivance label from DailyMed and the EMA product page) · Wording corrected. The label dose of 60 micrograms per kg per day for 3 days before and 3 days after is given with its source, and the withdrawal of the European Union authorisation on 1 April 2016 is stated.
- 2026-11-17 · Oral care and oral mucositis prevention during cancer treatment: reasons given for steps: Several reasons were not in the sources: that alcohol dries the mouth, that thrush is easier to treat early, that the laser reduces inflammation, that young children cannot keep ice in the mouth, and that the levels of evidence run from I to V. (Second accuracy pass; all sources read as raw text) · Each reason was reworded to what its source says or removed. The statements on glutamine, sucralfate, topical GM-CSF and lidocaine rinse, which the page had left out, were added with their grades.
- 2026-11-17 · Hearing protection for children receiving cisplatin: gap before the next cisplatin dose: The page gave a gap of at least 10 hours before the next cisplatin dose as the only rule. That is the US label. The European label for Pedmarqsi asks for a minimum 6-hour gap. (Second accuracy pass; PEDMARK US label (DailyMed XML) and Pedmarqsi summary of product characteristics (European Medicines Agency PDF), both read as raw text) · The step, the summary and a new variant now give both gaps and say which label prints which.
- 2026-11-17 · Hearing protection for children receiving cisplatin: sodium thiosulfate dose: The page said the dose was by weight band without giving it, and said the European label had not been read. The US label prints 10, 15 and 20 g/m² for under 5 kg, 5 to 10 kg and over 10 kg. The European label prints 6.4, 9.6 and 12.8 g/m² for the same bands because it counts the anhydrous salt. (Second accuracy pass; PEDMARK US label (DailyMed XML) and Pedmarqsi summary of product characteristics (European Medicines Agency PDF), both read as raw text) · A dose step was added with both sets of figures and a warning not to mix them. The European label was added as a source and the out-of-date gap removed.
- 2026-11-17 · Hearing protection for children receiving cisplatin: hearing tests during treatment: The first step told teams to plan hearing tests during treatment for every child, and said the loss starts in pitches a child does not notice. The source only says that surveillance is usually mandatory for at-risk patients during treatment, sets no schedule, and does not say the child does not notice. (Second accuracy pass; IGHG 2019 author manuscript, XML from the NCBI E-utilities API) · The step now says to test on the schedule the treatment protocol sets, and the reason is worded as the source words it.
- 2026-11-17 · Hearing protection for children receiving cisplatin: reasons given for two steps: Two reasons were attributed to the wrong source or went beyond it. The statement that damage may already be done after a longer cisplatin infusion comes from the US label, not the guideline. The label does not say the gap keeps the protector out of the blood; it says the timing is set to minimise interference with the antitumour activity of cisplatin. (Second accuracy pass; Freyer 2020 author manuscript (XML from the NCBI E-utilities API) and PEDMARK US label (DailyMed XML)) · Both reasons were reworded and attributed to the label.
- 2026-11-17 · Hearing protection for children receiving cisplatin: best in class statement: The page called the 2020 guideline the only guideline on preventing cisplatin hearing loss in children. No source says that, and a search found other papers on ototoxicity in childhood. (Second accuracy pass; Europe PMC title searches on 17 November 2026) · Reworded to say the guideline has not been replaced and that no later edition was found.
- 2026-11-17 · Dental assessment before bone-modifying agents and jaw osteonecrosis prevention: how common it is: The patient summary said these medicines "can rarely" damage the jaw. No source gives that word for cancer doses. AAOMS 2022 puts the risk in cancer below 5 per cent in most studies with a range up to 18 per cent, and the European denosumab leaflet calls it common, up to 1 in 10 people. (Second accuracy pass; AAOMS 2022 open PDF, EMA Xgeva product information and the SDCEP March 2024 supplement, read as raw text) · Reworded to "in a small number of people", with the figures and who gives them in the summary and the differences named in the gaps.
- 2026-11-17 · Dental assessment before bone-modifying agents and jaw osteonecrosis prevention: the AAOMS position on drug holidays: The page said, on the Korean statement's report, that AAOMS 2022 deliberately removed fixed drug holidays. The AAOMS paper itself says its working group could not reach a consensus and was evenly split, and it does print a timing for denosumab in osteoporosis: surgery 3 to 4 months after the last dose and restart 6 to 8 weeks after surgery. (Second accuracy pass; AAOMS 2022 position paper, read in full for the first time from the society's open PDF) · Rewrote the best-in-class statement from the AAOMS text, added AAOMS and the Scottish dental guidance as their own variants, and noted the discrepancy in the gaps.
- 2026-11-17 · Dental assessment before bone-modifying agents and jaw osteonecrosis prevention: timing of the dental check: The first step said to complete dental treatment before the first dose and cited the MASCC/ISOO/ASCO abstract, which says only that preventive care includes comprehensive dental assessments when a bone-modifying agent is started. (Second accuracy pass; Europe PMC abstract, SDCEP guidance PDF, AAOMS 2022 PDF and the product labels, read as raw text) · The step now says what the abstract says. The timing is a new step from the Scottish dental guidance with its printed grade (strong recommendation, low quality evidence), and a new AAOMS step covers waiting for extraction sites to heal, for which no source gives days or weeks.
- 2026-11-17 · Dental assessment before bone-modifying agents and jaw osteonecrosis prevention: warning signs: The list of symptoms to report (jaw pain, loose teeth, numbness, exposed bone) was cited to the Korean statement, which names no symptoms, and the reason given, that early signs are easier to treat, is in no source read. (Second accuracy pass; EMA Xgeva package leaflet and SDCEP guidance, read as raw text) · The step now quotes the European denosumab leaflet (contact the doctor and dentist immediately about loose teeth, pain or swelling, sores that do not heal or discharge) and names what the Scottish guidance adds.
- 2026-11-17 · Dental assessment before bone-modifying agents and jaw osteonecrosis prevention: discussing risk: A step on discussing modifiable risk factors cited NICE NG101, whose recommendation is about the benefits and risks of adjuvant bisphosphonates for women with early breast cancer and does not mention risk factors. (Second accuracy pass; NICE NG101 recommendation 1.12.3, HTML read as raw text) · The step now says what NICE says and for whom; the risk factors are attributed to the labels and the Korean statement.
- 2026-11-17 · Dental assessment before bone-modifying agents and jaw osteonecrosis prevention: pausing and surgery: Two reasons went beyond the sources: that pausing the medicine "may help the jaw heal", and that surgery for established osteonecrosis "is for cases that do not settle". The sources call the benefit of a pause inconclusive or unsubstantiated, and AAOMS 2022 and the Korean statement both treat surgery as an option at every stage. (Second accuracy pass; AAOMS 2022 PDF and the Korean 2025 full text from Europe PMC) · Both reasons reworded to the source text; the surgery variant now carries the AAOMS position and the US label's caution.
- 2026-11-17 · Dental assessment before bone-modifying agents and jaw osteonecrosis prevention: which guidelines exist: The page called MASCC/ISOO/ASCO 2019 the only oncology-specific international guideline. ESMO published a bone health in cancer guideline in 2020, which could not be read, so "only" cannot stand. (Second accuracy pass; Crossref record for 10.1016/j.annonc.2020.07.019; publisher refused a script) · Reworded, and the ESMO guideline added as a source that could not be reached, with a gap saying so.
- 2026-11-17 · Anticancer treatment prescribing and pharmacy verification before each cycle: when the pharmacy check happens: The page said the prescription is checked before any drug is made up or prepared. The BOPA standard says preparation may happen in advance, and that the treatment must not be given or issued to the patient until verification is complete. (Second accuracy pass; BOPA Standards for the Pharmacy Verification of Prescriptions for Cancer Medicines version 5.0, PDF read as raw text) · The summary for patients, the step and the best practice statement now say the check comes before the treatment is given or issued, and that preparation may start earlier.
- 2026-11-17 · Anticancer treatment prescribing and pharmacy verification before each cycle: reasons given for the checks: Several reasons named things the BOPA standard does not print: a unit error a thousand times too large, damage to the heart or lungs, herbal products, a heart scan before an anthracycline and a DPD test before fluorouracil, and electronic systems calculating doses. (Second accuracy pass; BOPA version 5.0, PDF read as raw text) · Each reason was reworded to what the standard says. The page now states that BOPA prints no cumulative dose limits, no number of days for blood results and no percentage for weight change, and leaves these to the treatment protocol and local policy.
- 2026-11-17 · Anticancer treatment prescribing and pharmacy verification before each cycle: paper prescriptions: The page said both standards accept electronic or preprinted order sets. BOPA requires an electronic prescribing system and allows non-electronic prescribing only in exceptional circumstances defined by local policy, such as system downtime. (Second accuracy pass; BOPA version 5.0, PDF read as raw text) · BOPA's rule is now stated on its own, with the exception.
- 2026-11-17 · Anticancer treatment prescribing and pharmacy verification before each cycle: pharmacy technician verification and capping: The page said technicians verify flat-dosed or well-tolerated drugs, and that body surface area is capped usually in people with obesity. BOPA says simple prescriptions agreed locally, from cycle 2, such as flat-dosed treatment that is usually well tolerated, and does not mention obesity in its note on capping. (Second accuracy pass; BOPA version 5.0, PDF read as raw text) · The variant now gives BOPA's own examples and its minimum requirements for a technician, worded as guidance; the obesity clause was removed.
- 2026-11-17 · Anticancer treatment prescribing and pharmacy verification before each cycle: weight checks in children: The page said both standards leave how often weight is rechecked to local policy, and listed BOPA's paediatric tables as 2, 3 and 5. BOPA asks for a child's weight before every cycle, and tables 2 to 6 all carry paediatric checks. (Second accuracy pass; BOPA version 5.0, PDF read as raw text) · The gap is now scoped to adults with the rule for children stated, and the tables are given as 2 to 6.
- 2026-11-17 · Anticancer treatment prescribing and pharmacy verification before each cycle: what the US standard asks of an order: The page said the order must state the dose calculation, told prescribers to use only a standard regimen, and asked for a second prescriber to authorise an exception. The ASCO-ONS standards put the dose calculation in the patient's medical record, provide for exception orders outside the standard regimen, and ask for a supporting reference, authorisation by a second licensed practitioner, or both. (Second accuracy pass; ASCO-ONS 2024 standards, full text read from the Oncology Nursing Society's open PDF) · The three statements were reworded to the standards' own terms in the steps, the summary and the best practice statement.
- 2026-11-17 · Anticancer treatment prescribing and pharmacy verification before each cycle: reasons given for four rules: Four steps gave reasons the ASCO-ONS standards do not print, for example that a spoken order leaves no record and that a trailing zero makes a decimal point easy to miss. The text says spoken orders offer more room for error than written ones and gives no reason for the rule on zeros. (Second accuracy pass; ASCO-ONS 2024 standards, full text read from the Oncology Nursing Society's open PDF) · Each reason now says what the standards or their glossary print, and the gaps note that no reason is given for the rule on abbreviations and zeros.
- 2026-11-17 · Bone health on aromatase inhibitors, ovarian suppression and androgen deprivation therapy: jaw and dental step: The step told teams to discuss jaw osteonecrosis and arrange a dental check before a bisphosphonate or denosumab, and said the drugs rarely cause it, resting on a NICE sentence that covers only adjuvant bisphosphonates in women and mentions neither a dental check, denosumab nor a frequency. (Second accuracy pass; NICE NG101 recommendations page and the Prolia US label from DailyMed, read as raw text) · Split into a NICE step on discussing the risks of bisphosphonates and a label-sourced step on a mouth examination before denosumab; the word rarely was removed.
- 2026-11-17 · Bone health on aromatase inhibitors, ovarian suppression and androgen deprivation therapy: treatment thresholds: The page said the UK expert group statement that NICE points to could not be found and gave no bone density threshold, no drug doses, no calcium or vitamin D figures and no repeat scan interval. NICE links to the statement, which treats at a T-score below -2 and below -1 on ovarian suppression with an aromatase inhibitor. (Second accuracy pass; the 2008 UK expert group PDF linked from NICE NG101, read with pdftotext) · Thresholds, bisphosphonate doses, calcium and vitamin D figures and the 24 month repeat scan were added, each attributed to its source.
- 2026-11-17 · Bone health on aromatase inhibitors, ovarian suppression and androgen deprivation therapy: NICE osteoporosis guidance: The page referred to NICE's osteoporosis guidance without reading it. NICE replaced that guideline on 29 July 2026 with NG259, which suggests considering treatment at a T-score of -1.5 or less in people taking an aromatase inhibitor or androgen deprivation therapy. (Second accuracy pass; NICE NG259 chapters, read as raw text from nice.org.uk) · NG259 was added as a source with a step on its treatment criteria and its rule on repeat scans.
- 2026-11-17 · Bone health on aromatase inhibitors, ovarian suppression and androgen deprivation therapy: denosumab dose and stopping: The page said to use denosumab at the osteoporosis dose without giving it and did not mention that fracture risk rises after denosumab is stopped. The label gives 60 mg under the skin every 6 months and says patients should move to another antiresorptive drug if it is stopped. (Second accuracy pass; Prolia US prescribing information from DailyMed, label dated 26 June 2026, read as raw text) · Two steps were added, on the dose and on not stopping without a follow-on drug.
- 2026-11-17 · Bone health on aromatase inhibitors, ovarian suppression and androgen deprivation therapy: reasons given for steps: Several reasons said more than the cited source: a list of risk factors and an instruction to advise everyone on lifestyle were attributed to an ASCO abstract that says only that bone health may benefit, ASCO was said to test bone density only at substantial risk, and bisphosphonates were said to prevent fractures in men on hormone therapy, which NICE does not say. (Second accuracy pass; ASCO 2019 abstract from Europe PMC and NICE NG131, read as raw text) · Each reason was reworded to what its source says, and the lifestyle step now rests on the 2025 joint position statement.
- 2026-11-17 · Bone health on aromatase inhibitors, ovarian suppression and androgen deprivation therapy: men on androgen deprivation therapy: The page gave only NICE's advice to consider a fracture risk assessment for men on androgen deprivation therapy. The UK National Osteoporosis Guideline Group makes a FRAX assessment for all such men a strong recommendation, and the NICE quote carried a stray space that is not in the source. (Second accuracy pass; NOGG 2024 guideline full text from Europe PMC and NICE NG131, read as raw text) · The NOGG recommendation was added with its grade beside the NICE one, and the quote was corrected.
- 2026-11-17 · Vaccination before, during and after cancer treatment: plain-language summary: The page said flu and other injected vaccines are usually safe during treatment. MMR and chickenpox vaccines are injected and live, and ASCO lists them as contraindicated with cancer treatment; what ASCO calls safe is non-live vaccines. (Second accuracy pass; ASCO 2024 full text from PubMed Central, read raw) · Reworded to say that vaccines containing no live germs can usually be given, and that live vaccines such as MMR and chickenpox cannot.
- 2026-11-17 · Vaccination before, during and after cancer treatment: how long to wait for live vaccines after chemotherapy: The page gave 6 months as the only interval. That is the Green Book and Indian Academy of Pediatrics figure. ASCO, for solid tumours, says live vaccines may be considered at least 3 months after treatment is completed, and ECIL 7 says usually at least 3 months after chemotherapy or 6 months from rituximab (graded D III). (Second accuracy pass; ASCO 2024 Table 4 and the ECIL 7 public slide set, read raw) · The summary, best-in-class statement and a new variant now state each body's interval.
- 2026-11-17 · Vaccination before, during and after cancer treatment: live vaccines for household contacts: The page said the vaccine viruses in MMR, varicella and rotavirus vaccines are not known to harm close contacts. The Green Book says this of MMR and rotavirus, but says varicella vaccine virus may rarely pass on from a vaccinated person who develops a rash, and advises limiting contact until the rash is dry and crusted. (Second accuracy pass; Green Book chapter 6 PDF, pdftotext) · The reason was reworded and the rash precaution added to the step.
- 2026-11-17 · Vaccination before, during and after cancer treatment: school and classmates: The page said vaccinated classmates do not pass vaccine viruses on in a way that harms the child. The Green Book says that for most live vaccines in the UK schedule transmission does not occur or can be minimised by simple precautions. (Second accuracy pass; Green Book chapter 6 PDF, pdftotext) · Reworded to the source's statement.
- 2026-11-17 · Vaccination before, during and after cancer treatment: live vaccines after rituximab: The reason given, that B-cell depleting antibodies weaken immunity for longer than chemotherapy, is not in the Green Book chapter cited, which gives the 12-month interval without a reason. (Second accuracy pass; Green Book chapter 6 PDF and ASCO 2024 full text, read raw) · The reason now says only that the Green Book sets a longer interval and that ASCO reports vaccine responses can stay absent for up to a year after B-cell targeted treatment.
- 2026-11-17 · Vaccination before, during and after cancer treatment: who must not have live vaccines: The page said live vaccines are contraindicated in anyone with leukaemia or lymphoma. The Green Book says most live vaccines should not be given to people with immunosuppression due to acute and chronic leukaemias and lymphoma, and the step left out that the bar applies during chemotherapy as well as for 6 months after. (Second accuracy pass; Green Book chapter 6 PDF, pdftotext) · Step and summary reworded to the source.
- 2026-11-17 · Vaccination before, during and after cancer treatment: COVID-19 vaccine after B-cell depleting therapy: The summary said revaccination for COVID-19 is at 6 months. ASCO says no sooner than 6 months after completion of treatment. (Second accuracy pass; ASCO 2024 Table 1, read raw) · Reworded to no sooner than 6 months.
- 2026-11-17 · Vaccination before, during and after cancer treatment: best-supported version and newer guidance: The page called ASCO 2024 the most recent graded guideline and said no graded international guideline for children could be read. ECIL 7 grades recommendations for children with acute lymphoblastic leukaemia in a public slide set, ECIL 11 (September 2026) and IDSA (published 2026) are later for parts of the subject, and the CDC advice behind ASCO's vaccine table has changed: a 21-valent pneumococcal vaccine was added in June 2024, pneumococcal vaccine was recommended from age 50 in October 2024, and RSV vaccine moved to all adults aged 75 and over and adults aged 60 to 74 at increased risk. (Second accuracy pass; ECIL slide sets (pdftotext), ACIP statements in MMWR (Europe PMC full text or abstract), IDSA 2025 abstracts) · Best-in-class statement reworded; variants added for the US schedule changes, ECIL 7, ECIL 11 and the UK shingles chapter; sources and gaps updated.
- 2026-11-17 · Anticancer treatment administration checks at the chair: tracing the lines: The page said tracing the lines finds clamps left closed. The ASCO-ONS standards say it reduces errors related to unclamped tubing, which is the opposite fault, and also checks that lines run at the prescribed rate. (Second accuracy pass; ASCO-ONS 2024 standards, read in full as raw text from the Oncology Nursing Society co-publication PDF) · The reason now lists what the standards list: tubing left unclamped, loose or disconnected lines, a drug on the wrong pump, the wrong tubing or filter, and a line not running at the prescribed rate.
- 2026-11-17 · Anticancer treatment administration checks at the chair: what patients are told to report: The patient summaries listed itching, breathlessness and flushing and said to report them at once, and named temperature, pulse and blood pressure. The standards name only hypersensitivity symptoms or pain at the infusion site as examples, give no time limit, and say vital signs without listing them. (Second accuracy pass; ASCO-ONS 2024 standards, read in full as raw text from the Oncology Nursing Society co-publication PDF) · The summaries now give the source's examples only, and the page says that the standards give no fuller symptom list or time limit.
- 2026-11-17 · Anticancer treatment administration checks at the chair: reasons given for steps: Several reasons stated things the sources do not: that most preparation errors are a wrong vial strength, volume or diluent, that reactions and leaks need treatment within minutes, and that the patient is often first to notice a reaction. (Second accuracy pass; ASCO-ONS 2024 standards and BOPA version 5.0, both read in full as raw text) · Each reason was rewritten to what the standards and their glossary say.
- 2026-11-17 · Anticancer treatment administration checks at the chair: who checks independently: The page said the last two verifications are each made by two licensed clinicians independently. The standards print the word independently for the third verification only, and say at least two clinicians for both. (Second accuracy pass; ASCO-ONS 2024 standards, read in full as raw text from the Oncology Nursing Society co-publication PDF) · The best-in-class statement and the steps now follow the wording of each verification, and a step was added for who may give the drug.
- 2026-11-17 · Anticancer treatment administration checks at the chair: the 2016 standards: The page said the 2016 standards had no separate check at the patient's side. The 2016 standards were not read, and the 2024 text says only that earlier versions required three independent verifications. (Second accuracy pass; ASCO-ONS 2024 standards, read in full as raw text from the Oncology Nursing Society co-publication PDF) · The variant now reports what the 2024 text says about earlier versions and states that the 2016 standards were not read.
- 2026-11-17 · Anticancer treatment administration checks at the chair: evidence gap: The page said no trial compares two-person checks with one-person checks, barcode scanning or smart pumps. No source read says that; the panel says there is a paucity of randomised trials and limited controlled data on technology. (Second accuracy pass; ASCO-ONS 2024 standards, read in full as raw text from the Oncology Nursing Society co-publication PDF) · The gap now uses the panel's own description.
- 2026-11-17 · Oral anticancer medicines at home: who the standards cover: The page said ASCO-ONS covers hormonal therapies where they are listed as antineoplastic, and that children's oral chemotherapy lay outside the sources. The standards say hormonal therapies are not included in their definition of antineoplastic agents, and they cover adult and paediatric oncology. (Second accuracy pass; ASCO-ONS 2024 standards, full text of the Oncology Nursing Forum co-publication read as raw text) · Population and summary corrected to the standards' own definition and scope.
- 2026-11-17 · Oral anticancer medicines at home: adherence checks: The page said to ask about missed doses at every contact, to offer coaching or motivational interviewing, and that these had helped in small trials. The ONS guideline sets no interval for ongoing assessment (its studies ranged from weekly to months apart), only suggests these interventions as conditional recommendations, and says motivational interviewing may improve adherence on low certainty evidence. (Second accuracy pass; ONS adherence guideline 2022, full text through NCBI E-utilities) · Step reworded to ongoing assessment with no fixed interval; coaching and motivational interviewing changed to consider; reasons reworded to the guideline.
- 2026-11-17 · Oral anticancer medicines at home: leftover medicine: The page told patients to return leftover tablets to the pharmacy. The ONS toolkit says never to bin, pour away or flush cancer drugs, and to ask the healthcare team whether unused medicine can be returned to the doctor's office, pharmacy or treatment centre. (Second accuracy pass; ONS safe handling toolkit 2018, PDF read as raw text) · Step, summary and patient text now say to ask the team where leftovers can be returned.
- 2026-11-17 · Oral anticancer medicines at home: crushing tablets and original containers: A rule not to crush, split or open any tablet was cited to the capecitabine product information, which speaks only of capecitabine, with a reason about drug dust that it does not give. Keeping tablets in the original container was credited to ACS and Macmillan, but ACS says only that some chemotherapy has to be kept in its bottle or box and Macmillan does not say it. (Second accuracy pass; Xeloda EU product information, ACS and Macmillan pages, and the MASCC Oral Agent Teaching Tool, all read as raw text) · The general rules are now cited to the MASCC teaching tool, which states them for all oral cancer medicines, and the capecitabine text is given as an example with the reactions it reports.
- 2026-11-17 · Oral anticancer medicines at home: missed or vomited doses: The step said to tell the team about a missed dose rather than making it up, wording the ACS page does not use, and said nothing about being sick after a dose. Macmillan says to contact the hospital nurse or doctor straight away if a dose is forgotten, if you are sick after taking one, or if you cannot take the medicine. (Second accuracy pass; Macmillan and ACS pages read as raw text) · Step re-based on the Macmillan sentence, quoted, with the ACS advice kept in the reason.
- 2026-11-17 · Oral anticancer medicines at home: body-fluid precautions: Patient text gave two days where ACS says at least 48 hours, a reason said contact can harm skin where ACS says it is not likely to cause harm but that being exposed often may lead to problems, and the page did not say that ACS describes these precautions for traditional chemotherapy. A gap also said no source settles gloves for carers, though the ONS toolkit says a no-touch technique or gloves. (Second accuracy pass; ACS Chemotherapy Safety page and ONS toolkit read as raw text) · Wording changed to at least two days, reasons reworded to ACS, scope stated in population and gaps, and the gap about gloves removed.
- 2026-11-17 · Oral anticancer medicines at home: reasons given for steps: Nine reasons stated things no cited source says, among them that automatic repeat prescriptions let people take a drug too long, that cycles with days off are a common cause of dosing errors, and a list of stop-and-call symptoms with no source. (Second accuracy pass; each cited source read as raw text, with the capecitabine package leaflet and the BOPA 2004 statement added) · Each reason reworded to what a source read states; the stop-and-call symptoms are now the capecitabine leaflet's own list, given as an example.
- 2026-11-17 · Oral anticancer medicines at home: UK position on who prescribes and dispenses: The page said no UK standard had been read and gave only the US position. BOPA's 2004 statement says prescribing and dispensing should remain with the hospital specialist and hospital pharmacy other than in agreed exceptions, and the MHRA's report of the 2008 NPSA alert says risks are increased when non-specialists prescribe, dispense or administer. (Second accuracy pass; BOPA position statement in the copy NICE publishes, and the MHRA Drug Safety Update page, read as raw text) · A step, a variant, two grades entries and a summary paragraph added; the NPSA report itself could not be read and the gap says so.
- 2026-11-17 · Oral anticancer medicines at home: source dates: The NCI booklet was dated 2023 and the capecitabine product information 2026. The booklet prints January 2024 and the EMA file is dated February 2025. (Second accuracy pass; the two PDFs read as raw text) · Years corrected to 2024 and 2025.
- 2026-11-17 · Eye care with antibody-drug conjugates that harm the cornea: what the patient summary said about recovery: The page told patients that most eye problems improve. The US belantamab label reports that 42 per cent of people with grade 2 to 4 findings had improved to grade 1 or better at their last event, and 22 per cent had returned to normal. (Second accuracy pass; BLENREP US label section 5.1, DailyMed SPL XML read raw) · The sentence was removed and the belantamab step and the gaps now give each label's own figures.
- 2026-11-17 · Eye care with antibody-drug conjugates that harm the cornea: European examination schedules and steroid drops: The page gave only the US schedules and said the European texts had not been read. The EU texts differ: belantamab examinations before each of the first 4 doses and then as indicated, mirvetuximab steroid drops only after a corneal reaction of grade 2 or higher, and tisotumab examination by an eye specialist before the first infusion and as indicated. (Second accuracy pass; EMA product information PDFs for Blenrep, Elahere, Tivdak and Padcev, read with pdftotext) · A European variant was added, and the summary and best in class now state both regulators' positions.
- 2026-11-17 · Eye care with antibody-drug conjugates that harm the cornea: belantamab grade 4 rule: The page gave the rule for grade 2 and 3 findings and said only that the drug stops after a recurrent grade 4 finding. The US label also says to consider permanent discontinuation at a first grade 4 finding, and gives the dose for restarting. (Second accuracy pass; BLENREP US label Table 2, DailyMed SPL XML read raw) · A grade 4 step was added with the label's sentence quoted.
- 2026-11-17 · Eye care with antibody-drug conjugates that harm the cornea: reasons given for the eye drops and cold packs: Several reasons were not in any label: that steroid drops can mask infection or raise eye pressure, that a gap stops the lubricant washing the steroid out, that vasoconstrictor drops and cold packs narrow vessels so less drug reaches the eye, and that contact lenses worsen corneal damage. (Second accuracy pass; ELAHERE, TIVDAK and BLENREP US labels, DailyMed SPL XML read raw) · Each reason now says only what the label says, and the gaps note that the labels do not explain how the measures work.
- 2026-11-17 · Eye care with antibody-drug conjugates that harm the cornea: enfortumab compared with the other drugs: The page said dry eye with enfortumab is usually milder than with the other three drugs and affects about a third. No label makes that comparison; the PADCEV label reports dry eye symptoms in 30 per cent of 384 people. (Second accuracy pass; PADCEV US label section 5.5, DailyMed SPL XML read raw) · The comparison was removed and the figure and its denominator stated.
- 2026-11-17 · Eye care with antibody-drug conjugates that harm the cornea: mirvetuximab dose rules and timing: The page gave no dose rules for mirvetuximab and said eye problems usually start in the first weeks. The label gives hold, reduce and stop rules by corneal finding, and a median onset of 5.1 weeks with a range of 0.1 to 68.6 weeks. (Second accuracy pass; ELAHERE US label sections 2.4 and 5.1, DailyMed SPL XML read raw) · A dose rule step was added and the timing now gives the median and range.
- 2026-11-17 · Eye care with antibody-drug conjugates that harm the cornea: a quotation from the belantamab dose table: The quoted sentence had a stray space before a full stop, so it was not character for character the label's text. (Second accuracy pass; BLENREP US label Table 2, checked by script) · The quotation was corrected to the label's text.
- 2026-11-17 · Safe handling of hazardous anticancer drugs where they are given: medical surveillance: The page set ASCO (routine surveillance unsupported) against HSE only, and treated HSE's monitoring of exposure as if it were medical surveillance. The NIOSH Alert says to make medical surveillance an important part of any safe handling programme, OSHA says exposed workers should be monitored in a medical surveillance programme, and ONS and the European Commission guidance recommend it too. HSE's monitoring is the testing of controls and exposure. (Second accuracy pass; NIOSH Alert 2004 (PDF from the UNT government web archive), OSHA page, ONS toolkit, European Commission guidance and HSE page, all read as raw text) · Summary, best in class and the surveillance variant now say who recommends surveillance and who calls its role unclear, with HSE's wording stated separately.
- 2026-11-17 · Safe handling of hazardous anticancer drugs where they are given: body fluids: The short summary said some drug is in body fluids for a day or two. NIOSH, OSHA and ONS say 48 hours and for some drugs 7 days, the American Cancer Society says at least 48 hours, and the European Commission guidance says generally 7 to 14 days. (Second accuracy pass; the same sources plus the ACS page, read as raw text) · The short summary now says at least two days after a dose and longer for some drugs, and the European figure is on the page.
- 2026-11-17 · Safe handling of hazardous anticancer drugs where they are given: closed-system devices: The page said USP chapter 800 requires closed-system transfer devices for administration without the condition that goes with it. ONS reports the requirement as applying when the dosage form allows, and OSHA words the same point as should be used. (Second accuracy pass; ONS toolkit (2018) PDF and OSHA page read as raw text. USP chapter 800 itself could not be read) · The condition is added to the step, summary and variant, and the page says the chapter was not read.
- 2026-11-17 · Safe handling of hazardous anticancer drugs where they are given: European guidance: The page said no European Union standard could be read. The European Commission's guidance on hazardous medicinal products at work (2023) is open, and differs from US practice: single gloves are normally sufficient for administration and excreta are treated as contaminated generally for 7 to 14 days. (Second accuracy pass; European Commission guidance PDF from osha.europa.eu read with pdftotext) · Added as a source, a variant and two quoted grade rows.
- 2026-11-17 · Safe handling of hazardous anticancer drugs where they are given: reasons and spill respirators: Several reasons given for steps said more than the sources: that skin contact is the main route for nurses, that priming sprays drug, that harm came to staff working without protection, that drugs harm in tiny amounts, and that a respirator is for spills that make airborne particles. (Second accuracy pass; NIOSH Alert 2004 and ONS toolkit read as raw text) · Each reason is reworded to what NIOSH, ONS or HSE states, and the spill steps now follow the ONS spill procedure.
- 2026-11-17 · Safe handling of hazardous anticancer drugs where they are given: patient reassurance: The short summaries told patients they are not dangerous to be near and are safe to touch or hug, with no source. (Second accuracy pass; American Cancer Society page Chemotherapy Safety (revised 15 May 2025) read as raw text) · Reworded to the ACS statement that patients can usually be around family and friends, and the page is cited.
- 2026-11-17 · Aseptic compounding of hazardous anticancer drugs in the pharmacy: plain-language summary: The page said the drugs are made up in a sealed cabinet and that staff use sealed transfer devices. NIOSH describes a Class II cabinet as open-fronted, and says only to consider closed-system transfer devices. (Second accuracy pass; NIOSH Alert 2004, full PDF text read from a government web archive copy) · Now says a ventilated cabinet or an isolator, and the transfer-device claim is removed from the summary for patients.
- 2026-11-17 · Aseptic compounding of hazardous anticancer drugs in the pharmacy: summary: The summary said compounding carries the highest exposure risk of any handling task. No source read says so. (Second accuracy pass; NIOSH Alert 2004 and the HSE page, read as raw text) · Sentence removed.
- 2026-11-17 · Aseptic compounding of hazardous anticancer drugs in the pharmacy: what ASCO says: The page said ASCO endorses the 2004 NIOSH Alert, found no long-term clinical evidence for external ventilation, and rated the evidence for closed-system devices as low quality. The ASCO full text could not be re-read; its abstract names OSHA, USP chapter 800 and ONS, not NIOSH. (Second accuracy pass; ASCO 2019 abstract through Europe PMC, publisher blocked) · Those statements were taken off the page and listed under gaps; ASCO is cited only for what its abstract says.
- 2026-11-17 · Aseptic compounding of hazardous anticancer drugs in the pharmacy: closed-system transfer devices: The page said USP chapter 800 requires closed-system devices for giving drugs and recommends them for compounding, citing ASCO. The chapter is behind a login, and the two readable reports differ: OSHA says should be used for both, ONS says recommended for compounding and required for giving. (Second accuracy pass; OSHA hazardous drugs page and ONS toolkit 2018, read as raw text) · Both reports are now given with their wording, and the chapter is marked as not read.
- 2026-11-17 · Aseptic compounding of hazardous anticancer drugs in the pharmacy: identifying hazardous drugs: The step quoted the 2024 NIOSH list, which could not be read again to confirm the sentence. (Second accuracy pass; cdc.gov refused the request and the Internet Archive was unavailable) · The step now quotes the 2004 NIOSH Alert, which says each organisation should create its own list, and the 2024 list is marked as not reachable.
- 2026-11-17 · Aseptic compounding of hazardous anticancer drugs in the pharmacy: employer duty: The page said employers must keep exposure as low as possible and that monitoring applies where carcinogens are handled. HSE says exposure should be prevented or adequately controlled, and that monitoring is normally necessary where there is potential for exposure to carcinogens. (Second accuracy pass; HSE page on cytotoxic drugs, read as raw text) · Reworded to HSE's terms.
- 2026-11-17 · Aseptic compounding of hazardous anticancer drugs in the pharmacy: reasons given for steps: Several reasons went beyond the source: that gloves become more permeable with time, that tested gloves and gowns block the drugs, that priming sprays droplets, and that surface residue is a main route of exposure. NIOSH states none of these. (Second accuracy pass; NIOSH Alert 2004, full PDF text) · Each reason now says only what NIOSH, HSE or OSHA states, and gloves are described as labelled chemotherapy gloves as NIOSH words it.
- 2026-11-17 · Aseptic compounding of hazardous anticancer drugs in the pharmacy: UK position: The page gave no UK standard for the cabinet and said the NHS aseptic standards had not been read. They and the joint HSE and MHRA guidance are open and differ from the United States position: either a positive or a negative pressure isolator may be used after a documented risk assessment. (Second accuracy pass; HSE and MHRA MS37 (2015), QAAPS Part A fifth edition (2016) and EU GMP Annex 1 (2022), full PDF texts) · Added UK steps on choosing an isolator, residue monitoring at least annually and a one-week expiry limit, and a variant setting the UK and United States positions side by side.
- 2026-11-17 · Drug, herbal and food interaction review for anticancer treatment: reconciliation before early-phase trials: The page said reconciliation in 501 trial patients found 346 interactions, more than half of the prohibited medicines being herbal. The full text reports 501 reconciliations (some patients had more than one) and 360 interactions in 189 patients, involving 346 treatments or products; 62.7% of the prohibited ones were self-medication or alternative medicine products. The first figures came from the abstract, which differs from the paper. (Second accuracy pass; Malifarge 2024 full text read as XML from Europe PMC (PMC11005900)) · Counts corrected to the full text, the abstract's differing figures attributed to the abstract, and the timing changed from before enrolment to before the first dose of trial treatment.
- 2026-11-17 · Drug, herbal and food interaction review for anticancer treatment: herb study figures: The page said 4% had a potentially relevant herb-drug interaction, half judged clinically relevant, without saying half of what. The paper found 13 potentially relevant interactions in 8 of 202 patients (4%), and 6 of those interactions, in 4 patients, were judged clinically relevant. (Second accuracy pass; Brokaar 2025 full text read as XML from Europe PMC (PMC11982495)) · Counts and denominators given in the summary and the variant; source now marked as read in full.
- 2026-11-17 · Drug, herbal and food interaction review for anticancer treatment: St John's wort: The statement that St John's wort lowers irinotecan's active metabolite and imatinib rested on another OnCo record, not on a source read for this page. (Second accuracy pass; current US labels for Camptosar and Gleevec read as XML from DailyMed) · Both labels added as sources and the wording matched to them: exposure to irinotecan or SN-38 substantially reduced, and a 30% fall in imatinib exposure. A step on irinotecan and strong CYP3A4 inducers was added with the label's sentence.
- 2026-11-17 · Drug, herbal and food interaction review for anticancer treatment: Cancer Drug Interactions checker: The page said the checker has been closed since 2023. The original site was taken offline in September 2023, but an archived checker last updated in June 2022 still answered on 17 November 2026, although its home page says it would close on 31 January 2025. (Second accuracy pass; cancer-druginteractions.org home page and checker page read as HTML) · Summary and gap reworded to say what the site states and what answered today.
- 2026-11-17 · Drug, herbal and food interaction review for anticancer treatment: unsupported reasons and generalisations: Several lines said more than their sources: that people often do not count herbal products as medicines (the standards say some patients, at times), that pharmacokinetic interactions matter most and oral targeted drugs carry most of them, that general checkers over-flag and miss interactions, that inducers can make treatment fail, that new prescriptions from other doctors are a common source of interactions, and that the check is repeated on admission, transfer or discharge. Two label instructions for single drugs were written as rules for whole drug classes. (Second accuracy pass; ASCO-ONS 2024 (ONS open PDF), BOPA version 5.0 PDF, the van Leeuwen 2022 abstract and the ibrutinib and erlotinib labels, all read as raw text) · Each line reworded to what its source states or removed; the label steps now name ibrutinib and erlotinib as the examples, and the summary gives the erlotinib label's timing (10 hours after an H2 blocker and at least 2 hours before its next dose).
- 2026-11-17 · Drug, herbal and food interaction review for anticancer treatment: who checks: The page called the BOPA check the pharmacist's verification and left out its qualifier for later cycles. BOPA calls it pharmacy verification, which pharmacy technicians may do at some levels, and says the later-cycle check is made when the information is accessible. (Second accuracy pass; BOPA version 5.0 PDF read with pdftotext) · Summary, best-in-class statement and step reworded.
- 2026-11-17 · Central line care for children with cancer: fever at home: The page told families to ring the treatment centre at once for a fever and cited a CDC sentence that is about reporting changes at the line site and says nothing about fever. CCLG's community resource states the advice: parents and carers are advised to telephone the unit if the child's temperature is 38°C or above on one occasion, even if the child seems well, between 37.5°C and 37.9°C with other concerns, or if the child seems unwell regardless of fever. No source read sets a time limit for the call. (Second accuracy pass; CCLG community care resource, febrile neutropenia page (last updated June 2025), read as raw HTML) · The fever advice is now its own step citing the CCLG page with its temperatures, the short summaries match it, and a gap says no source read sets a time within which to ring.
- 2026-11-17 · Central line care for children with cancer: chlorhexidine dressings under 18: The page said CDC makes no recommendation on chlorhexidine dressings for anyone under 18. CDC's July 2017 statement is limited to short-term, non-tunnelled central lines, which are not the tunnelled lines and ports this page is about. (Second accuracy pass; CDC/HICPAC 2011 guideline (last update October 2017), PDF read with pdftotext) · The summary, best in class, variant and grade row now give the scope, with CDC's sentence quoted, and a row was added for its Category IC advice against these dressings in premature neonates.
- 2026-11-17 · Central line care for children with cancer: the children's hospital bundle: The page said the list of what the Children's Hospital Association bundle asks for could not be read, including any mask or shielding of the child's face. The two papers now read in full. The bundle asks for sterile gloves and a mask for the person changing a dressing or port needle, the child's face shielded from the site, the cap scrubbed with alcohol for 15 seconds and dried for 15 or with chlorhexidine for 30 seconds and dried for 30 to 60, and port needles and clear dressings changed every 7 days. (Second accuracy pass; Rinke 2012 and Rinke 2013 in Pediatrics, full text from PubMed Central read as raw HTML) · The elements are now given in the summary and best in class, a variant covers the mask and face shielding and says they are expert consensus, and the port needle variant shows the bundle's 7 days beside CCLG's two weeks.
- 2026-11-17 · Central line care for children with cancer: flushing: The flushing step said 0.9% sodium chloride and cited CCLG. CCLG says the line requires flushing with sodium chloride and gives no strength or volume. (Second accuracy pass; CCLG community care resource, central lines page (last updated January 2026), read as raw HTML) · The step and summary now say sodium chloride, and a gap says no guideline read sets a flush volume or a heparin concentration for children. The heparin concentrations a Cochrane review calls typical and those one trial used are given in the heparin variant.
- 2026-11-17 · Central line care for children with cancer: baths and showers: The page said to keep a tunnelled line out of baths. CDC says not to submerge the catheter or its site and allows showering with an impermeable cover. CCLG says children with a tunnelled line cannot swim and must bathe with care, with showers more advisable. (Second accuracy pass; CDC/HICPAC 2011 PDF and the CCLG central lines page, both read as raw text) · The step now says the line and its site should not go under water and gives CCLG's wording on bathing.
- 2026-11-17 · Central line care for children with cancer: scrubbing the hub: The step told staff to let the hub dry and cited the SHEA passage, which sets at least 5 seconds of friction and does not mention drying. Drying comes from epic3 (at least 15 seconds, then allowed to dry) and from the children's hospital bundle. (Second accuracy pass; SHEA 2022 and epic3 full texts from Europe PMC, and Rinke 2012 from PubMed Central) · The step now gives each source's time and says which ones ask for drying.
- 2026-11-17 · Central line care for children with cancer: who antimicrobial locks are for: The page said SHEA lists antimicrobial locks for patients with long-term lines and a history of repeated infection. SHEA names three groups: patients with long-term dialysis lines and recurrent infection, patients with limited venous access and recurrent infection, and patients at heightened risk of severe harm from a line infection. (Second accuracy pass; SHEA 2022 full text from Europe PMC) · Best in class now lists the groups as SHEA does, and adds CDC's Category II suggestion for long-term lines with several infections despite the best aseptic care.
- 2026-11-17 · Central line care for children with cancer: reasons given for thirteen steps: Thirteen reasons said more than the sources: that most infections and breakages come from handling, that most line days are spent at home, that hands carry germs more often than anything else, that trouble is easier to treat when found early, that an ordinary needle would damage a port, that many breaks can be repaired, and others like them. None of these sentences is in the source cited. (Second accuracy pass; CCLG pages, CDC/HICPAC 2011, SHEA 2022 and epic3, all read as raw text) · Each reason now says what the cited source says.
- 2026-11-17 · Central line care for children with cancer: lock trials and swimming: The ethanol trial was called Dutch and the first taurolidine trial Danish, which their abstracts do not state. The swimming variant said some programmes allow swimming with a tunnelled line without saying these were home parenteral nutrition programmes, not cancer units. (Second accuracy pass; Europe PMC abstracts of Schoot 2015 and Handrup 2013, and Miller 2014 full text through NCBI E-utilities) · The nationalities are removed and the swimming variant names the programmes surveyed.
- 2026-11-17 · Central line care for children with cancer: the Children's Oncology Group document: The page said the Children's Oncology Group archived its guideline in 2018 without saying what it was. The archived document is the group's October 2016 endorsement of a 2013 guideline for patients with cancer, which uses no formal grading. (Second accuracy pass; the archived PDF on childrensoncologygroup.org, read with pdftotext) · The summary and gaps now describe the document, and it is listed as a source.
- 2026-11-17 · High blood pressure and protein in the urine on VEGF-pathway inhibitors: axitinib urine protein rule: The page said the axitinib label tells prescribers to withhold for moderate to severe proteinuria without giving a figure, and left axitinib out of the labels that withhold at 2 grams. The label's dose table says to withhold at 2 or more grams of protein per 24 hours until below 2 grams, then resume at a reduced dose. (Second accuracy pass; the Inlyta US prescribing information, DailyMed SPL XML read as raw text) · Axitinib is now listed with the other four labels at the 2 gram threshold, with its restart rule.
- 2026-11-17 · High blood pressure and protein in the urine on VEGF-pathway inhibitors: which labels stop for nephrotic syndrome and test before treatment: The page said all five labels stop treatment for nephrotic syndrome and all five test urine before and during treatment. The axitinib label gives no nephrotic syndrome rule, the bevacizumab label asks for serial dipstick tests during treatment only, and the ramucirumab label gives no timing. (Second accuracy pass; the five US labels on DailyMed, SPL XML read as raw text) · Each rule is now attributed to the labels that state it.
- 2026-11-17 · High blood pressure and protein in the urine on VEGF-pathway inhibitors: controlling blood pressure before the first dose: The summary said every label read says to control blood pressure before starting. The axitinib, lenvatinib, ramucirumab and fruquintinib labels say so; the bevacizumab label does not. (Second accuracy pass; the five US labels on DailyMed, SPL XML read as raw text) · The summary and the step now name the four labels and say the bevacizumab label does not.
- 2026-11-17 · High blood pressure and protein in the urine on VEGF-pathway inhibitors: diltiazem and verapamil: The step told teams to avoid the two medicines or use them only after a pharmacist's check, citing the AHA statement, which says only to use them with caution and names no pharmacist. It also missed that the ESC guideline does not recommend them (Class III, Level C), with a footnote allowing them in selected people intolerant of multiple other medicines. (Second accuracy pass; ESC 2022 Recommendation Table 33, read as text from the published PDF, and the AHA statement from PubMed Central XML) · The step now cites the ESC recommendation with its grade and footnote, and gives the AHA wording as caution.
- 2026-11-17 · High blood pressure and protein in the urine on VEGF-pathway inhibitors: ESC grades for the choice of blood pressure medicine: The page said the ESC class and level for ACE inhibitors, ARBs and calcium channel blockers could not be read. Recommendation Table 33 gives Class I, Level B for an ACE inhibitor or ARB first, and Class I, Level C for a dihydropyridine calcium channel blocker second and for the combination. (Second accuracy pass; ESC 2022 Recommendation Table 33, read as text from the published PDF in an open university repository) · The grades are now on the steps, in the variants and in the grades list, and the gap is rewritten.
- 2026-11-17 · High blood pressure and protein in the urine on VEGF-pathway inhibitors: ESC Figure 33: The page said the figure sets home systolic thresholds and treats only from 160 mmHg where prognosis is under one year. The figure is headed home blood pressure, not systolic, and marks 140 to 159 mmHg as may treat in that group. (Second accuracy pass; ESC 2022 Figure 33, read as text from the published PDF) · The figure is now described cell by cell with its own words.
- 2026-11-17 · High blood pressure and protein in the urine on VEGF-pathway inhibitors: ramucirumab hold rule: The best-in-class statement said ramucirumab is withheld for severe hypertension not controlled medically. The label says to withhold for severe hypertension until it is controlled, and to stop if it cannot be controlled. (Second accuracy pass; the Cyramza US prescribing information, DailyMed SPL XML read as raw text) · Corrected, and the grade 4 stop rules of the axitinib and fruquintinib labels were added.
- 2026-11-17 · High blood pressure and protein in the urine on VEGF-pathway inhibitors: emergency symptoms and urgency: The page told readers that severe headache, confusion, fits or loss of vision need emergency help and hospital treatment the same day, and told teams to arrange an urgent brain scan. The bevacizumab label lists the symptoms and says a scan is necessary to confirm the diagnosis, but gives no time limit. (Second accuracy pass; the Avastin label and Macmillan Cancer Support's bevacizumab page, both read as raw text) · The patient advice now follows Macmillan (contact the hospital straight away), which is added as a source, and the clinical step follows the label.
- 2026-11-17 · High blood pressure and protein in the urine on VEGF-pathway inhibitors: reasons given for steps: Several reasons were not in the cited sources: that protein in the urine is the first sign of filter damage, that a clinic reading checks the home machine, that the complications can recur and be fatal, that falls follow when tablets are not reduced, and that amlodipine and a validated machine were ESC advice. ESC says none of these; amlodipine is named by ESMO and a validated device by the AHA statement. (Second accuracy pass; ESC 2022, the AHA statement, ESMO 2020 and the labels, each read as raw text) · Each reason was reworded to what its source says and attributed to that source.
- 2026-11-17 · High blood pressure and protein in the urine on VEGF-pathway inhibitors: when to start treatment, and what the guidelines agree on: The step said to start or step up treatment at 140/90 mmHg; IC-OS says the threshold for starting can be extended to 140/90 mmHg and sets it above 130/80 mmHg for people with heart disease, proteinuric kidney disease or diabetes. The page also said all four guidelines agree on a pause at 180/110 mmHg; only ESC and IC-OS set it. The summary called hypertension the commonest cardiovascular effect where ESC says the most reported adverse event. (Second accuracy pass; IC-OS 2022 from the PubMed Central page, ESC 2022 from the published PDF, ESMO 2020 from PubMed Central XML) · Reworded to each source's own terms.
- 2026-11-17 · QT prolongation on cancer medicines: evidence level of the ESC recommendations: The page said every ESC recommendation on QT prolongation is Level C. The guideline grades QTc monitoring at baseline and at 14 and 28 days on ribociclib as Class I, Level A, and monitoring with any ribociclib dose increase as Class I, Level B. Only the management recommendations in Recommendation Table 32 are all Level C. (Second accuracy pass; 2022 ESC cardio-oncology guideline, publisher PDF from an open university repository read with pdftotext) · Limited the statement to Recommendation Table 32, added the two ribociclib grades to the grades list, the summary and the schedule variants, and rewrote the first gap.
- 2026-11-17 · QT prolongation on cancer medicines: threshold and source for pausing treatment: The step said to pause the medicine above 500 ms and cited the IC-OS statement, which only recommends considering a change of cancer therapy above 500 ms. The instruction to interrupt, with an ECG every 24 hours, is the ESC recommendation, and its threshold is 500 ms or more (Class I, Level C). (Second accuracy pass; 2022 ESC guideline Recommendation Table 32 read as text; IC-OS author accepted manuscript read in full) · Re-sourced the step to ESC with its own sentence and grade, and corrected the threshold to 500 ms or more.
- 2026-11-17 · QT prolongation on cancer medicines: a rise of more than 60 ms: The page said only the ribociclib label acts on a rise of more than 60 ms, and only with an arrhythmia. The adagrasib label withholds for an increase of more than 60 ms from baseline and resumes at the next lower dose, and the revumenib label interrupts for a rise of more than 60 ms and restarts at the lower dose. The step also added "and there are no symptoms", which ESC does not say. (Second accuracy pass; Krazati and Revuforj US prescribing information, DailyMed SPL XML read raw) · Added both label rules to the step, the summary, the best-in-class statement, the hold-rule variant and the gaps; reworded the step to ESC wording (should not routinely affect treatment decisions).
- 2026-11-17 · QT prolongation on cancer medicines: starting limit for ribociclib: The page said ribociclib may not be started above 450 ms. The label allows a start only in patients with QTcF values less than 450 ms. (Second accuracy pass; Kisqali US prescribing information, DailyMed SPL XML read raw) · Stated each label limit exactly: ribociclib only below 450 ms, quizartinib and revumenib not above 450 ms.
- 2026-11-17 · QT prolongation on cancer medicines: emergency step and patient warning signs: The last step told staff to call the emergency team, attach a monitor and check potassium and magnesium at once, and the patient summary said fainting or a racing heartbeat needs emergency help. No source read says this. ESC says symptomatic ventricular arrhythmias need a dose reduction or discontinuation and referral to a cardiologist and that treatment follows general guidelines. The labels tell patients to report dizziness, light-headedness, fainting or a fast or irregular heartbeat right away. (Second accuracy pass; 2022 ESC guideline section 6.4.2 (Internet Archive capture of the publisher page); five US labels on DailyMed read raw) · Reworded the step to the ESC text, reworded the patient lines to the labels, and named the missing emergency procedure in the gaps.
- 2026-11-17 · QT prolongation on cancer medicines: restarting at a reduced dose: The summary said treatment is usually restarted at a reduced dose. ESC says restarting may be considered, ideally at a reduced dose (Class IIb, Level C). (Second accuracy pass; 2022 ESC guideline Recommendation Table 32 read as text) · Reworded with the hedge and the Class IIb grade.
- 2026-11-17 · QT prolongation on cancer medicines: agreement between the guidelines: The page said ESC, IC-OS and ESMO agree on 500 ms as the level at which treatment changes, that ESMO prefers Fridericia and that all three say to avoid other QT-prolonging medicines. ESMO gives no QTc threshold, says Fridericia may be preferable, and says such medicines call for caution. (Second accuracy pass; ESMO 2020 consensus recommendations, PubMed Central XML through NCBI E-utilities) · Rewrote the best-in-class statement source by source.
- 2026-11-17 · QT prolongation on cancer medicines: grade beside the 480 to 500 ms step: The step carried Class I, Level C beside a narrative sentence about how common QT prolongation is, which has no grade. The grade belongs to the recommendation for weekly ECG monitoring in asymptomatic patients with QTcF 480 to 500 ms. (Second accuracy pass; 2022 ESC guideline Recommendation Table 32 read as text) · Replaced the quotation with the graded recommendation and added that it applies to people without symptoms.
- 2026-11-17 · QT prolongation on cancer medicines: adagrasib evidence row: The row on ECG and electrolyte monitoring for adagrasib quoted a different sentence, about avoiding the medicine in congenital long QT syndrome. (Second accuracy pass; Krazati US prescribing information, DailyMed SPL XML read raw) · Replaced the quotation with the label sentence on monitoring ECGs and electrolytes.
- 2026-11-17 · QT prolongation on cancer medicines: reasons given for the steps: Twelve of the reasons gave explanations no source states, for example that correcting other causes often brings the interval back down, that the interval usually shortens once the medicine is paused, that low blood salts make a dangerous rhythm much more likely, and that Bazett leads to medicines being stopped needlessly. (Second accuracy pass; All eight sources re-read as raw text) · Rewrote each reason to what the cited guideline or label says.
- 2026-11-17 · QT prolongation on cancer medicines: IC-OS quotations: Two quotations from the IC-OS consensus statement could not be confirmed, because the published text could not be read on 17 November 2026 and the author accepted manuscript words the same sentences slightly differently. (Second accuracy pass; IC-OS author accepted manuscript from the University of Glasgow repository, read with pdftotext) · Removed both quotations, kept the claims the manuscript supports, and said so in the gaps.
- 2026-11-17 · Atrial fibrillation, bleeding and high blood pressure on BTK inhibitors: anticoagulation grade: The step on anticoagulation carried the ESC grade Class I, Level C beside a narrative sentence that also covers the lower scores (1 in men, 2 in women), which ESC grades Class IIa, Level C in its recommendation table. (Second accuracy pass; 2022 ESC cardio-oncology guideline, Recommendation Table 31, published PDF read as text) · The quote is now the graded table sentence, and the page says anticoagulation is recommended at 2 or more in men or 3 or more in women (Class I, Level C) and should be considered at 1 in men or 2 in women (Class IIa, Level C).
- 2026-11-17 · Atrial fibrillation, bleeding and high blood pressure on BTK inhibitors: choice of anticoagulant: The page said only that ESC notes drawbacks of vitamin K antagonists and drug interactions with direct oral anticoagulants. The full ESC table, which the first writer could not read, says a direct oral anticoagulant should be considered in preference to low molecular weight heparin and vitamin K antagonists in people without a high bleeding risk, significant drug interactions or severe renal dysfunction (Class IIa, Level B). (Second accuracy pass; 2022 ESC cardio-oncology guideline, Recommendation Table 31, published PDF read as text) · The ESC recommendation and its grade were added to the summary, the warfarin step, the best in class statement, the variant and the grades.
- 2026-11-17 · Atrial fibrillation, bleeding and high blood pressure on BTK inhibitors: ventricular arrhythmia or cardiac arrest: The step said to avoid ibrutinib after a ventricular arrhythmia or cardiac arrest. The international consensus says BTK inhibitors as a class, especially ibrutinib, should be avoided and prefers a BCL-2 inhibitor to any BTK inhibitor. (Second accuracy pass; Awan et al., Blood Advances 2022, Europe PMC full-text XML) · The step now says to avoid BTK inhibitors, especially ibrutinib, and adds the Australia and New Zealand grade (Class III, Level C).
- 2026-11-17 · Atrial fibrillation, bleeding and high blood pressure on BTK inhibitors: rate control: Rate control with a beta-blocker was written as an order. ESC says a rate control strategy, preferably with beta-blockers, should be considered in people whose atrial fibrillation is well tolerated (Class IIa, Level C). (Second accuracy pass; 2022 ESC cardio-oncology guideline, Recommendation Table 31, published PDF read as text) · The step and the summary now say consider, give the scope and print the grade.
- 2026-11-17 · Atrial fibrillation, bleeding and high blood pressure on BTK inhibitors: baseline checks and blood pressure schedule: Two steps cited the ibrutinib label for a baseline ECG and blood pressure and for blood pressure at every visit with home readings weekly then monthly. The label prints none of these; it says to evaluate cardiac history and function at baseline and to monitor blood pressure. The schedule is from ESC, which says home readings should be considered (Class IIa, Level C). (Second accuracy pass; Imbruvica US label revised October 2025 on DailyMed, ESC Recommendation Table 12 and the Australia and New Zealand consensus Table 1, all read as raw text) · The baseline step now cites the Australia and New Zealand consensus (Class I, Level B) and the blood pressure step cites ESC (Class I, Level B), with home readings worded as consider.
- 2026-11-17 · Atrial fibrillation, bleeding and high blood pressure on BTK inhibitors: baseline echocardiogram: The step ordered an echocardiogram but cited the international consensus, which says one can be considered. The recommendation comes from ESC, for high-risk patients as ESC defines them (Class I, Level C). The reason given, about weak heart muscle and valve disease, was not in either source. (Second accuracy pass; 2022 ESC cardio-oncology guideline, Recommendation Table 12, and Awan et al. 2022, read as raw text) · The step now cites ESC with its grade and definition of high risk, and states the consensus position as can be considered.
- 2026-11-17 · Atrial fibrillation, bleeding and high blood pressure on BTK inhibitors: urgency wording: The page told patients that fainting, chest pain or heavy bleeding need urgent help, told staff to treat bleeding signs as urgent and collapse as an emergency, and said bleeding in the brain had occurred on every BTK inhibitor. No cited source said these things. The acalabrutinib patient information says to tell the healthcare provider right away about signs of bleeding or of a heart rhythm problem; the Australia and New Zealand consensus asks for urgent cardiology review if ventricular tachycardia is suspected. (Second accuracy pass; Calquence US label revised February 2026 with its patient information, and the other three labels, DailyMed SPL XML) · The wording now follows the acalabrutinib patient information (report straight away, with its list of signs), the emergency wording was replaced by the consensus sentence, and a gap says no source read sets a time limit or says when to call an ambulance.
- 2026-11-17 · Atrial fibrillation, bleeding and high blood pressure on BTK inhibitors: blood pressure medicine after the FLAIR signal: A gap said no source read changes blood pressure treatment because of the FLAIR finding on ACE inhibitors. The international consensus table on new hypertension lists a medicine other than an ACE inhibitor in the first instance. (Second accuracy pass; Awan et al., Blood Advances 2022, Table 4, Europe PMC full-text XML) · The gap and the best in class statement now say so and set it beside the ESC recommendation of an ACE inhibitor or ARB first (Class I, Level B).
- 2026-11-17 · Atrial fibrillation, bleeding and high blood pressure on BTK inhibitors: unsourced reasons and absolutes: Several reasons went beyond the sources: that a blood thinner prevents most strokes, that high blood pressure adds strain on the heart, that many people can stay on treatment, that dual antiplatelet therapy carries a high risk of serious bleeding, that the trials (rather than most trials) excluded warfarin, and that no randomised trial has tested an anticoagulant or a pause length. The acknowledgment in the international consensus was also described as medical writing, which the paper does not say. (Second accuracy pass; all seven readable sources read as raw text) · Each was reworded to what the source prints, and the absolute statements now say that none of the sources read cites such a trial.
- 2026-11-17 · Raised liver enzymes on targeted anticancer drugs: when to admit: The step said to admit and involve a liver specialist for jaundice with abnormal clotting or confusion. AASLD says people with coagulopathy or mental status changes may require hospitalisation, and recommends early transfer to a liver transplant centre for acute liver failure. (Second accuracy pass; AASLD 2023 practice guidance, Europe PMC full-text XML) · The step now says to consider admission and to arrange early transfer to a transplant centre if acute liver failure develops.
- 2026-11-17 · Raised liver enzymes on targeted anticancer drugs: steroids on sotorasib: The step said to consider steroids particularly when a checkpoint inhibitor was given in the previous three months, and that liver injury was more than twice as common after immunotherapy. The label says only to consider systemic corticosteroids, and prints 40%, 18% and 17%; it prints no ratio. (Second accuracy pass; Lumakras US label, DailyMed SPL XML) · The emphasis and the computed ratio were removed and the label's own figures given.
- 2026-11-17 · Raised liver enzymes on targeted anticancer drugs: share of liver injury caused by the drug: The page gave 30% as AASLD's figure for drug injury among causes of liver injury in advanced cancer, as a general finding. The sentence is in AASLD's section on checkpoint inhibitors and says recent studies suggest it. (Second accuracy pass; AASLD 2023 practice guidance, Europe PMC full-text XML) · The figure is now scoped to that section and hedged as the source hedges it, and a gap says no source read gives the share for targeted drugs.
- 2026-11-17 · Raised liver enzymes on targeted anticancer drugs: how fast to report symptoms: The page told patients to report symptoms the same day and listed itching under the pazopanib label. The labels say right away or immediately and set no time limit; itching is in the tucatinib, lapatinib and trastuzumab emtansine labels, not the pazopanib list. (Second accuracy pass; twelve US labels, DailyMed SPL XML) · The wording is now straight away, the step's list follows the pazopanib label, and the other labels are named for itching.
- 2026-11-17 · Raised liver enzymes on targeted anticancer drugs: extra tests during close observation: The step ordered clotting and a split bilirubin with every repeat test. The FDA guidance says additional tests as appropriate, for example INR and direct bilirubin, and lets retesting ease to weekly or less once results stabilise. (Second accuracy pass; FDA guidance of July 2009, PDF read with pdftotext) · The step now says as appropriate and adds the easing rule.
- 2026-11-17 · Raised liver enzymes on targeted anticancer drugs: monitoring sentences: The summary said all twelve labels add as clinically indicated to their schedule. The trastuzumab emtansine monitoring sentence does not. (Second accuracy pass; Kadcyla US label, DailyMed SPL XML) · Corrected to all except trastuzumab emtansine.
- 2026-11-17 · Raised liver enzymes on targeted anticancer drugs: Gilbert's syndrome and weekly tests: A pazopanib rule on Gilbert's syndrome was written as a rule for every drug, and weekly tests during a mild rise were attributed to many labels. Only the pazopanib and vorasidenib labels ask for weekly tests, and the Gilbert's rule is in the pazopanib label alone. (Second accuracy pass; Votrient and Voranigo US labels, DailyMed SPL XML) · Both steps are now scoped to the labels that say so.
- 2026-11-17 · Raised liver enzymes on targeted anticancer drugs: reasons given for steps: Ten reasons stated things no cited source says, for example that most rises come without symptoms, that another medicine is often the real cause, that the injury is not usually an immune attack, and that a lower dose makes a second rise less likely. The page also said the injury is usually silent and that AASLD reserves steroids for certain features, where AASLD says they may be of benefit in selected patients. (Second accuracy pass; all sources re-read as raw text) · Each reason now states what the cited label or guidance says.
- 2026-11-17 · Interstitial lung disease on antibody-drug conjugates and targeted drugs: severe (grade 3 and 4) steroid step: One step gave methylprednisolone 1 to 2 mg/kg a day for grade 3 or 4 and told staff to admit. In the expert opinion that dose is for grade 3 only. For grade 4 it says severe pneumonia is usually treated with intravenous methylprednisolone or prednisolone 2 mg/kg a day, with a pulse of 500 to 1000 mg a day for 3 days as an alternative, and for grade 3 it says severe symptoms may lead to hospitalisation. (Second accuracy pass; Conte and colleagues (ESMO Open 2022), Europe PMC full-text XML read raw) · The step is now for grade 3, admission is described as possible, oxygen is for low oxygen levels, and the grade 4 regimen is stated separately.
- 2026-11-17 · Interstitial lung disease on antibody-drug conjugates and targeted drugs: when to report symptoms: The page told patients to report a cough, breathlessness or fever the same day. The labels read say to report immediately (trastuzumab deruxtecan, osimertinib) or promptly (everolimus) and none sets a same-day limit. (Second accuracy pass; DailyMed label XML for Enhertu, Tagrisso and Afinitor read raw) · Reworded to straight away in the summary line, the plain summary and the step, and the missing time limit is named in the gaps.
- 2026-11-17 · Interstitial lung disease on antibody-drug conjugates and targeted drugs: holding the drug on suspicion: A step said to hold the next dose whenever lung inflammation is suspected, citing the ribociclib label. Only the ribociclib, datopotamab deruxtecan, sotorasib, olaparib and osimertinib labels say to withhold on suspicion. The everolimus label continues treatment with few or no symptoms and the abemaciclib label asks for no change at grade 1. (Second accuracy pass; eight DailyMed labels read raw) · The step now names the labels that ask for a hold, and the best-supported statement says a hold applies where the label asks for one.
- 2026-11-17 · Interstitial lung disease on antibody-drug conjugates and targeted drugs: steroid regimen for symptomatic cases: A step gave prednisolone 1 mg/kg a day or more for at least 14 days as the rule for symptomatic inflammation on any of the drugs. The figure is the example regimen in the trastuzumab deruxtecan and datopotamab deruxtecan labels only, and the stated reason that steroids calm the inflammation is not in a source, which says their efficacy is not clearly established. (Second accuracy pass; Enhertu and Datroway labels and Conte and colleagues read raw) · The step is scoped to the two deruxtecan drugs and worded as the labels' example, and the reason now reports what the expert opinion says.
- 2026-11-17 · Interstitial lung disease on antibody-drug conjugates and targeted drugs: abemaciclib at grade 2: The best-supported statement said abemaciclib may resume at a lower dose after recovery at grade 2. The label asks for no dose change at grade 2 unless it persists beyond 7 days or recurs, and only then withholds and lowers the dose. (Second accuracy pass; Verzenio label (DailyMed XML) read raw) · Statement corrected.
- 2026-11-17 · Interstitial lung disease on antibody-drug conjugates and targeted drugs: steroids at grade 1, grade row: The row saying steroids are not generally needed at grade 1 carried a quotation about steroid efficacy that does not say so. The sentence that states it is the next one in the paper. (Second accuracy pass; Conte and colleagues, Europe PMC full-text XML, quotes checked by script) · The row now carries the sentence that states the claim, and the efficacy sentence has its own row.
- 2026-11-17 · Interstitial lung disease on antibody-drug conjugates and targeted drugs: reasons given for steps: Twelve reasons said more than the sources do, for example that steroids can worsen an untreated infection, that symptomless cases on everolimus often do not progress, that milder cases on CDK4/6 inhibitors often settle with a pause, and that chest radiotherapy is a risk factor. None of these is in the text read. (Second accuracy pass; Conte, Powell and the eight labels read raw) · Each reason was replaced with what its source states, or removed.
- 2026-11-17 · Interstitial lung disease on antibody-drug conjugates and targeted drugs: expert opinion on rechallenge and second-line drugs: The page said rechallenge after grade 2 may be considered for most targeted agents after resolution, and listed infliximab, mycophenolate or immunoglobulin for no response in 48 to 72 hours. The paper says the drug should in most cases be stopped permanently, sets four conditions for considering rechallenge, includes tocilizumab at grade 3, and gives 48 hours at grade 4. (Second accuracy pass; Conte and colleagues, Europe PMC full-text XML read raw) · Both passages reworded to the paper's conditions and grades.
- 2026-11-17 · Fluid retention, swelling and pleural effusion on kinase inhibitors and MET-targeted drugs: diuretics: The step said to add a diuretic when swelling interferes with daily life and cited a European LeukemiaNet sentence that sets no threshold. The MET inhibitor review says to consider a diuretic if oedema is interfering with quality of life. (Second accuracy pass; Nishio et al. 2022 (Targeted Oncology), Europe PMC full-text XML read as raw text) · Step, summary and best-in-class statement now say consider, and the step quotes the review's own table.
- 2026-11-17 · Fluid retention, swelling and pleural effusion on kinase inhibitors and MET-targeted drugs: weight gain and warning signs: The page said to record a baseline weight, to investigate rapid weight gain the same week, and that patients should call the same day. The imatinib label says to weigh regularly and investigate unexpected rapid weight gain carefully, and the dasatinib label says to seek medical attention promptly. No source read gives a time limit. (Second accuracy pass; Gleevec and Sprycel US labels, current DailyMed SPL XML read as raw text) · Reworded to the labels' words (regularly, carefully, promptly) and the missing time limit named in gaps.
- 2026-11-17 · Fluid retention, swelling and pleural effusion on kinase inhibitors and MET-targeted drugs: looking for other causes: The page told the reader to check kidney, liver and thyroid function and a heart failure marker and listed low albumin as a cause. The review says these tests may help and does not list low albumin. (Second accuracy pass; Nishio et al. 2022, Europe PMC full-text XML) · Step reworded to consider, low albumin removed.
- 2026-11-17 · Fluid retention, swelling and pleural effusion on kinase inhibitors and MET-targeted drugs: capmatinib, tepotinib and amivantamab dose rules: The page gave permanent discontinuation at grade 4 for tepotinib only, left out that withholding at intolerable grade 2 is something to consider, and omitted the amivantamab grade 4 row. The capmatinib label also says to permanently discontinue at grade 4, and the amivantamab label says to withhold at grade 4 until grade 1 or baseline and resume at a reduced dose if recovery is within 4 weeks. (Second accuracy pass; Tabrecta, Tepmetko and Rybrevant US labels, current DailyMed SPL XML) · Step, summary, best-in-class statement and variant corrected, and the amivantamab quote replaced with the sentence that carries the claim.
- 2026-11-17 · Fluid retention, swelling and pleural effusion on kinase inhibitors and MET-targeted drugs: dasatinib pleural effusion: The steps said to drain a large effusion, that most effusions shrink once the drug is paused, and that a lower dose or another drug usually controls the leukaemia as well. The sources say a severe effusion may require drainage and that effusions are potentially reversible, and make no statement on leukaemia control. The 2025 European LeukemiaNet recommendations, which replace the 2020 edition, were not on the page. (Second accuracy pass; Steegmann et al. 2016 and Apperley et al. 2025 (Leukemia), Europe PMC full-text XML; Sprycel US label) · Steps reworded to the sources, the first-episode rule restored, and the 2025 edition added with its statement that recurrent effusions despite dose reductions call for a change of drug.
- 2026-11-17 · Fluid retention, swelling and pleural effusion on kinase inhibitors and MET-targeted drugs: skin care, reasons and scope: The page gave skin-care details (moisturise, report redness or heat), several reasons, and a statement that the same principles apply to other kinase inhibitors, none of which a cited source states. It also described the amivantamab clot warning without saying it applies to use with lazertinib, and gave that label's year as 2026 when it was revised in November 2025. (Second accuracy pass; Nishio et al. 2022 and the six US labels read as raw text) · Reworded to what the sources say, scope of the sources stated, label year corrected.
- 2026-11-17 · High blood sugar on PI3K, AKT and mTOR inhibitors: gedatolisib monitoring and dose rule: The page said all four PI3K and AKT inhibitor labels repeat HbA1c every 3 months and interrupt the medicine above 250 mg/dL. The gedatolisib label says HbA1c at regular intervals, and at 251 to 500 mg/dL asks for hydration and a retest, withholding only if glucose stays above 250 mg/dL. (Second accuracy pass; Revtorpyk US label, DailyMed SPL XML read as raw text) · Summary and best-in-class statement now state the gedatolisib rules separately.
- 2026-11-17 · High blood sugar on PI3K, AKT and mTOR inhibitors: capivasertib stop rule: The capivasertib variant said to stop for good above 500 mg/dL only if the value persists after 24 hours. The label also stops it for good for life-threatening effects of hyperglycaemia at any glucose level. (Second accuracy pass; Truqap US label, DailyMed SPL XML read as raw text) · The stop rule now includes life-threatening effects, and the 71 day median onset with abiraterone is given beside the 15 days with fulvestrant.
- 2026-11-17 · High blood sugar on PI3K, AKT and mTOR inhibitors: ketone checks and the reasons for avoiding insulin: The summary said both expert papers ask for ketone checks on an SGLT2 inhibitor and both hold insulin back because of pathway reactivation. The alpelisib panel did not require ketone monitoring, and the capivasertib paper advises avoiding insulin secretagogues because of hypoglycaemia. (Second accuracy pass; Gallagher 2024 and Iyengar 2025, Europe PMC full text XML) · Each position is attributed to the paper that holds it, in the summary, the steps and the best-in-class statement.
- 2026-11-17 · High blood sugar on PI3K, AKT and mTOR inhibitors: steps written for every medicine: Several steps applied one label's rule to all six medicines: metformin beforehand, interrupting above 250 mg/dL, monitoring after an episode, and HbA1c every 3 months. The step above 250 mg/dL also said to give fluids if dehydrated, where the alpelisib label says to administer intravenous hydration. (Second accuracy pass; Piqray, Truqap, Itovebi and Revtorpyk US labels, DailyMed SPL XML) · Each step now names the medicine its source covers and says how the other labels differ.
- 2026-11-17 · High blood sugar on PI3K, AKT and mTOR inhibitors: orders where the source says consider: The page said to set up home glucose testing and that the UK guideline starts gliclazide. The capivasertib label says to consider home monitoring, and the UK guideline's summary says to consider commencing gliclazide 40 mg daily. (Second accuracy pass; Truqap US label and JBDS 17 PDF read with pdftotext) · Both are now worded as things to consider.
- 2026-11-17 · High blood sugar on PI3K, AKT and mTOR inhibitors: ketoacidosis step and warning signs: The ketoacidosis step listed hospital tests and treatment, and a same-day time limit, that the capivasertib label does not give, and the patient summary said fast breathing needs urgent help. The label says to withhold at once and its leaflet says ketoacidosis needs hospital treatment; the labels list difficulty breathing and say to contact the healthcare provider immediately. (Second accuracy pass; Truqap, Piqray and Itovebi US labels and Medication Guides, DailyMed SPL XML) · The step and the patient wording now follow the labels, and the gaps say no label gives a time limit in hours.
- 2026-11-17 · High blood sugar on PI3K, AKT and mTOR inhibitors: reasons no source states: Eleven reasons given for steps were not in the cited sources, among them that a raised starting blood sugar is the strongest warning, that blood sugar often rises before symptoms, and that the dose rules were tested on fasting values. (Second accuracy pass; all thirteen sources re-read as raw text) · Each reason was replaced with what the cited label or paper says, including the SOLAR-1 metformin schedule and the age 75 figures from the alpelisib label.
- 2026-11-17 · High blood sugar on PI3K, AKT and mTOR inhibitors: everolimus and inavolisib figures: The page gave grade 3 hyperglycaemia on everolimus as up to 15%, a figure the label gives for raised glucose, cholesterol and triglycerides together, and said grade 3 hyperglycaemia stayed frequent in the inavolisib phase 1 study despite metformin started first. The paper says hyperglycaemia remained frequent in patients with risk factors despite early metformin treatment. (Second accuracy pass; Afinitor US label and Gambardella 2025 full text) · Both statements reworded to the sources.
- 2026-11-17 · Low magnesium on cetuximab and panitumumab: whether any guideline grades replacement: The page said replacement advice was ungraded and that no guideline existed. The Japanese 2022 guideline on kidney injury during anticancer drug therapy (clinical question 8) weakly recommends additional magnesium when low magnesium develops on an EGFR antibody, with certainty of evidence D (very weak) and agreement of all 28 voters. (Second accuracy pass; the guideline's English full text (Clinical and Experimental Nephrology 2024) read as Europe PMC XML, found from the reference list of the systematic review already cited) · The guideline was added as a source, a step and a graded row, and the summary, best in class statement and gaps were rewritten.
- 2026-11-17 · Low magnesium on cetuximab and panitumumab: magnesium sulfate dose for grade 3 or 4: The summary gave one dose, 6 to 10 g daily to twice weekly, for both the 2008 commentary and the 2021 nephrology review. The commentary says about 6 to 10 g daily to twice weekly; the review says 6 to 10 g daily. The review also limits parenteral magnesium to 1 g an hour, while the commentary describes 8 g over 4 hours, and the page did not show that the two differ. (Second accuracy pass; Fakih 2008 read from the cancernetwork.com page and Workeneh 2021 from its PubMed Central page, both as raw text) · Each source's dose and rate is now stated separately, and the difference is named in the step, the variant and the gaps.
- 2026-11-17 · Low magnesium on cetuximab and panitumumab: cetuximab heart warning: The step told teams to keep to electrolyte checks strictly in anyone with coronary disease, heart failure or arrhythmia, and said low electrolytes may contribute to cardiac arrest. The US label scopes the warning to head and neck cancer treated with radiotherapy or with platinum and fluorouracil, asks for careful consideration of use in those with such a history, and does not say low electrolytes caused the deaths. The 2008 commentary says that link is unclear. (Second accuracy pass; Erbitux US prescribing information, DailyMed XML version 58, sections 5.2 and 5.6) · The step now keeps the label's scope and wording, gives the label's figure (2% of 208 patients) and says the link is unclear.
- 2026-11-17 · Low magnesium on cetuximab and panitumumab: grades in mmol/L: The page said no source gave a threshold in mmol/L and showed grades only in mg/dL. CTCAE version 5.0 prints both: grade 1 down to 0.5 mmol/L, grade 2 to 0.4, grade 3 to 0.3 and grade 4 below 0.3 mmol/L. (Second accuracy pass; the National Cancer Institute's CTCAE version 5.0 spreadsheet, unpacked by script (version 6.0 prints the same)) · CTCAE version 5.0 was added as a source and the thresholds are in the summary in both units.
- 2026-11-17 · Low magnesium on cetuximab and panitumumab: steps for mild falls, severe falls and pausing the antibody: Several steps said more than their sources. Regular infusions were to be added if magnesium kept falling, where the review says at grade 2 or 3. Pausing the antibody was offered for any uncontrolled fall, where the review says grade 4, until magnesium improves to grade 2 or less. A kidney specialist referral for amiloride, magnesium corrected first, a baseline ECG especially with heart disease, and recovery within weeks were not in the sources, which say weeks to a few months. (Second accuracy pass; Siddiqui 2021 (journal PDF through pdftotext) and Workeneh 2021 (PubMed Central page)) · Each step was reworded to its source, and unsourced reasons were removed or replaced with what the source states.
- 2026-11-17 · Low magnesium on cetuximab and panitumumab: how long to keep checking after treatment: The page said checks continue for 8 weeks after the last dose. The US cetuximab label says at least 8 weeks and the panitumumab texts say up to 8 weeks. (Second accuracy pass; DailyMed XML for both US labels and the EMA product information PDFs through pdftotext) · Both wordings are now given in the step and the best in class statement.
- 2026-11-17 · Low magnesium on cetuximab and panitumumab: symptoms patients are asked to report: The patient summary listed symptoms to report (cramps, tingling, twitching, great tiredness, palpitations) with no source. Cancer Research UK's panitumumab page asks patients to tell their doctor or nurse about cramping in arm or leg muscles, tingling or numbness, palpitations or feeling faint, and gives no time limit. (Second accuracy pass; Cancer Research UK patient pages on panitumumab and cetuximab, fetched as raw HTML) · The two pages were added as sources and the wording was matched to them. The gaps say that no source sets a time limit.
- 2026-11-17 · High phosphate on FGFR inhibitors (erdafitinib, pemigatinib, futibatinib): monitoring, diet, binders and dose rules: phosphate above 10 mg/dL: The page said every one of the three medicines is withheld as soon as phosphate goes above 10 mg/dL. The pemigatinib label says to start phosphate-lowering therapy and withhold only if phosphate is not 10 mg/dL or less within 1 week. (Second accuracy pass; US Pemazyre prescribing information section 2.3 and the EU text, read as raw text from DailyMed and the EMA) · The step and the best-in-class statement now give each medicine's own rule.
- 2026-11-17 · High phosphate on FGFR inhibitors (erdafitinib, pemigatinib, futibatinib): monitoring, diet, binders and dose rules: stopping binders in a treatment break: The page said all three EU texts say to stop binders and the diet during treatment breaks. The futibatinib text says to discontinue them if treatment is stopped or phosphate falls below normal, the pemigatinib text says this should be considered, and the erdafitinib text says to discontinue them only if phosphate falls below normal. (Second accuracy pass; the three EMA product information PDFs read with pdftotext) · The step, summary and best-in-class statement now say what each text says.
- 2026-11-17 · High phosphate on FGFR inhibitors (erdafitinib, pemigatinib, futibatinib): monitoring, diet, binders and dose rules: acetazolamide dose: The page gave acetazolamide as 250 mg two or three times a day and said it is added. The review's table prints a starting dose of 250 mg once a day with 250 mg two or three times a day as the maximum, and its text says one can consider adding it. (Second accuracy pass; Kommalapati 2021 full text from Europe PMC) · Both doses are given as printed and the wording is now that it can be considered.
- 2026-11-17 · High phosphate on FGFR inhibitors (erdafitinib, pemigatinib, futibatinib): monitoring, diet, binders and dose rules: orders where the sources hedge: Several lines read as orders where the source says consider or may: starting a binder above 7 mg/dL (erdafitinib warnings and EU pemigatinib text), a binder at 5.5 to 6.9 mg/dL and raising the binder dose (Kommalapati review), calcium carbonate (Subbiah review says it may be used), and restarting erdafitinib at a reduced dose (US label says may restart). (Second accuracy pass; labels, EU texts and both reviews read as raw text) · Each line now carries the source's own hedge.
- 2026-11-17 · High phosphate on FGFR inhibitors (erdafitinib, pemigatinib, futibatinib): monitoring, diet, binders and dose rules: details no source states: The page asked for calcium and kidney function before the first dose, a dietitian, phosphate checks early in the first weeks and monthly once stable for all three medicines, and action the same day on cramps or skin lesions. No source read states these; monthly checks are the erdafitinib rule only and the labels tell patients to inform their team immediately. (Second accuracy pass; all nine sources read as raw text) · Removed or reworded to the labels' wording, with the missing points named in the gaps.
- 2026-11-17 · High phosphate on FGFR inhibitors (erdafitinib, pemigatinib, futibatinib): monitoring, diet, binders and dose rules: reasons given for steps: Several reasons were not in any source: that a binder only works with meals, that cramps suggest low calcium, that phosphate falls when the medicine is paused, that diet may be enough for a small rise, and that eye changes are found on a scan before vision is lost. (Second accuracy pass; all nine sources read as raw text) · Each reason now says only what the cited label or review says.
- 2026-11-17 · High phosphate on FGFR inhibitors (erdafitinib, pemigatinib, futibatinib): monitoring, diet, binders and dose rules: what the labels agree on: The page said all texts agree on a binder with food, weekly checks while phosphate is raised and a warning about painful skin lesions. Taking the binder with food and the skin lesion warning are in the erdafitinib texts, and weekly checks are written into each medicine's rule differently. It also called futibatinib a cholangiocarcinoma medicine where the label says intrahepatic cholangiocarcinoma. (Second accuracy pass; the six regulatory texts read as raw text) · The best-in-class statement and the population were rewritten to match.
- 2026-11-17 · Underactive thyroid on VEGFR-targeted tyrosine kinase inhibitors: starting dose of levothyroxine: The page said no source it had read gave a levothyroxine starting dose. The 2011 review it cites does: 1.6 micrograms per kg a day is called reasonable for people without cardiac disease, and a lower dose such as 50 micrograms a day for the first few weeks in coronary artery disease. (Second accuracy pass; Hamnvik 2011 read in full from its PubMed Central page, with the ATA 2014 guideline and NICE NG145 read raw) · The gap is corrected and the starting doses are now a variant, each figure attributed to the review, the ATA discussion or NICE.
- 2026-11-17 · Underactive thyroid on VEGFR-targeted tyrosine kinase inhibitors: pazopanib monitoring: The page said the pazopanib label asks for thyroid tests 'periodically'. The label says to monitor thyroid tests at baseline, during treatment and as clinically indicated. It also said every label asks for tests before treatment, where the cabozantinib label asks for an assessment for signs with testing as clinically indicated. (Second accuracy pass; US labels read as DailyMed SPL XML) · Pazopanib and cabozantinib are now described in their labels' own terms in the summary, the best in class statement and the variant.
- 2026-11-17 · Underactive thyroid on VEGFR-targeted tyrosine kinase inhibitors: continuing the cancer medicine: A step said to continue the kinase inhibitor at the same dose. The French expert opinion says only that it should not be discontinued because of thyroid dysfunction, and the cabozantinib label refers from its thyroid section to its general dose-modification rules. (Second accuracy pass; Drui 2018 abstract from Europe PMC and the Cabometyx US label) · The step now says not to stop the medicine because thyroid dysfunction has appeared, and the reason given is the source's.
- 2026-11-17 · Underactive thyroid on VEGFR-targeted tyrosine kinase inhibitors: mildly raised TSH and the review's schedule: Two steps were written as orders where the review hedges: it calls a low-dose levothyroxine trial for a TSH of 5 to 10 'reasonable', and it calls its own monitoring schedule not evidence based. The first step also offered 'recheck in a few weeks', which the review does not say. (Second accuracy pass; Hamnvik 2011 read in full) · Both steps now say 'consider', and the unsourced alternative is removed.
- 2026-11-17 · Underactive thyroid on VEGFR-targeted tyrosine kinase inhibitors: symptoms to watch for: The symptom list had no source and included a slow pulse, low mood, palpitations and tremor, which no source read lists. The step also said to ask at each visit, where the label says to monitor closely. (Second accuracy pass; Lenvima EU product information with its patient leaflet, read with pdftotext) · The list now matches the EU lenvatinib patient leaflet: tiredness, weight gain, constipation, feeling cold and dry skin, or a rapid heart rate, sweating and weight loss.
- 2026-11-17 · Underactive thyroid on VEGFR-targeted tyrosine kinase inhibitors: overactive phase and levothyroxine steps: Several statements went beyond the sources: that an underactive phase 'usually' follows an overactive one (the French opinion says thyrotoxicosis precedes hypothyroidism in 20% to 40% of cases), that levothyroxine improves tiredness in about half of people (one sunitinib study, at least half), that people on levothyroxine should 'expect' a dose increase, and that starting low in older people carried the guideline's grade when it is in the discussion. (Second accuracy pass; Drui 2018 abstract, Hamnvik 2011 and ATA 2014 full texts) · Each statement is reworded to its source, and nine 'why' lines with reasons no source states are replaced by what the sources say.
- 2026-11-17 · Underactive thyroid on VEGFR-targeted tyrosine kinase inhibitors: regorafenib figures: The page gave regorafenib a hypothyroidism rate of 6.4% in one trial, from a label it did not list as a source. The same label reports 18% against 6% on placebo in its GIST trial. (Second accuracy pass; Stivarga US label (DailyMed XML) and EU product information (EMA PDF), both read in full) · Both figures are given, the two regorafenib texts are added as sources, and the gap saying the EU text could not be fetched is updated.
- 2026-11-17 · Differentiation syndrome on IDH, FLT3 and menin inhibitors and on tretinoin or arsenic trioxide: restarting the drug: The step said to restart any of the drugs once signs had improved to grade 2 or lower. The revumenib label says to resume at the same dose when signs improve to grade 1 or lower, and the olutasidenib label says to resume after resolution. (Second accuracy pass; US labels for all six drugs, DailyMed SPL XML read as text) · The step now gives each label's restart point: grade 2 or lower on ivosidenib, enasidenib, gilteritinib and ziftomenib, grade 1 or lower on revumenib, after resolution on olutasidenib.
- 2026-11-17 · Differentiation syndrome on IDH, FLT3 and menin inhibitors and on tretinoin or arsenic trioxide: leukapheresis and hydroxyurea: The page said leukapheresis was an option on the non-APL labels and that hydroxyurea is added for leukocytosis throughout. Leukapheresis appears only on the ivosidenib, revumenib and ziftomenib labels. The enasidenib and olutasidenib labels name hydroxyurea only, and the gilteritinib label gives no rule for leukocytosis. (Second accuracy pass; eight US labels, DailyMed SPL XML read as text) · The step, summary and best-in-class statement now name which label says what, with each label's white count threshold.
- 2026-11-17 · Differentiation syndrome on IDH, FLT3 and menin inhibitors and on tretinoin or arsenic trioxide: length of steroid treatment: The page said every label and the European LeukemiaNet panel state at least 3 days of steroids followed by a taper. The enasidenib label sets no minimum, and the panel says only that steroids can be discontinued once the syndrome has resolved. (Second accuracy pass; US labels (DailyMed SPL XML) and Sanz et al. 2019 Table 1 (PubMed Central page text)) · Summary and best-in-class statement rewritten: seven of eight labels set a 3 day minimum, enasidenib none, and the panel's wording is given as printed.
- 2026-11-17 · Differentiation syndrome on IDH, FLT3 and menin inhibitors and on tretinoin or arsenic trioxide: antibiotics: The step ordered empirical antibiotics at the same time as steroids. The FDA authors say empirical antibiotics would be equally important to consider. (Second accuracy pass; Norsworthy et al. 2020, PubMed Central XML through NCBI E-utilities) · Step, summary and best-in-class statement now say consider.
- 2026-11-17 · Differentiation syndrome on IDH, FLT3 and menin inhibitors and on tretinoin or arsenic trioxide: reasons given for steps: Nine why lines gave reasons no cited source states, among them that a high white count makes the syndrome more dangerous, that organs can fail within hours, that steroids take time to work and that the syndrome shows the drug is working. One credited steroids alone with cutting deaths from 30% to 5%, where the review credits early chemotherapy together with dexamethasone and says appear to have reduced. (Second accuracy pass; labels, Norsworthy 2020 and Montesinos and Sanz 2011 (Europe PMC full-text XML)) · Each why line now says what its source says, with the review's own hedge and figures.
- 2026-11-17 · Differentiation syndrome on IDH, FLT3 and menin inhibitors and on tretinoin or arsenic trioxide: monitoring step: The step listed blood pressure, pulse, oxygen, fluid balance, weight and kidney function as the monitoring to do. The labels say haemodynamic monitoring and do not define it, and only the enasidenib label recommends hospital admission for lung or kidney signs. (Second accuracy pass; eight US labels, DailyMed SPL XML read as text) · Step reworded to haemodynamic monitoring, with the admission advice attributed to the enasidenib label.
- 2026-11-17 · Differentiation syndrome on IDH, FLT3 and menin inhibitors and on tretinoin or arsenic trioxide: advice to patients: The short summaries told patients to tell the team the same day, said the reaction comes days or weeks after starting and said it needs treatment in hospital. The labels say to report immediately and to call or go to the nearest emergency room right away, record onset up to 3, 5 and 18 months after starting, and say the team may monitor in hospital. (Second accuracy pass; label sections 5.1 and 17 and medication guides, DailyMed SPL XML read as text) · Reworded to the labels' wording. The urgency was not lowered.
- 2026-11-17 · Differentiation syndrome on IDH, FLT3 and menin inhibitors and on tretinoin or arsenic trioxide: summary figures and scope: The summary gave leukocytosis as absent in a fifth to two fifths of cases, a fraction no source prints, and said the US label recommends prednisone prophylaxis for everyone on arsenic trioxide with tretinoin. The FDA paper prints leukocytosis in 79% and 61% of cases, and the label section is for newly diagnosed low-risk APL. (Second accuracy pass; Norsworthy 2020 and the Trisenox label section 2.1) · Printed percentages used; scope of the prophylaxis recommendation added.
- 2026-11-17 · Differentiation syndrome on IDH, FLT3 and menin inhibitors and on tretinoin or arsenic trioxide: revumenib label year: The revumenib label was dated 2026. The label text prints Revised 10/2025. (Second accuracy pass; Revuforj label, DailyMed SPL XML) · Year set to 2025 and the revision month added to the source label.
- 2026-11-17 · Capillary leak syndrome on tagraxofusp, dinutuximab, aldesleukin and gemcitabine: dinutuximab beta figure: The page said the EU label reports severe capillary leak in 5.2% of patients. The label says that overall 5.2% of capillary leak syndrome was severe (grade 3 to 4), and gives capillary leak at 35% across investigator-sponsored studies. (Second accuracy pass; Qarziba EU product information, EMA PDF read as raw text) · Reworded to the label's own sentence and added the 35% figure.
- 2026-11-17 · Capillary leak syndrome on tagraxofusp, dinutuximab, aldesleukin and gemcitabine: reasons given for steps: Thirteen steps gave reasons that no cited source states, for example that a fall in albumin is often the first sign of leak, that albumin holds fluid in the vessels, that steroids calm the leak, and that saline moves into the lungs. (Second accuracy pass; the four US labels (DailyMed XML), the Qarziba product information and the Izzedine 2022 review, read as raw text) · Each reason now says only what the label or review says, and says so where the source gives no reason.
- 2026-11-17 · Capillary leak syndrome on tagraxofusp, dinutuximab, aldesleukin and gemcitabine: patient advice: The page told patients on any of these medicines to call at once, citing the tagraxofusp label, which says to contact the health care professional and sets no time limit. The aldesleukin label is the one that says immediately. The plain summary also described the tagraxofusp checks as if they applied to all four medicines. (Second accuracy pass; Elzonris and Proleukin US labels section 17, DailyMed XML read as raw text) · The tagraxofusp step now matches its label, a step quoting the aldesleukin label was added, and the plain summaries were reworded.
- 2026-11-17 · Capillary leak syndrome on tagraxofusp, dinutuximab, aldesleukin and gemcitabine: checks for other causes and saline: A step on looking for infection and heart or lung causes was written for every drug, but the consensus paper says it only about tagraxofusp, in its discussion. The saline step said to keep saline for restoring blood pressure, where the review says saline given to restore volume should be limited. The page also said the consensus paper follows the label scheme, when it lists precautions without thresholds or doses. (Second accuracy pass; Pagano 2023 and Izzedine 2022, Europe PMC full-text XML) · Scoped the step to tagraxofusp, reworded the saline step and the best-in-class statement.
- 2026-11-17 · Capillary leak syndrome on tagraxofusp, dinutuximab, aldesleukin and gemcitabine: aldesleukin wording and Qarziba year: The page said high-dose aldesleukin, a term the US label does not use, and dated the Qarziba product information 2017, the year of first authorisation, when the text read was last updated on 29 October 2026. It also said flatly that no randomised trial exists, which no source states. (Second accuracy pass; Proleukin US label (DailyMed XML) and the EMA Qarziba page, read as raw text) · Dropped high-dose, corrected the year, and reworded to say none of the sources read cites a randomised trial.
- 2026-11-17 · Tumour flare on bispecific antibodies and lenalidomide in lymphoma: pausing mosunetuzumab: The page said the mosunetuzumab labels only ask for monitoring and that, among the labels, only the US glofitamab label pauses treatment for flare. The EU mosunetuzumab label says patients with grade 3 or 4 reactions, naming tumour flare, should have treatment temporarily withheld until symptoms are resolved. (Second accuracy pass; Lunsumio EU product information section 4.2, EMA PDF read as raw text) · Added a step with the EU rule quoted, and corrected the summary, the best supported statement, the variant and the gap.
- 2026-11-17 · Tumour flare on bispecific antibodies and lenalidomide in lymphoma: which labels ask for monitoring: The page said all the bispecific antibody labels, US and EU, ask for close monitoring of people with bulky disease near an airway or vital organ. The US epcoritamab label has no tumour flare warning and gives no such advice; the US glofitamab and mosunetuzumab labels and the three EU labels do. (Second accuracy pass; Epkinly, Columvi and Lunsumio US labels from DailyMed XML and the three EMA PDFs, read as raw text) · Named the labels that say it and stated that the US epcoritamab label is silent.
- 2026-11-17 · Tumour flare on bispecific antibodies and lenalidomide in lymphoma: lenalidomide figures: The summary said flare on lenalidomide was reported in about 10% of patients in the mantle cell and follicular or marginal zone trials, mostly grade 1 or 2 and in cycle 1. The US label gives 10% in the mantle cell trial (all grade 1 or 2, all in cycle 1), 10.8% in AUGMENT with one grade 3 case and no statement about the cycle, and 4.1% in MAGNIFY. It also says fatal reactions occurred in investigational use for chronic lymphocytic leukaemia and lymphoma, not the leukaemia alone. (Second accuracy pass; Revlimid US label section 5.11, DailyMed XML read as raw text) · Stated each trial separately and corrected the sentence on fatal reactions.
- 2026-11-17 · Tumour flare on bispecific antibodies and lenalidomide in lymphoma: lenalidomide in chronic lymphocytic leukaemia: A step said not to use lenalidomide for this leukaemia outside a trial because fatal tumour flare happened in it. The US label says it is not indicated and not recommended outside controlled trials, and the reason it gives is more deaths than on chlorambucil in a randomised trial. (Second accuracy pass; Revlimid US label sections 5.5 and 5.11, DailyMed XML read as raw text) · Reworded the step to the label words and gave the label reason.
- 2026-11-17 · Tumour flare on bispecific antibodies and lenalidomide in lymphoma: orders where the labels hedge: Steps said to give analgesia, to give the standard emergency treatment for compression, and to class a scan as an indeterminate response. The EU glofitamab label says corticosteroids and analgesics should be considered, the US mosunetuzumab label says standard treatment with no mention of emergency, and the LYRIC abstract describes provisional recommendations for clinical trials. (Second accuracy pass; Columvi EU product information, Lunsumio US label, LYRIC abstract from Europe PMC, read as raw text) · Reworded each step to the strength and scope of its source.
- 2026-11-17 · Tumour flare on bispecific antibodies and lenalidomide in lymphoma: reasons given for steps: Twelve reasons said things no cited source states, among them that swelling would be harmless in the groin, that a further dose could make pressure worse, that flare usually starts at home, and that steroids settle it over a few days. The patient summary said lumps swell for a few days before they shrink; the EU mosunetuzumab label gives a median duration of 36 days. (Second accuracy pass; all eight labels read as raw text) · Replaced each reason with what the labels state, and changed the patient summary to days or sometimes weeks.
- 2026-11-17 · Tumour flare on bispecific antibodies and lenalidomide in lymphoma: date of the US lenalidomide label: The US lenalidomide label was dated 2026. The current DailyMed version prints a revision date of March 2023. (Second accuracy pass; Revlimid US label, DailyMed XML read as raw text) · Year corrected to 2023.
- 2026-11-17 · Constipation in adults with cancer: rectal examination when counts are low: The page told staff not to examine the rectum in patients with a low white cell count, and one passage extended that to anyone at risk of low platelets, low white cells or mucositis. The NCI says manipulation of the rectum and anus should be avoided in immunocompromised patients; its sentence about patients at risk of low counts or mucositis covers rectal agents only, and its pain summary says suppositories and enemas are generally avoided in neutropenia or thrombocytopenia. (Second accuracy pass; NCI PDQ Gastrointestinal Complications and Cancer Pain, raw page text) · Each rule now carries the population and strength the NCI gives it, and the two are stated separately.
- 2026-11-17 · Constipation in adults with cancer: opioid antagonists: The page said to add or switch to a peripherally acting opioid antagonist when laxatives fail and called it the next step. NICE recommends naloxegol and naldemedine as an option, and the Cochrane review says naldemedine and methylnaltrexone may improve bowel function. (Second accuracy pass; NICE TA345 and TA651 recommendations and the Cochrane 2022 abstract, raw text) · The step now says consider, the summary says an option, and the evidence line says may improve.
- 2026-11-17 · Constipation in adults with cancer: the MASCC recommendations: The page said the MASCC recommendations could be read only as an abstract and reported none of them. An open author-accepted manuscript is held by UCL Discovery. MASCC recommends polyethylene glycol first, says opioid antagonists should always be considered in opioid-induced constipation (level I), and says diet and exercise changes have a limited role in advanced cancer (level V), each of which differs from what the page presented. (Second accuracy pass; Davies et al. 2020, author-accepted manuscript read in full with pdftotext) · MASCC added to the best supported version, the variants and the grades with its levels of evidence, and a new variant on fluids, fibre and exercise.
- 2026-11-17 · Constipation in adults with cancer: digital rectal examination: The step told staff to do a digital rectal examination. The NCI lists it among tests that may be part of the evaluation, and MASCC says the examination should ideally include one. (Second accuracy pass; NCI PDQ Gastrointestinal Complications raw text and the MASCC manuscript) · The step now says consider.
- 2026-11-17 · Constipation in adults with cancer: spinal cord compression step: The step listed signs of cord compression, called it an emergency and said hours matter for walking, citing an NCI sentence that only lists cord compression among causes of constipation. (Second accuracy pass; NICE NG234 recommendations, raw page text) · The step now cites NICE NG234 recommendation 1.3.2 and its box of signs, and the unsourced reason is removed.
- 2026-11-17 · Constipation in adults with cancer: warning signs for patients: The plain summaries told patients to ring at once about vomiting, a swollen sore tummy, leaking stool or weak legs with no source, and said a daily laxative prevents most constipation, which no source states. The page also said no source gives a figure for when to report. The NCI patient page asks carers to tell the team about fewer than three bowel movements in a week, severe abdominal pain, vomiting and other signs of impaction. (Second accuracy pass; NCI Constipation and Cancer Treatment patient page and NICE NG234, raw page text) · Warning signs reworded to those two sources, the efficacy claim removed, and the gap rewritten with the NCI figure.
- 2026-11-17 · Constipation in adults with cancer: reasons given for steps: Several reasons were not in any source: that medicines are the commonest changeable cause, that the history usually shows more than one cause, that most opioid constipation is under-treated, that impacted stool will not pass with oral laxatives, and that vincristine with colicky pain needs a radiograph arranged where the label says a flat film is useful. (Second accuracy pass; NCI PDQ summaries, NICE CG140 and the vincristine label, raw text) · Each reason reworded to what its source says, and the radiograph changed to consider.
- 2026-11-17 · Radiotherapy dose and fractionation by cancer site: lung SABR schedule: The page gave 54 Gy in 3 fractions as the SABR schedule for medically inoperable early lung cancer. The RCR lists three schedules on the same grade (54 Gy in 3, 55 Gy in 5 and 60 Gy in 8, each Grade B) and does not rank them. (Second accuracy pass; RCR Radiotherapy dose fractionation, fourth edition, PDF read as raw text) · All three schedules are now listed with their overall times.
- 2026-11-17 · Radiotherapy dose and fractionation by cancer site: prostate schedule: The page showed 60 Gy in 20 fractions as the only prostate schedule and said it matched an older seven-and-a-half-week course. The RCR also lists stereotactic radiotherapy of 36.25 Gy in 5 fractions (Grade A) and prints no seven-and-a-half-week figure. (Second accuracy pass; RCR Radiotherapy dose fractionation, fourth edition, PDF read as raw text) · The five-fraction option was added and the reason reworded to what the RCR says.
- 2026-11-17 · Radiotherapy dose and fractionation by cancer site: comparison with ASTRO: The page said ASTRO lists four bone metastasis schedules as equally recommended, that its 2018 breast guideline does not prefer 26 Gy in 5, and that ultrahypofractionation is to be considered in high-risk prostate cancer. The abstracts say the schedules are recommended without ranking them, do not mention 26 Gy in 5, and say may be considered. (Second accuracy pass; the three ASTRO abstracts read as raw text from Europe PMC) · The comparisons were reworded to the abstracts' own terms.
- 2026-11-17 · Radiotherapy dose and fractionation by cancer site: reasons given for schedules: Several reasons went beyond the RCR: that temozolomide is usually given, that a quarter of people need retreatment (the RCR says around 25% may), and that a single dose for cord compression is chosen to spare visits. (Second accuracy pass; RCR Radiotherapy dose fractionation, fourth edition, PDF read as raw text) · Each reason was reworded to the RCR's text.
- 2026-11-17 · Radiotherapy dose and fractionation by cancer site: commissioning caveat: The page attributed the commissioning caveat to the RCR document and told prescribers to check local commissioning. The sentence is on the RCR publication page, not in the document, and it gives no such instruction. (Second accuracy pass; RCR publication page and PDF read as raw text) · The caveat is attributed to the page and the step says only what it says.
- 2026-11-17 · Radiotherapy dose and fractionation by cancer site: cord compression: The page did not show that the RCR and NICE differ. The RCR reports that NICE recommends a single 8 Gy dose for all patients, while the RCR keeps longer courses for ambulant patients with a good prognosis or after surgery. (Second accuracy pass; RCR Radiotherapy dose fractionation, fourth edition, PDF read as raw text) · The difference was added as a variant and a gap.
- 2026-11-17 · Release of patients after radionuclide therapy or a permanent implant: telling the radiation safety officer: The page said to tell the radiation safety officer at once if a patient who cannot be released has an emergency or dies, and cited 10 CFR 35.310. The duty is in 10 CFR 35.315(b), which says as soon as possible and names the radiation safety officer or designee and an authorised user. (Second accuracy pass; 10 CFR 35.310 and 35.315, read as raw text from the eCFR) · The step now says the officer and an authorised user are notified, as soon as possible, and names the right section.
- 2026-11-17 · Release of patients after radionuclide therapy or a permanent implant: the British regulation: The page called the Ionising Radiation (Medical Exposure) Regulations 2017 a UK rule and left out that written instructions are required where appropriate. The regulations extend to England, Wales and Scotland. (Second accuracy pass; regulation 12, latest revised text, read as XML from legislation.gov.uk) · The summary, step and source label now give the extent and the words where appropriate; a gap says the Northern Ireland regulations were not read.
- 2026-11-17 · Release of patients after radionuclide therapy or a permanent implant: who is responsible for release: The page said the treating physician (the practitioner in the UK, the authorised user in the US) is responsible for release. The US rule says the licensee may authorise release, and the British regulation places the duty on the employer's procedures. (Second accuracy pass; 10 CFR 35.75 from the eCFR and regulation 12 from legislation.gov.uk, read raw) · The population now says what each source says.
- 2026-11-17 · Release of patients after radionuclide therapy or a permanent implant: must and should: The summary said the calculation's assumptions must match the written instructions, and that written instructions are needed when anyone could receive more than 1 mSv. IAEA SSG-46 says the assumptions should be consistent, and the US rule says is likely to exceed 1 mSv. (Second accuracy pass; IAEA SSG-46 paragraph 4.247 (PDF text) and 10 CFR 35.75 (eCFR)) · Both sentences now use the sources' own strength of wording.
- 2026-11-17 · Release of patients after radionuclide therapy or a permanent implant: how the dose is estimated: The page said the dose is estimated from a measured dose rate and that this gives a realistic estimate rather than a worst case. IAEA SSG-46 says an acceptable method is the time integral of the ambient dose equivalent rate and gives no such reason. (Second accuracy pass; IAEA SSG-46 paragraph 4.247, PDF read as text) · The step and summary now describe the method as the guide does.
- 2026-11-17 · Release of patients after radionuclide therapy or a permanent implant: the 1 mSv constraint for children and pregnancy: The page listed the 1 mSv per episode constraint for children and for the embryo or fetus as a Safety Guide recommendation, and the statement on dose monitoring likewise. SSG-46 says the approach is often taken and that published data suggest monitoring is not necessary. (Second accuracy pass; IAEA SSG-46 paragraphs 2.49 and 4.247, PDF read as text) · The grade labels and the summary now say often and published data suggest.
- 2026-11-17 · Release of patients after radionuclide therapy or a permanent implant: dose against retained activity: The page said the IAEA and ICRP both hold that the dose approach is the more fundamental, that many European countries apply activity limits in law and bind patients by them, and that fixed limits rest on worst-case models. SRS-63 reports the first as the ICRP's view, says national regulations may not mirror it, and says some models are unduly conservative. (Second accuracy pass; IAEA Safety Reports Series No. 63 sections 3.3 and 6, PDF read as text) · The best-in-class statement, the step and the variant now carry the source's attribution and hedges.
- 2026-11-17 · Release of patients after radionuclide therapy or a permanent implant: permanent implants and the home radionuclide examples: The summary said patients with implants get a wallet card and advice about death, and the page named radium-223 and palladium-103. SSG-46 says a wallet card is useful and that precautions after death are for the radiation protection officer to determine; no source read names those two radionuclides. (Second accuracy pass; IAEA SSG-46 paragraphs 5.260, 5.261 and 5.289, PDF read as text) · The summary follows each paragraph and the two examples were taken out of the prose and named in the gaps.
- 2026-11-17 · Release of patients after radionuclide therapy or a permanent implant: newer ICRP publication: A gap said ICRP and IAEA guidance on release predates lutetium-177 therapy. ICRP Publication 140, Radiological Protection in Therapy with Radiopharmaceuticals, was published in 2019. (Second accuracy pass; the ICRP publication page read raw, DOI resolved through Crossref) · Publication 140 was added as a source (abstract and key points only) and the gap rewritten.
- 2026-11-17 · Release of patients after radionuclide therapy or a permanent implant: pending changes to the US rule: The page gave the US rule with no word that it is under revision, and dated 10 CFR 35.2075 to 2005. Two NRC proposed rules of July 2026 would rewrite 35.75 and 35.2075 and remove 35.315, and a draft Revision 2 of Regulatory Guide 8.39 is out; section 35.2075 dates from 2002. (Second accuracy pass; 91 FR 43456, 91 FR 47042 and 91 FR 51749 read as raw text from federalregister.gov, and the eCFR part 35 source note) · The US variant, gaps and sources now describe the proposals and say the rule in force was unchanged on 17 November 2026; the year was corrected.
- 2026-11-17 · Release of patients after radionuclide therapy or a permanent implant: patient summary: The patient summary promised a small, safe dose to others and a stay in a single room. The sources speak of dose limits and constraints, and the single room is a US requirement that also allows sharing with another treated patient. (Second accuracy pass; IAEA SSG-46 and 10 CFR 35.315 read raw) · The summary now says a dose within set limits and a stay in hospital.
- 2026-11-17 · Radiation safety at home after radioactive iodine (I-131) for thyroid cancer: how long to avoid pregnancy: The page said the UK committee ARSAC advises avoiding pregnancy for 6 months after treatment. That figure came from a 2009 IAEA report quoting ARSAC's 2006 notes. The current ARSAC notes (July 2026) say 4 months for iodine-131 in all thyroid indications up to 6,000 MBq, with longer advice where the clinical picture warrants it. The 2022 UK patient leaflet says six months. (Second accuracy pass; ARSAC Notes for Guidance, July 2026, table 7.1, the open PDF from gov.uk read as raw text, and the UK patient leaflet read as raw text) · The step, summary and comparison now give each figure with who states it: 4 months (ICRP and current ARSAC), six months (2022 UK patient leaflet), 6 to 12 months in some practice. The words "at least" were removed and the 6,000 MBq limit added.
- 2026-11-17 · Radiation safety at home after radioactive iodine (I-131) for thyroid cancer: US rule on written instructions: The page said US rules require written instructions where anyone could receive more than 1 mSv. The regulation says instructions are required where the dose to another person is likely to exceed 1 mSv; "could exceed" applies only to its clause on a nursing infant or child. (Second accuracy pass; 10 CFR 35.75 read as raw text from the eCFR) · Corrected in the summary, the US variant and the grades table, where the sentence of the regulation is now quoted.
- 2026-11-17 · Radiation safety at home after radioactive iodine (I-131) for thyroid cancer: written instructions before leaving: The page said that in the UK the patient must have written instructions, and gave this as a step for everyone, citing one country's law. The regulations apply as respects Great Britain and say the employer's procedures must provide for written instructions where appropriate. (Second accuracy pass; IR(ME)R 2017 regulation 12 as amended in 2024, read as raw text from legislation.gov.uk, and IAEA SSG-46 para 4.248 read as raw text) · The step now rests on the IAEA Safety Guide, which says the patient should be given written instructions, and names the rule in Great Britain and the US rule beside it.
- 2026-11-17 · Radiation safety at home after radioactive iodine (I-131) for thyroid cancer: where the home advice comes from: The home precautions were taken from an IAEA example leaflet written for an overactive thyroid and presented for thyroid cancer with the limit mentioned only under gaps. The IAEA report applies that advice to cancer patients who have spent a few days in hospital and retain less than about 600 MBq, says patients released straight after treatment need additional advice, and says kissing and sex should be avoided for at least two days where more than 600 MBq is retained. (Second accuracy pass; IAEA Safety Reports Series No. 63, section 5.3 and Annex V, read as raw text) · The limit is now stated in the summary, the population and the gaps.
- 2026-11-17 · Radiation safety at home after radioactive iodine (I-131) for thyroid cancer: UK guidance and US practice: The page listed professional bodies as authors of the 2023 UK guidance when its leaflet says they endorsed it, said the authors found over-restriction widespread when the abstract says some patients are affected, and said most US patients are now treated as outpatients when the source is a survey of patient association members. (Second accuracy pass; the guidance abstract through Europe PMC, its patient leaflet read in full, and the Wu 2020 survey read in full on the journal's page) · Reworded to what each source states, with the survey's counts (956 of 1,102 respondents).
- 2026-11-17 · Radiation safety at home after radioactive iodine (I-131) for thyroid cancer: reasons given for the steps: Several reasons were not in the cited report: that thyroid cancer treatment can usually wait until after pregnancy (the report says often), that iodine in milk could destroy a baby's thyroid (the report says permanent hypothyroidism or a high risk of later thyroid cancer), that adults at home receive very small doses (the report says doses to families and carers rarely approach or exceed a few millisieverts), and four explanations about distance, nights, babies and nearby organs. The carer dose constraints of 5 mSv and 1 mSv were stated flatly where the IAEA guide calls them commonly used. (Second accuracy pass; IAEA SRS-63 and SSG-46 read as raw text) · Each reason was reworded to the report's own statement and the hedges restored.
- 2026-11-17 · Radiation safety at home after lutetium-177 PSMA or dotatate therapy: days of distance after Lutathera: The page said only the US Pluvicto label states how many days to keep a distance and sleep apart, and that the Lutathera label and the guidelines leave the numbers to the hospital. The EU Lutathera product information suggests 7 days of limited close contact (under 1 metre), under 15 minutes a day with children and pregnant women for 7 days, and a separate bedroom for 7 days (15 days from children and pregnant women). The NANETS/SNMMI 2019 standard gives 3 days as basic guidance. (Second accuracy pass; EMA product information for Lutathera (PDF read with pdftotext) and the NANETS/SNMMI 2019 procedure standard (PubMed Central full text)) · Added both sources, a step for Lutathera with each source named, and a variant setting the US label, the EU text and NANETS/SNMMI side by side.
- 2026-11-17 · Radiation safety at home after lutetium-177 PSMA or dotatate therapy: newer guideline for lutetium-177 dotatate: The page said no guideline newer than 2013 was found for lutetium-177 dotatate. The NANETS/SNMMI Procedure Standard (J Nucl Med 2019) is newer and gives home precautions, 6 months of contraception and breastfeeding advice. (Second accuracy pass; Europe PMC search, then the full text from PubMed Central) · Added the 2019 standard as a source and stated where it differs from the 2013 guidance.
- 2026-11-17 · Radiation safety at home after lutetium-177 PSMA or dotatate therapy: catheter for incontinence: The page gave catheterisation for 2 days as the single rule for an incontinent patient. That is the 2013 IAEA/EANM/SNMMI guidance; the NANETS/SNMMI 2019 standard says routine bladder catheterisation is not recommended and suggests disposable undergarments. (Second accuracy pass; both guidelines read in full) · The step now names the 2013 source and states the 2019 position; a variant sets out all three guidelines.
- 2026-11-17 · Radiation safety at home after lutetium-177 PSMA or dotatate therapy: breastfeeding after Lutathera: The step said not to breastfeed after Lutathera with no period. The US label says during treatment and for 2.5 months after the last dose; the EU product information says to avoid breast-feeding during treatment and gives no period afterwards. (Second accuracy pass; DailyMed label XML and EMA product information) · Breastfeeding has its own step with the 2.5 months and the source of each statement.
- 2026-11-17 · Radiation safety at home after lutetium-177 PSMA or dotatate therapy: written instructions: The step told every team to give written instructions, citing a US regulation that requires them only where another person is likely to receive more than 1 mSv. (Second accuracy pass; 10 CFR 35.75 from the eCFR) · The step is scoped to the United States and the 1 mSv condition, with the label and EU wording on explaining precautions before release.
- 2026-11-17 · Radiation safety at home after lutetium-177 PSMA or dotatate therapy: fluids: The summary attributed drinking more for 3 to 4 days to the EANM/SNMMI guideline and both US labels. Only the guideline gives 3 to 4 days; the Lutathera labels say before, on the day of and the day after treatment, and the EU Pluvicto leaflet says 2 days. (Second accuracy pass; guideline full text, DailyMed label XML and EMA product information) · Each period is now attributed to its own source, and the 6 to 10 hours of frequent voiding is stated.
- 2026-11-17 · Radiation safety at home after lutetium-177 PSMA or dotatate therapy: reasons given for the precautions: Six reasons were not in any cited source: that patients remember little on treatment day, that distance cuts the dose sharply, that nights account for much of a partner's dose, that soiled surfaces are the commonest route of exposure, that the staff dose is small, and that label contraception periods derive from half-life and genotoxicity. (Second accuracy pass; all five original sources re-read as raw text) · Each reason was replaced with what the cited passage says, or removed.
- 2026-11-17 · Radiation safety at home after lutetium-177 PSMA or dotatate therapy: ward and population wording: The page said a shielded ward where the guideline says a radioisotope ward, and scoped the page to adults although the US Lutathera label covers patients aged 12 years and older. The guideline's grading was described as strong or weak; it also has a moderate-strong level. (Second accuracy pass; EANM/SNMMI 2023 full text and the Lutathera label) · Wording corrected in the step, the variant, the population and the grading scheme.
- 2026-11-17 · Radiation precautions at home after permanent prostate seed brachytherapy: summary for patients: The page said almost all of the radiation is absorbed inside the patient. Prostate Cancer UK says most radiation is absorbed by the prostate and surrounding tissue. (Second accuracy pass; Prostate Cancer UK page (updated September 2026), fetched and read as raw text) · Reworded to "most", with the three months the source prints for when most radiation is released.
- 2026-11-17 · Radiation precautions at home after permanent prostate seed brachytherapy: UK figures: The two-month contact period and the condom advice were written as general rules, and the condom advice said "the first few ejaculations". They are UK patient advice from one charity, which says the first five ejaculations and does not name the isotope. (Second accuracy pass; Prostate Cancer UK page, read as raw text) · The summaries now name them as UK patient advice, give five, and say the reader's own hospital instructions come first.
- 2026-11-17 · Radiation precautions at home after permanent prostate seed brachytherapy: wallet card and airport alarms: The step ordered a wallet card and said it explains a radiation detector alarm at an airport. IAEA SSG-46 and ICRP 98 say a card "is useful", and ICRP 98 speaks of "certain types of security radiation monitors" where the card "may prove to be helpful"; neither mentions airports. (Second accuracy pass; IAEA SSG-46 para 5.261 (PDF text) and the ICRP 98 abstract (ICRP page)) · The step now tells the patient to inform the surgeon and calls the card useful, with the ICRP wording on security monitors.
- 2026-11-17 · Radiation precautions at home after permanent prostate seed brachytherapy: release check: The step said doses to the public and to carers must stay within limits. SSG-46 says public doses should not exceed the dose limits and should be unlikely to exceed dose constraints for the public and for carers; carers have constraints, not limits. Its reason was not in the source. (Second accuracy pass; IAEA SSG-46 paras 5.257 and 5.258, PDF text) · Step and reason reworded to the guide's terms.
- 2026-11-17 · Radiation precautions at home after permanent prostate seed brachytherapy: iodine-125 and palladium-103 precautions: One US centre's finding for a typical patient was written as an instruction, with the reason that palladium-103 decays faster, which the readable text does not state. The paper reports a lower measured dose rate for palladium-103, and gives 84 and 42 days at the median rate. (Second accuracy pass; Dauer 2010 author manuscript, full text through the NCBI E-utilities API) · Step reworded as what a typical patient should need, the centre's own periods come first, and the reason now gives the measured dose rates.
- 2026-11-17 · Radiation precautions at home after permanent prostate seed brachytherapy: reasons given for four steps: Four reasons said more than the sources: that being around others "is safe" (ICRP 98 says doses are well below 1 mSv a year in the vast majority of cases), that a passed seed "should be handled properly", that family should be told how cremation is handled, and that contraception still matters. (Second accuracy pass; ICRP 98 abstract and IAEA SSG-46, read as raw text) · Each reason now repeats what the source says, and the cremation step states the 12 month, 3 month and 20 month intervals with who gives each.
- 2026-11-17 · HDR brachytherapy: treatment-room safety and the stuck-source emergency: emergency procedure at the console: The page said the manufacturer's one-page emergency procedure is posted at the console under US rule 10 CFR 35.610. That rule requires the licensee's own written procedures at the console and does not mention the manufacturer. The manufacturer's single page comes from IAEA SSG-46 para 5.136, which says only that it is suitable for posting in an appropriate place. (Second accuracy pass; 10 CFR 35.610 from the eCFR API and the SSG-46 PDF, both read as raw text) · The two sources are now stated separately, in the summary, the step and the best in class statement.
- 2026-11-17 · HDR brachytherapy: treatment-room safety and the stuck-source emergency: US rules written as universal: Four steps (the list of daily checks, the procedure and telephone numbers at the console, the monitor check on entering the room and the yearly drill) were written as general rules. They are requirements of the US regulation 10 CFR Part 35. The monitor check on entry was also cited to the IAEA accident report, which does not state it. (Second accuracy pass; 10 CFR 35.610, 35.615 and 35.643 from the eCFR API, read as raw text) · Each step now says it is the US rule, the monitor check is its own step citing 35.615(c), and the population and gaps say no other country's regulation was read.
- 2026-11-17 · HDR brachytherapy: treatment-room safety and the stuck-source emergency: drills: The best in class statement cited IAEA SSG-46 para 5.316 for annual drills. The IAEA guide says staff should regularly participate in drills and gives no interval. Only the US rule says initially and at least annually. (Second accuracy pass; SSG-46 PDF read with pdftotext) · Drills are now given as regular under IAEA and at least annual under the US rule.
- 2026-11-17 · HDR brachytherapy: treatment-room safety and the stuck-source emergency: stuck-source sequence: The sequence left out steps the IAEA guide lists: observing the radiation levels in the room after entry, surveying the afterloader as well as the patient, the further survey after the patient is taken out, and informing the maintenance personnel, the radiation protection officer and the regulator. The reason given, that each step is quicker and safer than the next, is not in the source. (Second accuracy pass; SSG-46 para 5.136 read with pdftotext) · The step and the summary now carry the whole sequence in the guide's order, and the reason is the guide's own.
- 2026-11-17 · HDR brachytherapy: treatment-room safety and the stuck-source emergency: staffing variant: For medium and pulsed dose rate units the page said a physicist and a trained physician must be immediately available while treatment continues. The rule says the physicist and either an authorised user or a supervised individual trained to remove the source applicators, who need not be a physician. The page also called the authorised user a radiation oncologist and gave the rule a rationale its text does not state. (Second accuracy pass; 10 CFR 35.615(f) and 35.2 from the eCFR API, read as raw text) · The variant, the step and the summary now use the rule's terms, with its definition of an authorised user.
- 2026-11-17 · HDR brachytherapy: treatment-room safety and the stuck-source emergency: patient summary and population: The patient summary said the source is tiny, stays in for a few minutes, that a stuck source is rare and that staff watch throughout, and the population listed eight treatment sites. No source read gives the size, the time, a frequency or the sites, and the sources require the means to see and speak to the patient rather than continuous watching. (Second accuracy pass; all cited sources read as raw text) · Those words were removed or reworded to what SSG-46 para 5.35 and 10 CFR 35.615(d) say, and a gap says the sources give no frequency.
- 2026-11-17 · HDR brachytherapy: treatment-room safety and the stuck-source emergency: reasons and details not in the sources: Several reasons and details had no source: that each daily check is a safety barrier, that the emergency kit sits next to the patient (the guide puts the container close to the patient and the kit in the room), that everything is recorded after an event (the guide says the assessments), and that the evidence behind the rules is the record of accidents. (Second accuracy pass; SSG-46 paras 5.135 and 5.319 and 10 CFR 35.643, read as raw text) · Each was reworded to the source's own statement or removed.
- 2026-11-17 · Radiotherapy with a pacemaker or implanted defibrillator: planning scan step: The step told staff to leave a device more than 10 cm from a field edge out of the planning scan. The UK guidance says only that it is not necessary to include it, and that the scan should not be cut short if the area is needed. (Second accuracy pass; UK Radiotherapy Board guidance (January 2025), PDF from rcr.ac.uk read with pdftotext) · Reworded to say the device need not be included.
- 2026-11-17 · Radiotherapy with a pacemaker or implanted defibrillator: moving the device and sparing it: The step said do not compromise the plan and do not move the device below 5 Gy. The guidance says the plan should not be significantly compromised and that relocation is not recommended below 5 Gy. (Second accuracy pass; UK Radiotherapy Board guidance (January 2025), PDF from rcr.ac.uk read with pdftotext) · Reworded to the strength and scope the guidance gives.
- 2026-11-17 · Radiotherapy with a pacemaker or implanted defibrillator: flattening-filter-free beams: The page said the Japanese guideline raises the risk category by one level for flattening-filter-free beams. It says to consider raising it, depending on the patient's condition. (Second accuracy pass; JASTRO/JCS guideline, full text from Europe PMC (PMC7779359)) · Reworded in the best-in-class statement, the variant and the grades table.
- 2026-11-17 · Radiotherapy with a pacemaker or implanted defibrillator: monitoring of high-risk patients in Japan: The grades table said continuous monitoring at every session. The Japanese guideline says to monitor by electrocardiogram or pulse oximeter at every session and does not say continuous. (Second accuracy pass; JASTRO/JCS guideline, full text from Europe PMC (PMC7779359)) · Claim reworded to match.
- 2026-11-17 · Radiotherapy with a pacemaker or implanted defibrillator: monitoring quote: The quoted sentence carried two reference numbers fused into its words, printing radiotherapy2 and ICDs24. (Second accuracy pass; UK Radiotherapy Board guidance (January 2025), PDF from rcr.ac.uk read with pdftotext) · Replaced with two clean verbatim sentences and split into two steps, one on who is monitored and one on the extra trained person on the first day.
- 2026-11-17 · Radiotherapy with a pacemaker or implanted defibrillator: emergency treatment dose limit: The page gave one dose limit of 2 Gy for emergency treatment without device information. The guidance prints two, below 2 Gy in its executive summary and below 5 Gy in Appendix D, and Appendix D adds that a magnet must be placed over an ICD during the exposure. (Second accuracy pass; UK Radiotherapy Board guidance (January 2025), PDF from rcr.ac.uk read with pdftotext) · Both limits and the magnet rule are now stated, and the inconsistency is named in the gaps.
- 2026-11-17 · Radiotherapy with a pacemaker or implanted defibrillator: grading and other bodies' thresholds: The page said none of the guidance is graded because the evidence is case series and laboratory studies, and gave no European or American thresholds. The 2022 ESC cardio-oncology guideline grades its device recommendations Class I, Level C and sets 2 Gy for a pacemaker and 1 Gy for an ICD. The UK guidance tabulates the AAPM categories as below 2 Gy, 2 to 5 Gy and above 5 Gy or neutrons. (Second accuracy pass; ESC 2022 guideline, published PDF from a university repository read with pdftotext, and Appendix A of the UK guidance) · ESC added as a source with its grades and thresholds, AAPM thresholds added as tabulated by the UK guidance, and the unsourced reason removed.
- 2026-11-17 · Radiotherapy with a pacemaker or implanted defibrillator: UK risk categories: The page did not give the UK low, medium and high risk categories because they are printed only as a flowchart image. (Second accuracy pass; Figure 1 of the UK guidance, page rendered and read by eye) · Categories and what each requires added to the summary, marked as read by eye and not quoted.
- 2026-11-17 · Radiotherapy with a pacemaker or implanted defibrillator: reasons given for steps: Several reasons said more than the sources: that higher energies scramble the device's memory, that total dose is what raises the chance of a fault, that the risk is small, and that faults are rare. The guidance says malfunctions are mainly software faults that may lead to a reset, that they correlate only weakly with dose, and that relatively few events have been reported. (Second accuracy pass; UK Radiotherapy Board guidance (January 2025), PDF from rcr.ac.uk read with pdftotext) · Each reason reworded to what the guidance states.
- 2026-11-17 · Radiotherapy with a pacemaker or implanted defibrillator: Italian classification: The variant gave device dose above 10 Gy as the Italian high-risk rule. The classification as tabulated has further high-risk criteria that combine pacing dependency with dose or energy, and for an ICD low risk also requires no frequent ICD interventions. (Second accuracy pass; Appendix A of the UK guidance and Table 1 of the JASTRO/JCS guideline) · Variant completed.
- 2026-11-17 · Stereotactic radiosurgery and SBRT: planning and delivery safeguards: staff at each session: The page said at least two trained therapists must be at every session and that a physicist directly supervises every later session. The guideline asks for two certified therapists of whom at least one is trained in the technique, requires the physicist in the room for the first session, and says later sessions should have a physicist on site and immediately available. (Second accuracy pass; AAPM-RSS MPPG 9.b (2025), full text read as Europe PMC XML) · Staffing reworded to the guideline's must and should.
- 2026-11-17 · Stereotactic radiosurgery and SBRT: planning and delivery safeguards: daily isocentre test: The page gave 1 mm for SRS and 1.5 mm for SBRT as a plain isocentre limit. The guideline gives it as a combined limit for isocentre coincidence and table repositioning, reported from MPPG 2.b, and its own table sets 2 mm for image-guided positioning in SBRT. MPPG 2.b's table prints the combined 1 mm test for SRS only. (Second accuracy pass; AAPM-RSS MPPG 9.b (2025), full text read as Europe PMC XML; MPPG 2.b (2021) full text read the same way) · Step and summary now give the combined limit with its attribution and the table figures, and a gap records the mismatch with MPPG 2.b.
- 2026-11-17 · Stereotactic radiosurgery and SBRT: planning and delivery safeguards: planning steps: Several planning points were written as fixed rules: obtaining earlier plans, a cumulative dose report, a 2 mm grid, a 1 mm grid for small targets and the oncologist's signature. The guideline says should for each, and says a 1 mm grid may be necessary for targets of 1 cc or less or beside critical organs. (Second accuracy pass; AAPM-RSS MPPG 9.b (2025), full text read as Europe PMC XML) · Summary and steps reworded to should and may, with the 1 cc threshold.
- 2026-11-17 · Stereotactic radiosurgery and SBRT: planning and delivery safeguards: margin trial: The page said a 1 mm margin gave less radionecrosis than 3 mm. The abstract reports radionecrosis in 6 lesions, 5 of them in the 3 mm group, with P=.10, low rates in both groups, and a conclusion resting on a higher potential risk. (Second accuracy pass; Kirkpatrick 2015 abstract read from Europe PMC) · The summary now gives the counts and P values as printed and the authors' judgement as theirs.
- 2026-11-17 · Stereotactic radiosurgery and SBRT: planning and delivery safeguards: reasons given for each step: Nine of the eleven reasons were not in the guideline, among them that position must repeat to within a millimetre or two, that a few millimetres off can miss the tumour, and that the second check is the main protection against a planning error. (Second accuracy pass; AAPM-RSS MPPG 9.b (2025), full text read as Europe PMC XML) · Each reason rewritten from what the guideline says at that point.
- 2026-11-17 · Stereotactic radiosurgery and SBRT: planning and delivery safeguards: who it covers: The page limited the protocol to adults, listed body sites, and said Gamma Knife and particle therapy have their own physics guidance. The guideline states no age range or site list and says only that it does not address gamma-ray or particle beam systems. (Second accuracy pass; AAPM-RSS MPPG 9.b (2025), full text read as Europe PMC XML) · Population reworded to the guideline's scope, including that it covers SBRT only on ring-mounted linacs.
- 2026-11-17 · Stereotactic radiosurgery and SBRT: planning and delivery safeguards: frame-based and frameless radiosurgery: The page said a head frame is fixed to the skull under local anaesthetic on the day of treatment and that the guideline treats both techniques as acceptable when commissioned and verified. No source read says either. (Second accuracy pass; AAPM-RSS MPPG 9.b (2025), full text read as Europe PMC XML) · The variant now says only what the guideline says about each technique.
- 2026-11-17 · Stereotactic radiosurgery and SBRT: planning and delivery safeguards: UK guidance: The page said UK SABR Consortium guidance could not be found yet covers the same ground. Version 6.1 (January 2019) is linked from the consortium's site and differs in places: the final dose grid must be 2 mm or finer, patient-specific QA measurements must be made for at least the first 10 patients, and its peripheral lung schedules are 18 Gy x 3, or 12 Gy x 5 or 11 Gy x 5 by the chest wall. (Second accuracy pass; UK SABR Consortium resource version 6.1, PDF read with pdftotext) · Added as a source, a variant and two graded claims, kept apart from the United States and ESTRO figures.
- 2026-11-17 · Stereotactic radiosurgery and SBRT: planning and delivery safeguards: ACR-ASTRO practice parameter: The source list dated the ACR-ASTRO SBRT practice parameter 2024 and marked it unreachable. The revision that could be confirmed is that of 2019, published in the American Journal of Clinical Oncology in 2020. (Second accuracy pass; Europe PMC record and abstract, and the reference list of MPPG 9.b) · Year, DOI and read status corrected; a gap says a later revision may exist.
- 2026-11-17 · Blood sampling from a central line or port: why hands are cleaned: The page said hands are the commonest route by which germs reach a line. CDC lists contact with hands as one of four recognised routes and names migration of skin organisms at the insertion site as the most common for short-term catheters. (Second accuracy pass; CDC/HICPAC 2011 guideline, PDF read as raw text) · The reason now says what CDC says: hands are one recognised route, and hand hygiene with aseptic technique protects against infection.
- 2026-11-17 · Blood sampling from a central line or port: status of the IDSA guideline: The page quoted the 2009 IDSA guideline without saying that IDSA now marks it as archived, and did not mention the newer German guideline for people with cancer (AGIHO/DGHO 2020). (Second accuracy pass; IDSA guideline page and the AGIHO/DGHO 2020 full text from Europe PMC, both read as raw text) · The summary, best-in-class statement and gaps now give the archived status. The AGIHO/DGHO guideline is added with its graded recommendations, including one that differs from IDSA on sampling every lumen.
- 2026-11-17 · Blood sampling from a central line or port: flushing after a sample: The step said to flush with 0.9% saline and that blood left in the line can feed germs. The ESMO sentence it quotes says saline with no strength, and no source read gives that reason. (Second accuracy pass; the ESMO guideline could not be re-read on 17 November 2026, so the stored quote was used, with the ASCO 2013 abstract and the Cancer Research UK PICC page read as raw text) · The strength was removed, the reason now follows Cancer Research UK (to clean the line and prevent clotting), and the gaps say that ESMO could not be re-read.
- 2026-11-17 · Blood sampling from a central line or port: discard volume: The page said local policies commonly discard a few millilitres and that the INS Standards set a discard volume. No source read says the first, and the INS Standards were never read. The summary also left out that the 2 ml study found agreement for every blood count except the white cell count. (Second accuracy pass; Pushparaj 2026 read in full from Europe PMC) · The page now reports what the study authors say (discarding is the most used technique and there is no agreed volume), adds the white cell exception and the study's limits, and no longer says what the unread INS Standards contain.
- 2026-11-17 · Blood sampling from a central line or port: reasons and scope not in the sources: Several lines said more than the sources: that the connector is a main route for germs, that tests should be grouped, that avoiding the line trades against needle sticks in people with poor veins, that children need their own local rules, and that the access-frequency studies were done in intensive care. The review covers hospital settings and names set sampling times. (Second accuracy pass; CDC 2011 full text and the Davis 2026 abstract from Europe PMC, read as raw text) · Each line was reworded to its source or removed, and the gaps name what no source addresses. The IDSA grading scheme is now named correctly.
- 2026-11-17 · Central line insertion bundle: preventing bloodstream infection at the time of placement: choice of insertion site: The page said the subclavian site is preferred "for non-tunnelled ICU lines", merging two recommendations. SHEA 2022 prefers it when the line is placed in intensive care (high quality evidence); CDC 2011 prefers it for non-tunnelled lines in adults (Category IB). (Second accuracy pass; SHEA 2022 full text (Europe PMC XML) and CDC 2011 PDF text) · Each recommendation is now stated with its own scope and grade, in the step, the summary and the best-in-class statement.
- 2026-11-17 · Central line insertion bundle: preventing bloodstream infection at the time of placement: choice of device and the short summary: The page told patients that for long treatments such as chemotherapy a tunnelled line or a port is usually the right kind of line. epic3 recommends those devices where long-term access is required (Class A), recommends a PICC for medium-term intermittent access (Class D/GPP), and does not name chemotherapy in either recommendation. (Second accuracy pass; epic3 full text (Europe PMC XML)) · The short summary now says one guideline recommends a tunnelled line or port for long-term use; the PICC recommendation was added and the unsourced reason removed.
- 2026-11-17 · Central line insertion bundle: preventing bloodstream infection at the time of placement: antibiotics at insertion: The step said not to give preventive antibiotics and that they do not lower line infections. The guidelines say not to give them routinely; epic3 reports no benefit shown in adults, notes a review that found possible benefit in some high-risk people with cancer, and advises against routine use to limit resistance. (Second accuracy pass; epic3 full text (Europe PMC XML) and CDC 2011 PDF text) · "Routinely" restored and the reason reworded to what epic3 reports.
- 2026-11-17 · Central line insertion bundle: preventing bloodstream infection at the time of placement: chlorhexidine in young infants: The page said chlorhexidine skin preparation in infants under 2 months is weighed case by case. CDC 2011 makes no recommendation for infants under 2 months (an unresolved issue); it is SHEA 2022 that says to use it in neonatal intensive care when benefits are judged to outweigh the potential risk. (Second accuracy pass; CDC 2011 PDF text and SHEA 2022 full text) · Reworded to state each source's position.
- 2026-11-17 · Central line insertion bundle: preventing bloodstream infection at the time of placement: ultrasound, checklist and emergency lines: The ultrasound step gave CDC's condition and reason under a SHEA citation and left out SHEA's warning that ultrasound may lead to a breach in aseptic technique. A checklist was written as required where SHEA asks for a process such as a checklist. The emergency step left out "as soon as possible" before "within 48 hours", and "dry completely" went beyond SHEA's "allowed to dry". (Second accuracy pass; SHEA 2022 full text and CDC 2011 PDF text) · Each step reworded to its source, with CDC and SHEA attributed separately.
- 2026-11-17 · Central line insertion bundle: preventing bloodstream infection at the time of placement: reasons given for the steps: Nine reasons were not in the cited sources, for example that hands carry the germs causing most early line infections, that chlorhexidine works only once dry, and that a second person catches steps a busy inserter misses. (Second accuracy pass; SHEA 2022, CDC 2011 and epic3 full texts) · Each reason replaced with what the cited guideline states.
- 2026-11-17 · Central line insertion bundle: preventing bloodstream infection at the time of placement: impregnated catheters and shared practices: The page said CDC recommends impregnated catheters "under the same conditions" as SHEA and that CDC and epic3 broadly share the checklist, observer, kit and ultrasound recommendations. CDC and epic3 set a condition of more than 5 days with a persistently high rate; SHEA sets three different conditions; the checklist, observer and kit are in SHEA alone, and epic3 does not recommend ultrasound. (Second accuracy pass; SHEA 2022, CDC 2011 and epic3 full texts) · The variant and the best-in-class statement now say which source recommends what.
- 2026-11-17 · Central line maintenance bundle: how often dressings are changed: The page gave one rule for every central line: transparent dressings at least every 7 days and gauze every 2 days. SHEA writes that rule for non-tunnelled lines. For tunnelled lines and ports, epic3 and NICE say every 7 days until the site has healed, epic3 adds that a dressing may then no longer be required, and CDC makes no recommendation on dressing a well-healed tunnelled site. NICE, for care at home, says gauze is changed every 24 hours. (Second accuracy pass; SHEA 2022 and epic3 (Europe PMC full text), CDC/HICPAC 2011 (PDF) and NICE CG139 (HTML), all read as raw text) · The step is scoped to non-tunnelled lines, the tunnelled and port rule and the NICE gauze interval are stated beside it, a NICE variant and an epic3 grade were added, and the patient summary says at least weekly while a dressing is needed.
- 2026-11-17 · Central line maintenance bundle: chlorhexidine dressings: The best-in-class statement listed chlorhexidine dressings in adults, while the step and SHEA say over 2 months of age. The summary and best-in-class statement also gave CDC's adult recommendation without its scope: CDC recommends them for short-term non-tunnelled lines. SHEA's note that the benefit is unclear on well-healed sites of long-term lines was missing. (Second accuracy pass; SHEA 2022 full text and the CDC/HICPAC 2017 dressing update, read as raw text) · The age is stated as SHEA prints it, CDC's scope is given wherever its recommendation appears, and SHEA's caveat is in the step.
- 2026-11-17 · Central line maintenance bundle: administration set intervals: The page gave 96 hours (epic3) to 7 days (SHEA) for continuous sets and left out NICE, its source for care at home, which says not more often than 72-hour intervals. The 12-hour blood set rule carried no grade; epic3 prints Class D/GPP and adds whichever is sooner. Two reasons given, that extra changes add handling and that blood and lipids support germ growth, are not in the sources read. (Second accuracy pass; NICE CG139 recommendations 1.4.4.12 to 1.4.4.14 and epic3 IVAD37 to IVAD39, read as raw text) · All three intervals are on the page with who says which, the epic3 grades are stated, and the reason now quotes the Cochrane finding epic3 reports.
- 2026-11-17 · Central line maintenance bundle: who preventive lock therapy is for: The page said antimicrobial locks are for patients with long-term lines and recurrent infection. SHEA names three groups: long-term haemodialysis catheters with recurrent CLABSI, limited venous access with recurrent CLABSI, and patients at heightened risk of severe consequences from a CLABSI. (Second accuracy pass; SHEA 2022 full text, Europe PMC) · The step, summary and grades give SHEA's groups, with epic3's position that locks are not routine.
- 2026-11-17 · Central line maintenance bundle: reasons given for eleven steps: Eleven reasons said more than the sources: that most line infections come from handling, that hands carry the germs that most often cause them, that swimming pools carry germs into the line, that preventive antibiotics breed resistance, that routine replacement adds insertion risks, and others like them. None of these sentences is in the source cited. The shower rule was also written as cover only, where CDC gives an impermeable cover as an example of a precaution. (Second accuracy pass; SHEA 2022, CDC/HICPAC 2011, epic3 and NICE CG139, all read as raw text) · Each reason now says what the cited source says and who says it.
- 2026-11-17 · Central line maintenance bundle: ethanol locks and children's guidance: The page said some children's centres fill the line with 70% ethanol between uses, with no source, and called the 2023 meta-analysis mostly before-and-after studies, which its abstract does not say. A gap said the Children's Cancer and Leukaemia Group guidance could not be found at a public address; the group publishes a community care page on central lines. (Second accuracy pass; Wolf 2018 and the 2023 meta-analysis (abstracts, Europe PMC) and the CCLG page read as raw HTML) · The variant now describes the trial regimen and the meta-analysis as its abstract does, and the gap is corrected.
- 2026-11-17 · Blocked central line or port: who gets the smaller alteplase dose: The page said children are dosed by lumen volume. The US label and the UK SmPC set the rule by weight, not age: under 30 kg the dose is 110% of the catheter's internal lumen volume, not exceeding 2 mg in 2 ml, and anyone of 30 kg or more gets 2 mg in 2 ml. (Second accuracy pass; Cathflo Activase US label (DailyMed label XML) and Actilyse Cathflo UK SmPC (emc HTML), read as raw text) · Population and dose step reworded to the under 30 kg rule, with the 3 ml aspiration volume for patients under 10 kg added.
- 2026-11-17 · Blocked central line or port: what to do when two doses fail: The page told the team to get imaging of the catheter after two failed doses and said the line was then likely misplaced, damaged or sheathed. No source read says either. The UK SmPC says to consider catheter replacement, and the US label says more than two doses have not been studied. (Second accuracy pass; Actilyse Cathflo UK SmPC (emc HTML) and Cathflo Activase US label, read as raw text) · Step now cites the UK SmPC sentence on considering catheter replacement. The imaging advice was removed and named in the gaps.
- 2026-11-17 · Blocked central line or port: causes to check first: The page listed a closed clamp among the causes and said many blocked lines are not clotted. Neither is in a source read. The label lists malposition, mechanical failure, constriction by a suture, and lipid deposits or drug precipitates. (Second accuracy pass; Cathflo Activase US label (DailyMed label XML), read as raw text) · Clamp and the unsourced proportion removed. A first step now quotes the label's own list of causes to consider before treatment.
- 2026-11-17 · Blocked central line or port: why force is avoided: The page said force can push a clot into the lungs and worded the warning as never forcing a flush. The label says vigorous suction should not be applied when testing the line, and excessive pressure when instilling could rupture the catheter or expel the clot into the circulation. (Second accuracy pass; Cathflo Activase US label (DailyMed label XML), read as raw text) · Step reworded to the label's two warnings and their stated risks.
- 2026-11-17 · Blocked central line or port: suspected clot in the vein: The page said ESMO recommends Doppler ultrasound when the arm, neck or face is swollen or painful, and that a vein clot needs anticoagulant treatment. The ESMO sentence on record says Doppler ultrasound is usually carried out, gives no symptom list, and could not be re-read. Patients were told to report a swollen face, which no source read lists. (Second accuracy pass; ESMO 2015 not readable on 17 November 2026 (publisher 403), Macmillan PICC lines page and both alteplase labels read as raw text) · Step reworded to the recorded sentence and dated. Patient advice now uses Macmillan's list (arm, hand, chest area or neck). The reason now cites the labels' caution about venous thrombosis near the catheter.
- 2026-11-17 · Blocked central line or port: flushing as prevention: The page called saline flushing the main prevention and said leftover blood or drug is the commonest cause of blockage. The recorded ESMO sentence only recommends routine flushing after any infusion or blood sampling. (Second accuracy pass; ESMO 2015 as recorded on 11 November 2026 (not readable on 17 November 2026) and Macmillan PICC lines page read as raw text) · Both unsourced claims removed. The step keeps the ESMO sentence and grade, dated.
- 2026-11-17 · Blocked central line or port: urokinase variant: The page said urokinase is used in countries where alteplase for catheters is not licensed. No source read says so, and the UK SmPC shows alteplase 2 mg is licensed for blocked central lines in the UK. (Second accuracy pass; Cochrane review full text (PubMed Central HTML) and Actilyse Cathflo UK SmPC (emc HTML), read as raw text) · Licensing claim removed. A UK product variant was added, and the Cochrane head-to-head result (alteplase favoured over urokinase 10,000 IU, very low quality) was added.
- 2026-11-17 · Blocked central line or port: types of blockage and later reviews: The page called a line that takes fluid but gives no blood a partial occlusion. The Cochrane review calls that a withdrawal occlusion and defines partial as reduced flow. The page also said no later systematic review existed. A 2019 meta-analysis in children with cancer and a 2025 umbrella review exist. (Second accuracy pass; Cochrane review full text (PubMed Central HTML) and Europe PMC search, read as raw text) · Definitions reworded to Cochrane's. Both later reviews named in the gaps as read in abstract only.
- 2026-11-17 · Central line and port removal: the removal technique: The page said no guideline that could be read describes how a line is removed, and gave the technique as uncited common practice, including breath-holding. The Association of Anaesthetists 2016 guideline is open access and states it: lie flat with the exit site below the heart, firm digital pressure for at least 5 minutes, then an occlusive dressing. Breath-holding is in no source read; a 2016 review gives a Valsalva manoeuvre or removal during active expiration. (Second accuracy pass; Association of Anaesthetists 2016 and McCarthy 2016, full text from Europe PMC) · The practice step was replaced by two sourced steps with the 5 minute pressure time, and the Valsalva advice is attributed to the review.
- 2026-11-17 · Central line and port removal: anticoagulation before removing a line with a clot: The page gave 3 to 5 days of anticoagulation before removal, grade [I, A], from the ESMO 2015 guideline, which could not be re-read (the publisher refuses a script and there is no open copy). No source read today gives that duration. (Second accuracy pass; ESMO 2015 not readable; Association of Anaesthetists 2025 read in full from Europe PMC) · The duration and grade were taken off the page and named in the gaps. The step now gives the 2025 statement that a line which is still required, working, well placed and not infected can stay while the person is anticoagulated.
- 2026-11-17 · Central line and port removal: infections that need the line out: The list of infections and the 48 to 72 hour figure rested on the unreadable ESMO 2015 text, with an unsourced reason. IDSA 2009 lists the same situations for long-term lines with more than 72 hours of effective antimicrobials, adds Pseudomonas aeruginosa, and grades them A-II. The AGIHO/DGHO 2020 guideline for people with cancer was not on the page. (Second accuracy pass; IDSA 2009 and AGIHO/DGHO 2020, full text) · Steps re-sourced to IDSA recommendations 32 and 44 with verbatim quotes, an AGIHO/DGHO step and variant added, and IDSA marked as archived.
- 2026-11-17 · Central line and port removal: what to report afterwards and how long to watch: The page told people to report chest pain and dizziness and said a review supports watching for at least two hours. No source read gives patients a symptom list, and the 2026 review sets no observation time; it reports neurological signs in 92 per cent and low oxygen in 90 per cent, with 32 per cent of cases starting more than two hours after removal. (Second accuracy pass; Pacini 2026, abstract from Europe PMC) · Wording changed to the signs the sources describe, the two hour claim removed, and the gaps now say that no source read gives a symptom list, an observation time or how long the dressing stays on.
- 2026-11-17 · Central line and port removal: suspected air embolism: The position and oxygen advice rested on the unreadable ESMO text and gave an unsourced reason. A 2016 review supports 100 per cent oxygen and the left-side head-down position for air in the veins, and says a patient with air in the arteries should be kept flat. (Second accuracy pass; McCarthy 2016 and Association of Anaesthetists 2016, full text from Europe PMC) · Step re-sourced, the reason replaced with the review's, and the caution about air in the arteries added.
- 2026-11-17 · Central line and port removal: reasons given for steps and scope: The page said a line should be reviewed at every contact, that an unneeded line adds a risk of clots, and that routine tip cultures are often contaminated; the CDC and IDSA passages cited say none of these. It also said non-tunnelled and dialysis lines have their own protocols, which do not exist. (Second accuracy pass; CDC 2011 and IDSA 2009, full text) · Each reason was reworded to what the cited passage says, and the scope sentence now says those lines are not covered.
- 2026-11-17 · Central venous catheter insertion for anticancer treatment: avoiding the femoral vein: The step telling staff to avoid the femoral vein cited the ITAC thrombosis guideline and quoted a sentence that does not mention the femoral vein. The reason given (more infection and clot risk in the groin) was not in that source either. CDC says it, for adults, as Category IA, and gives higher colonisation and a higher risk of deep vein thrombosis as the reasons. (Second accuracy pass; CDC/HICPAC 2011 (PDF text) and ITAC 2022 (Europe PMC full text), read raw) · Split into two steps. The femoral step now cites CDC with its own sentence and grade; the ITAC step keeps only what ITAC says (right side, jugular vein, tip at the junction of the superior vena cava and right atrium, grade 1B).
- 2026-11-17 · Central venous catheter insertion for anticancer treatment: ECG tip confirmation and NICE: The page said NICE supports ECG-based tip confirmation for PICCs and that it avoids a chest X-ray. NICE's guidance is about one named system (Sherlock 3CG), says it should be considered as an option, and says it usually avoids a confirmatory X-ray in patients who would otherwise have blind insertion, with cost savings subject to some uncertainty. (Second accuracy pass; NICE HTG368 recommendations page, read raw) · Reworded in the summary, the best-in-class statement and the variant to NICE's own strength and scope.
- 2026-11-17 · Central venous catheter insertion for anticancer treatment: where the guidelines differ: The page presented the right jugular vein and X-ray tip confirmation as settled. CDC prefers the subclavian site for non-tunnelled lines in adults (IB) and calls the site for tunnelled lines unresolved, SHEA prefers the subclavian site in intensive care, and the GAVeCeLT port bundle prefers the axillary vein. A 2025 UK consensus guideline, not on the page, asks for real-time imaging or ECG guidance for long-term catheters and calls fluoroscopy the gold standard for imaging, while GAVeCeLT calls fluoroscopy inaccurate and unsafe. (Second accuracy pass; CDC 2011, SHEA 2022, SIC-Port 2026 and the Association of Anaesthetists 2025 guideline, each read as full text) · Best-in-class statement and summary rewritten to give each body's position with its grade. The 2025 UK guideline added as a source, with a step on tip position and a step on clotting tests and platelet counts from its table. The GAVeCeLT PICC bundle, unreadable on 11 November, now reads and is added as a variant.
- 2026-11-17 · Central venous catheter insertion for anticancer treatment: figures resting on ESMO 2015: Two steps gave figures from the ESMO 2015 guideline alone: observations every four hours after insertion, and 100% oxygen with left-side, head-down positioning for suspected air embolism, written as an order. ESMO 2015 could not be re-read (publisher 403, no open copy). The UK anaesthetic guideline of 2016, which was read, says only that a head-down, left lateral tilt and aspiration of air through the line may be useful. (Second accuracy pass; ESMO 2015 not readable on 17 November 2026; AAGBI 2016 read as Europe PMC full text) · Both figures taken out of the steps and named in gaps. The air embolism step now cites the 2016 UK guideline and says to consider the position. Other ESMO statements are kept and marked as recorded on 11 November and not re-read.
- 2026-11-17 · Central venous catheter insertion for anticancer treatment: reasons given for steps: Several reasons said more than the sources: that each extra lumen is a route for clots, that preventive antibiotics encourage resistance, that ultrasound means fewer punctures of arteries or lung, that most early line infections come from skin germs carried in at insertion, that a checklist makes each step much more likely, and that head-down positioning keeps air from blocking the outflow of the heart. None is in the cited text. An ultrasound survey of the veins was attributed to an ESMO sentence that does not mention it. (Second accuracy pass; CDC 2011, SHEA 2022, SIC-Port 2026 and Association of Anaesthetists 2025, read raw) · Each reason reworded to what a source read states, or removed. The pre-procedure assessment step now cites the GAVeCeLT port bundle, which does describe the ultrasound survey.
- 2026-11-17 · Central venous catheter insertion for anticancer treatment: the patient summary: The summary for patients said the line is put in by a doctor, nurse or radiologist under local anaesthetic and told patients to say if they have chest pain. No cited source named a radiologist or stated the anaesthetic for all three devices, and the chest pain sentence came from an instruction to staff about when to order a chest X-ray. Macmillan's PICC page tells patients to contact their hospital doctor or nurse for shortness of breath or tightness in the chest. (Second accuracy pass; Macmillan PICC lines page, read raw; ESMO 2015 not readable) · Summary reworded to doctor or nurse, the anaesthetic claim removed, and the warning matched to Macmillan, which is added as a source. Gaps now say that no source read gives a time limit or a symptom list for ports and tunnelled lines.
- 2026-11-17 · Central venous catheter insertion for anticancer treatment: scope and evidence statements: The page said much of SHEA's evidence is from intensive care and that the GAVeCeLT port bundle has not been tested in a trial. SHEA says attention has focused on intensive care although most line infections occur outside it or in outpatients; the GAVeCeLT paper is an editorial that reports no trial and does not make that statement. (Second accuracy pass; SHEA 2022 and SIC-Port 2026 full texts, read raw) · Both reworded to what the papers say. SHEA's grading scheme line now includes its class of approaches that should not be routine.
- 2026-11-17 · Central line and port care in the community and at home: flush and lock: The step said to flush and lock with saline unless the device maker requires otherwise. NICE CG139 says saline should preferably be used, and separately that implanted ports or open-ended catheter lumens should be flushed and locked with heparin sodium flush solutions when the manufacturer recommends it. (Second accuracy pass; NICE CG139 recommendations 1.4.4.6 and 1.4.4.7, raw HTML read as text) · The step now states both recommendations, and says that neither NICE document gives a flush volume or frequency.
- 2026-11-17 · Central line and port care in the community and at home: dressings on tunnelled lines and ports: The page gave only the 7-day dressing change for central lines. For tunnelled or implanted devices NICE CG139 says dressings should be replaced every 7 days until the insertion site has healed. The skin clean with chlorhexidine in 70% alcohol at dressing changes was also missing. (Second accuracy pass; NICE CG139 recommendations 1.4.3.6 and 1.4.3.8, raw HTML read as text) · Two steps added, each with the verbatim recommendation and its year tag.
- 2026-11-17 · Central line and port care in the community and at home: who the guidance covers: The page scoped the guidance to adults with cancer and called CG139 the UK guideline covering England. CG139 covers children, young people and adults receiving NHS care in primary and community care and does not mention cancer. NICE says its quality standards apply in England and Wales. (Second accuracy pass; NICE CG139 and QS61 overview pages, raw HTML read as text) · Population, summary and best-in-class statement reworded to the scope the pages state.
- 2026-11-17 · Central line and port care in the community and at home: comparison with other guidelines: The page said the clinical steps match the CDC, epic3 and ESMO guidelines and that the 15-second hub clean in QS61 comes from epic3. No source cited on the page makes that comparison. QS61 says only that its list of procedures is adapted from CG139 and epic3. (Second accuracy pass; NICE QS61 statement 5, raw HTML read as text) · Both statements replaced with what QS61 says, and a gap added.
- 2026-11-17 · Central line and port care in the community and at home: reasons given for each step: Nine steps gave reasons that neither NICE document states, such as hands being the commonest route of infection, saline avoiding the risks of heparin, and community nurses seeing central lines only occasionally. The patient summaries also said a clean no-touch technique where NICE requires an aseptic technique, and promised teaching on who to call, which NICE does not mention. (Second accuracy pass; NICE CG139 section 1.4 and QS61 statement 5, raw HTML read as text) · Each reason now says only what NICE says, including its definition of aseptic technique and the QS61 rationale. The summaries were reworded.
- 2026-11-17 · Central line and port care in the community and at home: written protocol: The step said to audit against the written protocol. QS61 does not use the word. It suggests measuring evidence of a written protocol, the proportion of people for whom all specified procedures were completed, and the incidence of device-related bloodstream infection. (Second accuracy pass; NICE QS61 statement 5 quality measures, raw HTML read as text) · Step reworded to the three measures QS61 lists.
- 2026-11-17 · Implanted port access, flushing and care in adults having cancer treatment: needle change: The page said no national guideline read states how often the needle of an accessed port is changed. CDC/HICPAC 2011 addresses it directly: it makes no recommendation on how often the needle is replaced or how long it can stay in, listing both as unresolved issues. (Second accuracy pass; CDC/HICPAC 2011 PDF read as raw text) · Both CDC statements are now quoted with the grade Unresolved issue, and the summary, variant, best-in-class statement and gaps say so.
- 2026-11-17 · Implanted port access, flushing and care in adults having cancer treatment: dressing while accessed: The page gave a 7-day dressing change for an accessed port as NICE 1.4.3.4 and CDC Category II for implanted sites. The CDC rule, NICE 1.4.3.6 and epic3 IVAD22 concern the healing wound after a port is put in, and NICE 1.4.3.4 is written for central line insertion sites in general. No guideline read names the needle site of an accessed port. (Second accuracy pass; NICE CG139 recommendations page, CDC/HICPAC 2011 and epic3 read as raw text) · The step keeps the NICE rule and states its scope, the port-specific healing-wound rules are quoted in the evidence table, and a gap says no guideline covers the needle site.
- 2026-11-17 · Implanted port access, flushing and care in adults having cancer treatment: heparin or saline lock: The summary said the evidence favours saline. The Cochrane review found fewer occlusions with heparin on low-certainty evidence and concluded it was uncertain whether heparin results in fewer occlusions. The page also said a trial found saline non-inferior, which the abstract as read does not state. (Second accuracy pass; Cochrane CD008462.pub4 and Goossens 2013 abstracts read from Europe PMC) · Reworded to say heparin has not been shown to be clearly better. The trial is now reported by its rates (3.70% and 3.92% of accesses) and its authors conclusion.
- 2026-11-17 · Implanted port access, flushing and care in adults having cancer treatment: lock step: The step said to use heparin only where the maker or local policy requires it and that heparin can rarely cause bleeding. NICE says saline preferably and heparin when the manufacturer recommends it, and no source read gives a frequency for bleeding. (Second accuracy pass; NICE CG139 and epic3 read as raw text) · The step now follows the NICE wording and gives the side effects epic3 lists.
- 2026-11-17 · Implanted port access, flushing and care in adults having cancer treatment: heparin amount: The best-in-class statement gave an ESMO figure of more than 500 units of heparin. The ESMO guideline could not be re-read on 17 November 2026 and no other source read states that figure. (Second accuracy pass; ESMO 2015 not readable (publisher 403, no open copy)) · The figure was taken off the page and the gap names it. A new gap says no guideline read sets a heparin strength or volume for a port lock.
- 2026-11-17 · Implanted port access, flushing and care in adults having cancer treatment: masks and head turning: The page said some local policies require a mask on clinician and patient and the head turned away, and gave keeping droplets off the sterile field as the reason. Only one policy read requires both (Shared Health Manitoba, a paramedic procedure for all ages), it applies them from before access until the infusion is running, and it gives no reason. NHS Borders asks only the nurse for a face covering. (Second accuracy pass; Shared Health Manitoba P02.2 and NHS Borders protocol PDFs read as raw text, and epic3) · The variant now gives each policy with its exact scope and quotes, and adds what epic3 says about the lack of studies of masks in routine ward procedures.
- 2026-11-17 · Implanted port access, flushing and care in adults having cancer treatment: needle removal: The needle removal step was cited to NICE with a sentence about cleaning the hub, and gave preventing needlestick injury as the reason. NICE does not describe removing a port needle and no source read gives that reason. (Second accuracy pass; NICE CG139, NHS Borders and Manitoba texts read raw) · It is now a practice step naming the two local policies that describe it, and the NICE hub rule is in the evidence table.
- 2026-11-17 · Implanted port access, flushing and care in adults having cancer treatment: checks before access: Steps said not to use a port with signs of infection (cited to CDC), to wear sterile gloves (cited to NICE) and not to give vesicant treatment without blood return. CDC says to examine the site, NICE names no glove type, and the local rule on blood return covers all intravenous therapy. (Second accuracy pass; CDC/HICPAC 2011, NICE CG139, NHS Borders and Manitoba texts read raw) · Each step now says what its cited source says and names the local policy for the rest.
- 2026-11-17 · Implanted port access, flushing and care in adults having cancer treatment: reasons given for steps: Several reasons had no source: hands as the commonest route of infection, chlorhexidine working only once dry, an ordinary needle coring the septum, the connector as a main route, and infection in low white cells as an emergency. (Second accuracy pass; All cited sources read as raw text) · Each reason was replaced with what a cited source states, or removed.
- 2026-11-17 · Implanted port access, flushing and care in adults having cancer treatment: infection and blockage signs: The infection step ordered treatment to start, which the cited ESMO sentence does not, and the blockage step listed swelling of the face, which no source read lists. (Second accuracy pass; Macmillan and Cancer Research UK port pages and NHS Borders protocol read raw) · Signs now match Macmillan and NHS Borders, and the order to start treatment was removed.
- 2026-11-17 · Implanted port access, flushing and care in adults having cancer treatment: patient summary: The patient lines gave the port site as chest or arm, a flush about every four weeks and a list of warning signs with no source. Macmillan says every 4 to 6 weeks and lists the signs to report. (Second accuracy pass; Macmillan Implantable ports page read as raw text) · The patient lines now follow Macmillan, which is added as a source with Cancer Research UK.
- 2026-11-17 · Implanted port access, flushing and care in adults having cancer treatment: flushing interval studies: The page called the 2025 meta-analysis mostly observational, which its abstract does not say. (Second accuracy pass; Liu 2025 abstract read from Europe PMC) · Removed.
- 2026-11-17 · Midline catheters in cancer care: which drugs may go through a midline: The page said it had not been settled whether drugs that irritate or damage veins may be given through a midline and that no guideline read addressed it. The Association of Anaesthetists guideline of 2025 says blood flow at the tip of a midline is not sufficient for infusions of vesicants or irritants other than for short-term or emergency use. (Second accuracy pass; Association of Anaesthetists safe vascular access 2025, full text read as raw text from Europe PMC) · Added the guideline as a source and a step with its sentence quoted, and said that it is UK consensus without grades and not specific to cancer.
- 2026-11-17 · Midline catheters in cancer care: evidence in people having chemotherapy: The page said the only cancer-specific evidence was a study of continuous fluorouracil and showed nothing against midlines. An Italian prospective study of 555 people having chemotherapy found midlines failed more often than PICCs (5.89 against 1.40 failures per 100 patients per month) and concluded that a PICC must be preferred. (Second accuracy pass; Faccini et al., Journal of Vascular Access 2025, full text read as raw text from Europe PMC) · Added the study as a source and as a variant, and reworded the best-in-class statement.
- 2026-11-17 · Midline catheters in cancer care: bloodstream infection compared with a PICC: The page said flatly that a midline carries fewer bloodstream infections than a PICC and that thrombosis, failure, occlusion and phlebitis did not differ. The meta-analysis is mostly observational; in its three randomised trials the infection difference was not significant and PICCs had fewer failures and occlusions, and several secondary results changed when single studies were removed. (Second accuracy pass; Aljameie et al., Frontiers in Cardiovascular Medicine 2026, full text read as raw text from Europe PMC) · Removed the flat claim from the patient summary and stated the result with its limits.
- 2026-11-17 · Midline catheters in cancer care: palliative care evidence: The page presented a systematic review's complication rates of 0 to 2.46 per 1,000 catheter days as a finding about midlines and PICCs alike. Only two of the 17 studies included midlines, and the 2.46 figure comes from a study of PICCs. (Second accuracy pass; Gravdahl et al., Supportive Care in Cancer 2024, full text read as raw text from Europe PMC) · Stated how many studies included midlines and where the figure comes from.
- 2026-11-17 · Midline catheters in cancer care: reasons given for the care steps: Four steps gave reasons the CDC guideline does not state, for example that routine replacement has no benefit (CDC says no randomised study has tested it) and that the daily check looks for redness and swelling (CDC says to feel for tenderness and to look if the dressing is transparent). (Second accuracy pass; CDC/HICPAC 2011 guideline, full text PDF read with pdftotext) · Replaced each reason with what CDC says and matched the daily check to its wording.
- 2026-11-17 · Peripheral cannula for anticancer treatment: steel needles: The step said never to give anticancer drugs through a steel (butterfly) needle and cited CDC. CDC says to avoid steel needles for fluids and medication that might cause tissue necrosis if they leak, and does not mention butterfly needles. The absolute rule for chemotherapy is in the West Midlands network guidance. (Second accuracy pass; CDC/HICPAC 2011 (PDF) and West Midlands 2017 (PDF), read as raw text) · The step follows CDC's wording and scope, and the West Midlands rule is stated beside it with its source.
- 2026-11-17 · Peripheral cannula for anticancer treatment: when to replace a cannula: The page set CDC against the Japanese guideline on routine replacement and left out epic3, the English national guideline it already cited, which says peripheral cannulas should be re-sited when clinically indicated and not routinely (Class B). The reason given also said routine re-siting does not reduce leaks, while the Japanese guideline found it lowered infiltration but not extravasation. (Second accuracy pass; epic3 (Europe PMC full text), JSCN/JSMO/JASPO 2024 (Europe PMC full text) and CDC/HICPAC 2011 (PDF), read as raw text) · epic3's recommendation and grade are in the summary, the best-in-class statement, the variant and the grades, and the reason says what the Japanese guideline found.
- 2026-11-17 · Peripheral cannula for anticancer treatment: blood return checks: The check for blood return rested only on a UK network document, and the reason given was not in it. The Japanese guideline has a graded recommendation the page left out: checking blood backflow for early detection of a leak is weakly recommended, certainty D, with the timing and method unclear. (Second accuracy pass; JSCN/JSMO/JASPO 2024, Europe PMC full text including Table 1) · The recommendation and its grade were added to the step, the summary and the grades, and the uncertainty about timing is in the gaps.
- 2026-11-17 · Peripheral cannula for anticancer treatment: reasons given for the steps: Eight reasons said more than the sources: that arm veins carry less risk of clots than leg veins, that a big vein dilutes the drug quickly, that tendons are damaged, that steel needles slip out more easily, that a pump keeps pushing fluid after a cannula slips, that re-siting uses up veins, and others like them. None is in the source cited, and the Japanese guideline calls the link between pumps and leaks unclear. (Second accuracy pass; JSCN/JSMO/JASPO 2024, CDC/HICPAC 2011, epic3 and West Midlands 2017, all read as raw text) · Each reason now says what the cited source says and who says it.
- 2026-11-17 · Peripheral cannula for anticancer treatment: UK network practice written as fixed rules: The page gave a new cannula, a limit of two attempts, vesicants first and central access after a previous extravasation as rules, and the patient summary said what the nurse should do. The West Midlands guidance says ideally for the first three, allows exceptions to vesicants first, and asks only that central access be considered. It is one English network document for patients over 16, past its review date. (Second accuracy pass; West Midlands 2017 PDF from england.nhs.uk, read as raw text) · The hedges are restored, the patient summary names UK nursing guidance as the source, and the gaps say that no source read gives a gauge number, a cannula age limit in hours or a fixed limit on attempts.
- 2026-11-17 · Peripheral cannula for anticancer treatment: the Japanese guideline's title and certainty scale: The source was listed under a title that is not the paper's, and its certainty scale was described as A (strong) to D (very weak). The paper prints A high, B moderate, C low and D very low. The first step also said a central device should be checked especially for vesicants, long infusions or many cycles, which the guideline does not say. (Second accuracy pass; JSCN/JSMO/JASPO 2024, Europe PMC full text) · The title and scale are corrected, and the step gives the benefits and the 8% device complications the guideline lists.
- 2026-11-17 · PICC and tunnelled central line care: gauze dressing interval: The page said gauze dressings are changed every 2 days and gave NICE and epic3 as agreeing. NICE CG139 says to change a gauze dressing every 24 hours, and epic3 gives no interval for gauze. Only SHEA and CDC say 2 days. (Second accuracy pass; NICE CG139 recommendations, SHEA 2022, CDC 2011 and epic3 read as raw text) · The summary, the step, the best supported version and the gaps now give 2 days (SHEA, CDC) beside 24 hours (NICE).
- 2026-11-17 · PICC and tunnelled central line care: weekly dressing change: The step applied the SHEA interval to every line and dropped two words. SHEA says at least every 7 days and scopes the rule to non-tunnelled lines. (Second accuracy pass; SHEA 2022 full text from Europe PMC) · The step now says at least every 7 days for a PICC or other non-tunnelled line and gives the rule for tunnelled lines from CDC, epic3 and NICE.
- 2026-11-17 · PICC and tunnelled central line care: sutureless securement: The page gave sutureless securement as a CDC suggestion and did not say that the newer SHEA 2022 update lists it as an unresolved issue whose effect on line bloodstream infection is unknown. (Second accuracy pass; SHEA 2022 full text from Europe PMC) · Both positions are now on the step, in the summary, the grades and the gaps.
- 2026-11-17 · PICC and tunnelled central line care: heparin lock evidence: The page said the Cochrane review rated the evidence on how long a line stays open as very low certainty. The review rates it low certainty, and calls the finding of fewer blockages uncertain. The page also said heparin locks are common for tunnelled lines, which no source read says. (Second accuracy pass; Cochrane review CD008462.pub4 abstract from Europe PMC) · The certainty is corrected, the authors' caution added, and the unsourced phrase removed. NICE advice on heparin flushes was added.
- 2026-11-17 · PICC and tunnelled central line care: chlorhexidine dressings and the CDC update: The page said the CDC update of July 2017 on chlorhexidine dressings could not be read. Its text is printed in the guideline itself: recommended for short-term, non-tunnelled lines in adults (Category IA). The step also said consider where SHEA says use. (Second accuracy pass; CDC 2011 guideline PDF (last update October 2017) read with pdftotext) · The CDC recommendation and its scope are now on the page, and the step gives the SHEA and epic3 wording side by side.
- 2026-11-17 · PICC and tunnelled central line care: baths, swimming and showers: The page told patients to keep the line out of baths and swimming pools. CDC says not to submerge the catheter or its site. Macmillan says a bath or shower is safe with a waterproof cover and that swimming should usually be avoided. (Second accuracy pass; CDC 2011 PDF and Macmillan patient pages read as raw text) · The advice now says not to let the line or its site go under water, and the gaps set out how the sources differ.
- 2026-11-17 · PICC and tunnelled central line care: when to call the team: The page told patients to call at once with a fever and listed warning signs with no source. CDC says only to report changes at the site or new discomfort. Macmillan lists redness, swelling or pain, discoloured fluid and a high temperature, and says to contact the hospital doctor or nurse without a time limit. (Second accuracy pass; Macmillan PICC lines and Central lines pages read as raw text and added as sources) · The warning signs now follow Macmillan, and the gaps say that no source read sets a time limit.
- 2026-11-17 · PICC and tunnelled central line care: flushing after every use: The step said to flush after every infusion or blood sample. That timing rests only on the ESMO 2015 guideline, which could not be re-read. The quoted epic3 sentence covers only the choice of saline for lumens used frequently. (Second accuracy pass; epic3 and NICE CG139 read as raw text; ESMO 2015 not readable on 17 November 2026) · The step now covers flushing and locking with saline, the ESMO sentence is marked as not re-read, and the gaps say no source read gives a flush volume or interval.
- 2026-11-17 · PICC and tunnelled central line care: reasons given for steps: Nine reasons went beyond the sources, for example that hands are the commonest route for germs, that a few seconds of scrubbing are not enough when SHEA sets a 5 second minimum, and that a line left in raises the risk of clots. (Second accuracy pass; CDC 2011, SHEA 2022, epic3 and NICE CG139 read as raw text) · Each reason now says what a named source says.
- 2026-11-17 · PICC and tunnelled central line care: NICE Tegaderm CHG guidance: The source was listed under an old title, and the page said NICE supports the dressing in critically ill adults. The current title is Tegaderm CHG securement dressing for vascular access sites in critically ill adults, and NICE says it should be considered for adults in intensive care or high dependency units. (Second accuracy pass; NICE HealthTech guidance 379 read as raw text) · Title and wording corrected.
- 2026-11-17 · Choosing a venous access device for anticancer treatment: port rather than PICC, and where the line goes: One step joined two ITAC recommendations and showed only the grade 1B. ITAC grades the insertion site and tip position 1B, but its suggestion of ports over PICCs is expert consensus, printed as guidance, and the step said prefer where ITAC says suggest. (Second accuracy pass; ITAC 2022 full text, Europe PMC XML read as raw text) · Split into two steps, each with its own quote and grade, and the port step now says consider.
- 2026-11-17 · Choosing a venous access device for anticancer treatment: how much better ports did: The reason given was that ports had about a third fewer complications, a figure worked out by the writer that the trial does not print, and the step said to offer a port without the trial's scope. CAVA prints 29% against 43% with Hickman lines and 32% against 47% with PICCs, in adults having 12 weeks or more of treatment in the NHS. (Second accuracy pass; CAVA published abstract from Europe PMC and the authors' accepted manuscript from the University of Glasgow repository) · The printed rates replace the computed figure, and the step now says to consider a port first for 12 weeks or more of treatment.
- 2026-11-17 · Choosing a venous access device for anticancer treatment: reasons for a two-channel line and for kidney disease: Two steps gave reasons the source does not state: that a PICC or tunnelled line can be placed quickly and carries chemotherapy, blood products and antibiotics together, and that arm veins may be needed for dialysis. Only the MAGIC-ONC abstract was read and it gives no reasons; it recommends coordination with a nephrologist to preserve veins in the context of cancer prognosis. (Second accuracy pass; MAGIC-ONC abstract, Europe PMC record) · Both reasons removed and the steps reworded to the abstract, which is said to give no reasons.
- 2026-11-17 · Choosing a venous access device for anticancer treatment: PICC as the default device: The page said patient preference and waiting times can favour PICCs in practice, with no source. The CAVA manuscript reports the opposite on preference: a device-specific quality of life questionnaire favoured ports on 8 of 16 questions. It also reports more infection with ports than PICCs (12% against 8%) and more venous thrombosis with PICCs (11% against 2%). (Second accuracy pass; CAVA authors' accepted manuscript, open PDF read with pdftotext) · Unsourced sentence replaced with the trial's own findings on quality of life, infection, thrombosis and cost.
- 2026-11-17 · Choosing a venous access device for anticancer treatment: femoral vein and the ESMO guideline: The step to avoid the femoral vein, its grade and a statement about PICC longevity rested on the ESMO 2015 guideline, which could not be re-read. (Second accuracy pass; ESMO 2015 not readable on 17 November 2026; CDC 2011 guideline (last update October 2017) full text read instead) · The femoral step now rests on CDC (avoid the femoral vein in adults, Category IA), the PICC longevity statement was removed, and the gap is stated.
- 2026-11-17 · Choosing a venous access device for anticancer treatment: who it applies to: The page said children, people needing apheresis or transplant lines and dialysis patients have other considerations, and described treatment as lasting weeks to months, with no source. The trial studied 12 weeks or more, and only 89 of its 1061 participants had a haematological cancer. (Second accuracy pass; CAVA abstract and accepted manuscript, ITAC 2022 full text) · Population reworded to what the sources cover, and the gaps now say no source read gives a threshold in weeks for a cannula.
- 2026-11-17 · Bevacizumab and other anti-VEGF drugs around surgery: fruquintinib interval: The page listed fruquintinib with a 2-week pause before elective surgery. The US Fruzaqla label says not to give it for at least 2 weeks before major surgery; it does not use the words elective surgery. (Second accuracy pass; Fruzaqla US prescribing information, DailyMed XML read as raw text) · The summary, the step and the grade row now say at least 2 weeks before major surgery.
- 2026-11-17 · Bevacizumab and other anti-VEGF drugs around surgery: wound-healing problem during treatment: The step said to withhold any of these drugs until the wound healed and decide case by case about restarting. The labels differ: Zaltrap says to discontinue in patients with impaired wound healing, Inlyta says to resume at a reduced dose or discontinue, and Avastin says to withhold and to discontinue for necrotising fasciitis. (Second accuracy pass; Avastin, Zaltrap and Inlyta US prescribing information, DailyMed XML read as raw text) · The step now gives each label's own instruction, with two new quoted grade rows.
- 2026-11-17 · Bevacizumab and other anti-VEGF drugs around surgery: cabozantinib in the EU: The page gave only the US interval of 3 weeks for cabozantinib and said EU texts were less specific. The EU Cabometyx text, unread in the first pass, says at least 28 days before scheduled surgery if possible; the EU Cyramza and Zaltrap texts give at least 4 weeks. (Second accuracy pass; EMA product information PDFs for Cabometyx, Zaltrap and Cyramza, read with pdftotext) · Added the EU figures to the summary, a variant for Cabometyx and one for Cyramza, and quoted grade rows; both EU texts added as sources.
- 2026-11-17 · Bevacizumab and other anti-VEGF drugs around surgery: why the intervals differ: The page said the intervals follow each drug's half-life and that each tablet leaves the body at a different speed. No label read says how its interval was chosen; only the 2016 meta-analysis says so, and only for bevacizumab. (Second accuracy pass; twelve US labels and Zhang 2016 full text from Europe PMC) · The claim is now limited to bevacizumab and attributed to the meta-analysis; the gap says no label explains its interval.
- 2026-11-17 · Bevacizumab and other anti-VEGF drugs around surgery: minor surgery: The step on restarting after minor surgery once the wound has healed was written for every drug in the class. Only the Zaltrap texts state it. (Second accuracy pass; Zaltrap US label and EU product information read as raw text) · The step is scoped to ziv-aflibercept and says the other labels give no separate rule.
- 2026-11-17 · Bevacizumab and other anti-VEGF drugs around surgery: quotations: Four quotations were not the labels' words exactly: three had a space before a colon that the labels do not print, and one joined two table cells into one sentence. (Second accuracy pass; scripted verbatim check against the DailyMed XML text) · Replaced with verbatim sentences from the warnings sections of the Cyramza, Sutent, Nexavar and Avastin labels.
- 2026-11-17 · Bevacizumab and other anti-VEGF drugs around surgery: unsourced advice: The page said bevacizumab doubled wound complications, that a UK or EU team may use the US figure, that no guideline sets a different interval, and that biosimilars carry the same instructions. The meta-analysis reports an odds ratio of 2.32, and no source read states the other three. (Second accuracy pass; Zhang 2016 full text and the labels read as raw text) · Reworded to the odds ratio with its confidence interval; the other statements were removed or rewritten as things not found or not read.
- 2026-11-17 · Bleeding precautions at home when platelets are low: warning signs: The page told readers to call the team "straight away" and said the signs "need checking the same day" and could mean bleeding in the brain. NCI says only to call the doctor or nurse about these more serious problems and gives no time limit or diagnosis; Macmillan says to tell the doctor or nurse. (Second accuracy pass; NCI thrombocytopenia page, NCI booklet and Macmillan page read as raw text) · The time limits and the brain bleeding reason were removed from the short summary, the step and its reason, the list of signs was worded as NCI prints it, and a gap now says no source read gives a time limit.
- 2026-11-17 · Bleeding precautions at home when platelets are low: summary and best version: The page said NCI and Macmillan agree on every point except flossing. Macmillan gives only three tips: protective gloves, care not to bump into things or trip, and a soft toothbrush without flossing. (Second accuracy pass; Macmillan side effects of chemotherapy page read as raw text) · The summary and best version now say which source gives which advice.
- 2026-11-17 · Bleeding precautions at home when platelets are low: ASCO and NICE thresholds: The page said ASCO 2018 agrees with NICE on prophylactic platelet thresholds, and that adults after an autologous transplant "are transfused" at the first sign of bleeding. The ASCO abstract prints no thresholds and says such patients "may receive" a transfusion at the first sign of bleeding rather than prophylactically. (Second accuracy pass; ASCO 2018 abstract through Europe PMC (full text not readable) and NICE NG24 platelet chapter as raw text) · The agreement claim was removed, "may" was restored, and the page now says the abstract prints no thresholds.
- 2026-11-17 · Bleeding precautions at home when platelets are low: flossing and sports: The tooth care step said only to ask the team before flossing and the activity step said to avoid contact sports. The NCI booklet lists dental floss and toothpicks under things not to do, suggests a water flosser, and says not to play sports or do other activities in which you could get hurt. (Second accuracy pass; NCI Chemotherapy and You read with pdftotext) · Both steps now follow the booklet's wording.
- 2026-11-17 · Bleeding precautions at home when platelets are low: reasons given for steps: Six reasons were not in the sources: that a soft brush cleans without cutting, that cuts bleed for longer, that knocks cause bleeding that is hard to stop, that tampons and enemas damage delicate linings, that pressure helps a clot form, and that transfusion is decided on the count and on bleeding. The page also said the precautions lower the chance of bleeding and that no trial has tested them, which no source states. (Second accuracy pass; both NCI sources and the Macmillan page read as raw text) · Each reason now says what the source says, or that the source gives no reason; the benefit claim became "guidance advises" and the trial claim became "the sources cite no trial".
- 2026-11-17 · Bleeding precautions at home when platelets are low: sources: The NCI booklet was dated 2023, from the date the PDF file was made. The booklet prints January 2024. The population was given as adults, which the patient sources do not say. (Second accuracy pass; NCI Chemotherapy and You read with pdftotext) · The year is now 2024 and the population says people.
- 2026-11-17 · End of life care for children and young people with cancer: the Advance Care Plan and resuscitation: The step said to keep the Advance Care Plan separate from any decision about resuscitation. NICE NG61 says the plan records the management of life-threatening events, including plans for resuscitation or life support, and only that it should not be confused with a do-not-attempt-resuscitation order. (Second accuracy pass; NICE NG61 recommendations 1.2.5, 1.2.14 and 1.3.1, fetched raw and read as text) · The step now says not to confuse the plan with a do-not-attempt-resuscitation order, and that NICE says to attempt resuscitation unless such an order is in place.
- 2026-11-17 · End of life care for children and young people with cancer: fluids and food: The page said NICE NG61 makes no recommendation on stopping fluids and food. It has nine recommendations on hydration and nutrition (1.3.47 to 1.3.55), including whether starting or continuing tube or intravenous fluids and feeding is in the child's best interests. It has none on palliative sedation. (Second accuracy pass; NICE NG61 recommendations page, fetched raw and read as text) · The comparison with the Dutch guideline now states what NICE says on fluids and food and that it is silent only on palliative sedation.
- 2026-11-17 · End of life care for children and young people with cancer: pain, agitation and the named contact: Three things NICE words as "consider" or "think about" were written as orders: a stepwise approach to pain relief, a named first point of contact, and calm surroundings before medicines for agitation. NICE sets no order between the non-drug measures and medicines for agitation, and says medicines begin at low doses. (Second accuracy pass; NICE NG61 recommendations 1.3.27, 1.3.38 and 1.5.7, fetched raw and read as text) · The steps and the summary now carry NICE's own strength of wording.
- 2026-11-17 · End of life care for children and young people with cancer: what families are told to expect: The patient summary said a specialist team can treat symptoms at home, in a hospice or in hospital day and night, and that there is help day and night. NICE recommends that services provide consultant advice and paediatric nursing day and night, when needed, for a child cared for at home. No source read says how many families have it. (Second accuracy pass; NICE NG61 recommendation 1.5.9, fetched raw and read as text) · The summaries now say this is what NICE asks services to provide, and a gap says so.
- 2026-11-17 · End of life care for children and young people with cancer: timing of practical information: The step said to explain the practical arrangements before the death. NICE says to discuss them with parents and give them in writing and sets no time. (Second accuracy pass; NICE NG61 recommendation 1.4.1, fetched raw and read as text) · The words "before the death" were removed and "in a timely way" added for informing others, as recommendation 1.4.8 says.
- 2026-11-17 · End of life care for children and young people with cancer: reasons given for each step: Nineteen of the reasons given beside steps said things no cited source states, such as that agitation usually has a cause that can be eased, that children often know more than adults think, and that findings from a 2000 study are why symptom control is planned ahead. (Second accuracy pass; NICE NG61, Weaver 2015 (PubMed Central full text) and the Wolfe 2000 abstract, read as raw text) · Each reason was reworded to what its source says.
- 2026-11-17 · End of life care for children and young people with cancer: the variants: The variants said the psychosocial standard introduces the palliative care team early (it says palliative care concepts), that some services and audits count a home death as the mark of good care (no source read says so), that the charity pathway says the place may change (NICE says that), and that NICE NG61 is a consensus document and written for England (neither is in the text read). (Second accuracy pass; Weaver 2015 full text, Johnston 2021 abstract, Together for Short Lives pathway PDF and NICE NG61, read as raw text) · Each statement was reworded to its source or removed.
- 2026-11-17 · Ambulatory chemotherapy pumps at home: who sets and checks the rate: The page said the rate is set in the pharmacy or checked by two nurses, and that a second clinician independently checks the pump against the order before it is connected. No source read says the pharmacy sets the rate. ASCO-ONS requires at least two licensed clinicians to verify and document the rate, duration and pump settings in the presence of the patient; the word independently belongs to its third verification. (Second accuracy pass; ASCO-ONS 2024 read as raw text in the Oncology Nursing Forum copy) · Reworded the patient summary and the step to the fourth verification as printed.
- 2026-11-17 · Ambulatory chemotherapy pumps at home: when to call: The page told patients to call at once if the pump leaks, the line disconnects or it empties too fast or not at all, citing a study that only counts failure types. No source read gives those triggers or that urgency. Cancer Research UK says to contact the chemotherapy nurse with any problem or question and that a kit might be given for leaks. (Second accuracy pass; Cancer Research UK page on chemotherapy pumps, read as raw text) · Step and patient wording now follow Cancer Research UK, which is added as a source, and the gap is named.
- 2026-11-17 · Ambulatory chemotherapy pumps at home: ISMP Canada recommendations: The page said ISMP Canada recommends pharmacy-controlled rates, standardised volumes or concentrations and writing the duration into the order, and dated the death to 2006. ISMP Canada's bulletin says controlled-rate delivery (elastomeric pumps or preset maximum rates) until smart pumps are available, total volume and rate of infusion in the order, and that it was invited in 2006 to analyse the death. The full report could not be fetched again. (Second accuracy pass; ISMP Canada Safety Bulletin of 20 August 2007, PDF read as raw text) · Reworded to the bulletin, quoted it, added it as a source and said the recommendations are ungraded suggestions.
- 2026-11-17 · Ambulatory chemotherapy pumps at home: disconnection: The disconnection step said to flush and lock the line as for any access and to wear gloves. No source read covers flushing after disconnection. The residue study says personal protective equipment should be used and pumps disposed of in dedicated hazardous waste containers. (Second accuracy pass; Cardoso 2025 full text from Europe PMC) · Removed the flush and lock instruction, reworded to the study, and named the gap.
- 2026-11-17 · Ambulatory chemotherapy pumps at home: home teaching and reasons: The home teaching step added a spill kit and a 24-hour contact number that the cited ASCO-ONS standard does not list, and three steps gave reasons no source states, one of them that a rate safeguard means the error cannot happen. (Second accuracy pass; ASCO-ONS 2024 and the ISMP Canada bulletin, read as raw text) · Step reworded to the standard (safe handling, contact details with when and whom to call); reasons replaced with ISMP Canada's listed causes.
- 2026-11-17 · Iodinated contrast for CT: kidney function checks, kidney protection and metformin: restarting metformin: The step and summary said metformin restarts after an eGFR check at 48 hours. ESUR says to measure eGFR within 48 hours and restart only if renal function has not changed significantly. (Second accuracy pass; ESUR Contrast Media Safety Committee Guidelines 2026 (PDF read with pdftotext) and the 2018 ESUR paper (Europe PMC full text)) · The condition and timing were restored, and the ACR Manual's different rule (withhold 48 hours, restart only once kidney function is found normal) was added beside it.
- 2026-11-17 · Iodinated contrast for CT: kidney function checks, kidney protection and metformin: fluid regimen grade: The step giving the saline and bicarbonate regimens carried Level of evidence A. That grade belongs to the statement that the two fluids have similar efficacy; the regimens themselves are Level D, expert opinion. (Second accuracy pass; van der Molen et al. 2018, Table 2, Europe PMC full text) · The step now quotes the regimen sentence with Level D, and Level A stays with the efficacy statement.
- 2026-11-17 · Iodinated contrast for CT: kidney function checks, kidney protection and metformin: screening questions: The page said to test eGFR only if a screening answer is yes and to proceed without a test otherwise. NICE says to consider an eGFR test, and to consider proceeding without one in a stable person. (Second accuracy pass; NICE NG148 recommendations 1.1.10 to 1.1.12, page read as raw HTML) · Reworded to consider in the step, summary and variant.
- 2026-11-17 · Iodinated contrast for CT: kidney function checks, kidney protection and metformin: steps resting on the wrong source: Three statements were cited to a source that does not make them: alternative imaging and no extra dialysis were cited to NICE, and the list of N-acetylcysteine, statins, vitamin C and dialysis to the ESUR handbook, which names only statins and drug classes. (Second accuracy pass; NICE NG148, ESUR Guidelines 2026 and the 2018 ESUR paper, all read as raw text) · Alternative imaging and extra dialysis are now separate steps quoting the ESUR handbook; drug prophylaxis is cited to the 2018 paper with its grades (N-acetylcysteine Level A, the others Level B).
- 2026-11-17 · Iodinated contrast for CT: kidney function checks, kidney protection and metformin: contrast dose: The page gave the lowest diagnostic dose as an agreed rule. That is ESUR's statement; the ACR Manual does not recommend reducing the dose to lower kidney risk and recommends standard dosing. (Second accuracy pass; ACR Manual on Contrast Media 2026 edition (PDF read with pdftotext) and ACR/NKF 2020 (Europe PMC full text)) · Both positions are stated and a variant was added.
- 2026-11-17 · Iodinated contrast for CT: kidney function checks, kidney protection and metformin: nephrotoxic drugs and chemotherapy: The step said not to hold back chemotherapy or other nephrotoxic drugs, with the reason that stopping cancer treatment has its own cost. ESUR says only that there is insufficient evidence to recommend withholding them, and ACR/NKF recommend stopping non-essential nephrotoxic drugs when feasible at eGFR below 30 or in acute kidney injury. (Second accuracy pass; van der Molen et al. 2018 Table 1 and ACR/NKF 2020, Europe PMC full text) · Reworded to both sources and the unsourced reason removed.
- 2026-11-17 · Iodinated contrast for CT: kidney function checks, kidney protection and metformin: thresholds and emergency imaging: The page said ESUR, ACR/NKF and NICE all set the threshold at eGFR below 30 or acute kidney injury, and that emergency imaging is never delayed for a kidney result. NICE names only an eGFR below 30, and its emergency rule applies when the delay is likely to be clinically significant; ESUR also counts an eGFR below 45 in intensive care patients. (Second accuracy pass; NICE NG148 1.1.6 and 1.1.7, ESUR Guidelines 2026, ACR/NKF 2020, read as raw text) · Each body's threshold and emergency rule is now stated separately in its own words.
- 2026-11-17 · Iodinated contrast for CT: kidney function checks, kidney protection and metformin: reasons given for steps: Several reasons had no source: that most people need no fresh blood test, that fluid is the one measure shown to work, that less contrast means less strain on the kidneys, that metformin does not harm the kidneys, that fasting causes dehydration, and that calcium should be checked in myeloma. (Second accuracy pass; all cited sources read as raw text) · Each was replaced with what the cited source says, or removed.
- 2026-11-17 · Iodinated contrast for CT: kidney function checks, kidney protection and metformin: ACR Manual edition: The ACR Manual on Contrast Media was listed as a 2024 edition that could not be read. The current edition is 2026 and it now reads in full. (Second accuracy pass; ACR Manual 2026 PDF fetched on 16 November 2026 and read with pdftotext) · Year and label corrected, and the Manual's positions on metformin, contrast dose, fasting, myeloma and eGFR timing added.
- 2026-11-17 · Gadolinium contrast for MRI in kidney impairment: the American agent groups: The page gave gadoxetate as the American group III agent and told teams to check kidney function before it. That was the 2021 grouping. The current ACR Manual on Contrast Media (2026) moved gadoxetate into group II in 2024, lists no agent in group III, and gives gadopiclenol a provisional group II place. (Second accuracy pass; ACR Manual on Contrast Media 2026, PDF read as raw text (it could not be read on 11 November)) · The summary, the kidney test step and the gaps now give the 2021 grouping and the 2026 grouping side by side, and the manual is cited as a read source dated 2026 rather than 2024.
- 2026-11-17 · Gadolinium contrast for MRI in kidney impairment: time between repeat scans: The page told teams to space scans 7 days apart when eGFR is below 30. ESUR says only to consider that wait, and the page left out the different American advice: where delay is not medically inappropriate, more than 24 hours or a dialysis session in between may allow greater clearance, and an urgent repeat dose should not be delayed for fear of NSF. (Second accuracy pass; ESUR Guidelines 2026 PDF and the ACR/NKF 2021 full text from Europe PMC, both read as raw text) · The step now says to consider spacing, and the step, summary and best-in-class statement carry both bodies' figures with who says which.
- 2026-11-17 · Gadolinium contrast for MRI in kidney impairment: choice of agent: The step ordered a lowest-risk agent on the strength of a sentence that says only that the risk is very low, quoted with a broken superscript, and told all readers to keep liver-specific agents for liver imaging, which is a European licensing rule. The ACR/NKF paper says risk mitigation can include a group II agent. (Second accuracy pass; ACR/NKF 2021 full text and ACR Manual 2026, read as raw text) · The step now rests on the ACR Manual sentence that group II agents are strongly preferred in patients at risk, quoted verbatim, and the liver imaging rule is stated as European in the summary only.
- 2026-11-17 · Gadolinium contrast for MRI in kidney impairment: reasons given for steps: Seven reasons were not in the cited sources: that a blood test only adds delay, that a record helps later teams, that the choice is shared, that linear agents release gadolinium more easily and dialysis is the only way to clear them, that weak kidneys take days to clear the dye, that the dye crosses to the baby, and that little reaches breast milk (the last is in the ACR Manual, not in ESUR). (Second accuracy pass; ESUR Guidelines 2026, ACR/NKF 2021 and ACR Manual 2026, read as raw text) · Each reason now says what a named source says, including ESUR's statement that there is no evidence haemodialysis protects against NSF and the ACR figure of under 0.0004% for breast milk.
- 2026-11-17 · Gadolinium contrast for MRI in kidney impairment: patient summary and other wording: The patient summary called NSF rare, said today's types are used for most scans, and gave advice on what to tell the team with no source. The best-in-class statement said a scan is never delayed, which neither body says, and one variant described what some departments do with no source. (Second accuracy pass; ESUR Guidelines 2026, ACR/NKF 2021 and the NHS MRI scan page, read as raw text) · Rare and the claim about most scans were removed, the NHS page was added as the source for what patients are asked before a scan, the best-in-class statement uses each body's own words, and the variant now reports the FDA's 2010 boxed warning as ACR/NKF describe it.
- 2026-11-17 · Malignant ascites: paracentesis and indwelling peritoneal catheter drainage: warning signs with a long-term drain: The step telling patients what to report cited the NICE guidance, which lists complications seen in studies and gives no advice on what to report or how fast. Cancer Research UK's page on long term drains does: call the 24 hour advice line straight away for a high temperature, red or swollen skin around the drain, abdominal pain or a change in the colour of the fluid. (Second accuracy pass; NICE HealthTech guidance 282 and the Cancer Research UK page, both read as raw text) · The step now cites the Cancer Research UK page and uses its list and its urgency, and the gaps say NICE does not give this advice.
- 2026-11-17 · Malignant ascites: paracentesis and indwelling peritoneal catheter drainage: day-case and free drainage: The page said drainage is often a day case, told teams to plan a day case and to let the drain run freely, and said most patients prefer it. NICE says inpatient drainage is the most common. The one UK series behind these steps says a day case potentially represents the standard of care, chose people with good performance status, and found 5 of 7 preferred it. (Second accuracy pass; Harding 2012 full text from Europe PMC and NICE HealthTech guidance 282) · Both steps now say consider, the selection criteria and the actual counts are given, and the summary no longer says often.
- 2026-11-17 · Malignant ascites: paracentesis and indwelling peritoneal catheter drainage: the 2012 UK series: The summary said 21 people were drained without clamping. The paper reports 21 paracenteses in 18 women, and in 4 of them the drain was clamped because systolic pressure fell below 95 mm Hg. (Second accuracy pass; Harding 2012 full text from Europe PMC) · Corrected to 21 paracenteses in 18 women, with the 4 clamped drains stated.
- 2026-11-17 · Malignant ascites: paracentesis and indwelling peritoneal catheter drainage: paracentesis as first treatment: The page said paracentesis is the first-line treatment and told teams to offer it first. The Japanese guideline says it can be considered as the first-line treatment for massive ascites, in an ungraded general statement. (Second accuracy pass; Japanese guideline 2021 full text from Europe PMC) · Reworded to consider, with the scope of massive ascites, in the step, the summary and the best-in-class statement.
- 2026-11-17 · Malignant ascites: paracentesis and indwelling peritoneal catheter drainage: home drainage step: A sentence quoted as the NICE committee's statement of expected benefit is one of the benefits claimed by the device's sponsor, as NICE section 2.6 says. (Second accuracy pass; NICE HealthTech guidance 282 read as raw text) · The step now quotes the committee's own consideration on training community nurses, patients or carers, and its reason is what the committee was advised.
- 2026-11-17 · Malignant ascites: paracentesis and indwelling peritoneal catheter drainage: reasons given for steps: Several reasons were not in the cited sources: that shrinking the cancer stops the fluid returning, that routine drips keep people in hospital, that many people do not respond to diuretics, and that practice from cirrhosis is not supported. The sources say less: the diuretic data conflict, drainage speed and intravenous fluids are not sufficiently studied. (Second accuracy pass; Becker 2006 and Stephenson 2002 abstracts, Japanese guideline 2021 full text) · Each reason now says what its source says.
- 2026-11-17 · Malignant ascites: paracentesis and indwelling peritoneal catheter drainage: cost of the catheter: The summary gave only the estimated saving of £1,095 a person against inpatient drainage. The same NICE cost model estimates an additional cost of £896 a person against drainage as an outpatient procedure. (Second accuracy pass; NICE HealthTech guidance 282 section 3.37 read as raw text) · Both figures are now given.
- 2026-11-17 · Indwelling pleural catheter: draining fluid from the chest at home: why an IPC is offered: The page said the catheter avoids a hospital stay. The BTS guideline says an IPC appears to be associated with a shorter first hospital stay and fewer later inpatient days. (Second accuracy pass; BTS guideline and clinical statement with appendices 3 and 9, PDFs from the BTS document library read with pdftotext) · Reworded to a shorter first hospital stay.
- 2026-11-17 · Indwelling pleural catheter: draining fluid from the chest at home: reasons given for drainage steps: Five reasons were not in the cited source: that harder suction does harm, that pain often means the lung cannot expand (the statement says it may indicate this), that problems are treatable if caught early, that drained fluid carries drug residues, and that each system needs its own bottle to stay air-tight. (Second accuracy pass; BTS guideline and clinical statement with appendices 3 and 9, PDFs from the BTS document library read with pdftotext) · Each reason reworded to what the statement, its appendices or the leaflet say.
- 2026-11-17 · Indwelling pleural catheter: draining fluid from the chest at home: showering: The page said to shower once the dressing allows. The statement says patients should be able to shower normally provided the site is kept clean and dry. (Second accuracy pass; BTS guideline and clinical statement with appendices 3 and 9, PDFs from the BTS document library read with pdftotext) · Reworded to the statement's condition.
- 2026-11-17 · Indwelling pleural catheter: draining fluid from the chest at home: daily drainage and breathlessness: The page said two trials showed daily drainage roughly doubles pleurodesis without improving breathlessness. The ASAP abstract reports no breathlessness result, and the figures are 47% against 24% in ASAP and 44.2% against 15.9% in AMPLE-2. (Second accuracy pass; trial abstracts read from Europe PMC, BTS guideline PDF read with pdftotext) · The figures are given as printed, the breathlessness finding is attributed to AMPLE-2 and to the BTS evidence statement, and the word doubles was removed for talc as well (43% against 23%).
- 2026-11-17 · Indwelling pleural catheter: draining fluid from the chest at home: variants: Symptom-guided drainage was said to mean fewer drainages and lower cost, which no source read reports as a finding, and underwater seal drains were said to be used for inpatients where the statement says they may be considered. (Second accuracy pass; BTS guideline and clinical statement with appendices 3 and 9, PDFs from the BTS document library read with pdftotext) · Cost reworded to the BTS good practice point that daily drainage may carry a cost. Underwater seal reworded to may be considered.
- 2026-11-17 · Hot flushes on breast cancer endocrine therapy and prostate cancer hormone therapy: non-hormonal options: cognitive behavioural therapy: The step said to offer menopause-specific CBT and gave reasons (good evidence, no drug interactions, no effect on cancer treatment) that the source does not state. NICE NG23 recommendation 1.5.2 says to consider it. (Second accuracy pass; NICE NG23 recommendations page, raw HTML) · Changed to consider, and the reason now states the NICE, ASCO and NAMS positions.
- 2026-11-17 · Hot flushes on breast cancer endocrine therapy and prostate cancer hormone therapy: non-hormonal options: SSRIs with tamoxifen: The step and the best-in-class statement said only paroxetine and fluoxetine should be avoided with tamoxifen and credited NICE NG101 with an SSRI that does not interact with tamoxifen. NG101 1.14.24 says SSRIs are not for women taking tamoxifen and names no individual drug, venlafaxine or gabapentin; the paroxetine and fluoxetine rule is CCO only. (Second accuracy pass; NICE NG101 recommendations page and the CCO guideline full text on PubMed Central, raw HTML) · Both positions are now stated with who holds each, in the step, the best-in-class statement and the gaps.
- 2026-11-17 · Hot flushes on breast cancer endocrine therapy and prostate cancer hormone therapy: non-hormonal options: elinzanetant: The step told teams to consider elinzanetant, citing a trial abstract that reports a result and recommends nothing; no guideline read recommends it for people with cancer. The gaps also said a 2026 review did not name the approving agency, when it names both the EMA and the FDA. (Second accuracy pass; OASIS-4 abstract and the 2026 review via Europe PMC; US label via DailyMed; EMA medicine page, raw HTML) · Reworded as an option to discuss, with the trial dose of 120 mg once a day, the US label liver tests and warning signs, and the EU indication, which the EMA page shows includes hot flushes caused by adjuvant endocrine therapy for breast cancer.
- 2026-11-17 · Hot flushes on breast cancer endocrine therapy and prostate cancer hormone therapy: non-hormonal options: fezolinetant: The step said do not use fezolinetant during breast cancer treatment. NICE TA1143 says people having that treatment are not included in the marketing authorisation and that the committee could not make a recommendation for them. (Second accuracy pass; NICE TA1143 committee discussion page, raw HTML) · Reworded to say it is outside its UK marketing authorisation during treatment, with the NICE reason.
- 2026-11-17 · Hot flushes on breast cancer endocrine therapy and prostate cancer hormone therapy: non-hormonal options: reasons given for the breast cancer steps: Three reasons were not in the cited source: that information helps women cope and stay on treatment, that HRT could feed a hormone-sensitive cancer, and that soy, red clover and black cohosh have not been shown to work and may interact with cancer treatment. The first step also added before and during treatment, which NICE does not say. (Second accuracy pass; NICE NG101 and NG23 recommendations pages, raw HTML) · Each reason now says what NICE states: HRT is contraindicated and off-label after breast cancer, and NG23 says the safety of isoflavone and black cohosh preparations is uncertain and interactions with other medicines have been reported.
- 2026-11-17 · Hot flushes on breast cancer endocrine therapy and prostate cancer hormone therapy: non-hormonal options: summary for patients and neurokinin antagonists: The patient summary said hot flushes happen often and sometimes for years, which no source read states. The best-in-class statement called elinzanetant the only neurokinin antagonist with a phase 3 trial in this population, though the 2026 review lists a phase 3 trial of fezolinetant as ongoing. The ACP variant put a contraindication qualifier on the first-line recommendation, where the abstract puts it on the second line. (Second accuracy pass; Europe PMC abstracts and the 2026 review full text) · Reworded to can cause hot flushes, to first with phase 3 results, and to the ACP abstract wording.
- 2026-11-17 · Prostate biopsy: transperineal or transrectal route, antibiotic prophylaxis and aftercare: antibiotics for transrectal biopsy: The reason given said that choosing antibiotics by culture or covering more widely prevents more infections, and the step said to avoid fluoroquinolones without saying where. EAU says most studies of these options used fluoroquinolones and that how they apply to other antibiotics is unclear, and its strong rating for avoiding fluoroquinolones rests on their legal position in Europe. (Second accuracy pass; EAU Prostate Cancer Guidelines 2026, chapter 5, read as raw HTML text) · The reason now carries EAU's caveat, the fluoroquinolone advice is scoped to Europe, and the gaps say that no antibiotic, dose or timing is named because the recommendation names none.
- 2026-11-17 · Prostate biopsy: transperineal or transrectal route, antibiotic prophylaxis and aftercare: urgent symptoms after a biopsy: The page told men that fever, shivering or being unable to pass urine needs urgent care and cited NICE, which does not say so. (Second accuracy pass; NICE NG131 box 1 re-read, and Prostate Cancer UK and Cancer Research UK patient pages read as raw HTML text) · The advice is now its own step, quoted from Prostate Cancer UK, with both patient pages added as sources and the gap named.
- 2026-11-17 · Prostate biopsy: transperineal or transrectal route, antibiotic prophylaxis and aftercare: low-dose aspirin and blood thinners: The step said not to stop low-dose aspirin as routine and to plan other blood thinners with the team. EAU says only that low-dose aspirin is not an absolute contraindication, and no source read covers other blood thinners. (Second accuracy pass; EAU Prostate Cancer Guidelines 2026, chapter 5, read as raw HTML text) · The step now says what EAU says, and the gap on other blood thinners stays.
- 2026-11-17 · Prostate biopsy: transperineal or transrectal route, antibiotic prophylaxis and aftercare: who still needs antibiotics: Immunosuppression and a recent urinary infection were listed as named exceptions to omitting antibiotics. EAU names diabetes and a history of urinary retention as examples; the other two are exclusion criteria of the TRANSLATE trial. The summary for patients also said most men need no antibiotics, a proportion no source gives. (Second accuracy pass; EAU 2026 chapter 5 as raw HTML text and the TRANSLATE abstract from Europe PMC) · Each item is attributed to its source and the patient summary now says men with no extra infection risk.
- 2026-11-17 · Prostate biopsy: transperineal or transrectal route, antibiotic prophylaxis and aftercare: what NICE says about route: The page said the NICE committee found no evidence to separate the two routes. NICE says the committee identified a gap in the evidence on transperineal (non-mapping) biopsy and made a research recommendation. (Second accuracy pass; NICE NG131 recommendations page read as raw HTML text) · Reworded in the summary, the best-in-class statement and the NICE variant; the 2025 review and the update in development are noted.
- 2026-11-17 · Prostate biopsy: transperineal or transrectal route, antibiotic prophylaxis and aftercare: reasons given for skin and rectal cleansing, route and local anaesthetic: Several reasons described mechanisms that EAU does not state, such as cleaned skin keeping bacteria off the needle, and one grade label called a sentence from the text a summary of evidence. A sentence saying many centres still give one antibiotic dose had no source. (Second accuracy pass; EAU Prostate Cancer Guidelines 2026, chapter 5, read as raw HTML text) · The reasons now give what EAU prints (its stated reasons, the relative risk of 0.47 for povidone-iodine, more pain with the transperineal route), the grade label is corrected, and the unsourced sentence is replaced by the cohort counts EAU gives.
- 2026-11-17 · Surgical site infection prevention: why patients are kept warm: The page said mild cooling reduces blood flow to the wound and the body's defences. WHO says it is unclear how keeping body temperature normal reduces wound infection. (Second accuracy pass; WHO global guidelines second edition 2018 (PDF read with pdftotext) and NICE NG125 recommendations page (raw HTML), both re-read on 17 November 2026) · Replaced with what WHO reports: two randomised trials found fewer wound infections with normal body temperature, and the mechanism is unclear.
- 2026-11-17 · Surgical site infection prevention: population: The page said chemotherapy, steroids, malnutrition and previous radiotherapy all raise the risk. Neither source read says this; the WHO text does not mention chemotherapy or radiotherapy at all. (Second accuracy pass; WHO global guidelines second edition 2018 (PDF read with pdftotext) and NICE NG125 recommendations page (raw HTML), both re-read on 17 November 2026) · Removed the sentence.
- 2026-11-17 · Surgical site infection prevention: NICE details attributed to WHO: Steps citing WHO carried details only NICE gives: showering the day before or on the day, clipping on the day of surgery, which operations need antibiotic prophylaxis, and the mucous membrane exception for chlorhexidine in alcohol. WHO makes no recommendation on the timing of hair removal and says the list of operations is outside its scope. (Second accuracy pass; WHO global guidelines second edition 2018 (PDF read with pdftotext) and NICE NG125 recommendations page (raw HTML), both re-read on 17 November 2026) · Each detail is now attributed to NICE in the step and summary, and WHO's own position is stated.
- 2026-11-17 · Surgical site infection prevention: reasons given for steps: Several reasons were not in either source: that informed patients come back early, that shaving makes cuts where germs grow, that chlorhexidine in alcohol kills skin germs best, that coated sutures help a little, that a wound is sealed after two days, and that early treatment stops spread. (Second accuracy pass; WHO global guidelines second edition 2018 (PDF read with pdftotext) and NICE NG125 recommendations page (raw HTML), both re-read on 17 November 2026) · Each reason now says what the source reports, for example the trial comparison WHO reviewed, or says that NICE gives no reason.
- 2026-11-17 · Surgical site infection prevention: antibiotics after the operation: The page said continuing antibiotics adds side effects and resistance. WHO speaks of possible harm and a risk of resistance, and notes low to very low quality evidence of possible benefit in cardiac, vascular and orthognathic surgery. (Second accuracy pass; WHO global guidelines second edition 2018 (PDF read with pdftotext) and NICE NG125 recommendations page (raw HTML), both re-read on 17 November 2026) · Reworded with WHO's hedge and the exception it noted; the drain statement is marked as a separate conditional recommendation.
- 2026-11-17 · Surgical site infection prevention: best in class: The page said WHO and NICE agree on single-dose prophylaxis, a dressing at closure and showering from 48 hours. The dressing and showering advice are NICE only, NICE says to consider a single dose, and WHO suggests not using advanced dressings in place of standard ones. (Second accuracy pass; WHO global guidelines second edition 2018 (PDF read with pdftotext) and NICE NG125 recommendations page (raw HTML), both re-read on 17 November 2026) · Each body's position is stated separately and the difference on dressings is added, with WHO's grade and quote.
- 2026-11-17 · Surgical site infection prevention: oral antibiotics with bowel preparation: The variant said the day before elective colorectal cancer surgery. WHO's recommendation is for adults having elective colorectal surgery and gives no day and no cancer restriction. (Second accuracy pass; WHO global guidelines second edition 2018 (PDF read with pdftotext) and NICE NG125 recommendations page (raw HTML), both re-read on 17 November 2026) · Reworded to WHO's scope; the high oxygen summary also now says adults and if feasible.
- 2026-11-17 · Surgical site infection prevention: patient summary and warning signs: The plain summary said wound infections are common and mostly preventable, that a single antibiotic dose is given, and listed leaking pus among signs to report. WHO says they are among the most preventable hospital infections; NICE says to consider a single dose; NICE's definition lists heat, redness, pain, swelling and fever. (Second accuracy pass; WHO global guidelines second edition 2018 (PDF read with pdftotext) and NICE NG125 recommendations page (raw HTML), both re-read on 17 November 2026) · Reworded to the sources; the warning signs now follow NICE's definition and a gap says no source read gives patients a symptom list or a time limit.
- 2026-11-17 · Sentinel node tracers and blue dye: injection, allergic reaction readiness and aftercare: watching after blue dye: The page said every patient is watched for at least an hour after blue dye and that the label says reactions start late. The 60 minutes is the US label's rule for isosulfan blue only, and the label says nothing about late onset; the UK audit NAP6 says onset was delayed with Patent Blue dye and gives no time. (Second accuracy pass; isosulfan blue US label (DailyMed SPL XML) and the NAP6 abstract (Europe PMC), read raw) · The 60-minute watch is scoped to isosulfan blue and the US label, the late onset is attributed to NAP6, and the gaps say no source read gives a figure for Patent Blue.
- 2026-11-17 · Sentinel node tracers and blue dye: injection, allergic reaction readiness and aftercare: pulse oximetry: The page told staff to confirm a low oxygen reading with an arterial blood gas and said the effect lasts up to a few hours. The label says a blood gas may be needed, and that the effect is greatest at 30 minutes and generally minimal by 4 hours. (Second accuracy pass; isosulfan blue US label (DailyMed SPL XML), read raw) · Step reworded to "may be needed" with the label's two times.
- 2026-11-17 · Sentinel node tracers and blue dye: injection, allergic reaction readiness and aftercare: resuscitation readiness: The page applied the Lymphoseek label's sentence on resuscitation equipment to every tracer and gave a reason (treatment within minutes) the label does not state. (Second accuracy pass; Lymphoseek and isosulfan blue US labels (DailyMed SPL XML), read raw) · Step scoped to tilmanocept, with the isosulfan blue label's own sentence on trained staff named beside it.
- 2026-11-17 · Sentinel node tracers and blue dye: injection, allergic reaction readiness and aftercare: urine colour: The plain summary said urine may be blue or green. The label says blue. (Second accuracy pass; isosulfan blue US label (DailyMed SPL XML), read raw) · Now says blue.
- 2026-11-17 · Sentinel node tracers and blue dye: injection, allergic reaction readiness and aftercare: who it applies to: The page scoped itself to adults and listed vulval, cervical, endometrial and penile cancers, which no source read covers. The Lymphoseek label covers breast cancer, melanoma and mouth cancer and lymphatic mapping from one month of age; the isosulfan blue label says safety in children is not established. (Second accuracy pass; both US labels (DailyMed SPL XML), read raw) · Population rewritten from the labels and the unread cancers named in the gaps.
- 2026-11-17 · Sentinel node tracers and blue dye: injection, allergic reaction readiness and aftercare: reasons and variants: Several reasons were not in the cited source: that NICE recommends two tracers because they find the node more often (NICE gives no reason; a separate review does), that the team can choose another tracer for people at risk, that the pooled study was the largest and its studies observational, that indocyanine green is used alone or with isotope, and that Patent Blue is closely related to isosulfan blue. (Second accuracy pass; NICE NG101 (plain curl), the labels and three meta-analyses (Europe PMC full-text XML), read raw) · Each reworded to what its source says or removed.
- 2026-11-17 · Sentinel node tracers and blue dye: injection, allergic reaction readiness and aftercare: quoted figures: Two quoted abstract sentences were not character for character: the dashes in the confidence intervals and a space had been changed. (Second accuracy pass; Europe PMC full-text XML of both reviews, checked by script) · Quotes restored to the source's own characters.
- 2026-11-17 · Mouth care and oral mucositis prevention during head and neck radiotherapy: glutamine dose and the erratum: The page said the 2021 erratum to the MASCC/ISOO guideline could not be retrieved and that the printed glutamine dose of 10 to 30 mg daily could not be checked. The erratum reads, and corrects the unit to 10 to 30 g daily in the two supporting trials. (Second accuracy pass; the erratum in Cancer 2021, read as raw text from PubMed Central) · The variant now gives the trials' dose in grams with the erratum cited, and says it describes the trials and is not a recommended dose.
- 2026-11-17 · Mouth care and oral mucositis prevention during head and neck radiotherapy: doxepin rinse: The page said the 2014 statement on doxepin rinse for pain was carried forward unchanged. The 2020 update changed doxepin mouthwash and transdermal fentanyl from a suggestion to no guideline possible. (Second accuracy pass; MASCC/ISOO 2020 full text from Europe PMC, read as raw XML) · The note on carried-over statements now names the ones the text lists and says doxepin and fentanyl are no longer suggested.
- 2026-11-17 · Mouth care and oral mucositis prevention during head and neck radiotherapy: sucralfate and chlorhexidine: The page said no sucralfate for prevention of mucositis and gave chlorhexidine as a flat prohibition. The guideline recommends against sucralfate, combined topical and systemic, for preventing or treating mucositis pain, and only suggests that chlorhexidine not be used for prevention. (Second accuracy pass; MASCC/ISOO 2020 full text from Europe PMC, Table 1 read row by row) · The sucralfate step and summary carry the scope and level, and the chlorhexidine step is worded as a suggestion.
- 2026-11-17 · Mouth care and oral mucositis prevention during head and neck radiotherapy: reasons and timings: Several statements were not in the sources: that soreness usually starts in the second or third week, that dental work is safer before the jaw is irradiated, that benzydamine starts with the first fraction, that the light is red, that pain should be treated early and stepped up, and that some centres still give zinc. (Second accuracy pass; MASCC/ISOO 2020 and the 2024 MASCC/ISOO Delphi study, both full text from Europe PMC) · Each was removed or reworded to what the guideline says, and the gaps now name what the guideline paper does not print.
- 2026-11-17 · Mouth care and oral mucositis prevention during head and neck radiotherapy: when to call: The plain summary told patients to tell the team early if it hurts to eat or swallow, with no source. (Second accuracy pass; US National Cancer Institute patient page on mouth and throat problems, read as raw text) · A step and the plain summary now follow the page: call if pain makes it difficult to eat, drink or sleep, or with a fever of 38 °C or higher.
- 2026-11-17 · Mouth care and oral mucositis prevention during head and neck radiotherapy: expert opinion grades: Three steps gave level III to the panel's expert opinion on dental assessment, patient education and bland rinses. Level III is printed for the statements that no guideline was possible, and the opinion itself is ungraded; the education statement was about blood cancer patients. The page also called MASCC/ISOO the only international guideline, when ESMO published one in 2015. (Second accuracy pass; MASCC/ISOO 2020 Table 1 and the Europe PMC record of the ESMO guideline) · The grades now read no guideline possible with expert opinion, the scope is stated, and the ESMO guideline is named as older and not read.
- 2026-11-17 · Dental assessment before head and neck radiotherapy and osteoradionecrosis prevention: the early dental assessment: The step on booking a dental assessment as early as possible quoted a sentence from the US National Cancer Institute summary that is written for stem cell transplant patients, not for head and neck radiotherapy. That summary covers head and neck radiation only in a table. (Second accuracy pass; NCI PDQ Oral Complications re-read as raw HTML, and the Royal College of Surgeons of England dental guidance (February 2026) read in full as a PDF) · The step now cites the UK dental guidance, which says the assessment should be scheduled as early as possible after diagnosis, and the transplant sentence is no longer used.
- 2026-11-17 · Dental assessment before head and neck radiotherapy and osteoradionecrosis prevention: the UK dental guideline: The page said the UK dental guideline on cancer patients could not be found and that no source gave a recommendation on fluoride or on how long to leave between extraction and radiotherapy. The guideline reads today and was updated in February 2026: it says to remove teeth of dubious prognosis at least 10 days before treatment and ideally 14 to 21 days before radiotherapy, recommends fluoride and lifelong follow-up, and says to avoid extraction from irradiated sites where possible. (Second accuracy pass; Royal College of Surgeons of England, Faculty of Dental Surgery guidance (February 2026), full PDF read as text) · The guidance was added as a source with new steps on timing, fluoride, lifelong follow-up and extractions after radiotherapy, each marked as UK guidance that is not individually graded.
- 2026-11-17 · Dental assessment before head and neck radiotherapy and osteoradionecrosis prevention: hyperbaric oxygen: The step said "do not use hyperbaric oxygen routinely" and gave as reasons that trials had shown no clear benefit and that it is time-consuming and expensive. The guideline abstract says only that its use "remains largely unjustified, with limited evidence", and NICE says to consider it only within a clinical trial. (Second accuracy pass; ISOO-MASCC-ASCO 2024 abstract from Europe PMC and NICE NG36 recommendations page) · The order was replaced by two steps in the sources' own terms and the unsourced reasons were removed. The 2026 UK dental guidance, which makes no recommendation on hyperbaric oxygen, is now stated beside them.
- 2026-11-17 · Dental assessment before head and neck radiotherapy and osteoradionecrosis prevention: reasons and patient advice: Several reasons and pieces of patient advice had no source: that radiotherapy lowers the jawbone's ability to heal for life, that patients should tell every dentist about their radiotherapy, that extraction is the main trigger for bone damage, that dead bone does not heal on its own, and a referral to a specialist team attributed to NICE. (Second accuracy pass; all cited sources re-read as raw text, with the Cancer Research UK page on radiotherapy side effects for mouth cancer added) · Each was reworded to what a source read states, or removed. The page now says in its gaps that no source read tells patients in those words to inform every dentist.
- 2026-11-17 · Dental assessment before head and neck radiotherapy and osteoradionecrosis prevention: treating established osteoradionecrosis: The page gave only the NICE position on medicines for osteoradionecrosis (trial only) and described a hyperbaric oxygen course "before and after" extraction that no source read sets out. The 2026 UK dental guidance differs from NICE: it describes pentoxifylline with vitamin E as an option, prints a suggested daily dosage, and says no definitive recommendation can be made without randomised trials. (Second accuracy pass; Royal College of Surgeons of England guidance (February 2026) and NICE NG36, both read in full) · Both positions are now on the page with who holds each, the dosage is attributed as the guidance's suggestion, and the unsourced schedule was removed.
- 2026-11-17 · Bowel effects after pelvic radiotherapy (pelvic radiation disease): assessment and management: referral for leakage: The page told staff to refer faecal incontinence to a specialist team. The guidance prints that referral only for incontinence unrelated to radiotherapy; for leakage with radiation proctopathy and pelvic floor dysfunction it lists exercises, stool bulking and other first-line measures. (Second accuracy pass; Andreyev et al. 2015 (Frontline Gastroenterology), full text read as raw XML from Europe PMC) · Referral line removed and the step now gives the measures the guidance lists for this group.
- 2026-11-17 · Bowel effects after pelvic radiotherapy (pelvic radiation disease): assessment and management: six weeks of exercises: The page gave a minimum of six weeks of pelvic floor exercises for frequency or leakage and for pelvic floor dysfunction in general. The guidance prints the six-week minimum only for radiation proctopathy with frequency of defecation. (Second accuracy pass; Andreyev et al. 2015 (Frontline Gastroenterology), full text read as raw XML from Europe PMC) · The six-week figure is now scoped to proctopathy with frequency in the step and the summary.
- 2026-11-17 · Bowel effects after pelvic radiotherapy (pelvic radiation disease): assessment and management: bile acid malabsorption: The page said to treat moderate or severe malabsorption with a sequestrant and a lower-fat diet, and that binding bile often helps a great deal. The guidance says sequestrants are generally required for moderate malabsorption, that most patients with severe malabsorption need one with long-term fat reduction advice, and gives no statement of how well they work. (Second accuracy pass; Andreyev et al. 2015 (Frontline Gastroenterology), full text read as raw XML from Europe PMC) · Step reworded to the guidance's own scope and hedges and the unsupported reason removed.
- 2026-11-17 · Bowel effects after pelvic radiotherapy (pelvic radiation disease): assessment and management: irritable bowel syndrome: The page ordered staff not to settle for a diagnosis of irritable bowel syndrome because specialist units almost always find a specific cause. The guidance states only that specialist centres very rarely reach a new diagnosis of irritable bowel syndrome in this group. (Second accuracy pass; Andreyev et al. 2015 (Frontline Gastroenterology), full text read as raw XML from Europe PMC) · Step and reason reworded to the stated principle.
- 2026-11-17 · Bowel effects after pelvic radiotherapy (pelvic radiation disease): assessment and management: reasons given for steps: Several reasons were not in the source: that biopsy causes ulcers or fistulas, that radiotherapy stiffens the rectum, that sequestrants cause vitamin deficiency, that stopping blood thinners lets the rectum heal, and that emergencies need same-day action. The guidance gives a general principle of increased risk in irradiated tissue, and says emergencies need immediate action. (Second accuracy pass; Andreyev et al. 2015 (Frontline Gastroenterology), full text read as raw XML from Europe PMC) · Each reason reworded to what the guidance prints.
- 2026-11-17 · Bowel effects after pelvic radiotherapy (pelvic radiation disease): assessment and management: choice of ablation: The page said the UK guidance prefers hyperbaric oxygen and that the ASGE uses endoscopic thermal therapy as the usual first ablative step. The guidance lists advantages and disadvantages of three options without a preference, and the ASGE abstract says nothing about sequence. (Second accuracy pass; Andreyev 2015 full text and the ASGE 2019 abstract from Europe PMC) · Best-in-class statement and the variant reworded and the variant renamed.
- 2026-11-17 · Bowel effects after pelvic radiotherapy (pelvic radiation disease): assessment and management: patient summary: The patient summary said most bowel effects have a cause that can be found and treated and listed the cancers concerned. No cited source says either. (Second accuracy pass; Andreyev et al. 2015 (Frontline Gastroenterology), full text read as raw XML from Europe PMC; Cancer Research UK page read as text) · Summary reworded to the guidance's principles, with the advice to report bleeding taken from Cancer Research UK, now cited.
- 2026-11-17 · Skin care during radiotherapy and acute radiation dermatitis: film dressings, steroid cream and photobiomodulation: The page gave only the MASCC position and said the UK radiographers' guideline could not be read. That guideline (second revised edition, April 2020) now reads in full. It recommends steroid cream only for patients at high risk of high-grade dermatitis and not as early prevention in breast radiotherapy, found the film and dressing studies not significant enough to recommend a change in practice, and asked for more research before photobiomodulation is used widely. A 2025 ESTRO radiation therapist committee paper, not on the page, grades each of these. (Second accuracy pass; Society and College of Radiographers 2020 guideline, PDF from sor.org read as text; ESTRO RTT 2025, Europe PMC full text) · Both sources added and read in full. The summary, best-in-class statement, variants and grades now state the MASCC, UK and ESTRO positions side by side with the ESTRO grades.
- 2026-11-17 · Skin care during radiotherapy and acute radiation dermatitis: washing: The page told people to wash with mild soap. No source read says mild soap. The meta-analysis says washing with or without soap, and the UK guideline says there should be no restriction to a specific type of soap. (Second accuracy pass; Chan 2023 abstract from Europe PMC; Society and College of Radiographers 2020 guideline read as text) · Reworded to washing gently with usual products or soap and water, in the patient summary, the step and the best-in-class statement.
- 2026-11-17 · Skin care during radiotherapy and acute radiation dermatitis: deodorant: The deodorant step rested on a meta-analysis abstract that does not carry the exception the step stated, and it left out that deodorant should stop if the skin breaks. The UK guideline says to continue normal deodorant unless it irritates and to discontinue if the skin is broken. ESTRO 2025 says deodorant is allowed unless the skin is broken. (Second accuracy pass; ESTRO RTT 2025 full text from Europe PMC; Society and College of Radiographers 2020 guideline read as text) · Step re-sourced to ESTRO 2025 with its printed grade, with both conditions stated. The meta-analysis is kept in the grades and scoped to breast radiotherapy.
- 2026-11-17 · Skin care during radiotherapy and acute radiation dermatitis: when to stop a steroid cream: The step said to stop a steroid cream if the skin starts to weep or peel. The MASCC statement says to discontinue it if moist desquamation or other complications arise. Peeling alone is wider than that. (Second accuracy pass; MASCC 2026 steroid statement abstract from Europe PMC) · Reworded to moist desquamation (broken, weeping skin) or another complication.
- 2026-11-17 · Skin care during radiotherapy and acute radiation dermatitis: dressings for established dermatitis: The step said to use a soft silicone foam dressing such as Mepilex Lite especially for moist desquamation, and gave reasons no source read states. The MASCC abstract says only that Mepilex Lite dressings were recommended for management. (Second accuracy pass; MASCC 2023 Delphi abstract and MASCC systematic review full text from Europe PMC; Society and College of Radiographers 2020 guideline read as text) · Step reduced to what MASCC says, with the review finding as the reason. A separate step for broken skin now quotes the UK guideline on non-adhesive or low-adhesion silicone dressings.
- 2026-11-17 · Skin care during radiotherapy and acute radiation dermatitis: barrier-forming agents: The step called the product a gel and explained how it works. The MASCC abstract says barrier-forming topical agents, for patients at considerable risk of high-grade dermatitis, particularly at irregular contours, and gives no mechanism. The summary had left out the risk condition. (Second accuracy pass; MASCC 2026 barrier statement abstract from Europe PMC) · Reworded to the source's terms and scope. The unsourced reason was removed.
- 2026-11-17 · Skin care during radiotherapy and acute radiation dermatitis: reasons given for steps: Several reasons were not in any source read: that film dressings only help if started at the beginning, that a film protects from friction, that steroid calms the inflammation behind redness and itching, and that photobiomodulation needs a device and staff time. The risk assessment and the peak after treatment were cited to abstracts that do not state them. (Second accuracy pass; all cited abstracts from Europe PMC; Society and College of Radiographers 2020 guideline read as text) · Unsourced reasons removed or replaced with findings the sources print. Risk assessment and the peak around 10 to 14 days after the last session are now their own steps, quoted from the UK guideline.
- 2026-11-17 · Skin care during radiotherapy and acute radiation dermatitis: variants: The variants said that older information told people to keep the skin dry for the whole course, that some departments give steroid cream to everyone from the first day, and that many departments and patients use aloe vera and similar creams. No source read says these things. The UK guideline describes traditional advice not to wash or to use water without soap, and reports a 2014 UK survey in which 74% of departments restricted washing and 55% advised against deodorant. (Second accuracy pass; Society and College of Radiographers 2020 guideline read as text) · Variants reworded to what the sources describe, with the 2014 UK survey figures given as reported.
- 2026-11-17 · Heart risk assessment and monitoring before, during and after heart-toxic cancer treatment: the main guideline: The page said the 2022 ESC cardio-oncology guideline could not be read, so its graded recommendations and scan intervals were missing. It reads today: for example an echocardiogram every two cycles on anthracyclines at high or very high risk (Class I, Level C) and every 3 months on HER2-targeted therapy given before or after surgery (Class I, Level B). (Second accuracy pass; 2022 ESC guideline, published PDF from an open university repository read as text) · Added the ESC recommendations with their class and level to the steps, summary and best-in-class statement, and marked the source as read in full.
- 2026-11-17 · Heart risk assessment and monitoring before, during and after heart-toxic cancer treatment: baseline tests before immunotherapy and CAR-T: The page gave a baseline echocardiogram for everyone starting checkpoint inhibitors or CAR-T as the only position, and called heart inflammation from these treatments rare. HFA-ICOS 2020 do recommend it for all, citing severity and giving no frequency; ESC 2022 recommends the echocardiogram for high-risk patients before checkpoint inhibitors and for people with existing cardiovascular disease before CAR-T, and says it may be considered for all before checkpoint inhibitors. (Second accuracy pass; HFA-ICOS 2020 full text from Europe PMC and ESC 2022 PDF text) · Both positions are now stated with the ESC class and level, and the word rare was removed.
- 2026-11-17 · Heart risk assessment and monitoring before, during and after heart-toxic cancer treatment: three-monthly scans on HER2 therapy: The page said ESMO 2020 scans every 3 months regardless of baseline risk. As BSE/BCOS tabulate it, ESMO says every 3 months and that higher-risk patients may require more frequent monitoring. (Second accuracy pass; BSE/BCOS 2021 Table 4, full text from the NCBI E-utilities API) · Removed the phrase and added the ESMO qualification and the ESC 2022 schedule.
- 2026-11-17 · Heart risk assessment and monitoring before, during and after heart-toxic cancer treatment: imaging for survivors: The page told teams to offer imaging surveillance to higher-risk survivors and said no source read set a schedule. The ASCO 2017 abstract says routine imaging may be warranted for certain higher-risk survivors; BSE/BCOS recommend an echocardiogram 3 to 12 months after treatment in all patients; ESC 2022 sets a yearly review and scan intervals by risk. (Second accuracy pass; ASCO 2017 abstract from Europe PMC, BSE/BCOS 2021 full text, ESC 2022 PDF text) · Restored the ASCO wording, removed the gap and added the BSE/BCOS and ESC follow-up recommendations with their grades.
- 2026-11-17 · Heart risk assessment and monitoring before, during and after heart-toxic cancer treatment: reasons given for steps: Five reasons were not in the cited sources, among them that for most people the cancer is the greater threat, and that many people with an ejection fraction of 40 to 49 per cent can finish treatment safely. BSE/BCOS say only that there are increasing data to suggest this. (Second accuracy pass; HFA-ICOS 2020, IC-OS 2022 and BSE/BCOS 2021 full texts, ASCO 2017 abstract) · Each reason was replaced with what the cited source or the ESC guideline says.
- 2026-11-17 · Heart risk assessment and monitoring before, during and after heart-toxic cancer treatment: symptoms to report: The short summary told readers to tell any doctor about new breathlessness, swelling or chest pain, and no source read stated that. ESC 2022 recommends teaching patients to recognise early signs of cardiovascular disease, says survivors should report them promptly, and names breathlessness, ankle swelling and fatigue as the cardinal symptoms of heart failure. (Second accuracy pass; ESC 2022 PDF text) · Reworded to the ESC advice and added it as a cited step (Class I, Level C).
- 2026-11-17 · Heart scans after anthracyclines or chest radiotherapy for cancer in childhood, adolescence or young adulthood: the cost-effectiveness variant: The page headed this variant as supporting longer intervals for high-risk survivors. The paper says the opposite: it supports scanning high-risk survivors more often than every 5 years, found the 2-year interval the strategy most often preferred, and found no strategy cost-effective for low-risk survivors. (Second accuracy pass; Ehrhardt et al., J Clin Oncol 2020, full text read from PubMed Central (the first pass had the abstract only)) · Variant renamed and rewritten: scans every 2 years for high-risk survivors and none for low-risk survivors, with the 2015 risk groups and the model limits stated.
- 2026-11-17 · Heart scans after anthracyclines or chest radiotherapy for cancer in childhood, adolescence or young adulthood: grade row on borderline heart function: The claim said an ejection fraction between 40 per cent and the lower end of normal. The source prints between 40% and the upper limit of normal. (Second accuracy pass; IGHG cardiomyopathy surveillance recommendations 2022, PDF read with pdftotext) · The claim now gives the source words.
- 2026-11-17 · Heart scans after anthracyclines or chest radiotherapy for cancer in childhood, adolescence or young adulthood: steps on cardiac MRI, the moderate-risk schedule, lifelong scans and pregnancy: Five steps were written as orders (use, do, keep, offer, keep scanning). IGHG words each as reasonable or may be reasonable, a moderate recommendation, and the lifelong and MRI ones rest on expert opinion. (Second accuracy pass; IGHG cardiomyopathy surveillance recommendations 2022, PDF read with pdftotext) · Each step now says consider, and its reason gives the IGHG wording and grade. The summary gives the moderate-risk schedule as reasonable.
- 2026-11-17 · Heart scans after anthracyclines or chest radiotherapy for cancer in childhood, adolescence or young adulthood: reasons given for the steps: Sixteen reasons were not in any source read: why the panel ended low-risk scans (worry and extra tests), how pregnancy loads the heart, what a cardiologist does, that medicines reverse weakening when started early, that exercise protects the heart, and others. (Second accuracy pass; IGHG 2022 recommendations and COG Long-Term Follow-Up Guidelines Version 6.0 sections 34 and 77, PDFs read with pdftotext) · Each reason replaced with what IGHG or COG states. The page now says the documents read do not print the reasons of the panel for ending low-risk scans.
- 2026-11-17 · Heart scans after anthracyclines or chest radiotherapy for cancer in childhood, adolescence or young adulthood: symptom questions and telling the GP: The page said to ask at each visit and about nausea in children and young adults, and to write to the GP. COG says yearly and gives nausea and vomiting for those under 25. IGHG says health care providers should be aware and names no GP letter. (Second accuracy pass; COG Version 6.0 section 34 and IGHG 2022 recommendations, PDFs read with pdftotext) · Reworded to each year, under 25, and the health professionals who look after the survivor.
- 2026-11-17 · Heart scans after anthracyclines or chest radiotherapy for cancer in childhood, adolescence or young adulthood: COG variant and best in class: The page said IGHG uses the prescribed chest dose, that COG follows IGHG, that COG suggests a coronary consultation, and gave yearly history and examination without the 15 Gy condition of the radiation section. IGHG says only chest RT dose. COG prints the same bands without saying it follows IGHG, says the consultation may be reasonable, and limits the radiation section checks to 15 Gy or more. (Second accuracy pass; COG Version 6.0 and IGHG 2022 recommendations, PDFs read with pdftotext) · Reworded to what each source prints.
- 2026-11-17 · Heart scans after anthracyclines or chest radiotherapy for cancer in childhood, adolescence or young adulthood: ESC 2022 guideline: The page said the ESC 2022 cardio-oncology guideline could not be read and compared nothing with it. It now reads from an open repository copy: for adult survivors it recommends follow-up according to IGHG and says echocardiography should be considered every 2 years for high risk and every 5 years for moderate risk (Class IIa, Level B). (Second accuracy pass; 2022 ESC Guidelines on cardio-oncology, published PDF from a university repository, section 8.1.1 and Recommendation Table 40) · Source marked as read, and a variant, a grade row and a revised gap added.
- 2026-11-17 · Heart scans after anthracyclines or chest radiotherapy for cancer in childhood, adolescence or young adulthood: who the page covers: The page said valve disease after radiotherapy has its own IGHG guideline. The only such guideline named in a source read is for coronary artery disease. (Second accuracy pass; ESC 2022 reference list (IGHG coronary artery disease surveillance, 2021)) · Reworded so only coronary artery disease is tied to an IGHG guideline.
- 2026-11-17 · Chemotherapy nerve damage (peripheral neuropathy): compression variant: The page said the POLAR trial did not compare cooling and compression with each other. It did: people were randomised to one or the other, and the paper reports no significant difference between the two. (Second accuracy pass; POLAR (JAMA Oncology 2025), full text from PubMed Central read as raw text) · The sentence now says people were randomised between the methods and that no significant difference was found.
- 2026-11-17 · Chemotherapy nerve damage (peripheral neuropathy): who the trials were in: The page said CryoPac was in women with breast cancer, and singled out oxaliplatin among platinum drugs. The CryoPac abstract names neither sex nor cancer, and no source read singles out oxaliplatin. (Second accuracy pass; CryoPac abstract from Europe PMC and the Australian consensus paper, read as raw text) · Population reworded to what the abstract says; the oxaliplatin phrase removed.
- 2026-11-17 · Chemotherapy nerve damage (peripheral neuropathy): steps written as orders: Four steps were firmer than their sources: plan an assessment where the consensus says should be considered, refer where it says may warrant referral, offer duloxetine with good evidence where the ASCO abstract says appropriate evidence and limited benefit, and consider cooling or compression on the strength of one trial that no guideline has graded. (Second accuracy pass; Australian CIPN pathway consensus (full text, Europe PMC), ASCO 2020 abstract, POLAR full text) · Each step now carries the source's own hedge, and the cooling step says it is a trial-tested option, not a guideline recommendation.
- 2026-11-17 · Chemotherapy nerve damage (peripheral neuropathy): reasons given: Several reasons were not in any cited source: that a dose change in time prevents lasting harm, that cooling is thought to reduce drug reaching the small nerves (POLAR gives no mechanism), that symptoms are easy to dismiss, and that offering cooling is reasonable where tolerated. (Second accuracy pass; all cited sources read as raw text, and the Macmillan peripheral neuropathy page added) · Reasons replaced with what the sources say; the cooling mechanism is now attributed to Macmillan, and the advice to patients follows Macmillan's wording.
- 2026-11-17 · Food safety during cancer treatment and the evidence against neutropenic diets: the Ma 2022 meta-analysis: The page called it a meta-analysis of 6 randomised trials. The paper pooled 6 studies in 1,114 patients, of which 4 used randomisation and one was a case-control study, and its mortality result came from 2 studies. (Second accuracy pass; Ma 2022 (Frontiers in Oncology), full text from Europe PMC read as raw text) · The summary and the grades table now say 6 studies, 4 of them randomised, and that mortality was similar in the 2 studies that reported it.
- 2026-11-17 · Food safety during cancer treatment and the evidence against neutropenic diets: how long the 2009 transplant diet lasts: The summary and best in class statement said the food-safety diet after engraftment runs until immunosuppression stops. The guideline says 3 months after autologous transplant and, after allogeneic transplant, until all immunosuppressive drugs are stopped and live virus vaccines have been given, with the transplant doctor deciding. (Second accuracy pass; CDC, IDSA and ASBMT HCT guideline 2009, full text from Europe PMC read as raw text) · Both durations and the doctor's final say are now stated.
- 2026-11-17 · Food safety during cancer treatment and the evidence against neutropenic diets: reheating leftovers: The step said to reheat leftovers until steaming hot throughout. The guideline gives a temperature instead, above 74 °C (165 °F) throughout, and uses steaming only for hot dogs and deli meats. (Second accuracy pass; CDC, IDSA and ASBMT HCT guideline 2009, full text from Europe PMC read as raw text) · The step now says to reheat right through and gives the guideline's figure as the guideline's, noting that local advice may word it differently.
- 2026-11-17 · Food safety during cancer treatment and the evidence against neutropenic diets: reasons given for the food handling steps: Six reasons said things no cited source says: that teaching works best before the first course, that raw meat is the commonest source of food poisoning, that washing removes most germs, that colour is an unreliable guide, that germs multiply at room temperature, and that every guideline agrees on local advice. (Second accuracy pass; HCT guideline 2009 and POGO guideline 2025, full texts from Europe PMC read as raw text) · Each reason now says what its source says, and the page states that the handling steps come from a guideline written for transplant recipients.
- 2026-11-17 · Food safety during cancer treatment and the evidence against neutropenic diets: variants and summary wording: The page said some leukaemia units restrict uncooked foods, that many centres still use a protective diet, that the pooled trials show no overall benefit, that the 2009 handling advice is what most teaching uses, and that food safety advice agrees everywhere on a list of foods. The sources give one 2008 trial, a survey from 2015 and 2016, six trials with no consistent improvement and no pooling, nothing on teaching, and a statement that advice should be consistent. (Second accuracy pass; POGO guideline 2025 and Nutrients 2025 review, full texts from Europe PMC read as raw text) · Each statement was reworded to its source or removed.
- 2026-11-17 · Protective measures and daily life during neutropenia: sport and low platelets: One step joined two recommendations and printed the weaker grade for both, and gave the platelet figure without its unit. The Dutch guideline grades encouraging sport as a strong recommendation on expert opinion, and the limit on contact sports as weak on expert opinion, for a platelet count below 50 x 10^9/L. (Second accuracy pass; Dutch Children's Oncology Group guideline 2024, full text from Europe PMC read as raw text) · Split into two steps, each with its own grade, the first quoted, the second with the unit.
- 2026-11-17 · Protective measures and daily life during neutropenia: gardening and soil after transplant: The page graded the transplant guideline's advice to avoid gardening and soil as CIII only and wrote the mask and gloves as an instruction. The guideline's safe living section grades avoiding gardening and soil in the first 6 months and during substantial immunosuppression as BIII; CIII is the grade in its Nocardia section, where it says those who continue gardening may reduce exposure with a mask and gloves. (Second accuracy pass; CDC, IDSA and ASBMT guideline 2009, full text from Europe PMC read as raw text) · Both grades are given with their sections, and the protective gear is worded as an option.
- 2026-11-17 · Protective measures and daily life during neutropenia: cat litter after transplant: The page said someone else should change the cat litter for the first 6 months. The guideline says the first 6 months and also during periods of substantial immunosuppression. (Second accuracy pass; CDC, IDSA and ASBMT guideline 2009, full text from Europe PMC read as raw text) · The second period was added.
- 2026-11-17 · Protective measures and daily life during neutropenia: reasons given for each step: Nine reasons said more than the guidelines do: that hands carry most germs, that mould spores cause serious lung infection, that the infection risk at school is small, that no study found pets raise infection, that swimming was studied in children with tunnelled lines, that crowds spread respiratory viruses, and an outbreak on a cancer ward credited to the Dutch guideline when it is in the transplant guideline. The Dutch panel gives narrower reasons, mostly one small study or its own judgement. (Second accuracy pass; Dutch Children's Oncology Group guideline 2024 and CDC, IDSA and ASBMT guideline 2009, full texts read as raw text) · Each reason was reworded to what the panel wrote.
- 2026-11-17 · Protective measures and daily life during neutropenia: who the advice covers: The short summary told every reader to avoid building sites and soil and listed visitors among those the Dutch guideline asks to clean their hands. The Dutch recommendation on building sites and soil is for children with neutropenia and prolonged contact, its hand hygiene statement names patients, caregivers and medical personnel, and the visitor rules are the transplant guideline's. (Second accuracy pass; Dutch Children's Oncology Group guideline 2024 and CDC, IDSA and ASBMT guideline 2009, full texts read as raw text) · The summary and the steps now carry each source's own scope.
- 2026-11-17 · Protective measures and daily life during neutropenia: best in class and transplant rooms: The page called the Dutch guideline the only evidence-graded guideline on the subject, and said transplant centres place their highest-risk patients in filtered rooms. The authors say only that it is, to their knowledge, the first; the transplant guideline says all recipients should ideally have such rooms, with priority for the highest risk when rooms are short. (Second accuracy pass; both guidelines, full texts read as raw text) · Both statements were reworded to the sources.
- 2026-11-17 · MRI safety screening for implants, devices and medicine patches: patient summary: The summary said most ports can be scanned safely and told readers to bring an implant card. The MHRA guidance names no port and gives no such instruction to patients; it says device details should be available before the appointment. (Second accuracy pass; MHRA Safety Guidelines for MRI Equipment in Clinical Use, version 5.0, PDF read as text) · Removed the claim about ports and the instruction; the page now says to tell the unit about any implant at booking, and a gap says no source read gives the implant card advice.
- 2026-11-17 · MRI safety screening for implants, devices and medicine patches: medicine patches: A step ordered staff to remove medicine patches that contain metal. The guidance says sites should have a documented process for the removal and reapplication or disposal of patches. (Second accuracy pass; MHRA Safety Guidelines for MRI Equipment in Clinical Use, version 5.0, PDF read as text) · Reworded the step, summary and best in class statement to the documented site process.
- 2026-11-17 · MRI safety screening for implants, devices and medicine patches: generic implant safety procedures: The page said these procedures are for low-risk passive implants. The guidance says low-risk implants and does not limit them to passive ones. (Second accuracy pass; MHRA Safety Guidelines for MRI Equipment in Clinical Use, version 5.0, PDF read as text) · Removed the word passive in three places.
- 2026-11-17 · MRI safety screening for implants, devices and medicine patches: clothing and removable items: A step said to remove anything removable and change the person into unit clothing. The guidance says to remove items where it is safe to do so and that changing is the ideal. (Second accuracy pass; MHRA Safety Guidelines for MRI Equipment in Clinical Use, version 5.0, PDF read as text) · Restored both qualifications.
- 2026-11-17 · MRI safety screening for implants, devices and medicine patches: off-label risk assessment team: The summary listed a specialist such as a cardiologist as part of the team. The guidance sets a minimum of an MR radiographer, an MR safety expert and the MR clinician and says a specialist may be appropriate. (Second accuracy pass; MHRA Safety Guidelines for MRI Equipment in Clinical Use, version 5.0, PDF read as text) · Reworded to the minimum team, with a specialist where appropriate.
- 2026-11-17 · MRI safety screening for implants, devices and medicine patches: reasons given for steps: Three reasons were not in the guidance: that things change between scans, that a sedated person cannot warn staff, and that many common implants are safe under known conditions. (Second accuracy pass; MHRA Safety Guidelines for MRI Equipment in Clinical Use, version 5.0, PDF read as text) · Replaced each with the reason the guidance gives.
- 2026-11-17 · MRI safety screening for implants, devices and medicine patches: generative AI: The grades table said do not rely on generative AI for MR safety advice. The guidance says caution is advised. (Second accuracy pass; MHRA Safety Guidelines for MRI Equipment in Clinical Use, version 5.0, PDF read as text) · Reworded the claim to caution.
- 2026-11-17 · ReSPECT plans and DNACPR decisions in the UK: explaining a decision where CPR would not work: The summary, a step and a grade row said such a decision must be explained. The guidance says there must be a presumption in favour of explaining it, and that telling a patient may be delayed or avoided where it is likely to cause physical or psychological harm. (Second accuracy pass; the joint BMA, Resuscitation Council UK and RCN guidance (2016), full PDF text read raw) · Reworded to the presumption, with the exception for harm, in all three places.
- 2026-11-17 · ReSPECT plans and DNACPR decisions in the UK: when no decision is recorded: The page said the absence of a plan must never be read as a decision not to treat. The guidance sets an initial presumption in favour of CPR but adds that a carefully considered decision not to start inappropriate CPR should be supported, for example in the advanced stages of a terminal illness. (Second accuracy pass; the joint BMA, Resuscitation Council UK and RCN guidance (2016), full PDF text read raw) · Reworded to state the presumption and its exception.
- 2026-11-17 · ReSPECT plans and DNACPR decisions in the UK: what an attorney can decide: The page said flatly that an attorney with authority over life-sustaining treatment can decide. The NHS website for England says so; the joint guidance says only that it is likely, and untested in the courts. (Second accuracy pass; the NHS website DNACPR page and the 2016 joint guidance, both read raw) · Each source is now reported in its own words and the NHS statement is scoped to England.
- 2026-11-17 · ReSPECT plans and DNACPR decisions in the UK: blanket decisions: The page called group rules discriminatory, which is not what the sources say. The joint guidance calls blanket policies unethical and probably unlawful; the NHS website for England says they are unlawful. (Second accuracy pass; the 2016 joint guidance (section 3) and the NHS website, read raw) · Reworded to both sources, each attributed.
- 2026-11-17 · ReSPECT plans and DNACPR decisions in the UK: reasons given for five steps: Reasons were given that no source states: that a calm conversation gives a better decision, that people cannot weigh treatment without knowing their illness, that a DNACPR decision misread as wider withdrawal has caused harm (the guidance cites a common public fear), what a second opinion achieves, and that illness, age or disability do not mean incapacity. (Second accuracy pass; the 2016 joint guidance and the ReSPECT clinicians guide, read raw) · Each reason replaced with what the cited source says.
- 2026-11-17 · ReSPECT plans and DNACPR decisions in the UK: law by UK nation: The steps on capacity and on advance refusals did not say which nation the law applies to, and a right to be told was stated without a source. Advance decisions are statutory in England and Wales only; the guidance says they are likely to be binding under common law in Scotland and Northern Ireland; the right to be informed is stated by the NHS website for England. (Second accuracy pass; the 2016 joint guidance (sections 9 and 10) and the NHS website, read raw) · Nation named in each legal statement.
- 2026-11-17 · ReSPECT plans and DNACPR decisions in the UK: reports not read: The page said the sources name a Care Quality Commission review of 2021. The Resuscitation Council UK policy briefing names the Commission interim report of 2020 and the Ombudsman report of 2024. (Second accuracy pass; the Resuscitation Council UK policy briefing (February 2026), PDF text) · Named as the briefing names them.
- 2026-11-17 · Insomnia in adults with cancer: who CBT-I is recommended for: The page told readers to offer CBT-I, online CBT-I and mindfulness therapy first without saying who the recommendation covers. ESMO makes all three recommendations for cancer survivors and says it remains unclear whether CBT-I is effective in patients in active treatment. (Second accuracy pass; ESMO 2023 full text, read as raw XML from Europe PMC) · The steps, summary and best-in-class statement now say the recommendations are for survivors and state the uncertainty during active treatment.
- 2026-11-17 · Insomnia in adults with cancer: brief behavioural therapy: The page said brief behavioural therapy is recommended by ESMO for patients in active treatment. ESMO says it should be considered, and calls the evidence weak. (Second accuracy pass; ESMO 2023 full text, read as raw XML from Europe PMC) · Reworded to "should be considered" with ESMO's description of the evidence.
- 2026-11-17 · Insomnia in adults with cancer: reasons given for each step: Several reasons went beyond the guideline: that insomnia can last for years, that CBT-I has lasting effects, that a severity score shows whether treatment is working, and that exercise improves sleep. ESMO says insomnia frequently follows a chronic course, gives no durability claim for CBT-I, attributes tracking of treatment effects to the sleep diary, and rates the exercise evidence as weak to moderate. (Second accuracy pass; ESMO 2023 full text, read as raw XML from Europe PMC) · Each reason reworded to what ESMO states, including the Insomnia Severity Index score bands it prints.
- 2026-11-17 · Insomnia in adults with cancer: sleep hygiene and the clinical interview: The page described sleep hygiene as advice on screens and the bedroom, and the interview as looking for pain, low mood, anxiety and fatigue. ESMO lists caffeine and alcohol late in the day, daytime naps and irregular bedtimes for sleep hygiene, and for the interview lists night-time symptoms, daytime symptoms, other conditions and daily habits. (Second accuracy pass; ESMO 2023 full text, read as raw XML from Europe PMC) · Both descriptions replaced with the items ESMO lists.
- 2026-11-17 · Insomnia in adults with cancer: simple summary, trigger and gaps: The simple summary called CBT-I a short course that helps most, the trigger added "whenever a person reports poor sleep", and a gap implied a caution about sleep restriction in people at risk of falls or seizures. None of these is in a source read. (Second accuracy pass; ESMO 2023 full text and supplement, read as raw text) · Simple summary reworded, the trigger trimmed to ESMO's wording, and the gap rewritten to say what ESMO does and does not cover.
- 2026-11-17 · Chronic pain in cancer survivors: assessment, non-opioid treatment first and careful opioid use: the psychosocial and exercise step: One step put cognitive behavioural therapy, exercise and bringing in other professionals under a single grade of 1B. The Japanese statement grades psychosocial approaches 1B and exercise 1B in separate sentences, cognitive behavioural therapy 2C and a multidisciplinary approach 2C; involving other professionals for complex needs is an ASCO 2016 recommendation. (Second accuracy pass; Japan Society of Pain Clinicians 2025 full text from Europe PMC and the ASCO 2016 abstract, read as raw text) · Split into three steps, each with its own source, quotation and grade.
- 2026-11-17 · Chronic pain in cancer survivors: assessment, non-opioid treatment first and careful opioid use: non-opioid medicines: The step named NSAIDs, paracetamol, gabapentinoids and duloxetine and cited ASCO 2016, whose abstract names no drugs. The drugs and their weak 2B grades come from the Japanese statement. (Second accuracy pass; ASCO 2016 abstract and JSPC 2025 full text from Europe PMC) · The drugs are now attributed to the Japanese statement with their grades, and its note that NSAIDs and paracetamol are unproven for nerve pain is added.
- 2026-11-17 · Chronic pain in cancer survivors: assessment, non-opioid treatment first and careful opioid use: Japanese opioid limits: The best-in-class statement said the Japanese statement caps opioid dose and duration for survivors. The 60 mg and 90 mg a day limits and the 3 and 6 month limits apply only to pain not caused by cancer; pain caused by cancer follows the WHO analgesic guidelines. (Second accuracy pass; JSPC 2025 full text from Europe PMC) · The limits are scoped to pain not caused by cancer wherever they appear, and a gap says they are Japan's rule and not a general limit.
- 2026-11-17 · Chronic pain in cancer survivors: assessment, non-opioid treatment first and careful opioid use: reasons given for each step: Seven reasons were not in any source read, among them that pain after cancer is common and lasts for years, that acupuncture reduced joint pain in trials and helps people stay on hormone treatment, and a list of opioid risks credited to the ASCO abstract. (Second accuracy pass; the three ASCO abstracts and JSPC 2025 full text from Europe PMC) · Each reason now says what its source says; the opioid risks are those the Japanese statement lists; gaps name what no source read states.
- 2026-11-17 · Chronic pain in cancer survivors: assessment, non-opioid treatment first and careful opioid use: who it covers and the patient summary: The page said the guidance covers adults after treatment or living long term with cancer, and the patient summary listed physiotherapy. The Japanese statement defines a survivor as anyone from diagnosis to the end of life, and no source read names physiotherapy. (Second accuracy pass; JSPC 2025 full text and the ASCO 2016 abstract from Europe PMC) · Population rewritten with each source's own scope; physiotherapy removed.
- 2026-11-17 · Chronic pain in cancer survivors: assessment, non-opioid treatment first and careful opioid use: best-in-class statement: It said no recent graded guideline covers chronic pain in survivors, although the Japanese statement of 2025 is graded and covers exactly that. (Second accuracy pass; JSPC 2025 full text from Europe PMC) · Reworded to say the Japanese statement is the most recent and the only one read with a grade on each recommendation.
- 2026-11-17 · Bowel preparation for colonoscopy: instructions step: The step told units to give instructions "spoken and written" to every patient and to ward staff, quoting an ESGE recommendation that says "verbal or written" and sits under the heading on inpatients. Its reason, that most poor preparations come from unclear instructions, is not in the guideline. (Second accuracy pass; ESGE 2019 guideline PDF from esge.com, read as raw text) · Split into a general step on enhanced instructions and an inpatient step worded "spoken or written", each with its own quote, and the reason replaced by the pooled trial finding.
- 2026-11-17 · Bowel preparation for colonoscopy: enemas and prokinetics: One step ruled out routine enemas and routine prokinetic drugs together under the grade "strong, moderate". ESGE grades the prokinetic statement "weak, low", and the reason given cited trials where ESGE cites one. (Second accuracy pass; ESGE 2019 guideline PDF, read as raw text) · The step now covers enemas only and cites one randomised trial; the prokinetic statement stays in the grades list with its own grade.
- 2026-11-17 · Bowel preparation for colonoscopy: after an inadequate preparation: The step ordered a repeat "within a year, or sooner after extra preparation, with the next regimen adjusted" under the grade "strong, moderate". Only the repeat within 1 year carries that grade; ESGE says a same-day or next-day repeat "may be suggested" as a weak recommendation on very low evidence. (Second accuracy pass; ESGE 2019 guideline PDF, read as raw text) · The action is now the repeat within 1 year unless clinically contraindicated; the weaker advice is given in the reason and in the grades list with its own grade.
- 2026-11-17 · Bowel preparation for colonoscopy: how the dose is split: The page said to take half the laxative the evening before and half on the day, citing ESGE, which never defines the split. The US Task Force calls that the most common convention and notes other splits are used. (Second accuracy pass; ESGE 2019 PDF and the US Task Force 2025 full text (American Journal of Gastroenterology copy), read as raw text) · The step and the patient summary now say a split dose with the last part on the day; the convention is described in the gaps with its source.
- 2026-11-17 · Bowel preparation for colonoscopy: how the two guidelines differ: The page said ESGE addresses same-day dosing only for afternoon lists, when its text says the data favour split dosing for a morning colonoscopy. It also left out a difference on timing: ESGE says start the last dose within 5 hours, the US Task Force says begin the second portion 4 to 6 hours before. (Second accuracy pass; both guidelines read as raw text, the US one in full for the first time) · The comparison is rewritten with each body's wording and grade on timing, diet, volume and morning colonoscopy.
- 2026-11-17 · Bowel preparation for colonoscopy: laxative choice for people at risk: The step named frailty as a risk group, which ESGE does not; it names chronic kidney insufficiency, congestive heart failure and liver failure with ascites. A variant said the relative safety of each agent in kidney or heart failure had not been read, although the guideline read in full states it. (Second accuracy pass; ESGE 2019 guideline PDF, read as raw text) · The risk groups are ESGE's own, and the variant now reports what ESGE and the US Task Force say about PEG and hyperosmotic regimens in these groups.
- 2026-11-17 · Bowel preparation for colonoscopy: patient summary and reasons: The summaries said a colonoscopy "can only find" polyps if the bowel is clean and that a clean bowel means "nothing is missed"; the guideline reports lower detection and more missed adenomas, not certainty. Four reasons gave mechanisms the guideline does not state (fibre residue hiding the lining, bubbles hiding small polyps, avoiding an evening dose, fluid shifts mattering most for weak hearts). (Second accuracy pass; ESGE 2019 guideline PDF, read as raw text) · Reworded to fewer missed polyps, and each reason replaced by the trial finding ESGE reports.
- 2026-11-17 · Bowel preparation for colonoscopy: US Task Force grades: The US Task Force entry was read as an abstract only and its grade was described as "GRADE terms", a word the abstract does not use. The full text grades each statement strong or weak on high to very low quality evidence by committee consensus. (Second accuracy pass; US Task Force 2025 full text, American Journal of Gastroenterology copy, tags stripped by script) · Six US recommendations are now quoted with their printed grades, and the source is marked as read in full.
- 2026-11-17 · Routine tests before planned surgery, by size of operation and fitness: who it covers: The page said the guideline does not cover lung cancer surgery. NICE NG45's overview excludes cardiothoracic procedures, but its recommendations list lung operations among the examples of major or complex surgery, so the source does not support that statement. (Second accuracy pass; NICE NG45 overview and recommendations pages, fetched raw and read as text) · Removed the statement and named the unexplained difference in the gaps.
- 2026-11-17 · Routine tests before planned surgery, by size of operation and fitness: clotting tests: The page said to consider clotting tests in chronic liver disease and to make an individual plan for anyone on anticoagulants, with no limit. NICE's tables say this only for ASA grade 3 or 4 before intermediate, major or complex surgery, and ask for an individual plan, in line with local guidance, if anticoagulant treatment needs modifying. Before minor surgery clotting tests are not routine at any ASA grade. (Second accuracy pass; NICE NG45 Tables 1 to 3, fetched raw and read as text) · Step, grade and summary now carry the ASA grade, the surgery grade and the condition.
- 2026-11-17 · Routine tests before planned surgery, by size of operation and fitness: ECG in healthy people over 65: The step said to consider an ECG before major surgery in fit people over 65. NICE's table adds a condition: if no ECG results are available from the past 12 months. (Second accuracy pass; NICE NG45 Table 3, fetched raw and read as text) · Condition added to the step and its grade.
- 2026-11-17 · Routine tests before planned surgery, by size of operation and fitness: reasons given: Seven reasons were given that the NICE pages read do not state, for example that diuretics make particular tests worthwhile, that poorly controlled diabetes can be improved before surgery, and that some cancer surgery would be planned differently in pregnancy. The summary also said people with cancer often have recent blood tests. (Second accuracy pass; NICE NG45 recommendations and context pages, fetched raw and read as text) · Reasons reworded to what NICE states; the unsourced sentences removed and the gap named.
- 2026-11-17 · CT-guided and ultrasound-guided needle biopsy of the lung: clotting thresholds: The page said the INR below 1.5 and the platelet count of 70,000 per microlitre were the SIR 2019 consensus as the Korean guideline reports it. The Korean guideline cites the SIR 2012 and 2019 documents together for both figures, and neither SIR document could be read. (Second accuracy pass; Korean Society of Thoracic Radiology guideline 2020, full text read raw from Europe PMC) · The figures are now attributed to both SIR documents as the Korean guideline cites them, and the gaps say the 2019 edition's own figures are not confirmed here.
- 2026-11-17 · CT-guided and ultrasound-guided needle biopsy of the lung: lung function test in severe COPD: The step told the team to measure lung function under a grade A recommendation. The guideline only suggests the test (grade B, evidence level III); grade A applies to the multidisciplinary discussion. (Second accuracy pass; Korean Society of Thoracic Radiology guideline 2020, full text read raw from Europe PMC) · The step now gives the discussion as the recommendation and the lung function test as a suggestion with its own grade.
- 2026-11-17 · CT-guided and ultrasound-guided needle biopsy of the lung: ultrasound guidance: The step said to guide the needle with ultrasound when the lesion lies against the chest wall. The guideline says ultrasound can be considered as the primary guidance there. (Second accuracy pass; Korean Society of Thoracic Radiology guideline 2020, full text read raw from Europe PMC) · The step now says to consider ultrasound, and its reason gives what the guideline reports.
- 2026-11-17 · CT-guided and ultrasound-guided needle biopsy of the lung: when to call for help: The patient summary said to call the hospital at once for breathlessness, chest pain or coughing up blood, with no source. Cancer Research UK says a little blood for a few days can happen, to contact the team if it does not go away or there is a lot, and to contact a doctor for shortness of breath or chest pain. (Second accuracy pass; Cancer Research UK lung biopsy page, read raw) · The wording now follows Cancer Research UK, which is added as a source with a step of its own.
- 2026-11-17 · CT-guided and ultrasound-guided needle biopsy of the lung: complication rates for patients: The patient summary gave about one in four and about one in eighteen, fractions worked out by the writer. The meta-analysis prints 25.3% and 5.6%. (Second accuracy pass; Heerink 2017 meta-analysis, full text read raw from Europe PMC) · The printed percentages are now shown.
- 2026-11-17 · CT-guided and ultrasound-guided needle biopsy of the lung: reasons given for steps: Eleven reasons said things the guideline does not, for example that lung cannot be pressed on, that a sudden movement can tear the lung, that most air leaks cause no symptoms at first and that serious complications are rare. (Second accuracy pass; Korean Society of Thoracic Radiology guideline 2020, full text read raw from Europe PMC) · Each reason now says what the guideline says.
- 2026-11-17 · CT-guided and ultrasound-guided needle biopsy of the lung: grading scheme and evidence levels: The page said the table defining the grades was not in the text read, and that the guideline grades nothing above evidence level II. The table is in the text, and one recommendation listed on the page (biopsy for molecular profiling) is evidence level I. (Second accuracy pass; Korean Society of Thoracic Radiology guideline 2020, full text read raw from Europe PMC) · The grading scheme is now described from the guideline's table and the statement about evidence levels is limited to the steps.
- 2026-11-17 · CT-guided and ultrasound-guided needle biopsy of the lung: correcting clotting, and haemothorax in the meta-analysis: A step dropped the guideline's scope of a moderate or higher bleeding tendency, and a gap said the meta-analysis studies did not report haemothorax when a few did. (Second accuracy pass; Korean guideline and Heerink 2017, read raw from Europe PMC) · Both now match the sources.
- 2026-11-17 · Image-guided core needle biopsy of the breast and liver: preparation and aftercare: grades in the summary and best-in-class statement: The page gave the "of 1000 people like you" wording for risks a Strong grade with high quality evidence, and put Strong on the statement that an INR of 1.5 to 2.0 is not a contraindication. The guideline grades the first Moderate with weak evidence and the second Moderate with moderate evidence. (Second accuracy pass; BSG, RCR and RCPath liver biopsy guideline (Gut 2020), full text read as Europe PMC XML) · Each recommendation now carries its own printed grade, and the grades list gained the missing rows.
- 2026-11-17 · Image-guided core needle biopsy of the breast and liver: preparation and aftercare: platelet threshold quote: The quote for the platelet threshold printed "10 9 /L" where the guideline has a superscript, so it was not the source's text. (Second accuracy pass; BSG, RCR and RCPath liver biopsy guideline (Gut 2020), full text read as Europe PMC XML) · The quote was replaced with the guideline's verbatim plasma recommendation, which carries the same grade, and the platelet threshold is stated in prose with its unit.
- 2026-11-17 · Image-guided core needle biopsy of the breast and liver: preparation and aftercare: reasons given for the liver steps: Several reasons were not in the guideline: that aspirin adds a little to bleeding risk, that bleeding usually shows in the first few hours, that image guidance avoids blood vessels, and that written information makes people worry less. The guideline says none of these. (Second accuracy pass; BSG, RCR and RCPath liver biopsy guideline (Gut 2020), full text read as Europe PMC XML) · Each reason now says what the guideline says, for example that pulse and blood pressure are checked to detect complications early.
- 2026-11-17 · Image-guided core needle biopsy of the breast and liver: preparation and aftercare: shorter observation variant: The page said some units observe for 1 hour after a liver biopsy. The guideline reports one study suggesting 1 hour is enough and does not say any unit does this. (Second accuracy pass; BSG, RCR and RCPath liver biopsy guideline (Gut 2020), full text read as Europe PMC XML) · The variant now reports the study and the guideline's own recommendation of at least 3 hours.
- 2026-11-17 · Image-guided core needle biopsy of the breast and liver: preparation and aftercare: breast biopsy statements: The page said breast biopsies are done under local anaesthetic in clinic, that B3 lesions are usually managed by vacuum-assisted excision, and that the most suspicious lymph node is biopsied. The screening guidance does not mention anaesthetic, says the team decides between vacuum-assisted excision and diagnostic surgery, and allows fine needle aspiration or core biopsy of a node with core preferred where possible and safe. (Second accuracy pass; NHS Breast Screening Programme assessment guidance (updated 22 January 2026), read as raw HTML from gov.uk, and the NHS biopsy page (reviewed 11 June 2025)) · The statements were reworded to the guidance, and the NHS biopsy page was added as the source for local anaesthetic and for general warning signs after any biopsy.
- 2026-11-17 · Image-guided core needle biopsy of the breast and liver: preparation and aftercare: who the page applies to and what could not be read: The page scoped itself to adults and to breast biopsy in general, described the unread SIR 2019 consensus as if it had been read, and said no newer open lung biopsy guideline existed. The breast source covers screening assessment only, the SIR paper still cannot be read, and open lung biopsy guidelines exist and are used in the lung biopsy record. (Second accuracy pass; NHS Breast Screening Programme assessment guidance (updated 22 January 2026), read as raw HTML from gov.uk, and a failed fetch of the SIR paper) · The population and gaps were reworded, and the page now links the lung biopsy record.
- 2026-11-17 · Bone marrow aspirate and trephine biopsy: how often problems happen: The page said serious problems happen in about 1 in 1,250 tests and that haemorrhage ran at about 0.08% of procedures. The 2003 UK survey prints 16 adverse events of all kinds in 19,259 procedures (0.08%), 11 of them haemorrhage, and prints no 1 in 1,250 figure. (Second accuracy pass; Bain, J Clin Pathol 2005, abstract read raw from Europe PMC) · The summary, short summary and best-in-class statement now give 16 in 19,259 (0.08%) as all adverse events, with haemorrhage as 11 of the 16.
- 2026-11-17 · Bone marrow aspirate and trephine biopsy: how many people have severe pain: The page said 35% of people, or a third, have severe pain under local anaesthetic. The source says pain is reported in 70% of patients, of whom 35% rate it as severe. (Second accuracy pass; methoxyflurane trial, BMJ Supportive and Palliative Care 2014, abstract read raw from Europe PMC) · The summary and the pain relief step now say 35% of those who report pain.
- 2026-11-17 · Bone marrow aspirate and trephine biopsy: platelet transfusion before the test: The page said no source sets platelet thresholds and gave no guidance. NICE recommendation 1.7.8 says not to offer prophylactic platelet transfusions to people having bone marrow aspiration and trephine biopsy, and the British Society for Haematology summary lists transfusion before it as not indicated. (Second accuracy pass; NICE NG24 platelet chapter and the BSH platelet guideline summary, read as raw text) · Added the NICE recommendation as a step scoped to England, with its quotation, and both documents as sources. The gap now says only that no source sets a count below which the test should not be done.
- 2026-11-17 · Bone marrow aspirate and trephine biopsy: the safety surveys: The page said no national figure later than 2003 was found. The same survey series reported 17 adverse events in 13,506 procedures for 2002 and 15 in 20,323 for 2004. (Second accuracy pass; Bain, Clin Lab Haematol 2004 and Haematologica 2006, abstracts read raw from Europe PMC) · Added both surveys as sources with their figures and corrected the gap.
- 2026-11-17 · Bone marrow aspirate and trephine biopsy: the laboratory guideline: The page cited the 2008 ICSH guideline without saying it has a successor. ICSH Guidance on Bone Marrow Examination and Reporting was published on 4 September 2026 as an extension and refinement of the 2008 text. (Second accuracy pass; Europe PMC and Crossref records) · Added the 2026 guidance as a source read as an abstract only and named it in the gaps. No step rests on either.
- 2026-11-17 · Bone marrow aspirate and trephine biopsy: steps written as orders: The steps told staff to offer extra pain relief, offer distraction such as music or conversation, and aspirate rapidly. The sources are a trial background statement, a meta-analysis whose authors say staff should consider these measures and whose abstract does not name music or conversation, and one trial whose authors propose the rapid technique. (Second accuracy pass; abstracts read raw from Europe PMC) · The three steps now say consider, name only virtual reality as a tested distraction, and give the trial's figures.
- 2026-11-17 · Bone marrow aspirate and trephine biopsy: reasons no source gives: Several reasons were not in the sources: that distraction has no side effects, that rapid aspiration makes a repeat test less likely, that most oozing stops with pressure, that sedatives affect judgement for longer than people feel, that local anaesthetic needs no fasting, and that the warning signs point to bleeding under the skin. (Second accuracy pass; Cancer Research UK page and trial abstracts read as raw text) · Each reason was replaced with what the cited source says.
- 2026-11-17 · Bone marrow aspirate and trephine biopsy: biopsy site and children: The page said the sample is usually taken from the back of the hip (posterior iliac crest) and that children usually have sedation or general anaesthesia. Cancer Research UK says usually the hip and that children and teenagers often have sedation; no source read mentions general anaesthesia. (Second accuracy pass; Cancer Research UK page and REVEH full text read as raw text) · Reworded to usually the hip, noting that the REVEH trial used the posterior iliac crest, and to children and teenagers often having sedation.
- 2026-11-17 · Bone marrow aspirate and trephine biopsy: smaller wording faults: The summary gave the aspiration trial's pain difference as 1 point on a 10-point scale, the ache as lasting a few days, gas and air as used alongside sedation, and aftercare as consensus practice; a variant described how a powered driver works. The sources say 1 point on a visual analogue scale, a couple of days, instead of sedation, and nothing on consensus or mechanism. (Second accuracy pass; abstracts and Cancer Research UK page read as raw text) · Reworded each to the source.
- 2026-11-17 · Diagnostic flexible bronchoscopy and EBUS: preparation, sedation and aftercare: symptoms to report afterwards: The plain summary told readers to get help for breathlessness or coughing up a lot of blood with no source behind it. The BTS guideline says written advice should cover likely symptoms but does not list them in its text. (Second accuracy pass; BTS 2013 read as raw text, and the Cancer Research UK bronchoscopy page (reviewed 30 May 2025) fetched raw) · Reworded to Cancer Research UK's advice (contact a doctor or the hospital advice line straight away for breathlessness or chest pain at home, and ring if blood in the spit does not go away or gets worse), added that page as a source and a step quoting it, and named the gap.
- 2026-11-17 · Diagnostic flexible bronchoscopy and EBUS: preparation, sedation and aftercare: grades beside the steps: Four steps showed a single grade for several recommendations of different strength: Grade B beside the midazolam syringe limit and the 1 mg/mL rule (both Grade D), Grade A beside using the lowest lidocaine dose (Grade D), Grade C beside recording vital signs (Grade D), and Grade B beside the anticholinergic advice (Grade A). The best-in-class statement also gave low-strength midazolam Grade B. (Second accuracy pass; BTS 2013 guideline, PDF from the BTS document library read as raw text) · Each step now lists the grade of every recommendation it covers, and the statement gives the 1 mg/mL rule Grade D.
- 2026-11-17 · Diagnostic flexible bronchoscopy and EBUS: preparation, sedation and aftercare: fasting and driving in the short summary: The short summary said people usually stop eating 4 hours before and must not drive for 24 hours. The guideline says 4 hours without food and clear fluids to 2 hours is sufficient, on Grade D and with insufficient evidence for a specific time, and that people who had sedation should be advised not to drive. A gap note also said many units ask for 6 hours, which no source read states. (Second accuracy pass; BTS 2013 guideline, PDF from the BTS document library read as raw text) · Reworded to what the guideline says, scoped the driving advice to sedation, told readers to follow their own unit's instructions, and removed the 6 hour figure.
- 2026-11-17 · Diagnostic flexible bronchoscopy and EBUS: preparation, sedation and aftercare: reasons given for steps: Seven reasons said more than the guideline: that aspirin alone adds little bleeding, that a long fast is harmful, that routine blood tests do not change what happens, that sedation slows breathing, what a safety checklist catches, which symptoms mean a collapsed lung, and that sedatives impair judgement for longer than people feel. The guideline states none of these. (Second accuracy pass; BTS 2013 guideline, PDF from the BTS document library read as raw text) · Each reason now says what the guideline's evidence statements say, for example that routine clotting tests do not predict bleeding and that clopidogrel caused bleeding at lung biopsy.
- 2026-11-17 · Diagnostic flexible bronchoscopy and EBUS: preparation, sedation and aftercare: staging CT and anticholinergics: The summary gave Grade D to a staging chest CT and said anticholinergic premedication is not used. The guideline's Grade D recommendation is for a chest CT (the 2014 quality standards ask for a staging CT), and it says anticholinergics should not routinely be used. (Second accuracy pass; BTS 2013 guideline, PDF from the BTS document library read as raw text and the BTS 2014 quality standards PDF) · Reworded both, with the staging CT attributed to the quality standards.
- 2026-11-17 · Diagnostic flexible bronchoscopy and EBUS: preparation, sedation and aftercare: best in class statement: The statement said propofol and general anaesthesia are increasingly used for EBUS and robotic bronchoscopy and that 2026 German and United States documents address this. No source read says so, and neither document could be read beyond a title or an abstract of its scope. (Second accuracy pass; Crossref and Europe PMC records for both papers, both behind subscriptions) · Removed the trend claim and said only that the two newer documents exist and could not be read.
- 2026-11-17 · Sex, contraception and barrier protection during and after chemotherapy: asking about sex when blood counts are low: The page told people to ask about sex especially when blood counts are low, because very low platelets or white cells change the risks of bleeding and infection, and cited the NCI booklet. The booklet says only to ask the doctor or nurse whether it is okay to have sex, and that most women and most men can. No source read says anything about blood counts and sex. (Second accuracy pass; NCI Chemotherapy and You, PDF read as raw text) · The blood count wording was removed from the step, the reason and the population, and the gap is named on the page.
- 2026-11-17 · Sex, contraception and barrier protection during and after chemotherapy: how long to prevent pregnancy afterwards: The summary said NCI advises preventing pregnancy for a year after chemotherapy, as if for everyone. The booklet says it in its section for women; for men it speaks only of the time on treatment. The page also said all sources agree on contraception for a period after treatment, which the NHS page and the Society of Family Planning abstract do not say. (Second accuracy pass; NCI booklet, NHS page and the Society of Family Planning abstract read as raw text) · The year is now scoped to women, the agreement is limited to contraception during treatment, and FDA guidance for US labels (2019) was added with its periods by sex and type of drug.
- 2026-11-17 · Sex, contraception and barrier protection during and after chemotherapy: sources and their dates: The NCI booklet was dated 2023. Its back cover prints January 2024 (NIH Publication No. 24-7157); 2023 was the date the file was made. The summary also said Macmillan, the NHS and NCI all mention a pregnancy test before treatment and that pregnancy remains possible when fertility is reduced; Macmillan does not mention a test, and only Macmillan makes the second point. (Second accuracy pass; all three patient sources read as raw text) · The year was corrected to 2024 and each statement is now attributed to the sources that make it.
- 2026-11-17 · Sex, contraception and barrier protection during and after chemotherapy: reasons given for the choice of method: Three reasons had no source: that barrier methods are safe for almost everyone, that oestrogen-containing methods work against oestrogen-blocking treatment, and that fertility choices must be made before treatment. Macmillan gives no reason for preferring barrier methods, and ASCO says fertility preservation should be discussed before treatment. (Second accuracy pass; Macmillan and Cancer Research UK pages read as raw text, Society of Family Planning and ASCO abstracts from Europe PMC) · The reasons now say what the sources say, with Cancer Research UK added for why hormonal methods might not be safe with a hormone dependent cancer or a risk of blood clots.
- 2026-11-17 · Sex, contraception and barrier protection during and after chemotherapy: oestrogen and oestrogen-blocking therapy: The step tied the advice to avoid oestrogen-containing contraceptives with oestrogen-blocking therapy to breast cancer. In the abstract it is a separate Best Practice sentence, placed among the ovarian cancer recommendations and not limited to one cancer. The page also said the series advises against an IUD with active cervical cancer, where the abstract suggests against it (GRADE 2C), and called melanoma, meningioma and follicular lymphoma hormone-sensitive, which the abstracts do not. (Second accuracy pass; Society of Family Planning 2025 parts 2 and 3, abstracts from Europe PMC) · The step and grade now say the statement is separate and general, the variant uses the abstract's strength of wording, and a gap says the full text was not read.
- 2026-11-17 · Sex, contraception and barrier protection during and after chemotherapy: how long to use a barrier: A gap said the amount of drug reaching semen or vaginal fluid is poorly measured and that a few days of condom use is expert advice; no source read says either. Macmillan says small amounts can get into body fluids and a doctor may advise condoms for a few days after certain drugs; Cancer Research UK says it is not known for sure and gives about a week after treatment. (Second accuracy pass; Macmillan and Cancer Research UK pages read as raw text) · The gap now states the two positions, and Cancer Research UK was added as a source and a variant.
- 2026-11-17 · Fertility discussion and preservation before cancer treatment: letrozole during ovarian stimulation: The step told teams to add letrozole in oestrogen-sensitive cancers and gave a reason the source does not state. ESHRE made this a good practice point only, "probably recommended" in 2020, and its 2025 ovarian stimulation update says an anti-oestrogen such as letrozole or tamoxifen "can be considered". (Second accuracy pass; ESHRE 2020 and ESHRE ovarian stimulation update 2025, full text from Europe PMC) · Step reworded to "consider", re-quoted from the 2025 update, with both wordings shown.
- 2026-11-17 · Fertility discussion and preservation before cancer treatment: when to start ovarian stimulation: Starting at any point in the cycle was filed under the strong grade of the GnRH antagonist recommendation. ESHRE 2020 graded random start separately as a weak recommendation on low quality evidence, and the 2025 update, which had not been read, now says stimulation should be started irrespective of the cycle phase (strong, very low quality evidence). (Second accuracy pass; ESHRE 2020 and ESHRE ovarian stimulation update 2025, full text from Europe PMC) · Made a separate step on the 2025 text with its own grade; the 2020 position kept as a variant.
- 2026-11-17 · Fertility discussion and preservation before cancer treatment: best supported version: The page said NICE, ESHRE and ASCO agree that GnRH agonists do not replace freezing, and that NICE and ASCO agree on ovarian tissue freezing. The NICE chapter makes no recommendation on GnRH agonists, and NICE says to consider ovarian tissue cryopreservation where ASCO lists it among established methods that should be offered. (Second accuracy pass; NICE NG257 chapter read as HTML, ASCO 2025 abstract from Europe PMC) · Agreement limited to ESHRE and ASCO, NICE's silence stated, and the difference on ovarian tissue written out. The summary now says NICE "considers" ovarian tissue cryopreservation.
- 2026-11-17 · Fertility discussion and preservation before cancer treatment: psychological support: The step gave psychological support the grade of the decision aid recommendation (strong, low quality evidence). ESHRE grades psychological support and counselling strong on very low quality evidence and notes its benefit has not been studied. (Second accuracy pass; ESHRE 2020, full text from Europe PMC) · Both grades stated and the psychological support recommendation added to the grades table with its quote.
- 2026-11-17 · Fertility discussion and preservation before cancer treatment: reasons given for the steps: Several reasons were not in any source: that the window closes once treatment starts (ASCO says preservation may be offered after treatment), that sperm banking is quick, that chemotherapy damages sperm made in the following weeks, that a GnRH agonist rests the ovaries, and that storage depends on renewed consent. The sperm step was also an order to collect before treatment where ASCO says to advise of the risk. (Second accuracy pass; NICE NG257, ESHRE 2020 and the ASCO 2025 abstract, read raw) · Each reason reworded to what the cited guideline says; the sperm step now reads "advise".
- 2026-11-17 · Fertility discussion and preservation before cancer treatment: who it applies to, and the medicines named: The page listed chemotherapy, pelvic or total body radiotherapy and surgery as the treatments concerned, said NICE NG257 applies to England, called the GnRH agonist an injection, and listed goserelin and leuprolide as the protocol's drugs. None of the sources read says any of these. (Second accuracy pass; NICE NG257, ESHRE 2020 and the ASCO 2025 abstract, read raw) · Examples and country removed, the plain-language summary reworded, the two medicines moved to related records, and a gap added that no source read names a GnRH agonist or dose.
- 2026-11-17 · Fertility preservation for children and young people before cancer treatment: referral timing: The page told teams to refer families straight away and called the referral immediate, under a strong grade. The graded recommendation carries no timing; as soon as possible comes from a separate good practice statement that has no grade. (Second accuracy pass; COG endorsement document of ASCO 2025, PDF read as raw text) · The referral step, summary and best practice statement now give the graded recommendation and attribute the timing to the good practice statement.
- 2026-11-17 · Fertility preservation for children and young people before cancer treatment: ovarian tissue freezing: The page said to offer ovarian tissue freezing. The guideline says it may be offered as an established method, and adds a theoretical risk of returning cancer cells at transplant. (Second accuracy pass; COG endorsement document of ASCO 2025, PDF read as raw text) · Reworded to may be offered and the caveat added.
- 2026-11-17 · Fertility preservation for children and young people before cancer treatment: ovarian transposition: The page did not say the recommendation is for patients of reproductive age, or that the guideline calls the procedure unsuitable where the risk of spread to the ovary is moderate or high or where chemotherapy that harms the ovaries is given at the same time. (Second accuracy pass; COG endorsement document of ASCO 2025, PDF read as raw text) · Scope and the two exclusions added to the step, summary and grades.
- 2026-11-17 · Fertility preservation for children and young people before cancer treatment: GnRH agonists and follow-up wording: The summary said GnRH agonists must not replace established methods and that fertility is revisited at follow-up. The guideline says should not on a conditional recommendation, and that further discussions may be offered yearly. (Second accuracy pass; COG endorsement document of ASCO 2025, PDF read as raw text) · Both reworded to the guideline's strength.
- 2026-11-17 · Fertility preservation for children and young people before cancer treatment: reasons given for steps: Seven reasons said more than the guideline: that the team may be needed within days, that frozen sperm is the most reliable way, that risk is unknown for newer treatments, that egg collection never depends on the cycle, and others. (Second accuracy pass; COG endorsement document of ASCO 2025, PDF read as raw text) · Each reason reworded to what the guideline states; within days and most reliable removed.
- 2026-11-17 · Fertility preservation for children and young people before cancer treatment: PanCareLIFE and IGHG recommendations: The page said these paediatric recommendations could not be read and so did not say where they differ from ASCO. The IGHG posts them as open documents; they tie each method to the planned treatment, with thresholds by cyclophosphamide equivalent dose. (Second accuracy pass; IGHG recommendation documents dated 2020, open PDFs read as raw text) · Added as a source and as a variant with its thresholds and strengths; the gap rewritten.
- 2026-11-17 · Fertility protection before pelvic radiotherapy: the ovarian transposition step: The step told teams to offer ovarian transposition to premenopausal women. ESHRE says women may be offered it where pelvic radiotherapy without chemotherapy is planned, a weak recommendation on low-quality evidence, which the guideline says requires discussion and shared decision-making. (Second accuracy pass; the ESHRE 2020 guideline, full text from Europe PMC) · The step now says to discuss transposition as an option that may be offered, and the reason given says the recommendation is weak.
- 2026-11-17 · Fertility protection before pelvic radiotherapy: transposition with chemotherapy: The page said the guidelines differed only between ESHRE and ASCO. The 2026 review it cites says transposition should be offered to premenopausal women aged 35 to 40 or younger having pelvic radiotherapy with or without chemotherapy, the opposite of ESHRE's limit, and the page had not said so. (Second accuracy pass; Lim et al. 2026, full text from Europe PMC) · Both positions are now stated with who holds each and that the review is ungraded.
- 2026-11-17 · Fertility protection before pelvic radiotherapy: effect on sperm production: The patient summary said pelvic radiotherapy can damage sperm production. No source read for this page states the effect of pelvic radiotherapy on the testes. (Second accuracy pass; all six sources, read as raw text or abstract) · The sentence was removed and the gap is named.
- 2026-11-17 · Fertility protection before pelvic radiotherapy: who it applies to: The page listed bladder cancer among the cancers concerned, and said ovarian doses were far below those for rectal or anal cancer. No source names bladder cancer or gives a rectal or anal dose. (Second accuracy pass; Lim et al. 2026 and the 2022 meta-analysis, full text from Europe PMC) · The cancers are listed as the sources name them, and the dose comparison now gives the review's 45 to 50 Gy for postoperative gynaecological radiotherapy.
- 2026-11-17 · Fertility protection before pelvic radiotherapy: ovarian tissue freezing: The step added lack of time for stimulation as a reason to consider ovarian tissue freezing. NICE gives one example, girls before puberty. (Second accuracy pass; NICE NG257 section 1.53, full text) · The added example was removed.
- 2026-11-17 · Fertility protection before pelvic radiotherapy: reasons given for steps: Eight steps gave reasons no source states, among them that moving ovaries that may hold cancer can spread it and that tissue can be stored without hormone treatment. (Second accuracy pass; NICE NG257, ESHRE 2020 and Lim et al. 2026, read in full) · Each reason now says only what its source says.
- 2026-11-17 · Fertility protection before pelvic radiotherapy: variants: The page called placing the ovaries behind the uterus an older technique backed by historical case series, said many women have chemotherapy with their radiotherapy, and said ASCO scoped transposition to premenopausal women having pelvic radiotherapy. None of these is in the sources. (Second accuracy pass; Lim et al. 2026 full text and the ASCO 2025 abstract) · The wording now follows the review and the abstract.
- 2026-11-17 · Pregnancy and breastfeeding checks before radiotherapy: legal scope: The page said IR(ME)R 2017 is the law in the UK. It extends to England, Wales and Scotland; Northern Ireland has its own 2018 regulations. (Second accuracy pass; legislation.gov.uk regulation 11 and Schedule 2 and the RCR foreword, read as raw text) · Summary, population and best-in-class statement now name the nations, and the Northern Ireland regulations are listed as not read.
- 2026-11-17 · Pregnancy and breastfeeding checks before radiotherapy: who is asked and when: Two steps ordered staff to ask every person in the age range, at least before planning and first treatment. The guidance says the employer's procedure defines when the question is relevant, and that asking at those two points may be considered sufficient. (Second accuracy pass; RCR IR(ME)R guidance second edition (October 2026), PDF read as raw text) · Steps reworded to the guidance's own hedges.
- 2026-11-17 · Pregnancy and breastfeeding checks before radiotherapy: discovered pregnancy: The step said to stop and tell the practitioner, who reconsiders with the medical physics expert, and gave fetal shielding as an outcome. The guidance says the operator should alert the practitioner, and that the expert should be involved during a course or when treatment proceeds; it does not say stop or mention shielding. (Second accuracy pass; RCR IR(ME)R guidance second edition (October 2026), PDF read as raw text) · Step and reason reworded.
- 2026-11-17 · Pregnancy and breastfeeding checks before radiotherapy: treating a pregnant patient: The step asked for a fetal dose risk assessment by the medical physics expert and an obstetrician where needed, for any radiotherapy. The guidance, speaking of external beam radiotherapy, says the expert must be involved in a prior risk assessment and that an approach with an obstetrician and neonatologist may need to be considered. (Second accuracy pass; RCR IR(ME)R guidance second edition (October 2026), PDF read as raw text) · Step, summary and grade claim scoped and reworded.
- 2026-11-17 · Pregnancy and breastfeeding checks before radiotherapy: irregular cycles and testing: The page said not to rely on the ten-day rule, never to rely on menstrual history, and described departments testing everyone in the age range. The guidance says the rule may be ineffective, allows exceptions for some localised treatments, and mentions routine testing only for irregular cycles after a risk assessment. (Second accuracy pass; RCR IR(ME)R guidance second edition (October 2026), PDF read as raw text) · Steps, best-in-class statement and variant reworded; exceptions named.
- 2026-11-17 · Pregnancy and breastfeeding checks before radiotherapy: reasons given: Five reasons were not in any source, for example that a test settles the question and that a clear rule means nobody is missed. (Second accuracy pass; RCR IR(ME)R guidance second edition (October 2026), PDF read as raw text) · Reasons replaced with what the guidance says.
- 2026-11-17 · Pregnancy and breastfeeding checks before radiotherapy: breastfeeding question: The step ordered staff to ask everyone who has recently given birth. The guidance says this may be considered relevant, and adds that delaying treatment until lactation has ceased should be considered. (Second accuracy pass; RCR IR(ME)R guidance second edition (October 2026), PDF read as raw text) · Step reworded and the delay added.
- 2026-11-17 · Pregnancy and breastfeeding checks before radiotherapy: advice to patients: The plain summary told patients to tell the team straight away with no source. The NHS page on radiotherapy says to contact your care team immediately if you think you may be pregnant. (Second accuracy pass; nhs.uk What happens during radiotherapy, read as raw text) · NHS page added as a source and the sentence worded to match.
- 2026-11-17 · Pregnancy and breastfeeding checks before radiotherapy: aliases and grading note: The page listed the 28-day rule as another name, which no source read mentions, said most departments use ages 12 to 55 where the guidance says many, and said the guidance defines must, should and may, which it does not. (Second accuracy pass; RCR IR(ME)R guidance second edition (October 2026), PDF read as raw text) · Alias removed, wording corrected, and a gap added that neither rule is defined.
- 2026-11-17 · Advance care planning for adults with cancer: what the conversation covers: The step on thinking ahead said the discussion includes who the person wants involved in decisions and their preferences for treatment, support and where they live. NICE NG108 says that of the written document, which may include those things (recommendation 1.1.41), not of the discussion (1.1.39). (Second accuracy pass; NICE NG108 recommendations page, read raw) · The list moved to the step on the written record, worded as what the document may include.
- 2026-11-17 · Advance care planning for adults with cancer: changing the plan: The plain summary said a plan can be changed at any time, and the patient summary said it is revisited when health or wishes change and written in the person's own words. NICE NG108 says a plan is reviewed at reviews of treatment or support and amended while the person has capacity, and that notes are agreed with the person. (Second accuracy pass; NICE NG108 recommendations page, read raw) · Both summaries reworded to what NICE says.
- 2026-11-17 · Advance care planning for adults with cancer: the facilitator model tested in ACTION: The page said a facilitator, often not the person's doctor, holds one or more conversations with the patient and a family member. The trial report says 39 facilitators were certified, predominantly nurses, and that the intervention involved 1 or 2 conversations, with or without a personal representative present. (Second accuracy pass; ACTION trial full text from Europe PMC (PMC7665676), read raw) · Description corrected, and the paper's figure that 67 per cent of patients considered the conversations helpful added beside the 16 per cent who found them distressing.
- 2026-11-17 · Advance care planning for adults with cancer: reasons given for the steps: Nine steps gave reasons no source read states, for example that unidentified people have decisions made in an emergency, that most people need help picturing choices, and that a paper copy reaches places computer systems do not. One step added "before the discussion" and another turned NICE's description of binding and non-binding documents into an instruction to explain it. (Second accuracy pass; NICE NG108 and NG142 recommendations and the NG142 rationale page, read raw) · Each reason replaced with NICE's own wording or its committee's rationale; the two actions reworded to the recommendations.
- 2026-11-17 · Advance care planning for adults with cancer: when a person lacks capacity: The page said an attorney with the right authority is consulted. Section 9 of the Mental Capacity Act 2005 gives an attorney authority to make decisions, and section 4(7) separately asks a decision-maker to take account of an attorney's views. (Second accuracy pass; Mental Capacity Act 2005 sections 4, 9 and 26 on legislation.gov.uk, read raw) · Step reworded and section 9 cited.
- 2026-11-17 · Advance care planning for adults with cancer: what the guidance rests on: The page said the guidance rests on the person's right to be involved and on consensus, which no source read says, and described the ESMO grading scheme from a supplementary table that was not read. It also called the NICE guidelines written for England and said Scotland and Northern Ireland have different capacity law, neither of which the pages read state. (Second accuracy pass; NICE NG142 rationale page, ESMO 2021 full text and Mental Capacity Act 2005 section 68, read raw) · Replaced with the NG142 committee's statement that the evidence was unclear, the ESMO guideline's own sentence on ungraded statements, and the Act's extent to England and Wales only.
- 2026-11-17 · Follow-up after treatment for early breast cancer: best-in-class statement: The page said NICE NG101 advises against routine screening for distant recurrence. The follow-up section of NG101 does not address blood tests or imaging for distant recurrence; it says only not to use ultrasound or MRI routinely for surveillance. The recommendation against those tests is ASCO's. (Second accuracy pass; NICE NG101 recommendations page and the ASCO 2026 pocket guide, both read as raw text) · The statement now attributes the recommendation to ASCO 2026 with its printed grade and says NICE does not address it.
- 2026-11-17 · Follow-up after treatment for early breast cancer: ASCO 2026 mammography interval: The page said mammography every 1 to 2 years "can be considered" for the lower-risk group. ASCO's Recommendation 1.1.2 says it "is recommended" for people aged 50 or older with the listed low-risk features who had breast-conserving surgery and are recurrence-free at 3 years, and that staying on annual mammography is an option. (Second accuracy pass; ASCO 2026 pocket guide on Guideline Central, read in full as raw text) · Reworded to the source's strength and conditions, and added as a step with the verbatim sentence and its printed grade.
- 2026-11-17 · Follow-up after treatment for early breast cancer: ASCO 2026 steps: Two steps resting on the ASCO 2026 pocket guide had no quotation or grade, and one gave an instruction (agree which guideline defines the plan) that the guide does not contain. The step against routine tests for distant recurrence quoted only the superseded 2013 abstract and gave a reason (false alarms and extra procedures) that source does not state. (Second accuracy pass; ASCO 2026 pocket guide and the ASCO 2013 abstract from Europe PMC, read as raw text) · The steps now carry ASCO 2026's own sentences and printed grades for the intensity groups, visit intervals, mammography, distant imaging and blood tests; the unsourced reason was removed.
- 2026-11-17 · Follow-up after treatment for early breast cancer: care plan and primary care steps: The care plan step listed "review dates" and a comparison said NICE relies on "open access" to the specialist team. NICE lists dates for review of any adjuvant therapy and contact details for immediate referral to specialist care. The primary care step said to support the person to keep taking endocrine therapy where the guideline says to monitor adherence. (Second accuracy pass; NICE NG101 recommendation 1.15.4 and the ACS/ASCO 2016 abstract, read as raw text) · Reworded to what each source says.
- 2026-11-17 · Follow-up after treatment for early breast cancer: symptoms to report: A note listed symptoms to report between visits (a new lump, bone, breathing or neurological change) with no source, and no source read gives such a list. (Second accuracy pass; NICE NG101 and the Cancer Research UK follow-up page, read as raw text) · The list was removed. The page now gives NICE's requirement that the plan lists signs and symptoms and Cancer Research UK's advice to report any new symptom without waiting for the next visit, and the gap is named.
- 2026-11-17 · Follow-up after treatment for early breast cancer: notes on markers, harms and NCCN: Three statements had no readable source: that normal tumour markers cannot exclude recurrence, that extra testing carries radiation exposure, and an NCCN follow-up schedule taken from a SEER teaching presentation that was not linked and could not be found. (Second accuracy pass; every cited abstract and page re-read as raw text, and one search for the presentation) · The two unsourced statements and the NCCN schedule were removed; harms are now given as ASCO 2026 lists them and the NCCN gap is named.
- 2026-11-17 · Breast radiotherapy positioning, immobilisation and daily set-up verification: breast immobilisation step: The step said not to adopt a breast immobilisation device routinely. The guideline does not prohibit it: it says there is insufficient evidence to support the widespread adoption of any specific type, and that the effect on skin dose must be evaluated locally before use and monitored afterwards. (Second accuracy pass; ESTRO-ACROP positioning guideline (Mast et al. 2023), Europe PMC full text read raw) · Reworded the step, its reason, the summary and the best-in-class statement to the guideline's wording.
- 2026-11-17 · Breast radiotherapy positioning, immobilisation and daily set-up verification: arm position step: The step said to use one arm up if the patient cannot raise both, and gave as the reason that after surgery some people cannot lift both arms for long. The guideline says one arm up may be considered, and gives comfort reported in cohort studies as the reason. (Second accuracy pass; ESTRO-ACROP positioning guideline (Mast et al. 2023), Europe PMC full text read raw) · Step now says to consider one arm up; the reason is the guideline's own.
- 2026-11-17 · Breast radiotherapy positioning, immobilisation and daily set-up verification: reasons given for five steps: Reasons were given that the sources do not state: that lying on the back is the simplest position to repeat each day, that tattoos are a lasting reminder of cancer, and that a deep breath moves the heart away from the treated area. (Second accuracy pass; ESTRO-ACROP positioning guideline (Europe PMC full text) and NICE NG101 recommendations page, both read raw) · Each reason now says only what its source says.
- 2026-11-17 · Breast radiotherapy positioning, immobilisation and daily set-up verification: summary for patients: The summary for patients said you lie on a sloping board. The guideline accepts both flat and inclined positions. (Second accuracy pass; ESTRO-ACROP positioning guideline (Mast et al. 2023), Europe PMC full text read raw) · Now says a flat or sloping board.
- 2026-11-17 · Breast radiotherapy positioning, immobilisation and daily set-up verification: temporary skin marks: The page called SuPPORT 4 All a qualitative study and said temporary marks are renewed during the course. The guideline says neither; it says only that the study reported permanent tattoos may affect wellbeing. (Second accuracy pass; ESTRO-ACROP positioning guideline (Mast et al. 2023), Europe PMC full text read raw) · Both descriptions removed.
- 2026-11-17 · Breast radiotherapy positioning, immobilisation and daily set-up verification: summary and best in class: The page called the guideline the only international guideline on breast set-up, said its authors call most studies small and old, and scoped it to adults. The guideline makes no such claim, says only that prone against supine comparisons are mostly more than 10 years old, and sets no age scope. (Second accuracy pass; ESTRO-ACROP positioning guideline (Mast et al. 2023), Europe PMC full text read raw) · Reworded to the only guideline found, to what the authors say about study age, and to people having this treatment.
- 2026-11-17 · Breast screening after chest radiotherapy at a young age: how common breast cancer is: The summary gave the 13 to 20 per cent and 35 per cent figures for survivors treated with 10 Gy or more. The guideline gives them for chest radiotherapy without a dose threshold, the first for survivors of childhood cancer and the second for survivors of Hodgkin lymphoma only. (Second accuracy pass; IGHG 2020, full text read raw from PubMed Central) · The figures now carry the groups the guideline gives them for.
- 2026-11-17 · Breast screening after chest radiotherapy at a young age: patient summary: The patient summary merged two rules into one, saying MRI and mammograms should start at 25 or 30. IGHG recommends both tests from 25 or 8 years after radiotherapy. The NHS programme in England starts at 25 or 30 by age at radiotherapy and uses MRI alone before 40. (Second accuracy pass; IGHG 2020 full text and the NHS very high risk pages on GOV.UK, read raw) · The summary now gives the international recommendation and the English programme separately.
- 2026-11-17 · Breast screening after chest radiotherapy at a young age: total body irradiation: The page listed total body irradiation among the fields that lead to screening without qualification. IGHG could make no recommendation below a prescribed dose of 10 Gy, and NHS England says it will review whether total body irradiation qualifies. (Second accuracy pass; IGHG 2020 full text and the NHS guidance page, read raw) · Both positions are now stated in the population, summary, variant and gaps.
- 2026-11-17 · Breast screening after chest radiotherapy at a young age: reasons given in the steps: Five reasons said more than the sources: that survivors may not know unless told, that surveillance finds cancers earlier, that the two tests are better together in young women with dense breasts, that the risk after upper abdominal radiotherapy is less clear, and that women would otherwise be lost to follow-up. IGHG says it is still unclear whether surveillance improves outcomes, and that the balance of benefits and harms is what is less clear. (Second accuracy pass; IGHG 2020 full text and the NHS guidance page, read raw) · Each reason now says only what its source says; "tell them about their raised risk" and "before starting" were removed from two actions.
- 2026-11-17 · Breast screening after chest radiotherapy at a young age: anthracycline step: The quotation printed the dose as "mg/m 2", which is not how the abstract prints it, and the evidence line said "no association" below 100 mg/m² where the abstract says "no significant association". The page also said the NHS programme screens girls irradiated below 10 "unlike IGHG", which the IGHG text read does not state. (Second accuracy pass; IGHG 2026 abstract from Europe PMC, IGHG 2020 full text) · The quotation was replaced with a verbatim sentence, the evidence line reworded, and the comparison removed and named as a gap.
- 2026-11-17 · Breast screening after chest radiotherapy at a young age: sources: The two NHS pages were cited under titles they no longer carry. (Second accuracy pass; GOV.UK pages read raw, both updated 11 November 2026) · The labels now give the current titles: The Surveillance of women at very high risk of developing breast cancer, and Eligibility criteria and screening protocols for women at very high risk of breast cancer.
- 2026-11-17 · Deep inspiration breath-hold for left-sided breast radiotherapy: summary, breath-hold duration: The summary said a breath-hold of about 20 to 30 seconds is achievable for more than 90% of breast patients. The guideline makes two separate statements: breath-hold is often restricted to 20 to 30 seconds, a pragmatic compromise achievable by many patients, and reports suggest more than 90% of breast cancer patients can achieve deep inspiration breath-hold. (Second accuracy pass; ESTRO-ACROP breath-hold guideline (Aznar et al. 2023), open published PDF read raw) · The two statements are now given separately, and the step quotes the sentence that prints 20 to 30 seconds.
- 2026-11-17 · Deep inspiration breath-hold for left-sided breast radiotherapy: heart dose target step: The step said to check each plan against a mean heart dose target such as under 2 Gy. The RCR statement is a target for a department: more than 90% of left-sided patients not having internal mammary chain treatment should have a mean heart dose under 2 Gy, and centres agreed a target would help departments to implement breath-hold. (Second accuracy pass; RCR UK consensus statements (2016), PDF read raw) · Step and reason reworded as a department target; a gap notes that no source read gives a limit for an individual plan.
- 2026-11-17 · Deep inspiration breath-hold for left-sided breast radiotherapy: grades of the RCR statements: The statement that every centre should have breath-hold for internal mammary chain radiotherapy was graded only as a consensus statement; the RCR prints Unanimous support. The sentence on lower-half left breast tumours was labelled a consensus statement; it is background text that was not voted on. (Second accuracy pass; RCR UK consensus statements (2016), PDF read raw) · Both grades corrected, and the voted statement on a target mean heart dose (Strongly supported) added.
- 2026-11-17 · Deep inspiration breath-hold for left-sided breast radiotherapy: steps written as orders: Three steps were firmer than the guideline: give take-home material (the guideline says it may be useful), set up with free-breathing reference marks (it says they can be applied), and re-image and re-plan when breathing changes (it lists the ability to do so as part of an ideal set-up). (Second accuracy pass; ESTRO-ACROP breath-hold guideline (Aznar et al. 2023), open published PDF read raw) · Each step now carries the guideline's own strength.
- 2026-11-17 · Deep inspiration breath-hold for left-sided breast radiotherapy: reasons given for eight steps: Several reasons were not in the cited passage, among them that heart damage rises with every unit of dose (cited to NICE, which says only that breath-hold lowers mean heart dose), that no one should feel they have failed, and that daily checks stop dose moving onto the heart. (Second accuracy pass; NICE NG101 recommendations page and ESTRO-ACROP breath-hold guideline, both read raw) · Each reason now says only what its source says.
- 2026-11-17 · Deep inspiration breath-hold for left-sided breast radiotherapy: who it covers and the patient summaries: The page scoped the recommendation to the left breast or chest wall and said the RCR covers internal mammary chain treatment on either side; NICE says people with left-sided breast cancer and the RCR names no side. The short summary said about half a minute and that the team watches your breathing on a screen. (Second accuracy pass; NICE NG101 and RCR UK consensus statements (2016), both read raw) · Scope reworded to the sources; the summary now says often 20 to 30 seconds and that the breath-hold is checked with images.
- 2026-11-17 · Deep inspiration breath-hold for left-sided breast radiotherapy: variants: Voluntary breath-hold was called equivalent to and as reproducible as the spirometry device where the trial says comparable; device descriptions (skin marks or a laser, a mouthpiece, nose clip and valve, goggles) were not in any source read. (Second accuracy pass; HeartSpare abstract (Europe PMC), BRAVEHeart full text (Europe PMC) and the ESTRO-ACROP guideline main text) · Wording set to comparable; device descriptions cut back to what the sources read say, with the unread supplement named in the gaps.
- 2026-11-17 · Cancer-related fatigue: screening, treatable causes and management: tai chi or qigong during treatment: The step said tai chi or qigong "may also be recommended". The ASCO-SIO 2024 guideline says clinicians should recommend it, on moderate evidence, as a strong recommendation. (Second accuracy pass; ASCO-SIO 2024 Table 3, author manuscript in PubMed Central read as raw XML) · The step now says to recommend it, and notes that after treatment the evidence was found insufficient.
- 2026-11-17 · Cancer-related fatigue: screening, treatable causes and management: who the evidence comes from: The page said most trial participants had breast cancer and that the evidence was mostly from women with breast cancer. The guideline says many trials were in mixed cancer types and about 36% of the studies were in breast cancer only. (Second accuracy pass; ASCO-SIO 2024 participant characteristics, read as raw XML) · The population line and the gap now give the 36% figure and say the trials did not include people on immunotherapy.
- 2026-11-17 · Cancer-related fatigue: screening, treatable causes and management: treatable causes: The page listed thyroid and other hormone problems among the general causes to look for, limited the step to moderate to severe fatigue and said to treat causes first. The cited sentence lists pain, depression, anxiety, sleep disturbance, nutrition, activity, anaemia and medicines, with no threshold or ordering; thyroid problems appear only as an immunotherapy side effect. (Second accuracy pass; ASCO-SIO 2024 full text, read as raw XML) · The step and the patient summary now list what the sentence lists; thyroid stays in the checkpoint inhibitor step.
- 2026-11-17 · Cancer-related fatigue: screening, treatable causes and management: when to ask about fatigue: The page said to ask at the first visit, at intervals and whenever tiredness is reported. The 2014 ASCO abstract speaks only of evaluation after primary treatment; the APTA 2022 guideline says to screen at each encounter. (Second accuracy pass; ASCO 2014 abstract and APTA 2022 full text from Europe PMC) · The trigger and first step now give each source its own timing.
- 2026-11-17 · Cancer-related fatigue: screening, treatable causes and management: APTA tool grades: The grading scheme described grade P as limited evidence from one study and grade R as not recommended. The guideline defines P as best practice, may be used, and R as research, unable to recommend, and also has a grade C. (Second accuracy pass; APTA 2022 Table 1 footnote, Europe PMC full text XML) · The scheme is rewritten from the footnote.
- 2026-11-17 · Cancer-related fatigue: screening, treatable causes and management: reasons given for steps: Several reasons were not in the sources: that fatigue goes unreported, that a questionnaire catches what a question misses, that fatigue rarely improves while causes go untreated, that fatigue can be the first sign of gland damage, that exercise has the most consistent evidence of any treatment, that stimulant drugs do not help enough to outweigh side effects, and that steroid side effects build up. (Second accuracy pass; ASCO-SIO 2024 and APTA 2022 full texts, read as raw XML) · Each reason is reworded to what the guideline says.
- 2026-11-17 · Cancer-related fatigue: screening, treatable causes and management: CBT delivery and drug grades: The page applied "in person or online" to CBT during treatment, where the guideline says it only for after treatment, and gave one grade (strong, against) for a step that also covers methylphenidate and other phases, which are conditional. (Second accuracy pass; ASCO-SIO 2024 Tables 3 to 5) · The delivery note is scoped to after treatment and the grade line gives each grade.
- 2026-11-17 · Cancer-related fatigue: screening, treatable causes and management: American ginseng: The page said the guideline suggests twice-daily dosing and that a pharmacist should check for interactions. The guideline says there is no conclusive evidence on schedule and twice-daily dosing may be considered; it reports possible drug interactions and advises avoiding methanol extracts in hormone receptor-positive cancers, and does not mention a pharmacist. (Second accuracy pass; ASCO-SIO 2024 Table 3 footnote and section 1.6) · Reworded to the guideline text.
- 2026-11-17 · Cancer-related fatigue: screening, treatable causes and management: children and NCCN: The page said children have a separate guideline with a 2023 update and called the NCCN guideline not reproducible. No source read says either. (Second accuracy pass; All five sources re-fetched on 17 November 2026) · The page now says the sources read cover adults only and that the NCCN guideline was not read.
- 2026-11-17 · Follow-up after curative treatment for colorectal cancer: length of follow-up: The step said to offer follow-up for at least the first 3 years. NICE NG151 recommendation 1.6.1 says for the first 3 years and does not say at least. (Second accuracy pass; NICE NG151 recommendations page, read as raw HTML) · Removed at least, and replaced the reason with the one NICE gives.
- 2026-11-17 · Follow-up after curative treatment for colorectal cancer: grade on the CT step: The grade 2A was printed beside a sentence comparing ASCO with NCCN, which carries no grade. The American Cancer Society prints 2A on its Recommendation 5. (Second accuracy pass; American Cancer Society 2015, full text from PubMed Central through the NCBI API) · The step now quotes Recommendation 5, and the ASCO difference is given in the reason.
- 2026-11-17 · Follow-up after curative treatment for colorectal cancer: where the guidelines agree: The page said all sources agree on CEA and CT and on a colonoscopy within a year, and that colonoscopy is 1 year after surgery in all of them. NICE leaves colonoscopy out of scope, the BSG counts a year from diagnosis, and the German guideline relies on ultrasound with CT marked once. (Second accuracy pass; NICE NG151, BSG and ACPGBI 2020 and the German S3 summary 2025, each read as raw text) · Agreement and differences are now stated source by source.
- 2026-11-17 · Follow-up after curative treatment for colorectal cancer: what the trials support: The page said the Cochrane review supports the shorter, lighter end of the schedules for most people. The review's authors conclude only that there is no overall survival benefit from intensifying follow-up. (Second accuracy pass; Cochrane review 2019, abstract from Europe PMC) · Replaced the inference with the authors' conclusion.
- 2026-11-17 · Follow-up after curative treatment for colorectal cancer: when to stop colonoscopy: The step ordered surveillance to stop at about 75 or a life expectancy under 10 years. The BSG and ACPGBI statement is a weak one (we suggest), in general, written for surveillance after polyp removal. (Second accuracy pass; BSG and ACPGBI 2020, full text from Europe PMC) · Reworded as a general rule, with its strength and scope stated.
- 2026-11-17 · Follow-up after curative treatment for colorectal cancer: first colonoscopy: The step cited the BSG for a colonoscopy within a year of diagnosis or surgery, sooner if the whole bowel was not seen. The BSG says only within a year of diagnosis; the other timings come from ASCO and the German guideline. (Second accuracy pass; BSG and ACPGBI 2020, ASCO 2013 abstract and German S3 summary 2025, read as raw text) · The action now states the BSG timing, and the ASCO and German timings are attributed to them.
- 2026-11-17 · Follow-up after curative treatment for colorectal cancer: reasons given for steps: Several reasons were not in the cited sources: that most returns happen in the first few years, that PET and routine blood tests add cost and false alarms, that bowel problems after rectal surgery are common, and that tests are missed when nobody knows whose job they are. (Second accuracy pass; NICE NG151 rationale, American Cancer Society 2015 and ASCO 2013 abstract, read as raw text) · Each reason was replaced with what the cited source says, or with a statement that the source gives no reason.
- 2026-11-17 · Follow-up after curative treatment for colorectal cancer: the follow-up plan: The step asked for a written plan saying who does each test and when. ASCO says the specialist's plan should have clear directions on appropriate follow-up by a non-specialist. (Second accuracy pass; ASCO 2013 abstract from Europe PMC) · Reworded to what ASCO says.
- 2026-11-17 · Follow-up after curative treatment for colorectal cancer: reporting new symptoms: The page told patients to report new symptoms between check-ups with no source. Cancer Research UK states this advice. (Second accuracy pass; Cancer Research UK follow-up page, read as raw HTML) · Added the Cancer Research UK page as a source and a step quoting it.
- 2026-11-17 · Follow-up after curative treatment for colorectal cancer: proctoscopy and second colonoscopy in the American schedule: The page gave proctoscopy every 6 months for rectal cancer without its scope, and the second colonoscopy as 3 years after the first. The American Cancer Society scopes proctoscopy to people who had a low anterior resection, for 5 years, and gives the second colonoscopy as at 3 years in its text and in year 4 in its table. (Second accuracy pass; American Cancer Society 2015, full text) · Scope added and both wordings given; the undefined 2A label is named in the gaps.
- 2026-11-17 · Consent and education before anticancer treatment: who it applies to: The page said children and young people need assent and the consent of those with parental responsibility, as if that held everywhere. The ASCO-ONS glossary says consent must be obtained from the parents or guardian under the applicable United States state and federal laws, and UK law on consent for under-18s was not read. (Second accuracy pass; ASCO-ONS 2024 standards, read as raw text from the Oncology Nursing Society's open PDF) · The statement is now scoped to ASCO-ONS and the United States, and the page says the rule differs by country and that UK law was not read.
- 2026-11-17 · Consent and education before anticancer treatment: contact details: The patient summary promised a 24-hour number. ASCO-ONS asks for contact information with availability and when and whom to call (standard 2.3.11), and separately for an organisational policy giving 24/7 triage to a licensed practitioner (standard 1.12). No source read says each patient is given a 24-hour number. (Second accuracy pass; ASCO-ONS 2024 standards, read as raw text from the Oncology Nursing Society's open PDF) · The summary now promises contact details, and the 24/7 triage policy is described as the standard states it.
- 2026-11-17 · Consent and education before anticancer treatment: reasons given for the steps: Ten steps gave reasons no cited source states, among them that emergencies need treating within hours, that infertility can only be prevented before treatment starts, that drugs at home can harm carers, children and pets, and that people remember little of one conversation. (Second accuracy pass; ASCO-ONS 2024 read as raw text from the open ONS PDF and NICE NG197 fetched from nice.org.uk) · Each reason now says only what ASCO-ONS, NICE NG197 or BOPA states, with the source named.
- 2026-11-17 · Consent and education before anticancer treatment: who takes consent: The step on documenting consent said consent should be taken by a trained professional and cited ASCO-ONS, which does not say this. It is a UK standard from BOPA (3.6.1): a trained and appropriately experienced healthcare professional as determined by local, regional or national policy. (Second accuracy pass; ASCO-ONS 2024 and BOPA version 5.0, both read as PDF text) · The requirement is now attributed to BOPA and to the UK.
- 2026-11-17 · Consent and education before anticancer treatment: consent forms and the written summary: The page said UK consent forms are often specific to the regimen and signed before the first cycle, which no source read states, and that NICE asks for a written summary afterwards, where NICE NG197 asks that people are offered resources in their preferred format and gives a printout as one possibility. (Second accuracy pass; BOPA version 5.0 PDF text and NICE NG197 from nice.org.uk) · Both statements now follow the sources' wording.
- 2026-11-17 · Consent and education before anticancer treatment: adults who lack capacity: The page said people who may lack capacity are assessed and decided for under the law that applies where they are, with no source read that states it. NICE NG197 says only that it does not cover an adult who lacks capacity to make the decision at the time and points to a separate guideline. (Second accuracy pass; NICE NG197 overview page from nice.org.uk) · The sentence now says what NICE states, and the gaps say the law on capacity is not set out here.
- 2026-11-17 · Diet, alcohol and weight during cancer treatment: best supported version: The page said the ESPEN and ASCO guidelines agree on screening weight from diagnosis, avoiding calorie restriction and avoiding high-dose supplements. Those are ESPEN recommendations; the ASCO abstract, which is all that could be read, covers none of them. It also said both call the evidence for weight loss during treatment limited, which only ASCO says. (Second accuracy pass; ESPEN 2021 full text (PDF read as text) and the ASCO 2022 abstract from Europe PMC) · Rewritten to say which guideline says what, with the scope of each exercise recommendation.
- 2026-11-17 · Diet, alcohol and weight during cancer treatment: alcohol step: The page told readers to limit or avoid alcohol with low platelets or diarrhoea and said alcohol adds to bleeding risk and worsens dehydration. NCI says people may be advised to limit or avoid it if platelets are low, Macmillan says it can help to avoid it with diarrhoea, and neither mentions dehydration. The advice to ask before drinking had no source. (Second accuracy pass; NCI bleeding page, NCI Chemotherapy and You, NCI Eating Hints and Macmillan, read as full text) · Hedges restored, dehydration removed, and the NCI chemotherapy booklet added as the source for checking before drinking.
- 2026-11-17 · Diet, alcohol and weight during cancer treatment: neutropenic diet step: The step carried an exception for transplant or haematology teams and said such diets make eating harder. The ASCO abstract states neither. (Second accuracy pass; ASCO 2022 abstract from Europe PMC) · Both removed; a gap says the abstract does not address transplant recipients.
- 2026-11-17 · Diet, alcohol and weight during cancer treatment: oral anticancer drugs step: The page said to ask whether tablets are taken with or without food and that food changes how much drug is absorbed. The ACS page says to ask what to avoid and whether tablets can be mixed with food or liquid, and gives no mechanism. (Second accuracy pass; American Cancer Society, Getting Oral Chemotherapy, full text) · Reworded to the ACS page.
- 2026-11-17 · Diet, alcohol and weight during cancer treatment: reasons given for the steps: Four reasons went beyond the sources: that weight loss is harder to reverse once advanced, that keeping weight up makes treatment easier to complete, that high-dose supplements have not been shown to help, and that food poisoning is more serious during treatment. (Second accuracy pass; ESPEN 2021 full text and NCI Eating Hints (PDF read as text)) · Each reason reworded to what ESPEN or NCI states.
- 2026-11-17 · Diet, alcohol and weight during cancer treatment: short summaries and scope: The summaries said to keep weight steady, that this is not the time to cut calories, and that no special restricted diet is needed; the scope said transplant recipients follow their team's food rules. NCI's weight advice is about the time before treatment, ESPEN's rule on calorie restriction is for people malnourished or at risk, ASCO speaks only of neutropenic diets, and no source read gives the transplant rule. (Second accuracy pass; ESPEN 2021, ASCO 2022 abstract and NCI Eating Hints) · Reworded to the sources; transplant food rules now point to the food safety protocol.
- 2026-11-17 · Diet, alcohol and weight during cancer treatment: food list and transplant variants: The page said the NCI food list goes further than the trials support, scoped it to high infection risk, and said some transplant units keep food restrictions for weeks. None of these is in a source read. (Second accuracy pass; NCI Eating Hints and ESPEN 2021 full text) · Now says the list goes further than ASCO's recommendation, uses NCI's own scope, and gives ESPEN's commentary on diets after transplant.
- 2026-11-17 · Distress screening with the distress thermometer, and referral by need: thoughts of suicide or self-harm: The page said no source it had read spells out how urgently thoughts of suicide are assessed. The ASCO 2014 guideline, already cited, says a patient identified as at risk of harm to self or others should be referred to appropriate services for emergency evaluation, with a safe environment and one-to-one observation, and that teams should check their hospital's preferred pathway for a mental health emergency. (Second accuracy pass; ASCO 2014 adaptation, full text read from PubMed Central as raw HTML (only the abstract had been read before)) · Added a step quoting the ASCO sentence and rewrote the population note.
- 2026-11-17 · Distress screening with the distress thermometer, and referral by need: the Calgary screening trial: The page gave the trial as 1,134 patients and said full screening and triage lowered distress against the thermometer alone. The abstract prints 585 patients with breast cancer and 549 with lung cancer, not a total, and reports the results separately: in lung cancer only the triage arm differed, in breast cancer both the full screening and triage arms did. (Second accuracy pass; Carlson 2010, abstract read through Europe PMC) · Replaced the total with the two printed figures and stated the result for each cancer as the abstract does.
- 2026-11-17 · Distress screening with the distress thermometer, and referral by need: the cut-off of 4: The page called the cut-off a convention with known limits and gave only the one study in which it was 95 per cent sensitive and 45 per cent specific. NCCN 2019 also reports a meta-analysis of 42 studies with a pooled sensitivity of 81 per cent and specificity of 72 per cent at a cut-off of 4, and says the thermometer has shown good sensitivity and specificity. (Second accuracy pass; NCCN Distress Management Version 3.2019, full text read through NCBI E-utilities) · Removed the framing and put the meta-analysis figures beside the single study.
- 2026-11-17 · Distress screening with the distress thermometer, and referral by need: treatment by severity: The step said to offer a psychological therapy first for moderate or severe symptoms, and the summary gave one list for moderate symptoms. ASCO 2023 gives separate lists: for moderate depression the options include mindfulness-based stress reduction and structured physical activity, for moderate anxiety they include structured physical activity and acceptance and commitment therapy, and a medicine may also be offered where one has worked before. (Second accuracy pass; ASCO 2023 guideline update, abstract read through Europe PMC) · Reworded the step and split the summary into the lists as printed.
- 2026-11-17 · Distress screening with the distress thermometer, and referral by need: referral after a positive screen: The page said every positive screen should lead to a referral. NCCN 2019 says a score of 4 or more should trigger further evaluation and referral to an appropriate resource if needed. (Second accuracy pass; NCCN Distress Management Version 3.2019, full text read through NCBI E-utilities) · Reworded to further evaluation, referral where needed and follow-up.
- 2026-11-17 · Distress screening with the distress thermometer, and referral by need: when to screen: The page gave the end of treatment as one of NCCN's moments for screening and wrote set or regular intervals. NCCN 2019 says appropriate intervals and lists remission, recurrence, progression and treatment-related complications. Screening after treatment and at the transition to palliative care comes from ASCO 2014. (Second accuracy pass; NCCN 2019 and ASCO 2014 full texts read as raw text) · Gave each list to the guideline that prints it.
- 2026-11-17 · Distress screening with the distress thermometer, and referral by need: editions and sources: The page treated NCCN Version 1.2026 as current, said the ASCO 2014 guideline could be read only as an abstract, and said NICE's 2004 guidance does not grade. NCCN's website now lists Version 1.2027, which sits behind a login. The ASCO 2014 full text is readable on PubMed Central. The 2004 NICE text grades its evidence from A to C. (Second accuracy pass; NCCN guideline page, PubMed Central and the retired NICE PDF read as raw text) · Named the newer NCCN edition as unread in the gaps, marked ASCO 2014 as read in full, and corrected the note on NICE grading.
- 2026-11-17 · Driving, flying and travel during cancer treatment: live vaccines for life: The page named leukaemia and HIV-related cancer as the conditions after which live vaccines may have to be avoided for life, and left out lymphoma, which Macmillan names first. (Second accuracy pass; Macmillan, Cancer and travel vaccinations, fetched raw and read as text) · Lymphoma added, and the wording now follows Macmillan: some people have to avoid live vaccines for life, and people with these three conditions should always ask their cancer doctor.
- 2026-11-17 · Driving, flying and travel during cancer treatment: medicines and driving: The page said the drug-driving law covers prescribed opioids and benzodiazepines in general, and cited the law page for advice about sleepiness. The GOV.UK page lists named medicines (morphine and other opioids, methadone, six named benzodiazepines and amphetamine) and says nothing about sleepiness. (Second accuracy pass; GOV.UK, Drugs and driving: the law, fetched raw and read as text) · Reworded to the medicines the law lists and to impairment. The advice not to drive when sleepy stays, attributed to the Macmillan chemotherapy page that states it.
- 2026-11-17 · Driving, flying and travel during cancer treatment: who must tell the DVLA: The short summary said most people with cancer need not tell the DVLA and gave three of the four exceptions. The DVLA page does not say most people, applies the rule to car and motorcycle licences, and has a fourth exception for drivers restricted to certain or adapted vehicles. (Second accuracy pass; GOV.UK, Cancer and driving, fetched raw and read as text) · Short summary rewritten for a car or motorcycle driver with all four exceptions. The GOV.UK page that sends Northern Ireland drivers to the Driver and Vehicle Agency was added as a source.
- 2026-11-17 · Driving, flying and travel during cancer treatment: flying and clots: The page said low platelets, raised pressure in the brain and recent surgery can make flying unsafe, and told readers to ask how to reduce their clot risk. Macmillan says people may be advised not to fly, lists six reasons including breathlessness, anaemia and ear or sinus problems, and says only that the doctor can explain the clot risk. (Second accuracy pass; Macmillan, Can you fly with cancer?, fetched raw and read as text) · Reworded to Macmillan's full list and its wording, and to asking the doctor about the risk of a clot.
- 2026-11-17 · Driving, flying and travel during cancer treatment: travel vaccines: The page said to see a travel health professional at least 8 weeks before travel and described the advice as drawing on the UK Green Book. Macmillan says to speak to the GP, practice nurse or a travel health professional, if possible at least 8 weeks ahead, and only lists the Green Book among sample references for its travel pages. The page also said no source gave a waiting time before a live vaccine. (Second accuracy pass; Macmillan, Cancer and travel vaccinations, fetched raw; CDC Yellow Book 2026 edition, Immunocompromised Travelers, read from an Internet Archive capture of 4 October 2026) · Reworded with "if possible" and all three professionals. The CDC waiting periods (3 months after chemotherapy or radiotherapy, at least 6 months after lymphocyte-depleting drugs, 1 month after immunosuppressive steroid courses) added as a United States variant.
- 2026-11-17 · Driving, flying and travel during cancer treatment: a quotation: The quotation about food and drink carried a space before a comma that the Macmillan page does not have. (Second accuracy pass; Macmillan, Taking care while you are on holiday, fetched raw and checked by script) · Replaced with the sentence that follows it on the page, checked character for character.
- 2026-11-17 · Early palliative care alongside cancer treatment: referral of carers: The step told clinicians to offer carers a referral. The ASCO 2024 abstract says clinicians "may refer the caregivers" for additional support, which is permission and not an instruction. (Second accuracy pass; ASCO 2024 guideline update, abstract read as raw text from Europe PMC) · The step now says to consider referring carers, and its reason is limited to what ASCO and the Hui 2022 review say.
- 2026-11-17 · Early palliative care alongside cancer treatment: reasons given under the steps: Seven steps gave reasons that the cited text does not state, for example that early referral was compared with being seen only when problems became severe, that a patient is often the first to know, and that people in phase 1 trials carry a heavy symptom load. The trials compared early palliative care with standard care, and the ASCO abstract gives no reasons. (Second accuracy pass; ASCO 2024 abstract, Hui 2016 full text, Hui 2022 full text and NICE NG142, all read as raw text) · Each reason was reworded to what its source says, or replaced by a statement that the source gives no reason.
- 2026-11-17 · Early palliative care alongside cancer treatment: the consensus criteria: The page said any one major criterion is reason to refer and that the thresholds behind the criteria could not be read. The consensus paper, now read in full, says one major criterion alone could be sufficient to trigger a referral, defines a severe symptom as one scored 7 to 10 on a ten-point scale, and says the criteria should add to clinical judgement and not replace it. A request from the family was also listed as a trigger, and only a patient request is a criterion. (Second accuracy pass; Hui et al. 2016, published PDF read with pdftotext from the University of Milan open repository) · The wording now follows the paper, the severity score and three quoted statements were added, and the family request was removed from the trigger.
- 2026-11-17 · Early palliative care alongside cancer treatment: trial descriptions: The page said Temel 2010 was run at one US centre, that the STEP study was in Toronto, that STEP was not randomised, that most trials were at large centres, and that the stepped care trial compared with monthly visits. None of the first four is in the abstracts read, and the stepped care abstract says visits every 4 weeks. (Second accuracy pass; trial abstracts read as raw text from Europe PMC, and the Lancet Oncology Commission accepted manuscript) · The unsupported details were removed or attributed (the Commission describes the Temel 2010 site as a highly specialised institution), and monthly was changed to every 4 weeks.
- 2026-11-17 · Early palliative care alongside cancer treatment: the NICE position: The page said the NICE evidence review on the timing of referral had not been read, and described the holistic needs assessment in words NICE does not use. The review now reads in full: the committee concluded that the evidence on early referral was not sufficient to formulate a recommendation on timing. (Second accuracy pass; NICE NG142 recommendations page and evidence review B, PDF read with pdftotext) · The committee's conclusion was added with a verbatim quote and a new source entry, and the assessment is described in NICE's words.
- 2026-11-17 · Enhanced recovery after breast reconstruction: quoted recommendation on smoking, weight and alcohol: The quotation printed the body mass index unit as "30 kg/m 2", a mangled superscript that is not in the source, which prints kg/m2 with a superscript. (Second accuracy pass; ERAS Society consensus review 2017, PubMed Central full text read as XML through NCBI E-utilities) · Quote cut to its first verbatim sentence; the BMI of 30 or less stays in the step and the alcohol sentence is quoted separately in the grades.
- 2026-11-17 · Enhanced recovery after breast reconstruction: which items rest on high evidence: The best-in-class line said only opioid-sparing analgesia and conventional sutures rest on high-quality evidence. The review also prints a high evidence level for weight reduction in obesity. (Second accuracy pass; ERAS Society consensus review 2017, full text) · Obesity added to the high-evidence items in the statement and the summary.
- 2026-11-17 · Enhanced recovery after breast reconstruction: age of the evidence: The page said the guideline rests on literature to about 2016. Its methods say the databases were searched through January 2015. (Second accuracy pass; ERAS Society consensus review 2017, full text) · Corrected to January 2015.
- 2026-11-17 · Enhanced recovery after breast reconstruction: reasons given for steps: Several reasons said more than the review does: that smoking, obesity and heavy drinking each cause flap loss, wound breakdown and infection; that getting cold triples wound infection; that movement prevents clots; that early return to theatre saves a flap; and that mapping in general, not CT angiography compared with Doppler ultrasound, shortens the operation. The review reports different harms for each risk factor, an association for hypothermia, reduced risks and improved salvage. (Second accuracy pass; ERAS Society consensus review 2017, full text) · Each reason reworded to the review's own statement.
- 2026-11-17 · Enhanced recovery after breast reconstruction: carbohydrate drink and fasting: The page gave the carbohydrate drink as settled and said it reduces the body's stress response. The review says thirst, anxiety and catabolic losses, grades the evidence low, and notes the fasting advice has not been applied to people at risk of delayed gastric emptying; NICE NG180 says only to consider carbohydrate drinks before abdominal major or complex surgery. (Second accuracy pass; ERAS Society consensus review 2017 full text and NICE NG180 recommendations page read raw) · Reason reworded, the caveat added, and the NICE position added as a variant.
- 2026-11-17 · Enhanced recovery after breast reconstruction: grading scheme and scope: The page called the evidence levels GRADE and said reconstruction is most often a DIEP flap, immediate or delayed. The review assessed study quality with the Oxford Centre for Evidence Based Medicine criteria, used GRADE only for strong or weak, and does not make either scope statement. (Second accuracy pass; ERAS Society consensus review 2017, full text) · Grading scheme and population reworded.
- 2026-11-17 · Enhanced recovery after breast reconstruction: what the page does not cover: The gaps named antibiotic-coated implants, which the review does not discuss (it discusses coated sutures and antibiotic irrigation of the implant pocket), and implied the trial of regional anaesthesia and recurrence had reported, which no source read says. (Second accuracy pass; ERAS Society consensus review 2017, full text) · Both gaps reworded; gaps added on the length of clot prevention (the recommendation says until walking or discharged, the text says at least 7 to 10 days) and on fasting from solids.
- 2026-11-17 · Enhanced recovery after colorectal cancer surgery: the whole pathway: The page said the current ERAS Society 2025 guideline could not be read and built every step on the 2024 Korean guideline. The 2025 guideline is open access and reads in full, and it differs on several items. (Second accuracy pass; ERAS Society colorectal guideline 2025 (Surgery 184:109397), open-access PDF from the University of Melbourne repository read as raw text) · Steps, grades, summary and best-in-class statement rebuilt on the 2025 guideline with its printed quality of evidence and recommendation grade; the Korean guideline kept as a named variant.
- 2026-11-17 · Enhanced recovery after colorectal cancer surgery: fasting before surgery: The fasting step cited a carbohydrate-loading recommendation and gave no interval for food. The ERAS Society recommends solids until 6 hours and clear liquids until 2 hours before anaesthesia (strong). (Second accuracy pass; ERAS Society colorectal guideline 2025 (Surgery 184:109397), open-access PDF from the University of Melbourne repository read as raw text) · New fasting step with both intervals and the grade.
- 2026-11-17 · Enhanced recovery after colorectal cancer surgery: carbohydrate drink: The drink was presented as a standard part of the package. The 2025 guideline downgraded it from strong to weak and says it offers no advantage in comfort or clinical outcomes over short fasting. (Second accuracy pass; ERAS Society colorectal guideline 2025 (Surgery 184:109397), open-access PDF from the University of Melbourne repository read as raw text) · Reworded to a consider step with the weak grade; the EUPEMEN volumes are given in a variant.
- 2026-11-17 · Enhanced recovery after colorectal cancer surgery: prehabilitation: The page told teams to offer prehabilitation. The ERAS Society 2025 guideline makes no recommendation for a specific regimen because trial results conflict; only the Korean guideline recommends it (conditional, moderate). (Second accuracy pass; ERAS Society colorectal guideline 2025 (Surgery 184:109397), open-access PDF from the University of Melbourne repository read as raw text) · Reworded to a consider step that states both positions.
- 2026-11-17 · Enhanced recovery after colorectal cancer surgery: bowel preparation: The page implied bowel preparation with antibiotic tablets is routine and described an odds ratio of 0.51 as roughly halving wound infections. The ERAS Society says mechanical preparation should not be used routinely in colonic surgery and may be considered in rectal surgery with a diverting stoma, with oral antibiotics added if it is used. (Second accuracy pass; ERAS Society colorectal guideline 2025 (Surgery 184:109397), open-access PDF from the University of Melbourne repository read as raw text; Korean guideline, Europe PMC full text) · Step rewritten with scope and both grades; the halving claim removed.
- 2026-11-17 · Enhanced recovery after colorectal cancer surgery: blood clot prevention: The page said only to start a drug before surgery and that duration was not covered. The ERAS Society recommends mechanical plus pharmacological prevention, heparin started before surgery or within 24 hours after it, and continuation until 4 weeks after surgery in patients at risk (strong, high). (Second accuracy pass; ERAS Society colorectal guideline 2025 (Surgery 184:109397), open-access PDF from the University of Melbourne repository read as raw text) · Two steps added with the timing window, the 4-week duration and the grades.
- 2026-11-17 · Enhanced recovery after colorectal cancer surgery: urinary catheter, eating and walking: Removal the day after surgery and eating and walking from the day after were written as the rule. The ERAS Society scopes catheter removal to within 24 hours after uncomplicated minimally invasive colectomy and 48 hours after rectal surgery, asks for oral intake within hours, and mobilisation ideally on the day of surgery. (Second accuracy pass; ERAS Society colorectal guideline 2025 (Surgery 184:109397), open-access PDF from the University of Melbourne repository read as raw text) · Steps reworded with the scope; the Korean day-after position kept in the grades and variant.
- 2026-11-17 · Enhanced recovery after colorectal cancer surgery: reasons given for steps: Several reasons had no support in the cited source, for example that early walking reduces chest problems and clots (the guidelines found no clear effect on complications) and that early catheter removal gets people home sooner. (Second accuracy pass; ERAS Society colorectal guideline 2025 (Surgery 184:109397), open-access PDF from the University of Melbourne repository read as raw text; Korean guideline, Europe PMC full text) · Each reason rewritten from the source text.
- 2026-11-17 · Enhanced recovery after colorectal cancer surgery: epidural variant: The page said it could not tell whether the 2025 edition kept the 2018 positions on thoracic epidurals and a catheter for 1 to 3 days. It did not: an epidural should be considered in open surgery (weak) and not used in minimally invasive surgery (strong), and the catheter may come out after 2 days with an epidural. (Second accuracy pass; ERAS Society colorectal guideline 2025 (Surgery 184:109397), open-access PDF from the University of Melbourne repository read as raw text) · Variant rewritten with the 2025 grades.
- 2026-11-17 · Enhanced recovery after colorectal cancer surgery: NICE quotation: A quoted NICE recommendation had a stray space before the full stop and so was not verbatim. (Second accuracy pass; NICE NG180 recommendations page read as raw text) · Quote corrected to the published wording.
- 2026-11-17 · Enhanced recovery after cytoreductive surgery with or without HIPEC: two NICE quotations: Both quotations from NICE NG180 (recommendations 1.2.1 and 1.4.2) carried a stray space before the full stop, so neither was the source's text character for character. (Second accuracy pass; NICE NG180 recommendations page read as raw text and checked by script) · Both quotations now match NICE exactly.
- 2026-11-17 · Enhanced recovery after cytoreductive surgery with or without HIPEC: standing of the ERAS guideline: The page called the ERAS Society guideline the only specialty guideline for this operation. The Indian consensus it cites also draws on a 2019 guideline from the Society of Onco-Anaesthesia and Perioperative Care on the same operation. (Second accuracy pass; INDEPSO-ISPSM consensus full text from Europe PMC) · It is now called the dedicated ERAS guideline for this operation, in the summary and the best in class statement.
- 2026-11-17 · Enhanced recovery after cytoreductive surgery with or without HIPEC: drains in the Indian consensus: The page said the Indian panel recommends drains for every patient and places them in every patient. The consensus statement says drains should be considered for all patients (79.17% of votes in round two). (Second accuracy pass; INDEPSO-ISPSM consensus full text from Europe PMC) · Summary, best in class and the variant now say considered, with the vote.
- 2026-11-17 · Enhanced recovery after cytoreductive surgery with or without HIPEC: reasons given for the steps: Eleven reasons were not in the cited source, among them that clot risk lasts after discharge, that chlorhexidine reduces wound infection, that a cold or hot body causes complications, that a catheter left in raises infection, and that most people do not need a nose tube. (Second accuracy pass; INDEPSO-ISPSM consensus full text from Europe PMC and NICE NG180 read as raw text) · Each reason now says what the source says, for example a 30 to 50% clot risk without prophylaxis, or says that the source gives no reason.
- 2026-11-17 · Enhanced recovery after cytoreductive surgery with or without HIPEC: fluids, mobilisation and catheter steps: The fluid step ordered arterial line and cardiac output monitoring where the consensus says goal-directed management should be attempted and monitoring may assist. The mobilisation step said more than 6 hours a day where the source says more than six hours. The catheter step dropped the words as early as. (Second accuracy pass; INDEPSO-ISPSM consensus full text from Europe PMC) · The three steps now carry the source's hedges and wording.
- 2026-11-17 · Enhanced recovery after cytoreductive surgery with or without HIPEC: who the page applies to: The population named adults, elective surgery and five primary cancers. None of the sources read states an age range or lists the primary cancers for this guideline. (Second accuracy pass; ERAS Society abstracts from Europe PMC and the INDEPSO-ISPSM full text) · The population now gives the ERAS panel as the abstract describes it and says what the sources do not state.
- 2026-11-17 · Enhanced recovery after cytoreductive surgery with or without HIPEC: fasting, clot prevention and feeding figures: The fasting times, the four weeks of clot prevention and the start of feeding were not clearly tied to the body that states each, and the page did not say that the ERAS guideline's own figures could not be read. (Second accuracy pass; INDEPSO-ISPSM consensus full text; the ERAS guideline full text returned 403 on 17 November 2026) · Each figure is now attributed with the Indian panel's vote, the Indian day 1 start for clear liquids is shown beside the ERAS day of surgery, and a gap names what could not be read.
- 2026-11-17 · Enhanced recovery after gastrectomy (ERAS pathway for stomach cancer surgery): when eating starts: The patient summary said drinking and eating start the day after surgery for everyone. The ERAS item is written for total gastrectomy, and the Japanese guideline starts solid food on days 2 to 4 and reports more complications with earlier eating. (Second accuracy pass; Yamagata et al. 2019 Table 1 and the Japanese guidelines 2025, both read as full text from Europe PMC) · The summaries now scope next-day eating to removal of the whole stomach and say hospitals differ on the day.
- 2026-11-17 · Enhanced recovery after gastrectomy (ERAS pathway for stomach cancer surgery): Japanese pathway: The Japanese pathway was described from the 2018 guideline through a review. A 2025 (7th) edition exists and can be read. (Second accuracy pass; Japanese gastric cancer treatment guidelines 2025, full text from Europe PMC) · The variant now quotes the 2025 edition directly, with its days for tube removal, fluids, solid food, drain removal and discharge.
- 2026-11-17 · Enhanced recovery after gastrectomy (ERAS pathway for stomach cancer surgery): reasons given for steps: Eight steps gave reasons the cited source does not state, such as a nose tube not preventing leaks or vomiting, early eating being safe, and feeding adding nothing for well-nourished people; two dropped the source's "may". (Second accuracy pass; Yamagata et al. 2019, Liu et al. 2025, NICE NG83 and NG180 read as raw text) · Each reason now says only what its source says.
- 2026-11-17 · Enhanced recovery after gastrectomy (ERAS pathway for stomach cancer surgery): scope of the page: The page said general elements such as warming and clot prevention were taken from general sources, but no step or source on the page covers them. (Second accuracy pass; the ERAS 2014 full text again refused a script, so its general items were not read) · The population and gaps now say these elements are not covered and that the fasting steps come from NICE guidance for England.
- 2026-11-17 · Enhanced recovery after gastrectomy (ERAS pathway for stomach cancer surgery): quotations: Two NICE quotations had a space before the full stop that is not in the source, and one ERAS abstract quotation began with the heading word "Results". (Second accuracy pass; quotes checked by script against NICE NG180 and the Europe PMC abstract) · The three quotations now match the sources character for character.
- 2026-11-17 · Enhanced recovery after gastrectomy (ERAS pathway for stomach cancer surgery): wound catheters and drains in Japan: The summary said there was no gastrectomy evidence for wound catheters, which the source says only of TAP block, and drains were called common in Japan where the review only presumes this. (Second accuracy pass; Yamagata et al. 2019 full text from Europe PMC) · Wound catheters and TAP block are now graded separately, and the statement about Japan is given as the review's presumption.
- 2026-11-17 · Enhanced recovery after gynaecological cancer surgery: evidence behind stopping the drip: The page said stopping intravenous fluids within 24 hours rests on high-quality evidence. The ERAS Society 2016 guideline grades it moderate evidence, strong recommendation. (Second accuracy pass; ERAS Society 2016 Part II, full text read as XML from Europe PMC) · The best-in-class statement now says moderate.
- 2026-11-17 · Enhanced recovery after gynaecological cancer surgery: blood clot prevention: The page said four weeks of injections that more than halve the risk. The guideline recommends extended prophylaxis for 28 days after laparotomy for abdominal or pelvic cancer without naming injections or a drug, and reports clots in 6.1% against 14.3% in a review of four trials. (Second accuracy pass; ERAS Society 2016 Part II, full text read as XML from Europe PMC) · The summary lines now say 28 days of clot prevention, the reason gives the two percentages the guideline prints, and a gap says the drug and dose are not named.
- 2026-11-17 · Enhanced recovery after gynaecological cancer surgery: blood glucose threshold: The step gave 10 to 11 mmol/l beside the guideline's 180 to 200 mg/dL. No source read prints the mmol/l figure; it was the writer's own conversion. (Second accuracy pass; ERAS Society 2016 Part II, full text read as XML from Europe PMC) · The conversion was removed, the step now gives the guideline's instruction (insulin infusion above 180 to 200 mg/dL with regular checks), and a gap says the guideline prints mg/dL only.
- 2026-11-17 · Enhanced recovery after gynaecological cancer surgery: reasons given for five steps: The reasons for the enhanced recovery programme, stopping the drip, avoiding drains and early walking said more than the sources do (that the programme shortens recovery, that drip fluid causes swelling and slows the bowel, that drains delay recovery, and that walking reduces clots and chest infection as a fact where the guideline calls the benefits hypothesised). (Second accuracy pass; ERAS Society 2016 Part II full text from Europe PMC and NICE NG180 recommendations page) · Each reason was reworded to what the source states.
- 2026-11-17 · Enhanced recovery after gynaecological cancer surgery: catheter timing and epidural wording: The summary lines said the catheter comes out within a day where the guideline says a short period, preferably under 24 hours, and said an epidural often needs added intravenous opioids where the recommendation says patients may require them. The summary lines also did not say that the recommendations come from the 2016 edition. (Second accuracy pass; ERAS Society 2016 Part II, full text read as XML from Europe PMC) · The hedges were restored and the short summary now says the 2026 update could not be read.
- 2026-11-17 · Enhanced recovery after gynaecological cancer surgery: NICE quotation: The quotation of NICE recommendation 1.2.1 had a stray space before its full stop, so it was not character for character. (Second accuracy pass; NICE NG180 recommendations page, raw HTML) · The space was removed and the quotation verified by script.
- 2026-11-17 · Enhanced recovery after major head and neck cancer surgery with free flap reconstruction: the elements around the operation: The page said a carbohydrate drink, no prolonged fasting, antibiotics and short-acting anxiolytics only. The abstract, the only part of the ERAS review that could be read, says preoperative carbohydrate treatment, the avoidance of preoperative fasting, antibiotics in clean-contaminated procedures and short-acting anxiolytics, with no form, length of fast or "only". (Second accuracy pass; ERAS Society review 2017, abstract read raw from Europe PMC) · Step, summary and TL;DR reworded to the abstract's own terms
- 2026-11-17 · Enhanced recovery after major head and neck cancer surgery with free flap reconstruction: reasons given for the ERAS elements and flap checks: The page gave reasons (less stress response, sickness and pain; a clot in the flap must be found quickly; early movement prevents chest infection and clots) that are not in the text that could be read. (Second accuracy pass; ERAS Society review 2017, abstract read raw from Europe PMC) · Reasons replaced with what the abstract says, and the page now says the reasons, doses, times and grades are in the unread full text
- 2026-11-17 · Enhanced recovery after major head and neck cancer surgery with free flap reconstruction: choosing a feeding tube: The page said a gastrostomy suits feeding of more than about 4 to 6 weeks and that a nasal tube is otherwise usually enough, decided by duration alone. ASPEN says a gastrostomy may be more appropriate for longer durations (commonly more than 4 to 6 weeks), that otherwise a nasogastric tube should be considered, and that the choice also rests on the clinical situation, symptoms, treatment plan and psychosocial situation. (Second accuracy pass; ASPEN 2026 full text, Europe PMC XML) · Step reworded with the guideline's hedges and its other criteria
- 2026-11-17 · Enhanced recovery after major head and neck cancer surgery with free flap reconstruction: feeding after surgery: The page said to start nutrition within 24 hours as a flat instruction and that feeding starts within a day. ASPEN suggests it (weak recommendation, very low certainty) and says early eating by mouth may be inappropriate in some situations. (Second accuracy pass; ASPEN 2026 full text, Europe PMC XML) · Reworded to an aim, with tube feeding kept until enough is taken by mouth
- 2026-11-17 · Enhanced recovery after major head and neck cancer surgery with free flap reconstruction: reasons for dietitian, screening and swallowing steps: The page said eating can be affected for weeks, that the protein and energy amounts both meet the needs of most people, and that a swallowing assessment reduces swallowing problems. ASPEN gives no duration, says "the needs of most" of protein only, and says speech pathology may improve or prevent decline in swallowing. (Second accuracy pass; ASPEN 2026 full text, Europe PMC XML) · Each reason reworded to the guideline's rationale
- 2026-11-17 · Enhanced recovery after major head and neck cancer surgery with free flap reconstruction: how strong the nutrition evidence is: The page said most of the ASPEN recommendations it uses are expert opinion graded strong. Four of the ten are; three are weak recommendations, and the arginine grade lacked its strength (weak). (Second accuracy pass; ASPEN 2026 full text, Europe PMC XML) · Counts corrected and each recommendation's grade named
- 2026-11-17 · Enhanced recovery after major head and neck cancer surgery with free flap reconstruction: NICE quotation: The quoted NICE recommendation had a stray space before its full stop, so it was not the source's exact text. (Second accuracy pass; NICE NG180 recommendations page, raw HTML) · Quote corrected; NICE's 2-hour clear-fluid advice and its narrower advice on carbohydrate drinks added
- 2026-11-17 · Enhanced recovery after liver surgery (ERAS pathway for hepatectomy): pain relief: The page said wound infiltration and a TAP block both give pain relief equal to an epidural with fewer complications, and that an epidural is not preferred. The guideline says this of continuous wound infiltration only, calls a TAP block a supplement to standard analgesia, and does not rule out an epidural. (Second accuracy pass; ERAS Society liver surgery recommendations 2022, full text re-read from Europe PMC) · The step, its reason and the best in class statement now say what the guideline says of each technique.
- 2026-11-17 · Enhanced recovery after liver surgery (ERAS pathway for hepatectomy): fluids and warming: One step joined low central venous pressure and fluids with keeping the patient warm, and carried the evidence grade for warming (moderate). The guideline grades the fluid recommendation high. (Second accuracy pass; ERAS Society liver surgery recommendations 2022, full text re-read from Europe PMC) · Split into two steps, each with its own quotation and grade.
- 2026-11-17 · Enhanced recovery after liver surgery (ERAS pathway for hepatectomy): reasons given for steps: Several reasons were not in the guideline: that early movement reduces chest infections, that counselling makes people take part in recovery, that an antibiotic dose reduces wound infection in liver surgery (the guideline says hard data are lacking), that tubes and drains add discomfort, and that feeding beforehand reduces complications. (Second accuracy pass; ERAS Society liver surgery recommendations 2022, full text re-read from Europe PMC) · Each reason reworded to what the guideline reports.
- 2026-11-17 · Enhanced recovery after liver surgery (ERAS pathway for hepatectomy): patient summaries: The summaries said eating starts on the day of surgery and that everyone is helped to get fit. The guideline says early eating without naming a day, and recommends prehabilitation for high-risk patients only, on a weak grade. (Second accuracy pass; ERAS Society liver surgery recommendations 2022, full text re-read from Europe PMC) · Both summaries reworded. The simple summary now tells readers to follow their own team's fasting times.
- 2026-11-17 · Enhanced recovery after liver surgery (ERAS pathway for hepatectomy): nutrition and steroid wording: The page said 7 to 14 days of supplements and a single 500 mg dose of methylprednisolone. The guideline says enteral supplementation for at least 7 to 14 days, and gives the dose without saying it is single. (Second accuracy pass; ERAS Society liver surgery recommendations 2022, full text re-read from Europe PMC) · Wording matched to the guideline.
- 2026-11-17 · Enhanced recovery after liver surgery (ERAS pathway for hepatectomy): quotations: Two NICE quotations carried a stray space before the full stop, four grade lines a stray space before the colon, and one ERAS quotation a broken subscript. (Second accuracy pass; NICE NG180 recommendations page and the ERAS full text, checked by script) · Quotations and grade lines corrected to the printed text, and the broken quotation replaced.
- 2026-11-17 · Enhanced recovery after lung cancer surgery (ERAS pathway for lung resection): how the two guidelines relate: The page said the Italian PACTS consensus agrees with and grades every element the ERAS abstract names. PACTS makes no recommendation on preventing blood clots and gives no recommendation favouring one anaesthetic agent (volatile against propofol is graded I). (Second accuracy pass; PACTS 2020 parts 1 and 2, Europe PMC full-text XML read raw) · The statement now says PACTS grades most of the elements and names the two it does not.
- 2026-11-17 · Enhanced recovery after lung cancer surgery (ERAS pathway for lung resection): reasons given for nine steps: The reasons for fasting limits, avoiding benzodiazepines, warming, fluid balance, nausea prevention, nerve blocks, tubes and catheters, drain suction and early mobilisation stated mechanisms the consensus does not give (for example that cold raises the risk of infection, bleeding and heart problems, or that too much fluid injures the lung). (Second accuracy pass; PACTS 2020 parts 1 and 2, Europe PMC full-text XML read raw) · Each reason now says what the consensus rationale says, with the trial figures it prints.
- 2026-11-17 · Enhanced recovery after lung cancer surgery (ERAS pathway for lung resection): protective ventilation: The page said protective ventilation reduces injury to the remaining lung as settled fact. The consensus says the trials are few and small: one found lung problems in 4% against 22%, another found no advantage. (Second accuracy pass; PACTS 2020 parts 1 and 2, Europe PMC full-text XML read raw) · The reason now gives the trial result and the caveat, and the step states the tidal volume limit of 6 mL/kg ideal body weight or less.
- 2026-11-17 · Enhanced recovery after lung cancer surgery (ERAS pathway for lung resection): fasting and drinking after surgery: The fasting limits were given for everyone, but the consensus scopes them to patients not at risk of delayed gastric emptying. Drinking 4 to 6 hours after surgery was written as an order, where the consensus says oral intake can be started then and should be adapted to tolerance. Paracetamol was written as part of the order, where the consensus says it may add benefit. (Second accuracy pass; PACTS 2020 parts 1 and 2, Europe PMC full-text XML read raw) · The scope and the hedges are now on the page.
- 2026-11-17 · Enhanced recovery after lung cancer surgery (ERAS pathway for lung resection): who PACTS covers and the variants: The page said PACTS covers thoracic surgery generally; it says it is specifically about lung resection for cancer and was written by six Italian societies. The drain variant called the PACTS threshold lower than the ERAS one, which depends on body weight, and two variants described practices in words the source does not use. (Second accuracy pass; PACTS 2020 parts 1 and 2, Europe PMC full-text XML read raw) · Population corrected and the three variants reworded to what PACTS says.
- 2026-11-17 · Enhanced recovery after oesophagectomy (ERAS pathway for oesophageal cancer surgery): reasons given for steps: Thirteen steps gave reasons the cited sources do not state, such as cold raising the risk of infection and bleeding, gut feeding being safer than vein feeding, and a clot risk lasting weeks after discharge. The summary table read lists the recommendations without reasons. (Second accuracy pass; Halliday and Moorthy 2022 read as full text from Europe PMC, NICE NG83 and NG180 recommendations pages) · Each reason now says only what the source says, or says that the source gives no reason.
- 2026-11-17 · Enhanced recovery after oesophagectomy (ERAS pathway for oesophageal cancer surgery): patient summaries: The patient text promised early mouth feeding, removal of tubes in general and injections against clots, and left out the caution about solid food for people with swallowing difficulty. The source recommends early tube feeding, gives a criterion for chest drains only, names no drug or route for clot prevention, and attaches the caution to the six hour rule. (Second accuracy pass; Halliday and Moorthy 2022 Table 1, full text from Europe PMC) · The summaries now say feeding through a tube, chest drains, treatment to prevent clots, and carry the swallowing caution.
- 2026-11-17 · Enhanced recovery after oesophagectomy (ERAS pathway for oesophageal cancer surgery): comparison with other ERAS pathways: The page said nasogastric decompression and epidural analgesia differ deliberately from most ERAS pathways. No source read makes that comparison. (Second accuracy pass; ERAS Society 2019 abstract and Halliday and Moorthy 2022 full text, Europe PMC) · The comparison was removed; the page now reports what the guideline abstract says about sections new to ERAS programmes.
- 2026-11-17 · Enhanced recovery after oesophagectomy (ERAS pathway for oesophageal cancer surgery): fluids, feeding target and study type: The page said balanced crystalloid where the source says a balanced fluid replacement strategy using crystalloid, said feeding reaches full calories by day 3 to 6 where the source says aim to, and called the study of pathway elements a meta-analysis where the review cites a component analysis. (Second accuracy pass; Halliday and Moorthy 2022 full text, Europe PMC) · All three were reworded to the source.
- 2026-11-17 · Enhanced recovery after oesophagectomy (ERAS pathway for oesophageal cancer surgery): variants and scope: The variants described what some centres do with feeding routes, early oral intake, exercise testing and pyloroplasty, and the population named adults, elective surgery and chemoradiotherapy. The sources read do not say these things. A NICE recommendation to offer nutrition was written as an order to start it. (Second accuracy pass; Halliday and Moorthy 2022 full text and NICE NG83 recommendation 1.6.2) · Variants and population now follow the sources, the step says offer, and NICE is labelled as UK guidance.
- 2026-11-17 · Enhanced recovery after oesophagectomy (ERAS pathway for oesophageal cancer surgery): NICE NG180 quotation: The quoted recommendation 1.2.1 had a stray space before its full stop, so it was not character for character the source. (Second accuracy pass; NICE NG180 recommendations page read as raw text) · The quotation now matches the page exactly.
- 2026-11-17 · Enhanced recovery after radical cystectomy: drains: The page said the 2025 surgeons' consensus was against leaving a drain at the operation site. The full text says the opposite: the panel was against leaving the drain out and recommends draining the pelvis after cystectomy. The first pass had only the abstract, whose wording is ambiguous. (Second accuracy pass; Albisinni et al., Eur J Surg Oncol 2025, publisher's version read in full as PDF text from the Amsterdam UMC repository) · Summary and variant corrected, and the reading of the abstract explained in the gaps.
- 2026-11-17 · Enhanced recovery after radical cystectomy: who wrote the 2025 update: The page called the 2025 paper the ERAS Society's own update. It is by the EAU Young Academic Urologists urothelial carcinoma working group and the ERUS society, a panel of 23 urological surgeons, and its authors say it lacks the wider range of specialists of the 2013 guideline. (Second accuracy pass; the same paper, full text) · Attributed correctly in the summary, best-in-class statement, source entry and gaps.
- 2026-11-17 · Enhanced recovery after radical cystectomy: clot prevention: The step said to give LMWH and the patient summary said four weeks of injections. The EAU recommendation is to offer pharmacological prophylaxis, such as LMWH, for at least four weeks, and the 2025 update allows heparin or an oral factor Xa inhibitor. (Second accuracy pass; EAU 2026 guideline, online full text read raw, and the 2025 update full text) · Reworded to offer, at least four weeks, with LMWH as the example; the oral option added as a variant.
- 2026-11-17 · Enhanced recovery after radical cystectomy: prehabilitation and reasons given: Prehabilitation was written as something to offer, where the EAU text only says it may help and that trials are missing. Reasons were given that no source states: that bowel preparation causes dehydration and discomfort, that too much fluid slows the bowel, and that people on ERAS had a little more pain (the source prints scores of 3.1 against 1.1). Early mobilisation was credited to the EAU guideline, which does not name it. (Second accuracy pass; EAU 2026 guideline and the 2025 update full text) · Step reworded to consider; unsourced reasons replaced with what the sources report; early mobilisation and early oral diet moved to a step resting on the 2025 update.
- 2026-11-17 · Enhanced recovery after radical cystectomy: nasogastric tube and item grades: The step on removing the nasogastric tube at the end of surgery left out the source's exception for complications during the operation, and the steps resting on the 2025 update carried no evidence level because only its abstract had been read. (Second accuracy pass; the 2025 update full text) · Exception added; each step now quotes the full text with its printed level of evidence and expert consensus; steps added for medical optimisation, fasting times (6 hours solids, 2 hours liquids) and early diet.
- 2026-11-17 · Enhanced recovery after radical cystectomy: Hong Kong source title: The Hong Kong consensus was listed under a title ending in muscle-invasive bladder cancer. Its title ends in muscle-invasive and advanced urothelial carcinoma. (Second accuracy pass; Crossref record and Europe PMC full text) · Title corrected.
- 2026-11-17 · Enhanced recovery after pancreatoduodenectomy: the ERAS Society guideline: The page said only the abstract of the ERAS Society 2019 recommendations could be read, gave no item grades, and took the drain, nasogastric tube, biliary drainage and somatostatin items from a narrative review. The full text is openly available from a university repository and prints an evidence level and a strong or weak grade for each of 27 items. (Second accuracy pass; ERAS Society pancreatoduodenectomy recommendations 2019 (World J Surg 2020), publisher PDF from the Amsterdam UMC repository read as text) · Steps, grades and variants rebuilt on the guideline's own words, with its figures for fasting, antibiotic timing, clot prevention, warming and drain removal.
- 2026-11-17 · Enhanced recovery after pancreatoduodenectomy: drain removal: The page said to place a drain and remove it within 72 hours. The guideline says early removal at 72 hours when drain amylase on day 1 is below 5000 U/L, and says only that routine drain placement might be recommended while the evidence on leaving no drain conflicts. (Second accuracy pass; ERAS 2019 full text, item 19) · Wording changed to removal at 72 hours, with the hedge on routine placement and the printed grade (evidence High, recommendation Strong).
- 2026-11-17 · Enhanced recovery after pancreatoduodenectomy: nutrition before surgery: The page said feeding beforehand helps. The guideline recommends nutritional intervention for severe weight loss (more than 15%, or BMI below 18.5) and not as a general measure, and says it remains unproven that such support reduces complications or speeds recovery. (Second accuracy pass; ERAS 2019 full text, item 5) · Reason reworded to say the benefit is unproven; the BMI criterion and the limit to severe weight loss added.
- 2026-11-17 · Enhanced recovery after pancreatoduodenectomy: biliary stent: The page attached NICE's metal stent recommendation to four situations (not yet fit, very high bilirubin, chemotherapy first, cholangitis). NICE makes it only for people not yet fit enough for surgery; the other three are the ERAS guideline's reasons for drainage, and that guideline found neither plastic nor metal stents better for complications after the operation. (Second accuracy pass; NICE NG85 recommendation 1.7.3 and ERAS 2019 item 3, both read as raw text) · The step is scoped to NICE's wording, the ERAS reasons for drainage have their own step, and the difference on stent type is stated.
- 2026-11-17 · Enhanced recovery after pancreatoduodenectomy: somatostatin analogues: The page said the ERAS guideline finds moderate evidence that somatostatin analogues may reduce fistula severity. The guideline reports a Cochrane review in which clinically significant fistulas were not significantly fewer, and says systematic use cannot be recommended (evidence Moderate, recommendation Weak). (Second accuracy pass; ERAS 2019 full text, item 20) · Variant rewritten from the guideline.
- 2026-11-17 · Enhanced recovery after pancreatoduodenectomy: reasons given for steps: Several reasons were not in the cited sources: that long fasting leaves people weaker, that a drain detects a leak, that gut feeding helps the bowel recover, and that most items rest on cohort studies. The order to act on compliance data was cited to a meta-analysis that reports only an association. (Second accuracy pass; ERAS 2019 full text, NICE NG85 and NG180, and the 2023 meta-analysis abstract) · Reasons reworded to what the sources say; the audit step now cites the guideline's audit item.
- 2026-11-17 · Enhanced recovery after pancreatoduodenectomy: NICE quotations: Two quoted NICE NG180 recommendations had a stray space before the full stop and so were not verbatim. (Second accuracy pass; NICE NG180 recommendations page read as raw text) · Quotes corrected to the published wording.
- 2026-11-17 · Exercise during and after cancer treatment: assessment, dose and referral: who needs clearance first: The page said people with recent lung or abdominal surgery need a doctor's clearance. The ACSM roundtable's triage table lists lung or abdominal surgery and does not say recent; it also lists a worsening or changing physical condition, which the summary left out. (Second accuracy pass; ACSM 2019 roundtable full text (PubMed Central author manuscript through NCBI efetch), Table 4) · Removed the word recent, completed the list, and said in the gaps that the table does not say how recent.
- 2026-11-17 · Exercise during and after cancer treatment: assessment, dose and referral: the dose: The patient summary said about 30 minutes three times a week such as brisk walking, starting low, and the clinical summary said 8 to 12 weeks. The roundtable says at least 30 minutes, at least three times a week, for at least 8 to 12 weeks; it does not mention brisk walking, and its start low advice is about resistance training with lymphoedema. (Second accuracy pass; ACSM 2019 roundtable full text, section 6) · Restored at least to each figure, removed brisk walking and starting low, and removed the unsourced if possible from the resistance training advice.
- 2026-11-17 · Exercise during and after cancer treatment: assessment, dose and referral: avoid inactivity: The first step cited the ASCO 2022 guideline for telling every survivor to avoid inactivity, and gave as a reason that being inactive slows recovery. The ASCO abstract says neither; avoid inactivity is the ACSM roundtable's conclusion. (Second accuracy pass; ASCO 2022 abstract (Europe PMC) and ACSM 2019 full text) · Split the step: the ASCO step now carries only ASCO's recommendation, and a new step cites and quotes the roundtable.
- 2026-11-17 · Exercise during and after cancer treatment: assessment, dose and referral: reasons given for triage and referral: Four reasons were not in the cited sources: that most people can simply start, that the higher-risk group risks fracture or a heart problem and needs a supervised programme, and that an exercise professional tailors the programme and keeps it safe. The referral step also added where one is available to an Australian recommendation. (Second accuracy pass; ACSM 2019 Table 4 and the COSA 2018 abstract (Europe PMC)) · Reworded each reason to what the table or abstract says, named COSA as Australian, and removed the added condition.
- 2026-11-17 · Exercise during and after cancer treatment: assessment, dose and referral: the colon cancer trial: The page called CHALLENGE the first phase 3 evidence of a survival effect, described its end point as the cancer coming back or dying, gave its target as about 150 minutes a week, and said ESMO recommends a structured programme. ESMO calls it the first high-level evidence of a disease-free survival benefit in resected high-risk colon cancer; the end point also counts a new primary cancer; ESMO equates the target to at least 150 minutes; and ESMO says participation can be recommended. (Second accuracy pass; ESMO Express Update full text (Europe PMC) and the CHALLENGE abstract (Europe PMC)) · Reworded all four to the sources, and noted that ESMO's text gives 962 patients where the trial abstract gives 889.
- 2026-11-17 · Accreditation standards for stem cell transplant and CAR T-cell programmes (FACT-JACIE): minimum patient numbers: The page gave only one floor for CAR T-cell and other immune effector cell therapy, 3 new patients a year, which is the figure for a programme that also transplants. The immune effector cell standards print a different floor for a programme that does not transplant: 5 new patients at a single site treating adults or children. (Second accuracy pass; FACT-JACIE Immune Effector Cells standards version 3.1, Appendix I, PDF from ebmt.org read as raw text) · Both figures are now given, each with the kind of programme it applies to.
- 2026-11-17 · Accreditation standards for stem cell transplant and CAR T-cell programmes (FACT-JACIE): registry data: Collecting complete registry data was listed among the firm requirements for every programme. It is a requirement in the transplant standards and only a recommendation in the immune effector cell standards for programmes that do not transplant. (Second accuracy pass; standard B9.2 in both documents, PDFs from ebmt.org read as raw text) · The summary, the step and the best-supported statement now say which document requires it and which recommends it, with the wording of each quoted.
- 2026-11-17 · Accreditation standards for stem cell transplant and CAR T-cell programmes (FACT-JACIE): survival outcomes: The recommendation to reach an expected one-year survival was written as if it belonged to the standards in general. It is in the transplant standards only. The immune effector cell standards have no survival standard. (Second accuracy pass; section B4.7 of both documents, PDFs from ebmt.org read as raw text) · The statement is now scoped to the transplant standards, and the page says what the immune effector cell standards ask for instead.
- 2026-11-17 · Accreditation standards for stem cell transplant and CAR T-cell programmes (FACT-JACIE): patient summary: The short summary said a programme must have round-the-clock access to medicines and intensive care. The standards ask for a pharmacy with 24-hour availability, and separately for access to an intensive care unit or emergency services with no 24-hour wording. (Second accuracy pass; standards B2.11 and B2.12 of the haematopoietic standards version 9.1, PDF from ebmt.org read as raw text) · The summary now says medicines available at any hour and access to intensive care or emergency services.
- 2026-11-17 · Accreditation standards for stem cell transplant and CAR T-cell programmes (FACT-JACIE): reasons given for each step: Nine steps gave a reason the standards do not state, for example that complications can become serious within hours, that a plan catches near misses, and that regular practice keeps skills. The standards give requirements without reasons and no evidence for their numbers. (Second accuracy pass; both standards documents, PDFs from ebmt.org read as raw text) · Each reason was reworded to what the standard itself asks for.
- 2026-11-17 · Accreditation standards for stem cell transplant and CAR T-cell programmes (FACT-JACIE): which standards exist: The page said there is one international standard and no other. FACT also publishes a first edition of standards for immune effector cells in community clinics, and says the joint immune effector cell standards are for programmes that do not transplant. (Second accuracy pass; FACT standards page at factglobal.org read as raw text) · The statement was reworded, the FACT page was added as a source, and a gap says the community standards were not read.
- 2026-11-17 · FDG PET-CT scan preparation: which guidelines exist: The page said no newer adult oncology FDG guideline from EANM or SNMMI had been found. Two exist: a joint immunotherapy guideline (2022), which says to fast for at least 6 hours, and a joint breast cancer guideline (2024). The United States guideline (2006) also prints different figures and was not on the page. (Second accuracy pass; Europe PMC title search and the full text of each guideline read as raw text) · The statement now says EANM 2015 is the latest general procedure guideline, and the three other guidelines are added as sources and variants with their own fasting, glucose and uptake figures kept apart.
- 2026-11-17 · FDG PET-CT scan preparation: high blood sugar: The reason given for the 11 mmol/L limit was that very high sugar makes the scan less accurate. For a scan that is indicated in someone with unstable or poorly controlled diabetes, the EANM guideline says fasting hyperglycaemia does not hamper the clinical value of FDG PET. (Second accuracy pass; EANM 2015 full text from Europe PMC) · The reason now says what EANM says, and the supporting quote was replaced with the sentence that states it.
- 2026-11-17 · FDG PET-CT scan preparation: diabetes tablets: The page said tablets do not interfere with the scan in the way insulin does. EANM does not say that, and notes that bowel uptake of the tracer may be increased in people taking metformin. EANM's 48 hour metformin rule with contrast was also left out. (Second accuracy pass; EANM 2015 full text from Europe PMC) · Reason reworded to the source, with the 48 hours and the word preferably for the late morning slot.
- 2026-11-17 · FDG PET-CT scan preparation: insulin timing: The step gave only the 4 hour gap after rapid-acting insulin. EANM also gives 6 hours after short-acting insulin and does not recommend the injection on the same day after intermediate or long-acting insulin. (Second accuracy pass; EANM 2015 full text from Europe PMC) · Both rules added to the step and the summary.
- 2026-11-17 · FDG PET-CT scan preparation: reasons given for each step: Several reasons said more than the guideline: that the tracer is pulled away from the cancer, that water lowers the bladder dose, that exercise can hide or mimic disease, that high sugar competes with the tracer and low sugar is unsafe, that chewing works muscles, and that uptake keeps rising with time. The patient summary said to avoid hard exercise the day before, where EANM says at least 6 hours and preferably 24. (Second accuracy pass; EANM 2015 full text from Europe PMC) · Each reason reworded to what the guideline states, with its lower glucose limit of 4 mmol/L added.
- 2026-11-17 · Financial toxicity: screening, the cost conversation and referral for money, benefits and work support: two-stage screening: The page told teams to screen in two stages as a settled instruction. ESMO calls a sequential strategy a reasonable approach that requires further validation and says it might be used. (Second accuracy pass; ESMO 2024 consensus statements, full text read as Europe PMC XML) · The step now says to consider two-stage screening and states that ESMO says it still needs validation. The same caveat was added to the best-supported summary.
- 2026-11-17 · Financial toxicity: screening, the cost conversation and referral for money, benefits and work support: work and sick pay: The page gave asking about work and sick pay and supporting paid leave as a step for the clinic, and said lost earnings are the largest money problem in many countries. ESMO addresses its statement on return to work and workplace adjustments to governments and employers, and does not say lost earnings are the largest problem. (Second accuracy pass; ESMO 2024 consensus statements, full text read as Europe PMC XML) · The step now asks about work and points to leave and adjustments where they exist, says who ESMO's statement is addressed to, and gives ESMO's own reason. The gap is named.
- 2026-11-17 · Financial toxicity: screening, the cost conversation and referral for money, benefits and work support: reasons given for steps: Several reasons were not in any source read: that help is easier before debts build up, that one question is quick enough to ask everyone, that an unaffordable plan does not work, and that every extra test costs travel and time off work. The cost conversation step also added costs the person meets themselves, which the ASCO abstract does not mention. (Second accuracy pass; ESMO 2024 full text, ASCO 2009 abstract and the NCI summary, read as raw text) · Each reason now says what the source says, with ESMO's caveat that the evidence on timing is not consistent and the NCI figures on cost discussions named.
- 2026-11-17 · Financial toxicity: screening, the cost conversation and referral for money, benefits and work support: what drives costs in each system: The summary for patients said heating costs arise in every health system, the UK section rested on treatment being free at the point of use, and the US section said costs fall above all on oral cancer medicines. The sources read report heating only for the UK, do not state the free at the point of use premise, and say oral medicines often carry high coinsurance without ranking them first. (Second accuracy pass; UK scoping review (Europe PMC full text), NCI summary (cancer.gov HTML) and ESMO 2024) · Reworded to lost wages and travel costs, to the two reasons the UK review gives, and to high coinsurance for many oral cancer medicines.
- 2026-11-17 · Financial toxicity: screening, the cost conversation and referral for money, benefits and work support: when to ask again: The trigger listed later prompts for asking again, such as a missed appointment or an unfilled prescription, as if a source set them. No source read lists them. (Second accuracy pass; all cited sources re-read as raw text) · The trigger now says ESMO names diagnosis or the start of treatment and that the later prompts are common sense with no source.
- 2026-11-17 · Geriatric assessment before systemic cancer therapy in older adults: when the assessment is done: The page said the assessment is done before treatment and applied to people being considered for it. The recommendations say patients receiving systemic therapy and fix no time point; the two trials behind them enrolled people receiving a new regimen. (Second accuracy pass; ASCO 2023 guideline update, author manuscript in PubMed Central, read as raw XML through the NCBI E-utilities API) · Timing wording removed from the first step, the population and the trigger, and the missing time point named in the gaps.
- 2026-11-17 · Geriatric assessment before systemic cancer therapy in older adults: G8 screen step: The step told teams to use the G8 where a full assessment cannot be done for everyone, and gave a reason. The guideline only lists the G8 in a table, with a score of 0 to 14 leading to a full assessment. (Second accuracy pass; ASCO 2023 guideline update, author manuscript in PubMed Central, read as raw XML through the NCBI E-utilities API) · Step reworded to apply if a screen is used, and described as a table entry with no grade.
- 2026-11-17 · Geriatric assessment before systemic cancer therapy in older adults: CARG toxicity tool step: The step ordered a toxicity estimate with the CARG tool. The 2023 update lists the tool in a table with what each risk band leads to and has no graded recommendation to use it. (Second accuracy pass; ASCO 2023 guideline update, author manuscript in PubMed Central, read as raw XML through the NCBI E-utilities API) · Step reworded to apply if the tool is used, with the table's own description of its 11 items.
- 2026-11-17 · Geriatric assessment before systemic cancer therapy in older adults: life expectancy step: The quotation had spaces added inside the brackets round the web address, the grade left out the evidence quality, and the reason given was not in the source. (Second accuracy pass; ASCO 2023 guideline update, author manuscript in PubMed Central, read as raw XML through the NCBI E-utilities API) · Quotation replaced with the source's exact text, the grade given in full, and the reason reworded to the source's point about non-cancer life expectancy.
- 2026-11-17 · Geriatric assessment before systemic cancer therapy in older adults: sharing results: The summary said clinicians should share results with patients and caregivers. The source says the panel suggests the information be provided to them. (Second accuracy pass; ASCO 2023 guideline update, author manuscript in PubMed Central, read as raw XML through the NCBI E-utilities API) · Reworded to suggests, and the two recommendations carried over unchanged from 2018 are labelled as such.
- 2026-11-17 · Geriatric assessment before systemic cancer therapy in older adults: scope: The page said assessment before surgery or radiotherapy alone is outside the guideline's scope. The text read prints no such exclusion. (Second accuracy pass; ASCO 2023 guideline update, author manuscript in PubMed Central, read as raw XML through the NCBI E-utilities API) · Reworded to say the update makes no recommendation about it.
- 2026-11-17 · Geriatric assessment before systemic cancer therapy in older adults: grading scheme: The scheme was described as evidence quality high, moderate, low or insufficient. The guideline's appendix defines high, moderate, low and very low, and strength as strong or weak. (Second accuracy pass; ASCO 2023 guideline update, author manuscript in PubMed Central, read as raw XML through the NCBI E-utilities API) · Grading scheme rewritten from the appendix table.
- 2026-11-17 · Geriatric assessment before systemic cancer therapy in older adults: reasons and best in class: Three reasons (problems often missed, a screen finds those most in need, the point is what is done with it) and the claim that this is the most recent guideline were not in the source. A gap said NICE has no such guidance without a search. (Second accuracy pass; ASCO 2023 guideline update, author manuscript in PubMed Central, read as raw XML through the NCBI E-utilities API) · Reasons reworded to the guideline's own statements, the claim limited to what was checked, and the gap changed to say NICE was not searched.
- 2026-11-17 · Hand hygiene in cancer care: teaching patients and carers: The step quoted a NICE recommendation that sits in the section on long-term urinary catheters and presented it as general advice to teach hand cleaning before discharge. The general NICE recommendation (1.1.1.3) asks that patients and carers are educated about the benefits, techniques and timing of hand decontamination and their role in maintaining staff standards, and does not mention discharge. The reason given, that most cancer treatment is outpatient, is in no source. (Second accuracy pass; NICE CG139 recommendations page, read as raw text) · The step, its quote, its reason and the summary now rest on recommendation 1.1.1.3.
- 2026-11-17 · Hand hygiene in cancer care: soap and water for diarrhoea: The step put infectious diarrhoea under the WHO grade IB recommendation and said alcohol does not kill spores and washing removes them. WHO's recommendation covers exposure to spore-forming germs that is strongly suspected or proven, says antiseptics are not reliably sporicidal and that washing may help remove spores. Other germs that cause diarrhoeal illness are named by NICE, not WHO. (Second accuracy pass; WHO 2009 guideline, full text of the PDF, and NICE CG139) · The step now says strongly suspected or proven, credits the diarrhoea wording to NICE, and words the reason as WHO does.
- 2026-11-17 · Hand hygiene in cancer care: visitors: The step told every unit to ask every visitor to clean their hands, with no grade. The transplant guideline says centre policies should encourage visitors to perform hand hygiene before and after each visit, graded BIII, and it is written for transplant centres. (Second accuracy pass; CDC, IDSA and ASBMT guideline 2009, full text from Europe PMC) · The step now quotes the graded sentence, gives BIII and names the setting.
- 2026-11-17 · Hand hygiene in cancer care: short summary for patients: The summary called clean hands the single most important way to stop infection and told patients to ask staff to clean their hands. Neither wording is in a source read. The transplant guideline calls hand hygiene the mainstay of infection prevention in hospital, NICE asks that patients are educated about their role in maintaining staff standards, and WHO says the effect of patients reminding staff remains to be assessed. (Second accuracy pass; WHO 2009, NICE CG139 and the 2009 transplant guideline, read as raw text) · The summary and the variant on patients reminding staff were reworded to the sources, with WHO's caveats.
- 2026-11-17 · Hand hygiene in cancer care: reasons given for each step: Six reasons said more than the sources do: that most line infections come from the patient's skin germs, that bed rails and pumps carry germs (credited to NICE), that germs survive on missed fingertips, that jewellery and sleeves harbour germs, that visitors carry germs harmless to themselves, and that reminders change behaviour. (Second accuracy pass; WHO 2009 guideline, NICE CG139 and the 2009 transplant guideline, read as raw text) · Each reason was reworded to what its source says, and the step on the patient's surroundings now rests on WHO's graded indication.
- 2026-11-17 · Hand hygiene in cancer care: endorsement and comparison with NICE: The page said the Dutch guideline was endorsed by the Children's Oncology Group without a source, and that NICE describes five circumstances equivalent to WHO's five moments. The endorsement is real (March 2025) and is stated in the Children's Oncology Group document, not the paper. NICE lists five circumstances of its own, and says soap and water must be used where WHO calls it the preferred means. (Second accuracy pass; Children's Oncology Group endorsement document (version date 28 April 2025), NICE CG139 and WHO 2009, read as raw text) · The endorsement document was added as a source and the comparison reworded.
- 2026-11-17 · Immobilisation mask for head and neck and brain radiotherapy, including mask anxiety: music and relaxation step: The step told staff to offer music and guided relaxation to people with mask anxiety, alongside explanation and breaks. The single trial behind it enrolled 216 people having head and neck radiotherapy who were not selected for anxiety, and its abstract says nothing about explanation or breaks. (Second accuracy pass; Supportive Care in Cancer 2025 trial, abstract read raw from Europe PMC) · Reworded to consider offering music with guided relaxation in the first sessions, with the trial population stated and the note that no guideline recommends it yet.
- 2026-11-17 · Immobilisation mask for head and neck and brain radiotherapy, including mask anxiety: how common mask anxiety is: The summary said 26% of 100 consecutive patients had mask anxiety. The study sampled 100 patients in three one-week recruitment windows during treatment and does not call them consecutive. (Second accuracy pass; Nixon 2018, full text read raw from Europe PMC) · Now says a cohort of 100 patients, and the gaps explain that the 26% and 27.3% figures measure different things.
- 2026-11-17 · Immobilisation mask for head and neck and brain radiotherapy, including mask anxiety: open-face masks and the ESTRO guideline: The summary and the best-in-class statement said that open-face masks depart from ESTRO advice against cutting the mask. The guideline says only that cutting out a mask should be avoided except for bite blocks or respiratory devices, and no source read says this conflicts with purpose-made open-face masks. (Second accuracy pass; ESTRO ACROP 2017 and the OPEN methods paper, full texts read raw from Europe PMC) · Now says the ESTRO guideline does not discuss open-face masks and quotes its one statement on cutting out.
- 2026-11-17 · Immobilisation mask for head and neck and brain radiotherapy, including mask anxiety: screening: The page said both Australian studies recommend routine screening, that staff do not always notice anxiety, and that the four-item screen is validated. The 2018 study says screening should be considered, neither study says staff miss it, and the screen was developed and tested in the same patients at one centre, with its authors calling external validation essential. (Second accuracy pass; Nixon 2018, Nixon 2019 and the R-MAQ paper, full texts read raw from Europe PMC) · Reworded each statement to what the studies say and added the limitation to the gaps.
- 2026-11-17 · Immobilisation mask for head and neck and brain radiotherapy, including mask anxiety: reasons and scope: Several reasons were not in the guideline: that metal disturbs the scan and dose, that the breathing should be slow, that a mask fitting at the scan will fit at treatment. The population also said brain masks lack the shoulder section and that children usually need play preparation or anaesthesia, which no source read states. (Second accuracy pass; ESTRO ACROP 2017, full text read raw from Europe PMC) · Reasons reworded to the guideline's own, and the two unsourced population statements removed.
- 2026-11-17 · Image-guided verification of daily radiotherapy set-up: summary and best in class: The page said the UK Radiotherapy Board, the IAEA and the AAPM all agree that automatic matches are checked by eye, that gross errors are investigated before treating and that the oncologist prescribes the imaging. Only On target 2 states the first two and the IAEA the third; the AAPM guideline covers the imaging systems and staff duties. (Second accuracy pass; On target 2 and IAEA Human Health Reports No. 16 read in full from their PDFs, AAPM MPPG 2.b from Europe PMC) · Each statement now names the body that makes it.
- 2026-11-17 · Image-guided verification of daily radiotherapy set-up: choice of scan or flat X-ray: The step told staff to use a three-dimensional scan where available. On target 2 says such scans should be considered where possible, and that flat X-ray imaging can still be justified in a limited number of cases. (Second accuracy pass; On target 2 read in full from the PDF) · The step now says consider, with the source's own qualification.
- 2026-11-17 · Image-guided verification of daily radiotherapy set-up: gross errors: The step ordered a fresh set-up or a repeat image and gave reasons the source does not give. On target 2 says a fresh set-up or reimaging may be required, that the cause is always investigated before the decision to treat, and lists the possible causes. (Second accuracy pass; On target 2 sections 6.1.1 and 6.4.2 read from the PDF) · The step now says may be needed and the reason gives the source's list of causes.
- 2026-11-17 · Image-guided verification of daily radiotherapy set-up: surface guidance: The step allowed surface guidance to replace internal imaging where a site protocol says it may. On target 2 gives no such exception: it says surface-based approaches are not a replacement for other forms of image guidance. (Second accuracy pass; On target 2 section 4.5 read from the PDF) · The exception was removed.
- 2026-11-17 · Image-guided verification of daily radiotherapy set-up: online and offline imaging: The page said On target 2 notes a shift to daily online imaging. Its words are a shift towards more online verification, and its appendix says offline protocols are now much less commonly used. (Second accuracy pass; On target 2 section 5.2 and Appendix 14.1 read from the PDF) · Reworded, and the appendix statement added.
- 2026-11-17 · Image-guided verification of daily radiotherapy set-up: reasons given for steps: Seven reasons were the writer's own and not in the cited source, among them that a second check matters because one wrong match is a large share of a short treatment, and that imaging dose under 5% is small beside the treatment. (Second accuracy pass; On target 2, the IAEA report, MPPG 2.b and the TG-180 abstract re-read as raw text) · Each reason now says what the source says.
- 2026-11-17 · Image-guided verification of daily radiotherapy set-up: UK dose reference levels: The page said it had not been checked whether national dose reference levels for cone-beam CT verification had been published. They have: the UK Health Security Agency lists values adopted in 2025 for adults, from 3.5 mGy for brain to 21 mGy for prostate and gynaecological scans, as a weighted cone beam dose index. (Second accuracy pass; GOV.UK national diagnostic reference levels page read raw) · Source added, and the gap and the summary updated.
- 2026-11-17 · Image-guided verification of daily radiotherapy set-up: who reviews images: The page implied the AAPM guideline follows the 2025 ACR-ARS practice parameter. It cites the 2009 ACR-ASTRO edition, and the IAEA duty of the radiation oncologist to review and approve images was missing. (Second accuracy pass; MPPG 2.b from Europe PMC and the IAEA report section 5.1.3) · The edition is named and the IAEA duty added.
- 2026-11-17 · Linear accelerator quality assurance: what happens when a check fails: The page said that if a test fails the machine is not used until it is fixed and checked again. AAPM MPPG 8.b says the qualified medical physicist must decide whether the machine is removed from service or limited to certain treatments, and that some failures are not clinically significant for some treatments. (Second accuracy pass; AAPM MPPG 8.b (2023), full text read raw from Europe PMC) · Reworded the plain summary and the one-line summary, and added a step quoting the guideline's sentence.
- 2026-11-17 · Linear accelerator quality assurance: one-line summary: The page said these checks are why a machine fault very rarely reaches a patient. No source read says this; MPPG 8.b says only that hardware and software risks are lower than risks from human processes. (Second accuracy pass; AAPM MPPG 8.b (2023), full text read raw from Europe PMC) · Removed the sentence.
- 2026-11-17 · Linear accelerator quality assurance: grade of the log file step: The step saying log files alone cannot replace a measured leaf test was graded "must". The guideline's sentence carries neither "must" nor "should". (Second accuracy pass; AAPM MPPG 8.b (2023), full text read raw from Europe PMC) · The grade now says it is a statement in the guideline's text, and the "must" in the grades list covers only the weekly leaf test.
- 2026-11-17 · Linear accelerator quality assurance: MPPG 8.a variant: The page said MPPG 8.a dropped tests that scored low in its risk analysis. Its full text says the committee in some cases included a lower ranking test or excluded a high ranking one. (Second accuracy pass; AAPM MPPG 8.a (2017), full text read raw from Europe PMC for the first time) · Rewrote the variant from the full text and added how its limits differ from 8.b: output within 3% daily and 2% monthly as "should", leaf positions monthly to 1 mm, against 8.b's "must" and weekly 0.5 mm.
- 2026-11-17 · Linear accelerator quality assurance: best in class: The page said MPPG 8.b updates TG-142 and TG-198. It is the 5-year update of MPPG 8.a only; it reviewed the tests in the two task group reports without replacing them. It also said most UK centres meet or exceed IPEM Report 81, where the survey says centres did so for the majority of tests. (Second accuracy pass; AAPM MPPG 8.b (2023) and the 2025 UK survey, full texts read raw from Europe PMC) · Reworded both statements, and said that MPPG 8.b is a United States guideline.
- 2026-11-17 · Linear accelerator quality assurance: technical bulletins step: The step said technical bulletins and service reports should be read promptly. MPPG 8.b says bulletins must be reviewed immediately and maintenance reports promptly. (Second accuracy pass; AAPM MPPG 8.b (2023), full text read raw from Europe PMC) · The step now gives each its own word.
- 2026-11-17 · Linear accelerator quality assurance: weekly, monthly and annual steps: The weekly leaf test left out that it may be done in arc mode instead of at several gantry angles. The monthly output step added a review of daily results that the guideline does not ask for under output. The annual isocentre step gave a Winston-Lutz test as the only method, where the guideline allows star shots within a 2 mm circle if that test is not feasible. The annual calibration was said to need a calibrated chamber, where the guideline says equipment calibrated by an accredited secondary standards laboratory. (Second accuracy pass; AAPM MPPG 8.b (2023), full text and Tables 4 to 8 read raw from Europe PMC) · Each step and the summary reworded to the guideline's terms.
- 2026-11-17 · Linear accelerator quality assurance: reasons given for the steps: Ten of the reasons beside the steps gave explanations the guideline does not state, such as that log files record what the machine thinks it did, or that the annual calibration ties the machine to the national standard of dose. (Second accuracy pass; AAPM MPPG 8.b (2023), full text read raw from Europe PMC) · Each reason reworded to what the guideline says.
- 2026-11-17 · Linear accelerator quality assurance: other variants and gaps: The TG-142 variant described separate tables for non-IMRT, IMRT and stereotactic machines, which its abstract does not say. The manufacturer tools variant said they are used for some daily and monthly checks, which the cited abstract does not say, and that MPPG 8.b allows them where the physicist understands them, where it requires independent verification first. A gap said there is no trial evidence for any test frequency and that physicists disagreed about common delivery failures; the guideline reports 184 responses on 11 IMRT delivery failures and says its tests were chosen by consensus. (Second accuracy pass; abstracts of TG-142 and the IPEM topical report from Europe PMC, the 2025 UK survey and MPPG 8.b full texts) · Each reworded to its source; the UK survey's figure that 40% of centres do the daily output check with a manufacturer-integrated device replaces the unsupported sentence.
- 2026-11-17 · Long-term follow-up after childhood, teenage and young adult cancer: when follow-up starts: The page said follow-up by treatment received begins "about two years" after treatment, as if every body agreed. The Children's Oncology Group says its guidelines are for use from two or more years after treatment; PanCareSurFup says long-term follow-up should start no later than 5 years after treatment or 5 years from diagnosis; single IGHG checks start sooner. (Second accuracy pass; COG Version 6.0 PDF, PanCareSurFup accepted manuscript and IGHG documents, read as raw text) · Each body's starting point is now given separately in the summary, population, trigger, the step and the comparison.
- 2026-11-17 · Long-term follow-up after childhood, teenage and young adult cancer: heart scans: The step ordered heart scans every 5 years at moderate risk. IGHG recommends scans at high risk but only calls them reasonable at moderate risk (low-quality evidence, moderate recommendation). (Second accuracy pass; IGHG cardiomyopathy recommendations 2022, PDF read as raw text) · The step now says to consider scans every 5 years at moderate risk, and that high-risk scans start no later than 2 years after treatment.
- 2026-11-17 · Long-term follow-up after childhood, teenage and young adult cancer: hearing tests: The step said to test "from the end of treatment" and to refer "any loss". IGHG says to begin no later than the end of treatment and to refer for symptoms of hearing loss or a loss of more than 15 dB at 1000 to 8000 Hz. The reasons given were not in the document read. (Second accuracy pass; IGHG ototoxicity recommendations 2019, PDF read as raw text) · Wording and referral threshold now match IGHG, and the reason is limited to what IGHG and COG state.
- 2026-11-17 · Long-term follow-up after childhood, teenage and young adult cancer: bowel screening after radiation: The step said "lower spine" and "begin early". COG lists radiation to the abdomen, pelvis, spine (lumbar, sacral or whole) and total body irradiation, and starts screening 5 years after radiation or at age 30, whichever is later. A sentence said COG starts earlier than it otherwise would, which COG does not say. (Second accuracy pass; COG Version 6.0 section 86, PDF read as raw text) · Fields and start time are now stated as COG prints them, and the unsupported comparison is removed.
- 2026-11-17 · Long-term follow-up after childhood, teenage and young adult cancer: breast screening comparison: The comparison gave the 10 Gy threshold as IGHG's without saying that COG Version 6.0 has no minimum dose and lists radiation to the chest or axilla or total body irradiation. It also gave the 2025 doxorubicin indication without its scope (no chest radiation, from age 30, at the frequency of local moderate-risk guidelines). (Second accuracy pass; COG Version 6.0 section 73 and IGHG breast recommendations 2020 and 2025, PDFs read as raw text) · Both positions are now stated with their scope.
- 2026-11-17 · Long-term follow-up after childhood, teenage and young adult cancer: yearly psychosocial questions: The step listed mood, relationships and money under a single score of 1. COG's yearly history covers school or work progress and social withdrawal (section 1, score 1) and depression, anxiety, post-traumatic stress and suicidal ideation (section 2, score 2A). (Second accuracy pass; COG Version 6.0 sections 1 and 2, PDF read as raw text) · The items are listed as COG lists them and both scores are given.
- 2026-11-17 · Long-term follow-up after childhood, teenage and young adult cancer: reviewing the plan with a clinician: A step told clinics to go through each plan with a late effects clinician, resting on a COG sentence that is about something else: COG encourages survivors who choose to read the professional guideline to do so with a knowledgeable professional. (Second accuracy pass; COG Version 6.0 preface, PDF read as raw text) · The step now says what COG says.
- 2026-11-17 · Long-term follow-up after childhood, teenage and young adult cancer: which guideline is stronger: The page called IGHG "the more rigorous" and "the stronger" source and said COG Version 6.0 "adopted the IGHG 2022 heart schedule". No source read ranks them; COG says it aimed to align better with IGHG, and its review says it dropped low-risk heart scans in concordance with IGHG. It also said COG names the guideline's length as a barrier, where COG names clinicians' time and the effort of finding the relevant recommendations. (Second accuracy pass; COG Version 6.0 PDF and the COG review (author manuscript in PubMed Central), read as raw text) · The ranking is removed and the sentences now report what COG and its review state.
- 2026-11-17 · Long-term follow-up after childhood, teenage and young adult cancer: reasons given for steps: Fifteen reasons were plausible but stated by no cited source, for example that heart damage appears "silently years later", that a semen test at the wrong time "can worry without helping", and that knowing the risk of a second cancer "shortens the time to diagnosis". (Second accuracy pass; all cited sources re-read as raw text) · Each reason now reports what the cited guideline says, or names the source it comes from.
- 2026-11-17 · Long-term follow-up after childhood, teenage and young adult cancer: thyroid checks: The page said COG examines the thyroid yearly "after neck radiation". COG section 67 covers radiation to the head or brain, neck, spine (cervical or whole) and total body irradiation. (Second accuracy pass; COG Version 6.0 section 67, PDF read as raw text) · The fields are now named.
- 2026-11-17 · Medicines reconciliation at the start of anticancer treatment and at each move between care settings: the trial enrolment variant: The page said pharmacist reconciliation found 158 prohibited interactions and uncovered 35% of the medicines patients were taking. The paper says 158 treatments were prohibited by the trial protocol, and that 35.2% of the treatments patients were taking had not otherwise been known to the clinical team. (Second accuracy pass; Malifarge 2024, full text read as Europe PMC XML) · Both figures reworded to what the paper counts, and the source is now marked as read in full.
- 2026-11-17 · Medicines reconciliation at the start of anticancer treatment and at each move between care settings: the first-cycle step: The page told staff to take a full drug history before the first cycle and called the check the pharmacist's. The BOPA standard asks for a check of the patient's concurrent medicines for interactions, contraindications and duplication at pharmacy verification, which trained pharmacy technicians may also carry out at set levels. (Second accuracy pass; BOPA standards version 5.0, PDF read with pdftotext) · Step and summary reworded to the standard's own check.
- 2026-11-17 · Medicines reconciliation at the start of anticancer treatment and at each move between care settings: repeating reconciliation and involving families: The page ordered reconciliation to be repeated at ward transfer and discharge and families to be involved. NICE says to recognise that it may need to be carried out more than once, and to involve family members or carers where appropriate. (Second accuracy pass; NICE NG5 recommendations 1.3.2 and 1.3.6, page read as raw text) · Both steps reworded with NICE's hedges.
- 2026-11-17 · Medicines reconciliation at the start of anticancer treatment and at each move between care settings: the restart plan step: A step on restarting medicines stopped during acute illness was presented as general, with the reason that paused medicines are often never restarted. It is a practice point from a kidney disease guideline, which says failure to restart may lead to unintentional harm. (Second accuracy pass; KDIGO 2024 practice points 4.3.2 and 4.3.3, PDF read with pdftotext) · Scope stated in the step, the population and the gaps, and the reason reworded.
- 2026-11-17 · Medicines reconciliation at the start of anticancer treatment and at each move between care settings: the reasons given for steps: Several reasons were not in the cited sources, among them that medicines are most often lost when people move between settings, and an example of a new anticoagulant at discharge. NICE's introduction says there is a greater risk of unintended changes at transfer, and that between 30% and 70% of patients have an error or unintentional change. (Second accuracy pass; NICE NG5 introduction, ASCO-ONS 2024 and the WHO High 5s protocol, read as raw text) · Reasons reworded to what each source states.
- 2026-11-17 · Medicines reconciliation at the start of anticancer treatment and at each move between care settings: the Canadian variant: The page said one centre reached 4.7% full reconciliations within 30 days of starting treatment. The paper says up to 4.7% had their medicines reconciled, measured within 30 days either side of starting systemic therapy. (Second accuracy pass; Powis 2023, full text read as Europe PMC XML) · Figure and window reworded, and the source is now marked as read in full.
- 2026-11-17 · Medicines reconciliation at the start of anticancer treatment and at each move between care settings: advice to patients: The plain-language line told patients to bring all their medicines or a list to every stay, with no source. The WHO protocol asks that patients keep an up-to-date list or bring their medicines to hospital and clinics, and speak up if they believe a mistake has been made. (Second accuracy pass; WHO High 5s medication reconciliation protocol, version 3 of September 2014, PDF read with pdftotext) · Line reworded to the WHO text, and the protocol added as a source and a variant.
- 2026-11-17 · Medicines reconciliation at the start of anticancer treatment and at each move between care settings: the ASCO-ONS quotation: The quoted standard was printed with a straight apostrophe where the source has a typographic one, and the page said NICE NG5 had not been updated when its update page lists link changes in 2019 and 2021. (Second accuracy pass; ASCO-ONS 2024 in the Oncology Nursing Forum PDF, and the NICE update information page) · Quotation made character for character and the gap reworded.
- 2026-11-17 · Nutrition screening and support during cancer treatment: energy and protein: The energy figure of 25 to 30 kcal/kg/day was written as a target and carried the grade of the protein recommendation (moderate evidence, strong consensus). ESPEN recommends assuming energy expenditure to be similar to healthy people and generally 25 to 30 kcal/kg/day, graded strong on low evidence, consensus. (Second accuracy pass; ESPEN practical guideline 2021, full text PDF read as raw text) · Step, summary and grades now state the energy recommendation as an assumption with its own grade, beside the protein grade.
- 2026-11-17 · Nutrition screening and support during cancer treatment: who it covers: The page said the ESPEN guideline covers paediatric nutrition. The guideline is for adult cancer patients and has no recommendation for children. (Second accuracy pass; ESPEN practical guideline 2021, full text PDF read as raw text) · Population and gaps now say children are not covered by either source.
- 2026-11-17 · Nutrition screening and support during cancer treatment: advanced cancer and exercise grades: The advanced cancer step carried only the grade of the recommendation on dying patients (strong consensus), and resistance exercise sat under the strong, high evidence grade for physical activity. ESPEN grades the advanced cancer recommendation strong, low evidence, consensus, and only suggests resistance exercise (weak, low evidence). (Second accuracy pass; ESPEN practical guideline 2021, full text PDF read as raw text) · Both steps now give each recommendation its own grade, and resistance exercise is written as a weak suggestion.
- 2026-11-17 · Nutrition screening and support during cancer treatment: screening tool and NICE scope: The screening step called MUST a validated tool and read as a universal rule. NICE CG32 says MUST may, for example, be used, and it is a UK guideline for all adults that says support should be considered, not started, at its BMI and weight loss criteria. (Second accuracy pass; NICE CG32 recommendations page read as raw HTML) · Step, summary and best in class reworded; inpatient screening points added; NICE scoped to the UK.
- 2026-11-17 · Nutrition screening and support during cancer treatment: omega-3 and radiotherapy: The omega-3 suggestion was given for anyone on chemotherapy at risk of weight loss, and radiotherapy support as counselling and supplements. ESPEN limits the omega-3 suggestion to advanced cancer during chemotherapy, and says counselling and/or supplements during radiotherapy. (Second accuracy pass; ESPEN practical guideline 2021, full text PDF read as raw text) · Both reworded to the source.
- 2026-11-17 · Nutrition screening and support during cancer treatment: reasons given for steps: Eight reasons were not in either source, among them a list of symptoms that stop eating, that tube feeding is safer than feeding into a vein, that refeeding shifts are dangerous, and advice to ask for a dietitian. (Second accuracy pass; ESPEN practical guideline 2021 and NICE CG32 read as raw text) · Each reason reworded to what the guideline's recommendation or commentary says.
- 2026-11-17 · Lymphoedema after breast cancer: arm precautions: The page said blood pressure readings on the treated arm were not associated with arm swelling in the Ferguson cohort. The full paper reports that having blood pressure readings was associated on univariate analysis (P = .034) and not on multivariate analysis, and its authors conclude only that these exposures may not be associated. (Second accuracy pass; Ferguson 2016 full text read from PubMed Central) · The summary, best in class statement and precautions variant now give both results and the authors' hedged conclusion.
- 2026-11-17 · Lymphoedema after breast cancer: why early compression: Two steps said a sleeve worn early reduced progression to lasting lymphoedema in a randomised trial. PREVENT gave the sleeve to everyone who crossed the trigger and compared two ways of measuring, not a sleeve against no sleeve. (Second accuracy pass; PREVENT full text read from Europe PMC) · The reasons now say what the trial tested and rest the sleeve on the MASCC and APTA recommendations.
- 2026-11-17 · Lymphoedema after breast cancer: UK guidance missing: The page presented clinic surveillance and preventive sleeves as the best supported approach without NICE. NICE guideline NG101, updated April 2025, asks for information and self-monitoring and does not recommend scheduled clinic measurement, and says compression therapy should not be offered as a risk-reducing measure. (Second accuracy pass; NICE NG101 recommendations 1.14.1 to 1.14.11 read as raw text) · NICE added as a source, a variant, grades and in the best in class statement, with the disagreement stated.
- 2026-11-17 · Lymphoedema after breast cancer: warning signs: The advice to seek treatment quickly for redness or heat, and to tell the team at once, cited a cohort study that gives no such advice. (Second accuracy pass; NHS lymphoedema prevention page and NICE NG101 read as raw text) · The step now rests on NICE 1.14.1 and quotes the NHS wording: contact a GP as soon as possible about symptoms of a possible skin infection. The NHS page's older advice to avoid injections and blood pressure readings in the affected area is stated beside NICE.
- 2026-11-17 · Lymphoedema after breast cancer: surveillance steps: Steps wrote hedged recommendations as orders and added details no source states: measuring both arms, a fixed schedule for at least 3 and up to 5 years cited to a paper that gives only the 3 years, a fitted sleeve, and most early swelling appearing within nine months where the source says more than half. (Second accuracy pass; MASCC 2024, APTA 2020 and the PREVENT timing abstract read from Europe PMC) · Actions reworded to the sources' strength, schedules attributed to APTA and MASCC, and more than half restored.
- 2026-11-17 · Lymphoedema after breast cancer: reasons given for steps: Several reasons were not in the cited sources: that bioimpedance detects fluid before the arm looks swollen, that symptoms often come before measurable change, that a garment controls early swelling for most people, that cellulitis is a proven risk, and that exercise does not worsen lymphoedema. (Second accuracy pass; MASCC 2024, APTA 2017 and 2020 and the 2024 evidence map read in full) · Each reason replaced with what the cited guideline says, including that MASCC treats bioimpedance as an option and not a standard of care.
- 2026-11-17 · Lymphoedema after breast cancer: other cancers: The page listed vulval, cervical, endometrial and penile cancer and leg lymphoedema after groin surgery as covered by the ACS/LANA statement. The abstract names only gynaecological cancers, melanoma and head and neck cancer. (Second accuracy pass; ACS/LANA 2024 abstract from Europe PMC) · Population and gap reworded to the abstract's list.
- 2026-11-17 · Lymphoedema after breast cancer: sources: The APTA 2020 grading scheme was described as running from A to D with Expert Opinion, but that guideline uses A, B, C and Best Practice. The ASBrS entry was labelled Parts 1 and 2 although its DOI is Part 1. Two quotes did not match their sources character for character. (Second accuracy pass; APTA 2020 Table 5 and Europe PMC records) · Grading scheme and label corrected and both quotes replaced with the sources' text.
- 2026-11-17 · Consent for radiotherapy: IR(ME)R information duty: The page called IR(ME)R 2017 UK radiation law and cited the text as first made. The regulations extend to England, Wales and Scotland only, and Schedule 2 was amended from 1 October 2024. The paragraph on information about benefits and risks is unchanged. (Second accuracy pass; legislation.gov.uk, latest revised Schedule 2 and regulation 1, read raw) · Scoped to England, Wales and Scotland, cited the revised version, and said in the gaps that the Northern Ireland regulations were not read.
- 2026-11-17 · Consent for radiotherapy: sending the form before the appointment: The page said the RCR FAQs discuss sending forms in advance and called this RCR practice advice. The FAQ page lists the question and prints no answer to it. (Second accuracy pass; RCR consent forms page, HTML read raw) · The variant now says the question is raised and not answered, and that the RCR gives no recommendation either way.
- 2026-11-17 · Consent for radiotherapy: electronic consent step: The step told departments to check that an e-consent system reproduces the RCR form exactly and transfers the whole signed form, and said such systems have been found to drop details. The RCR asks for a comprehensive local risk assessment and review arrangements, and says only that there is some evidence that certain commercial forms lose critical details on transfer. (Second accuracy pass; RCR position statement dated 30 October 2026, HTML read raw) · Step, reason, summary and variant reworded to the statement's own terms.
- 2026-11-17 · Consent for radiotherapy: reasons given for steps: Four reasons were not in any source: a singer's voice as an example, weeks passing between clinic and treatment, talking to family, and different legal rules for children. One step ordered use of the RCR form although the RCR advises and does not mandate it. (Second accuracy pass; RCR Implementation (April 2021), PDF read with pdftotext) · Reasons reworded to what the guidance says, and the form step now begins where the RCR form is used.
- 2026-11-17 · Consent for radiotherapy: summary of the forms and the law: The summary said the current forms date from 2025 and that the patient must be told of alternatives including no treatment under Montgomery. Not every listed form is a 2025 version, and the judgment speaks of reasonable care and of reasonable alternative or variant treatments. (Second accuracy pass; RCR forms page and the judgment at the National Archives, read raw) · Summary names the one form version read (breast, January 2025 version 4) and words the duty as paragraph 87 does. The judgment is now a source read in full.
- 2026-11-17 · Question prompt lists before key appointments, and a recording or written summary of the consultation: anxiety: The page said neither tool has been shown to lower anxiety. A 2018 systematic review of consultation recordings (Rieger and colleagues) reports statistically significant support for a beneficial effect on anxiety and depression, while the 2008 Cochrane review found no difference in any of ten studies. (Second accuracy pass; Rieger 2018 and Pitkethly 2008 abstracts, Europe PMC, read as raw text) · The summary for patients, the best-supported version and the gaps now say the findings on anxiety are mixed and name each review's position.
- 2026-11-17 · Question prompt lists before key appointments, and a recording or written summary of the consultation: when a patient asks to record: The step told clinicians to agree to a request to record, and said a refusal can lead to secret recording. The British Medical Association encourages doctors to support such requests and says patients whose doctors agree are less likely to record covertly. (Second accuracy pass; BMA page fetched as raw HTML) · The step now says to support the request where possible and gives the BMA's reason as the BMA states it.
- 2026-11-17 · Question prompt lists before key appointments, and a recording or written summary of the consultation: recording other patients, and local policies: The step said not to record other patients, and a variant said hospitals will have local policies. The BMA says other patients must not be recorded without their explicit consent, and that NHS facilities will have local policies. (Second accuracy pass; BMA page fetched as raw HTML) · Both now carry the BMA's wording, and the step says the advice is about the United Kingdom.
- 2026-11-17 · Question prompt lists before key appointments, and a recording or written summary of the consultation: how robust the prompt list results are: The summary said the authors of the 2025 meta-analysis rate the robustness of the pooled results as suboptimal. They say that of two outcomes only, the number of questions asked and consultation length, and call the other results robust. (Second accuracy pass; Han 2025 full text, Europe PMC) · The caveat now names the two outcomes, and is added where the page says consultations were not lengthened.
- 2026-11-17 · Question prompt lists before key appointments, and a recording or written summary of the consultation: what recordings and summaries achieve: The page said a recording or summary helps most people remember and that most use it to tell family. The Cochrane review reports better recall in five of nine studies and that 41.5 to 94.4 per cent used a recording to inform family and friends. (Second accuracy pass; Pitkethly 2008 abstract, Europe PMC) · The figures are now given as the review prints them.
- 2026-11-17 · Question prompt lists before key appointments, and a recording or written summary of the consultation: reasons given for five steps: The reasons for preparing beforehand, bringing a companion, agreeing an agenda, the take-away summary and the letter were statements no source read makes, such as that questions are otherwise left to the last minute. (Second accuracy pass; NICE NG197 recommendations 1.2.3 to 1.2.21 read as raw HTML) · Each reason now says what NICE gives as the purpose, and the examples of who needs extra help are attributed to the BMA, which lists them.
- 2026-11-17 · Question prompt lists before key appointments, and a recording or written summary of the consultation: what NICE recommends: The summary said NICE recommends a printout and encouraging a recording. NICE gives the printout as an example of a resource and recording as one of three possible forms of extra support. The letter step also cited recommendation 1.2.20 alone when the contact details come from 1.2.19. (Second accuracy pass; NICE NG197 read as raw HTML) · Summary and step reworded; both recommendation numbers given.
- 2026-11-17 · Question prompt lists before key appointments, and a recording or written summary of the consultation: quality of life and survival: The page said no review found an effect on quality of life and none measured survival. Only the Cochrane review of recordings and the review in advanced cancer report looking at quality of life, and only the Cochrane review says no study evaluated survival. (Second accuracy pass; abstracts from Europe PMC) · The statements are now scoped to the reviews that make them.
- 2026-11-17 · Question prompt lists before key appointments, and a recording or written summary of the consultation: a shortened quotation: A sentence quoted from the 2019 overview of reviews was cut short inside its quotation marks, dropping the words about patient-held records. (Second accuracy pass; Licqurish 2019 abstract, Europe PMC) · The full sentence is quoted.
- 2026-11-17 · Bladder filling and rectal preparation for prostate radiotherapy: which guidance exists: The page said no guideline covers bladder filling or rectal preparation. The UK Radiotherapy Board's On target 2 (2021) does, without grades: it says bowel-emptying protocols should be in place, recommends a bladder volume of 150 to 300 ml if achievable, written guidance for patients, and a repeat planning scan after voiding if the rectum is more than 4 cm front to back. (Second accuracy pass; On target 2 section 10.10, open PDF read as text) · Added the guidance as a source with four quoted steps, and stated where it and the 2026 Royal Marsden study point different ways on routine rectal preparation.
- 2026-11-17 · Bladder filling and rectal preparation for prostate radiotherapy: the Royal Marsden study: The page said micro-enema preparation did not change movement of the prostate during treatment. The study measured movement during treatment only in 40 patients who used no preparation, and found that a large rectum did not increase it. The page also called the study a retrospective cohort of 295 patients; one dataset was sampled prospectively and the paper does not print 295. (Second accuracy pass; Alexander et al. 2026, full text from Europe PMC) · Corrected the finding, gave the patient counts as printed (255 and 40) and described the design as a single-centre service evaluation and audit, not randomised.
- 2026-11-17 · Bladder filling and rectal preparation for prostate radiotherapy: routine rectal preparation: A step told teams not to prescribe rectal preparation to everyone by default, resting on one centre's study, and the summary for patients said newer studies suggest it is not needed. The study's authors say preparation is not universally required; UK guidance says bowel-emptying protocols should be in place. (Second accuracy pass; Alexander et al. 2026 full text and On target 2 read as text) · Reworded to "consider whether every patient needs it", named the study as one centre's, and put the UK guidance beside it.
- 2026-11-17 · Bladder filling and rectal preparation for prostate radiotherapy: antiflatulent trials: The page said two trials found no reduction in rectal volume, gas or planning rescans. The 78-patient trial measured rescans and rectal diameter, not volume; the 30-patient SPoRT trial measured rectal and gas volume, not rescans. (Second accuracy pass; both abstracts from Europe PMC) · Each trial is now reported with the outcomes it measured.
- 2026-11-17 · Bladder filling and rectal preparation for prostate radiotherapy: steps with instructions no source gives: Steps told staff to make the patient wait or drink more if the bladder differed on the scan, to tell the patient no rectal method is best, and not to ask for large drinks. The sources say only that a scan before each treatment is useful for checking filling, that no method has robust evidence over another, and that 540 ml varied less than 1080 ml in one trial. A bladder scanner variant also said the patient drinks until a target is reached, which the review does not say. (Second accuracy pass; Nasser et al. 2021 full text, McNair 2014 and Mullaney 2014 abstracts) · Removed the unsourced instructions and reworded the steps to what each source states.
- 2026-11-17 · Bladder filling and rectal preparation for prostate radiotherapy: hedges and counts: The page said a full bladder lifts the bowel out of the pelvis, that 300 to 400 ml gave the best reproducibility, that an empty bladder is not recommended when nodes are treated, and that a laxative trial had 17 patients. The sources say a full bladder can potentially move bowel out of the treated region, that 300 to 400 ml potentially gives the best reproducibility, that a full bladder is preferable when nodes are treated, and that the trial had groups of 8 and 9. (Second accuracy pass; Nasser 2021 and Chen 2021 full texts, Weston 2020 abstract) · Restored the sources' hedges and printed counts, and added the meta-analysis's mean differences and waiting times from its full text.
- 2026-11-17 · Patient identification and pre-treatment time-out in radiotherapy: where the law applies: The page said the identity check is the law in the UK. The regulations cited, IR(ME)R 2017, extend to England, Wales and Scotland; Northern Ireland has its own regulations. (Second accuracy pass; IR(ME)R 2017 Schedule 2 on legislation.gov.uk, read as raw text) · The summary, the first step, its grade and the best-practice statement now say England, Wales and Scotland, and the gap list says the Northern Ireland regulations were not read.
- 2026-11-17 · Patient identification and pre-treatment time-out in radiotherapy: current UK guidance: The page said no UK document newer than 2008 covered patient identification in radiotherapy, and that how to identify people who cannot answer was left wholly to local procedure. The RCR guidance on IR(ME)R in radiotherapy, second edition, October 2026, has a chapter on it and gives examples of alternatives. (Second accuracy pass; RCR IR(ME)R guidance second edition, PDF text read with pdftotext) · The guidance was added as a source with four steps quoted from it: three questions matched to the primary source data, a six-point check before the first exposure, alternatives with no exposure if identity cannot be established, and a record of who did the check.
- 2026-11-17 · Patient identification and pre-treatment time-out in radiotherapy: the WHO risk profile: The page said the WHO manual rates incorrect identification as a high-impact risk at every stage and lists a unique patient identifier as a solution, and a step asked for a unique patient number at the machine. The manual has no such row at the decision to treat stage, and lists a unique identifier at the assessment stage only. (Second accuracy pass; WHO Radiotherapy Risk Profile PDF from who.int, read with pdftotext) · The summary now says most stages including treatment delivery, and the step says the unique identifier is listed at assessment only.
- 2026-11-17 · Patient identification and pre-treatment time-out in radiotherapy: grades: Four sentences from the body text of Towards Safer Radiotherapy were graded as its consensus recommendation, and the statement that mistakes are more likely later in treatment was graded as an observation from UK incident data. Only one boxed recommendation covers identification, and the source gives no data or reference for the statement. (Second accuracy pass; Towards Safer Radiotherapy PDF from the Internet Archive capture, read with pdftotext) · The grades now say body text or boxed recommendation, and the statement is marked as carrying no data or reference.
- 2026-11-17 · Patient identification and pre-treatment time-out in radiotherapy: side and site: A step told staff to confirm the side and site at every fraction and cited Towards Safer Radiotherapy, which speaks only of the correct side. (Second accuracy pass; Towards Safer Radiotherapy section 5.10.1, read as raw text) · The step now says side only; site and laterality appear in the new six-point check step, as the RCR guidance words it.
- 2026-11-17 · Patient identification and pre-treatment time-out in radiotherapy: reasons and added details: Several reasons and details were not in any source read: that this is one of the most important safety checks, that nervous or hard of hearing people say yes to another name, that familiarity makes staff skip the check, that the time-out is done as a team, that the photograph is taken at planning, and that the check covers brachytherapy and children by name. (Second accuracy pass; all five readable sources re-read as raw text) · Each was removed or replaced with what the cited source says; the population now follows the scope and examples of the RCR guidance.
- 2026-11-17 · Radiotherapy planning scan (CT simulation): identity and consent step: The step told staff to complete the pregnancy enquiry and cited the ESTRO ACROP head and neck guideline, which asks only for departmental consent and identification procedures and never mentions pregnancy. (Second accuracy pass; ESTRO ACROP head and neck guideline (2017), full text from Europe PMC) · Pregnancy enquiry removed from the step and the best-practice statement; the page now says it is covered in its own linked protocol.
- 2026-11-17 · Radiotherapy planning scan (CT simulation): short summary: The patient summary said the appointment is usually a CT scan with sometimes a cushion, and that a full or empty bladder or breath-hold practice may be needed. None of the sources cited on the page says any of this; the NCI booklet says x-rays or scans, a body mould or a mask. (Second accuracy pass; NCI booklet (April 2021) PDF text and the two ESTRO guidelines, full text) · Summary reworded to what the sources describe, with a pointer to the linked pages for preparation by area and a named gap.
- 2026-11-17 · Radiotherapy planning scan (CT simulation): contrast step: The step said to screen for previous reactions and gave as its reason that reactions are rare. The guideline says to screen for potential anaphylaxis, document it and check creatinine clearance, and gives no frequency. (Second accuracy pass; ESTRO ACROP head and neck guideline (2017), full text from Europe PMC) · Step reworded to the guideline's terms and the unsourced reason removed.
- 2026-11-17 · Radiotherapy planning scan (CT simulation): slice thickness and scan dose step: Recording the scan dose was called a legal requirement for imaging. The guideline says only that documenting it is in line with European Commission directive 97/43 (Euratom), which is a European rule. (Second accuracy pass; ESTRO ACROP head and neck guideline (2017), full text from Europe PMC) · Reason scoped to the directive the guideline names, and a gap added saying current national rules were not read.
- 2026-11-17 · Radiotherapy planning scan (CT simulation): positioning and bolus step: The step put marking of scars and nodes after immobilisation, and another step said a wrong orientation could flip the plan. The guideline says any marking should be done before immobilisation and gives no such reason for the orientation check. (Second accuracy pass; ESTRO ACROP head and neck guideline (2017), full text from Europe PMC) · Order corrected and the unsourced reason replaced with the guideline's own.
- 2026-11-17 · Radiotherapy planning scan (CT simulation): skin marks step: The quote carried a reference number and did not match the source character for character, and the step was graded as a consideration although discussing the options, and using skin marks where there is no surface guidance, are recommendations in the guideline. The trial result was given as no difference where the guideline says no significant difference. (Second accuracy pass; ESTRO-ACROP breast guideline (2023), full text from Europe PMC) · Step now quotes the recommendation verbatim, carries the condition about surface guidance, and names the considerations as considerations.
- 2026-11-17 · Radiotherapy planning scan (CT simulation): virtual-reality explanation: The page said the trial enrolled 196 women and described a staff-led walkthrough of the room, machine and session in a headset or on a visit. The abstract says 196 patients with breast cancer and an immersive virtual-reality explanation, and gives no more detail. (Second accuracy pass; Shin et al. 2023, abstract from Europe PMC (full text is subscription only)) · Wording cut back to the abstract, arm sizes added, and a gap added for what the abstract does not say.
- 2026-11-17 · Radiotherapy planning scan (CT simulation): sources: The page said the AAPM report on CT simulators (2003) could not be read and did not know it had been updated. A 2026 update (Task Group 66U1, report 83.B) exists and is free to read. (Second accuracy pass; AAPM Task Group 66U1 report, full text from PubMed Central, DOI resolved through Crossref) · Report added as a source and the gap rewritten; it covers quality assurance of the scanner, so no step rests on it.
- 2026-11-17 · Risk-based quality management of the radiotherapy process (TG-100 failure mode and effects analysis): how common process errors are: The page said most radiotherapy errors come from the process and not the machines. The TG-100 report says many errors, not most. (Second accuracy pass; AAPM TG-100 (2016), full text read from PubMed Central) · Changed to many in the short summary and in the first step, and the examples now match ones the report gives.
- 2026-11-17 · Risk-based quality management of the radiotherapy process (TG-100 failure mode and effects analysis): which failures count as high risk: The page said studies commonly treat the top quarter of scores and any severity of 8 or more as high risk, and wrote that as the rule to follow. The 2025 review found thresholds varied, with a score above 100 the commonest fixed cut-off and the top 20% the commonest percentage. TG-100 gives the highest 20% to 25% and a severity of 8 as examples only. (Second accuracy pass; Medical Physics 2025 systematic review and AAPM TG-100, both read in full) · The summary now gives the review's figures, the step quotes TG-100 itself and presents its numbers as examples, and the top-quarter rule is attributed to the one 2026 exercise that used it.
- 2026-11-17 · Risk-based quality management of the radiotherapy process (TG-100 failure mode and effects analysis): other safety measures: The page listed incident learning, checklists, training and staffing as the commonest other recommendations. The mapping paper's order is training, staffing, documentation, incident learning, with checklists sixth. (Second accuracy pass; Dunscombe 2012, Europe PMC full text) · The list now follows the paper's order and adds that prospective risk assessment was cited in three of the seven documents.
- 2026-11-17 · Risk-based quality management of the radiotherapy process (TG-100 failure mode and effects analysis): action priority compared with the risk priority number: The page said the action priority table picked out the same critical failures as the older method. In the exercise it selected 7 failure modes and the older method selected those 7 and 23 more, and the authors say only that it appeared less affected by scoring differences. (Second accuracy pass; Rangel et al. 2026, Europe PMC full text) · Reworded to the counts and the authors' own hedged wording.
- 2026-11-17 · Risk-based quality management of the radiotherapy process (TG-100 failure mode and effects analysis): steps written as orders: Three steps were instructions that their cited passages did not give: who should be on the team rested on a count of who took part in published studies, preferring prevention to detection rested on one study's discussion with an added warning against extra checks that the study does not make, and analysing before a new technique treats patients rested on a sentence saying the analysis can be done during development. (Second accuracy pass; AAPM TG-100 full text and Towards Safer Radiotherapy (2008), read from the Internet Archive capture of the RCR PDF) · The team and the choice of tools now quote TG-100's own recommendations with their hedges, and the new-technique step quotes the UK recommendation and is marked as a UK recommendation.
- 2026-11-17 · Risk-based quality management of the radiotherapy process (TG-100 failure mode and effects analysis): reasons given for steps: Eight of the reasons given beside steps were not in the cited sources, for example that harm happens when two small failures line up, or that departments differ in equipment and staffing. (Second accuracy pass; all six sources read in full) · Each reason was rewritten from the cited passage, and five steps were moved to cite TG-100 directly now that its full text can be read.
- 2026-11-17 · Risk-based quality management of the radiotherapy process (TG-100 failure mode and effects analysis): evidence that it reduces harm: The page stated as fact that no trial or study shows risk analysis reduces patient harm, and that patients benefit. No source read says either. TG-100 says the benefit should eventually be validated at the local clinic. (Second accuracy pass; AAPM TG-100, the 2025 review and Dunscombe 2012, read in full) · Now says none of the sources read reports such a study, with what each does say, and the statement that patients benefit was removed.
- 2026-11-17 · Risk-based quality management of the radiotherapy process (TG-100 failure mode and effects analysis): prescriptive quality assurance: The page said TG-100 argues that prescriptive, device-focused quality assurance cannot keep pace with modern processes. TG-100 says growing complexity challenges prescriptive methods and that they will continue to play a role. (Second accuracy pass; AAPM TG-100 full text) · Reworded to both statements.
- 2026-11-17 · Risk-based quality management of the radiotherapy process (TG-100 failure mode and effects analysis): quotations: Four quotations had the publisher's hyphen character replaced by a keyboard hyphen, so they were not character for character. (Second accuracy pass; quotes checked by script against the Europe PMC texts) · The quotations were cut again from the source text by script and keep the publisher's characters.
- 2026-11-17 · Breathing motion in lung and upper abdominal radiotherapy: 4D CT, gating, breath-hold and compression: summary: The page said the US guideline MPPG 9.b requires end-to-end testing with a moving phantom. The guideline says the treatment chain should be assessed that way, and it uses should for a prudent practice, not a requirement. (Second accuracy pass; AAPM-RSS MPPG 9.b (2025), Europe PMC full text) · Reworded to say the guideline lists 4D CT capability among the resources for thoracic and abdominal SBRT services and says the chain should be tested end to end.
- 2026-11-17 · Breathing motion in lung and upper abdominal radiotherapy: 4D CT, gating, breath-hold and compression: motion threshold: The page told teams to use a lower threshold than 5 mm for SBRT and gave only the US figure. AAPM TG-76 says its 5 mm figure may be reduced for SBRT. The UK SABR Consortium prints different figures for stereotactic treatment: more than 10 mm for lung and more than 5 mm for liver or abdominal sites. (Second accuracy pass; AAPM Report 91 (TG-76) PDF read with pdftotext, and the UK SABR Consortium resource version 6.1 read as PDF text) · The step now says the figure may be reduced, and the UK figures are given beside the US one as each report prints them.
- 2026-11-17 · Breathing motion in lung and upper abdominal radiotherapy: 4D CT, gating, breath-hold and compression: measuring motion: The page said TG-76 recommends measuring three-dimensional motion for every patient, usually with 4D CT. TG-76 says motion should be measured when possible, over several breathing cycles if possible. The phrase about 3D mobility is a quotation from an EORTC guideline, and TG-76 does not name 4D CT as the usual method. (Second accuracy pass; AAPM Report 91 (TG-76) PDF read with pdftotext) · Step and summary reworded with the report's own hedges. The 93% use of 4D CT is now attributed to the 2020 TG-324 survey.
- 2026-11-17 · Breathing motion in lung and upper abdominal radiotherapy: 4D CT, gating, breath-hold and compression: verification imaging: The page gave daily then weekly verification imaging as a general rule. TG-76 wrote it for gated or breath-hold treatment guided by an external monitor, and it describes radiographs. (Second accuracy pass; AAPM Report 91 (TG-76) section V.B.5, PDF read with pdftotext) · The step and summary are scoped to gated or breath-hold treatment and say radiographs. UK advice on 4D cone-beam CT is given separately.
- 2026-11-17 · Breathing motion in lung and upper abdominal radiotherapy: 4D CT, gating, breath-hold and compression: reasons given for steps: Five reasons were not in the source: that coaching makes breathing more regular, that a back-up scan avoids a second visit, that problems are easiest to fix at the start, that simpler calculations get the lung dose wrong, and that larger movements are worth controlling to spare healthy organs. (Second accuracy pass; AAPM Report 91 (TG-76) PDF read with pdftotext) · Each reason was replaced with what the report says at that passage.
- 2026-11-17 · Breathing motion in lung and upper abdominal radiotherapy: 4D CT, gating, breath-hold and compression: population and details: The page said the guidance is for adults, that compression uses a plate or belt, that the plate presses gently, and that TG-76 predates routine cone-beam CT and surface guidance. None of these is stated in the sources read. TG-76 gives no age range and describes a plate. (Second accuracy pass; AAPM Report 91 (TG-76) PDF and the TG-324 survey full text) · Reworded to the scope TG-76 states. The page now says the AAPM formed Task Group 324 to update TG-76 and that only its survey was found.
- 2026-11-17 · Scalp cooling during chemotherapy: who it does not suit: The page gave only the 2015 DigniCap list of contraindications (cold disorders and blood cancers treated to cure) and said later FDA documents were not read. The 2017 DigniCap and 2018 Paxman summaries, which widened use to solid tumours, also exclude children, head and neck cancers, brain and spinal cord cancers, squamous cell and small cell lung cancer, skin cancers, tumours likely to spread in transit, planned bone marrow ablation and skull radiotherapy, and Paxman adds scalp metastases and severe liver or kidney disease. (Second accuracy pass; FDA 510(k) summaries K170871 and K173032, PDFs read as raw text) · Both summaries added as sources, a step added with the longer list, and the summary, population and patient line updated.
- 2026-11-17 · Scalp cooling during chemotherapy: taxanes with anthracyclines: The page said results are poorer with taxane and anthracycline combinations. The device instructions go further: cooling has not been shown to be successful with taxanes plus anthracyclines and the device should not be used in these patients. (Second accuracy pass; FDA decision summary DEN150010 and 510(k) summaries K170871 and K173032, read as raw text) · The step and summary now give the instructions' wording, and the gaps note that the trial and registry nevertheless included such regimens.
- 2026-11-17 · Scalp cooling during chemotherapy: why people stop: The page said chills, cap heaviness and headache are the main reasons people stop, and are worst in the first part of the first session. The meta-analysis says the commonest reason for stopping was continuing hair loss (15 per cent), with headache, cold intolerance and discomfort each at about 4 per cent. Macmillan says the cap is hardest in the first 15 to 30 minutes. (Second accuracy pass; Breast Cancer Research and Treatment 2025 abstract from Europe PMC, and the Macmillan page, read raw) · Reason corrected with the source's figures and Macmillan's wording.
- 2026-11-17 · Scalp cooling during chemotherapy: risk of spread to the scalp: The page said the risk is very low, about 1 or 2 people in 1000, a figure the writer worked out. The source prints 0.15 per cent with a range of 0.05 to 0.47, and the device instructions say the long-term risk has not been fully studied and may be higher in the palliative setting with metastatic cancer. (Second accuracy pass; meta-analysis abstract from Europe PMC and FDA summaries, read raw) · The source's figure and the instructions' warnings are now stated.
- 2026-11-17 · Scalp cooling during chemotherapy: cooling time after the infusion: The steps ordered 20 minutes after taxanes and about 90 minutes after anthracyclines. The sources are trials, not guidelines: 20 against 45 minutes was randomised for 3-weekly docetaxel and weekly paclitaxel only, the 90-minute comparison used registry patients, and the one anthracycline trial was in FEC, where 150 minutes did not change head-cover use but moderate to complete hair loss was reported less often (33 against 49 per cent). (Second accuracy pass; trial abstracts from Europe PMC and the paclitaxel trial full text, read raw) · Steps reworded to consider 20 minutes where the unit's protocol allows, scoped to the regimens tested, with the FEC trial's second result added.
- 2026-11-17 · Scalp cooling during chemotherapy: wetting the hair: The page told units to treat wetting as optional and said it added little. The registry study found no convincing evidence either way, with a significant association for one of five regimens (FEC with docetaxel), and its authors gave no instruction. (Second accuracy pass; Supportive Care in Cancer 2023 abstract from Europe PMC, read raw) · Step reworded to follow the unit's instructions for the cap, with the finding stated as the source states it.
- 2026-11-17 · Scalp cooling during chemotherapy: reasons given for steps: Three reasons were not in the sources cited: that the cap must touch the scalp evenly (Macmillan says weaves and braids strain the follicles), that the scalp must be cold when drug levels are highest, and that damp hair conducts cold better. (Second accuracy pass; Macmillan page, paclitaxel trial full text and registry abstract, read raw) · Each reason replaced with what a source says.
- 2026-11-17 · Scalp cooling during chemotherapy: study in Black patients: The page said the abstract's figures were cut off. They were not: the first reading had deleted them. Hair loss and distress both rose (p < .001 and p = .04), only one participant prevented significant hair loss, and most stopped because of more than 50 per cent hair loss. (Second accuracy pass; Oncologist 2021 abstract from the Europe PMC record, read raw) · Figures added and the note corrected.
- 2026-11-17 · Scalp cooling during chemotherapy: Macmillan source date: The Macmillan page was dated 2026. It says it was reviewed on 1 December 2022. (Second accuracy pass; Macmillan page read raw) · Year corrected to 2022.
- 2026-11-17 · Sexual health after cancer: the evidence base: The page said the guideline's evidence was mostly from breast, gynaecological and prostate cancer and from heterosexual couples. The Cancer Care Ontario guideline says its evidence primarily concerned breast or prostate cancer and does not mention heterosexual couples. (Second accuracy pass; Cancer Care Ontario 2017 full text read raw from PubMed Central) · Population and gaps reworded to the guideline's own statement.
- 2026-11-17 · Sexual health after cancer: reasons given for each step: Six steps gave reasons no source states, among them that moisturisers are safe and need no prescription in most countries, that vaginal oestrogen works well, that tablets help many men, and that counselling improved intimacy in the studies. For men the guideline says no study showed a significant improvement in intimacy. (Second accuracy pass; Cancer Care Ontario 2017 full text and ASCO 2018 abstract, read raw) · Each reason replaced with what the panel itself says.
- 2026-11-17 · Sexual health after cancer: vaginal dilators and vaginal oestrogen: The page said to offer dilators and listed low-dose vaginal oestrogen as a fixed next step. The guideline says dilators can be of benefit, and the ASCO abstract says low-dose vaginal oestrogen, lidocaine and dehydroepiandrosterone may be considered in some cases. (Second accuracy pass; Cancer Care Ontario 2017 full text and ASCO 2018 abstract, read raw) · Hedges restored in the steps, the summary and the patient summary. ASCO's wording on lidocaine and dehydroepiandrosterone added with no dose.
- 2026-11-17 · Sexual health after cancer: two quotations on PDE5 inhibitors: Two quotations printed a stray space inside the abbreviation for PDE5 inhibitors, so they were not character for character. (Second accuracy pass; Cancer Care Ontario 2017 full text, quotes checked by script) · Both quotations corrected to the source text.
- 2026-11-17 · Sexual health after cancer: drugs for low desire: The variant named only NCCN and SOGC as including drugs for low desire. The 2025 comparison says all the guidelines except ESO-ESMO include them, which takes in ASCO. (Second accuracy pass; Supportive Care in Cancer 2025 abstract from Europe PMC) · ASCO added to the variant name, with a note that the abstract does not give the strength of each recommendation.
- 2026-11-17 · Sexual health after cancer: counselling as a couple, and who is outside the guideline: The counselling step cited the ASCO abstract for couple counselling, which it does not mention, and the page said children, adolescents and fertility were outside the guideline. The guideline covers adults and leaves out adults previously treated for a childhood cancer. (Second accuracy pass; ASCO 2018 abstract and Cancer Care Ontario 2017 full text, read raw) · Couple counselling attributed to the Cancer Care Ontario findings and the population reworded.
- 2026-11-17 · Skin and nail care during chemotherapy and targeted therapy: how long minocycline is given: The page gave doxycycline or minocycline "for the first 6 weeks" and graded "systemic tetracyclines". Six weeks is the guideline's figure for doxycycline; for minocycline it reports a study over the first 8 weeks, and its table lists tetracycline itself as not recommended. (Second accuracy pass; MASCC 2011 full text read as XML from Europe PMC) · Six weeks now belongs to doxycycline, minocycline's 8-week study is quoted, and the grade names minocycline or doxycycline.
- 2026-11-17 · Skin and nail care during chemotherapy and targeted therapy: what causes sore nail folds: The page said pressure and small injuries trigger nail fold inflammation. The guideline says injury is not believed to cause it but to aggravate it. (Second accuracy pass; MASCC 2011 full text read as XML from Europe PMC) · The reason now says injury is thought to aggravate the inflammation, not cause it.
- 2026-11-17 · Skin and nail care during chemotherapy and targeted therapy: hand and foot care: The page told readers to protect hands and feet from pressure and heat and to wear well-fitting shoes, and said friction and heat worsen hand-foot syndrome. The cited abstract says only to avoid chemical and physical stress and use alcohol-free moisturiser; the EGFR guideline says protective footwear and covered fingertips to avoid friction. (Second accuracy pass; 2026 guideline comparison abstract from Europe PMC and MASCC 2011 full text) · Reworded to what the two sources say, and the missing advice is named in the gaps.
- 2026-11-17 · Skin and nail care during chemotherapy and targeted therapy: what to report and how fast: The page cited a booklet for reporting sore or blistered hands and feet, which the booklet does not mention, and said to get urgent help for a rash with wheezing, where the booklet says to tell the doctor or nurse right away. (Second accuracy pass; NCI Chemotherapy and You PDF and the NCI skin and nail changes page, read as text) · The NCI page is now a cited source for what to tell the team, and the urgency wording follows the booklet.
- 2026-11-17 · Skin and nail care during chemotherapy and targeted therapy: reasons given for steps: Several reasons were in no source read: that hot water strips the skin's oils, that thick creams hold water better than lotions, that the rash is not ordinary acne, and that early treatment makes the rash milder. (Second accuracy pass; MASCC 2011 full text and NCI patient material read as text) · Each reason now gives what the guideline or NCI says.
- 2026-11-17 · Skin and nail care during chemotherapy and targeted therapy: the most recent guidance: The page called the unread ESMO 2021 guideline the most recent comprehensive one and dated the NCI booklet 2023. A 2025 European and US expert consensus on the rash is newer and favours a medium-potency steroid cream, doxycycline at 100 or 200 mg a day and SPF 30 or higher; the booklet is printed January 2024. (Second accuracy pass; J Eur Acad Dermatol Venereol 2025 position paper read as full text from Europe PMC; NCI booklet back cover) · The consensus is added as a source and a variant, and the booklet year corrected.
- 2026-11-17 · Smoking cessation at cancer diagnosis and during treatment: plain-language summary: The page said stopping is not a test a person has to pass to get treatment, as a general rule. The only source for it is NICE's lung cancer guideline, which says not to postpone surgery for lung cancer to allow people to stop smoking. (Second accuracy pass; NICE NG122 section 1.3, HTML read as raw text) · Reworded to what NICE says about lung cancer surgery and attributed to NICE.
- 2026-11-17 · Smoking cessation at cancer diagnosis and during treatment: behavioural support step: The step told teams to arrange behavioural support from a trained adviser as a flat order. NICE says to offer it and arrange it if the person agrees, and the recommendation does not mention adviser training. (Second accuracy pass; NICE NG209 recommendation 1.14.5, HTML read as raw text) · Restored "if the person agrees" in the step and the summary and removed "trained adviser".
- 2026-11-17 · Smoking cessation at cancer diagnosis and during treatment: reasons given for nine steps: Nine steps gave reasons no cited source states, for example that medicines work best when in place before the last cigarette, that waiting for someone to stop lets the cancer grow, and that the first weeks carry the strongest cravings. (Second accuracy pass; NICE NG209, NICE NG122 and the IASLC 2022 statement, HTML read as raw text) · Each reason now says what the cited source says, including the timing NICE gives for starting each medicine and the medicines NICE names as affected by stopping.
- 2026-11-17 · Smoking cessation at cancer diagnosis and during treatment: surgery step: The step said smoking raises the risk of lung complications after an operation. NICE says this of lung cancer surgery only. (Second accuracy pass; NICE NG122 section 1.3 and NICE NG209 recommendation 1.14.1, HTML read as raw text) · Scoped to lung cancer surgery, and added NICE's general statement that people who stop in the 8 weeks before surgery can benefit significantly.
- 2026-11-17 · Smoking cessation at cancer diagnosis and during treatment: relapse step: The step said to offer treatment again after a relapse and cited a NICE sentence about record keeping. NICE says to encourage people who have not met their goal to try again and to discuss which option to use next. (Second accuracy pass; NICE NG209 recommendations 1.11.6, 1.11.7 and 1.17.7, HTML read as raw text) · Reworded to NICE's wording, with the NCCN abstract's "retreatment as needed" named as the other source.
- 2026-11-17 · Smoking cessation at cancer diagnosis and during treatment: sustained counselling variant: The page said the trial gave eleven counselling sessions. The abstract prints four weekly, four fortnightly and three monthly sessions and no total. (Second accuracy pass; Park 2020 abstract from Europe PMC) · Removed the total and kept the schedule as printed.
- 2026-11-17 · Smoking cessation at cancer diagnosis and during treatment: opt-out referral: The page said opt-out programmes had been studied in cancer centres and through programme evaluations. One programme at one cancer centre was read, and it had no comparison group. (Second accuracy pass; Wakeman 2026 full text from Europe PMC) · Reworded to one cancer centre programme and added that abstinence was self-reported among the 37.5% of referred patients who completed follow-up.
- 2026-11-17 · Smoking cessation at cancer diagnosis and during treatment: cutting down variant: The page said the quit date is within six weeks. NICE says it should normally be within 6 weeks of starting behavioural support. (Second accuracy pass; NICE NG209 recommendation 1.15.9, HTML read as raw text) · Restored "normally" and the starting point.
- 2026-11-17 · Smoking cessation at cancer diagnosis and during treatment: NCCN edition: The page cited NCCN version 3.2022 and said a later version could not be checked. The NCCN website lists version 2.2026 as current. (Second accuracy pass; nccn.org guideline listing, HTML read as raw text) · The gaps now name version 2.2026 and say it needs a login and was not read.
- 2026-11-17 · Stoma care after cancer surgery: teaching before and after the operation: The page said a specialist stoma nurse should teach everyone before and after the operation. The ASCRS guideline scopes this to elective stoma creation and says a specialised provider such as a wound, ostomy and continence nurse. (Second accuracy pass; ASCRS 2022 full text read raw from the publisher) · The step, the short summaries and the grade row now say a planned stoma and a specialised provider such as a stoma nurse.
- 2026-11-17 · Stoma care after cancer surgery: the new ileostomy pathway: The step ordered that a new ileostomy be managed on a care pathway with three named parts. ASCRS says only that a pathway may decrease readmission, a weak recommendation (2B), and the three parts are those of one retrospective pathway it describes. (Second accuracy pass; ASCRS 2022 full text read raw from the publisher) · The step now says consider, and attributes the parts to the one pathway the guideline describes.
- 2026-11-17 · Stoma care after cancer surgery: signs of dehydration and skin problems: The page said no readable source states the signs of dehydration or when to call, and told readers to report skin problems with no source. The NHS page on ileostomy complications states both, for people in England. (Second accuracy pass; NHS page read raw) · The NHS page was added as a source. A new step gives its signs of dehydration and its advice on when to seek help, and the skin step now rests on it.
- 2026-11-17 · Stoma care after cancer surgery: reasons given for the steps: Six reasons were not in the cited sources: that an expected stoma is easier to accept, that a well placed stoma leaks less, that waste runs under the seal of a flat stoma, that a nurse can usually fix the fit, the list of topics a specialist answers, and that dehydration is treated more simply when caught early. (Second accuracy pass; ASCRS 2022, NICE NG151 and Vogel 2022 read raw) · Each reason was rewritten to what its source says.
- 2026-11-17 · Stoma care after cancer surgery: what NICE asks before surgery: The step said NICE asks that people are told before surgery how likely a stoma is. NICE recommendation 1.2.2 asks that people are advised of possible reasons their treatment plan might change, the likelihood of a stoma among them, and does not say when. (Second accuracy pass; NICE NG151 recommendations page read raw) · The step was reworded to the recommendation.
- 2026-11-17 · Stoma care after cancer surgery: the RNAO nursing guideline: The page described what the RNAO guideline covers, including follow-up in the first four weeks at home, although the guideline could not be read. (Second accuracy pass; rnao.ca returned 403 again) · The description was removed. The page now reports only what the Canadian position statement says of it.
- 2026-11-17 · Stoma care after cancer surgery: variants: The rehydration drink was described as given every day for about six weeks, where the trial abstract says 40 days after discharge. The high output variant said many hospitals act on a daily volume, which the consensus paper does not say. The ERAS Society 2025 recommendations were listed as unreadable. (Second accuracy pass; Europe PMC abstract, Carr 2025 full text, ERAS 2025 open copy read raw) · The drink is now described as in the abstract, the hospital claim was replaced with the paper's own wording, and the ERAS 2025 guideline is recorded as read with what it does and does not say about stomas.
- 2026-11-17 · Sun protection during and after cancer treatment: evidence gap: The page said no trial has tested sun protection advice during anticancer treatment. The MASCC guideline it cites refers to a randomised placebo-controlled trial (N05C4, 2010) of SPF 60 sunscreen twice a day for 28 days in people starting an EGFR inhibitor, which found no difference in rash (78% against 80%). (Second accuracy pass; the MASCC 2011 full text from Europe PMC and the trial abstract from Europe PMC) · The gap now names the trial and what it tested, and the trial is listed as a source read as an abstract.
- 2026-11-17 · Sun protection during and after cancer treatment: NCI chemotherapy booklet year: The booklet was dated 2023. Its back cover prints NIH Publication No. 24-7157, January 2024. (Second accuracy pass; the NCI PDF read as text) · Year set to 2024 throughout.
- 2026-11-17 · Sun protection during and after cancer treatment: hours to avoid the sun: The page said NCI advises avoiding direct sun from 10 a.m. to 4 p.m. NCI says to avoid direct sunlight and that the summer sun tends to be strongest from 10 a.m. until 4 p.m. (Second accuracy pass; the NCI chemotherapy booklet PDF read as text) · Reworded in the summary, the variant and the gaps, and the UK and US hours are attributed to their sources.
- 2026-11-17 · Sun protection during and after cancer treatment: radiotherapy skin after treatment: The page told readers to keep treated skin out of the sun and said treated skin of any kind stays sensitive for years. Macmillan says to take extra care to protect radiotherapy skin, especially for the first year, and that it stays sensitive for many years; NCI says it will always burn quickly. (Second accuracy pass; the Macmillan page and the NCI radiotherapy booklet read as text) · Step, plain summary and one-line summary reworded to the sources, with both durations given.
- 2026-11-17 · Sun protection during and after cancer treatment: reasons given for the steps: Five reasons were not in the cited sources: that most ultraviolet light arrives at midday, that clothing protects more reliably than sunscreen, that products can worsen the radiotherapy reaction or interfere with treatment, that sun worsens radiation recall, and that sunbeds give the same light as the sun. (Second accuracy pass; the NHS page and both NCI booklets read as text) · Each reason replaced with what the source says; where NCI gives no reason the page says so.
- 2026-11-17 · Sun protection during and after cancer treatment: applying sunscreen: The page told everyone to apply sunscreen 30 minutes before going out and again just before. The NHS says this for people who plan to be out long enough to risk burning. (Second accuracy pass; the NHS page fetched and read as text) · Condition added, with reapplication straight after water, towel drying or sweating as the NHS words it.
- 2026-11-17 · Sun protection during and after cancer treatment: what the sources agree on: The page said NCI and Macmillan agree on five points and that NCI's radiotherapy booklet gives SPF 30. Each point comes from one publisher, and NCI says only that the team may suggest SPF 30 or higher. The page also said no graded guideline exists although the ESMO 2021 guideline could not be read. (Second accuracy pass; the NCI booklets and the Macmillan page read as text) · Each point attributed to its source, the hedge restored, and the statement limited to the sources read.
- 2026-11-17 · Sun protection during and after cancer treatment: which drugs and for how long: The page said the drugs that cause sun sensitivity and how long it lasts are set out in each drug's label, which no cited source says. Macmillan says sensitivity after some chemotherapy can sometimes last several years. (Second accuracy pass; the Macmillan page read as text) · Population and gap reworded; Macmillan's statement added.
- 2026-11-17 · Surgical drain care at home after cancer surgery: removal volume: The summary called a removal volume of 30 to 50 ml in 24 hours common. No source prints that. WHO describes usual practice as waiting until the amount falls to a set volume, typically less than 30 mL to 100 mL, or removing at a set time, and recommends neither. (Second accuracy pass; WHO global guidelines 2018, section 4.26, PDF read as text) · Replaced with WHO's wording, and the page now says that no source read gives a volume or a longest number of days.
- 2026-11-17 · Surgical drain care at home after cancer surgery: when to call: The page gave a hospital leaflet's call thresholds (a 100 mL change in output, fever of 38.3 °C) without its two levels of urgency, and listed infection signs under NICE that NICE does not print (spreading redness, pus). The leaflet says to call right away for fever, bright red fluid, increased redness, swelling, tenderness or pus, and in office hours for a change in output. NICE's definition lists heat, redness, pain, swelling and fever. (Second accuracy pass; Memorial Sloan Kettering leaflet and NICE NG125 read as raw text) · The step now carries NICE's signs, a new variant gives the leaflet's two lists with their urgency and the NHS advice for England, and a gap says no guideline sets these.
- 2026-11-17 · Surgical drain care at home after cancer surgery: emptying the drain: The page said no trial read states how often to empty a drain. One trial now read in full (Degnim 2013) told all patients to strip the tubing, empty the bulb and record the volume at least twice daily, as routine and not as the thing tested. (Second accuracy pass; Degnim 2013 author manuscript read in full from PubMed Central) · The practice step and the variant now say so.
- 2026-11-17 · Surgical drain care at home after cancer surgery: information and teaching: The page said to give written information on the drain site, citing NICE, and to teach drain care, citing the early discharge trial. NICE says to offer information and advice on wound care and does not name drains or writing, and the trial's readable text does not describe teaching. (Second accuracy pass; NICE NG125 and Wells 2004 full text read as raw text) · Reworded to what each source says, with the teaching attributed to the trial that describes it.
- 2026-11-17 · Surgical drain care at home after cancer surgery: reasons given for steps: Several reasons had no source: that people remember little on the day of discharge, that infection is easier to treat when caught early, that advice on showering varies between hospitals, and that antibiotics are continued most often after implant reconstruction. The removal step also added "as judged by the surgical team, rather than on a fixed day", which WHO does not say. (Second accuracy pass; every cited source read as raw text) · Each replaced with a statement a source makes, or removed.
- 2026-11-17 · Surgical drain care at home after cancer surgery: sources: The American Society of Breast Surgeons document was titled a Consensus Guideline; it is a Resource Guide that calls itself a consensus document. The title of the Sattar 2023 trial had been altered. (Second accuracy pass; the society's PDF and statements index, and Europe PMC) · Both labels corrected.
- 2026-11-17 · End of treatment: treatment summary, survivorship care plan and handover to primary care: timing of the GP cancer care review: The page said the Quality and Outcomes Framework requires the cancer care review within 12 months of the end of acute treatment. The indicator Macmillan prints (CAN004, 2023/24) counts 12 months from the date of diagnosis, with a first contact within 3 months of diagnosis; Macmillan says the 12 months is so the review can coincide with the end of treatment. (Second accuracy pass; Macmillan Cancer Care Review page, raw HTML read on 17 November 2026) · Step, summary and the England variant now say within 12 months of diagnosis, the 3-month contact is added, the step quotes the Macmillan sentence, and the page is added as a source.
- 2026-11-17 · End of treatment: treatment summary, survivorship care plan and handover to primary care: what to do about the written plan: The page presented as the protocol, and as best supported practice, that effort should go into a needs assessment, a named responsible clinician and follow-through, and ordered pairing the plan with a discharge conversation. No source read says this. The 2018 review reports limited but more positive findings when a plan comes with counselling and asks for research on making sure recommendations are acted on; the breast cancer trial found the plan no better than a standard discharge visit. (Second accuracy pass; Jacobsen 2018 and Grunfeld 2011 abstracts and Hill 2020 full text, read raw on 17 November 2026) · Step, summary, patient summary and best-in-class statement reworded to what the reviews report, with the hedge kept and the statement marked as ungraded.
- 2026-11-17 · End of treatment: treatment summary, survivorship care plan and handover to primary care: strength of the trial evidence: The page called the evidence consistently negative, said the breast trial had one gain, and described the 2020 meta-analysis as eight trials. The 2018 review says generally negative with positive findings in single studies, including satisfaction with care; the breast trial was read only as an abstract; the meta-analysis counts eight articles. (Second accuracy pass; Europe PMC abstracts and Hill 2020 full text from PubMed Central, read on 17 November 2026) · Reworded to generally negative with the single-study positives named, the abstract's one reported difference, and eight articles.
- 2026-11-17 · End of treatment: treatment summary, survivorship care plan and handover to primary care: who and when: The trigger included the start of long-term endocrine therapy and the population was scoped to curative intent, neither stated by the guidance read. The NCI page was dated 2026 but prints an update date of 2 December 2024. Several reasons given for steps (problems being common, a named person preventing gaps) were not in the cited pages, and the NCI's may was written as an instruction. (Second accuracy pass; NCI and Macmillan pages as raw HTML and Tevaarwerk 2019 full text through NCBI E-utilities, read on 17 November 2026) · Trigger and population reworded to the sources, the NCI year set to 2024, and the reasons and the instruction reworded to what each cited page says.
- 2026-11-17 · End of treatment: treatment summary, survivorship care plan and handover to primary care: which country: The page said the treatment summary goes to the patient and GP in England. Macmillan's page names the NHS Long Term Plan but no nation; only the framework specification it links to is NHS England's. (Second accuracy pass; Macmillan pages, raw HTML read on 17 November 2026) · The handover is now attributed to the NHS as Macmillan describes it, England is kept for the framework indicators, and the limit is stated in the gaps.
- 2026-11-17 · Urgent suspected cancer referral, safety netting and the waiting time standards: review for people not referred: The step, the summary and the grades table said people with a symptom linked to cancer who do not meet the criteria are offered a review, or that a review is agreed. NICE NG12 recommendation 1.15.2 says to consider a review. (Second accuracy pass; NICE NG12 (last updated 15 April 2026), HTML read as raw text) · Reworded to consider a review, with the two forms NICE names.
- 2026-11-17 · Urgent suspected cancer referral, safety netting and the waiting time standards: immediate referral: The step said people with emergency features are sent for assessment the same day. NICE NG12 defines immediate as an acute admission or referral within a few hours, or even more quickly if necessary. (Second accuracy pass; NICE NG12 terms, HTML read as raw text) · Reworded to within a few hours, with the example NICE gives.
- 2026-11-17 · Urgent suspected cancer referral, safety netting and the waiting time standards: discussion with a specialist: The step told the clinician to discuss an uncertain case with a specialist. NICE NG12 recommendation 1.16.2 says discussion should be considered. (Second accuracy pass; NICE NG12, HTML read as raw text) · Reworded to consider discussing.
- 2026-11-17 · Urgent suspected cancer referral, safety netting and the waiting time standards: what the guidelines share: The page said the steps of referral are the same in the guidelines read and that NICE NG12 rests on a 3% threshold. Referral within 1 working day and telling people that most referred do not have cancer are in NICE only, and the 3% threshold is stated in the Scottish guidelines, which call it in line with NICE. (Second accuracy pass; NICE NG12 and the Scottish Referral Guidelines 2025, HTML read as raw text) · Each statement now names the guideline that makes it, and steps resting on the Scottish guidelines say so.
- 2026-11-17 · Urgent suspected cancer referral, safety netting and the waiting time standards: reasons given for steps: Nine reasons were not in any source read, among them that people told why they are referred are more likely to attend, and that people who miss appointments have the most in their way. The outcome step also said to tell the person as soon as it is known, which no source states. (Second accuracy pass; NICE NG12 and the Scottish Referral Guidelines 2025, HTML read as raw text) · Each reason now says only what the cited guideline says.
- 2026-11-17 · Urgent suspected cancer referral, safety netting and the waiting time standards: England's standards: The page said England's 31-day and 62-day standards and clock starts could not be described, and stated without a source that the 28-day definition replaced a two-week appointment. (Second accuracy pass; the Handbook to the NHS Constitution for England (GOV.UK, updated 24 January 2025) and Cancer Research UK's cancer waiting times page (reviewed 9 October 2026), HTML read as raw text) · England's 28-day, 31-day and 62-day waits and where each starts are now given from the Handbook, the October 2023 change from Cancer Research UK, and the percentages stay named as a gap.
- 2026-11-17 · Superior vena cava obstruction in cancer: when to seek urgent help: The second accuracy pass removed the unsourced advice to be assessed the same day and to ring 999, which left the page with no sourced statement of which symptoms a patient should report and how fast. (Second accuracy pass follow-up; Macmillan Cancer Support page on superior vena cava obstruction (reviewed 31 January 2025), fetched raw and read as text) · A first step now lists the symptoms Macmillan lists and says to contact the cancer doctor or nurse straight away, with Macmillan's sentence quoted. The short summary and one-line version use the same words. The page says that no source read tells a patient when to go to an emergency department instead.
- 2026-11-17 · Raised intracranial pressure in cancer: when to seek urgent help: The page told patients to contact their cancer team urgently on the strength of a professional guideline alone, with no patient source saying which symptoms to report or what to do after a fit. (Second accuracy pass follow-up; Cancer Research UK page on symptoms of brain tumours (last reviewed 6 March 2026), fetched raw and read as text) · A first step now lists the symptoms Cancer Research UK gives for increased pressure, and says to see a doctor or go to A&E immediately after a seizure and to see a doctor for the others, with the sentence quoted. The one-line version adds the advice on fits. The page says that no source read sets a time limit for the other symptoms.
- 2026-11-17 · Malignant pericardial effusion: when to seek urgent help: The page had no sourced list of warning signs for patients, and its one-line version told readers to ring 999 for severe breathlessness, which no source read states. (Second accuracy pass follow-up; US National Cancer Institute PDQ patient summary on cardiopulmonary syndromes (updated 15 May 2025), fetched raw and read as text) · A first step now lists the signs the summary lists and says to check with your doctor, with the sentence quoted, and notes that tamponade is life-threatening and must be treated right away. The one-line version was reworded to match and no longer names 999. The page says that no UK patient page on this problem was found and that no source read names an emergency number.
- 2026-11-17 · Metastatic spinal cord compression: when to seek urgent help: After the second accuracy pass the page had no sourced step telling a patient which symptoms need help and how fast. NICE's time limits are written for the professional who contacts the coordinator, not for the patient. (Second accuracy pass follow up; Macmillan Cancer Support page on metastatic spinal cord compression (reviewed 1 September 2024), fetched and read as raw text, quote checked by script; NICE NG234 recommendations fetched again) · Added a patient step in Macmillan's words: get medical advice immediately for any symptom of cord compression, even at a weekend or on a holiday, from the usual hospital team, and go to the nearest emergency department if nobody can be reached. The step lists Macmillan's warning signs, and the patient summaries now use its wording.
- 2026-11-17 · Immune-mediated hepatitis from checkpoint inhibitors, by grade: when to seek urgent help: After the second accuracy pass the page told patients only to mention yellow skin or eyes, with no other symptoms and no urgency, because neither guideline states them. (Second accuracy pass follow up; Macmillan Cancer Support pembrolizumab page (reviewed 1 August 2026) and the US Medication Guides for pembrolizumab, nivolumab and ipilimumab from DailyMed, read as raw text, quote checked by script) · Added a patient step: contact the hospital straight away on the 24 hour number for yellow skin or eyes, feeling very sleepy, dark urine, unexplained bleeding or bruising, or pain in the right side of the tummy, during treatment or after it ends. The reason gives the Medication Guides' own list and their words, call or see your healthcare provider right away. The gaps say which leaflets were not read.
- 2026-11-17 · T-cell engaging bispecific antibodies: step-up dosing, monitoring and outpatient delivery: where the US and EU labels agree: The page said the US and EU labels for the three BCMA bispecifics agree on the doses. Only the step-up doses agree: the teclistamab days differ (US days 1, 4 and 7; EU days 1, 3 and 5) and some premedication doses differ. (Independent dose re-read; current US prescribing information (DailyMed SPL XML) and EU product information PDFs, read as raw text) · Reworded to say the labels agree on the step-up doses, with the differences named.
- 2026-11-17 · T-cell engaging bispecific antibodies: step-up dosing, monitoring and outpatient delivery: talquetamab, glofitamab, epcoritamab and mosunetuzumab: Four bispecifics were linked to the page although no source read covered them and the page itself said their labels had not been checked. (Independent dose re-read; current US prescribing information for Talvey, Columvi, Epkinly and Lunsumio, DailyMed SPL XML read as raw text) · The four US labels were added as sources with their step-up doses and hospital instructions as a variant and four quoted label rows; the EU product information for these four is named as not read.
- 2026-11-17 · ICANS after CAR T-cell therapy: steroid doses from the axicabtagene ciloleucel label: The page gave only the label doses that SITC 2020 reported from an earlier package insert (dexamethasone 10 mg every 6 hours up to grade 3 and methylprednisolone 1000 mg daily for 3 days at grade 4). The current US label, revised 07/2026, prints dexamethasone 10 mg four times a day at grade 2, methylprednisolone 1000 mg once daily at grade 3 and 1000 mg twice a day at grade 4. (Independent dose re-read; YESCARTA US prescribing information revised 07/2026, read as pdftotext of the fda.gov PDF and as the DailyMed text) · The current label doses are now given beside the SITC report in the summary, a variant and the gaps, with the label added as a source.
- 2026-11-17 · Strong opioids for cancer pain: rescue dose as a share of the daily dose: The page said that the WHO rescue dose of 50 to 100% of the regular 4-hourly dose works out at roughly 8 to 17% of the daily dose. WHO prints only the 50 to 100% figure and no source read prints the percentage of the daily dose, so it was a figure computed on the page. (Independent dose re-read; WHO 2018 cancer pain guideline section 6.2.5 and Annex 6, the PDF read as text with pdftotext) · The computed percentage was removed and the page now says WHO prints no share of the daily dose. The routes of the WHO annex range of 2 to 20 mg every 4 hours were also stated in full.
- 2026-11-17 · Catheter-related thrombosis in cancer: drug links: The page linked apixaban, rivaroxaban and edoxaban as its drugs. No source read names any of them for a clot around a central line in adults: the ITAC guideline suggests low molecular weight heparin and mentions direct oral anticoagulants only as a class that has not been compared, the ASH guideline names no drug, and none of the three US labels covers this use in adults. (Independent re-read of drug links; ITAC 2022 and ASH 2021 saved raw texts and the three US labels from DailyMed) · The three links were removed and the gap says why.
- 2026-11-17 · Oral anticancer medicines at home: drug class link: The page linked the CDK4/6 inhibitor class, which no source read on the page names. (Independent re-read of drug links; the page's sources as saved raw text) · The link was removed and the gap reworded.
- 2026-11-17 · Stem cell mobilisation and collection: plerixafor maximum dose in kidney impairment: The page gave plerixafor as never more than 40 mg a day and 0.16 mg/kg when creatinine clearance is 50 mL/min or less, with no separate maximum. The US label says that at that level of kidney function the dose is 13 mg or 0.16 mg/kg at 83 kg or less and should not exceed 27 mg a day. (Independent dose re-read; Mozobil US prescribing information section 2.3 and Table 1, DailyMed label XML fetched 17 November 2026) · The summary and the plerixafor step now give the fixed 13 mg option and the 27 mg a day maximum for creatinine clearance of 50 mL/min or less.
- 2026-11-17 · Hearing protection for children receiving cisplatin: sodium thiosulfate vial strength: The page said neither label relates the US and European dose figures. The US label does say its 100 mL vial holds the equivalent of 12.5 g of sodium thiosulfate pentahydrate (125 mg/mL), provided as 8 g of the anhydrous salt, and the European label gives 80 mg/mL of the anhydrous salt. (Independent dose re-read; PEDMARK US prescribing information sections 2.1, 3 and 11, DailyMed label XML fetched 17 November 2026, and the Pedmarqsi product information PDF from the European Medicines Agency) · The summary, the dosing step and the gap now give both vial strengths and the US label's statement that PEDMARK is not substitutable with other sodium thiosulfate products.
- 2026-11-17 · Clostridioides difficile in children with cancer: IDSA and SHEA doses in the summary: The summary gave the metronidazole and vancomycin doses without saying they are for a non-severe episode and without their maximum doses. The table gives 500 mg a dose as the maximum for metronidazole and 125 mg a dose for oral vancomycin in a non-severe episode. (Independent dose re-read; IDSA and SHEA 2017 guideline Table 2, PubMed Central text) · The summary now says the doses are for a non-severe first episode and gives the maxima.
- 2026-11-17 · GVHD prophylaxis: tacrolimus starting dose in BMT CTN 1703: A gap said the trial gives tacrolimus starting doses in accordance with institutional practices, which read as if it printed none. The trial prints 0.05 to 0.06 mg/kg a day by mouth or 0.02 to 0.03 mg/kg a day intravenously. (Independent dose re-read; BMT CTN 1703 full text, PubMed Central author manuscript) · The gap now gives the printed starting doses and says they are the trial's protocol, not a recommendation.
- 2026-11-17 · Diarrhoea from chemotherapy, targeted therapy and radiotherapy: loperamide daily maximum in NCI PDQ: The page gave about 12 mg a day as the only loperamide maximum in NCI PDQ and set it against the 16 mg of eviQ. PDQ prints about 12 mg a day in its diarrhoea section and also says, in its section on acute radiation enteritis, that the daily total should not exceed 16 mg. (Independent dose re-read; NCI PDQ Gastrointestinal Complications health professional version, fetched as raw HTML from cancer.gov on 17 November 2026) · Added the 16 mg limit for radiation enteritis beside the 12 mg figure in the loperamide step, the summary and the best in class statement.
- 2026-11-17 · Renal dose adjustment of anticancer drugs in reduced kidney function: methotrexate at eGFR 45 to 59 and 30 to 44: The page made keeping an exposure threshold a condition for full-dose methotrexate of 500 mg/m2 or more and said consider, and it left out when ADDIKD chooses an alternative protocol over a dose cut. ADDIKD says to give the full dose where intent is curative, performance status is good and there are no nephrotoxic co-medicines, noting this matters most in primary CNS lymphoma. Below 500 mg/m2 it says to consider an alternative where intent is curative with both poor performance status and nephrotoxic co-medicines, and a 25% reduction in all other patients. At eGFR 30 to 44 it says a 50% reduction except where an alternative is considered for curative intent or for poor performance status with nephrotoxic co-medicines. (Independent dose re-read; ADDIKD guideline version 1 (eviQ 2022), section 4.34, PDF read as raw text) · The methotrexate variant now states each condition as the guideline does and says the reductions are for non-transplant settings.
- 2026-11-17 · Renal dose adjustment of anticancer drugs in reduced kidney function: cisplatin 50 mg/m2 or below at eGFR 45 to 59: The page said to consider full dose. ADDIKD says the clinical consensus is to administer full dose cisplatin where intent is curative, performance status is good and there are no nephrotoxic co-medicines. (Independent dose re-read; ADDIKD guideline version 1 (eviQ 2022), section 4.11, PDF read as raw text) · The cisplatin variant now says the consensus is to give the full dose in that group.
- 2026-11-17 · Linear accelerator quality assurance: UK notification level for standard output: The page gave the commonest UK notification level for standard output as 1.5% in 16 of 37 centres. In the survey's Table 4 that count is the electron-beam row; the photon-beam row prints 17 of 40 centres, and suspension at 2% in 26 of 38 (photons) and 22 of 38 (electrons). (Independent dose re-read; Tassano-Smith et al., British Journal of Radiology 2025, full text from Europe PMC read as raw text) · Both rows are now given: 17 of 40 centres for photon beams and 16 of 37 for electron beams, with the suspension counts for each.
- 2026-11-17 · Scalp cooling during chemotherapy: contraindications in the Paxman summary: The page listed the contraindications of the 2018 Paxman summary but left out three that it prints: haematological malignancies (leukaemia, non-Hodgkin and other generalised lymphomas) or haematological malignancies being treated for cure, cold migraine, and post-traumatic cold dystrophy. The page had described the exclusion of blood cancers as limited to curative-intent treatment. (Independent dose re-read; FDA 510(k) summary K173032, PDF read with pdftotext, beside K170871 and DEN150010) · The three contraindications were added to the step, and the population now gives the Paxman wording on blood cancers beside the DigniCap and Macmillan wording.
- 2026-11-17 · Bleeding precautions at home when platelets are low: when to seek urgent help: The page listed bleeding signs to call about but gave no urgency, because the earlier wording "straight away" and "the same day" had no source and was removed. (Second accuracy pass follow-up; Cancer Research UK, Your blood, bone marrow and cancer drugs, last reviewed 22 July 2026, full text fetched with curl and the quote matched by script) · Added a step, with the source, saying that a person whose blood count is low because of cancer drug treatment should call the advice line, GP or specialist nurse immediately on noticing any bleeding or bruising. The short summaries and the gaps now say that this is the only source read that gives an urgency.
- 2026-11-17 · MOUNTAINEER: The page showed the registry details of two studies: its own (NCT03043313, 117 participants, completed) and those of its phase 3 successor, MOUNTAINEER-03 (NCT05253651, 366 sites), because any ClinicalTrials.gov link on a record was read as that record's registration (ClinicalTrials.gov API v2 for both numbers) · The fetcher now treats a linked registration that another record owns as a citation. MOUNTAINEER-03's details appear only on its own page. The same rule removed the ICONIC registration from the Cedars-Sinai ivermectin trial
- 2026-11-17 · START-B: The page showed the registry record of START Trial A (NCT00005588). The ISRCTN record covers both trials and lists only Trial A's ClinicalTrials.gov number (ClinicalTrials.gov API v2: NCT00005588 is titled "Standardization of Breast Radiotherapy: Trial A", NCT00005587 "Trial B") · START-B now carries NCT00005587, and the fetcher refuses the cross-reference
- 2026-11-17 · FOxTROT: The page showed the registry record of FOxTROT2France (NCT06293625), a separate French study of 150 people aged 70 or more, which the ISRCTN record lists (ClinicalTrials.gov API v2) · The fetcher refuses the cross-reference; FOxTROT2France keeps its own page
- 2026-11-17 · SEQUOIA, SEQUOIA (pancreatic cancer): One record held two unrelated trials that share an acronym: zanubrutinib in chronic lymphocytic leukaemia (NCT03336333, BeiGene) and pegilodecakin with FOLFOX in pancreatic cancer (NCT02923921, Eli Lilly). The page mixed their drugs, cancers, sponsors and sites (ClinicalTrials.gov API v2 for both numbers; Crossref for the pancreatic trial's report (10.1200/JCO.20.02232)) · The pancreatic trial is now its own record. The existing address stays with the leukaemia trial, which is what its text described
- 2026-11-17 · KontRASt-02, JDQ443: The trial was shown as active. The registry lists it as terminated, "Business reasons and not based on any safety concerns.", with 95 participants (ClinicalTrials.gov API v2, record last updated 6 November 2026) · Status, summary, plain-language line and translations corrected; the drug's page now says so too
- 2026-11-17 · CAMBRIA-1 & CAMBRIA-2: The record is named for two trials but carried only CAMBRIA-2's registration, described it as leading both, and gave 10,000 as a registry enrolment figure (ClinicalTrials.gov API v2: CAMBRIA-1 is NCT05774951 (recruiting, 4,300 planned), CAMBRIA-2 is NCT05952557 (closed to recruitment, 5,511 enrolled)) · The record names both registrations, says whose registry details it shows, gives 9,811 as the sum of the two and links CAMBRIA-1's own record
- 2026-11-17 · Trial statuses and phases across the corpus: 206 registry entries shown on trial pages carried a recruitment status the registry has since changed, and 234 trial records carried a lifecycle status the registry no longer gives (122 more than at the last pass on 19 October). Twelve registry-ingested trials and two hand-written ones (Beamion LUNG-1, CARD) carried a phase the registry does not give. Three trials running only in Austria were among them: LIQUIPANC (NCT06391892, shown as recruiting phase 3; the registry says phase 2, no longer recruiting), NCT05866432 and NCT04875585 (both shown as recruiting; no longer recruiting) (ClinicalTrials.gov API v2 for all 6,418 registration numbers in the corpus) · Every changed entry was read again and replaced; phases, and three trial titles the sponsors had revised, now follow the registry. NHWD-870's phase 2 trial (NCT06527300), filed as historic while its status was unknown, is recruiting again and says so
- 2026-11-17 · Francisco Berrospi Espinoza: The plain-language line and eight translations said he heads Peru's INEN (INEN's pages on gob.pe: its directory of officials lists Wilder Genaro Calmet Berrocal as Jefe Institucional and Francisco Berrospi Espinoza as Director General de la Dirección de Cirugía) · The lines now give his present post. The Arabic and Hindi lines drop the clause
- 2026-11-17 · IPO Coimbra: The record said the institute hosts the regional cancer registry (Lei n.º 53/2017, read as PDF text: "Os dados existentes nos Registos Oncológicos Regionais (ROR) são integrados no RON", in force from 1 January 2018) · The record now describes the national registry, as the IPO Lisboa record already did
- 2026-11-17 · Cervical cancer screening programme in Myanmar: The record called the programme national. IARC's fact sheet does not; the only thing it describes as national is the information system (The text file behind IARC's CanScreen5 fact sheet for Myanmar) · "National" removed, and the record says the fact sheet gives no geographic reach
- 2026-11-17 · INCAN, Guatemala: The record said the Liga keeps a national cancer registry. Its UICC page says it registers cancer cases "through its Department of National Cancer Registry since 1975", and Guatemala told WHO in 2021 that it has no population-based cancer registry (UICC's member page read as raw text; WHO Global Health Observatory API) · The record now quotes the page and gives WHO's answer beside it
- 2026-11-17 · Hamad Medical Corporation: The record said the corporation was established in 1979 and that its cancer centre "has been accredited by the Joint Commission". HMC's pages give no founding year, and they disagree on the accreditation: the centre's home page says "JCI Accredited in 2006", its About page "since 2011" (Four pages of hamad.qa read as raw text) · The year is removed, and the record quotes both statements and says they conflict
- 2026-11-17 · Kuwait Cancer Control Center: The record said the centre was founded in 1968. Its strategic plan says "In 1968, the old KCCC building for Radiation Oncology started. The Kuwait Cancer Control Center was established in 1982" (The centre's Strategic Plan 2019 to 2023, read as PDF text) · The record gives both dates from the plan and notes that the Arab Times gave 1968
- 2026-11-17 · Country ranking rows for Mexico, Iran, France and Taiwan: Funder and programme entries had no source. Mexico's named CONAHCYT and listed a hospital as a programme (The rows in src/data/country-extras.ts against the sources the country pages cite) · Entries with no source were removed; France keeps INCa and its ten-year strategy, Taiwan its cancer registry, each with a source
- 2026-11-17 · Countries: the trials column: The source note did not say that the count is of ClinicalTrials.gov studies of every condition, not cancer studies (scripts/fetch-countries.ts, which queries by location only) · The note shown under the table now says so. The column header is in a component and is unchanged
- 2026-11-17 · Norway, Bulgaria: Norway's count of 238 trial records included MOUNTAINEER-03 twice (The participation data after the corrections above) · Norway's page now gives 237 records and 236 registrations; Bulgaria's comparison follows
- 2026-11-17 · NewYork-Presbyterian Hospital: The one-line description called it a not-for-profit hospital system; none of the hospital pages the record cites says so. (Institution record re-check; nyp.org, its cancer page and its about page read in raw text) · Removed "not-for-profit" from the description
- 2026-11-17 · Beth Israel Deaconess Medical Center: The record said the hospital's home page describes a joint breast cancer care model with Dana-Farber; the home page describes the collaboration but not a breast cancer care model. (Institution record re-check; bidmc.org read in raw text) · The sentence now says only what the page says: a collaboration with Dana-Farber and Harvard Medical Faculty Physicians to widen access to cancer care and research
- 2026-11-17 · Chonnam National University Hwasun Hospital: The record named a "Gwangju-Jeonnam Regional Cancer Center" as linked from the hospital's site; the site names it 전남지역암센터, the Jeonnam Regional Cancer Center. (Institution record re-check; cnuhh.com read in raw text) · Renamed in the summary and the programme list
- 2026-11-17 · Abbott Northwestern Hospital: The one-line description called it a private not-for-profit hospital; neither Allina Health page the record cites says so. (Institution record re-check; the Allina Health hospital and cancer institute pages read in raw text) · Removed "private not-for-profit" from the description
- 2026-11-17 · Guam Regional Medical City: The record said the hospital "diagnoses and treats cancer in both children and adults". The hospital's radiation oncology, medical oncology and lung screening pages do not mention children or adults. (Institution record re-check; grmc.gu radiation oncology, medical oncology and lung cancer screening pages read in raw text) · The claim was removed from the one-line description and the summary, which now says the pages do not state whether children are treated.
- 2026-11-17 · Instituto Guatemalteco de Seguridad Social (IGSS): The record said the IGSS home page states that it protects more than 3.8 million people. The home page carries no such figure; the number comes from the institute's 80th anniversary statement, which gives it for August 2026. (Institution record re-check; igssgt.org home page, the 30 October 2026 anniversary statement and the 2025 services portfolio read in raw text) · The figure is now attributed to the anniversary statement, as more than 3.88 million people in August 2026, and the summary notes that IGSS's own documents date its founding decree to both 28 and 30 October 1946.
- 2026-11-17 · Centro Internacional de Cáncer (El Salvador): The record listed the centre's equipment in the IAEA directory as two linear accelerators and a CT scanner. The directory also lists a Gamma Knife unit there. (Institution record re-check; IAEA DIRAC equipment report read as raw data) · The Gamma Knife unit was added to the summary.
- 2026-11-17 · National Oncology Centre, The Royal Hospital, Muscat: The record called its Times of Oman source "undated" and gave visit numbers for "the year to August" without a year. The article is dated 23 September 2023, so the figures are for the year to August 2023. (Institution record re-check; the Times of Oman article read in raw text) · The summary and the link label now give the article's date, and the visit figure is tied to August 2023.
- 2026-11-17 · Registre des Cancers de Parakou: The record gave the population the registry covers as about 255,000. The African Cancer Registry Network page gives about 295,000, by the estimates of the 2013 census. (Institution record re-check; the African Cancer Registry Network page for the Parakou registry read in raw text) · The figure was corrected to about 295,000.
- 2026-11-17 · Department of Health Rizal Cancer Registry: The record said the Philippine Cancer Society turned its 1968 tumour registry into a population-based registry in 1984. The Department of Health report it cites gives 1983; 1984 is the year the Manila and Rizal registries began collecting data actively together. (Institution record re-check; Department of Health 2024 terminal report on population-based cancer registries read in raw text) · Year corrected to 1983 and attributed to the report.
- 2026-11-17 · Ministry of Health, Viet Nam: The record said a new insurance medicines list would add 84 medicines, 30 of them for cancer, but neither article it linked gives the figure 84 (Institution record re-check; SGGP (26 September 2026), Nguoi Lao Dong (June 2026) and Dan Tri (28 May 2026) read in Vietnamese in raw text) · The figure is right and comes from Dan Tri's report, which the Viet Nam page already cited; that report and the Nguoi Lao Dong one are now linked on the record
- 2026-11-17 · Bucharest Oncology Institute Prof. Dr. Alexandru Trestioreanu: The record said the OECD and European Commission's 2025 profile found that over half of the patients of Romania's three cancer institutes travel long distances from their home counties. The profile gives no such figure; it says the three institutes treat most cancer patients and that uneven distribution of staff and equipment means patients travel long distances. (Institution record re-check; Country Cancer Profile Romania 2025 read in raw text) · The sentence now says only what the profile says, and the profile is linked from the record.
- 2026-11-17 · Spanish Network of Cancer Registries (REDECAN): The record gave 3 December 2025 as the publication date of the network's third survival report. The report itself carries only the month, December 2025. (Institution record re-check; REDECAN survival report 2003-2017 read in raw text) · The day was removed; the record now says the report is dated December 2025.
- 2026-11-17 · Ministry of Health of Indonesia: The one-line description said the free health check programme includes breast and cervical screening, but none of the record's sources says so (Institution record re-check; the national cancer plan PDF and the ministry's releases read in raw text) · The clause was removed; the record now says only that the ministry runs the free health check programme
- 2026-11-17 · National Health Insurance Authority (NHIA): The office address was given as 297 Shehu Yar'adua Way, Utako, Abuja; the Authority's own site gives Plot 297, POW Mafemi Crescent, off Solomon Lar Way, Utako District (Institution record re-check; the Authority's website read in raw text) · Address replaced with the one the Authority publishes
- 2026-11-17 · Sociedad Anticancerosa de Venezuela: The record said the society's membership of the Union for International Cancer Control could not be confirmed, because UICC's page had refused an automated request on 11 November 2026 (Institution record re-check; UICC's member page read in raw text) · The page now answers and lists the society as a full member; the record says so
- 2026-11-17 · National Center for Disease Control and Public Health (Georgia): The record gave five-year survival for all cancers diagnosed in 2017 as 55.0%. The report prints two figures: 55.0% in its survival table and 56.3% in its summary. (Institution record re-check; NCDC report "Cancer in Georgia 2015-2021" read in raw text) · The record now gives both figures and says the report does not explain the difference, as the Georgia country page already did.
- 2026-11-17 · Dental and Pharmaceutical Benefits Agency (TLV): The record said that after three-party talks a company repays part of a medicine's cost to the regions. TLV's page says the repayment goes to the state and the regions. (Institution record re-check; TLV's page on three-party talks read in raw text) · Now reads "to the state and the regions".
- 2026-11-17 · Regional Cancer Centres in Collaboration (RCC i samverkan): The record said the 148,441 care pathways completed in 2025 were "across 27 diagnoses". The annual report gives 27 as the number of pathways measured against the inclusion goal, of which 16 met it, and does not tie the total to 27 diagnoses. (Institution record re-check; RCC's annual report on standardised care pathways for 2025 (PDF) read in raw text) · Reworded to say what the report says; RCC's own screening pages were added as sources for the screening programmes the record names.
- 2026-11-17 · Ethiopian Food and Drug Authority (EFDA): The record gave the authority's earlier full name as a fact from its sources. The pages it cites give only the abbreviation: the reliance guideline calls it the "former EFMHACA". (Institution record re-check; EFDA's home page, about page and reliance guideline (PDF) read in raw text) · The summary now says the cited pages give only the abbreviation.
- 2026-11-17 · Addis Ababa City Cancer Registry: The record said IARC applies the registry's rates to the whole country but cited no page that says so. (Institution record re-check; IARC's GLOBOCAN 2024 fact sheet for Ethiopia (PDF) read in raw text) · The claim stands ("Registry (subnational) ... Subnational rates applied to 2024 population") and the fact sheet is now cited on the record.
- 2026-11-17 · Federal Agency for Medicines and Health Products (FAMHP): The record said the agency's pages could not be re-read on 17 November 2026 because of a bot challenge. They answered, and the decree, the two programmes, the dashboard, the 654 staff, the chief executive and the annual report date were all confirmed. (Institution record re-check; three FAMHP pages read in raw text) · The sentence now says the pages were read again on 17 November 2026.
- 2026-11-17 · Democratic Republic of the Congo: lede and page description: The description said a public radiotherapy centre was "being financed", although the OPEC Fund page names only a "Kinshasa Cancer Center Project" and does not describe the facility. The lede said people were displaced "by conflict", which UNHCR's figures do not state, and gave the absence of a registry in the present tense from a 2021 review (Second accuracy pass; the OPEC Fund project page, UNHCR's population API and the imPACT Review summary report read in raw text) · The description now gives the 2023 foundation stone and the 2026 loan as two separate facts; the lede dates the registry finding to 2021 and says UNHCR records millions displaced from their homes
- 2026-11-17 · Democratic Republic of the Congo: About 70,000 new cancers a year in IARC's model: The card said the numbers were modelled "because the country has no cancer registry", crediting that to a fact sheet which only leaves its country data source blank (Second accuracy pass; IARC's GLOBOCAN 2024 fact sheet read as PDF text) · The card now says the fact sheet names no Congolese data source
- 2026-11-17 · Democratic Republic of the Congo: Where care is: The card said many cancer medicines were not on the essential medicines list and counted one surgical oncologist. The review says medicines "are not all part of the essential medicines list" and mentions "a surgical oncologist" without giving a count (Second accuracy pass; the IAEA, WHO and IARC imPACT Review summary report read as PDF text) · Reworded to "not all cancer medicines" and "a surgical oncologist"; the Ministry record carried the same medicines wording and was corrected too
- 2026-11-17 · Democratic Republic of the Congo: No population-based or hospital-based registry: The card said the country "has never had" a population-based cancer registry. The review says only that it did not have one in 2021 (Second accuracy pass; the imPACT Review summary report read as PDF text) · The sentence now reports the 2021 finding, and the table caption is dated the same way
- 2026-11-17 · Democratic Republic of the Congo: Millions of people displaced inside the country: The card stated that cancer services compete with epidemics for staff and money and that displaced people are far from the places that treat cancer. No cited source says either (Second accuracy pass; UNHCR's population API, the ACP dispatch of 6 January 2026 and the imPACT Review read in raw text) · Replaced with what the sources say: the outbreaks ACP listed, the review's finding that Ministry departments were redeployed during COVID-19, and a statement that no source read measures the effect of displacement on cancer care
- 2026-11-17 · Democratic Republic of the Congo: Universal health coverage began with childbirth: The half fee for children with cancer at the Cliniques Universitaires de Kinshasa was written as a fact reported by Radio Okapi. It is a statement by the national cancer centre's director that Radio Okapi quoted (Second accuracy pass; the Radio Okapi report of 19 February 2026 read in raw text on allAfrica) · The sentence now attributes the statement to the director
- 2026-11-17 · Democratic Republic of the Congo: A regulator since 2020: The card said ACOREP replaced a ministry directorate. The source says the directorate's equipment and infrastructure became ACOREP's assets (Second accuracy pass; NIAID ClinRegs' country profile read in raw text) · Reworded to say ACOREP took over the directorate's equipment and infrastructure
- 2026-11-17 · Democratic Republic of the Congo: One private radiotherapy centre since 2020: The card credited the IAEA with saying the only radiotherapy centre was in Kinshasa. The IAEA article gives no city (Second accuracy pass; the IAEA article of 22 February 2022 read in raw text through the Internet Archive) · The location was removed from the sentence credited to the IAEA
- 2026-11-17 · Democratic Republic of the Congo: Training capacity exists: The title and text said oncology was not a specialty and that the country had one surgical oncologist in 2021. The review recommends making medical oncology an independent specialty and mentions "a surgical oncologist" without a count; it also uses the abbreviations CUKs, CULs and CUKIs without spelling them out (Second accuracy pass; the imPACT Review summary report read as PDF text) · The title and text now report the recommendation, say the review gives no count of oncologists, and mark the hospital names as OnCo's reading of the abbreviations
- 2026-11-17 · Democratic Republic of the Congo: Patient groups and hospitals read only through secondary pages: The gap card said OnCo could not reach a website belonging to the patient groups. The Congo Cancer Foundation's own site, whose address is on its UICC member page, answered (Second accuracy pass; cancer.cd and the archived UICC member page read in raw text) · The card now says the site answers, summarises its home page and links it
- 2026-11-17 · Centre National de Lutte contre le Cancer (CNLC), Democratic Republic of the Congo: The record said the centre was to write the country's "first" national cancer plan and listed the plan as "in preparation". The review says only that the centre was created to develop a National Cancer Control Plan (Second accuracy pass; the imPACT Review summary report read as PDF text) · "First" was removed and the programme line now says the plan has been its task since 2020 and none was found published
- 2026-11-17 · Cliniques Universitaires de Kinshasa, Cliniques Universitaires de Lubumbashi: The French Africa Pediatric Oncology Group's protocols were called trials. ClinicalTrials.gov registers all of them as observational studies (Second accuracy pass; ClinicalTrials.gov API records for NCT04425421, NCT04425434, NCT04423484 and NCT03803735 read as JSON) · "Trials" changed to "studies"
- 2026-11-17 · Cervical precancer and HPV types among women in Kinshasa: The plain-language line said the finding is why WHO recommends earlier and more frequent screening for women living with HIV. The paper's abstract does not mention WHO recommendations (Second accuracy pass; the abstract for DOI 10.1186/s13027-015-0015-z read through Europe PMC) · The line now states only what the study found
- 2026-11-17 · Screening for hypertension, diabetes and cervical cancer among women in the 2023 to 2024 DRC Demographic and Health Survey: The record said women had "ever" been screened and compared the figure with a WHO target of 70%. The abstract says women "were screened" and names no target (Second accuracy pass; the abstract for DOI 10.1038/s41598-026-62712-y read through Europe PMC) · "Ever" and the WHO target were removed; the line now gives the authors' own recommendation
- 2026-11-17 · Cervical screening by visual inspection photographed on a smartphone, Monkole Hospital, Kinshasa: The plain-language line said the method is valid where HPV testing is not yet available. The abstract concludes it is valid and affordable for screening programmes in low-income areas (Second accuracy pass; the abstract for DOI 10.1136/ijgc-2022-003592 read through Europe PMC) · Reworded to the authors' conclusion
- 2026-11-17 · Quality of life of people with cancer in Lubumbashi: The comparison group was called Spanish patients and the difference "much higher". The abstract calls them matched Spanish controls and says symptom burden was "more pervasive" (Second accuracy pass; the abstract for DOI 10.1136/bmjopen-2025-114083 read through Europe PMC) · Changed to "controls" and to the abstract's wording
- 2026-11-17 · Tanzania: A trial needs national ethics clearance and TMDA authorisation within 60 working days: The card and the research institute's record said that foreign researchers need a research permit from the Tanzania Commission for Science and Technology, which implied local researchers do not. (Second accuracy pass; NIAID ClinRegs' Tanzania profile read in raw text, which says the Commission must issue permits for all research, for both foreign and local researchers.) · Both now say that researchers, foreign or local, need the permit.
- 2026-11-17 · Tanzania: Ocean Road Cancer Institute, the national cancer institute since 1996: The card listed cervical screening among the institute's services where its page says cancer screening, said the hospital became a radiotherapy hub in 1980 where the page records a decision that year, and credited a newspaper report of one official's remark to Zanzibar's health ministry. (Second accuracy pass; Ocean Road Cancer Institute's own pages and The Guardian (Tanzania) report of 17 January 2025 read in raw text.) · The service is given as cancer screening, 1980 is given as the year of the Ministry's decision, the two remarks are attributed to the newspaper reports that carry them, and those reports are linked.
- 2026-11-17 · Tanzania: Screening card on cervical cancer: A sentence about expanding screening in a phased, resource-sensitive way was written as a recommendation of the September 2025 review. It is a quotation from an IARC scientist in the IAEA's article. The profile card also credited a late-presentation figure to the review when it is background in the article. (Second accuracy pass; the IAEA's article of 10 December 2025 read in raw text from its Internet Archive copy.) · The quotation is attributed to IARC's Partha Basu in his own words, and the late-presentation figure to the article.
- 2026-11-17 · Tanzania: About 300 cancers a year recorded in Zanzibar: The card said radiotherapy is still on the mainland and that patients needing it travel to Dar es Salaam. No source read says that in those terms. It also wrote a newspaper report of an official's briefing as the ministry's own report. (Second accuracy pass; three reports in The Guardian (Tanzania) and the IAEA's December 2025 article read in raw text.) · The card now gives what the sources say: many patients seek treatment at Ocean Road, the health minister called radiotherapy difficult to access locally, and the IAEA lists Zanzibar among planned radiotherapy sites.
- 2026-11-17 · Tanzania: Radiotherapy in Dar es Salaam, Mwanza and, since February 2026, Moshi: The card and the hospital's record described newspaper previews as reports at the opening of the Kilimanjaro centre and said the President was reported to open it on 24 June 2026. Both articles were written before the event. The card also said Bugando now treats over 1,000 patients a year from an undated IAEA sheet. (Second accuracy pass; The Guardian (Tanzania) of 24 June 2026, The Citizen of June 2026 and the IAEA country sheet read in raw text.) · The opening is described as due on 24 June 2026, the reports as written before it, the Bugando figure is attributed to the IAEA without a date, and The Citizen report that gives the 25 February start date is linked.
- 2026-11-17 · Tanzania: A first national registry report in 2025, and the gaps it names itself: The card and the Dodoma hospital's record repeated the report's statement that 18% of Dodoma cases were diagnosed through death certificates without saying that the report's own table 12 records none. The registry table's note left out that the report's Mbeya section gives 1,619 cases and a 2017 start, and that its Mwanza section covers two districts, not three. (Second accuracy pass; the Ministry of Health's Tanzania Population-Based Cancer Registry Report 2019 to 2023 read in raw text.) · The contradictions inside the report are now stated on the card, in the table note and in the two hospital records.
- 2026-11-17 · Tanzania: The Universal Health Insurance Act of 2023: The card's title said compulsory cover has not started. The only source is the commencement notice of August 2024, which shows the compulsory clause was left out then and cannot show what has happened since. (Second accuracy pass; Government Notice 700A of 2024 read in raw text.) · The title now says the compulsory clause was left out when the law took effect in 2024.
- 2026-11-17 · Tanzania: profile note and gap card on national estimates: The page said the IAEA and the Ministry of Health cite about 40,000 new cases a year. The IAEA says over 40,000, and the Ministry's figure was read only in a newspaper. (Second accuracy pass; the IAEA's article of 10 December 2025 and The Guardian (Tanzania) of 24 June 2026 read in raw text.) · The two figures are now given separately, and the Ministry's is attributed to the newspaper report.
- 2026-11-17 · Bugando Medical Centre: The record and the paper record said 1,116 cases were recorded in 2016. The abstract gives 1,116 cases against an expectation adjusted to 2016 and does not say when they were recorded. The record also treated the registry in that paper and the hospital registry the Ministry's report dates to 2015 as one without saying so. (Second accuracy pass; the abstract of Olson et al. (Cancer Epidemiology 2020) read through Europe PMC and the Ministry's registry report read in raw text.) · The comparison is worded as the abstract gives it, and the two sources are reported separately.
- 2026-11-17 · National Health Insurance Fund, Tanzania: The summary line said members can be treated for cancer at the Aga Khan hospitals since September 2026. The Fund's page reports the signing of an agreement, not the start of treatment. (Second accuracy pass; the Fund's news page of 25 September 2026 read in raw text.) · The line now says an agreement announced in September 2026 lets members be treated there.
- 2026-11-17 · Zanzibar Health Services Fund: The summary line called the founding law an Act of the Revolutionary Government of Zanzibar. The source says only that the fund was established by the act in March 2023. (Second accuracy pass; GIZ's story of 25 September 2023 read in raw text.) · The line now says established by an Act in March 2023.
- 2026-11-17 · Kenya: Cervical screening reaches about 31% of women; breast screening became national in 2021: Said the 2023-2027 strategy will pilot and roll out HPV testing for cervical screening, when the strategy lists that piloting and roll-out among the achievements of the 2017-2022 strategy and plans only to expand HPV testing platforms; and credited a GLOBOCAN 2020 share of cancers to the 2018 screening guidelines when it is the strategy's own sentence (Second accuracy pass; the Ministry of Health's National Cancer Control Strategy 2023-2027 (WHO Africa copy) read in raw PDF text) · The card now says HPV testing was piloted and rolled out under the earlier strategy and that the new one plans to strengthen and expand it, and credits the 43.1% and 36.9% figures to the strategy
- 2026-11-17 · Kenya: The Pharmacy and Poisons Board registers medicines for five years, with full, reliance and fast-track routes: Said the Board publishes no target time for its own scientific review; its flow chart marks 24 months against first in, first out scientific review and 90 days for fast-tracked reliance, and only the assessor's own review time has no timeline (Second accuracy pass; the Board's marketing authorisation flow chart read in raw PDF text and as a rendered page) · The card, the gap card and the Pharmacy and Poisons Board record now give the 24 months and limit the missing timeline to the assessor's review time
- 2026-11-17 · Kenya: The National Cancer Institute inspects and certifies cancer centres, and only certified centres are paid: Said the 2012 Act gives the Institute the job of setting standards and designating institutions, which is not among the functions in section 5; and said the Social Health Authority has contracted only certified hospitals since October 2026, when the joint notice of that month states the rule without saying it was new and the tariff page already required certification (Second accuracy pass; the Cancer Prevention and Control Act, 2012, the joint notice and the Authority's benefit tariffs page read in raw text) · The card now gives the functions the Act lists and attributes inspection and accreditation to the Institute's own description of its directorate; the card, the certify-then-pay card and the National Cancer Institute of Kenya record now say the two bodies stated the rule in October 2026, to facilities intending to contract
- 2026-11-17 · Kenya: Sources that refused a script or could not be read: Said IARC's Cancer Incidence in Five Continents holds the Nairobi and Eldoret registries' 2013 to 2017 data and could not be read, and that Kenya Law refused every request; the volume's summary download is readable and gives Nairobi for 2013 to 2016 and Eldoret for 2014 to 2017, and Kenya Law's legislation index now answers, though its document pages still return a browser challenge. The page also called the registries' 2004 to 2008 rates their latest published figures (Second accuracy pass; the registry list and case file in IARC's Volume XII summary download, and the first page of Kenya Law's legislation index, read in raw text) · The gap cards, the registry table, the GLOBOCAN note and the verify text now give the Volume XII periods and all-sites case counts (9,675 for Nairobi, 2,335 for Eldoret), say that rates were not worked out from them, and record that the index lists two sets of rules made under the Cancer Prevention and Control Act in 2026 whose texts were not read
- 2026-11-17 · Kenya: HPV vaccination since 2019, now one dose for girls aged 10 to 14, with coverage rising from 25% to 79%: Credited the coverage estimates to WHO and UNICEF; the cited data service names only a WHO indicator (Second accuracy pass; WHO Global Health Observatory API response and indicator definition read as raw JSON) · Attribution changed to WHO
- 2026-11-17 · Gladys Chesumbai: Used a gendered pronoun that no source read supports, said Chesumbai had long represented the Eldoret registry, and gave a Moi University affiliation that the papers do not state (Second accuracy pass; the National Cancer Institute of Kenya's conference report and the Europe PMC author affiliations read in raw text) · Rewritten without a pronoun and with the affiliation the papers give, the Eldoret Cancer Registry at Moi Teaching and Referral Hospital
- 2026-11-17 · Myanmar: page description and "what to verify": The description said IARC's 2024 estimate rests on the two regional registries, and the verify note said the figures rest on two regional registries. IARC's fact sheet names one unnamed subnational registry as its incidence source. (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet for Myanmar read as PDF text, which gives "Country-specific data source: Registry (subnational)".) · Both passages now say the estimate is built from subnational registry data and that two regional registries count cancers directly.
- 2026-11-17 · Myanmar: HPV vaccination cards: The page said girls had been vaccinated with a single dose since 2020. WHO's data give 2020 as the year of introduction and one dose as the current schedule, and do not say the schedule was one dose from the start; a 2023 commentary cites coverage for first and second doses. (Second accuracy pass; WHO immunization data API (MT_HPV) read as JSON and the Thinn commentary read in full text on Europe PMC.) · The description and three cards now say vaccination began in 2020 and that WHO records a one-dose schedule, without dating the single dose to 2020. The authors of the commentary are no longer described as involved in the programme, which the paper does not say.
- 2026-11-17 · Myanmar: Lung, liver and breast cancer lead: The card divided the registry's cancers into a tobacco group and an infection group and credited that split to the registry's authors. Their lists overlap, with liver and cervical cancer on both. (Second accuracy pass; Shwe and colleagues, APJCP 2022, read in full text on Europe PMC.) · The card now gives the authors' own figures: tobacco-related cancers 70.6% of men's cases and 48.1% of women's, infection-related cancers 41.9% of all cases.
- 2026-11-17 · Myanmar: Two regional registries and no national count: The card said the Nay Pyi Taw registry began on the advice of an international review. The review visited in 2015 and found registration already being developed there; the registry's data start in 2013. The Yangon report was also said to be quoted and prepared with Vital Strategies, where the review only cites it as published by Vital Strategies and partners. (Second accuracy pass; Shwe and colleagues, APJCP 2022, and Win and colleagues, J Epidemiol Glob Health 2025, read in full text on Europe PMC.) · The card, the related card on the registry and the registry record now give the 2015 date and what the review found and recommended, and describe the Yangon report as cited.
- 2026-11-17 · Myanmar: Conflict, displacement and an earthquake: The card gave the health cluster's figure as 9.2 million people in need on June 2026 data; the cluster's page now gives 9.3 million on September 2026 data. OCHA's page could not be re-read, so its figures were unconfirmed. (Second accuracy pass; WHO Health Cluster page for Myanmar read as raw HTML; OCHA's Myanmar page answered a scripted request with a bot challenge.) · The health cluster figure is corrected, and the card and the lede now say the OCHA figures are as read on 11 November 2026 and were not re-verified on 17 November 2026.
- 2026-11-17 · Myanmar: Attacks on health workers and facilities: The plain text said two databases recorded hundreds of attacks, when one recorded 443 and the other 1,492. The card also said most health service provision stopped after the coup, a finding the source makes about providers outside the state system. (Second accuracy pass; Usmany and colleagues, Popul Health Metr 2025, and Than and colleagues, Int J Equity Health 2024, read in full text on Europe PMC.) · The plain text gives both counts and the sentence is limited to non-state providers.
- 2026-11-17 · Myanmar: Screening by testing for the virus, begun as a pilot: The card said IARC's CanScreen5 records a national programme and that its pages could not be read. The fact sheet does not call the programme national, and its text can be read. It also records no breast screening policy or protocol as of 2025, which the page did not say. (Second accuracy pass; the CanScreen5 cervix and breast fact sheet text files for Myanmar read directly from the site.) · The word national is removed, the breast statement is reported beside WHO's 2019 entry, and the two gap cards are updated.
- 2026-11-17 · Myanmar: What WHO recorded about cancer services in 2019: The card said the profile was built from WHO's 2019 country capacity survey. The profile does not name its survey. (Second accuracy pass; WHO Cancer Country Profile 2020 for Myanmar read as PDF text.) · The clause is removed and the items are given with the year the profile prints. A quotation from the same profile on another card now matches its capital letter.
- 2026-11-17 · Myanmar: Who paid for health care, 2019 to 2023: The table note said the shares do not add to 100 because voluntary prepaid schemes and other private spending are not shown. The data source gives no breakdown of the remainder. (Second accuracy pass; World Bank API indicator series for Myanmar read as JSON.) · The note now says other sources of health spending are not shown.
- 2026-11-17 · Myanmar: Care given outside the state system: The card described a growing convergence with ethnic health systems from 2015 to 2020. The study describes the start of a convergence that stayed largely unfinished. (Second accuracy pass; Than and colleagues, Int J Equity Health 2024, read in full text on Europe PMC.) · The card now says the start of a convergence in the years before 2021.
- 2026-11-17 · Assessing the role of non-state actors in health service delivery and health system resilience in Myanmar: The record said the interviews and focus groups were mostly collected in 2022 and called the study qualitative. The abstract says mostly qualitative data were collected in 2022. (Second accuracy pass; the paper's abstract and full text read on Europe PMC.) · The summary and caveat now say the data, mostly qualitative, were collected in 2022.
- 2026-11-17 · Nay Pyi Taw Cancer Registry (population-based), Myanmar: The record said cases were also collected from Yangon and Mandalay General Hospitals and other centres outside the territory. The paper names those two hospitals and the two cities' children's hospitals. (Second accuracy pass; Shwe and colleagues, APJCP 2022, read in full text on Europe PMC.) · The record names the four hospitals and gives 2015 as the year of the international review.
- 2026-11-17 · Sudan: A National Cancer Registry since 2009: The page, its description and its incidence note said as fact that Khartoum's 2009 to 2010 figures were the only population-based figures Sudan had published, and that they were much higher on the world standard IARC uses. No source says nothing else was published, the Ministry's strategy describes a second population-based registry for Gezira State, and the registry paper does not say which standard IARC uses. (Second accuracy pass; the registry paper (Saeed and colleagues, Cancer Medicine 2014) read in full through Europe PMC and the National Cancer Strategy 2012 to 2016 read as PDF text.) · Reworded to say these are the only figures OnCo found, that they are higher on the 1966 world standard and about fifteen years older, and added the 2022 paper's statement that Sudan had no functioning national cancer registry.
- 2026-11-17 · Sudan: New cancer patients at centres outside Khartoum: The table said its counts were reported by the centres to the editorial's authors and that Khartoum's centres saw most patients before the war. The editorial says neither; it gives the counts without saying how they were gathered and calls two of them averages. A card also said, citing the editorial, that public centres treated without charge, where the editorial says as part of universal health coverage. (Second accuracy pass; the editorial by Alrawa and colleagues (ecancermedicalscience 2023) read in full through Europe PMC.) · Caption and note reworded to what the editorial says, the unsupported clause removed, and the card now uses the editorial's own wording.
- 2026-11-17 · Sudan: No organised cancer screening, and no HPV vaccine: The card gave 82% of women at Gedaref diagnosed at stage III or IV, but that study's 394 patients included men (96% were women). It also said screening mammography was not linked to treatment, where the strategy says not necessarily linked. (Second accuracy pass; the abstract and full text of Ahmed and colleagues (ecancermedicalscience 2024) through Europe PMC and the strategy's PDF text.) · Changed to patients and restored the strategy's qualifier.
- 2026-11-17 · Sudan: Hepatitis B vaccine for infants since 2004: The card said coverage was above 90% before 2020. WHO and UNICEF estimates were above 90% only from 2011 to 2019, and 75% in 2010. (Second accuracy pass; WHO immunisation coverage rows for Sudan read as JSON from the public API.) · Now says above 90% from 2011 to 2019.
- 2026-11-17 · Sudan: One board registers medicines, sets prices and licenses clinical trials: The card and the regulator's record said the Board runs a pharmacovigilance centre. Its pages show a medicine safety section and requirements for good pharmacovigilance practice, and name no centre. (Second accuracy pass; the National Medicines and Poisons Board's Arabic home and about pages read as raw HTML.) · Reworded to what the pages show.
- 2026-11-17 · Sudan: A health system under attack: The card said WHO's 2025 situation analysis warned that many oncology patients arrived at advanced stages. That sentence in the analysis is about patients with non-communicable diseases in general. (Second accuracy pass; WHO's public health situation analysis of 10 March 2025 read as PDF text.) · Qualifier added.
- 2026-11-17 · National Cancer Registry, Sudan: The record said the registry was established under the Ministry's non-communicable diseases directorate, and the Ministry's record said that directorate runs it. The strategy places only the cancer control programme in that directorate; the registry paper says the registry is under the auspices of the Ministry. (Second accuracy pass; the National Cancer Strategy 2012 to 2016 as PDF text and the registry paper in full through Europe PMC.) · Both records now say under the auspices of the Federal Ministry of Health.
- 2026-11-17 · Cancer incidence in Khartoum State, 2009 to 2010: A caveat said both biases the authors name would raise Khartoum's rates. The authors say patients seeking alternative treatment go uncounted, and only out-of-state patients would raise the rates. (Second accuracy pass; the paper read in full through Europe PMC.) · Caveat corrected.
- 2026-11-17 · East Oncology Centre, Gedaref: The record said the centre treats patients from Gedaref State and the east of the country, which its source does not say, and the linked paper's note spoke of women where the study counted patients. (Second accuracy pass; Ahmed and colleagues (ecancermedicalscience 2024) read in full through Europe PMC.) · Now says it was established in 2016 and is described as the main cancer treatment centre in its region; the paper's note says patients.
- 2026-11-17 · Collapse of cancer care under the conflict in Sudan: The record said the paper documents the loss of the two main centres in the first year of the war; the abstract gives no such timing. A card also named St Jude Children's Research Hospital where the abstract says St Jude Hospital. (Second accuracy pass; the abstract read from Europe PMC (the publisher's page returned a bot challenge).) · Timing removed and the abstract's own name used.
- 2026-11-17 · Uganda: A 2026 Act to replace the 1993 law, with compassionate use written in: The card, titled "A 1993 law replaced in 2026", and the gap card beside it said the old medicines law had been replaced and that the text of the new Act had not been read. The Act had been assented to but takes effect only on a date the Minister appoints. (Second accuracy pass; the National Drug and Health Products Authority Act, 2026 was read in raw text from the PDF on Parliament's bill tracker. Section 1 reads "This Act shall come into force on a date appointed by the Minister, by statutory instrument.") · The title, plain text, detail, the verification note and the gap card now say the Act was assented to on 29 April 2026 and comes into force on a date the Minister appoints, that OnCo found no commencement instrument, and what section 21 allows for unregistered drugs.
- 2026-11-17 · National Drug Authority, Uganda: The record said a 2026 law "replaces" the authority, without saying the Act is not in force until the Minister appoints a date. (Second accuracy pass; the National Drug and Health Products Authority Act, 2026 read in raw text (sections 1, 21 and 119).) · The summary now gives the commencement rule, says no commencement instrument was found, and describes section 21 in the Act's own terms.
- 2026-11-17 · Ministry of Health, Uganda: The record said the 2026 Act had replaced the 1993 law, that the Uganda Cancer Institute coordinates the national cancer plan, and that the institute works under the Minister's written directions. (Second accuracy pass; the 2026 Act, the plan slides on the ICCP portal and the Uganda Cancer Institute Act 2016 read in raw text. The slides give coordination to a national secretariat "with technical guidance of Uganda Cancer Institute", and the 2016 Act says the Minister "may give policy directions".) · Reworded to the commencement rule, to the secretariat under the institute's technical guidance, and to directions the Minister may give.
- 2026-11-17 · Uganda: Measuring how HIV, and HIV treatment, change a country's cancers: The card said Kaposi sarcoma has fallen since antiretroviral drugs became available, and the profile card said it declined as those drugs arrived. The registry papers tie the fall to falling HIV prevalence and say rates in men were steady after about 2002. (Second accuracy pass; the abstracts of Bukirwa et al. 2021 and Wabinga et al. 2014 read in raw text from Europe PMC.) · Both cards now say the decline is consistent with falling HIV prevalence, and the link between antiretroviral therapy and the fall in non-Hodgkin lymphoma is given as the authors' "possibly".
- 2026-11-17 · Trends in cancer incidence in Kampala, Uganda, 1991 to 2010: The one-line summary said Kaposi sarcoma fell as antiretroviral drugs arrived, which the abstract does not say. (Second accuracy pass; the abstract read in raw text from Europe PMC.) · Now says Kaposi sarcoma in men fell until about 2002, over years in which HIV prevalence dropped.
- 2026-11-17 · Uganda: A public cancer institute built in partnership with a US research centre: The card said the institute had worked with Fred Hutch since 2004 and that Fred Hutch had trained more than a dozen Ugandan physician-scientists since 2004. Fred Hutch's page dates only its training of African medical professionals to 2004. (Second accuracy pass; Fred Hutch's global oncology page read in raw text.) · Reworded to what the page says, without a start year for the partnership.
- 2026-11-17 · Uganda: In 2016 Uganda's only working radiotherapy machine broke down; it now has eight: The card applied a five-year timescale to the plan for two linear accelerators at each regional centre. The institute gave five years only for the five machines at Mulago. (Second accuracy pass; the Daily Monitor report of 6 October 2026 read in raw text.) · The five-year timescale now applies to Mulago only.
- 2026-11-17 · Uganda: Other figures not found or not verified: The gap card said Hospice Africa Uganda's website could not be reached and repeated, without a source, that trained nurses prescribe oral morphine. (Second accuracy pass; the charity's website answered and was read in raw text.) · The card now reports what the site says about its three sites and its morphine production in Kampala, and the unsourced statement is removed.
- 2026-11-17 · Uganda: opening summary and the note on IARC's estimate: The summary said most people pay without insurance, and the note said as fact that Uganda has two population-based registries and that a city with easier diagnosis records more cancer. The sources give an insurance figure, a list of two member registries and a statement that diagnostic access may influence rates. (Second accuracy pass; the Daily Monitor report of 4 November 2026 and the African Cancer Registry Network pages read in raw text.) · Reworded to each source's own claim, with OnCo's inference labelled as one.
- 2026-11-17 · Projected incidence of the five commonest cancers in Kampala, Uganda, to 2030: The one-line summary said cases would rise by a third to a half when the paper's range reaches 57.7%, and a caveat said the text and table disagree on prostate cancer's starting rate when the text uses the table's own 2001 to 2005 value. (Second accuracy pass; the full text read in raw form from Europe PMC (tables 1 and 2).) · The summary now gives 33% to 58% and the caveat is corrected.
- 2026-11-17 · Iraq: A national registry since 1976, run by the Iraqi Cancer Board, and Kurdistan's own registry data: The card and the registry table gave the Kurdistan Region study's incidence as rising from 73 to 174 per 100,000 women and from 36 to 85 per 100,000 men. Those are the both-sexes rates for Erbil and for Duhok; the paper's abstract mislabels them. The card also credited the CanReg5 software to IARC, which the registry report does not do. (Second accuracy pass; the full text of M-Amen et al. (Asian Pac J Cancer Prev 2022) read as raw text from Europe PMC, table 1 and results, and the Cancer Registry of Iraq annual report 2022 read from the Ministry of Health PDF.) · The card now gives 73 to 174 per 100,000 people in Erbil and 36 to 85 in Duhok and says the abstract differs; the table has one 2019 row for each governorate with the rates by sex from table 1 (Erbil 192 women, 157 men; Duhok 93 women, 77 men); the software is named without attribution.
- 2026-11-17 · Cancer incidence in Erbil and Duhok, Kurdistan Region of Iraq, 2013 to 2019: The first finding repeated the abstract's labelling of 73 to 174 and 36 to 85 per 100,000 as rates in women and men, when the paper's own table shows them to be the rates for Erbil and Duhok. The summary line called the incidence crude, which the paper does not, and said the rise also reflects better registration, which the paper does not say. (Second accuracy pass; the paper's full text read as raw text from Europe PMC.) · The finding now gives the rates by governorate and by sex from table 1, a caveat records that the abstract differs, the word crude is removed, and the explanation is limited to the one the authors give: displaced people arriving in Erbil in 2014 as a plausible reason for the sharp rise between 2013 and 2014.
- 2026-11-17 · Iraq: The 2020 health insurance law exempts people with cancer from every fee and co-payment: The card said the scheme began in Baghdad at the end of 2023 and is being extended province by province, and gave the 2.5 million figure as the health minister's own statement. The source says the law came into effect at the end of 2023 and is being rolled out in four phases, the first covering Baghdad, and it reports the minister's words second hand from the state news agency. (Second accuracy pass; the Rudaw report of 14 November 2025 read as raw text.) · Both statements now follow Rudaw's wording and name Rudaw and the state news agency as the route. The same change was made on the Health Insurance Authority record.
- 2026-11-17 · Iraq: Cancer written into the insurance law as a reason to pay nothing: The card said Rudaw listed 285 types of surgery, 40 types of consultation and all medical tests as covered. Rudaw says they will be covered. (Second accuracy pass; the Rudaw report of 14 November 2025 read as raw text.) · The sentence now says they will be covered.
- 2026-11-17 · Iraq: Broken equipment, missing guidelines and too few trained staff, in a 2021 review and its 2023 follow-up: The plain summary said the review found no national treatment guidelines. The IAEA's words are a lack of cancer treatment guidelines. (Second accuracy pass; the IAEA report of 30 March 2023 read as raw text from its Internet Archive copy.) · Reworded to a lack of cancer treatment guidelines.
- 2026-11-17 · Iraq: Oncology hospitals in Baghdad and radiotherapy centres in the Kurdistan Region: The title and summary called all three centres public. Only Zhianawa Cancer Center is described as public in its source; the IAEA report does not say it of Al-Amal National Oncology Hospital, nor the UICC profile of Rizgary Oncology Center. (Second accuracy pass; the IAEA report of 7 February 2022 and both UICC member profiles read as raw text from their Internet Archive copies.) · The word public now applies only to Zhianawa, as its own description. The same word was removed from the Al-Amal National Oncology Hospital record.
- 2026-11-17 · Directorate of Technical Affairs, Ministry of Health, Iraq (medicines regulator): The record said the regulator tests every batch, with exemptions for products holding FDA, EMA or United Kingdom certificates. The report says its laboratory tests the quality of medicines before registration and release, and that companies holding any of the recognised certificates are exempt from batch testing; it does not say every batch is tested. (Second accuracy pass; Al-Jumaili et al. (East Mediterr Health J 2021) read as raw text from the journal PDF.) · Reworded to what the report says.
- 2026-11-17 · KIMADIA, State Company for Marketing Drugs and Medical Appliances (Iraq): A programme line said KIMADIA clears customs for public-sector medicines. The report says its Customs Clearance Department and the Ministry of Trade regulate pharmaceuticals at the federal level for both the public and private sectors. (Second accuracy pass; Al-Jumaili et al. (East Mediterr Health J 2021) read as raw text from the journal PDF.) · The line now reads customs clearance of medicines at the federal level, with the Ministry of Trade.
- 2026-11-17 · Pharmaceutical regulation in Iraq, from medicine approval to post-marketing: A caveat stated that the rules have changed since 2020. No source read for the page says so. (Second accuracy pass; the paper read as raw text from the journal PDF.) · The caveat now says the rules may have changed since.
- 2026-11-17 · Tahseen Al-Rubai: The summary line said he is General Secretary of the Iraqi Cancer Board, in the present tense, on the strength of the Board's report for 2022. (Second accuracy pass; the Cancer Registry of Iraq annual report 2022 read as raw text from the Ministry of Health PDF.) · The line now says he is listed as General Secretary in the annual report for 2022.
- 2026-11-17 · Iraq: page introduction: The introduction stated as plain fact that chemotherapy and radiotherapy are free in government hospitals. The only source for it is a newspaper report. (Second accuracy pass; The National's report of 4 August 2023 read as raw text.) · The sentence now says a 2023 news report said so.
- 2026-11-17 · Afghanistan: Households pay about three quarters of health spending: The card said a 2026 assessment found limited access to essential cancer medicines. The assessment says essential medicines, without the word cancer. (Second accuracy pass; the abstract of Abdullah et al. (Int J Cancer 2026, doi 10.1002/ijc.70431) read in raw text from Europe PMC and Crossref.) · Reworded to say that the assessment of cancer control found limited access to essential medicines.
- 2026-11-17 · Afghanistan: No national plan, no radiotherapy, almost no palliative care: The card said the questionnaire survey and the desk and literature review followed the 2021 site visits. The source says they complemented the visits and continued to January 2025, and gives no order. (Second accuracy pass; the abstract of Abdullah et al. (Int J Cancer 2026, doi 10.1002/ijc.70431) read in raw text from Europe PMC and Crossref.) · Reworded to the source's own description.
- 2026-11-17 · Afghanistan: No trial in OnCo's records has a site in Afghanistan: The card spoke of studies of oesophageal and gastric cancer at a Kabul referral centre. There is one paper, a retrospective case series. (Second accuracy pass; Europe PMC search results read as raw JSON (doi 10.1186/s12885-026-15945-z).) · Reworded to a case series.
- 2026-11-17 · Yemen: Who counts cancer: The section heading said Yemen has two regional registries, and the Hadhramout card and the table note said the Hadhramout registry gives no rates. The Aden registry's own report counts four registries in Yemen (Aden, the National Oncology Center, Hadhramout and Taiz, the last three run as hospital registries), and the Hadhramout paper does calculate a childhood incidence of 1.9 per 100,000. (Second accuracy pass; the open full text of Bawazir (BMC Cancer 2018) and Jawass and colleagues (Saudi Medical Journal 2016) read as raw text through Europe PMC.) · Heading, both cards and the table note now give the four registries as the report lists them, attribute the hospital-registry description to that report, and say the Hadhramout paper gives a childhood rate but no all-cancer rate.
- 2026-11-17 · Yemen: The Aden Cancer Registry: The card said the registry counted 6,974 cancers and was set up by Aden University's cancer centre. The report says 7,413 records were reported and the 6,974 complete ones were analysed, and that the registry is supervised by the centre. (Second accuracy pass; the open full text of Bawazir (BMC Cancer 2018) read as raw text through Europe PMC.) · The card, the table note, the registry record and the paper record now give both numbers and say supervised.
- 2026-11-17 · Yemen: No organised screening programme: The 2025 survey of 996 people was presented without saying how they were recruited, and the card said WHO delivered two mammography machines in 2019. The survey was an online questionnaire shared on social media, 42.0% of respondents worked in health care, and its authors say the sample is younger, more educated and more urban than Yemen as a whole; WHO's page of 4 February 2019 says the machines had been delivered and gives no year. (Second accuracy pass; the open full text of Al-Jamei and colleagues (Conflict and Health 2025) through Europe PMC and the raw text of WHO's page of 4 February 2019.) · The card and the paper record now describe the sample and its limits, call the survey published in 2025, and report the mammography delivery as WHO stated it.
- 2026-11-17 · Yemen: Patients travel far, and many cannot complete treatment: Two travel costs were written as what WHO described families paying; each is one relative's own account. A sentence about treatment being available in about 20% of health facilities was credited to WHO's representative when it is the release's own text. (Second accuracy pass; raw text of WHO's page of 4 February 2019 and an Internet Archive copy of WHO's release of 4 February 2018 on ReliefWeb, which refused a direct request.) · The costs are now given as a mother's and a son's accounts, and the 20% figure is credited to the release and names the diseases it covers.
- 2026-11-17 · Yemen: Cancer medicines supplied by WHO and donors: The card said WHO has supplied the National Oncology Centre since 2018, in the present tense, and that WHO said each time that supplies cover only acute shortages. The newest WHO page read is from September 2020, and the acute-shortage wording appears only in the November 2018 and February 2019 pages. The page description had the same present tense. (Second accuracy pass; raw text of WHO's pages of 20 November 2018, 4 February 2019 and 20 September 2020.) · The card and the description now give the period 2018 to 2020 and say which WHO page said what.
- 2026-11-17 · Yemen: A board chaired by the health minister: The card title said the medicines board had not yet been assessed by WHO, and the card said medicines are registered and controlled by the board. WHO's list shows only which regulators reached maturity level 3 or 4, and none of the sources read sets out the board's functions. (Second accuracy pass; text of WHO's list of regulators at maturity level 3 and 4 (August 2026, PDF), the Cairo 24 report of 22 January 2023 in Arabic, and the Yemen Monitor report of July 2026.) · The title now says the board is not on WHO's list, and the card says the sources read do not set out its powers.
- 2026-11-17 · Supreme Board of Drugs and Medical Appliances (Yemen): The record said the board regulates medicines and listed registration and import control as its programme, which no source read for the record states. (Second accuracy pass; the Cairo 24 report of 22 January 2023, the Yemen Monitor report of July 2026 and the abstract of Al-Worafi (Australasian Medical Journal 2014) read as raw text.) · The record now says the sources do not set out the board's legal powers, and lists the sourced 2023 memorandum with Egypt's Drug Authority instead.
- 2026-11-17 · Yemen: Radiotherapy: The card title said radiotherapy is concentrated in Sana'a and that Aden's centre is still being equipped, and the card placed the IAEA-trained National Oncology Center in Sana'a. No source read says where the machines are; Saba reported a meeting that discussed providing the Aden centre's requirements; and the IAEA page names no city. The link labels also said Saba's pages refused a request, and both now answer. (Second accuracy pass; text of the IAEA's country page for Yemen (PDF) and the raw text of Saba's items of 21 January 2024 and 28 October 2025.) · The title now gives WHO's ratio and the planned unit in Aden, the IAEA sentence says the page names no city, and the Saba labels no longer say the pages are unreadable.
- 2026-11-17 · Yemen: Cancer charities that run centres: The card opened by saying much of Yemen's cancer care is run or paid for by Yemeni charities, which no source read measures. The details of the Shabwah Cancer Control Foundation could not be re-read. (Second accuracy pass; UICC's site refused requests, so its member page and 2021 article were read in Internet Archive copies; no copy of the Shabwah page exists.) · The opening now says charities run and pay for some cancer services, and the Shabwah details are marked as not re-verified on 17 November 2026.
- 2026-11-17 · Yemen: A national cancer strategy for 2025 to 2030: The card said the Minister said the Ministry would build cancer treatment centres with the governorates. The report says he stressed the importance of working with all governorates to establish them. (Second accuracy pass; raw text of the Yemen Monitor report of 31 December 2024.) · Reworded to what the report says, and the reported figure is given as 30,000 as the report has it.
- 2026-11-17 · National Oncology Centre, Sana'a: The record said the IAEA supported a move to three-dimensional radiotherapy planning; the IAEA page says radiotherapy treatment and does not name the centre's city. (Second accuracy pass; text of the IAEA's technical cooperation page for Yemen (PDF).) · The word planning is removed and the record says the IAEA page gives no city.
- 2026-11-17 · Public knowledge and uptake of cancer screening in Yemen: The record did not say the survey was run online through social media or that 42.0% of respondents worked in health care. (Second accuracy pass; the paper's open full text read through Europe PMC.) · The summary and the caveat now say so, in the authors' terms.
- 2026-11-17 · Cancer incidence in Aden and three neighbouring governorates, 1997 to 2011: The record said 6,974 cancers were recorded and called the series the longest in Yemen. The paper reports 7,413 records, of which 6,974 complete ones were analysed, and makes no such comparison. (Second accuracy pass; the paper's open full text read through Europe PMC.) · The finding gives both numbers and the series is described as a 15-year series.
- 2026-11-17 · Morocco: page description and summary: The summary said outright that Morocco has no national cancer registry, the description said cancer care in public centres is free, and it gave 45 linear accelerators with no date. (Second accuracy pass; IARC's GLOBOCAN 2024 fact sheet, the Hespress report of 15 July 2026 and the 2020 to 2029 national plan read in raw text. IARC says only that its source is subnational registries, the minister's word was "largement", and the accelerator count is the plan's own.) · The summary now says the figures do not come from a national registry, the description says care is largely free and dates the accelerator count to the 2020 plan.
- 2026-11-17 · Morocco: About 48,000 new cancers a year, breast cancer first: The card described nasopharyngeal cancer as common across North Africa. (Second accuracy pass; IARC's GLOBOCAN 2024 fact sheet for Morocco read in raw text. It gives 666 cases and makes no regional statement, and no other cited source does.) · The regional clause was removed; the count stays.
- 2026-11-17 · Morocco: Two city registries stand in for a national one: The card said Morocco counts cancer in two cities and that the Rabat report was written at the national oncology institute. The registry table labelled the plan's 2008 to 2012 rates as both sexes and said the plan compared them with a 2004 rate on a different standard. (Second accuracy pass; the Europe PMC record of Tazi and colleagues (2013) and the 2020 to 2029 national plan read in raw text. Only the first author's affiliation is recorded, no source says there are only two registries, and the plan gives rates per 100,000 inhabitants without naming a sex or the standard of the 2004 rate.) · The card now speaks of the two registry reports read for the page and of a first author at the institute; the table row and its note say what the plan states and no more.
- 2026-11-17 · Morocco: Cancer care in public centres is largely free: The card credited the statement on free care to the ministry, said the medicines budget was for public oncology centres, and said the first plan put standard preparation units in every regional centre. (Second accuracy pass; the Hespress report of 15 July 2026 and the 2020 to 2029 national plan read in raw text. The statement is the minister's as reported, the budget is for all oncology centres, and the plan speaks of standards for preparing medicines, not units.) · The statement is attributed to the minister as reported by Hespress, the budget wording follows the report, and the plan's sentence is given as standards.
- 2026-11-17 · Morocco: A new medicines agency replaced the ministry's directorate: The card said the House of Representatives adopted bill 10-22 in February 2023. (Second accuracy pass; the Le360 report of 1 March 2023 read in raw text. Its headline says the House adopted the bill the day before, while its text says the House's social sectors committee did, by majority.) · The card now reports both statements and attributes them to Le360.
- 2026-11-17 · Morocco: Breast and cervical early detection since 2010, free in public care: The card said women aged 40 to 69 are offered a breast examination every two years. (Second accuracy pass; the 2020 to 2029 national plan and the Fondation Lalla Salma's early detection page read in raw text. Both speak of breast screening and neither names the test.) · The card now says breast screening.
- 2026-11-17 · Morocco: HPV vaccine since October 2022, reaching about 3% of girls: The card's opening line said the vaccine was introduced for girls aged 11. (Second accuracy pass; the Le360 report of 19 September 2022 on the joint circular read in raw text. The circular recommends it for all girls aged 11 and over.) · The line now says girls from the age of 11.
- 2026-11-17 · Morocco: The insurers' and the medicines agency's own sites could not be read: The card said Internet Archive copies could not be fetched. (Second accuracy pass; the four sites and the Internet Archive were requested again on 17 November 2026. The sites still did not answer, but the Archive did, and its copy of the medicines agency's home page of 24 March 2025 was read in raw text.) · The card now records the second attempt, says what the archived home page is (a menu of services and lists) and that nothing behind it was read.
- 2026-11-17 · Morocco: No national registry found, no recent Rabat report, no counted deaths: The card stated as fact that Morocco has no national cancer registry and no reliable count of cancer deaths. (Second accuracy pass; IARC's GLOBOCAN 2024 fact sheet and WHO's 2020 cancer country profile read in raw text. IARC names subnational registries as its source and WHO rates death registration very low for 2007 to 2016; neither says no national registry exists.) · The card now says that no national registry was found and gives WHO's rating in WHO's terms.
- 2026-11-17 · Agence Marocaine du Médicament et des Produits de Santé (AMMPS): The record said bill 10-22 was adopted by the House's social sectors committee and then by the House in February 2023. (Second accuracy pass; the Le360 report of 1 March 2023 read in raw text. It describes one adoption the day before, by the House in its headline and by the committee in its text.) · The record now reports what the headline and the text each say.
- 2026-11-17 · Caisse Nationale de Sécurité Sociale (CNSS), Morocco: The record said the fund's count of 3.95 million people in AMO Tadamon matches the ministry's figure for principal insured persons and not its 11.1 million beneficiaries. (Second accuracy pass; the Hespress reports of 4 September and 7 November 2026 read in raw text. The fund's communiqué as reported says only that the scheme covered 3.95 million people.) · The record now gives both sets of figures and says the communiqué does not state what its count includes.
- 2026-11-17 · Centre Mohammed VI pour le traitement des cancers, CHU Ibn Rochd, Casablanca: The short description said the centre hosts a gynaecological and breast oncology centre of excellence and that Casablanca has one of the country's two paediatric haemato-oncology centres. (Second accuracy pass; the 2020 to 2029 national plan read in raw text. The plan names the city, not the building, and counts two centres working with two more being opened in Fès and Marrakech.) · The description now says what the plan lists for Casablanca and that it does not say which buildings house them.
- 2026-11-17 · Cancer incidence in Rabat, Morocco, 2006 to 2008: The record said Rabat and Casablanca are the two Moroccan cities with published population-based registry data. (Second accuracy pass; the paper's abstract read through Europe PMC. It does not say so, and registry studies from other Moroccan regions exist.) · The record now says these are the two cities whose registry reports were read for OnCo.
- 2026-11-17 · Awareness and use of the Casablanca cancer registry among Moroccan health stakeholders: The record said the registry is known to only about half of the people working in cancer control. (Second accuracy pass; the paper's abstract read through Europe PMC. The figure is 56.4% of 319 respondents in a convenience sample.) · The record now gives the figure as a share of those who answered the survey.
- 2026-11-17 · Karima Bendahhou: The record used a gendered pronoun that no source states and spoke of co-authored papers where one was checked. (Second accuracy pass; the registry's 2013 to 2017 report and the Europe PMC record of the 2019 Breast Disease paper read in raw text.) · The pronouns were removed and the record now cites the one paper resolved.
- 2026-11-17 · Angola: HPV vaccination appears in WHO's data in 2025, at 61% of the girls targeted: The card credited the figure to WHO and UNICEF and gave it as 61% of girls. The source is WHO's own HPV estimate of programme coverage: 1,264,990 first doses against a target of 2,088,292, so it is a share of the girls the programme targeted, not of all girls. The card also said no primary source on breast screening had been found, when the Ministry of Health publishes a clinical protocol for breast cancer with a screening section. (Second accuracy pass; WHO's immunisation data rows for Angola read as raw JSON, and the Ministry of Health's news pages and breast cancer protocol (PDF) read as raw text.) · The figure is now attributed to WHO and described as coverage of the targeted girls, in the description, the title and the card. The ministry's reports of the national campaign for girls aged 9 to 12, launched on 27 October 2025, and the screening steps in its breast cancer protocol are cited.
- 2026-11-17 · Angola: Ministry, regulator and cancer plan: The page said the Ministry of Health's search returned only its home page and that its cancer content could not be searched. The search answers when the query is sent as the site's own form sends it, and returned 13 news items with cancer in the title; the documents list holds a clinical protocol for breast cancer. (Second accuracy pass; minsa.gov.ao home page, news search results and document lists read as raw HTML on 17 November 2026.) · The gap card and the paying card now say what the ministry's site holds and what it does not: no cancer plan and no essential medicines list among the documents listed, and nothing found on what patients pay. The line on what to verify was narrowed to match.
- 2026-11-17 · Angola: The national institute's records, and Lubango's registry from 2022: The card called the Luanda hospital registry Angola's longest-running cancer records, said the 2019 paper asked for a population-based registry with stronger quality control, and said the Lubango registry reported limited knowledge of pathology. The paper's authors call theirs the oldest hospital-based registry, and ask for more rigorous quality control at the hospital registry; the Lubango page reports difficulties with pathology, not limited knowledge of it. (Second accuracy pass; the abstract and full text of Miguel et al. 2019 from Europe PMC, and the African Cancer Registry Network's Lubango page and membership criteria, read as raw text.) · Each statement was reworded to the source's own claim and attributed. The card now says that the words population-based come from the network's membership criteria, not from the registry's page.
- 2026-11-17 · Angola: Building oncology outside Luanda, and teaching it: The card said a 2024 paper proposed making oncology a compulsory subject in medical schools, and that it described cancer services being built outside the capital. The abstract proposes an oncology programme for undergraduate medical students and notes that schools have no compulsory oncology discipline; it says nothing about services outside the capital. (Second accuracy pass; the abstract of da Rocha et al. 2024 (J Cancer Educ) read from Europe PMC as raw JSON.) · Reworded to what the abstract says, and the closing sentence now refers only to the Lubango pilot and the Lubango registry.
- 2026-11-17 · The hospital-based cancer registry of the Instituto Angolano de Controlo do Cancro, Luanda, 2012 to 2016: The summary line said prostate cancer, Kaposi sarcoma and lymphoma led in men, and stated as fact that the registry is Angola's oldest and best organised. The abstract gives shares for men (19.3% prostate, 7.5% each Kaposi sarcoma and non-Hodgkin lymphoma) without ranking them, and the description of the registry is the authors' own. (Second accuracy pass; the abstract of Miguel et al. 2019 read from Europe PMC as raw JSON.) · The line now gives the shares as the abstract states them, and the description of the registry is attributed to its authors.
- 2026-11-17 · Uzbekistan: IARC note, registry table and the gap card on national counts: The page called two lower counts "Uzbek sources" and presented "more than 20,000 new cancers registered in 2020" as the IAEA's own registered count, and "about 15,000 a year" as a WHO count of new cancers. Both come from IAEA summaries: the first sits in a sentence the IAEA attributes to IARC, and the second reads "WHO estimates that 15 000 people are afflicted by cancer". (Second accuracy pass; the IAEA imPACT Review summaries of 2014 and 2022 read as PDF text.) · The note, the table row and the gap card now say these are statements in IAEA summaries, name the attribution to IARC and to WHO, and say neither is an Uzbek registry table.
- 2026-11-17 · Uzbekistan: A cancer register assigned to the national centre in 2017, and Cancer-registr software announced in 2024: The title said Uzbekistan had "a national cancer register since 2017" and that the Cancer-registr software was "built in 2024". The 2017 resolution assigns the task of forming a register, and the UZINFOCOM news item of 8 November 2024 reports that the software had been developed without saying when. (Second accuracy pass; resolution PP-2866 on lex.uz and the UZINFOCOM news item read as raw page text.) · The title, the plain text and the IARC note now say the register was assigned in 2017 and that the software was reported in November 2024.
- 2026-11-17 · Uzbekistan: At the national cancer centre, care is free on an electronic referral, paid by the State Medical Insurance Fund: The title stated as a general fact that cancer care is free on an electronic referral, and the text said state insurance was "piloted in Syrdarya region from 2020". The only source for free care is the national cancer centre's own page, headed "Draft, not yet approved" and speaking of care at the centre; the pilot was ordered on 12 November 2020 and its funding moved to the Fund from 1 July 2021. (Second accuracy pass; the cancer centre's free-care page and the president.uz resolution of 12 November 2020 read as raw page text.) · The title and plain text now attribute free care to the centre's draft page and say the pilot was ordered in November 2020.
- 2026-11-17 · Uzbekistan: HPV vaccine reaching nine in ten girls or more, hepatitis B vaccine since 2001, and new early detection from 2027: The title and the page description said the vaccine reaches "more than nine in ten girls". The last-dose coverage Uzbekistan reported to WHO for 2021 is exactly 90%. (Second accuracy pass; WHO immunisation coverage rows for Uzbekistan read from the public API as JSON.) · Both now say "nine in ten girls or more".
- 2026-11-17 · Uzbekistan: Working radiotherapy machines rose from 5 in 2014 to 19 in 2022, and cobalt units are being replaced: The card said the 2022 review would guide an Uzbekistan and Islamic Development Bank project "under the IAEA's Rays of Hope initiative". The IAEA summary places the project under a partnership agreement the IAEA and the bank signed in 2019. (Second accuracy pass; the IAEA imPACT Review summary of 26 September 2022 read as PDF text.) · The sentence now names the 2019 partnership, as the summary does.
- 2026-11-17 · Uzbekistan: A fast lane for new cancer medicines at the national centre: The plain text said the resolution covers "blood cancer" medicines. The resolution says haematological medicines, which is wider. (Second accuracy pass; resolution PP-238, item 8, read on lex.uz as raw page text.) · Changed to "haematology".
- 2026-11-17 · Uzbekistan: No trial in OnCo's records has a site in Uzbekistan: The card gave ClinicalTrials.gov counts of 92 studies with an Uzbek site and 9 matching cancer, and stated the nine in the present tense. The registry now returns 93 and 10. (Second accuracy pass; the ClinicalTrials.gov API queried for studies with a site in Uzbekistan and read as JSON.) · The card gives the 17 November counts and keeps the 11 November ones beside them.
- 2026-11-17 · Uzbekistan: the gap cards on national counts and on figures not found: The gap cards said the national statistics agency's health tables could not be reached and that the number of oncologists was not found. The agency's social statistics page answers, and it carries a table of the number of oncologists (1,066 in 2023, 944 in 2024, 582 in 2025, from Ministry of Health administrative data). It lists no table of new cancers or cancer deaths that OnCo could find. A third gap card called the IAEA's 2022 summary two pages long; it is three. (Second accuracy pass; the National Statistics Committee's page and its data file read as raw text, and the IAEA PDF's page count checked.) · The cards now say what the statistics page holds, give the oncologist figures with a caution that the swings are unexplained, and correct the page count.
- 2026-11-17 · Republican Specialized Scientific-Practical Medical Center of Oncology and Radiology: The record said the centre's new complex is at 14A Tashkent Ring Road. The centre's site gives that address without saying it is the new complex. (Second accuracy pass; the centre's About page read as raw page text.) · Reworded to say the site gives this as the centre's address.
- 2026-11-17 · State Medical Insurance Fund of Uzbekistan: The record said as fact that the Fund pays for hospital cancer care on electronic referral. The source is the national cancer centre's draft free-care page, which describes care at that centre. (Second accuracy pass; the centre's free-care page read as raw page text.) · The sentence now attributes the statement to the centre's draft page.
- 2026-11-17 · Uzbekistan national cancer registry (Cancer-registr): The record said the register was being moved onto software "built with UZINFOCOM in 2024". The news item reports the software as developed and being prepared for launch, with no build date and no statement that the register had moved. (Second accuracy pass; the UZINFOCOM news item of 8 November 2024 read as raw page text.) · Reworded to say UZINFOCOM reported developing the software with the centre in November 2024.
- 2026-11-17 · Republican Specialized Scientific-Practical Medical Center of Maternal and Child Health: The record said the centre runs the HPV test reference laboratory and trains gynaecologists. Resolution PP-402 orders both; nothing read shows them in operation. (Second accuracy pass; resolution PP-402 read on lex.uz as raw page text.) · Reworded to say the centre is charged with these tasks.
- 2026-11-17 · Scientific-Practical Medical Center of Paediatric Oncology, Haematology and Immunology: The record said the Samarkand and Ishtikhan pilot sets the fast-track route for children with suspected cancer. In resolution PP-186 the fast-track route is a priority of the whole programme; the pilot introduces a new system for routing patients between levels of care. (Second accuracy pass; resolution PP-186, items 5 and 6, read on lex.uz as raw page text.) · Reworded to describe the pilot as the resolution does.
- 2026-11-17 · Oesophageal cancer in Uzbekistan, 2000 to 2018: The record's caveat, and the country page's table, called the incidence "crude" and the database "passive". The abstract uses neither word; it gives a mean incidence per 100,000 without saying whether it is age-standardised. (Second accuracy pass; the abstract read through Europe PMC as JSON.) · The caveat and the table now say the abstract does not state whether the rate is age-standardised.
- 2026-11-17 · Institut National d'Oncologie, Rabat: Said the Rabat registry's 2006 to 2008 report "was written" at the institute; the record gives an affiliation for the first author only (Second accuracy pass; Europe PMC record for DOI 10.3332/ecancer.2013.338 read in raw text) · Now says the report's first author is based there
- 2026-11-17 · Democratic Republic of the Congo: One private radiotherapy centre since 2020, a first public one begun in 2023, and a cancer centre loan in 2026: The page said the IAEA DIRAC directory was not read and gave no machine count. The registry section and card titles and one paper record stated the absence of a registry without the date of the review that found it. (Second accuracy pass; IAEA DIRAC public feed (JSON) read in raw text; the registry wording was compared with the dated finding of the 2021 imPACT Review already cited on the page.) · The radiotherapy card and the gap note now give DIRAC's listing read on 17 November 2026: one centre, Nganda Hospital Center in Kinshasa, one linear accelerator, no cobalt-60 unit, no brachytherapy, entry last updated in 2021. The registry titles now say no registry was found in 2021.
- 2026-11-17 · Tanzania: The IAEA's pages were read through the Internet Archive, and DIRAC was read later, on 17 November 2026: The gap note said DIRAC was not read and that the page could give no current count of linear accelerators or cobalt units. (Second accuracy pass; IAEA DIRAC public feed (JSON) read in raw text.) · The gap note and the radiotherapy card now give DIRAC's listing: six centres, nine linear accelerators and three cobalt-60 units, brachytherapy at four centres, each centre named with the year its entry was last updated (2017 to 2026). The page still says no staff count was read.
- 2026-11-17 · Kenya: 19 radiotherapy machines in 10 centres, about one for every 3 million people: The card and a gap note said DIRAC's list of individual centres requires a sign-in, so the centres were not named. It does not. (Second accuracy pass; IAEA DIRAC public feed (JSON) read in raw text.) · The card now names the 10 centres by city, gives the 19 machines as 17 linear accelerators and 2 cobalt-60 units, says six centres list brachytherapy, and gives the entry years (2018 to 2026). The sign-in statement is withdrawn in the gap note.
- 2026-11-17 · Myanmar: Nine radiotherapy centres with 23 treatment machines: The card and two gap notes said DIRAC's list of centres requires an IAEA sign-in. It does not. The page also stated as fact that Myanmar has no national cancer registry, where the sources read only describe none. (Second accuracy pass; IAEA DIRAC public feed (JSON) read in raw text; the registry wording was compared with the 2025 review already cited, which says the two regional registries cover only 17% of the population.) · The card now names the nine centres, gives the 23 machines as 18 linear accelerators and 5 cobalt-60 units, and says four of the nine entries were last updated in 2017. The lede, the registry card, a table note, the IARC note and a gap now say that no national registry is described in the sources read.
- 2026-11-17 · Sudan: One working machine per seven million people before April 2023, and one centre by September 2025: A gap note said the IAEA DIRAC directory was not read. (Second accuracy pass; IAEA DIRAC public feed (JSON) read in raw text.) · The radiotherapy card now gives DIRAC's listing beside the 2026 census paper's counts: three centres with radiotherapy in entries last updated in 2022, with four cobalt-60 units and two linear accelerators and no brachytherapy listed. The gap note is corrected.
- 2026-11-17 · Uganda: In 2016 Uganda's only working radiotherapy machine broke down; it now has eight: Two notes said the IAEA DIRAC directory was not read and that the machine count rested only on the institute's statement to a newspaper. (Second accuracy pass; IAEA DIRAC public feed (JSON) read in raw text.) · The radiotherapy card and the gap note now give DIRAC's listing: one centre, the Uganda Cancer Institute's radiotherapy department, with three linear accelerators, two cobalt-60 units and brachytherapy in an entry last updated in 2026, and one brachytherapy unit in the country table against the three the institute reported.
- 2026-11-17 · Iraq: Oncology hospitals in Baghdad and radiotherapy centres in the Kurdistan Region: The card said no current public count of radiotherapy machines was found and that DIRAC returned no figures to a script; a gap note repeated it. (Second accuracy pass; IAEA DIRAC public feed (JSON) read in raw text.) · The card and the gap note now give DIRAC's listing: 31 centres with radiotherapy by city, 54 linear accelerators, no cobalt-60 unit, three Gamma Knife units, brachytherapy at five centres, and the entry years (29 of 31 last updated in 2025).
- 2026-11-17 · Afghanistan: No national plan, no radiotherapy, almost no palliative care: A gap note said the IAEA DIRAC directory was not read. (Second accuracy pass; IAEA DIRAC public feed (JSON) read in raw text.) · The card and the gap note now say that DIRAC, read on 17 November 2026, has no row for Afghanistan in either its table of countries or its list of centres, and that a missing row is not by itself a finding that no machine exists.
- 2026-11-17 · Yemen: How many radiotherapy machines work now is not confirmed: The gap card said DIRAC does not show country data without its interactive tool. The lede gave the absence of a national registry as the reason IARC models Yemen's figures, which IARC's fact sheet does not say, and a table note stated that absence as fact. (Second accuracy pass; IAEA DIRAC public feed (JSON) read in raw text; the registry wording was compared with the IARC fact sheet and WHO 2020 profile wording already quoted on the page.) · The gap card and the radiotherapy card now give DIRAC's listing: three centres in entries last updated in 2026, two in Sana'a and one in Mukalla, with four linear accelerators, no cobalt-60 unit and brachytherapy at one centre. The lede now says IARC's fact sheet names no Yemeni data source, and the table note says no national registry was found.
- 2026-11-17 · Morocco: 45 linear accelerators, more than half of them private: The card and a gap note said the IAEA's directory of radiotherapy centres could not be queried, so no count later than the national plan's was given. (Second accuracy pass; IAEA DIRAC public feed (JSON) read in raw text.) · The card now gives DIRAC's listing beside the plan's 45: 60 centres with radiotherapy in 23 cities, 91 linear accelerators, no cobalt-60 unit and brachytherapy at 24 centres, every entry last updated in 2026. The gap note is corrected.
- 2026-11-17 · Angola: Four radiotherapy centres and seven treatment machines in the IAEA's directory: The card's DIRAC figures were right, but a gap note said the centre-by-centre report was not read and the centres were not named. (Second accuracy pass; IAEA DIRAC public feed (JSON) read in raw text.) · The card now names the four centres, says all seven machines are linear accelerators and that three centres list brachytherapy, with entries last updated in 2026. The gap note is corrected.
- 2026-11-17 · Uzbekistan: Working radiotherapy machines rose from 5 in 2014 to 19 in 2022, and cobalt units are being replaced: The card gave only the IAEA review summary's count of 19 machines in 15 institutions; DIRAC's listing, which differs, was not on the page. (Second accuracy pass; IAEA DIRAC public feed (JSON) read in raw text.) · The card now gives DIRAC's listing beside the review's count: 15 centres with radiotherapy, 18 cobalt-60 units and 2 linear accelerators, brachytherapy at eight centres, every entry last updated in 2022.
- 2026-11-17 · Ghana: How many radiotherapy machines are treating patients is not settled: The page set two conflicting counts side by side and said neither named the hospitals, without the IAEA directory's listing. The lede gave the single city registry as the reason the national figures are IARC estimates, and the registry card stated as fact that Ghana has no national registry. (Second accuracy pass; IAEA DIRAC public feed (JSON) read in raw text; the registry wording was compared with the 2025 review and IARC fact sheet wording already on the page.) · The gap card and the radiotherapy card now add DIRAC's listing: three centres in entries last updated in 2026, with three linear accelerators and one cobalt-60 unit. The lede and registry card now state only what the review says and what OnCo found.
- 2026-11-17 · Saudi Arabia: No Saudi national count of radiotherapy machines beyond the IAEA's directory, and no published national cancer plan: The gap card and the radiotherapy card said the IAEA DIRAC directory could not be read and returned no data. (Second accuracy pass; IAEA DIRAC public feed (JSON) read in raw text.) · The gap card now gives DIRAC's listing: 18 centres with radiotherapy by city, 47 linear accelerators under one heading and 51 under another, one Gamma Knife, no cobalt-60 unit, brachytherapy at nine centres, and entry years from 1999 to 2026. The card title and plain text are corrected.
- 2026-11-17 · Ghana: page description and opening summary: The page said the Ghana Medical Trust Fund began paying for cancer care in 2026 and was starting to pay for cancers outside national insurance. The sources read say only that payouts in general had begun by December 2025 and that the Fund's Administrator said in May 2026 that its first support, for cancer, would start the following month. (Second accuracy pass; Graphic Online's reports of 1 December 2025 and 6 May 2026 read in raw text.) · The description and opening summary now say what the Administrator said in May 2026 and that the Fund is meant to pay for what insurance does not, starting with cancer.
- 2026-11-17 · Ghana: Four radiotherapy machines in operation in 2026, and about a quarter of patients who need radiotherapy treated: The page said the international reviewers counted four radiotherapy machines. The IAEA's article says only that four machines were in operation nationwide and does not say the review team counted them. The card also called a paper published in 2022 on 2020 data a 2020 study. (Second accuracy pass; the IAEA's article of 11 June 2026 read in raw text from the Internet Archive copy, and the paper's abstract read through Europe PMC.) · The figure is now given as what the IAEA wrote after the review, in the card, the opening summary and the Ministry of Health record, and the study is described as a study of 2020.
- 2026-11-17 · Ghana: The Kumasi Cancer Registry: population-based since 2012, run on a teaching hospital's budget: Two statements were credited to the registry's manager that the news report makes in its own voice or credits to the registry: that it had been run from the hospital's funds since 2012 and that 1,070 cancers were recorded in 2012 to 2014. The manager's quoted words mention traditional healing and prayer camps, not herbal centres. The summary of national figures also set the registry's 2015 rates beside IARC's age-standardised rates without saying the two may not be measured the same way. (Second accuracy pass; MyJoyOnline's report of 24 April 2015 and the registry's 2019 paper abstract (Europe PMC) read in raw text.) · The two statements are now credited to MyJoyOnline's report and the registry, the manager's words are given as quoted, and the comparison with IARC's rates carries the qualifier.
- 2026-11-17 · Kumasi Cancer Registry: The record said the registry was drawing on private pathology laboratories, private clinics and Ghana Health Service facilities, where its network page only lists these as identified data sources, and it credited the funding and case-count statements to the registry's manager. (Second accuracy pass; the African Cancer Registry Network page dated March 2017 (Internet Archive copy) and MyJoyOnline's report of 24 April 2015 read in raw text.) · The record now says the page lists these as identified data sources and credits the launch statements to the news report and the registry.
- 2026-11-17 · Ghana: WHO maturity level 3 for medicines since 2020, and level 4 for clinical trial oversight since 2024: The card described maturity level 3 as a stable and well-functioning system and said WHO rated the Authority in 2020. WHO's feature uses neither phrase and gives no date for the rating; it reports the level in an article dated 13 May 2020. (Second accuracy pass; WHO's feature of 13 May 2020 read in raw text.) · The card and the Food and Drugs Authority record now say WHO reported in May 2020 that the Authority had reached level 3, the second-highest of four.
- 2026-11-17 · Ghana: The Ghana Medical Trust Fund: launched in 2025, paying for cancer care first: The card said the Health Minister told the November 2025 National Cancer Forum that the Fund would cover cancer treatment from 2026, without saying the speech was read on the Minister's behalf by the Ministry's Chief Director. (Second accuracy pass; the Ghanaian Times report of 19 November 2025 read in raw text.) · The card and the Ministry of Health record now say the speech was read on the Minister's behalf.
- 2026-11-17 · Saudi Arabia: A notifiable disease by royal decree, counted by trained registrars in oncology centres across the Kingdom: The card said the registry had covered the whole country since 1992, that every government and private hospital must report cancer, and that registrars work in every oncology centre. The page's opening line also said cancer had been counted nationally since 1992. The registry's report gives 1992 only as the year it was set up, with annual reports since 1994, and says it strives for full access to data from all hospitals, clinics and laboratories. (Second accuracy pass; the Saudi Cancer Registry's Cancer Incidence Report 2024 read in raw PDF text.) · Reworded the card, its title and the opening line to the report's own statements: set up in 1992, cancer notifiable by royal decree, and the registry striving for full access to data.
- 2026-11-17 · Saudi Arabia: About 28,000 cancers reported in 2024, more in women than men, with a median age at diagnosis in the fifties: The card said more cancers were found in Saudi women than men mainly because of breast and thyroid cancer, a reason the report does not give. The note on IARC's figures described 5,728 breast cancers as being in women of all nationalities, which the report's table does not state. (Second accuracy pass; the Saudi Cancer Registry's Cancer Incidence Report 2024 read in raw PDF text (tables 2.5 and 6.1).) · The card now gives the shares the report states, and the note says 5,728 breast cancers registered in women, 4,601 of them in Saudi nationals.
- 2026-11-17 · Saudi Arabia: Citizens are treated free in government hospitals, and the system is being split into health clusters: The card said free government care includes cancer treatment and that the Ministry is moving hospitals into regional health clusters. The source is a 2021 plan written in the future tense, which promises citizens free cover of all healthcare requirements without naming cancer, and does not call the clusters regional. (Second accuracy pass; the Ministry of Health's Health Sector Transformation Program delivery plan read in raw PDF text.) · Reworded as what the 2021 plan sets out, and the five test clusters are now described as to be launched.
- 2026-11-17 · Saudi Arabia: Private health insurance is compulsory for the groups the Council of Health Insurance defines: The card said insurance is mostly bought by employers and that the Council runs the NPHIES insurance exchange. The Council's page says neither: it says the Council oversees the operation of the NPHIES platform, and the cited paper speaks only of the reform's aim to rely more on employer insurance. (Second accuracy pass; the Council of Health Insurance's archived who-we-are page read as raw HTML text and the Althabaiti abstract read through Europe PMC.) · Removed the employer statement and reworded the NPHIES clause to the Council's words.
- 2026-11-17 · Saudi Arabia: Saudi hospitals take part in multinational phase 3 cancer trials, but in one survey few patients had been in one: The card called the patient survey a 2026 survey of Saudi cancer patients. It ran from December 2023 to June 2024, was published in 2026, covered 367 patients, and its abstract does not give their nationality. (Second accuracy pass; the abstract of Alkaiyat and colleagues read through Europe PMC.) · The card and its title now say a survey of 367 cancer patients in Saudi Arabia, published in 2026.
- 2026-11-17 · Saudi Arabia: No national count of radiotherapy machines, and no published national cancer plan: The card said the 2019 proton therapy guideline described the Saudi Particle Therapy Center as setting up a proton unit; the guideline says the centre had recently established one, in the same paragraph that says the Kingdom lacked proton facilities. The card also said the 2018 radiotherapy equipment report could not be downloaded, and that King Faisal Specialist Hospital's pages do not mention proton therapy when only its home page was read. (Second accuracy pass; the 2019 guideline read in raw PDF text, the 2018 report downloaded and inspected, and the hospital's home page read as raw HTML text.) · The card now gives both statements from the 2019 guideline, says the 2018 report is a 35-page scan with no text layer from which no figures are given, and refers to the hospital's home page only.
- 2026-11-17 · Saudi Arabia: A national cancer centre that sets guidelines and publishes the numbers: The card said the Center's goals include tracking what cancer costs patients and the system. Its goals page says only that it follows up on the financial burden of cancer. (Second accuracy pass; the National Cancer Center's strategic goals page read as raw HTML text.) · Reworded to the goal as the Center states it.
- 2026-11-17 · Saudi Arabia: Verify before relying on it: The closing advice said screening is offered through health clusters and the vaccine schedule through primary care. Neither is in the sources read; the registry's report names Ministry screening centres, mobile screening trucks and partner health sectors. (Second accuracy pass; the Saudi Cancer Registry's Cancer Incidence Report 2024, part VI, read in raw PDF text.) · Replaced with the report's own list and removed the sentence on where the vaccine is given.
- 2026-11-17 · Council of Health Insurance, Saudi Arabia (CHI): The record said the Council operates the NPHIES platform, described it as a national insurance data exchange, and said its site publishes monthly statistics. The Council's page says it oversees the operation of the NPHIES platform and lists statistics without saying how often. (Second accuracy pass; the Council's archived who-we-are page read as raw HTML text.) · Reworded to the Council's own description and removed the word monthly.
- 2026-11-17 · Colorectal cancer incidence in Saudi Arabia, 2002 to 2023, by age: The record said the author argues for reliable diagnostic pathways after a positive test, which the abstract does not say. (Second accuracy pass; the abstract read through Europe PMC.) · Reworded to the abstract's conclusion: age-targeted screening, stronger early detection pathways and continuity of diagnostic services.
- 2026-11-17 · Radiotherapy capacity, staffing and waiting times in Saudi Arabia: The record's summary line said public centres had the fewest machines per patient; the abstract says fewer. (Second accuracy pass; the abstract read through Europe PMC.) · Corrected to fewer machines per patient.
- 2026-11-17 · Cancer screening among privately insured people in Saudi Arabia, 2023 to 2024: The record's summary line described a dip of about a third around Ramadan and Eid; the abstract gives April counts 32.4% to 41.6% below predicted values. (Second accuracy pass; the abstract read through Europe PMC.) · Reworded to a third or more below prediction.
- 2026-11-17 · Peru: The Lima and Callao population registry, with its latest report for 2013 to 2015: The card called the 2013 to 2015 report the registry's sixth report and said its cases also come from specialist centres. The report is volume 6 but says it is the seventh report published since 1990, and it names pathology and haematology laboratories and private practices, not specialist centres. (Second accuracy pass; INEN's Registro de Cáncer de Lima Metropolitana 2013 to 2015, volume 6, read as raw PDF text.) · Now says latest report, volume 6, and lists the sources of cases as the report does.
- 2026-11-17 · Peru: What Peru's own registries report: The table note and a gap card stated as fact that no national incidence count exists. No source read says so; the first pass had only failed to find one. (Second accuracy pass; IARC's GLOBOCAN 2024 fact sheet for Peru and the gob.pe page of the Lima registry, read as raw text.) · Both now say that OnCo found no published national count.
- 2026-11-17 · Peru: The 2021 cancer law: universal, free and priority cover, and SIS enrolment for anyone without insurance: The card said the law protects the cancer budget programme and declares central purchasing of cancer medicines of national interest, dated the regulation 30 March 2022, and said private insurers must cover treatments in international guidelines. The law declares both points of national interest and speaks of strategic resources for the public sector; the decree was signed on 29 March 2022; and the duty on private insurers applies to policies that include cancer cover. (Second accuracy pass; Ley 31336 as printed in El Peruano and Decreto Supremo 004-2022-SA, read as raw PDF text.) · Reworded to the law's terms, both dates given and the qualifier added.
- 2026-11-17 · Peru: FISSAL pays for seven cancers, automatically from diagnosis, and for travel when treatment is far away: The card said cancers outside the seven are covered through the essential and complementary plans under the 2022 regulation. The regulation says only that cancer cover runs through whichever insurance plan applies, and FISSAL's page does not address other cancers. (Second accuracy pass; article 7 of Decreto Supremo 004-2022-SA and FISSAL's institutional page on gob.pe, read as raw text.) · Reworded to what the two sources say.
- 2026-11-17 · Peru: A 45-day decision for medicines already registered in a high-surveillance country, or approval by default: The card said the 2025 law repealed the 45-day article that Ley 31738 had added. The law says only that it repeals article 9 of Ley 29698 as added by Ley 31738. A gap card also mentioned a 2025 draft of an implementing decree for which no source was given. (Second accuracy pass; Ley 32319 read from El Peruano's raw HTML.) · The repeal is now described in the law's own terms and the unsourced draft is removed.
- 2026-11-17 · Peru: Insurance follows the diagnosis: an uninsured person with cancer is enrolled in the SIS whatever their income: The card called FISSAL's travel cover Peru's answer to a cancer service concentrated in Lima and credited a 43% figure to an INEN statement. The first is an inference no source makes; the figure is from an El Peruano report of INEN's numbers. (Second accuracy pass; El Peruano's report of 5 February 2022, read as raw HTML.) · The inference is removed and the figure is attributed to El Peruano, with the 57% from Lima and Callao beside it.
- 2026-11-17 · Peru: A national cancer institute that runs the registry and leads a national oncology network: The card said INEN shares responsibility for the national cancer plan with the Ministry, which none of its sources says, and gave INEN's 2021 activity figures as INEN's own report. (Second accuracy pass; Ley 31336 and El Peruano's report of 5 February 2022, read as raw text.) · The unsupported phrase is removed and the figures are attributed to El Peruano.
- 2026-11-17 · Peru: Childhood cancer: an urgency law, and treatment abandonment cut from 18.6% to 8.5%: The card said the paper's authors ran the programme they evaluate. The abstract describes a participatory and documentary evaluation, and the authors' affiliations are INEN, the Ministry, the Pan American Health Organization, St. Jude Children's Research Hospital and others. (Second accuracy pass; the Europe PMC record for DOI 10.26633/RPSP.2023.132, read as raw JSON.) · Reworded to what the abstract and affiliations show.
- 2026-11-17 · Peru: 144 trials in OnCo's records list a site in Peru, nearly all phase 3 and industry-sponsored: The card gave 142 trials, 125 of them phase 3, and 50 active but not recruiting. OnCo's records now hold 144, 127 and 52. (Second accuracy pass; the ClinicalTrials.gov participation data in OnCo's graph, recounted.) · Counts updated and dated 17 November 2026.
- 2026-11-17 · Registro de Cáncer de Lima Metropolitana: The record called the 2013 to 2015 report the sixth report and listed doctors' practices and specialist centres among the sources of cases. The report is volume 6 and the seventh since 1990, and names laboratories and private practices. (Second accuracy pass; INEN's Registro de Cáncer de Lima Metropolitana 2013 to 2015, volume 6, read as raw PDF text.) · Reworded to the report's own description.
- 2026-11-17 · Ministerio de Salud del Perú (MINSA): The record dated Decreto Supremo 004-2022-SA 30 March 2022 and Decreto Supremo 016-2025-SA 10 September 2025. Those are the dates gob.pe lists; the decrees were signed on 29 March 2022 and 5 September 2025. (Second accuracy pass; the two decrees, read as raw PDF text and by OCR.) · Both dates are now given for each decree.
- 2026-11-17 · Seguro Integral de Salud (SIS), Peru: The record said the regulation gives SIS cancer patients the essential plan, the complementary plan and FISSAL cover. That provision applies to a patient left without cover during another insurer's latency period. (Second accuracy pass; article 9.3 of Decreto Supremo 004-2022-SA, read as raw PDF text.) · Reworded to the article's scope.
- 2026-11-17 · Seguro Social de Salud (EsSalud), Peru: The record described EsSalud as the insurer for salaried workers with its own hospitals, stated in the present tense that it has radiotherapy at two Lima hospitals, credited a list of five hospitals to EsSalud from a table whose headings cannot be read in the scan, and said a patient whose EsSalud cover lapses is enrolled in the SIS. (Second accuracy pass; EsSalud's institutional page, Plan Esperanza (Decreto Supremo 009-2012-SA), the 2025 to 2030 cancer plan and Ley 31336, read as raw text.) · Reworded to EsSalud's own description, the hospital list dated to 2012 with what each plan shows, and the law's wording on enrolment.
- 2026-11-17 · Mortality-incidence ratio as an estimate of five-year cancer survival in Metropolitan Lima: The record said Lima has no routine survival statistics, which the abstract does not say. (Second accuracy pass; the Europe PMC record for DOI 10.3332/ecancer.2018.799, read as raw JSON.) · Replaced with the authors' conclusion and what the registry's 2013 to 2015 report tabulates.
- 2026-11-17 · Peru's childhood cancer policy under the WHO Global Initiative for Childhood Cancer: The record said Peru joined the initiative in June 2019 and that the paper was written by the people who ran the programme. The abstract reports progress since June 2019 and describes a participatory evaluation. (Second accuracy pass; the Europe PMC record for DOI 10.26633/RPSP.2023.132, read as raw JSON.) · Reworded to the abstract and the authors' affiliations.
- 2026-11-17 · Dirección General de Medicamentos, Insumos y Drogas (DIGEMID), Peru: The record said Ley 32319 repealed the earlier article of Ley 31738. The law repeals article 9 of Ley 29698, which Ley 31738 had added. (Second accuracy pass; Ley 32319 read from El Peruano's raw HTML.) · Reworded to the law's own terms.
- 2026-11-17 · Wilder Genaro Calmet Berrocal: The record stated that Francisco Berrospi Espinoza was institutional head from 2022, which the cited directory does not show and which could not be re-read on an official page. (Second accuracy pass; the gob.pe directory of INEN officials, read as raw HTML.) · Now gives what the directory shows and says the 2022 date was not re-read on 17 November 2026.
- 2026-11-17 · Peru: Eight public hospitals with their own radiotherapy, and five that outsource it: The card and a gap card said the IAEA's DIRAC directory of radiotherapy centres could not be read because it serves its data through scripts. The directory has a public data feed, which lists 26 centres with radiotherapy in Peru, with 54 linear accelerators and 11 cobalt-60 units. (Second accuracy pass; the DIRAC equipment feed at dirac.iaea.org, read as raw JSON on 17 November 2026.) · The card now gives DIRAC's centres, machines by type and entry years beside the Ministry's count of public hospitals, and the gap text is corrected.
- 2026-11-17 · Nepal Population-Based Cancer Registry (NHRC): The record placed both East and West Rukum in Karnali Province. (Second accuracy pass; the Nepal Health Research Council's report Cancer Incidence and Mortality in Rukum Districts in 2018, read in raw PDF text, puts East Rukum in Province 5 and West Rukum in Karnali Province.) · The record now gives each district its own province.
- 2026-11-17 · Nepal: No organised screening programme; cervical screening is free but few women have had it: The card said only 63% of women who needed treatment after a screening result received it, and that screening is offered in some facilities. The strategy says only that 63% received treatment after a result, and that screening is done in camps and research programmes. (Second accuracy pass; the Nepal Cancer Control Strategy 2024 to 2030, section 3.2, read in raw PDF text with its legacy Nepali font decoded.) · Both sentences now follow the strategy's wording.
- 2026-11-17 · Nepal: HPV vaccine for girls nationwide from 2025: The title called 2025 the first campaign, the card gave the routine dose as for girls aged 10 to 14, and the checking note said the vaccine is given through schools and health facilities. The strategy records a campaign in all seven provinces in 2023, WHO's schedule gives one dose at age 10 with ages 10 to 14 being the out-of-school arm of the 2025 campaign, and no source read says where the vaccine is given. (Second accuracy pass; WHO immunisation schedule, introduction and coverage rows for Nepal read as raw JSON, and the strategy's section 3.1 read in raw PDF text.) · Title, card and checking note reworded to what the two sources say.
- 2026-11-17 · Nepal: Free childhood cancer medicines through a WHO and St. Jude platform: Two cards and the BP Koirala Memorial Cancer Hospital record said Nepal receives 35 childhood cancer medicines free, as a present fact, and one card gave no source for it. The source is a March 2024 announcement that Nepal will receive them. (Second accuracy pass; WHO Nepal's news release of 27 March 2024 read in raw HTML.) · Reworded as what WHO Nepal announced in March 2024, with the release linked from the card that lacked it.
- 2026-11-17 · Nepal: A national cancer strategy built on an international review: The card dated a follow-up session of the Ministry and WHO to May 2024. The WHO page is dated 13 May 2024 and does not say when the session was held. (Second accuracy pass; WHO Nepal's departmental update of 13 May 2024 read in raw HTML.) · Now says WHO Nepal reported the session on 13 May 2024.
- 2026-11-17 · Nepal: A grant of NPR 100,000 for poor patients: The card said the strategy names the grant and the Health Insurance Act as the main financial relief; the Ministry and Health Insurance Board records said the ministry runs the scheme and that it pays up to NPR 100,000. The strategy does not call them the main relief, both sources call the scheme the Government of Nepal's, and the board's report says aid worth NPR 100,000. (Second accuracy pass; the strategy's section 2 and the Health Insurance Board's 2022/23 factsheet and executive summary, both read in raw PDF text.) · The word main, the attribution to the ministry and the words up to are removed.
- 2026-11-17 · Nepal: About 24,500 new cancers a year: The card said the national strategy says most cancers are found late. The strategy gives late-stage figures for breast, lung, and mouth and throat cancers only. (Second accuracy pass; the strategy's section 3.2 read in raw PDF text.) · The sentence now names those cancers.
- 2026-11-17 · Nepal: Six government cancer hospitals, four of them in Kathmandu Valley: The card said the reviewers found that about 25% of patients do not complete planned treatment because of cost. The report gives the figure as an estimate. (Second accuracy pass; the IAEA, WHO and IARC imPACT Review summary report for Nepal read in raw PDF text.) · Reworded as an estimate the reviewers reported.
- 2026-11-17 · Nepal: 48 radiation oncologists, 20 medical oncologists and 18 medical physicists: The card said the reviewers recommended training medical and surgical oncologists through fellowships abroad. The report recommends national high-volume centres for medical oncologists, and medium-term or longer fellowships involving cancer centres abroad for surgeons. (Second accuracy pass; the IAEA, WHO and IARC imPACT Review summary report for Nepal read in raw PDF text.) · Reworded to the report's terms.
- 2026-11-17 · Nepal: No trial in OnCo's records has a site in Nepal: The card said all the registry papers were written with colleagues at IARC's regional hub at Tata Memorial Hospital. One of the four lists no author there. (Second accuracy pass; author affiliations of the four papers read in the Europe PMC full-text XML.) · Now says three of the four, at Tata Memorial Centre, Mumbai.
- 2026-11-17 · Urban and rural differences in cancer in Nepal: Kathmandu Valley and Rukum registries: The record's reading of the paper added a recommendation for cervical screening where breast cancer is less common, which the abstract does not make. (Second accuracy pass; the abstract read through Europe PMC.) · Replaced with the authors' own conclusion that urban and rural regions need different policy priorities.
- 2026-11-17 · Cancer risk across urban, suburban and rural Nepal from the population-based registries, 2019: The record called the paper the first registry-based picture spanning urban, suburban and rural Nepal. The abstract makes no such claim, and a paper on the same 2019 registry data was published two months earlier. (Second accuracy pass; both abstracts and publication dates read through Europe PMC.) · The word first is removed.
- 2026-11-17 · Cancer in Kathmandu Valley in 2018, from Nepal's first population-based cancer registry: The record said that until this registry cancer in Nepal was counted only in hospital registries and IARC estimates, which the abstract does not say. (Second accuracy pass; the abstract read through Europe PMC.) · Now says, as the authors do, that the country lacked a population-based registry until 2018.
- 2026-11-17 · Nepal: About ten radiotherapy machines against a need of about 30: The card gave only the national strategy's count of ten radiotherapy units and a 2020 paper's count, and did not give the IAEA directory's own listing, which differs. (Second accuracy pass; the IAEA DIRAC equipment feed read as raw JSON on 17 November 2026.) · The card now adds what DIRAC lists, seven centres with 10 linear accelerators and 3 cobalt-60 units and brachytherapy at four, with entry years from 2018 to 2026, beside the strategy's count, with both dated.
- 2026-11-17 · Uganda: page summary: Said IARC's estimate is built on the Kampala and Gulu registries; the fact sheet names one unnamed subnational registry and does not say how each was used (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet for Uganda read in raw text) · Now says the estimate is built from subnational registry data, and names the two registries separately
- 2026-11-17 · Afghanistan: no screening programmes and no HPV vaccination in WHO's data: Said WHO's immunisation data returned no rows for HPV in Afghanistan; that was true only of a narrow antigen filter (Second accuracy pass; WHO immunisation API queried for every HPV antigen and read as raw JSON) · Now says WHO holds rows for 2018 to 2025 and that every one is blank, with the wider query linked
- 2026-11-17 · Venezuela: lede, description and yearbook table note: The page said as fact that Venezuela "has not published" and "publishes no" national count of new cancers, that the yearbooks "for a time went no further than 2013" and that the 2017 yearbook came "after years without one". The sources only lack such a count (IARC's fact sheet says "No data", WHO's 2019 row says "No information"), and the Ministry's site does not show when each yearbook was released. (Second accuracy pass; IARC's GLOBOCAN 2024 fact sheet, WHO's 2020 cancer country profile and the Ministry of Health's yearbook index read in raw text.) · Reworded to "OnCo found no published national count of new cancers after 2014"; the two timing claims were removed, and the statement that little health data was published is now attributed to the two doctors who wrote it in 2019.
- 2026-11-17 · Venezuela: Counted deaths: 25,400 from cancer in 2017: Registered cancer deaths were rounded to "about 25,000 a year in 2016 and 2017", and the difference from IARC's 2024 estimate was explained by population ageing and unregistered deaths, which neither source says. (Second accuracy pass; the Ministry of Health's 2016 and 2017 mortality yearbooks (PDF text) and IARC's fact sheet.) · The card gives 25,836 for 2016 and 25,400 for 2017, and the note now says only that the two figures are seven years apart and not like for like.
- 2026-11-17 · Venezuela: What was published, and when: The card said the Ministry "posted weekly epidemiological bulletins again" in 2026, which the bulletins page does not say, and gave the list as ending at week 42. (Second accuracy pass; the Ministry of Health's bulletins page read in raw text on 17 November 2026.) · "Again" was removed and the list is given as it stood on 17 November 2026, to week 44 (1 to 7 November).
- 2026-11-17 · Venezuela: The Sociedad Anticancerosa's own projections: Two cards said the Sociedad made its projections because official figures had stopped appearing, that it "commissioned" them, that one partner was "a pharmaceutical company", and that no official source then gave state-level rates. The bulletin states none of these. (Second accuracy pass; the Sociedad Anticancerosa de Venezuela's 2018 bulletin (PDF text, ICCP portal copy).) · The motive and the claim about official sources were removed, the work is described as a joint effort, and the partners are named as the bulletin names them: Laboratorios Nolver and the Fundación Banco Mercantil.
- 2026-11-17 · Venezuela: 56 high-cost pharmacies that the Ministry says supply cancer medicines free: The card and its title said the Ministry describes cancer care in public hospitals and at the IVSS as free, and that "patients and doctors" reported shortages. The pages cited say only that the IVSS pharmacies supply cancer therapies free; the shortage reports come from interviewed oncologists and a survey of adults. (Second accuracy pass; the Ministry's IVSS anniversary page (through the site's public API), the IVSS pharmacy directory, and the Duma commentary in full text.) · Title and summary reworded to what each source says, with the shortage reports attributed to their sources.
- 2026-11-17 · Venezuela: The UN humanitarian plan covers people with chronic disease who lack medicines: The card gave the health response as 31% funded in June 2026. WHO's Health Cluster page now carries data dated September 2026 and shows 38% funded. The two OCHA plans could not be re-read because ReliefWeb answered with an automated challenge. (Second accuracy pass; WHO Health Cluster page read in raw text on 17 November 2026; ReliefWeb returned a challenge, which was not bypassed.) · The funding figure and its date were updated, the plan figures are labelled as read on 11 November 2026 and not re-read, and the Health Cluster's own statement that the 2026 response targets 5.2 million people is given beside the plan's 5.5 million.
- 2026-11-17 · Venezuela: A medicine needs a registro sanitario from the INHRR before it can be sold: The card said biotherapeutic products are "a category that includes many biological cancer medicines", which no INHRR page says. (Second accuracy pass; the three INHRR pages and the posted norm (a 17-page scan with no text layer).) · The clause was removed.
- 2026-11-17 · Venezuela: HPV vaccine not yet in the national schedule; hepatitis B coverage has fallen: The card said Venezuela "reported no HPV coverage". WHO's coverage table holds 44 HPV rows for Venezuela for 2018 to 2025, all blank. (Second accuracy pass; WHO immunisation data API queried for every HPV antigen for Venezuela and read as JSON.) · The card now says the rows exist and that coverage, doses and target are blank in every one.
- 2026-11-17 · Venezuela: 43 radiotherapy centres in IAEA records, but how many machines work is not published: The card gave "13 brachytherapy units" and dated its reading of the IAEA directory to 11 November from the country table page, which is a script shell. The directory's data feed gives a country brachytherapy count of 13 but marks only 9 centres as having brachytherapy, and it lists machine types and entry years the card did not give. (Second accuracy pass; IAEA DIRAC's public data feed (centre rows and country totals) read as JSON on 17 November 2026.) · The card now gives 43 centres, 59 megavoltage units (50 linear accelerators and 9 cobalt-60 units), both brachytherapy figures, the spread of entry years from 2014 to 2026, and the Razetti institute's own entry.
- 2026-11-17 · Venezuela: The Instituto Oncológico Luis Razetti, 90 years old, and the Hospital Padre Machado: The summary said the Hospital Padre Machado "offers radiotherapy, chemotherapy and surgery" as a present fact; the source is a ministerial statement of October 2023. (Second accuracy pass; the Ministry of Health's news page of 24 October 2023 read through the site's public API.) · The sentence is now attributed to the Ministry and dated.
- 2026-11-17 · Instituto Nacional de Higiene "Rafael Rangel" (INHRR): The programme list named a pharmacovigilance centre (CENAVIF) and a "public health laboratory network" that appear on none of the cited pages. (Second accuracy pass; the three INHRR pages read in raw text.) · The list now says "Pharmacovigilance" and "Public health laboratory diagnosis".
- 2026-11-17 · Sociedad Anticancerosa de Venezuela: The summary line said the Sociedad published projections "because recent official figures were not available", a reason its bulletin does not give. (Second accuracy pass; the Sociedad's 2018 bulletin (PDF text).) · The reason was removed.
- 2026-11-17 · Routine cystoscopy and rectosigmoidoscopy in staging locally advanced cervical cancer in Venezuela: The summary line said invasion was "almost always already suspected on MRI"; the abstract says on MRI or clinical symptoms. (Second accuracy pass; the abstract through Europe PMC.) · The words "or from symptoms" were added.
- 2026-11-17 · Sri Lanka: Colombo district: the first population-based registry, from 2012: The card gave Colombo district as 12.4% of Sri Lankans at the 2012 census, said the registry counted every cancer among residents, and gave the reason for its higher rates as a more extensive methodology. (Second accuracy pass; the scanned Colombo District 2013 to 2019 report was read by text recognition, and it gives 2,324,349 people as 11.4% of the population in two places and puts the higher rates down to data from multiple sources.) · The share now reads 11.4%, the registry is described as population-based without the word every, and the reason is given in the report's terms.
- 2026-11-17 · Sri Lanka: Radiotherapy machines: the IAEA directory gives a count, and OnCo found no official Sri Lankan one: The gap card and the capacity card said the IAEA's radiotherapy directory could not be read and that no current machine count was available. (Second accuracy pass; the directory's public data feed at dirac.iaea.org was read as raw JSON on 17 November 2026 and holds nine entries for Sri Lanka.) · Both cards now give the directory's nine centres by name with 16 linear accelerators and 13 cobalt-60 units, three centres recording brachytherapy and entry years 2022 to 2026, beside the national plan's 2020 figure of seven centres.
- 2026-11-17 · Sri Lanka: The national programme reports long waits and travel to Maharagama: The card said the programme attributed the waiting list to the concentration of specialised care at a few hospitals. (Second accuracy pass; the raw text of the National Cancer Control Programme's Annual Report 2024 states the waiting list and the concentration of facilities as two separate observations.) · The sentence now reports both observations without giving one as the cause of the other.
- 2026-11-17 · Sri Lanka: The 2020 to 2024 plan has ended, and OnCo found no published successor: The card said the 2024 report described a palliative care framework for 2025 to 2030 as being developed, and its title stated that the successor plan was not published. (Second accuracy pass; the raw text of the Annual Report 2024 says, under a palliative care heading, that work on a National Strategic Plan on Prevention and Control of Cancer for 2025 to 2030 began with a consultative meeting on 29 November 2024.) · The card now reports the passage as written, with the meeting date, and the title says OnCo found no published successor.
- 2026-11-17 · Sri Lanka: An independent authority since 2015 that registers medicines and sets their prices: The card said the regulator licenses pharmacies, checks quality in its own laboratory and fixes a price for every medicine it registers. (Second accuracy pass; the regulator's About Us page and its pricing guideline GL-033 were read as raw text and do not describe pharmacy licensing, and the pricing rule covers new registrations and renewals from July 2025.) · The card now says the authority regulates pharmacies, that a quality assurance laboratory works under it, and that the price rule dates from July 2025 and applies to new registrations and renewals.
- 2026-11-17 · Sri Lanka's 2022 economic crisis and chronic disease care in two provinces: The record described Uva as the poorer, more rural province, which the paper's abstract does not say. (Second accuracy pass; the abstract was read through Europe PMC.) · The description was removed and the sentence now names the two provinces.
- 2026-11-17 · Syria: lede and description: The opening said Syria has had no working national cancer registry since 2011, "so" its figures are IARC estimates borrowed from neighbouring countries, and that the IAEA had "begun working with Syria" after December 2024. IARC's fact sheet gives its method and no cause, and the IAEA was already reviewing Syria's cancer services in 2022. (Second accuracy pass; IARC's GLOBOCAN 2024 fact sheet, the IAEA's articles of 15 June 2023 and 5 June 2025 and Basha et al. 2024 read in raw text.) · The registry's halt, as a 2024 systematic review reports it, and IARC's stated method are now given as two separate facts, and the IAEA sentence describes the programme it announced in June 2025. The same wording was corrected in the description and the closing note.
- 2026-11-17 · Syria: HPV vaccine has not been introduced; hepatitis B vaccine is in the schedule: The card listed pneumococcal conjugate and rotavirus vaccines among those WHO records for Syria, and said WHO's HPV coverage table has no rows for Syria. WHO lists those two vaccines only as planned, for 2028 and 2027, and its coverage table holds HPV rows for 2018 to 2025 with every figure blank. (Second accuracy pass; WHO's vaccine introduction and coverage tables read as JSON from its public API, the coverage table queried for every antigen containing HPV.) · The card now says the two vaccines are planned, not introduced, and that the HPV rows exist but are blank.
- 2026-11-17 · Syria: Few machines and specialists, and no radiotherapy in the northwest during the conflict: The page gave no directory count of radiotherapy machines and its gap said only that the number could not be confirmed. The IAEA's DIRAC directory lists four centres with 7 linear accelerators and 6 cobalt-60 units, and already lists brachytherapy at Al-Bayrouni University Hospital. (Second accuracy pass; the IAEA DIRAC public data feed, fetched 17 November 2026, read as JSON.) · The card, the brachytherapy card and the gap now give DIRAC's centres, machines and entry years (one from 2026, three from 2020), and say that DIRAC does not show which machines are working.
- 2026-11-17 · Syria: A hospital-based national registry from 2002, halted by the conflict: The card said support from IARC became erratic "from 2007" and that hospitals keep their own registries; the table caption and a gap stated as fact that no registry has published since 2011. The review says support had become more erratic "by 2007" and mentions only Al-Bairouni's hospital registry. (Second accuracy pass; full text of Basha et al., BMC Cancer 2024, read through Europe PMC.) · The date is corrected, the hospital registry sentence is limited to Al-Bairouni, and the statements about the absence of a registry are attributed to the review.
- 2026-11-17 · Syria: Since 2025: an IAEA programme to rebuild radiotherapy, and Syria's first full brachytherapy suite: The card and the Ministry of Health record ran together the meeting with the President on 4 June 2025 and the signing of the memorandum, as if both were dated 4 June. The IAEA's article dates only the meeting and says the signing took place during the visit. (Second accuracy pass; the IAEA's article of 5 June 2025 read in raw text from the Internet Archive.) · Both now say the memorandum was signed during the same visit.
- 2026-11-17 · Syria: No trial in our records has a site in Syria, and most patients have never heard of one: The card called the registered cancer studies mostly small single-centre studies, called two Al-Hawash Private University studies surveys, and characterised Syria's published research as a whole. The registry records them as observational studies and none of the other statements was read from a source. (Second accuracy pass; the ClinicalTrials.gov API read as JSON on 17 November 2026.) · The unsupported descriptions were removed or reworded to what the registry records show.
- 2026-11-17 · Syria: A national cancer strategy the committee agreed to finalise after the 2022 review; none found published: The title said a strategy was "agreed in principle in 2022" and the card said the reviewers held meetings at 13 facilities. The IAEA wrote that the committee agreed to finalise the strategy and that meetings were held with stakeholders from 13 facilities. (Second accuracy pass; the IAEA's article of 15 June 2023 read in raw text from the Internet Archive.) · Title and sentence reworded to the IAEA's account.
- 2026-11-17 · Syria: No published registration rules, maturity level or medicines list: The gap said the Ministry of Health's website timed out. Its home page answered on 17 November 2026. (Second accuracy pass; the live home page of moh.gov.sy fetched and read in raw text.) · The gap now says what the home page shows and that only the home page was read.
- 2026-11-17 · Syria: About 27,000 new cancers a year, with breast, lung and bladder cancer leading: The card said hospital studies from Damascus find most cancers are diagnosed late. The two studies cover breast and lung cancer only, and the lung cancer study does not name its city. (Second accuracy pass; abstracts of Nahhat et al. 2023 and Hanafi et al. 2024 read through Europe PMC.) · Reworded to say two hospital studies found breast and lung cancers often diagnosed late.
- 2026-11-17 · Idlib Oncology Centre (Syrian American Medical Society): The summary line stated as fact that this was the only cancer centre in northwest Syria during the conflict, while the record itself notes two further centres opened in 2020. (Second accuracy pass; full texts of Basha et al. 2024 and Atassi et al. 2022 read through Europe PMC.) · The statement is now attributed to the 2024 review.
- 2026-11-17 · Cancer and cancer services in Syria during the conflict: a systematic review: The record said support from IARC became erratic from 2007 and stated the absence of a registry since 2011 as OnCo's own finding. (Second accuracy pass; the paper's full text read through Europe PMC.) · Corrected to "by 2007" and attributed the registry statement to the authors.
- 2026-11-17 · Cancers diagnosed in 2020 by a pathology laboratory serving northwest Syria: The record said no registry operated in northwest Syria. The paper says only that national reporting often excluded the area. (Second accuracy pass; the paper's full text read through Europe PMC.) · Reworded to the authors' statement.
- 2026-11-17 · Kazakhstan: Other figures not found or not verified: The gap card said the IAEA's Directory of Radiotherapy Centres had not been read, and left radiotherapy capacity resting on three national counts. The directory does answer: its table loads from a public data feed. (Second accuracy pass; the directory's public equipment feed for Kazakhstan read as raw JSON on 17 November 2026.) · The gap card and the radiotherapy card now give what the directory lists (26 centres, 31 linear accelerators, 18 cobalt-60 units, 2 gamma knife units, no proton unit, entries dated 2017 to 2026) beside the plan's, the institute's and the Ministry's counts, without choosing between them.
- 2026-11-17 · Kazakhstan: IARC's estimate and the national register: The page said IARC's estimate "is not built from the national register", that IARC "does not use" the register, that the two sets of figures differ because IARC's are modelled, and that IARC's colorectum figure includes the anus. The fact sheet says only that the country-specific incidence data source is "No data" and gives its method; it does not mention the register, explain the difference or define colorectum. (Second accuracy pass; the IARC GLOBOCAN 2024 fact sheet for Kazakhstan read as PDF text.) · The note, the sites table, the register card and the closing advice now state what the fact sheet says in its own words, and the unsupported reasons are removed.
- 2026-11-17 · Kazakhstan: Cancer is commonest in the north and east: The card's title said cancer is found earliest in the cities, but the region with the largest share found at stage I in 2023 was North Kazakhstan (38.6%), ahead of the city of Almaty. The card also attached the cancer plan's explanation for the weakest regions to 2023 figures, when the plan wrote it about 2022. (Second accuracy pass; the institute's Indicators for 2023 (PDF text) and the cancer plan on zakon.uchet.kz read in Russian.) · The title now says that early detection varies widely by region, and the plan's explanation is dated to 2022 and given in the plan's terms.
- 2026-11-17 · Kazakhstan: Breast, cervical and bowel screening by invitation: The card said screening became "free for everyone from 2026" and that from January 2026 it was paid from the state budget so that people without insurance could have it. The order shows every screening step marked as guaranteed free care in the text in force from 1 January 2026, but does not say what applied before, and free care covers citizens, kandas, refugees and permanent residents rather than everyone. The same claim stood in the card on who pays. (Second accuracy pass; order ҚР ДСМ-174/2020 and article 196 of the health Code read in Russian on zakon.uchet.kz.) · Both cards now say what the 2026 text shows and no longer state a change of funding that the sources do not describe.
- 2026-11-17 · Kazakhstan: HPV vaccine for girls nationwide since 2024: The card gave WHO's coverage only from the programme rows (38% in 2024, 47% in 2025). WHO also holds administrative rows for the same doses with a larger target, which give 33% and 44%. It also said deaths from cervical cancer fell by 17.0% in 2023; that figure is the fall in the death rate, and deaths went from 602 to 510. (Second accuracy pass; WHO's immunisation coverage rows for Kazakhstan queried without the narrow antigen filter, and the institute's Indicators for 2023, table 2.1.) · The card gives both sets of WHO rows with their targets, and the cervical cancer figure is given as a rate with the two counts.
- 2026-11-17 · Kazakhstan: What Kazakhstan's cancer register reports: The 2025 row placed a newspaper's summary of the Ministry's results under headings that say "excluding skin" and "crude", and the first card said the Ministry reported the 2025 count. The article says neither whether skin cancer is counted nor which kind of rate 61 per 100,000 is. (Second accuracy pass; Kazakhstanskaya Pravda's article of 5 January 2026 read as page text.) · The table note says what the article leaves unstated, and the 2025 figures are attributed to the newspaper's report of the Ministry's results.
- 2026-11-17 · Kazakhstan: One electronic register for every cancer patient: The card said 90.1% of new cancers were confirmed by histology and that the register is what the researchers cited on the page used. The report's figure is morphological confirmation, which includes cell examination, and the studies used the national electronic health system, oncology service data and screening reports. (Second accuracy pass; the institute's Indicators for 2023 (PDF text) and the three abstracts through Europe PMC.) · The card now says morphological confirmation and names the data each study used.
- 2026-11-17 · Kazakhstan: A green corridor: The card said the plan funds rapid cancer diagnosis centres and text message reminders. The plan mentions rapid diagnosis centres only in its introduction and marks text reminders as needing no funds; only the regional situation centres carry a budget. (Second accuracy pass; the cancer plan's action table read in Russian on zakon.uchet.kz.) · The sentence now says which of the three the plan budgets for.
- 2026-11-17 · Kazakh Institute of Oncology and Radiology (KazIOR): The record said 117,616 patients observed for five years or more were 55.4% of the 218,186 registered. The report gives both numbers and the 55.4% share but not the denominator, and 117,616 is 53.9% of 218,186. (Second accuracy pass; the institute's Indicators for 2023, section 3.9, read as PDF text.) · The record gives the share as the report states it and notes that the denominator is not given.
- 2026-11-17 · Akhmedullin et al., cancer mortality trends in Kazakhstan: The record tied the paper to the Kazakh Institute of Oncology and Radiology, said mortality was "stable to falling" and said diagnosis was disrupted in the COVID-19 years. Every author is at Nazarbayev University, and the abstract says overall trends remained stable and does not mention disrupted diagnosis. (Second accuracy pass; the Europe PMC record for the DOI, with author affiliations and abstract.) · The institute link is removed from the paper and the two sentences follow the abstract.
- 2026-11-17 · Semenova et al., evaluation of Kazakhstan's breast screening programme: The record tied the paper to the Kazakh Institute of Oncology and Radiology and to the Ministry of Health. Its authors are at Nazarbayev University, the National Research Oncology Center in Astana and other bodies, not those two. (Second accuracy pass; the Europe PMC record for the DOI, with author affiliations.) · The paper now links the national oncology centre in Astana only.
- 2026-11-17 · Dyussupov et al., radiation and cancer mortality around Semipalatinsk: The record tied the paper to the Kazakh Institute of Oncology and Radiology and said the effect of nuclear testing is "still measurable" in cancer deaths. Its authors are at Semey Medical University, Shakarim University and the Ministry of Health, and the abstract's highest risks are for 1965 to 1974. (Second accuracy pass; the Europe PMC record for the DOI, with author affiliations and abstract.) · The paper now links the Ministry of Health, and the plain-language sentence gives the dose relation and the years of highest risk.
- 2026-11-17 · Aimagambetova et al., cervical screening in Kazakhstan and Central Asia: The record tied the paper to the Kazakh Institute of Oncology and Radiology. Its authors are at Nazarbayev University, the University Medical Center in Astana and a perinatal centre in Aktobe. (Second accuracy pass; the Europe PMC record for the DOI, with author affiliations.) · The institute link is removed from the paper.
- 2026-11-17 · Ishkinin et al., COVID-19 and cancers covered by screening in Kazakhstan: The plain-language sentence said cancers missed in 2020 and 2021 may have been found later and at a later stage. The abstract reports detection rates only and says nothing about stage. (Second accuracy pass; the Europe PMC abstract for the DOI.) · The sentence now reports the fall in detection and says the abstract does not address stage.
- 2026-11-17 · National Scientific Oncology Center (Astana): The record called the centre a lead body in the cancer plan. The plan names it, by agreement, as one of several co-executors. (Second accuracy pass; the cancer plan's action table read in Russian on zakon.uchet.kz.) · The record now says co-executor, and adds what the IAEA's directory lists for the centre.
- 2026-11-17 · Cambodia: HPV vaccination at 85% to 87% of girls in WHO's estimates since 2023: The card said WHO's data had no HPV coverage figure for Cambodia before 2023 and no figure for boys, and credited the figures to WHO and UNICEF. WHO's data also hold administrative rows: a 2018 final dose of 96.18% for girls and 94.74% for boys, and a first dose in girls of 99.3%, 93.0% and 104.35% in 2023 to 2025, all higher than the programme estimates the card quoted. (Second accuracy pass; WHO's immunisation data API read as raw JSON for every HPV antigen, not only the programme-estimate rows the first query asked for.) · The card now gives both sets of figures, says which is which, says the estimates (not all of WHO's data) begin in 2023 and are blank for boys, and credits them to WHO's HPV estimates.
- 2026-11-17 · Cambodia: lede, registry section and gap on counted figures: The page said Cambodia has no population-based cancer registry in IARC's sources and gave that as the reason IARC models its figures. IARC's fact sheet names no Cambodian data source for incidence and states its method; it does not say there is no registry or give a reason. (Second accuracy pass; IARC's GLOBOCAN 2024 fact sheet read as text from the PDF, and WHO's Global Health Observatory row on the existence of a population-based cancer registry read as raw JSON (2021: No).) · The lede, the description, the registry section and the gap now state WHO's dated row and IARC's stated method as two separate facts, with no cause given.
- 2026-11-17 · Cambodia: Government sources not read: The gap said the IAEA's DIRAC radiotherapy directory was not read and that radiotherapy capacity was not described. DIRAC's public data feed does list Cambodia. (Second accuracy pass; DIRAC's data feed read as raw JSON on 17 November 2026.) · The gap now reports three radiotherapy centres, all in Phnom Penh, with one linear accelerator each (Calmette Hospital, entry dated 2020, also with brachytherapy; the Khmer-Soviet Friendship Hospital and the Orange Cancer Clinic, entries dated 2024), no cobalt-60 unit, and a note that entries are self-reported.
- 2026-11-17 · Cambodia: About one woman in seven screened for cervical cancer: The card said recent studies had tested HPV self-sampling in rural Cambodia. The paper linked is a costing study of a self-sampling programme, not a test of it. (Second accuracy pass; the paper's abstract read through Europe PMC.) · The card now says the study costed a community programme in Siem Reap Province in which 7,524 women were screened, and links the garment workers paper it mentions by title.
- 2026-11-17 · Cervical cancer at a national cancer centre in Phnom Penh, 2019: stage and histology: The plain-language line said most women reach the centre with advanced cancer. The study's 57.6% is of the 309 women with a recorded stage; 42 of 351 had none. (Second accuracy pass; the abstract read through Europe PMC and the full paper read from the journal's PDF.) · The line now says more than half of the women whose stage was recorded.
- 2026-11-17 · Khmer-Soviet Friendship Hospital, Phnom Penh: The record listed a gynaecologic oncology programme and credited the 2023 study to the gynaecology department and one Japanese centre. The paper names a gynaecology department and a Department of Obstetrics and Gynecology, with authors from five bodies. (Second accuracy pass; the paper's affiliations read from the journal's PDF and Europe PMC.) · The programme line and the authorship wording now follow the paper.
- 2026-11-17 · Ecuador: Radiotherapy machines as the IAEA's directory lists them: The gap said the IAEA's radiotherapy directory (DIRAC) returned no data, so no machine count was given, and that the full 2019 imPACT Review report is not published. (Second accuracy pass; the DIRAC public data feed (EquipmentReport and GetCountriesAndRegions) read as raw JSON, and WHO's Cancer Country Profile 2020 for Ecuador read as PDF text.) · The gap now gives what DIRAC lists on 17 November 2026: 15 centres, 12 with radiotherapy, 35 linear accelerators, 2 cobalt-60 units, 1 gamma knife and 5 centres with brachytherapy, each centre named with its entry year (2017 to 2026). It adds WHO's 2019 figure of 7.8 external beam units per 10,000 cancer patients, and says the full imPACT report was not found instead of not published.
- 2026-11-17 · Ecuador: HPV vaccine in the national schedule since 2014: The card called WHO's programme coverage figures the coverage Ecuador reported, and said boys were added in 2024, which no source read states. (Second accuracy pass; WHO immunisation data API (MT_AD_COV_LONG, Ecuador, all HPV antigen codes) read as raw JSON.) · The figures are now attributed to WHO's HPV estimates, the administrative figures Ecuador reported are given where they differ (70% for 2020, 28% for 2021, 81.86% for 2023), the 2024 and 2025 last-dose rows are described as repeating the first-dose figures without dose counts, and the title and text say only that WHO's rows for boys begin in 2024.
- 2026-11-17 · Ecuador: ARCSA's list of recognised regulators: The gap said ARCSA's list of high-surveillance countries was not reachable. (Second accuracy pass; the instruction document named on ARCSA's gob.ec procedure page returned an error page, and its Internet Archive copy of 27 July 2024 was read as PDF text.) · The gap now reports version 3.0 of the instruction (August 2021), whose annex 3 lists agencies in 43 countries and the European Medicines Agency, and says OnCo could not confirm it is the list in force.
- 2026-11-17 · Ecuador: About 31,000 new cancers a year: The card placed the stomach cancer death cluster in the Andean provinces of Bolívar, Los Ríos, Chimborazo, Tungurahua and Cotopaxi; the paper does not describe those five provinces that way, and its cluster analysis covers 2010 to 2015. (Second accuracy pass; the abstract of Montero-Oleas et al. (PMID 29096319) read through Europe PMC.) · The card now gives the cluster as the paper does, the most likely high-occurrence cluster for 2010 to 2015, and reports the authors' own remark that clusters lie along the Andes. The paper's record was reworded the same way.
- 2026-11-17 · Ecuador: IARC and registry figures (note under the profile): The note said the five high-quality registries cover about 41% to 45% of the population, joining two different measures, and stated as fact an inference about IARC's method; the opening line said cancer has been counted in Quito since 1984, the year the registry was created. A table note gave differences in diagnosis as a reason for the spread between registries, which the strategy does not say. (Second accuracy pass; the Ministry's 2017 strategy read as PDF text, the Quito registry page as HTML and Cueva et al. 2022 in full text through Europe PMC.) · The two coverage figures are now given separately with their sources (45% for the registries in the 2017 strategy, 41.4% for the five judged of high quality in the 2022 paper), the IARC sentence is worded as a reading of IARC's own description, the opening line says Quito has had a registry since 1984, and the table note says the strategy gives no explanation.
- 2026-11-17 · Ecuador: The Constitution promises free specialised care for catastrophic illness: The card said the 2017 strategy lists IESS and the armed forces and police institutes as the social security members of the public network; they appear only in its list of abbreviations. (Second accuracy pass; Acuerdo Ministerial 0059-2017 and the strategy read as PDF text.) · The card now gives the Constitution's own definition of the network and says the three institutes are named in the strategy's abbreviations. The Ministry's record was reworded the same way.
- 2026-11-17 · Ecuador: Where care is costly, people turn to cattle ivermectin: The plain text read the abstract's 19% as a share of all participants and said specialists found no scientific evidence; the abstract's wording is narrower and its denominator is unclear. (Second accuracy pass; the abstract of Jiménez-Gaona et al. (DOI 10.3390/nursrep13010030) read through Europe PMC.) · The plain text now keeps the abstract's wording: 19% of the participants diagnosed with cancer took ivermectin-based medicines, 81% used it for other diseases, and specialists said there is no scientific knowledge about its use in people.
- 2026-11-17 · Ecuador: From a 2017 strategy and a 2019 review to a five-year national cancer plan: The card said the strategy was prepared with clinicians from SOLCA, the social security hospitals and the registry; the document lists them as its external validation team, and the hospitals are those of the police and the armed forces. (Second accuracy pass; the strategy read as PDF text.) · Reworded to say its external validation team included clinicians from SOLCA, the police and armed forces hospitals and the Registro Nacional de Tumores.
- 2026-11-17 · Cancer incidence and mortality in Quito, 1985 to 2017: The summary line said incidence and mortality both rose by about 2% a year; mortality in women rose by 1.3% a year. (Second accuracy pass; the paper's abstract and full text read through Europe PMC (PMC9651078).) · The line now gives incidence at about 2% a year in both sexes and mortality at 2.0% a year in men and 1.3% in women.
- 2026-11-17 · Zimbabwe: HPV vaccination for girls since 2018: The page gave HPV vaccination as 90% of girls in 2025 with no qualifier, said there was no figure for 2020, and credited the figures to WHO and UNICEF. The 90% is WHO's estimate of programme coverage. The same WHO data give 63% of girls vaccinated by age 15 in 2025, an official last-dose figure of 55% and an administrative one of 42.9%, and they hold coverage by age 15 for 2020 (88% first dose, 64% last dose). (Second accuracy pass; WHO's immunisation data API read as raw JSON without the narrow antigen filter.) · The description, card title and text now say programme coverage, give the by-age-15, administrative and official rows beside it, say the 2020 programme rows are blank and that rows for boys exist and are blank, and name the source as WHO's HPV estimates.
- 2026-11-17 · Zimbabwe: the Ministry's cancer strategy, radiotherapy and what patients pay: The gap said no primary source on radiotherapy capacity was read and that the IAEA's DIRAC directory was not read. DIRAC's public data feed answers. (Second accuracy pass; the IAEA DIRAC equipment feed read as raw JSON on 17 November 2026.) · The gap now gives what DIRAC lists: three centres, with one linear accelerator at Oncocare Zimbabwe in Harare (entry dated 2022) and brachytherapy but no linear accelerator or cobalt-60 unit listed at Parirenyatwa (2021) and Mpilo Central Hospital in Bulawayo (2020), with a note that entries are self-reported and dated.
- 2026-11-17 · Zimbabwe: a registry that records HIV status: The card said childhood Kaposi sarcoma and lymphoma rose with HIV and fell with treatment, and that among adults on HIV treatment three cancers are the commonest. The first is the authors' interpretation of a peak in incidence, and the second comes from one clinic in Harare. (Second accuracy pass; both abstracts read in Europe PMC's raw record.) · The card now says the authors link the 1996 to 2001 peak to HIV prevalence and the availability of treatment, and that the cancer ranking is among people aged 16 and over starting treatment at one Harare clinic.
- 2026-11-17 · Zimbabwe: what WHO's health spending figures show: A study was described as one of 396 patients in Zimbabwe that found barriers lengthened the time to a first visit and that diagnosis took longer in less-resourced regions. The study had 1,021 participants in two countries, was cross-sectional, and its finding on less-resourced regions is not given for Zimbabwe alone. (Second accuracy pass; the abstract read in Europe PMC's raw record.) · The sentence now gives the study's size and both countries, and reports associations as the abstract does.
- 2026-11-17 · Zimbabwe: IARC's method note: The note said the Zimbabwe National Cancer Registry is the country's population-based registry, in a sentence that implied it is the registry IARC used. IARC's fact sheet names no registry. (Second accuracy pass; IARC's GLOBOCAN 2024 fact sheet read with pdftotext and the African Cancer Registry Network's page read as raw HTML.) · The note now states the two facts separately and says the fact sheet does not say which registry IARC used.
- 2026-11-17 · Ministry of Health and Child Care, Zimbabwe: The record gave the African Cancer Registry Network's page for the cancer registry as the Ministry's website. (Second accuracy pass; the page read as raw HTML is the registry's member page on afcrn.org, and the live record showed it under Website.) · The website field was removed; the page remains cited as the source of the summary.
- 2026-11-17 · Parirenyatwa Group of Hospitals: The record gave the African Cancer Registry Network's page for the cancer registry as the hospital's website. (Second accuracy pass; the page read as raw HTML is the registry's member page on afcrn.org.) · The website field was removed; the page remains cited as the source of the summary.
- 2026-11-17 · Chile: How Chile counts: regional population registries for adults, a national registry for children: The card and the page's opening said Chile has no national registry for adult cancer yet. The Ministry's own 2022 plan, which the page cites, says the Registro Nacional de Cáncer has been introduced step by step since a 2019 pilot, with 27 establishments reporting 7,481 records by July 2021, and that it was not yet representative of the country in December 2021. (Second accuracy pass; the Plan Nacional de Cáncer 2022-2027 read in raw text (pdftotext), chapter IV.) · Both passages now say what the plan says, with its dates, and that OnCo has not read a later account. The norm on population registries is now dated as its cover and decree date it (November 2024, approved April 2025).
- 2026-11-17 · Chile: Five-year survival of adults with cancer in four regional registries: The table gave thyroid cancer 76.3% observed and 94.1% relative survival. Those are the plan's figures for non-melanoma skin cancer; its thyroid figures are 86.6% and 91.4%. (Second accuracy pass; figure 134 of the Plan Nacional de Cáncer 2022-2027, read in raw text and checked on the rendered page of the PDF.) · Thyroid row corrected to 86.6 and 91.4.
- 2026-11-17 · Chile: Seven public hospitals give radiotherapy, four of them brachytherapy too: The card and a gap note said the IAEA's directory of radiotherapy centres could not be read, and the gap list said no count of machines outside the public network was found. The directory's public feed lists 24 centres in Chile with 43 linear accelerators, no cobalt-60 units and two gamma knife units, eight of them reporting brachytherapy. The card also called the plan's count of treatments courses, and cited the wrong sections of the plan. (Second accuracy pass; the IAEA DIRAC feed (EquipmentReport, read 17 November 2026) and table 18 of the Plan Nacional de Cáncer 2022-2027 in raw text.) · The card now gives DIRAC's count beside the plan's seven public hospitals, with the years of the entries, and the gap note says only that DIRAC does not label which centres are public.
- 2026-11-17 · Chile: HPV vaccine for girls at school since 2014 and for boys since 2019: The card said WHO and UNICEF estimate that 92% of girls in the target group were vaccinated in 2025. The series is WHO's estimate of first-dose coverage, and the source read does not name UNICEF. (Second accuracy pass; WHO's Global Health Observatory indicator SDGHPVRECEIVED and WHO's immunisation coverage data for Chile, both read as JSON.) · The figure is now described as first-dose coverage, with the last-dose series for girls beside it (87% in 2024, 92% in 2025).
- 2026-11-17 · Chile: 314 trials in OnCo list a site in Chile, almost all run by companies: The card named SPOTLIGHT and GLOW among stomach cancer trials with a site in Chile. Only SPOTLIGHT (NCT03504397) lists Chile; GLOW (NCT03653507) does not. (Second accuracy pass; OnCo's ClinicalTrials.gov capture in the repository, recounted, and the registry record for NCT03504397.) · The card now names SPOTLIGHT and says that GLOW lists no site in Chile. The counts (314 trials and their phases and sponsors) were recounted and stand.
- 2026-11-17 · Chile: A cancer foundation with its own cancer cover scheme: The card, then titled "A cancer charity that sells cancer cover", called Fundación Arturo López Pérez a non-profit institute that sells a prepaid plan. Neither of the foundation's pages cited says non-profit, prepaid or sells. (Second accuracy pass; the foundation's history page and its Convenio Oncológico benefits page read in raw text.) · Title and opening reworded to what the pages say: a cancer foundation that runs its own cancer cover scheme.
- 2026-11-17 · Chile: Preventive gallbladder removal for people aged 35 to 49 with gallstones, guaranteed since 2006: The card said gallbladder cancer is unusually common in Chile and that studies disagree on how much the guarantee has helped. The 2024 study cited speaks of one of the highest death rates in the world, and neither abstract says the studies disagree. (Second accuracy pass; the abstracts of Samaniego and colleagues (2024) and Mardones and Frenz (2019), resolved by DOI through Europe PMC.) · Reworded to the 2024 study's statement on mortality, and to say that the two studies read do not settle how much the guarantee has helped.
- 2026-11-17 · Chile: A ministry committee funds high-cost cancer drugs for public patients outside GES and Ricarte Soto: The card dated the drug list to December 2021. The plan's table is headed as updated in March 2021 and is sourced to a resolution of 6 December 2021. (Second accuracy pass; section 2.2.4 of the Plan Nacional de Cáncer 2022-2027 in raw text.) · Both dates are now given as the plan gives them.
- 2026-11-17 · Chile: Five in six people are in the public insurer, FONASA; one in eight in a private ISAPRE: The card said the 2.2% of residents outside FONASA and the ISAPREs include the armed forces, which have their own systems. The report says only that the remainder includes armed forces personnel and other people affiliated to neither. (Second accuracy pass; section 1.1 of the Superintendencia de Salud's 2025 report on the ISAPRE system in raw text.) · Reworded to the report's statement.
- 2026-11-17 · Chile: Organised screening for two cancers only, and coverage below target for both: The card said IARC's CanScreen5 records a national cervical programme. The fact sheet gives the year the programme started (1986) and that its policy is documented in law, for reporting year 2020; it does not use the word national. (Second accuracy pass; the CanScreen5 fact sheet content for Chile read in raw text.) · Reworded to what the fact sheet records, with its reporting year.
- 2026-11-17 · Tunisia: how the IARC estimate compares with the northern registry's: The note said the Northern Tunisia Cancer Registry's 2023 rate was higher than IARC's but included non-melanoma skin cancer, as if IARC's rate did not. IARC's rate includes it too. (Second accuracy pass; IARC's GLOBOCAN 2024 fact sheet for Tunisia read as raw text, where the incidence figures carry the footnote that they include non-melanoma skin cancer.) · The note now says both rates include non-melanoma skin cancer and that they differ in area, method and the standard population each source names.
- 2026-11-17 · Tunisia: The Northern Tunisia Cancer Registry: The card said the registry was run from the Institut Salah Azaïez, that its authors were all there, that the national figures were built from regional registries, and that a registry in Monastir had published breast cancer figures. The papers do not say where the registry is run, several authors work elsewhere, IARC names only a subnational registry, and the Monastir paper used the Cancer Register of the Centre. (Second accuracy pass; the abstracts and open-access full text of the three papers read through Europe PMC, and the IARC fact sheet.) · The title and text now give the author affiliations as published, the 2024 paper's own statement that Tunisia has three regional registries and not yet a national one, IARC's wording, and the right registry for the Monastir study.
- 2026-11-17 · Tunisia: HPV vaccination appears in WHO's data for the first time in 2025: The card credited the figure to WHO and UNICEF and said the cervical cancer paper called for vaccinating girls. The rows come from WHO's immunisation data, and the paper says females. (Second accuracy pass; WHO's immunisation data read as JSON with a wider query for every HPV row, and the paper's abstract through Europe PMC.) · The card now credits WHO's data, quotes the paper's word, and adds what the wider query shows: rows from 2010 that are blank before 2025, 2,977 first doses against a target of 104,151 girls in 2025, and blank rows for boys.
- 2026-11-17 · Tunisia: sources that could not be reached, and radiotherapy: The page said the Internet Archive's copy of the health insurance fund's site did not load, that Tunisia regulates medicines nationally with no source for it, and that radiotherapy capacity was not on the page because the IAEA directory had not been read. (Second accuracy pass; the sites requested again, the Internet Archive copy of 9 August 2025 read, and the IAEA DIRAC public data feed read on 17 November 2026.) · The page now says the archived home page loads but says nothing about cancer cover, drops the unsourced clause, and gives DIRAC's list: 17 centres with 21 linear accelerators and 11 cobalt-60 units, two with brachytherapy, with entry years from 2011 to 2026.
- 2026-11-17 · Estimated cancer incidence in northern Tunisia in 2023: The record said lung cancer led in men and breast cancer in women without the paper's qualifier that skin cancer was set apart, and summed the projections up as colorectal and breast cancer being the growing burden, which the paper does not say. (Second accuracy pass; the abstract read through Europe PMC.) · The qualifier is added and the projections are given as the abstract states them.
- 2026-11-17 · Myanmar: HPV vaccination of 9-year-old girls since 2020: The card credited WHO's HPV coverage estimates to WHO and UNICEF, though the data name only WHO; it gave 8% for 2022 as the share of girls vaccinated without saying it is the first-dose estimate, when the same data hold a last-dose estimate of 44% and a final-dose figure of 36.82% reported by Myanmar; and it said WHO's data do not show when the schedule became one dose, when the introduction rows mark one dose from 2023. (Second accuracy pass; WHO immunisation API queried for every HPV antigen and read as raw JSON; the introduction rows and the Global Health Observatory indicator were read the same way.) · The estimates are credited to WHO alone, the 8% is described as a first-dose estimate with the last-dose estimate and the reported figure beside it, and the one-dose schedule is dated from 2023 with links to the rows.
- 2026-11-17 · Uganda: HPV vaccine for girls nationwide since 2015: The card described WHO's own programme estimates (95% in 2024 and 69% in 2025) as the coverage Uganda reported to WHO. The figures Uganda reported are different: official first-dose coverage of 178% in 2024 and 83% in 2025. It also called the single-dose schedule an inference when WHO's introduction rows mark one dose for 2025. (Second accuracy pass; WHO immunisation API queried for every HPV antigen and read as raw JSON; the introduction rows were read the same way.) · The figures are credited to WHO's estimates, Uganda's reported figures are given beside them with a note on values above 100%, and the one-dose schedule is credited to WHO's rows.
- 2026-11-17 · Angola: HPV vaccination appears in WHO's data in 2025: The page summary called the 61% figure HPV vaccination first reported in 2025. It is WHO's programme estimate; the administrative and official rows, which carry what a country reports, are blank for Angola in every year from 2018 to 2025. (Second accuracy pass; WHO immunisation API queried for every HPV antigen and read as raw JSON.) · The summary now calls it WHO's estimate and the card says the administrative and official rows are blank, with a link to them.
- 2026-11-17 · Uzbekistan: HPV vaccine reaching nine in ten girls or more: The card described WHO's programme estimates of HPV coverage as the coverage Uzbekistan reports to WHO. The official figures Uzbekistan reported are a separate series and are higher, between 98.6% and 99.8% for the final dose in 2019 to 2025. (Second accuracy pass; WHO immunisation API queried for every HPV antigen and read as raw JSON.) · The quoted figures are credited to WHO's estimates and the reported official figures are given beside them, with a link to all the rows.
- 2026-11-17 · Ghana: HPV vaccine for girls nationwide since October 2025: The card said Ghana reported 79% HPV coverage for 2025 to WHO. The 79% is WHO's programme estimate; the administrative and official rows in the same data give first-dose coverage of 0.57% for 2025, from 2,496 doses against a target of 438,299 girls. (Second accuracy pass; WHO immunisation API queried for every HPV antigen and read as raw JSON.) · The 79% is credited to WHO's estimate and the administrative and official figure is given beside it, with a note that the data do not explain the difference.
- 2026-11-17 · Saudi Arabia: HPV vaccine for girls since 2022: The card described WHO's programme estimates of HPV coverage (62% first dose in 2025, 50% in 2024, 47% last dose in 2023) as figures Saudi Arabia reported. The figures it reported use smaller targets and are higher: 68.12% in 2025, 66.28% in 2024 and 65.2% final dose in 2023. (Second accuracy pass; WHO immunisation API queried for every HPV antigen and read as raw JSON.) · The estimates are credited to WHO and the reported administrative and official figures are given beside them with their targets, with a link to all the rows.
- 2026-11-17 · Sri Lanka: HPV vaccine for schoolgirls: The card and its table presented WHO's programme estimates as the coverage Sri Lanka reported, labelled them administrative, and said coverage fell to 31% in 2023. The administrative coverage Sri Lanka reported was 91.56% for the first dose and 91.71% for the final dose in 2023; only WHO's estimate fell. (Second accuracy pass; WHO immunisation API queried for every HPV antigen and read as raw JSON; the introduction rows were read the same way.) · The estimates are labelled as WHO's, the reported administrative figures for 2020 to 2025 are given beside them, the title and gap card no longer say coverage fell, and the 2017 start is credited to WHO's introduction rows.
- 2026-11-17 · Belgium: Radiotherapy is counted only from the IAEA's directory; workforce and waiting times are not on this page: The page said the IAEA's DIRAC radiotherapy database returned only a page shell to a script, so that Belgium's radiotherapy centres and machines could not be read and no figure was given. (Second accuracy pass; DIRAC's public data feed (EquipmentReport, on dirac.iaea.org) read as raw JSON on 17 November 2026.) · The card now gives what DIRAC lists: 36 centres, 91 linear accelerators, no cobalt-60 units and 20 centres reporting brachytherapy, with the years of the entries (32 from 2019, three from 2021, one from 2023), and says the earlier statement was wrong.
- 2026-11-17 · Belgium: HPV vaccination: about four in five girls and three in four boys get a first dose: The coverage figures were described as WHO data as reported for Belgium, when the rows quoted are WHO's own estimates; and the 77%, 69%, 71% and 63% values were said to cover 2021 and 2022 when the same rows give them for 2020 as well. (Second accuracy pass; WHO immunisation data API (MT_AD_COV_LONG) queried for every Belgian HPV row and read as raw JSON.) · The card names each coverage category: WHO's programme estimates (80%, 72%, 73%, 65% for 2025, the same by age 15), the official coverage rows (79.9%, 71.6%, 73.4%, 65.3%), and the administrative rows, which are blank for 2021 to 2025. The earlier step is now dated 2020 to 2022.
- 2026-11-17 · Belgium: A reimbursement commission, agreements with makers, and a reform under way: The card, its title and the page's closing advice said Belgium signs confidential agreements with the maker for many new cancer medicines, and that a medicine is listed after advice from the reimbursement commission. The insurer's pages say only that agreements under articles 81 and 111 are concluded for some innovative medicines and that their public parts are published. (Second accuracy pass; RIZIV-INAMI's pages on the reimbursement reform, on reimbursement of a medicine and on Early and Equitable Fast Access read as raw HTML in French.) · Reworded to what the pages say: agreements for some innovative medicines, of which only the public parts are published, and an ordinary route that runs through the commission.
- 2026-11-17 · Belgium: 912 cancer trials in OnCo's records have a site in Belgium, and EORTC is based in Brussels: The card gave 909 of 5,579 trials with a site in Belgium, 262 recruiting, 250 completed and 636 in phase 3. OnCo's trial data had changed since the card was written and no longer gave those counts. (Second accuracy pass; the counts were recomputed from OnCo's own ClinicalTrials.gov participation records on 17 November 2026.) · Now 912 of 5,578: 260 recruiting, 370 active but not recruiting, 255 completed, 19 terminated, four not yet recruiting and four of unknown status, with 640 in phase 3, and the count is dated.
- 2026-11-17 · Belgium: Survival is published for the whole country, and lung cancer survival is improving: The card said the registry can publish national survival because registration is compulsory and linked to vital status, and that survival was for patients followed from 2015 to 2024. The fact sheet says only that the Crossroads Bank for Social Security provides additional data on vital status, and its tables are headed 2015 to 2024 without saying what the period means. (Second accuracy pass; the Belgian Cancer Registry's Cancer Fact Sheet 2024 and its lung, breast, prostate and colorectal appendices read with pdftotext.) · The causal wording was removed and the period is given as the tables head it. The figures themselves were confirmed.
- 2026-11-17 · Belgium: how these figures compare with GLOBOCAN: The note said IARC's total was labelled ICD-10 C00-97 and called IARC's figure modelled. The fact sheet gives no ICD range; it marks the total as including non-melanoma skin cancer, names the national registry as its data source and gives its method as trends in national rates projected to 2024. (Second accuracy pass; IARC's GLOBOCAN 2024 fact sheet for Belgium read with pdftotext.) · The ICD label was removed, the data source and method are stated as the fact sheet gives them, and the breast figure is described as the one GLOBOCAN prints for women and for both sexes.
- 2026-11-17 · Belgium: A national institute that supplies isotopes for nuclear medicine worldwide: The card called IRE ELiT a subsidiary of the institute, said its gallium-68 generators are used for PET scans, and credited a news item about actinium-225 to IRE ELiT. The institute's home page says only that the group has two entities and does not mention PET. (Second accuracy pass; the home page of ire.eu read as raw HTML.) · Reworded to the page's own description, and the actinium-225 item is credited to the news on IRE's site dated 27 May 2026.
- 2026-11-17 · National Institute for Health and Disability Insurance (RIZIV-INAMI): The record said the insurer decides with the sickness funds which medicines are reimbursed, that its agreements for innovative medicines are confidential, and that it runs the BeNeLuxA collaboration. None of the three is stated on the insurer's pages that the record cites. (Second accuracy pass; RIZIV-INAMI's pages on its partners, the reimbursement reform and the BeNeLuxA initiative read as raw HTML in French.) · Reworded: its lists set which medicines are reimbursed, agreements under articles 81 and 111 have their public parts published, and Belgium takes part in BeNeLuxA.
- 2026-11-17 · Cuba: One national registry, published about three years after diagnosis: The page said the National Cancer Registry's figures appear about two years after diagnosis. The yearbook published in 2025 carries cancers diagnosed in 2022, and the one published in 2024 carried 2020, so the delay is about three to four years. The same card and the note under the IARC table also said that IARC does not use the registry's counts, which IARC's fact sheet does not say. (Second accuracy pass; the Ministry of Public Health's Anuario Estadístico de Salud 2024 (edition 2025) and 2023 (edition 2024) and IARC's GLOBOCAN 2024 fact sheet for Cuba were read as raw PDF text.) · The delay is now given as about three years with both editions named, on this card and on the card about yearly publication. The IARC sentences now give the fact sheet's own words: no country-specific data source for incidence, and the method it names.
- 2026-11-17 · National Cancer Registry of Cuba: The record said the registry's counts appear in the yearbook about two years after diagnosis. The 2022 counts were published in the 2025 edition. (Second accuracy pass; the Anuario Estadístico de Salud 2024 (edition 2025) was read as raw PDF text.) · The record now says about three years and names the edition.
- 2026-11-17 · Cuba: In 2023, 40% of the basic medicines list was reported missing, and five cancer drugs were affected: The card said OnCo had found no official shortage count later than July 2023, and a note on what is missing said BioCubaFarma's page of missing medicines showed a heading but no list. That page links a list for every month from January 2022 to January 2025. The card also credited BioCubaFarma's president with saying the national industry supplies 61% of the basic list, which the newspaper states in its own voice, and its title said the 251 missing medicines included five cancer drugs, which the report does not say. (Second accuracy pass; the Internet Archive copies of BioCubaFarma's missing medicines page and its January 2025 list (PDF) and of the Granma report republished by Guerrillero on 28 July 2023 were read as raw text.) · The card and the note now give the January 2025 list: 255 of the 395 medicines BioCubaFarma supplies were missing at the end of that month, with six cancer medicines named on its first-priority list. The 61% figure is credited to the newspaper and the title no longer says the five cancer drugs were among the 251.
- 2026-11-17 · BioCubaFarma: The record credited BioCubaFarma's president with the figures of 251 missing medicines and a 61% share of the basic list; the newspaper report gives both in its own voice. The record also called the group a state group, which the page cited does not say. (Second accuracy pass; the Internet Archive copies of the Granma report republished by Guerrillero on 28 July 2023 and of BioCubaFarma's own description were read as raw text.) · The two figures are now credited to the report, the president's statements are limited to what he is reported as saying, and the description follows the group's own words.
- 2026-11-17 · Cuba: HPV vaccine for girls aged nine from October 2025: The card said Cuba reported to WHO that 94% of the girls targeted in 2025 were vaccinated. The 94% is WHO's own programme coverage estimate; the administrative and official coverage rows give 93.65%. The card also said WHO recorded the vaccine as not introduced from 2018 to 2024, when its rows say so for every year from 2006. (Second accuracy pass; WHO's immunisation coverage and introduction rows for Cuba were read as raw JSON, using the wider query for every HPV antigen rather than the programme estimate rows alone.) · The card now names each coverage category: 93.65% administrative and official first-dose coverage (54,064 doses for 57,727 girls), 94% in WHO's programme estimate, and blank rows for final dose, coverage by age 15 and boys. The introduction years are corrected and the link now opens all the rows.
- 2026-11-17 · Cuba: Radiotherapy machines: the IAEA directory lists nine centres, with no official Cuban count: The page said a count of radiotherapy machines could not be read because the IAEA's Directory of Radiotherapy Centres did not return readable text. The directory's table loads from a public data feed that does. (Second accuracy pass; the IAEA DIRAC equipment feed was read as raw JSON on 17 November 2026 and summed by centre.) · The note and the workforce card now give what DIRAC lists for Cuba: nine centres, six linear accelerators (four of them at the national institute in Havana) and 12 cobalt-60 units, brachytherapy at five centres, each centre named with the year of its entry, which ranges from 2022 to 2026.
- 2026-11-17 · Cuba: No trial in OnCo's records has a site in Cuba: The card said Cuban trials are registered in the Cuban registry instead of ClinicalTrials.gov. ClinicalTrials.gov itself lists 27 studies for cancer with a location in Cuba; none is among the trials OnCo holds. The card about trial rules also said that WHO's listing makes Cuban trials findable through WHO's search portal, which the WHO page cited does not say. (Second accuracy pass; the ClinicalTrials.gov API was queried for cancer studies with a location in Cuba and its JSON read, and WHO's list of primary registries was read as raw page text.) · The card now gives the count of 27 and says none is in OnCo's data. The sentence about WHO's portal is removed from the card and from the record for the national trials coordinating centre.
- 2026-11-17 · Cuba: Cancer vaccines and antibodies made in Cuba: The card said all three Cuban products were tested with researchers from the Center of Molecular Immunology and that BioCubaFarma, called a state group, makes them. The nimotuzumab paper's listed affiliation is a Havana hospital, and BioCubaFarma's page does not say either thing. A note also said the regulator's entry for nimotuzumab had not been read. (Second accuracy pass; the three abstracts and affiliation lists were read through Europe PMC and the regulator CECMED's entries for CIMAvax-EGF, Vaxira and CIMAher were read as raw page and PDF text.) · The card now says what CECMED's entries show: the Centro de Inmunología Molecular holds the registration of all three and is a manufacturer of each, and nimotuzumab (CIMAher) has been registered in Cuba since 19 February 2002.
- 2026-11-17 · Nimotuzumab or placebo with radiotherapy for high-grade glioma: The record gave the absence of skin rash as a result of the trial; the abstract states it only as background about the antibody. It also said the trial lies behind the drug's use in Cuba and other countries that approve it, which the abstract does not say. (Second accuracy pass; the abstract was read through Europe PMC after resolving the DOI, and CECMED's entry for CIMAher was read as raw PDF text.) · The skin toxicity statement is now given as the abstract's background, and the closing line states only the Cuban registration and indication that CECMED's entry gives.
- 2026-11-17 · CIMAvax-EGF: The summary line said the survival gain was significant only in patients who completed the first four doses. The abstract also reports significance by a weighted log-rank test. The record also said the vaccine was developed at the Centro de Inmunología Molecular, where the regulator's entry names it as holder and manufacturer. (Second accuracy pass; the abstract was read through Europe PMC and CECMED's summary of product characteristics as raw PDF text.) · The summary line now gives all three results: not significant by the standard test, significant by the weighted test and in patients who had at least four doses. The centre is described as maker and registration holder.
- 2026-11-17 · Racotumomab: The record said the vaccine was in use in Cuba before its current registration entry under an earlier registration. The regulator's entry gives one date, 26 May 2025, and no history. (Second accuracy pass; CECMED's entry and summary of product characteristics for Vaxira were read as raw page and PDF text.) · The record now says that OnCo has not read whether there was an earlier Cuban registration.
- 2026-11-17 · Cuba: Cervical smear testing, counted since 1970, now reaches about a quarter of the women it reached in 2019: The card said testing began in 1970 and described stage 0 findings as precancers. The yearbook's table starts in 1970 without giving a start date for the programme, and it does not define its stages. (Second accuracy pass; tables 77 to 79 of the Anuario Estadístico de Salud 2024 were read as raw PDF text.) · The title now says counted since 1970 and the card gives stage 0 as the yearbook prints it.
- 2026-11-17 · Cuba: A shrinking, ageing population changes the rates: The card said an older population pushes crude rates up and that IARC's larger population figure raises its estimated counts. Neither source says so. (Second accuracy pass; the Anuario Estadístico de Salud 2024 and IARC's GLOBOCAN 2024 fact sheet were read as raw PDF text.) · The card now gives only the figures: deaths rose by 511 while the crude rate rose from 247.5 to 261.2 per 100,000, a quarter of people are aged 60 or over, and IARC's population is about 1.08 million larger.
- 2026-11-17 · Czechia: HPV vaccination for girls and boys, paid in full from 11 to 15: The card and the gap card said no coverage figure more recent than 2019 had been read from WHO's immunisation data, and the plain text called HPV the virus behind most cervical cancers, which the cited page does not say. (Second accuracy pass; WHO's immunisation data service queried for every HPV antigen and the JSON read in raw text, with the cervix.cz page for parents.) · The card now gives WHO's rows for 2025 by category: first-dose administrative and official coverage of 69.97% of girls and 49.94% of boys, WHO estimates of 70% and 50% for programme coverage and 70% and 43% by age 15, and blank final-dose rows for 2018 to 2025. The plain text now says HPV can cause these cancers.
- 2026-11-17 · Czechia: Why IARC's estimate and the Czech registry differ: The page said IARC's Czech figures are national registry rates projected to 2024. That is true of incidence only: the fact sheet gives WHO national data as the source for mortality. The note also inferred that IARC's model leaves out most of the registry's skin cancers, which no source says. (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet for Czechia read with pdftotext.) · The note and the gap card now give the two sources separately, the inference is removed, and the rate comparison cites the fact sheet's footnote that its incidence includes non-melanoma skin cancer.
- 2026-11-17 · Czechia: Radiotherapy: 38 centres and 67 high-energy treatment machines in the IAEA's directory: The card said the Czech figures were last updated in 2021 and gave no machine types. The country total is dated 2021, but the entries for the centres themselves are dated 2017 to 2020. (Second accuracy pass; IAEA DIRAC country and centre feeds read as raw JSON on 17 November 2026.) · The card now gives the centre rows: 59 linear accelerators in 24 centres, 7 cobalt-60 units and 1 gamma knife, 16 centres with brachytherapy, the entry years, and a note that one institute appears twice.
- 2026-11-17 · Czechia: Three established screening programmes and two newer early detection programmes: The card said the lung programme started on 1 January 2022 without saying that the national plan describes it as a pilot programme. (Second accuracy pass; English text of the National Oncology Plan 2030 read with pdftotext.) · The card now says it was launched as a pilot programme on that date.
- 2026-11-17 · Czechia: A network of comprehensive cancer centres, with separate centres for children and blood cancers: The plain text said the centres work with the hospitals near patients' homes, while the national plan reports large gaps in that cooperation, and it applied the plan's figure of more than 70% to all the listed centres rather than to the comprehensive cancer centres. (Second accuracy pass; English text of the National Oncology Plan 2030 read with pdftotext, and the Czech Oncological Society's list on Linkos.) · The clause is removed and the 70% figure is tied to the comprehensive cancer centres.
- 2026-11-17 · Czechia: Registry data, with stage, are open to anyone: The card said a matching data summary exists for relative survival, and opened with an unsourced comparison with other countries. The portal's list of registry summaries shows incidence, deaths and prevalence, by region and by district, and no survival summary. (Second accuracy pass; the NZIP list of registry data summaries and the incidence methodology note read in raw text.) · The card now lists the summaries the portal shows, and the comparison is removed.
- 2026-11-17 · Czechia: lede, stage, national plan, centre medicines and treatment programme cards: Five smaller statements went beyond their sources: that the registry has counted every cancer since 1976, that the projected rise in incidence is mainly from ageing, a paraphrase of the plan's fourth objective, that the plan expects patient numbers to grow, and that a doctor can propose a treatment programme. (Second accuracy pass; ÚZIS's registry page, the National Oncology Plan 2030 and section 49 of Act 378/2007 Coll. read in raw text.) · Each now follows the source's wording: a population-wide registry of individual tumour records since 1976, ageing together with unhealthy lifestyles, the plan's own objective, numbers that are increasing, and a legal or natural person as proposer.
- 2026-11-17 · Institute of Health Information and Statistics of the Czech Republic (ÚZIS ČR): The record's summary line said the National Screening Centre manages the cancer screening programmes. The national plan says it provides coordination, methods and infrastructure for introducing, running and evaluating them. (Second accuracy pass; English text of the National Oncology Plan 2030 read with pdftotext.) · The line now says the centre coordinates and evaluates the programmes.
- 2026-11-17 · Greece: First funded by the EU recovery fund, and not yet evaluated: The card said the EU recovery fund paying for the screening programmes had ended in December 2025 and that how they were funded afterwards was not established; it also credited to the OECD profile a statement that the text-message invitations draw on the COVID-19 vaccination approach, which the profile makes about mobile screening units. (Second accuracy pass; the Ministry of Health press release of 13 October 2026 and the OECD and European Commission Country Cancer Profile Greece 2025, both read in raw text.) · The card now gives the OECD statement in its own tense, the Ministry's statement that the programme is a permanent national policy financed entirely from national resources, and the profile's sentence about mobile units; the title and the two gap cards were changed to match.
- 2026-11-17 · Greece: Prolamvano: three cancer screening programmes for insured and uninsured people alike: The July 2025 totals were described as quoted from the Deputy Minister of Health, when they are in the article's own text and the official quoted is the Alternate Minister; bowel screening was described as sending free home self-tests, when the launch report said text messages offering free self-tests were expected from June 2024. (Second accuracy pass; CNN Greece reports of 26 April 2024 and 7 July 2025 and the Ministry of Health press release of 13 October 2026, read in raw text.) · Attribution and wording corrected, and the Ministry's own totals of October 2026 added with their date.
- 2026-11-17 · Greece: HPV vaccination for girls and boys, with coverage at about half: The page said current HPV coverage had been looked for and not found. WHO's immunisation data hold the figures Greece reported for 2025 and WHO's own estimates for 2022 to 2025. (Second accuracy pass; WHO immunisation data queried for every HPV row for Greece and read as raw JSON.) · The card now gives the reported 2025 figures for girls and boys, names them as administrative and official coverage, gives WHO's estimates separately, and says which rows are blank.
- 2026-11-17 · Greece: Seven radiotherapy machines per million people, and long waits in public hospitals: The card and a gap card said the IAEA's DIRAC directory had not been read. Its data feed is public. (Second accuracy pass; the DIRAC equipment feed as saved on 17 November 2026, read as raw JSON.) · The card now gives DIRAC's 30 centres, 54 linear accelerators, 3 cobalt-60 units and 1 gamma knife, with the years of the entries, beside the OECD's rate for 2022.
- 2026-11-17 · Greece: 440 cancer trials in OnCo's records have a site in Greece: The detail gave 439 trials, 341 of them phase 3 and 182 active but not recruiting, which disagreed with the card's own title. (Second accuracy pass; recount from the ClinicalTrials.gov participation data held in OnCo.) · The detail now gives 440 registered trials, 342 phase 3 and 183 active but not recruiting, with the date of the count and of the registry fetch.
- 2026-11-17 · Greece: GLOBOCAN note and death table: The note said IARC labels its total ICD-10 C00-97, which the fact sheet does not, stated as fact that Greece has published no national count of new cancers, and credited Eurostat's death figures to the Hellenic Statistical Authority, which Eurostat does not name. (Second accuracy pass; the IARC GLOBOCAN 2024 fact sheet for Greece, the Eurostat causes-of-death table and its metadata, and the OECD profile, read in raw text.) · The note now quotes the fact sheet's footnote on non-melanoma skin cancer and its stated data source and method, gives the OECD's statement on registry data, says no published count was found, and describes the deaths as Eurostat publishes them. The same wording was corrected in the description, the opening paragraph, the table and the gap card.
- 2026-11-17 · Greece: Sources that could not be read directly: The gap cards said the Ministry of Health's site did not answer and that the Government Gazette number of the registry decision was not read. The Ministry's site answered on 17 November 2026, and the cited News4Health report prints the Gazette issue. (Second accuracy pass; www.moh.gov.gr and the News4Health report of 9 February 2024, read in raw text.) · The cards now say what was read from the Ministry's site and what was not, give the Gazette issue as the report states it, and record that the registry's and the screening programme's own sites still did not answer.
- 2026-11-17 · Greece: Cancer care routed through the e-prescription system: The card said Greece builds these systems on e-prescribing rather than on hospital reports, a contrast no source makes, and credited the link between text-message invitations and COVID-19 vaccination to the OECD profile. (Second accuracy pass; News4Health report of 25 February 2025, CNN Greece launch report, the Ministry of Health press release dated 9 May 2024 and the OECD profile, read in raw text.) · The contrast was removed, and the text-message statement is now credited to the Minister of Digital Governance as CNN Greece reported it and to the Ministry's press release.
- 2026-11-17 · Greece: Cancer medicines: EOPYY pharmacies, case-by-case approval, and biomarker tests paid ad hoc: The card said many costly cancer medicines are collected from the insurer's pharmacies. The insurer's page speaks of high-cost medicines and names no disease. (Second accuracy pass; the insurer's page on dispatch points for high-cost medicines, read in raw text.) · The card now speaks of high-cost medicines and says the page does not list which cancer medicines are among them.
- 2026-11-17 · Committee for the Evaluation and Reimbursement of Medicines for Human Use (Greece): The record said the committee advises on which medicines enter the reimbursed list. The law cited gives only its opinion on exceptional approval of unlisted medicines, and the law that sets out its wider role was not read. (Second accuracy pass; Law 4931/2022 as published by the insurer, read in raw text.) · The claim was removed and the record now says the earlier law was not read.
- 2026-11-17 · National Organisation for the Provision of Health Services (EOPYY, Greece): The record said the insurer dispenses high-cost medicines such as many cancer drugs. The insurer's page names no disease. (Second accuracy pass; the insurer's page on dispatch points for high-cost medicines, read in raw text.) · The cancer wording was removed and the record says the page does not list the medicines.
- 2026-11-17 · National Registry of Patients with Neoplastic Diseases (Greece): The record placed the registry's management in a Ministry of Health unit, which the report does not say, and stated as fact that the registry went into production and had published nothing. (Second accuracy pass; News4Health reports of 9 February 2024 and 25 February 2025, read in raw text.) · The record now names the department as the report does, gives the Gazette issue the report prints, and says the start of production is a press report and that no published figures were found.
- 2026-11-17 · Portugal: 74 radiotherapy units, many past their working life, and a proton centre planned: The card said radiotherapy units over 15 years old were beyond "the 12-year lifespan WHO considers optimal". The OECD profile gives no number of years: the 12 was an axis label of the chart beside the sentence, picked up when the PDF was turned into text. The card also relied only on the OECD's quotation of IAEA figures. (Second accuracy pass; the OECD and European Commission Country Cancer Profile Portugal 2025 read in raw text and on the page image, and the IAEA DIRAC public data feed read on 17 November 2026.) · Removed the 12 years and said the profile gives no figure. Added what DIRAC lists: 24 centres, 56 linear accelerators, no cobalt-60 units, one gamma knife and 12 centres with brachytherapy, with the years of the entries.
- 2026-11-17 · Portugal: Sources that refused a script or could not be reached: The card said IAEA DIRAC could only be read by script and that the European Cancer Inequalities Registry rejected the request for the OECD profile. DIRAC's table loads from a public data feed that answers, and the registry's own address now returns the profile. (Second accuracy pass; the DIRAC feed and the profile's direct address fetched on 17 November 2026, the PDF compared byte for byte with the archived copy.) · The card now says what DIRAC lists for Portugal, that the direct address answered on 17 November 2026 with the same file, and which other sites still refused on that date.
- 2026-11-17 · Portugal: 316 trials in OnCo list a site in Portugal, almost all sponsored abroad: The card gave 96 trials recruiting and 71 completed. One record holds two registrations and the status was taken from the one with no Portuguese site. The count was also dated 11 November 2026 without saying the registry data were retrieved on 7 October 2026. (Second accuracy pass; recount from OnCo's ClinicalTrials.gov participation data on 17 November 2026.) · Corrected to 97 recruiting and 70 completed, taken from the registration that lists Portugal, and gave both the recount date and the retrieval date.
- 2026-11-17 · Portugal: Free cancer care in the NHS, but more than a quarter of health spending paid out of pocket: The card said the national health service "is paid for from taxes". None of the cited sources says how it is financed. (Second accuracy pass; the OECD and European Commission Country Cancer Profile Portugal 2025 read in raw text.) · Reworded to say only that it is the public national health service.
- 2026-11-17 · Portuguese National Health Service (Serviço Nacional de Saúde): The record called the service "tax-funded". None of the cited sources says how it is financed. (Second accuracy pass; the OECD and European Commission Country Cancer Profile Portugal 2025 read in raw text.) · Reworded to "public national health service".
- 2026-11-17 · Portugal: A national cancer strategy with numbered targets for 2030: The card said the strategy aims for 90% of cancers with curative drug treatment to start "within a guaranteed time" in 2026. The strategy speaks of timely access and sets no guaranteed time for drug treatment. (Second accuracy pass; Despacho 13227/2023 and its annex read in Portuguese in raw text.) · Reworded to timely drug treatment for 90% of cancers with an indication for curative drug treatment in 2026.
- 2026-11-17 · Portugal: HPV vaccination at age 10 for girls and boys: The card said a 2026 standard "moved" the vaccine to age 10; the standard recommends age 10 and does not state an earlier age. The 91% coverage figure was credited to WHO without saying it is WHO's own estimate. (Second accuracy pass; Norma 002/2026 read in raw text, and WHO's immunisation data queried for every HPV row on 17 November 2026.) · Reworded to "recommends", named the WHO category of the 91%, and added the figures filed as Portugal's official coverage, which agree with it for 2023.
- 2026-11-17 · Directorate-General of Health of Portugal (Direção-Geral da Saúde): The record described the directorate as the Ministry of Health's directorate for public health. No cited source describes it that way. (Second accuracy pass; Despacho 13227/2023 and Norma 002/2026 read in raw text.) · Reworded to what the sources show: it issues standards addressed to the whole health system, among them those of the national vaccination programme.
- 2026-11-17 · Portuguese National Cancer Registry (Registo Oncológico Nacional): The short description said every new cancer must be recorded within nine months of diagnosis. The law counts the nine months from when the service learns of the diagnosis. (Second accuracy pass; Law 53/2017, article 3, read in Portuguese in raw text.) · Reworded to nine months of learning of the diagnosis, as the longer text of the record already said.
- 2026-11-17 · Hungary: 476 cancer trials in OnCo's records have a site in Hungary: The card's title and first paragraph said 477 trials list a site in Hungary, while its own detail said 476, and the recruiting count carried no date. (Second accuracy pass; recounted from OnCo's ClinicalTrials.gov participation data, in which 476 trial records list Hungary, as retrieved from the registry on 7 and 8 October 2026.) · Title and paragraph corrected to 476, and the count and the registry retrieval are now dated.
- 2026-11-17 · Hungary: HPV vaccination at school for girls since 2014 and for boys since 2020: The card gave only the OECD profile's 2023 figures (76% of girls, 64% of boys by age 15) as the coverage. WHO's immunisation data now show higher figures for 2023: WHO's own estimate by age 15 is 80% for girls and 68% for boys, and the coverage Hungary reported is 80.51% and 68.41%. (Second accuracy pass; WHO immunisation data API rows for Hungary read in raw JSON, and the OECD and European Commission Country Cancer Profile Hungary 2025 read in raw PDF text.) · Both sets of figures are now given, each with its source and coverage category, with the 2025 figures.
- 2026-11-17 · Hungary: Fourteen radiotherapy centres, most machines renewed in the past decade: The page said the IAEA's DIRAC radiotherapy directory had not been read and gave the IAEA's machine count only at second hand. DIRAC has a public data feed. (Second accuracy pass; IAEA DIRAC data feed read in raw JSON on 17 November 2026.) · The card now says DIRAC lists 14 centres with 51 linear accelerators and 3 cobalt-60 units, brachytherapy at 12 of them, in self-reported entries dated 2020 to 2026, beside the profile's 64 units; the gap card is corrected.
- 2026-11-17 · Hungary: Breast, cervical and bowel screening, free and by invitation: The card said the OECD profile gives the 8% colorectal screening figure without naming a year, and that breast screening began in 2002 with 49 providers. The profile gives the 8% in the same passage as its 2021 figures and says the programme now includes 49 providers. (Second accuracy pass; OECD and European Commission Country Cancer Profile Hungary 2025, section 4, read in raw PDF text.) · Both statements reworded to what the profile says, and the IARC CanScreen5 date is marked as not re-read on 17 November 2026.
- 2026-11-17 · Hungary: Five-year survival from the registry: The title said survival was below the EU, which the cited profile does not say, and the summary gave the profile's colon cancer survival (57%) as bowel cancer. (Second accuracy pass; table 2 of the OECD and European Commission Country Cancer Profile Hungary 2025 read in raw PDF text.) · Title changed to say only that survival is improving, and the figure is labelled colon cancer.
- 2026-11-17 · Hungary: Compulsory social insurance that covers cancer care from screening to palliation: The summary said almost everyone is treated in the public system once cancer is diagnosed, and the page's opening said the death rate was driven by lung and bowel cancer. The profile says patients generally receive care in the public system, and names lung and colorectal cancer as leading causes of death. (Second accuracy pass; OECD and European Commission Country Cancer Profile Hungary 2025, sections 2 and 5.1, read in raw PDF text.) · Both sentences reworded to the profile's own statements.
- 2026-11-17 · Hungary: A population-based registry required by law, with an open online query: The warning that recent years' counts will change was credited to the registry's query tool; it appears on the National Institute of Oncology's registry page. The registry's counts could not be read again on 17 November 2026. (Second accuracy pass; onkol.hu registry page and the stat.nrr.hu page read in raw HTML.) · Attribution corrected, and the registry table and the comparison with IARC now say the counts were read on 11 November 2026 and not re-read.
- 2026-11-17 · National Institute of Public Health of Hungary: The record said the institute runs the mobile screening buses. Its own page says it operated the programme from 2018 to 2023 and that the National Directorate General for Hospitals has run it since 1 April 2023. (Second accuracy pass; the institute's page on the programme "Helybe visszük a szűrővizsgálatokat" read in raw HTML.) · Summary, short description and programme list corrected, with the page cited.
- 2026-11-17 · National Institute of Pharmacy of Hungary: The programme list included clinical trial authorisation, which none of the cited sources states. (Second accuracy pass; the public health centre's about page and notices and the OECD profile read in raw text.) · Item removed.
- 2026-11-17 · Hungarian League Against Cancer: The record said the League has local branches across the country and that its club meetings are run by local branches; the archived home page names branches and club meetings without saying either. (Second accuracy pass; Internet Archive copy of rakliga.hu of 14 August 2025 read in raw HTML.) · Reworded to what the page says.
- 2026-11-17 · Underlying reasons for post-mortem diagnosed lung cancer cases (HULC study): The paper's title had lost the word robust, and the record linked the paper to the National Cancer Registry, which no author gives as an affiliation. (Second accuracy pass; Europe PMC record for DOI 10.3389/fonc.2022.1032366 read in raw JSON.) · Title restored, and the link now points to the National Institute of Oncology, which two authors give as their affiliation.
- 2026-11-17 · National Childhood Cancer Registry of Hungary: The record linked the paper to the Hungarian National Cancer Registry. The paper is about the separate childhood registry, and its authors give the Hungarian Paediatric Oncology Network and Semmelweis University. (Second accuracy pass; Europe PMC record for DOI 10.1556/650.2024.33061 read in raw JSON.) · Institution link removed.
- 2026-11-17 · István Kenessey: The short description called him senior author of both registry studies; he is last author of the 2023 data quality evaluation and first author of the 2024 lung cancer study. (Second accuracy pass; Europe PMC records for both papers read in raw JSON.) · Reworded to an author of both.
- 2026-11-17 · Belarus: Most of the registry's own book has not yet been read into this page: The page said the Belarusian Cancer Registry's book was printed only, that no online copy existed and that the registry's national total could not be read. The cancer centre's website links four editions, and the 2015 to 2024 edition counts 59,058 new cases registered in 2024, a total that includes basal-cell skin cancer, against the 46,420 IARC estimate the page led with. (Second accuracy pass; omr.by answered a plain request, and its statistics page and the PDF text of Cancer in Belarus: figures and facts, 2015 to 2024 were read in raw text.) · The registry's 2024 total, its crude rate of 646.7 per 100,000 and its count of 2,620 womb cancers are now in the description, the opening lines, the comparison note and the registry table, beside IARC's figures and without an explanation neither source gives. The gap card now says which chapters remain unread.
- 2026-11-17 · Belarus: HPV vaccine introduced nationwide in 2025: The card gave 75% first-dose coverage for 2025 as the figure Belarus reports to WHO and said no rows are reported for boys. The 75% is WHO's own estimate against a target of 57,452 girls; Belarus's administrative and official rows give 84.36% against a target of 51,220, and rows for boys exist but are blank. (Second accuracy pass; WHO's immunisation coverage rows for Belarus were read as raw JSON under the wider HPV antigen filter.) · The card names the category of each figure, gives Belarus's reported 84.36% and 3.05% beside WHO's estimated 75% and 2%, and says the rows for boys are blank.
- 2026-11-17 · Belarus: Radiotherapy machines in 2011 and in the IAEA directory now, and palliative care in 2013: The capacity card said the most detailed published counts were more than ten years old, and a gap card said the IAEA's directory of radiotherapy centres did not return its data. The directory's public feed lists 13 centres in Belarus. (Second accuracy pass; the IAEA DIRAC equipment feed was read as raw JSON.) · The card now gives the directory's listing, read 17 November 2026: 13 centres, 23 linear accelerators, 11 cobalt-60 units and 1 Gamma Knife, eight centres with brachytherapy, entries last updated 2019 to 2024, beside the Ministry's 2011 count. The gap card is corrected.
- 2026-11-17 · Belarus: Medicines are registered by the state after review by the Center for Examinations and Tests: The card and the regulator's record said the centre registers manufacturers' maximum selling prices and keeps the state register of them. The archived page the card cites, captured 9 August 2026, and the live page do not list that work; only the capture of 20 July 2026 does. (Second accuracy pass; three copies of the centre's activity page were read in raw text: Internet Archive captures of 20 July and 9 August 2026 and the live page.) · The price work is now dated to the July capture, with a statement that later copies do not list it, on the card and on the regulator's record.
- 2026-11-17 · Belarus: opening lines and the note comparing IARC's figures with the registry: The opening lines said IARC projects both its cases and its deaths from the registry's counts, and the note said GLOBOCAN's womb cancer figure covers ICD-10 C54 only. IARC's fact sheet gives WHO national mortality data as the source for deaths and gives no ICD codes for its cancer headings. (Second accuracy pass; the PDF text of IARC's GLOBOCAN 2024 fact sheet for Belarus was read.) · The opening lines now give the two sources separately, and the note says the fact sheet does not give the codes.
- 2026-11-17 · Belarus: Counts from the Belarusian Cancer Registry, as published: The breast cancer rows were labelled as women only. The Ministry of Health's tables are headed malignant breast neoplasms in Belarus and state neither the sex nor the population behind the rate. (Second accuracy pass; annex A9 of IARC Working Group Report 6 was read in PDF text.) · The rows are labelled Breast and the note says what the tables do not state.
- 2026-11-17 · Belarusian Cancer Registry: The record said the second-cancer study of thyroid cancer patients was done with IARC, said the registry's book had no online copy, and listed as its key papers three papers that do not name the registry as their data source. (Second accuracy pass; each paper's abstract and affiliations were read through Europe PMC, and the full text of Zablotska et al. 2011.) · The second-cancer study is described by its authors' countries, the online editions of the book are linked, and the paper on stage at diagnosis is listed under the cancer centre instead. Two thyroid papers remain listed for a later decision.
- 2026-11-17 · Ministry of Health of the Republic of Belarus: The record said the breast screening pilot sites were named within the EU-funded BELMED project and listed two papers with no Ministry author as its key papers. WHO's news item says the radiologists' training was within BELMED. (Second accuracy pass; WHO Europe's news item of 26 February 2018 and the two papers' affiliations in Europe PMC were read in raw text.) · The BELMED sentence now refers to the training, and the two papers are listed under the N.N. Alexandrov National Cancer Centre, where their Belarusian author works.
- 2026-11-17 · N.N. Alexandrov National Cancer Centre of Belarus: The record said the Ministry of Health designated the centre as the national reference and training centre for screening in February 2011, and that one of its departments supplies registry data for national studies. The mission report says the Ministry told the experts the centre would serve in that role, and the papers show only that department staff are co-authors. (Second accuracy pass; section 4.2.1 of IARC Working Group Report 6 and the papers' affiliations in Europe PMC were read.) · Both statements are reworded to what the sources say.
- 2026-11-17 · Thyroid cancer risk in Belarus among people exposed to radioactive iodine as children after Chornobyl: The paper was attached to the Belarusian Cancer Registry, which its text never mentions; it is a screening cohort. (Second accuracy pass; the full text in Europe PMC (PMC3039791) was searched for the registry and its affiliations read.) · The paper is attached to the University of California San Francisco and the US National Cancer Institute, two of its authors' affiliations.
- 2026-11-17 · Breast and cervical screening practices in nine countries of Eastern Europe and Central Asia: The summary line said mammography screening in Belarus is organised at the national oncology centre. The abstract says screening was introduced or piloted and organised at national oncology or screening centres in three countries including Belarus. The paper was also attached to the registry, not to its Belarusian author's institution. (Second accuracy pass; the abstract and affiliations were read through Europe PMC.) · The summary line follows the abstract, and the paper is attached to the cancer centre, IARC and WHO.
- 2026-11-17 · Bulgaria: HPV vaccination reaches few girls: The page gave 14% as Bulgaria's HPV coverage in girls in 2025 without saying it is WHO's own estimate, credited it to WHO and UNICEF, and said the first row for boys was for 2025. Bulgaria's own reported figure for the same year is 5.88% for the first dose, because the same 8,885 doses are divided by a larger group of girls, and the rows name WHO alone. Rows for boys exist from 2010 but are blank until 2025. (Second accuracy pass; WHO's immunisation data rows for Bulgaria read as raw JSON with the wider HPV query, which returns the reported administrative and official rows as well as WHO's estimates.) · The card, its title and the page description now name WHO's estimate as such, give the reported figures beside it with both denominators, credit WHO alone and say the boys' rows are blank before 2025.
- 2026-11-17 · Bulgaria: The public insurer, reporting hospital medicines paid outside pathway prices: The card said the insurance fund pays for cancer medicines outside the price of each clinical pathway, called it the one public insurer, and said medicines for home treatment are tracked through electronic protocols. The two pages cited list reports on hospital medicines paid outside pathway prices and a list of anti-tumour medicines, but say none of those three things. (Second accuracy pass; the National Health Insurance Fund's pages on hospital medicines and on expert assessments for expensive medicines read in Bulgarian as raw HTML.) · The card and the page description now say what the pages list and state that they do not say which cancer medicines are paid outside the pathway price. The sentence on electronic protocols was removed.
- 2026-11-17 · National Health Insurance Fund (NHIF), Bulgaria: The record said the fund pays hospitals by clinical pathway, pays for cancer medicines outside the pathway price and tracks home-treatment prescriptions through electronic protocols, and listed Positive Drug List reimbursement among its programmes. None of this is on the two pages the record cites. (Second accuracy pass; the same two National Health Insurance Fund pages read as raw HTML.) · The record now describes the reports and lists the pages publish; the two unsupported programme entries were removed and the list of anti-tumour medicines added.
- 2026-11-17 · Bulgaria: A national council for medicine prices and the Positive Drug List: The card said the council decides which medicines enter the Positive Drug List and takes part in the Heads of HTA Agencies group. Its home page, the only source cited, is a menu that lists registers, decisions and assessment summaries and has a section on that group; it does not describe the council's powers. (Second accuracy pass; the home page of the National Council on Prices and Reimbursement of Medicinal Products read in Bulgarian as raw HTML.) · The card and its title now say what the council publishes and that the home page does not describe its powers.
- 2026-11-17 · National Council on Prices and Reimbursement of Medicinal Products, Bulgaria: The record said the council sets medicine prices, decides which medicines enter the list the insurance fund can pay for, and takes part in the Heads of HTA Agencies group, none of which its home page states. (Second accuracy pass; the council's home page read as raw HTML.) · The record now lists what the site publishes, including the price registers and the annexes of the Positive Drug List.
- 2026-11-17 · Bulgaria: Screening, the cancer plan, waiting times and radiotherapy: The gap card said the IAEA's DIRAC radiotherapy directory was not read and that radiotherapy capacity was not on the page. DIRAC's public data feed lists 26 centres in Bulgaria. (Second accuracy pass; DIRAC's public JSON feed read on 17 November 2026.) · The card now gives DIRAC's 26 centres by city, 47 linear accelerators, 5 cobalt-60 units, one gamma knife and 8 centres with brachytherapy, with the entry years 2019 to 2026 and a note that entries are self-reported.
- 2026-11-17 · Bulgaria: Many international cancer trials for a country of under seven million: The card gave 213 trials in OnCo's records with a site in Finland; a recount gives 212. The OpenAlex paper counts carried no date. (Second accuracy pass; trials recounted from OnCo's ClinicalTrials.gov participation data and paper counts checked against OnCo's stored OpenAlex file.) · Finland's count corrected to 212 and the counts dated 17 November 2026; the OpenAlex figures are now dated 8 September 2026, when OnCo took them.
- 2026-11-17 · Bulgaria: A national registry with roots in 1952: The section title said Bulgaria has had a national registry since 1952. The hospital's history page says a section opened in 1952 that is today the National Cancer Registry, not that a registry existed from that year. (Second accuracy pass; the Sofia oncology hospital's history page read in Bulgarian as raw HTML.) · The section title now says the registry has roots in 1952.
- 2026-11-17 · University Specialised Hospital for Active Treatment in Oncology, Sofia (USBALO): The record said the hospital takes part in the JANE-2 and CraNE joint actions and the EUnetCCC network. Its pages only list them in a menu of programmes and projects. (Second accuracy pass; the hospital's history and about pages read as raw HTML.) · The record now says the hospital lists them among its programmes and projects.
- 2026-11-17 · Austria: HPV vaccination free from 9 to 21, with about half of 14-year-olds fully vaccinated in 2025: The card said about 53% of 15-year-olds had both doses in 2021 and credited the figure to the WHO, placed Austria in the bottom third of EU countries, and said no current national coverage figure could be found. The OECD profile's own figure note gives the WHO figure as one for 15-year-old girls, WHO's rows show a much lower figure for boys, and the health ministry had published a coverage report for 2025. (Second accuracy pass; the OECD and European Commission's Country Cancer Profile Austria 2025, WHO's immunisation data service and the health ministry's short report on HPV coverage in 2025 (13 July 2026) read in raw text.) · The card now gives the ministry's 2025 figures (49% of 14-year-olds with both doses, 55% of girls and 43% of boys), WHO's estimates by age 15 (54% of girls and 39% of boys in 2021, 62% and 46% in 2025) and the official coverage Austria reported, each named by category, and states what the OECD profile says and where it differs. The title and the two gap cards were corrected to match.
- 2026-11-17 · Austria: 46 radiotherapy machines counted against 57 planned, and none in Burgenland: The page said the IAEA's radiotherapy directory had not been read for Austria and that no count newer than the 2018 equipment plan was found. The directory's public data feed lists Austria's centres. (Second accuracy pass; the IAEA DIRAC country and centre tables read as raw JSON on 17 November 2026.) · The card now says DIRAC lists 17 centres reporting 55 linear accelerators, no cobalt-60 unit, 1 gamma knife and 1 proton or ion therapy facility, with 12 centres marked as having brachytherapy, gives the years of the entries (2016 to 2026), and sets this beside the equipment plan's count without choosing between them. The gap cards were corrected.
- 2026-11-17 · Austria: 482 cancer trials in OnCo's records have a site in Austria, eight led from Austria: The card gave 123 trials as recruiting and 146 as completed, dated 11 November 2026, and called the Elisabethinen Hospital's pancreatic cancer trial a phase 3 trial. OnCo's data now give 124 and 145, and ClinicalTrials.gov lists the trial as phase 2. (Second accuracy pass; OnCo's ClinicalTrials.gov participation data recounted on 17 November 2026 and the registry record NCT06391892 read through the registry's data service.) · The counts were corrected and dated, with the date the registry was fetched (7 October 2026). The pancreatic trial is now described with both OnCo's recorded phase and the registry's, and its site in Linz.
- 2026-11-17 · Austria: A federal office whose staff come from the AGES medicines agency: The card said the BASG's three members are one from the ministry and two from the AGES medicines market surveillance division. The BASG's page says one is from the ministry, one from AGES, and the third is the head of that division. (Second accuracy pass; the BASG's own description read in raw text.) · The membership was corrected on the card and on the BASG record.
- 2026-11-17 · Austria: Bowel and cervical screening: widely used, but not yet organised programmes: The card gave 52% of men and 62% of women as a breakdown of the 64% of people who reported bowel screening in 2019. The OECD profile gives these as two separate statements from different data, and the two figures for men and women cannot average to 64%. (Second accuracy pass; the OECD and European Commission's Country Cancer Profile Austria 2025 read in raw text.) · The two statements are now given separately, with the survey named for the second, and the card says the profile does not reconcile them.
- 2026-11-17 · Austria: Outpatient medicines: the Erstattungskodex and its red, yellow and green boxes: The card said a medicine prescribed outside hospital is paid for only if it is listed in the Erstattungskodex. Section 30b of the General Social Insurance Act also allows reimbursement of an unlisted medicine in justified individual cases, with approval from the insurers' medical service. (Second accuracy pass; section 30b ASVG read in the JUSLINE consolidated text.) · The exception was added to the plain text and the detail.
- 2026-11-17 · Austrian Health Insurance Fund (Österreichische Gesundheitskasse): The record said benefits still differ because contract negotiations with the state medical chambers were not changed. The OECD profile speaks of state-level representations of healthcare providers and does not name the medical chambers. (Second accuracy pass; the OECD and European Commission's Country Cancer Profile Austria 2025 read in raw text.) · Reworded to what the profile says.
- 2026-11-17 · Four years of organised mammography screening combined with ultrasound in Tyrol, Austria: The record called the Tyrol programme the forerunner of the national one and said the authors run the programme they evaluate. Neither statement is in the paper's abstract, which says only that the authors evaluated what they call their own screening approach. (Second accuracy pass; the abstract read through Europe PMC (PubMed id 29209825).) · Both statements were replaced with what the abstract and the National Cancer Framework Programme say.
- 2026-11-17 · Austrian Cancer Aid (Österreichische Krebshilfe): The record and the page's card called Julius Hochenegg a surgeon. The Krebshilfe's history page, the only source cited, calls him a renowned physician and does not give his specialty. (Second accuracy pass; the Krebshilfe's history page read in raw text.) · Changed to physician on the record and the card.
- 2026-11-17 · Serbia: 8 of 56 recent cancer medicines available, most only patient by patient, against an EU average of 28: The card, the page description and the gap card said 8 of 56 recent cancer medicines were on Serbia's public list, some only in limited quantities. The survey's own breakdown puts 13% of the 8 under full public availability and 88% under limited availability on an individual-patient basis, and its definition of limited availability includes medicines that are not publicly listed. (Second accuracy pass; EFPIA Patients W.A.I.T. Indicator 2025 survey read in raw text and from the page images of its charts on pages 22, 23 and 27.) · Title, summary and detail now say 8 were counted as available, give the 13% and 88% split, and no longer say all 8 are on the list. The same wording was corrected on the named-patient card and the gap card.
- 2026-11-17 · Serbia: HPV vaccine free for girls and boys, but coverage is 3% in WHO's estimate and 7% in Serbia's own figures: Two cards and the page description gave first-dose coverage of about 3% as what WHO records, without saying the figure is WHO's own estimate. The administrative and official figures Serbia reported to WHO are higher: first dose 4.73% in 2023 and 7.27% in 2024, final dose 2.99% and 5.07%, against a target of 654,963. (Second accuracy pass; WHO immunisation data API queried for every HPV row for Serbia and the raw rows read.) · Both cards now name WHO's estimate as an estimate and give Serbia's reported figures beside it, with doses and targets, and say the reported rows for 2025 are blank.
- 2026-11-17 · Serbia: Eight radiotherapy centres and 15 new linear accelerators bought in 2016 to 2019: The page said IAEA radiotherapy data for Serbia had not been read and gave only the Belgrade institute's own count of eight linear accelerators. It also called that institute the national institute, which no source read says, and gave its MRI-based brachytherapy as for gynaecological cancers only. (Second accuracy pass; the IAEA DIRAC public data feed read in raw JSON, and the institute's radiotherapy page read in raw text.) · The card now reports DIRAC's rows, read 17 November 2026: the programme's eight centres with 34 linear accelerators and no cobalt-60 unit, 13 of them at the Belgrade institute against the eight on its own page, with each entry's year, and a 2015 entry for Pristina filed under Serbia. The institute is named plainly, prostate brachytherapy is added, and the gap card is corrected.
- 2026-11-17 · Serbia: In early 2018 fewer than half of patients with four common cancers started radiotherapy within 28 days: The title applied the 28-day figures to all patients when the source gives them for breast, cervical, prostate and lung cancer only. The card and the section title also said new cancer drugs are given only at tertiary level, in the present tense, from a programme written in 2020. (Second accuracy pass; the Programme for Improving Cancer Control 2020 to 2022 read in raw text, and the insurance fund's home page read in its Internet Archive copy of 17 July 2026.) · The title names the four cancers, the section title dates the statement to the 2020 programme, and the card adds the fund's notice of 16 December 2024 that patients on innovative breast cancer medicines can receive part of their therapy in general hospitals.
- 2026-11-17 · Serbia: 74 trials in OnCo's records have a site in Serbia, most of them international phase 3 studies: The card and the page description counted 75 trials with a site in Serbia, 21 of them recruiting, and described one of them as a trial that is not a cancer trial. That trial was removed from OnCo's records on 16 November 2026. (Second accuracy pass; trials recounted from OnCo's ClinicalTrials.gov participation data on 17 November 2026.) · The count is now 74, with 20 recruiting, and the card says when and why it changed and that the data were retrieved on 7 October 2026.
- 2026-11-17 · Serbia: No national cancer programme after 2022 was found: The gap card said the Ministry of Health's website refused a plain request, so whether a newer plan exists could not be checked. The site answers now. (Second accuracy pass; the Ministry of Health's page of programmes read in raw text on 17 November 2026.) · The card now says what that page lists: five programmes, none for cancer, and that one page is not proof that no newer plan exists.
- 2026-11-17 · Serbia: In 2013 to 2018 breast and bowel screening reached about one in nine of the people they were meant for: The card gave the 2019 survey's Pap test figures as shares of women without saying the survey questioned women aged 15 and over, which includes women outside the screening ages of 25 to 64. (Second accuracy pass; Djordjevic et al. 2023 read in full text through Europe PMC.) · The age range of the survey is now stated on the card.
- 2026-11-17 · Republic Health Insurance Fund (RFZO): The record said the survey found 8 of 56 recent cancer medicines on Serbia's public reimbursement list, and that the newest cancer medicines are prescribed only in tertiary hospitals, a statement from 2020 written as current. (Second accuracy pass; EFPIA Patients W.A.I.T. Indicator 2025 survey page 23 read from the page image, and the fund's archived home page read in raw text.) · The record now gives the survey's split of full and individual-patient availability, attributes the tertiary-only rule to the 2020 to 2022 programme, and adds the fund's December 2024 notice on breast cancer therapy in general hospitals.
- 2026-11-17 · Institute for Oncology and Radiology of Serbia (IORS): The record called the institute Serbia's national cancer institute, which none of its cited sources says, and gave only the institute's own count of eight linear accelerators. (Second accuracy pass; the institute's pages read in raw text and the IAEA DIRAC feed read in raw JSON.) · The record now describes it as a tertiary cancer institute named among the programme's eight radiotherapy centres, and gives DIRAC's 13 linear accelerators (entry dated 2025) beside the institute's eight.
- 2026-11-17 · Cervical screening uptake in Serbia by demographic and socioeconomic group: the 2019 National Health Survey: The record said organised screening has not yet reached a quarter of women in Serbia. The survey counted any Pap test, in women aged 15 and over, and the authors' conclusion is more general than the record said. (Second accuracy pass; the paper read in full text through Europe PMC.) · The record now states who was surveyed and gives the authors' conclusion as they wrote it.
- 2026-11-17 · Cancer incidence and mortality in Serbia, 1999 to 2009, from the two population registries: The record called the paper the first evidence that several cancers are commoner and deadlier in Serbia than in most of Europe, which the paper does not claim. (Second accuracy pass; the abstract and full text read through Europe PMC.) · The record now gives the authors' own conclusion.
- 2026-11-17 · Denmark: HPV vaccination: The card's title and text said coverage was back near 90 per cent and that WHO estimated coverage of the girls' target group at 89 to 92 per cent from 2020 to 2025, without saying this was first-dose coverage. WHO's estimate for the last dose of the course was 78 to 82 per cent in those years, and 16 per cent in 2016 against 47 per cent for the first dose. (Second accuracy pass; WHO's immunisation data (WIISE coverage rows for Denmark, all HPV antigens) and the Global Health Observatory series read as raw JSON.) · The title, text and the Statens Serum Institut record now say first dose, give WHO's last-dose estimates beside them, name the coverage category, and add Denmark's own official figures for 2025 for girls and boys.
- 2026-11-17 · Denmark: 421 trials in OnCo's data list a site in Denmark: The card said 422 trials listed a Danish site and 127 were recruiting. (Second accuracy pass; recount from OnCo's ClinicalTrials.gov capture of 7 October 2026 as the data stood on 17 November 2026, after stale trial entries had been removed.) · Corrected to 421 and 126, with the date of the count.
- 2026-11-17 · Denmark: What OnCo could not read or verify: The card said no authoritative public count of radiotherapy machines had been read. The IAEA's directory of radiotherapy centres does list Denmark. (Second accuracy pass; the IAEA DIRAC public data feed read as raw JSON on 17 November 2026.) · The card now gives what the directory lists: eight centres, 55 linear accelerators, no cobalt-60 units, one proton centre and three centres with brachytherapy, with every entry last updated in 2020. It also records that Sundhedsstyrelsen's site, eSundhed and Sundhedsdatabanken's cancer address still could not be read on that day.
- 2026-11-17 · Denmark: One council for five regions: The page's opening, this card and the Medicinrådet record said the council decides which new hospital medicines become standard treatment, and the card said most new cancer medicines are given in hospital. The council's terms of reference say it recommends and that its recommendations are not binding on the regions; no source read says most. (Second accuracy pass; Kommissorium for Medicinrådet version 1.7 read as PDF text.) · Reworded to recommends, with the statement that recommendations are not binding, in the opening, the card and the Medicinrådet record.
- 2026-11-17 · Denmark: New cancer medicines are approved for Denmark through the EU: The card stated that the time from European approval to treatment in a Danish hospital is set mainly by Medicinrådet's assessment and the price Amgros agrees. No source read says this. (Second accuracy pass; the EMA leaflet, the Danish Medicines Agency's page and Medicinrådet's terms of reference read as raw text.) · Replaced with what the terms of reference say: the recommendation rests on the council's assessment and on Amgros's price negotiations and tenders.
- 2026-11-17 · Denmark: Registers linked through one identity number: The card gave 66 per cent of people with lung cancer and 67 per cent with ovarian cancer in employment after twelve months as if for the whole period 2012 to 2022; these are the 2022 values, and earlier years were higher. It also said every cancer pathway is tracked against its standard time, and none of its sources stated that registers are linked through the identity number. (Second accuracy pass; table 11 of Sundhedsdatastyrelsen's analysis for Kræftplan V read as PDF text, and the abstract of Schmidt and colleagues, European Journal of Epidemiology 2014, read through Europe PMC after resolving its DOI.) · The year is given, every is removed, and the linkage statement now cites the 2014 paper; the same unsourced statement was taken off the registry card.
- 2026-11-17 · Denmark: Hospital care is run by five regions: The card said every cancer pathway figure, waiting time and screening rate on the page is reported region by region. The page gives no screening rates and only the pathway table is by region. (Second accuracy pass; the page read against itself.) · The sentence now says only that the cancer pathway figures are reported region by region.
- 2026-11-17 · Denmark: opening paragraph: The opening said Denmark has counted every new cancer since 1943 and gives each patient a legal maximum wait. The registry paper says data since 1943 with a high degree of completeness, and the regions' fact sheet says the maximum waits apply to most types of cancer patient. (Second accuracy pass; the abstract of Gjerstorff 2011 through Europe PMC and Danske Regioner's fact sheet of 12 April 2024 read as PDF text.) · Reworded to a registry holding data on new cancers since 1943 and to most cancer patients.
- 2026-11-17 · Denmark: Measured against the other Nordic countries: The card said Kræftplan V repeats Kræftplan IV's survival goal. Kræftplan IV's goal was to match the best Nordic neighbours by 2025; Kræftplan V's is to be at the top among the Nordic countries. (Second accuracy pass; the November 2025 agreement on Kræftplan V read as PDF text and Sundhedsstyrelsen's Kræftplan IV page read from its Internet Archive copy.) · Reworded so the two goals are stated separately.
- 2026-11-17 · Statens Serum Institut: The record and the HPV card said the institute runs the national vaccination register. Its page says it monitors uptake through the register and sends the reminders. (Second accuracy pass; Statens Serum Institut's childhood vaccination page read as raw text.) · Reworded to what the page says.
- 2026-11-17 · Finland: summary and IARC note: The summary said Finland "has registered every cancer since 1953", and the note comparing IARC's estimate with the registry's count said the registry's figure was published after IARC's projection was made. It did not say that IARC's total includes non-melanoma skin cancer while the registry's leaves out basal cell carcinoma of the skin. (Second accuracy pass; IARC's GLOBOCAN 2024 fact sheet for Finland and the Finnish Cancer Registry's Cancer in Finland 2024 report read in raw text. The report says some cases may go unreported; the fact sheet gives no date for its projection and marks its totals as including non-melanoma skin cancer (33,930 new cancers without it).) · The summary now says Finland has kept a national cancer registry since 1953. The timing claim is removed and the note says what each total includes.
- 2026-11-17 · Finland: The registry's own count for 2024: The table note said the counts include all cancers registered. (Second accuracy pass; Cancer in Finland 2024 read in raw text. It says basal cell carcinoma of the skin and some other tumours are recorded separately and are not in the overall figures, and that benign brain and central nervous system tumours are included.) · The note now says what the overall figures leave out and include.
- 2026-11-17 · Finland: About 39,000 new cancers a year: The card gave 1,175 and 1,717 as lung cancers in women and men. The registry's category is lung and tracheal cancer. (Second accuracy pass; Cancer in Finland 2024, chapter 6, read in raw text.) · The figures are now described as lung and tracheal cancers.
- 2026-11-17 · Finland: Run by the Cancer Society for the state, and faster than before: The card said stage at diagnosis was recorded for 42% of solid tumours, without the report's exclusion of brain and central nervous system tumours. (Second accuracy pass; Cancer in Finland 2024, foreword and chapter 5, read in raw text.) · The exclusion is now stated, on the card and on the registry's record.
- 2026-11-17 · Finland: Cervical screening since 1963, breast screening, and bowel screening rolled out from 2022: The card said bowel screening was launched in 2022 for people aged 60 to 70. The registry's page gives 60 to 70 as the age range now covered, not as the range in 2022, and the screening pages were quoted without their dates. (Second accuracy pass; the Finnish Cancer Registry's three screening pages read in raw text, with the date each was last changed.) · The age range is now given as the page states it, and each screening page is dated (July 2025, October 2023, January 2024).
- 2026-11-17 · Finland: HPV vaccine for girls and boys at school: Coverage figures were credited to WHO and UNICEF and presented without saying they are WHO's own estimates. The final-dose figures Finland reported differ from those estimates, and the card said the first rows for boys are for 2020 when earlier rows exist and are blank. (Second accuracy pass; WHO's immunisation data rows for Finland read as raw JSON with the wider HPV antigen filter. The rows name WHO alone; official coverage reported for 2025 is 81% and 75% for the first dose and 71% and 64% for the final dose, against WHO's estimates of 80%, 74%, 65% and 57%.) · The card now names each category of figure, gives Finland's reported figures beside WHO's estimates, and says WHO's estimates for boys start in 2020.
- 2026-11-17 · Finland: One national cancer centre made of five regional ones: The card said FICAN's own steering group had accepted the cancer strategy's roadmap, and described the founding decree in words FICAN's page does not use. (Second accuracy pass; FICAN's home page and its news item of 17 June 2026 read in raw text. The news item says the steering group for implementing the national cancer strategy accepted the roadmap on 15 June 2026.) · The card now names the right steering group and the date, links the news item, and uses FICAN's own name for the decree.
- 2026-11-17 · Finland: A national cancer strategy for 2026 to 2035: The card listed reducing inequalities as an aim of the strategy, which FICAN's page gives as a value, and repeated that FICAN's steering group had accepted the roadmap. It still described the roadmap only as due in 2026. (Second accuracy pass; FICAN's strategy page, roadmap page and news item of 17 June 2026 read in raw text.) · The aims are now those FICAN states, the steering group is named correctly, and the card says the roadmap is linked from FICAN's page, covers 2026 to 2031 and estimates EUR 140 million of additional investment.
- 2026-11-17 · Finland: 212 trials in OnCo's records have a site in Finland: The card gave 213 trials with a site in Finland. (Second accuracy pass; OnCo's ClinicalTrials.gov participation data recounted on 17 November 2026, giving 212 trial records that list Finland.) · The count is now 212 and dated 17 November 2026.
- 2026-11-17 · Finland: Other figures not found or not verified: The card said radiotherapy capacity was not on the page and that the IAEA's DIRAC directory had not been read. (Second accuracy pass; the IAEA's DIRAC public data feed read as raw JSON on 17 November 2026. It lists 13 centres in Finland with 50 linear accelerators and no cobalt-60 units, 6 centres marked as offering brachytherapy, every entry dated 2020.) · The card now reports what the directory lists, with the entry year and a caution that the list may be out of date.
- 2026-11-17 · Finnish Cancer Registry: The record said the registry plans and supports the screening programmes. The page cited says this only of bowel screening. (Second accuracy pass; the Finnish Cancer Registry's colorectal screening page read in raw text.) · The record now says the registry planned the bowel screening programme with the Ministry of Social Affairs and Health and clinical experts, and supports municipalities in starting it.
- 2026-11-17 · Finnish Medicines Agency (Fimea): The one-line description said new cancer medicines reach Finland mainly through the European route, a word neither cited page uses. (Second accuracy pass; the European Medicines Agency's page on the centralised procedure and the archived Fimea page read in raw text.) · The description now says medicines with a new active substance for cancer must go through the centralised procedure.
- 2026-11-17 · Norway: HPV vaccination for girls since 2009 and boys since 2018, with coverage of about 90% or more: The card credited its 2025 coverage figures to WHO and UNICEF and gave them as plain coverage. The rows name WHO alone and are WHO's own estimates (first dose 91% in girls and boys). The figures Norway itself reported for 2025 were not shown: 94.18% (girls) and 92.28% (boys) for the first dose, and 91.83% (girls) and 89.84% (boys) for the final dose, the last of which is below the "above 90%" in the title. The card also said the registry credits the vaccine for the fall in cervical cancer in the youngest women, where the report says the fall most likely reflects it. (Second accuracy pass; WHO's immunisation data rows for Norway read as raw JSON with the wider HPV query, and the foreword of Cancer in Norway 2024 read in the PDF text.) · The card names WHO's estimates as such, gives Norway's own reported figures beside them, says "about 90% or more" in the title and the page description, and keeps the registry's "most likely".
- 2026-11-17 · Norway: Other figures not found or not verified: The card said the IAEA's DIRAC directory of radiotherapy machines had not been read and that radiotherapy capacity was missing from the page. The directory's public data feed is readable. The card also said the Norwegian Institute of Public Health publishes pathway times, where the source says only that it calculates them. (Second accuracy pass; the DIRAC equipment feed read as raw JSON on 17 November 2026, and the Directorate of Health's chapter on cancer patient pathways read as raw HTML.) · The card now gives what DIRAC lists: 10 centres, 42 linear accelerators, no cobalt-60 unit, one gamma knife and brachytherapy at five centres, with the machines by hospital and the entry years (2018 to 2021), and says the Institute calculates pathway times.
- 2026-11-17 · Norway: What a cancer patient pays outside the cap: The card said Helsenorge's page on charges and exemption cards returned 404 and that no official statement of costs outside the cap had been found. The page answers, and lists charges that do not count towards the cap. (Second accuracy pass; Helsenorge's English page on the exemption card, last updated 1 August 2026, read as raw HTML on 17 November 2026.) · The card now reports the page's examples (dressings, anaesthetics and equipment, medical certificates, copies of medical records, private providers with no reimbursement agreement) and its note on reimbursed patient travel, and says the page is not written for cancer patients. The sentence pointing to this gap from the paying card was changed to match.
- 2026-11-17 · Norway: 38,811 new cancers and 347,867 people living after a diagnosis: The card's first sentence and the page description gave the 11,452 cancer deaths in 2024 as the Cancer Registry's own count. The report takes them from the Cause of Death Registry and calls them preliminary. The description also said hospital care is paid by the state, which none of the sources cited says. (Second accuracy pass; the summary of Cancer in Norway 2024 read in the PDF text, and Nye metoder's English page read as raw HTML.) · The sentence and the description now credit the deaths to the Cause of Death Registry, and the description says publicly funded hospital care.
- 2026-11-17 · Norway: 238 trial records in OnCo list a site in Norway: The card gave 238 trials with a site in Norway and gave the OpenAlex count for 2025 as a full year. The 238 are trial records: two registrations are each attached to two records, so they are 236 separate registered trials. The OpenAlex counts were taken on 8 September 2026, so 2025 is a part-year figure. (Second accuracy pass; OnCo's trial participation data and its stored OpenAlex country file recounted on 17 November 2026.) · The title and text say 238 records and 236 separate registered trials, dated 17 November 2026, and the 2025 figure is marked as part of a year.
- 2026-11-17 · Cancer Registry of Norway: The record listed the breast screening programme under the name BreastScreen Norway, which appears in none of the sources the record cites. (Second accuracy pass; Cancer in Norway 2024 read in the PDF text and the Institute's Mammografiprogrammet page read as raw HTML.) · The programme is listed as the breast cancer screening programme (Mammografiprogrammet), the names the sources use.
- 2026-11-17 · Ireland: HPV vaccine for girls and boys, with three quarters vaccinated in 2025: The card credited its figures to WHO and UNICEF and said coverage fell in 2020 and had not returned to an 87% first-dose figure for girls in 2019. Those figures are WHO's own estimates. The coverage Ireland reported to WHO was 75.65% of girls in 2019 and 84.14% in 2020, so the fall in 2020 appears only in WHO's estimates. (Second accuracy pass; WHO immunisation data rows for Ireland read as raw JSON with a query covering every HPV antigen code, which shows the coverage category of each row.) · The card now names both series, WHO's estimates and Ireland's reported administrative and official coverage, gives each year's figures for both, and says they differ in 2019 and 2020 without explaining why. The source link now returns all HPV rows.
- 2026-11-17 · Ireland: A law that will require alcohol to carry a cancer warning: The card said the start of the labelling requirement was put back three years in September 2025. The order sets the start at three years after 3 September 2025, which moves it from 22 May 2026 to 3 September 2028, about two years and three months later. (Second accuracy pass; the Irish Statute Book text of S.I. No. 422 of 2025 and S.I. No. 423 of 2025 read as raw text.) · The card now says the start was reset to three years after the September 2025 order, gives 3 September 2028, and quotes the words substituted in the labelling regulations.
- 2026-11-17 · Ireland: 241 trials in OnCo's records have a site in Ireland: The card counted 242 trials with a site in Ireland. One registration, NCT05253651, was counted twice because it is attached to two trial records, and the second record's own registration lists no Irish site. (Second accuracy pass; OnCo's ClinicalTrials.gov participation data recounted by distinct registration on 17 November 2026.) · The count is now 241, dated 17 November 2026, with the size of the breast and lung cancer groups stated.
- 2026-11-17 · Ireland: Other figures not found or not verified: The card said the IAEA's DIRAC radiotherapy directory was not read and that radiotherapy capacity was not on the page. The directory has a public data feed that lists Ireland's centres. (Second accuracy pass; the DIRAC equipment feed read as raw JSON on 17 November 2026.) · The card now gives what DIRAC lists: 13 radiotherapy centres with 40 linear accelerators and no cobalt-60 units, by city, eight with brachytherapy, every entry dated 2026.
- 2026-11-17 · Ireland: Survival up from 50% to 66%, and premature cancer deaths below the EU average: The card described the 66% five-year survival figure as that of people diagnosed in 2019 to 2023. The registry's figure for 2019 to 2023 is an age-standardised estimate made by the period method. (Second accuracy pass; the methods section and survival chapter of the registry's report Cancer in Ireland 1994 to 2023 read from the PDF text.) · The card now gives the figures as the registry does, for 1999 to 2003 and 2019 to 2023, and states the method.
- 2026-11-17 · North Korea: What this page can say, and what it cannot: The page said WHO's immunisation data held no HPV vaccination rows for North Korea and that no primary source said whether HPV vaccination or screening exists. WHO holds 44 HPV rows for 2018 to 2025, all blank, records the vaccine as not introduced in every year from 2006 to 2025, and holds the country's answer that it has a national cervical screening programme. (Second accuracy pass; WHO's immunisation data API queried without the narrow antigen filter, WHO's vaccine introduction data and WHO's Global Health Observatory API, each read as JSON.) · The card now says the rows exist and are blank, that WHO records HPV vaccine as not introduced, and no longer says the existence of screening or vaccination is unknown. The description was corrected to match.
- 2026-11-17 · North Korea: What North Korea has told WHO about its cancer services: The page said no primary source existed for cancer registration, cancer services, radiotherapy, screening or survival, and that Europe PMC held no study from inside the country. WHO's Global Health Observatory holds the country's own answers on a cancer plan, a population-based registry (no), services in the public sector and screening programmes, and Europe PMC holds two 2011 surveys about cervical cancer screening carried out in six provinces. (Second accuracy pass; WHO's Global Health Observatory API rows for North Korea read as JSON and the two abstracts read on Europe PMC.) · A new card sets out the answers with their years, says they are the country's own and cannot be checked by OnCo, and describes the two surveys. The gap card, the lede and the description were reworded to match.
- 2026-11-17 · North Korea: Two radiotherapy centres in the IAEA's directory, both in Pyongyang: The page said the IAEA's DIRAC radiotherapy directory had not been read and that no primary source existed for radiotherapy. The directory lists two centres in Pyongyang, each with one cobalt-60 unit and no linear accelerator, in entries dated 2016. (Second accuracy pass; the public data feed of the IAEA's DIRAC directory read as JSON.) · A new card names the two centres, their machines and the year of the entries, and says the entries are self-reported and may be out of date.
- 2026-11-17 · North Korea: No regulator source, no trial site in OnCo's records, three papers in OpenAlex: The card title said there were no trial sites without saying this was a count of OnCo's own records. ClinicalTrials.gov files one study location under North Korea, NCT03993873, with Seoul given as its city. (Second accuracy pass; ClinicalTrials.gov's API queried by location country and the study record read as JSON; OnCo's trial data recounted.) · The title now says the count is of OnCo's records, and the card states the registry entry, that its city is Seoul, and that it is not evidence of a trial site in North Korea.
- 2026-11-17 · China: key figures: The page said GLOBOCAN's rates for China were "on a different population standard" from the National Cancer Center's, and gave a count of registered trials beside the oncology paper count without saying it covers trials of every kind. (Second accuracy pass; the National Cancer Center report (J Natl Cancer Cent 2024) and the IARC fact sheet were read in raw text and both use the world standard population, and the fetcher script and the ClinicalTrials.gov API showed the count has no cancer filter (56,182 for all conditions against 21,683 for cancer on 17 November 2026).) · The sentence now says the IARC figures are a separate estimate for a later year on the same standard, and the trial figure is labelled as trials of all kinds. The registry sentence was also reworded to the paper's own description.
- 2026-11-17 · China: digestive cancers: The card gave causes and places the cited sources do not name, dated newborn hepatitis B vaccination to 1992 and 2005 without a source, called a trial of camrelizumab with rivoceranib chemo-immunotherapy, and dated the approval of the Claudin 18.2 CAR-T satricabtagene autoleucel to 2025. (Second accuracy pass; the National Cancer Center report and the Qidong trial (PLoS Med 2014) were read in full text, the CARES-310 abstract through Europe PMC, and CARsgen's announcement of 22 June 2026 in raw text.) · Risk factors and screening are now stated as the report gives them, the unsourced dates are removed, the trial is described correctly, and the approval is dated June 2026. The Qidong control group is now described as unvaccinated at birth, since two thirds had catch-up vaccination later.
- 2026-11-17 · China: survival: The card compared China's five-year survival with "roughly two-thirds" in the US, Japan and Western Europe and described urban-rural and east-west gaps; neither is in the cited abstract. (Second accuracy pass; the abstract of the cited review (Chin J Cancer Res 2025) was read through Europe PMC and the full text could not be reached.) · The comparison and the gaps were removed; the card gives the two survival figures the abstract states.
- 2026-11-17 · China: paying for innovation: The card said PD-1 antibodies were self-pay until 2017, before any was approved in China; described the 2017 price fall as an average cut; and stated unsourced claims on discounts above half, renewal caps, supplementary insurance, Western drugs not launched and cell therapies being unlisted. (Second accuracy pass; the two NRDL studies (BMJ Glob Health 2021, J Glob Health 2022) were read in full text, the renewal and price-trend abstracts through Europe PMC, and the trade report on CAR-T prices in raw text.) · The 48.9% figure is now given as a fall against the expected trend, the unsourced claims are removed, and the CAR-T prices are attributed and dated. The National Healthcare Security Administration record now gives the average cuts the study reports, 44% in 2017 and 57% in 2018.
- 2026-11-17 · China: rival drugs, nasopharyngeal cancer and the trial list: The page said a price war had left most PD-1 antibodies unprofitable, with no source; said nasopharyngeal cancer had no immunotherapy "until 2018" while dating the first approval to 2021; and listed HARMONi among programmes that reproduced Western PD-1 results. (Second accuracy pass; the lung cancer price study abstract, Junshi's 2025 annual results and openFDA approval records were read in raw text.) · The unprofitability claim was removed, the date is 2021, penpulimab's 2025 approval is identified as the US one, and HARMONi is described as the ivonescimab programme. The lede no longer says China imported every new cancer drug a decade ago.
- 2026-11-17 · Aumolertinib: The record said the European authorisation of Aumseqa was revoked in September 2026 after a review of protected data, citing the EMA's page, which does not say so. (Second accuracy pass; the EMA page was read in raw text on 17 November 2026 and an archived copy of June 2026 was read once.) · The record now says the medicine was authorised on 12 February 2026, that the EMA page today shows a new application with a positive CHMP opinion of 12 November 2026 and no Commission decision yet, and that the account of the revocation could not be re-read.
- 2026-11-17 · HARMONi-A: The record said the US application built on the global HARMONi trial had a November 2026 action date; the decision had been announced. (Second accuracy pass; Summit Therapeutics' release of 15 November 2026 was read in raw text.) · The record now says the FDA issued a complete response letter because overall survival was not statistically significantly improved.
- 2026-11-17 · ZGDH3 (donafenib vs sorafenib): The outcome table gave 334 patients in each arm beside survival figures that come from the full analysis set of 328 and 331, and the record called it the first trial in which a new tablet beat sorafenib on survival, which the paper's abstract does not say. (Second accuracy pass; the abstract (J Clin Oncol 2021) was read through Europe PMC.) · Arm sizes corrected and the "first" claims removed from the trial and from the donafenib record.
- 2026-11-17 · Equecabtagene autoleucel: The record said about 96% of infused patients responded; the figure is 97 of the 101 patients evaluable for efficacy, of 103 infused. (Second accuracy pass; the FUMANBA-1 abstract (JAMA Oncol 2024) was read through Europe PMC.) · Wording corrected in both places.
- 2026-11-17 · Sintilimab: The record said the FDA's advisers voted 14 to 1 that a single-country trial with a chemotherapy comparator did not fit US practice; the vote was on whether further trials showing the results apply to US patients should be required. (Second accuracy pass; the FDA committee minutes and Lilly's release of 24 March 2022 were read in raw text from archived copies.) · The record now gives the question voted on and the letter's recommendation of a multiregional trial.
- 2026-11-17 · Rivoceranib (apatinib): The record attributed all three FDA complete response letters to manufacturing and clinical-site inspections and called apatinib the first oral VEGFR2 inhibitor approved for gastric cancer anywhere, which its source does not say. (Second accuracy pass; the OncLive report of 13 July 2026 and Hengrui's exchange filing were read in raw text.) · The three letters are now described separately, the second for reasons not made public, and the "first" claim was removed.
- 2026-11-17 · JW Therapeutics: The record said relma-cel was not on the national reimbursement list and that provincial insurance filled some of the gap; its source gives list prices only. (Second accuracy pass; the trade report (Flcube, October 2024) was read in raw text.) · The reimbursement statements were removed and the price is attributed and dated.
- 2026-11-17 · United States: The country screens hard, except for the cancer that kills the most people: The card gave lung cancer screening uptake as 18.2 per cent of eligible adults in 2022, from 31.0 per cent in Rhode Island to 9.7 per cent in Wyoming, credited to an analysis of the American College of Radiology registry, and said lung screening prevents the most deaths of the four screening tests. The cited page gives none of those figures for screening and makes no such comparison; the only 18.2 per cent on it is a survival rate. (Second accuracy pass; the American Lung Association State of Lung Cancer key findings page read in raw text.) · Now gives the page's own figures: 24.4 per cent of those at high risk screened, on 2024 data, from 38.7 per cent in Massachusetts to 13.5 per cent in South Dakota. The claim about preventing the most deaths was removed.
- 2026-11-17 · United States: How the United States counts cancer: The card said as fact that SEER does not record recurrence or anything after the first course of treatment. The SEER page lists what the registries routinely collect and says nothing about what they leave out. (Second accuracy pass; the NCI SEER overview page read in raw text.) · Reworded to say what the SEER list contains and that recurrence and later treatment are not on it. The CDC statements are marked as not re-read, because cdc.gov refused the request.
- 2026-11-17 · United States: The FDA and its Oncology Center of Excellence: The 10 approvals under the 505(b)(2) pathway or as biosimilars were written as part of the 89 oncology drug and biologic approvals for 2024. The FDA report gives them in addition to the 89. (Second accuracy pass; the FDA 2024 OCE Annual Report page read in raw text.) · The sentence now follows the report: 89 approvals including 19 new products and 34 new indications, and separately 10 products under 505(b)(2) or as biosimilars and 76 devices.
- 2026-11-17 · United States: Financial toxicity is a measured phenomenon: The plain summary said Americans with cancer go bankrupt at more than twice the rate of others. The study covered western Washington State only. The national medical debt estimate was given without its date, and the CDC hardship figures were credited to the National Health Interview Survey when the report used the Medical Expenditure Panel Survey. (Second accuracy pass; the abstracts of Ramsey et al 2013 and 2016 read through Europe PMC, the abstract of Ekwueme et al 2019 read through Europe PMC, and the Peterson-KFF brief read in raw text.) · The summary now names western Washington State, the debt estimate is dated to survey data for December 2021, and the survey is named correctly.
- 2026-11-17 · United States: Who is covered, and who is not: The card described the 10.5 per cent with direct-purchase coverage as people buying a plan on an exchange. The Census Bureau page does not say how much of direct-purchase coverage is bought on an exchange. (Second accuracy pass; the Census Bureau P60-291 page read in raw text.) · Now says one in ten buys a plan directly.
- 2026-11-17 · United States: What happens to someone who is uninsured: The summary opened by saying uninsured people are diagnosed later. Neither paper cited on the card measures stage at diagnosis by insurance status. (Second accuracy pass; the abstracts of Ntowe et al 2026 and Han et al 2022 read through PubMed and Europe PMC.) · The clause was removed; the survival figures stand.
- 2026-11-17 · United States: What the openness costs: The title said the United States has the highest brand-name prices of any rich country and the summary said generics are cheaper than almost anywhere. The sources compare the United States with three European countries and with 33 OECD countries taken together, not with each country. (Second accuracy pass; the abstract of Vokinger et al 2021 read through Europe PMC; RAND's site refused the request, so its figures could not be re-read.) · Title and summary reworded to the comparisons actually made, and the RAND figures are marked as not re-read on 17 November 2026.
- 2026-11-17 · United States: The pathways, and why the FDA is often first: The card said a new cancer medicine is usually approved in the United States before anywhere else and gave the expedited pathways as the reason. The cited study took only drugs first approved by the FDA, so it cannot show how often that happens, and it gives no cause. (Second accuracy pass; the abstracts of Kim et al 2026 and da Silva et al 2026 read through Europe PMC.) · Now says a new cancer medicine often receives its first approval in the United States, without the causal claim.
- 2026-11-17 · United States: A publicly funded national trial network: The title and summary said no other country has a trial network at this scale and that most countries run academic trials through a handful of groups, and the enrolment card called the United States the country that runs the most cancer trials. No cited source makes those comparisons. (Second accuracy pass; the NCI National Clinical Trials Network page and the NCORP pages read in raw text.) · The comparisons were removed; the network figures stand.
- 2026-11-17 · United States: Black Americans: The summary called the pattern the signature of a difference in care rather than in biology, as if the report had found that. The report says the causes are multifactorial and largely stem from inequalities in the social determinants of health. (Second accuracy pass; the abstract of Saka et al 2025 read through Europe PMC.) · Reworded to the report's own attribution.
- 2026-11-17 · United States: American Indian and Alaska Native people: The summary said colorectal cancer is rising fastest in the young. The report describes steep increases in early-onset disease and makes no comparison of speed between age groups. (Second accuracy pass; the abstract of Kratzer et al 2022 read through Europe PMC.) · Now says it has risen steeply in adults under 50.
- 2026-11-17 · United States: Prior authorisation, step therapy and the pathway vendor: The card said 13 per cent of sampled prior authorisation denials should not have been denied, and called Elsevier ClinicalPath formerly Via Oncology. The inspector general says those requests met Medicare coverage rules and likely would have been approved under original Medicare; the Elsevier page does not mention the earlier name. (Second accuracy pass; the HHS OIG report page and the Elsevier ClinicalPath page read in raw text.) · Reworded to the inspector general's finding; the earlier product name was removed.
- 2026-11-17 · United States: What could not be sourced: Two gap entries said sources could not be read that can be: the CMS quarterly Part B payment limit file, and the United States survival figures in CONCORD-3. (Second accuracy pass; the CMS payment limit file for October to December 2026 read as CSV, and the open author manuscript of CONCORD-3 read through PubMed Central.) · The entries now give the pembrolizumab payment limit with the arithmetic marked as OnCo's, and United States five-year net survival for 2010 to 2014 for breast, colon, rectum, lung and prostate cancer.
- 2026-11-17 · India: Gallbladder cancer burden: The card's title said one gallbladder cancer in six worldwide is Indian. GLOBOCAN 2024 gives India 30,542 of the world's 126,384 cases, about one in four, as the card's own text already said; one in six was the 2022 ratio. (Second accuracy pass; the GLOBOCAN 2024 gallbladder file by country and sex and the IARC India fact sheet read in raw text.) · Title corrected to about one in four. The rank is now stated as both sexes, the age wording follows the review's own terms (fifth and sixth decades), and the reading of the migrant finding is given as the authors' own.
- 2026-11-17 · India: What India is doing about gallbladder cancer: The card said India runs the world's only randomised trials of treatment before surgery and counted RUGB among them. No linked source says they are the only ones, and the registry describes RUGB as a trial of radiotherapy added to systemic therapy for cancers that cannot be removed. The card also stated as fact that no screening programme exists anywhere. (Second accuracy pass; the ClinicalTrials.gov records for NCT02867865, NCT06712420 and NCT06214572 read as raw JSON.) · The word only was removed from the title and text, RUGB is described separately, POLCAGB's enrolment of 124 is given, and the screening sentence now says that we found no programme.
- 2026-11-17 · India: Why gallbladder cancer is common here: Poverty and hand-pump water were listed as risk factors found by the northern India study, which reports them only as possible factors. Four exposures in the north-east were called near-universal when the source uses that phrase for mustard oil alone, and the risk factors were said to cluster in the same districts, which the review does not say. (Second accuracy pass; the abstracts of Mishra 2021, Thomas 2026, Mhatre 2020 and Dutta 2019 read in raw text from Europe PMC.) · The significant and possible factors are separated, the exposures are called common with the authors' caution added, the confidence interval that just includes no effect is noted for mustard oil, and the clustering clause was replaced by the review's statement that the factors act together.
- 2026-11-17 · India: Tobacco, chewed as much as smoked: The card credited the National Tobacco Control Programme with about 3,500 tobacco deaths a day and said state gutka bans followed the 2003 Act. The programme's page gives 13.5 lakh deaths a year and neither linked source mentions state bans. (Second accuracy pass; the National Tobacco Control Programme home page and the full text of the PLoS One paper read in raw text.) · The figure now reads 13.5 lakh (1.35 million) deaths a year and the clause on state bans was removed.
- 2026-11-17 · India: How India counts cancer: The 12.8% rise was written as a rise by 2025 after the 2022 estimate; the paper gives it as 2025 compared with 2020. The registries were called urban-weighted, which the registry paper does not say. (Second accuracy pass; the abstract of Sathishkumar 2022 and the full text of Mathur 2020 read in raw text from Europe PMC.) · The comparison years are stated and the urban-weighted clause was removed.
- 2026-11-17 · India: Make it here: The card said Biocon Biologics' bevacizumab biosimilar Abevmy is approved in the US and EU. Abevmy is the EU product; the US product is Jobevne, approved in April 2025. The second CAR-T was said to come from Barcelona's ARI-0001, a name the linked report does not use, and the patient count for NexCAR19 was given without saying it is the company's own. (Second accuracy pass; the EMA pages for Ogivri, Abevmy and Ituxredi, the openFDA Drugs@FDA records, the AABB report and the ImmunoACT site read in raw text.) · Each biosimilar is given with its own name and approval year in each region, the CAR-T is described as licensed from Hospital Clínic de Barcelona, the patient count is attributed to the company, and the EMA and FDA records are linked.
- 2026-11-17 · India: Why cancer drugs cost less in India: The card said Indian biosimilars of rituximab, trastuzumab and bevacizumab were among the first anywhere, which none of its linked sources says, and gave the pooled procurement saving against list price where the paper says maximum retail prices. (Second accuracy pass; the abstract of Pramesh 2023 in the Bulletin of the World Health Organization and the two linked Wikipedia pages read in raw text.) · The first-anywhere claim was removed and the saving is stated against maximum retail prices, with the paper's note that savings were larger on generics.
- 2026-11-17 · India: Low-dose and metronomic trials: The low dose of nivolumab was described as roughly 6% of the standard dose, a figure the trial abstract does not give. (Second accuracy pass; the abstract of Patil 2023 read in raw text from Europe PMC; the publisher's page refused the request.) · The percentage was removed; the dose is given as a flat 20 mg every 3 weeks.
- 2026-11-17 · India: A different shape of cancer: The card said India has a high case-fatality because so many cancers are found late, a cause the GLOBOCAN fact sheet does not give. (Second accuracy pass; the IARC GLOBOCAN 2024 India fact sheet read as PDF text.) · Replaced by the plain comparison that deaths are more than half the number of new cases; the registry figures on late stage follow in the next sentence with their own source.
- 2026-11-17 · Japan: The high-cost medical expense benefit: The card and the table gave the August 2018 ceiling (80,100 yen plus 1 percent above 267,000 yen for a middle-income household under 70) as the current one, and the next card said the ceilings "are going up" in August 2026, after that date had passed. (Second accuracy pass; the health ministry's high-cost benefit page and its December 2025 reform paper read in raw text on 17 November 2026.) · The card now gives the ceiling for treatment from August 2026 (85,800 yen plus 1 percent above 286,000 yen, about 93,000 yen on a 1,000,000 yen month by the ministry's own example) with the earlier figure dated to July 2026, and the table column is labelled August 2018 to July 2026.
- 2026-11-17 · Japan: classification of stomach cancer: The card said a Japanese stage is not always a Western stage and that the Western system counts involved nodes "instead". The Japanese guidelines say the 15th edition of the Japanese classification is identical to the 8th edition of the UICC TNM classification. (Second accuracy pass; the full text of the Japanese Gastric Cancer Treatment Guidelines 2021 (Gastric Cancer 2023) read through Europe PMC.) · Title and text rewritten: the stages are the same; what is particular to Japan is that the extent of the operation (D1, D1+, D2) is defined by numbered lymph node stations. Two unsourced remarks were removed.
- 2026-11-17 · Japan: particle therapy: The card said locally advanced cervical cancer and oesophageal cancer are given as advanced medical care and paid for by the patient, and that pancreatic cancer was the most recent insured indication. The cited Gunma University page says neither; it is silent on cover for those diseases and gives no dates. (Second accuracy pass; Gunma University Heavy Ion Medical Center's indication page read in raw text.) · The card now lists only the indications the page says are insured, says the page is silent on the others, and tells the reader to ask the treating centre.
- 2026-11-17 · Japan: Helicobacter and hepatitis: The card presented Helicobacter eradication as removing the cause of stomach cancer and left out the national plan's own caution; it also said Japan had spent "two decades" on the two infections that cause "most" of these cancers and that hepatitis treatment is subsidised, none of which the cited sources state. (Second accuracy pass; the Fourth Basic Plan to Promote Cancer Control (PDF text) and the full text of the Helicobacter 2026 paper.) · The card now states the plan's position that there is not yet enough evidence that eradication in healthy people without symptoms prevents stomach cancer, and the unsupported phrases were reworded to what the plan says.
- 2026-11-17 · Japan: genomic medicine: The card called SCRUM-Japan industry-funded and named four arms including "LC-SCRN", a name that does not exist; it also said every insured test's result must go into the national database with individual consent, which the cited C-CAT page does not say. (Second accuracy pass; the SCRUM-Japan site and the C-CAT background page read in raw text on 17 November 2026.) · SCRUM-Japan is now described in its own terms, a project of hospitals and drug companies together with two projects, LC-SCRUM-Asia and MONSTAR-SCREEN, and the C-CAT sentences follow the C-CAT page.
- 2026-11-17 · Japan: what the corpus did not hold: The card said Japan comes tenth in OnCo's count of institutions and eleventh in its count of people. The corpus now holds 72 Japanese institutions, the third largest count, and people records carry no country to count by; the computed box on the same page already said third. (Second accuracy pass; counts recomputed from the corpus and from the OpenAlex file the page uses.) · The fixed ranks were removed; the card keeps the two figures that still hold (fifth by indexed papers, twenty-seventh on the composite score) and points to the computed box.
- 2026-11-17 · Japan: stage at diagnosis: The card and a table heading presented the registry's stage figures as the result of screening ("what screening buys"). The registry file gives stage and survival only, and its own figures cut both ways: bowel cancer, which is screened, is 45.6 percent localised, and liver cancer, which is not, is 59.2 percent. (Second accuracy pass; the national cancer registry survival workbook (xlsx) read cell by cell.) · The causal wording was removed from the card, the title and the table heading, and the bowel and liver figures were added. Every number in the stage table matched the workbook.
- 2026-11-17 · Japan: the stomach screening evidence: The card said the Korean study was "the largest study behind" the recommendation, that most people are still offered barium, that the barium grade rests on older Japanese studies, and that the grade is B because no trial was randomised. The guideline page states none of these. (Second accuracy pass; the National Cancer Center's gastric screening guideline page and the abstract of the Korean study (Gastroenterology 2017).) · Reworded to what the guideline page says: it cites the Korean study for endoscopy only and does not name the studies behind the barium grade. The odds ratios were confirmed and kept.
- 2026-11-17 · Japan: the drug lag: The card said the gap was "largely closed" and stated what closed it as fact. The first study reports associations, and the second concludes that substantial delays persist for drugs given accelerated approval in the United States. (Second accuracy pass; both abstracts read through Europe PMC after resolving the DOIs.) · Title and summary reworded to associations, with the persisting delay stated.
- 2026-11-17 · Japan: trials that tested standards set elsewhere: The card said Japanese trials "changed the Western standard", that JCOG0802 with CALGB 140503 made sublobar resection a standard option, and that the cranial irradiation trial "reversed" a European standard. The cited papers do not say so. (Second accuracy pass; the three cited abstracts read through Europe PMC.) · Reworded to each paper's own conclusion, including the post-hoc finding of more local recurrence after segmentectomy.
- 2026-11-17 · Japan: paying for care and the regulator: Several unsourced statements: that every scheme covers the same treatments at fixed prices with no deductible, that advanced medical care is "the" exception to the ban on mixed billing, and that PMDA's target is a harder measure than other agencies' clocks. (Second accuracy pass; the ministry's insurance slide pack, its advanced medical care page and PMDA's FY2024 performance report read in raw text.) · The statements were removed or narrowed to what the sources say. All scheme counts, co-payment rates, review times and the worked examples were confirmed.
- 2026-11-17 · Japan: surgery, endoscopy, JCOG and medical oncologists: Unsourced claims: that JCOG9501 is the reason D2 is the Japanese standard, that Japan invented endoscopic resection and its makers supply the world's endoscopes, that JCOG runs the trials industry will not pay for, and that most cancer drug treatment is given by organ specialists. (Second accuracy pass; the NEJM and Gastric Cancer abstracts, the JCOG 2026 report and history page and the JSMO history and president's pages read in raw text.) · Each was reworded to what its source states or removed; the figures on those cards were confirmed.
- 2026-11-17 · Japan: how Japan counts cancer, the cancer mix and the gaps: The registry card said clinics opt in and that the 2023 data were published in March 2026; the mix card gave breast cancer survival without saying it is for women only and claimed liver cancer rates the West does not see; the gap card described a paywalled system and a C-CAT search that the linked page does not contain. (Second accuracy pass; the National Cancer Center's registry and statistics pages, GLOBOCAN 2024 as held in the corpus, and the C-CAT page read in raw text.) · Reworded to the sources: all medical institutions that treat cancer notify, the pages carried the 2023 data from April 2026, the survival figure is marked women only, the title follows the GLOBOCAN rates, and the gap card says what was actually read.
- 2026-11-17 · Brazil: Lung cancer causes the most cancer deaths, then bowel and breast: The card said the latest all-cancer death total on INCA's statistics page was for 2021 (231,694 deaths). INCA updated that page on 5 November 2026 and it now gives 2024: 133,430 deaths in men and 126,828 in women. The card also said Brazil publishes no national survival series, which the source does not state. (Second accuracy pass; INCA's Estatísticas de câncer page read in raw text.) · Replaced the 2021 figures with the 2024 ones (260,258 in all, prostate 17,826 in men, breast 20,849 in women) and reworded the survival sentence to say OnCo found no such series on INCA's pages. The gap cards were changed to match.
- 2026-11-17 · Brazil: Private health plans: about 53 million people, and a list of what they must cover: The beneficiary counts for June 2026, December 2022, December 2024 and December 2025 no longer match ANS's page, which now runs to September 2026 with revised earlier figures, and coverage of 24.9 per cent was dated August rather than September 2026. The card also said oral cancer medicines are covered and that the law does not define an internationally renowned assessment body as plain facts. (Second accuracy pass; ANS's Dados gerais and Rol pages and the Schulze article read in raw text.) · Updated to ANS's current figures (53,108,097 in medical plans and 38,988,679 in employer plans in September 2026; 50,500,350, 51,902,469 and 52,802,820 for the three Decembers), reworded the oral medicines sentence to what ANS says, and attributed the remark on the undefined term to the federal judge who made it in 2023.
- 2026-11-17 · Brazil: Bowel cancer screening with a home stool test from 2026: The card gave bowel cancer as the second commonest cancer in Brazil as if that were INCA's estimate. INCA's Estimativa places colon and rectum third overall, after breast and prostate; second is the Ministry of Health announcement's wording. The card also said the Ministry expects to widen access for 40 million people where the announcement says it could. (Second accuracy pass; INCA's Estimativa 2026 synthesis and the 21 May 2026 announcement read in raw text.) · Both rankings are now given with their sources, and expects became could.
- 2026-11-17 · Brazil: LACOG: academic trials and the cohorts that measure Brazil's own care: The count of 47 studies led by LACOG was out of date. (Second accuracy pass; ClinicalTrials.gov API queried for studies with LACOG as lead sponsor.) · Now 48 studies, dated 17 November 2026, with the 3 November count kept beside it.
- 2026-11-17 · Brazil: Breast and prostate lead: The title said cervical, stomach and oral cancers are more common in Brazil than in richer countries, and the text said screening and vaccination can prevent most cervical cancer. Neither statement is in INCA's Estimativa or its release. (Second accuracy pass; INCA's Estimativa 2026 synthesis and 4 February 2026 release read in raw text.) · Title now states that cervical cancer is the third commonest cancer in women, and the prevention sentence uses INCA's own wording.
- 2026-11-17 · Brazil: A CAR-T cell therapy made in a public university: The card called CAR-T among the most expensive treatments in medicine and said the Nutera factory was built with the Instituto Butantan; the sources give a cost of up to US$500,000 a treatment abroad and describe Nutera as formed by groups from the medical school and Butantan. (Second accuracy pass; the Hemocentro de Ribeirão Preto's Nutera page and Agência FAPESP's 20 June 2022 report read in raw text.) · Reworded to the sources' statements and named the co-author of the October 2026 article.
- 2026-11-17 · Brazil: Rare diseases: a 60-day review after a mandatory pre-submission meeting: The resolution behind this card could not be read again: the Ministry of Health's legislation library refused the request. (Second accuracy pass; request for RDC 205/2017 returned HTTP 403.) · The card and the sources gap card now say the timelines could not be re-read on 17 November 2026.
- 2026-11-17 · Brazil: several cards (how Brazil counts cancer, medicines, radiotherapy travel, the courts, reliance, the 60-day law, the tumour atlas, the research consortium): Eight sentences stated more than their sources: that Brazil has no national cancer register, that private and public patients meet very different treatment options, that radiotherapy machines are unevenly spread and many patients travel, that many Brazilians go to court for medicines, that ANVISA made reliance permanent in 2024, that few countries put a waiting time into law, that most genetic data come from people of European ancestry, and that 11 institutions were brought together to build a network. (Second accuracy pass; INCA's methodology page, Law 12.732, CONITEC's judicialisation page, ANVISA's IN 289/2024 notice and INCA's 2026 news items read in raw text.) · Each sentence was reworded to what its source says or removed; the 60-day card now also says its waiting-time figures come from 1,709 women with stage I to III breast cancer.
- 2026-11-17 · Agência Nacional de Saúde Suplementar: The record carried the June 2026 beneficiary figures that ANS has since replaced, and said oral cancer drugs are covered as an inference from an exclusion. (Second accuracy pass; ANS's Dados gerais and Rol pages read in raw text.) · Updated to the September 2026 figures and reworded to ANS's statement about the exclusion.
- 2026-11-17 · Safety and efficacy of a new academic CD19-directed CAR-T cell (Donadel and colleagues, 2024): The record called the paper the first report of its kind, said the product was made entirely in a public university, and said the national product would cost a fraction of imported ones. None of these is in the paper's text or the centre's page. (Second accuracy pass; the paper's opening text on the publisher's page and the Hemocentro's Nutera page read in raw text.) · Reworded to a report of 20 patients and to the centre's stated aim of reducing cost. The same first was removed from two people records.
- 2026-11-17 · Russia: Biocad: Russia's own PD-1 antibody, and Russian-made versions of other antibodies: The card said the DOMAJOR lung cancer trial enrolled patients in Russia, China, Hungary and Slovakia, said Biocad "makes the immunotherapy Russian oncologists use", and its title said the company copies "most of the rest". (Second accuracy pass; the ClinicalTrials.gov record NCT03912389 read as raw JSON lists sites in Czechia, Georgia, Hungary, Romania, Russia and Slovakia and none in China, and no source read supports the two general statements.) · The countries now follow the registry record, the comparator is given as placebo plus chemotherapy, and the general statements are replaced by what the FLAT paper says: prolgolimab was the first Russian PD-1 inhibitor approved.
- 2026-11-17 · Russia: What could not be sourced: The card said the list of Russian registries in Cancer Incidence in Five Continents volume XII could not be retrieved. (Second accuracy pass; the registry list inside the summary data file on IARC's CI5 volume XII download page, read as raw text.) · The card now names the nine Russian registries listed for 2013 to 2017: Karelia, Samara, Arkhangelsk, Pskov, Kaliningrad, Komi Republic, Murmansk, Vologda Region and Orenburg, and notes that three carry an asterisk the file does not explain.
- 2026-11-17 · Russia: The Blokhin centre: the reference hospital: The card said the Organisation of European Cancer Institutes lists Ivan Stilidi as the centre's representative, and stated as fact that the centre is the country's largest cancer hospital and its federal reference centre. (Second accuracy pass; the OECI list of members in the Russian Federation and the centre's own home page, both read as raw HTML.) · OECI lists him as director, which the card now says; the size claim is now given as the centre's own description and the reference centre wording is removed.
- 2026-11-17 · Russia: What it reports: stage at diagnosis, one-year fatality, morphological verification and active detection: The card said late presentation is worst for oropharynx, mouth and rectum. (Second accuracy pass; the Herzen institute volume for 2025 read with pdftotext, which gives those figures for visually accessible sites only and gives pancreas 56.8 per cent at stage IV alone.) · The sentence now says the figures are for sites that can be seen or felt on examination.
- 2026-11-17 · Russia: The vital and essential medicines list is the boundary of what is free: The card said a drug not on the list is in practice the patient's problem, and described appendix 3 of the government order as a list for centrally supplied high-cost conditions giving blood cancers a federal supply route. (Second accuracy pass; article 80 of federal law 323-FZ and government order 2406-r read as raw HTML on the legal database.) · The card now gives part 3 of article 80, under which a medicine outside the list is not charged to the patient when a medical commission decides it is needed, and describes appendix 3 only by what its title says.
- 2026-11-17 · Russia: Cervical cancer: incidence and death rate both rising: The title said cervical cancer is rising in Russia where it is falling elsewhere, the summary called Russia a high-income health system, and the detail said modelling projected continued rises without vaccination and screening. (Second accuracy pass; the abstract of the Cancer Epidemiology 2018 paper read through Europe PMC, and IARC's population file, which classes Russia as upper middle income.) · The comparison and the income label are removed and the detail now reports what the authors wrote: risk rising in recent generations and a call for vaccination and screening.
- 2026-11-17 · Russia: The Herzen institute and the radiological centre: the statistics and the radiation: The card said Obninsk is where much of Russia's reactor-produced medical isotope work is based and that radiopharmaceutical research sits at the centre. (Second accuracy pass; the National Medical Research Radiological Centre's home page read as raw HTML, which says neither.) · Both statements are removed.
- 2026-11-17 · Russia: Petrov in Saint Petersburg, Rogachev for children, Tomsk for Siberia, and the societies: The card called the Rogachev centre the federal children's hospital for leukaemia, solid tumours and immunodeficiency and gave a congress abstract deadline of 25 October. (Second accuracy pass; the Rogachev centre's and RUSSCO's home pages read as raw HTML.) · The Rogachev centre is now described in its own words and the abstract deadline, which is past and no longer on the page, is removed.
- 2026-11-17 · Russia: R-Pharm, Mabscale and the rest of the domestic supply: The card said every trial named follows one pattern with the Russian regulator as the audience, and gave R-Pharm a founding year its site could not confirm. (Second accuracy pass; the ClinicalTrials.gov records for each trial read as raw JSON; the company's site returns only a script shell.) · The pattern is now limited to the comparative trials and to what the registry shows about sites, and the founding year is withdrawn with a note that the site could not be re-read on 17 November 2026.
- 2026-11-17 · Germany: What happens when the answer is no added benefit: The card said an orphan medicine whose added benefit is not proven after it crosses the revenue threshold must be priced appropriately below the comparator. Section 130b(3) SGB V applies that rule to a medicine whose added benefit counts as not proven because the company did not file the required evidence on time or in full. The card also credited a report to the German haematology and oncology society alone when it was published jointly with the AWMF. (Second accuracy pass; section 130b and section 35a SGB V read in raw text on gesetze-im-internet.de, and the report's imprint read in the PDF.) · Rule restated as the statute gives it, and the report attributed to both publishers.
- 2026-11-17 · Germany: Reimbursed on day one, judged afterwards, priced last: The card said the German list price anchors reference pricing across the continent. None of the card's sources says so. (Second accuracy pass; the G-BA procedure pages, sections 35a and 130b SGB V and the vfa report read in raw text.) · Claim removed; the sentence now says only that the company sets the price for the first six months.
- 2026-11-17 · Germany: The hospital exemption: The card said the Paul-Ehrlich-Institut's gene therapy table lists eighteen entries, seventeen of them with EU numbers, and that its other two tables carry four more national permissions. Read on 17 November 2026 the table lists nineteen entries, eighteen with EU numbers, and the other two tables carry six national numbers. (Second accuracy pass; the three Paul-Ehrlich-Institut tables read in raw text.) · Counts corrected and dated.
- 2026-11-17 · Germany: WiZen and what the registries are used for: Two cards said the WiZen study linked insurance claims to four cancer registries to compare survival. The survival comparison was run on the claims data and, separately, on the registry data; linking the two was a separate feasibility question whose results the final report calls of limited value. (Second accuracy pass; the Deutsches Ärzteblatt article and the final report to the Innovationsfonds read in raw text.) · Both cards now describe two separate, complementary analyses.
- 2026-11-17 · Germany: The argument about the method: The card presented a 2023 review of ten years of AMNOG as the German Cancer Society's own view. The paper is by a working group of the society's Section C, its authors are employees of pharmaceutical companies, and its abstract says it is written from the perspective of manufacturers of cancer drugs. (Second accuracy pass; the abstract and the full text read through Europe PMC.) · Authorship and perspective stated on the card.
- 2026-11-17 · Germany: nNGM and whether the network helps: The cards said the network's site lists 29 centres, that the national rollout began on 1 April 2018, and that Germany's largest statutory insurer paid for the outcome study. The site's menu lists 23 network centres, the cited paper gives only the year 2018, and the study was funded by the federal association of the AOK funds. (Second accuracy pass; the nNGM site and the two papers' abstracts read in raw text.) · All three statements corrected to what the sources say.
- 2026-11-17 · Germany: MASTER: The card reported a head-to-head comparison of whole-genome and panel sequencing without saying it covered 20 patients. (Second accuracy pass; the abstract read through Europe PMC.) · Sample size added.
- 2026-11-17 · Germany: Academic trial groups: The card said the German Breast Group's GeparSixto and pooled analyses defined HER2-low as a category. The pooled analysis used four other trials and concluded that HER2-low tumours can be identified as a subgroup distinct from HER2-zero. (Second accuracy pass; the Lancet Oncology abstract read through Europe PMC.) · Sentence reworded to the paper's own conclusion.
- 2026-11-17 · Germany: A national cancer information service: The card said the service is paid for from public funds, that most countries have no equivalent, and that it is the front door most German patients use. The service says it is financed predominantly from public funds; the other two statements have no source. (Second accuracy pass; the Krebsinformationsdienst pages read in raw text.) · Qualifier restored and the two unsourced statements removed.
- 2026-11-17 · Germany: What Germany does not publish, and what the corpus does not hold: The gap cards called the G-BA scheme the most rigorous in the world, with no source, and said the corpus has no record of the German Consortium for Hereditary Breast and Ovarian Cancer, which it now holds. (Second accuracy pass; the G-BA pages read in raw text and the corpus searched by record name.) · Superlative removed and the corpus statement corrected.
- 2026-11-17 · Germany: Regulators section: A sentence under the regulator cards said a German patient's access to an EU-authorised product depends on the benefit assessment. Statutory insurance pays from launch, and the assessment sets the price that applies from the seventh month. (Second accuracy pass; section 130b SGB V and the EFPIA survey's country definitions read in raw text.) · Sentence rewritten.
- 2026-11-17 · Indonesia: A registry covering about 14 per cent of the population: The card, the page's opening paragraph and its note on IARC's figures said that no Indonesian registry has been published in Cancer Incidence in Five Continents, gave that as the reason IARC's figures are estimates, and said IARC found the Jakarta pilot's coverage too low. The national plan reports only that the submissions for Volumes X, XI and XII did not meet the quality criteria, and does not say who evaluated the pilot. (Second accuracy pass; the Ministry of Health's Rencana Kanker Nasional 2024-2034 read as raw text with pdftotext, and IARC's GLOBOCAN 2024 fact sheet for Indonesia.) · The registry's status and IARC's method are now stated as two separate facts, each as its source gives it, and the evaluation is no longer credited to IARC.
- 2026-11-17 · Indonesian National Cancer Registry: The record said IARC judged the Jakarta pilot's coverage too low and its data quality below standard. The plan says an evaluation found this and does not name the evaluator. (Second accuracy pass; the Rencana Kanker Nasional 2024-2034 read as raw text with pdftotext.) · Reworded to what the plan says.
- 2026-11-17 · Indonesia: About 285 million people registered with the Jaminan Kesehatan Nasional: The opening paragraph gave 82 per cent active cover beside the share of the population registered, as if it were a share of the population. The source gives 81.5 per cent as a share of members, and its own figures (283 million members, about 266 million active) do not agree with that percentage. (Second accuracy pass; the Fortune IDN report of 7 April 2026 read as raw page text.) · The figure is now given as a share of members and the card notes the disagreement inside the report.
- 2026-11-17 · Indonesia: The Formularium Nasional decides which medicines JKN pays for: The card said a cancer medicine approved for sale but not on the list is generally not paid for by the insurance scheme. The cited ministry page says only that medicine services follow the list. (Second accuracy pass; the Ditjen Farmalkes page of September 2022 and the national cancer plan read as raw text.) · Reworded to what the ministry page says and to the plan's own proposal, which treats a medicine outside the list as paid for by another insurer or by the patient.
- 2026-11-17 · Indonesia: A free health check for every citizen around their birthday: The card described the Cek Kesehatan Gratis as planned around each person's birthday and separate from the insurance scheme. That description comes from the national cancer plan's free screening programme; the ministry's 2026 report on the Cek Kesehatan Gratis does not mention a birthday. (Second accuracy pass; the Kemenkes news page of 4 August 2026 and the national cancer plan read as raw text.) · The birthday description is now attributed to the plan, and the card says the ministry's report does not mention it.
- 2026-11-17 · Indonesia: Cervical screening: The card gave the plan's obstacles as hard-to-reach health centres and scarce cryotherapy. The plan names poor access to early detection in remote areas, and cryotherapy referral that is distant and scarce. (Second accuracy pass; the Rencana Kanker Nasional 2024-2034 read as raw text with pdftotext.) · Reworded to the plan's own terms.
- 2026-11-17 · Indonesia: 155 cancer studies on ClinicalTrials.gov list a site in Indonesia: The card described the 89 studies with no phase as observational or device studies. The registry lists 29 as observational and 60 as interventional studies with no phase, which covers more than devices. (Second accuracy pass; the ClinicalTrials.gov API v2 queried again on 17 November 2026 and all 155 records tallied.) · The description now gives the two counts as the registry records them.
- 2026-11-17 · Indonesia: oncology papers a year with an Indonesian author: The card gave a live OpenAlex reading of 3,388 papers for 2021 to 2025 with 896 for 2025, said the plan sets out four research needs, and said survival measurement depends on a linkage to death records that has not been done. The cited query now returns 4,324 with 1,136 for 2025; the plan lists five needs; and the plan says only that the missing linkage may be why survival studies are limited. (Second accuracy pass; the OpenAlex works API queried on 17 November 2026 and the Rencana Kanker Nasional 2024-2034 read as raw text.) · The live figures are replaced with the 17 November reading and the title widened to match, the fifth need (global research partnerships) is added, and the survival sentence carries the plan's own hedge.
- 2026-11-17 · Italy: One service, two countries: The card set a peak figure for Calabria and a 2023 figure for Molise against long-run averages for Lombardy, Friuli Venezia Giulia and Lazio as if all were for 2023, and the summary said the Calabrian figure approaches half. (Second accuracy pass; the interregional mobility chapter of I numeri del cancro in Italia 2025 read in raw text from the PDF.) · Each figure now carries its own period: 45.7 per cent for Calabria and 40.3 per cent for Molise in 2022 to 2023, against 1.9, 2.1 and 3.0 per cent, with the whole-period averages given separately.
- 2026-11-17 · Italy: What the Italian registry is: The summary, the gaps card and the section heading said the implementing regulation of the 2019 registry law was not in place. The source says only that the legislative steps to activate the national registry were not complete at the end of September 2025. (Second accuracy pass; the AIOM president's introduction to I numeri del cancro in Italia 2025 read in raw text from the PDF.) · Reworded to what the introduction says and attributed to it.
- 2026-11-17 · Italy: A ring-fenced national fund for innovative medicines: The card said AIFA's commission may take into account that a medicine was developed mainly in Italy. The criteria say it may consider, when judging how far trial results carry over to practice, that the research and development plan was designed and conducted in Italy. (Second accuracy pass; annex 1 to AIFA determination 966/2025 read in raw text from the PDF.) · Reworded to the criteria's own terms.
- 2026-11-17 · Italy: No cost-per-QALY threshold on AIFA's pages: The title and summary stated as fact that Italy publishes no cost-effectiveness threshold. The source is one AIFA page that names none. (Second accuracy pass; AIFA's page on economic evaluations read in raw text.) · The title and summary now say that AIFA's page names no threshold.
- 2026-11-17 · Italy: Research hospitals with state research status: The summary said most of Italy's leading cancer centres hold IRCCS status and that 27 of them belong to Alleanza Contro il Cancro. The network's site says its members include 27 IRCCS and makes neither of the other claims. (Second accuracy pass; the Alleanza Contro il Cancro home page read in raw text.) · Reworded to 27 IRCCS.
- 2026-11-17 · Italy: Records that should be in this corpus and are not: The card and the trials table caption said many academic trials registered in the EU system never reach ClinicalTrials.gov, so the table undercounts them. No source was cited for that. (Second accuracy pass; AIFA's page on the EU clinical trials regulation read in raw text.) · Now says that trials not also registered on ClinicalTrials.gov are missing from the count and that their number was not measured.
- 2026-11-17 · Italy: Five cancers in Italy's national report: The table caption did not say that the breast cancer survivor figure counts women only while the new cases include 621 in men, and said the report states for every site that 2023 death estimates were unavailable when it does so for bowel and lung cancer. (Second accuracy pass; the site chapters of I numeri del cancro in Italia 2025 read in raw text from the PDF.) · Caption corrected.
- 2026-11-17 · France: Three organised screening programmes: The card called organised screening for breast, bowel and cervical cancer free and said participation was below European targets. INCa's cervical screening page says that with an invitation the analysis of the test is paid in full and the consultation is reimbursed at 70 per cent of the agreed tariff, and the Panorama cites a European threshold only for bowel screening. (Second accuracy pass; INCa's cervical screening page and the Panorama des cancers 2026 PDF read in raw text.) · Removed "free", stated the cervical reimbursement rule as INCa gives it, and narrowed the title to bowel screening.
- 2026-11-17 · France: Early access: The title said early access was mostly used for cancer. The HAS four-year report counts 75 oncology decisions among 152 favourable first requests, which is 49 per cent. (Second accuracy pass; HAS four-year early-access infographic PDF read in raw text.) · Title now says cancer is the largest area of use.
- 2026-11-17 · France: 433,136 new cancers a year: The card put the doubling of new cancers since 1990 down to rising rates in women and to population growth and ageing. The estimate paper says the rise is mainly demographic and secondarily a rise in risk in both sexes, larger in women. (Second accuracy pass; the Bulletin épidémiologique hebdomadaire article read in raw text.) · Reworded with the paper's figures (up 98 per cent in men and 104 per cent in women).
- 2026-11-17 · France: 28 local registries: The card said the prostate and thyroid projections were not published. The paper says no projection was made for those two cancers. (Second accuracy pass; the Bulletin épidémiologique hebdomadaire article read in raw text.) · Reworded to say no projection was made and the last year shown is 2018.
- 2026-11-17 · France: A national cancer registry created by law: The card gave as the registry's purpose that national figures would stop being estimates. The Panorama gives the aim as improving national coverage of case collection. (Second accuracy pass; Panorama des cancers 2026 PDF read in raw text.) · Reworded to say the figures remain estimates until the registry reports.
- 2026-11-17 · France: An economic opinion for the most promising and costliest medicines: The card compared the HAS with NICE in England with no source for the comparison, and rested the absence of a threshold on a page that is silent about it. (Second accuracy pass; the HAS economic evaluation page and the CEESP doctrine PDF of July 2021 read in raw text.) · Removed the NICE comparison and cited the doctrine, which says the public decision-maker in France does not at present regulate prices by reference to a particular efficiency value.
- 2026-11-17 · France: New cancer medicines are licensed for the whole EU: The card said the ANSM's own page states that it authorises clinical trials and runs pharmacovigilance; that page does not say so in those terms, and the scientific advice role is on a different ANSM page. (Second accuracy pass; ANSM's "Qui sommes-nous" and early-access pages read in raw text.) · Removed the unsupported clause and linked the page that describes the scientific advice role.
- 2026-11-17 · France: What could not be read: The card listed the PFMG2025 programme site as unreadable. It now redirects to pfmg2025.fr and answers; its home page carries none of the figures used on the page. The health insurance and health ministry sites still serve a human-verification page. (Second accuracy pass; each site requested again as raw HTML.) · Card now names two unreadable sources, dated, and says the PFMG2025 site answers.
- 2026-11-17 · France: page introduction: The introduction, the page description and the overseas paragraph said the health insurance pays for cancer care in full. Service-public.fr says care is covered at 100 per cent of the social security tariff, with excess fees, flat contributions, the deductible and the daily hospital charge left to the patient. (Second accuracy pass; the service-public.fr page on long-term conditions read in raw text.) · The three phrases now say 100 per cent of the official tariff.
- 2026-11-17 · Satricabtagene autoleucel: The record and the regional approvals table gave China's approval as 2025 ("Sep 2025"). CARsgen's own announcement dates the approval to 22 June 2026, for Claudin 18.2-positive, HER2-negative disease after at least two lines. (Second accuracy pass on the China page; the CARsgen release of 22 June 2026 read in raw text.) · Year, date and indication corrected in both places, and the release linked from the record.
- 2026-11-17 · Camrelizumab: The summary said the US application with rivoceranib had received complete response letters in 2024 and 2025. A third letter was reported on 13 July 2026, and the report gives no public reason or year for the second. (Second accuracy pass on the China page; the OncLive report of 13 July 2026 read in raw text.) · Now describes the three letters as the report does.
- 2026-11-17 · Gallbladder cancer: geography: A finding said India runs "the world's only" randomised trials of pre-operative treatment and described RUGB as chemoradiation among them. No source says "only", and the registry describes RUGB as radiotherapy added to systemic therapy for unresectable disease. (Second accuracy pass on the India page; ClinicalTrials.gov records NCT02867865, NCT06712420 and NCT06214572 read through the registry's API.) · "Only" removed, RUGB described separately, and enrolment figures given as the registry states them.
- 2026-11-17 · United Kingdom: Different screening ages for the same cancer: The card said bowel screening in Northern Ireland starts at about 60 and runs from 60 to 74. It also said England changed its cervical screening interval after Scotland and Wales, which the cited sources do not state. (Second accuracy pass; nidirect's bowel cancer screening page read in raw text, which says screening is offered to people aged 59 to 74, and Cancer Research UK's cervical screening page, which gives only the current intervals.) · Northern Ireland's range corrected to 59 to 74 in the summary, the detail and the link label. The cervical sentence now says only that England, Scotland and Wales invite every five years and that England's protocol was published on 1 July 2025.
- 2026-11-17 · United Kingdom: Four waiting-time standards that cannot be compared: The card said Northern Ireland's republished 62-day series revised every quarter back to December 2024. (Second accuracy pass; the Department of Health's revised 62-day publication of 11 September 2026 read in raw text. Its table revises published quarters back to the one ending 31 March 2024.) · Changed to say the published quarters back to March 2024 were revised downward, by as much as 4.5 percentage points.
- 2026-11-17 · United Kingdom: What each nation publishes: The card called the Northern Ireland Cancer Registry's 94.7 percent staging of lung cancers the best stage completeness of the four nations. The registry's report makes no comparison with other nations. (Second accuracy pass; the Northern Ireland Cancer Registry lung cancer report for 1993 to 2023 read in raw text.) · The comparison was removed. The 94.7 percent figure stands.
- 2026-11-17 · United Kingdom: Ten national cancer audits: The card quoted the audit centre as saying it does "not routinely report hospital-level results". The phrase could not be found on the centre's about page, its questions page or either audit report. (Second accuracy pass; the NATCAN about and FAQ pages and the 2025 bowel and 2026 lung audit reports read in raw text.) · The quotation was removed. The sentence now says the two reports give results by NHS trust, health board and cancer alliance.
- 2026-11-17 · United Kingdom: Genomics as a national service: The card said one neuro-oncology centre in ten could not freeze tumour at all. The paper says 19 of 21 centres reported snap-freezing and does not say why the other two did not. (Second accuracy pass; the Neuro-Oncology Practice paper read in full text through Europe PMC.) · Changed to say two of the 21 centres did not report snap-freezing.
- 2026-11-17 · United Kingdom: One committee for four nations: The card said the TRANSFORM trial invites Black men from 45 against 50 for everyone else. Prostate Cancer UK's page gives no starting age of 50. (Second accuracy pass; Prostate Cancer UK's TRANSFORM page read in raw text, which says invitations go to selected men aged 45 to 74 and that at least 1 in 10 of those invited in stage 1 will be Black men.) · Reworded to what the page says.
- 2026-11-17 · United Kingdom: Last of seven comparable countries: A sentence read as if the government's technical annex advised checking for CONCORD-4 before relying on this page. The annex says only that CONCORD-4 is expected later in 2026. (Second accuracy pass; the National Cancer Plan technical annex read in raw text.) · The advice is now a separate sentence in the page's own voice.
- 2026-11-17 · United Kingdom: Roughly average spending: The card described the National Cancer Research Institute as the body that used to publish the UK Health Research Analysis. That analysis is a UK Clinical Research Collaboration publication. (Second accuracy pass; the UK Health Research Analysis 2022 report and the institute's letter of 26 June 2023 read in raw text.) · The institute is now described in its letter's own terms, as set up to connect the organisations that fund cancer research.
- 2026-11-17 · United Kingdom: The slowest country in its comparison group: The card gave participants in commercial research as falling from over 50,000 to just over 28,000, and its wording suggested the UK's share of a commercial trial subset had recovered when it fell. (Second accuracy pass; the O'Shaughnessy review final report, which says around 50,000 and around 28,000, and the Life Sciences Competitiveness Indicators 2026 data tables read in raw text.) · Both figures now say around. The share is now described as slipping from 3.0 to 2.7 percent.
- 2026-11-17 · United Kingdom: The three kinds of no: A sentence under the worked examples gave NICE's count as 596 cancer appraisals and 666 recommendations, while the card above it gave 600 and 672. (Second accuracy pass; NICE's cancer appraisal data page, last updated 14 October 2026, read in raw text. It gives 600 appraisals and 672 recommendations.) · The sentence now gives 600 and 672.
- 2026-11-17 · Canada: The pan-Canadian Pharmaceutical Alliance's negotiation database has not been counted: The gap card, and the last sentence of the card on the alliance, said the alliance's own website did not open and that its negotiation database could not be read. The site answers today. (Second accuracy pass; www.pcpacanada.ca was fetched as raw HTML on 17 November 2026 and returned its home page, its "browse negotiations" page and its page on the temporary access process.) · Both cards now say the site answered on 17 November 2026, describe what its database offers in the site's own words (six negotiation statuses and CSV or JSON exports), and say plainly that OnCo has not yet counted the cancer files in it. No figure was added.
- 2026-11-17 · Canada: A national cancer strategy run by a federally funded partnership, not by a national health service: The card called the Canadian Partnership Against Cancer an independent organisation and described the national strategy as a 10-year plan with eight priorities. The cited page says only that the Partnership is funded by the federal government, and gives neither the ten years nor the eight priorities. (Second accuracy pass; the Partnership's "what we do" page and its cancer strategy landing page were read as raw HTML on 17 November 2026.) · "Independent" was removed. The ten years are now quoted from the strategy landing page ("an ambitious 10-year action plan"), which is cited; the count of eight priorities was removed because neither page gives it.
- 2026-11-17 · Canada: Treatment is free at the point of care; whether a cancer drug is depends on how it is taken and where you live: The card said hospital and doctor care is publicly paid in every province, but neither source it cited says so. (Second accuracy pass; the 2022 Current Oncology review was read in full text from Europe PMC and the Council of the Federation page as raw HTML, and neither contains the statement. The Canada Health Act was then read on the Justice Laws Website.) · The card now cites the Canada Health Act and says what it provides: insured health services are hospital, physician and surgical-dental services, drugs count as hospital services only "when administered in the hospital", and the full federal contribution depends on no user charges or extra-billing.
- 2026-11-17 · Canada: What the review agency says yes to: most cancer drugs, almost always with conditions: The card said the conditions attached to positive recommendations are usually that the price comes down. The study it cites says something less specific. (Second accuracy pass; Meyers et al., JAMA Internal Medicine 2021, was read on PubMed Central, where it says "Conditional recommendations are usually conditional on cost-effectiveness".) · The sentence now says the conditions are usually about cost-effectiveness, attributed to the study.
- 2026-11-17 · Spain: The public health service makes and pays for its own CAR-T cells: The card said the health service pays for a Hospital Clínic CAR-T "for leukaemia and lymphoma". The funded use of ARI-0001 is relapsed or refractory B-cell acute lymphoblastic leukaemia in adults over 25; no lymphoma use is funded. The card also gave 10 February 2021 as the day of authorisation, which is the date of the regulator's announcement. (Second accuracy pass; the Ministerio de Sanidad's BIFIMED entries for ARI-0001 and ARI0002h and the AEMPS announcement, read in raw text.) · Now says one CAR-T for leukaemia and one for myeloma, gives both funded uses from BIFIMED with links, and says the AEMPS announced the authorisation on 10 February 2021.
- 2026-11-17 · Spain: SEOM's annual report: The gap card said SEOM's report Las cifras del cáncer en España 2026 could not be read and called it the most quoted summary. The report can be read at www.seom.org. (Second accuracy pass; the report's PDF fetched from www.seom.org and read in raw text.) · The card now says what the report gives (301,884 new cancers for 2026, 115,578 deaths from tumours in 2024, five-year net survival of 57.4% in men and 65.2% in women), that these match the REDECAN and INE figures on the page, and that SEOM's press page prints 301,184.
- 2026-11-17 · Spain: Why Spain's numbers are projections: The lede, the card and the REDECAN record said flatly that Spain has no single national cancer registry. REDECAN says only that national incidence is estimated from the provinces that have population registries and from deaths in all provinces, and a national childhood tumour registry exists. (Second accuracy pass; REDECAN's 2026 incidence report and the national cancer strategy, read in raw text.) · Reworded to say the national figures are estimates projected from the provinces that registries cover, not a count of every case.
- 2026-11-17 · Spain: A clinical trial country: The card listed an accelerated assessment procedure among the reasons the AEMPS gives for Spain's trial numbers. The agency's fifth reason is flexible regulation; the accelerated procedure is listed separately among its measures for 2025. The page description also said Europe's highest trial count where the agency says the European Union's. (Second accuracy pass; the AEMPS announcement of 12 January 2026 and its activity report announcement of 2 June 2026, read in raw text.) · Both sentences now follow the agency's wording.
- 2026-11-17 · Spain: Cancer medicines developed in Spain: The card called plitidepsin investigational. OnCo's own record gives it as approved in Australia in 2018 for myeloma and refused in the European Union. (Second accuracy pass; the card read against the plitidepsin record in the corpus.) · The card now says so.
- 2026-11-17 · Spain: Sources that could not be reached: The gap card and the ISCIII record said the Instituto de Salud Carlos III's website showed a maintenance page. It answers again. (Second accuracy pass; isciii.es fetched as raw HTML on 17 November 2026.) · Both now give the date of the failure and say the site answered on 17 November 2026, and that its budget and funding calls were still not read.
- 2026-11-17 · Spain: What a patient pays: The page said co-payment percentages and caps change by royal decree-law, and called fruquintinib a tablet. The law lets the Government update the caps by royal decree, and the financing report does not give the medicine's form. (Second accuracy pass; article 102 of the Ley de garantías in the BOE consolidated text and the Ministry's public report on fruquintinib, read in raw text.) · The note now says the two latest changes were by royal decree-law and that caps can be updated by royal decree; the card says medicines taken at home.
- 2026-11-17 · Valentín Ortiz-Maldonado: The summary said fractionated dosing cut severe cytokine release syndrome from 26.7% to 4.3%. The paper reports the two rates before and after a protocol amendment in a single-arm study, without attributing the difference. (Second accuracy pass; the abstract of the CART19-BE-01 report (Molecular Therapy 2021) read through Europe PMC.) · Reworded as a before and after comparison.
- 2026-11-17 · South Korea: National screening for six cancers: The card title, the page's opening paragraph and a paper record called the national screening programme free for everyone. The programme's own 2025 review says it is free for Medical Aid recipients and the lower-income half of the insured, that the upper half pay 10 per cent for stomach, breast, liver and lung screening, and that bowel and cervical screening are fully covered. The card also said barium X-ray is no longer a method, where the review still lists it as a selective alternative, and left out that the health minister can designate further people for lung screening. (Second accuracy pass; full text of Lee, Suh and Choi, J Prev Med Public Health 2025 (PMC12332392) and the cancer.go.kr screening rules page, both read as raw text.) · Title, plain text and detail now state who pays what, give both sources on the X-ray, and add the minister's designation and the 2019 start of lung screening.
- 2026-11-17 · South Korea: What could not be sourced: The gap card said the programme's review gives only that six cancers were offered by 2019 and so the start of lung screening was unknown. The review states that lung cancer screening was introduced in 2019. (Second accuracy pass; full text of the same review read as raw text.) · The gap now says only the month is missing and gives the year.
- 2026-11-17 · South Korea: Records that should be in this corpus and are not: The card listed the Korean Cancer Association as missing and said OnCo held no Korean trial networks. The corpus holds the Korean Cancer Association, and this page itself lists the cooperative groups KCSG and KGOG. (Second accuracy pass; the corpus files in src/data searched by id.) · The association is removed from the list and the sentence now names what is missing: the national trial support body and the trial registry.
- 2026-11-17 · South Korea: Stomach and liver cancer: The comparison of stomach cancer in men with the United States and United Kingdom was credited to GLOBOCAN 2024. The cancer.go.kr table it comes from uses IARC's 2022 estimates. The plain text also described falls in age-standardised rates as falls in deaths and cases. (Second accuracy pass; cancer.go.kr international comparison page and Table 5 of the registry's 2023 report, read as raw text.) · Attribution corrected to the 2022 estimates and the wording changed to rates.
- 2026-11-17 · South Korea: Cancer is a quarter of all deaths: A stray footnote on cancer.go.kr giving 79,153 deaths was said to belong to an earlier year. The page gives the footnote no year; it sits under the 2025 heading and disagrees with the table. (Second accuracy pass; cancer.go.kr deaths page read as raw HTML.) · The card now says the footnote does not match the page's own table, without assigning it a year.
- 2026-11-17 · South Korea: No written threshold, but a published range: The card and the HIRA record described the published cost-effectiveness figures as values HIRA had accepted. The trade press report describes them as the results for drugs that were assessed. HIRA's rule on thresholds was also quoted as if read directly when it was read in the same press report. (Second accuracy pass; Daily Pharm articles of 31 December 2024 and 8 January 2026 read as raw text.) · Wording changed to drugs assessed, and the rule is attributed to Daily Pharm's quotation.
- 2026-11-17 · South Korea: Confidential risk-sharing contracts since 2014: The plain text said most new cancer and rare-disease drugs have been listed through risk-sharing since 2014, where the study covers January 2014 to December 2020 only, and described 785 drugs as listed where the source says introduced. (Second accuracy pass; Europe PMC abstracts of Kim et al., Front Pharmacol 2023 and Yu, J Eval Clin Pract 2025.) · The period is now given as 2014 to 2020 and the wording follows the abstract.
- 2026-11-17 · South Korea: Gastric cancer trials led from Korea: The card title said the trials changed practice abroad and the text gave early detection in Korea as the reason they were run. Neither trial report says so. (Second accuracy pass; Europe PMC abstracts of KLASS-01 (JAMA Oncol 2019) and CLASSIC (Lancet 2012).) · Title and plain text now state only what the trials found and where they were run.
- 2026-11-17 · South Korea: Medicines approved in Korea first: lazertinib: Lazertinib was described as licensed to Janssen, which neither regulator's page says; they name Janssen as the company. A sentence about Alteogen rested on a site that could not be re-read. (Second accuracy pass; FDA approval notice and EMA Lazcluze page read as raw text; alteogen.com returned a script challenge on 17 November 2026.) · Reworded to what the regulators state, and the Alteogen sentence is labelled as not re-read on 17 November 2026.
- 2026-11-17 · South Korea: The Korea Central Cancer Registry: The plain text said Korea has counted every cancer diagnosed since 1999. The registry says it has reported nationwide statistics since 1999 and estimates its 2023 database to be 97.4 per cent complete. (Second accuracy pass; full text of the registry's 2023 report on e-crt.org.) · Reworded to the registry's own statement, with the completeness figure.
- 2026-11-17 · South Korea: About 289,000 new cancers a year: The card said the National Cancer Center attributes the growth in cases to ageing. Its page says the standardised rate has recently been flat and links that pattern to population ageing. (Second accuracy pass; cancer.go.kr incidence page read as raw text.) · Reworded to what the page says.
- 2026-11-17 · Cancer statistics in Korea: incidence, mortality, survival and prevalence in 2023: The record's plain-language note said most of the survival gain comes from finding cancers earlier. The authors say cancer control efforts contributed to early detection and better survival, and caution that lead time should be considered. (Second accuracy pass; full text on e-crt.org and the Europe PMC abstract.) · Note reworded to the authors' own conclusion.
- 2026-11-17 · Trends in the pricing and reimbursement of new anticancer drugs in South Korea: The record said a new cancer drug took about three years after approval to be covered. The abstract gives an average time to listing of 36.7 months without saying it is counted from approval. (Second accuracy pass; Europe PMC abstract (PMID 33275463).) · The words after approval removed.
- 2026-11-17 · Türkiye: GLOBOCAN note: The note said IARC's incidence estimate "in practice means the Ministry's provincial registries" and that the death estimate was derived "rather than from Turkish death certificates". IARC's fact sheet says neither. (Second accuracy pass; IARC's GLOBOCAN 2024 fact sheet for Türkiye read in raw text, which gives only "Registry (subnational)", "Subnational rates applied to 2024 population" and "National incidence estimates applied to model-based M:I ratios".) · The note now states IARC's source and method in the fact sheet's own terms and says the registries are not named. The same wording was corrected in the cancer deaths gap card.
- 2026-11-17 · Türkiye: What a patient pays: The card said section 4.1.11 of the Health Implementation Communiqué requires hospitals to supply listed chemotherapy medicines, so patients do not buy them at a pharmacy. Section 4.1.11 only provides for the published list; the supply duty for inpatient and day-case medicines is in section 4.1.2, which also lets a contracted pharmacy dispense a medicine the hospital cannot supply. (Second accuracy pass; the consolidated Sağlık Uygulama Tebliği PDF from mevzuat.gov.tr read in raw text.) · The card now cites both sections and includes the pharmacy route.
- 2026-11-17 · Türkiye: How many people take part: The card said the national cancer programme sets a target of at least 70% of the eligible population screened and lists it among its targets, and that the programme gives no breast coverage figure. The programme gives 70% as the old smear programme's target, as the coverage bowel screening must reach and as a cervical performance indicator, not as one target for all three programmes. (Second accuracy pass; the English edition of the 2021 Turkey Cancer Control Programme read in raw text from the Internet Archive copy.) · The card now says where each 70% figure appears and says no breast figure was found, not that none exists.
- 2026-11-17 · Türkiye: Cancer trials with a site in Türkiye: The card gave 568 trials and 25 phase 2/3 trials; OnCo's records now hold 567 and 24. It called 197 trials "active and not recruiting", which is OnCo's own status field and not the registry status, and listed prostate cancer straight after lung cancer although a generic metastatic cancer record comes second. (Second accuracy pass; the trial records in OnCo's own data recounted.) · Counts corrected and dated, the status field named, and the ranking limited to named cancer types.
- 2026-11-17 · Türkiye: Cancer deaths gap: The card said the Turkish Statistical Institute publishes a share and not a count of cancer deaths. Only the text of its 2024 bulletin was read, not its tables. (Second accuracy pass; the archived TÜİK bulletin for 2024 read in raw text.) · The card now says the bulletin's text gives a share and no count.
- 2026-11-17 · Türkiye: Time from licence to reimbursement gap: The card said the regulator and the insurer do not publish the time from licence to reimbursement or a cost-effectiveness threshold, and that reimbursement decisions appear without a published assessment. These were absences stated as facts. (Second accuracy pass; the insurer's announcement 2026/43, the Alternatif Geri Ödeme Yönetmeliği and the Sağlık Uygulama Tebliği read in raw text.) · The card now says these were not found in the texts read.
- 2026-11-17 · Australia: Lung cancer screening, new from 1 July 2025: The card and the screening table left out one eligibility rule, that a person must smoke now or have quit within the past ten years, and said a person stays eligible after quitting more than ten years ago, which is true only of someone already in the program. The card also said the screening register sends results letters and that the first mobile truck had begun its visits. (Second accuracy pass; the Department of Health's page How the National Lung Cancer Screening Program works read in raw text from the Internet Archive copy of 17 September 2026.) · Eligibility now reads as the department states it, in the card and the table; the register is described as sending reminders and next steps, with results going to the healthcare provider; the truck visits are given as the department's schedule.
- 2026-11-17 · Australia: How fast it actually moves: The title and summary called the 36 standard-route approvals new active substances, but the TGA's row also counts biosimilars. The summary said 13 medicines were provisionally registered where the report counts 13 approvals of all application types. The card and table note said working days exclude the time a company takes to answer questions, which the report does not say. (Second accuracy pass; TGA Performance Report July 2023 to June 2024, appendix tables 1, 3 and 8, read in raw text from the PDF.) · The row is now named as the report names it, the provisional figure is described as approvals of all types, and the unsupported definition of working days is removed.
- 2026-11-17 · Australia: Medicare, free as a public patient, a gap as a private one: The card called Medicare universal health insurance and said many people with cancer meet gap fees, neither of which the cited pages say. It described the Greatest Permissible Gap as a safety net that accumulates over a calendar year, when it is a per-item rule. The table note said uncounted costs are where many patients' costs fall, without a source. (Second accuracy pass; the Department of Health's Medicare costs and Medicare safety nets pages read in raw text from Internet Archive copies of 12 June and 27 August 2026.) · Reworded to the department's own description, the Greatest Permissible Gap is set apart from the two annual safety nets, and the unsourced remarks are removed.
- 2026-11-17 · Australia: Medicines, at most $25 a script: The summary said a person pays at most $25.00 however expensive the drug, and that many chemotherapy drugs are given only in hospitals. The PBS page says up to $25.00 for most PBS medicines, apart from brand premiums and certain other charges, and that some medicines needing supervision, such as chemotherapy drugs, are available only at specialised services. (Second accuracy pass; pbs.gov.au About the PBS read in raw text, live and from the Internet Archive copy of 14 May 2026.) · Both sentences reworded to the scope the PBS page gives.
- 2026-11-17 · Australia: The 2024 review of how Australia assesses health technology: The card called the review committee independent, gave 6 June 2023 as the close of the first consultation and placed one of the expert centres at a named university. None of the cited pages says any of these. (Second accuracy pass; the Department of Health's HTA Review pages, its research papers page, the terms of reference PDF and the PBS news item read in raw text.) · The committee is described as the reference committee that oversaw the review, the unsupported date is replaced by the start date the department gives, 27 October 2022, and the university is removed.
- 2026-11-17 · Australia: The Life Saving Drugs Program: The title said the program is not the route for cancer medicines and the text listed the practical routes to an unsubsidised cancer medicine. Neither cited page mentions cancer or those routes. (Second accuracy pass; the Department of Health's Life Saving Drugs Program page and its page about the program read in raw text from Internet Archive copies of 17 June and 12 April 2026.) · The card now says what the department says: the program is separate from the PBS, through which most medicines are subsidised, and covers ultra-rare conditions. The unsourced sentence is removed.
- 2026-11-17 · Australia: Melanoma: The card said the AIHW links the fall in melanoma among people under 40 to the Slip Slop Slap campaign. The AIHW sets the two side by side without stating a cause. (Second accuracy pass; AIHW Cancer data in Australia 2024 overview read in raw text from the Internet Archive copy of 26 October 2024.) · Reworded to the AIHW's own sentence, with a note that it does not state the campaign as the proven cause.
- 2026-11-17 · Australia: page introduction and GLOBOCAN note: The introduction said Australia aims to be the first country to eliminate cervical cancer; the national strategy sets 2035 and it is the Australian Centre for the Prevention of Cervical Cancer that anticipates Australia will be first. The GLOBOCAN note said IARC's skin cancer line is a model because Australian registries do not collect those cancers, a cause no source read gives. (Second accuracy pass; the Department of Health's national strategy page, the ACPCC's About page and the GLOBOCAN 2024 file this site renders, read in raw text.) · The introduction now states the 2035 aim, and the note states the two facts separately without the causal link.
- 2026-11-17 · Australia: Optimal Care Pathways: The card, and the Cancer Australia record, said proposals for new pathways go to Cancer Australia and that it coordinates the pathways. The source gives only a Cancer Australia address for guidance on developing a pathway. (Second accuracy pass; Cancer Council Australia's Optimal Care Pathways page read in raw text.) · Reworded to what the page says, in the card and in the record.
- 2026-11-17 · Australia: Records that should be in this corpus and are not: The gap list named the Australasian Lung Cancer Trials Group as sponsor of DREAM3R. The registry lists PrECOG as sponsor and the Thoracic Oncology Group Australasia as a collaborator. (Second accuracy pass; ClinicalTrials.gov record NCT04334759 read as JSON.) · The entry now names the collaborator as the registry does and gives the sponsor.
- 2026-11-17 · The projected timeframe until cervical cancer elimination in Australia: The record said the paper put a date on elimination for the first time in any country. The abstract makes no such claim; it says Australia is likely to be one of the first countries to reach the thresholds. (Second accuracy pass; the abstract read through Europe PMC after resolving the DOI.) · Reworded to the abstract's own statement.
- 2026-11-17 · Poland: Four free screening programmes, and lung screening new in October 2026: The card said the lung screening programme was for people who had stopped smoking no more than 15 years ago, which left out current smokers; it credited the programme page to the NFZ; it stated that booking had opened; and it gave the five-year HPV test interval without the condition that there are no risk factors. (Second accuracy pass; the pacjent.gov.pl lung programme page and the regulation of 7 March 2025 (Dz.U. 2025 poz. 298) read in raw text.) · Now says current smokers and people who stopped no more than 15 years ago qualify, attributes the page to the government patient portal, says first bookings were due in the second half of October 2026, and adds the risk-factor condition.
- 2026-11-17 · Poland: 904 cancer trials in OnCo's records have a site in Poland: The card gave 902 of 6,501 trial records, with 283 recruiting, 196 completed and 659 in phase 3, and said almost all were company trials with only a handful led from Poland. The counts had drifted from the corpus and the two general statements were not counted from anything. (Second accuracy pass; recounted from OnCo's own trial records on 17 November 2026.) · Counts updated to 904 of 6,525, with 281 recruiting, 199 completed and 661 in phase 3; the tie for third commonest cancer is stated; the uncounted generalisations are removed.
- 2026-11-17 · Poland: A single national health fund, paid for by a 9% health contribution: The card said work done on the cancer card is not limited by the contract's ordinary ceiling. The act says only that the amount is adjusted to health needs and to what was done. (Second accuracy pass; article 136 of the healthcare benefits act (Dz.U. 2025 poz. 1461) read in raw text.) · The inference is removed here and in the National Health Fund record.
- 2026-11-17 · Poland: Only hospitals in a graded national network may treat cancer on public money: The summary said the top level oversees the other levels' treatment plans. The act gives third-level centres oversight of first-level centres' plans only. (Second accuracy pass; article 8 of the National Oncology Network act as amended by Dz.U. 2026 poz. 1007, read in raw text.) · Reworded to first-level hospitals' treatment plans.
- 2026-11-17 · Poland: Most new cancer medicines are paid for through hospital drug programmes: The figures and the named medicines come from the Ministry of Health's summary of the 1 October 2026 reimbursement list, which could not be reached again. (Second accuracy pass; the Ministry's page on gov.pl did not answer on 17 November 2026 and the Internet Archive holds no copy.) · The card now says the summary could not be re-read on 17 November 2026 and that the paragraph stands as first read. The same label is on the regulator card and the Medical Fund card for what they take from gov.pl.
- 2026-11-17 · Poland: lede: The opening said Poland counts every cancer in one national registry. The registry's own bulletin says its counts still carry some under-registration. (Second accuracy pass; the registry's bulletin on 2023 read in raw text.) · Now says Poland has one national cancer registry, which says its counts still miss some cases.
- 2026-11-17 · Mexico: the IARC estimate and the page's opening lines: The page said Mexico "has no national incidence registry, so IARC's incidence figure is modelled", and that the only population-based counts of new cancers Mexico has published come from the nine-registry network. No source gives the first as a cause, and the second is wrong: the network's own paper says a registry report had already been published for Mérida. (Second accuracy pass; the IARC GLOBOCAN 2024 fact sheet for Mexico and the full text of Mohar and colleagues (Salud Publica Mex 2024) read as raw text.) · The note now quotes IARC's own data source ("No data") and method as separate facts, the opening no longer joins them with "so", and the Mérida report is named as published and not read for this page.
- 2026-11-17 · Mexico: 95,108 deaths from cancer in 2024: The national death rate of 73.1 per 100,000 was called a crude rate. INEGI's release calls it only a rate. (Second accuracy pass; INEGI's release 7/26 of 30 January 2026 read from the PDF text.) · The word "crude" was removed and the sentence about women's and men's rates now follows INEGI's wording.
- 2026-11-17 · Mexico: a national cancer registry in law since 2017: The card said as fact that the health ministry restarted the registry in September 2025. The only source is a press report that the ministry would take it up again; what is documented is the presentation of a module for children's cancers. The table note also said the registries started between 2018 and 2020, where the paper's table gives 2018 and 2019 for their establishment and 2018 to 2020 for data collection. (Second accuracy pass; the Expansión article of 24 September 2025 and Table I of Mohar and colleagues read as raw text.) · The restart is now given as a press report beside the presentation, and the table note gives both sets of years.
- 2026-11-17 · Mexico: a national fund pays for high-cost treatment: The card called the Fondo de Salud para el Bienestar a public trust run by IMSS-Bienestar. The law makes IMSS-Bienestar its settlor, in a development bank. (Second accuracy pass; article 77 bis 29 of the Ley General de Salud read from the Cámara de Diputados PDF.) · Reworded to a public trust of which IMSS-Bienestar is the settlor.
- 2026-11-17 · Mexico: what COFEPRIS reported for 2025: The card and three link labels cited the regulator's end-of-year statement as press release 38/2025. The statement as published carries no number. (Second accuracy pass; the gob.mx page of 31 December 2025 read as raw HTML.) · The number was removed; the statement is cited by its date.
- 2026-11-17 · Mexico: mammography every two years from 40 to 69, and cervical screening from 25 to 64: The short text gave a smear test every three years from 25 to 34 and an HPV test every five years from 35 to 64 as what the official standards recommend. The cervical standard sets neither age band; that scheme is how the federal audit office describes current practice. (Second accuracy pass; the 2007 modification of NOM-014-SSA2-1994 in the Diario Oficial and the Auditoría Superior de la Federación's audit 35 of the 2022 accounts read as raw text.) · The scheme is now attributed to the audit, and the standard is described by what it says: screening offered to women aged 25 to 64.
- 2026-11-17 · Mexico: relying on foreign approvals, but only on full ones: The card said a medicine with an accelerated or conditional approval abroad has to go through COFEPRIS's full review, and cannot use the short route until the foreign regulator converts the approval. Article 20 says only that such approvals cannot support an application; it says nothing about what follows. (Second accuracy pass; the guidelines published in the Diario Oficial on 18 July 2025 read as raw text.) · The card now states what article 20 says, marks the general point about cancer medicines as OnCo's reading, and says the guidelines do not describe what happens next.
- 2026-11-17 · Mexico: HPV vaccination extended to boys in 2025: The card said IMSS's press release could not be read and the card could not be rechecked. The release now downloads, and its text supports the card. (Second accuracy pass; IMSS press release 474/2025 downloaded and read from the PDF text.) · The note saying the source was unreadable was removed.
- 2026-11-17 · Mexico: counting children's cancers first: The card said as fact that the health ministry restarted the cancer registry in 2025 with a children's module. The restart is a press report; the presentation of the module is what is documented. (Second accuracy pass; the Expansión article of 24 September 2025 read as raw HTML.) · Reworded to the presentation of the module and the press report.
- 2026-11-17 · Mexico: 368 cancer trials in OnCo have a site in Mexico: The card gave 26 trials in hormone receptor-positive breast cancer. OnCo's records give 27. (Second accuracy pass; the count was made again from the trial records in the corpus.) · Corrected to 27.
- 2026-11-17 · Mexico: what is not known, and how to check: A gap card said Mexico publishes no national count of new cancers or survival, which states an absence as fact, and the closing note said IMSS, ISSSTE and IMSS-Bienestar each buy from the national formulary separately, which the law does not say. (Second accuracy pass; Mohar and colleagues (2024) and article 29 of the Ley General de Salud read as raw text.) · The gap now says OnCo found no such figures, and the note says each public institution draws its own list of medicines from the formulary.
- 2026-11-17 · Implementation of a population-based cancer registry network in Mexico 2017-2020: The record called the network Mexico's first national network of population-based registries and said it ran for under four years. The paper says neither; it speaks of a relatively short time of operation. (Second accuracy pass; the paper's full text read from the journal's PDF, and its DOI and PubMed id resolved through Crossref and Europe PMC.) · "First" was removed and the caveat now says the network ran for a short time.
- 2026-11-17 · Registro Nacional de Cáncer (Mexico): The one-line description said the health ministry restarted the registry in 2025 with children's cancers first, a press report written as fact. (Second accuracy pass; the Expansión article of 24 September 2025 read as raw HTML.) · It now says the health minister presented a module for children's cancers in September 2025, and that OnCo found no national count published from the registry.
- 2026-11-17 · Pakistan: A free HPV vaccine campaign for 13 million girls began in September 2025: The card and a gap card said no official coverage figure for HPV vaccination had been found from WHO or anyone else. WHO's immunisation data hold a 2025 figure for Pakistan. (Second accuracy pass; WHO's immunisation data feed read as raw JSON for every HPV antigen code, not only the narrow one.) · Both cards now give WHO's 2025 row (first dose, females: 7,032,118 doses, target number 17,580,296, coverage 40%) and say that the campaign's own final count against its 13 million target was still not found.
- 2026-11-17 · Pakistan: CAR-T cells made in a Pakistani hospital, from June 2026: The page description, the opening paragraph, the card and the centre's record said the Rawalpindi centre was reported to have given the country's first CAR-T treatment. The report says the centre carried out its own first successful CAR-T treatment. (Second accuracy pass; the Arab News report of 5 July 2026 read in raw HTML.) · All four places now say the centre was reported to have carried out its first successful CAR-T treatment, and keep the report's statement that Pakistan joined the countries able to offer it.
- 2026-11-17 · Pakistan: At the atomic energy cancer hospitals, most patients arrive with stage III or IV disease: The card said 327 of 8,816 women with breast cancer were diagnosed at stage I. The 8,816 are all breast cancer patients, 8,577 women and 239 men. The title also applied a figure from one hospital network to all patients. (Second accuracy pass; the PAEC Cancer Registry Report for 2018 and 2019 read with pdftotext from the Internet Archive copy.) · The sentence now says patients and gives the split by sex, and the title names the atomic energy cancer hospitals.
- 2026-11-17 · Pakistan: Health cards pay for inpatient cancer treatment, but the scheme now differs by province: The card said a planning body cleared the federal scheme and directed its renaming. The report says the body cleared it and referred it for formal approval, and that the planning minister gave the direction; it also quotes him saying the programme was first launched in 2015, not 2018. (Second accuracy pass; the Dawn report of 16 January 2026 read in raw HTML from its Internet Archive copy.) · The card now gives the referral for formal approval, attributes the renaming to the minister, records both launch dates and gives the publication date of each Dawn report.
- 2026-11-17 · Pakistan: WHO's 2020 profile records no national screening programme for breast or cervical cancer: The card stated as a present fact that Pakistan has no programme inviting people for breast, cervical or bowel screening. The source is WHO's 2020 profile, whose rows are dated 2019 and have no row for a bowel screening programme. (Second accuracy pass; WHO's Cancer Country Profile 2020 for Pakistan read with pdftotext.) · The title and text now report what WHO's dated rows record and say that no later official source was found.
- 2026-11-17 · Pakistan: Cancer statistics note: The note said IARC models deaths because Pakistan has no national registration of deaths by cause, and that most registries in the pooled national report are hospital-based. Neither source says so: IARC gives its method without a cause, and the report's abstract does not describe the registries. (Second accuracy pass; IARC's fact sheet, WHO's 2020 profile, the report's abstract in Europe PMC and the PAEC registry report for 2020 and 2021 read in raw text.) · IARC's method and WHO's mortality registration row are now given as two separate facts, and the hospital-based point is replaced by PAEC's own statement that its hospital-based registry supplied about 67% of the cases. The same wording is corrected in the profile card, the table note and the paper's record.
- 2026-11-17 · Pakistan: Charity and public hospitals carry much of the cost of treatment: Two cards said Shaukat Khanum was building a third hospital in Karachi, citing a page that does not say so. (Second accuracy pass; Shaukat Khanum's about page and its Karachi hospital page read in raw HTML.) · Both cards now quote the Karachi page, which is headed "Opening in December 2026", and link to it.
- 2026-11-17 · Pakistan: Seventeen registries, covering 19 of Pakistan's 129 cities: Three places said a 2025 review found the Karachi Cancer Registry the most productive in research, or to have the highest research output. The review says it was noted for its high research output. The opening paragraph also said flatly that Pakistan has no national cancer registry. (Second accuracy pass; the review's abstract and full text read from Europe PMC after resolving its DOI through Crossref.) · The cards and the paper's record now use the review's wording, and the opening paragraph attributes to the review its finding that there is no unified national registry.
- 2026-11-17 · Pakistan: Unregistered cancer medicines can be imported with a no-objection certificate: The card said a certificate takes 10 to 30 days for a hospital, and that the regulator does not publish its own figures. The letter does not tie the 10 to 30 days to hospitals, and the second statement was an absence given as a fact. (Second accuracy pass; the letter read in raw HTML on the journal's site.) · The card now follows the letter's wording and says that no figures published by the regulator were found.
- 2026-11-17 · Pakistan: The atomic energy commission runs a nationwide chain of cancer hospitals: The card explained the commission's focus on radiotherapy as following from its nuclear mandate, and stated its free or subsidised treatment as fact. The commission's page says only that it works under that mandate, and the pricing is its own account. (Second accuracy pass; the commission's medical services page read in raw HTML from its Internet Archive copy.) · The explanation is replaced by the commission's own sentence and the pricing is attributed to it.
- 2026-11-17 · Pakistan: records for the Lahore incidence study, the registries review and the National Institutes of Health: The Lahore study's caveats said the district is mostly urban and that cases are likely missed; the review's record said it explains why the numbers are estimates and counts only registries that have published; the institute's record left out that patient details may not go to unauthorised persons without prior written consent. None of these matched the sources. (Second accuracy pass; the study's full text and the review's abstract read from Europe PMC, and the Gazette of Pakistan text of Act XXIX of 2025 read with pdftotext.) · Each caveat now uses the authors' own points, the review's record states what it found, and the institute's record carries the Act's full condition.
- 2026-11-17 · Bangladesh: Following every screened woman electronically, through community clinics: The card said community clinics enrol more than half of the women and cited the Nessa paper, which gives 22.6% for community clinics and 51.8% for upazila health complexes. The figure of more than half comes from a different source and period, the programme's dashboard as archived in March 2025. (Second accuracy pass; the paper's full text read from Europe PMC and the archived dashboard read as raw text.) · Both figures are now given, each with its source and period, and the dashboard is linked from the card.
- 2026-11-17 · Bangladesh: A country that makes 98 per cent of its own medicines and exports to more than 150 countries: The card said the regulator's report states that most raw materials are still imported. The report says only that a park for raw material makers was built at Gazaria to make the country self-sufficient, with 44 plots allocated and three makers in production. (Second accuracy pass; the Directorate General of Drug Administration's Bengali annual report for 2024-25 read as raw text, sections 9.1 to 9.3.) · The sentence now says what the report says and notes that it gives no import share.
- 2026-11-17 · Bangladesh: Almost no international cancer trials: The card gave 6,501 trial records as all carrying registry countries, and called Incepta's study the only one from a Bangladeshi company. On 17 November 2026 OnCo held 6,525 trial records, of which 5,578 carry a list of countries, and the registry also lists two studies led by a Dhaka sponsor named Clinical Research Organization. (Second accuracy pass; OnCo's trial data counted in the repository and the ClinicalTrials.gov API read as raw JSON.) · The counts are corrected and dated, and the sentence on industry sponsors now reports the registry's own sponsor class for the 48 studies.
- 2026-11-17 · Bangladesh: No published approval times or prices, and no count of working radiotherapy machines: The card said no official count of linear accelerators or cobalt units in Bangladesh was found. IAEA's Directory of Radiotherapy Centres lists 22 centres with 26 linear accelerators and 5 cobalt-60 units, 16 of them with brachytherapy, every entry last updated in 2026. (Second accuracy pass; the directory's public data feed read as raw JSON on 16 November 2026.) · The card now gives the directory's listing with its date and keeps the point that no source says how many machines were working.
- 2026-11-17 · Bangladesh: A hospital register since 2005, with reports years apart: The card said Bangladesh has never had a population-wide cancer register. The source, the director's message in the 2025 registry report, says only that there is no nationwide population-based registry now. (Second accuracy pass; the Hospital Cancer Registry Report 2025 read as raw text.) · The sentence now reports the director's statement and no longer says never.
- 2026-11-17 · Bangladesh: A 2023 law, reliance on eight reference regulators, and no published approval times: The card said the reliance guideline covers biological medicines including antibodies used in cancer. The guideline speaks of vaccines and biological products and names neither antibodies nor cancer. (Second accuracy pass; the 2023 reliance guideline read as raw text.) · The added phrase was removed.
- 2026-11-17 · Viet Nam: About 177,400 new cancers a year, and how Viet Nam counts: The page said outright that Viet Nam has no national registry and publishes no national count, and gave that as the reason IARC's figures are estimates. No source read says either; IARC's fact sheet says only that its estimate is built from subnational registry rates. (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet and Decision 358/QD-BYT read in raw text.) · Now says that OnCo found no published national registry count, and states IARC's method separately, in the introduction, the GLOBOCAN note, two cards and the table note.
- 2026-11-17 · Viet Nam: What the city registries show: The card said nasopharyngeal cancer was falling and had given way to mouth cancer in men. The paper reports nasopharyngeal and sinonasal cancers together, as a share of cases (68% to 21% in men), not a rate for nasopharyngeal cancer alone. (Second accuracy pass; abstract of Liu et al., JCO Global Oncology 2026, read through Europe PMC.) · Title and text now give the combined group and the share of cases; the paper's own record carries the same wording.
- 2026-11-17 · Viet Nam: A list of 76 cancer and immune medicines: The Nguoi Lao Dong report was dated 18 June 2026; it was published on 19 June 2026. The card also said a medicine must be registered with the Drug Administration, where the circular accepts a valid registration or an import licence. (Second accuracy pass; the article's own publication date and Article 3 of Circular 37/2024/TT-BYT read in raw text.) · Date corrected in the text and link labels; the import licence added on the card and on the Ministry of Health record.
- 2026-11-17 · Viet Nam: A new list with 30 more cancer medicines, still in draft: The card said some medicines would be paid at 30% to 70% and credited this to the ministry's reply to voters. The newspaper reports it as a proposal described by the ministry's Health Insurance Department. The Ministry of Health record said the ministry had consulted on the list, which the sources do not say. (Second accuracy pass; Nguoi Lao Dong, 19 June 2026, and Dan Tri, 28 May 2026, read in raw text.) · Reworded as a proposal with the right attribution; the record now says the ministry drafted the list.
- 2026-11-17 · Viet Nam: 70 trials in OnCo's records have a site in Viet Nam: The count of 1,304 studies with a location in Viet Nam was credited to an OpenAlex snapshot. It is a ClinicalTrials.gov count held in OnCo's country snapshot. (Second accuracy pass; the snapshot file and the script that builds it.) · Attribution corrected; the number is unchanged.
- 2026-11-17 · Thailand: figures compared with GLOBOCAN: The note said the registry's years, 2019 to 2021, were years in which COVID-19 disrupted diagnosis. Cancer in Thailand Vol. XI does not mention COVID-19. (Second accuracy pass; Cancer in Thailand Vol. XI read in raw text (pdftotext) and searched for the claim.) · The clause was removed.
- 2026-11-17 · Thailand: liver and bile duct cancer and the national profile: Three passages said the national report attributes or credits the fall in liver and bile duct cancer to hepatitis vaccination, screening and liver fluke prevention. The report says the fall may be due to them. The profile also said cases rose with an older population, which the report says only of particular cancers. (Second accuracy pass; Cancer in Thailand Vol. XI read in raw text.) · Reworded to the report's own qualified wording in all three places, and the profile sentence now states only the fall in rates and the rise in estimated cases.
- 2026-11-17 · Thailand: how Thailand counts: The card stated as fact that Thailand has no single registry that counts every cancer. The report describes how the national estimate is built from 16 of 19 registries and says nothing more. (Second accuracy pass; Cancer in Thailand Vol. XI, chapters 1 and 2, read in raw text.) · Reworded to say the national figures are estimates, not a count of every case.
- 2026-11-17 · Thailand: Cancer Anywhere: The page called the cancelled April 2025 rules a tighter version and described the no-referral rule as the current rules. The source is a February 2025 report that calls the cancelled rules only a new version. (Second accuracy pass; Thai Post report of 6 February 2025 read in raw text.) · Now says a revised version was cancelled and that the rules restored in 2025 need no referral letter, on the page and in the National Health Security Office record.
- 2026-11-17 · Thailand: Thai FDA review routes: The card applied a footnote about when the review clock starts to every route, and described ASEAN zone IVb as hot and very humid. In the Thai FDA's slides the footnote belongs to the two collaborative procedures only, and the climate description is not there. (Second accuracy pass; the Thai FDA's symposium slides read in raw text (pdftotext).) · The clock sentence now names the collaborative procedures, the climate gloss was removed, and the gap card no longer says the 220 and 154 day targets run from a valid application.
- 2026-11-17 · Thailand: cost-effectiveness threshold: The card and two records said 160,000 baht was about 1.2 times gross national income per head and that a 2026 paper described price negotiation. HITAP's brief says gross national income without per head, and the paper's abstract says negotiation without price. (Second accuracy pass; HITAP's July 2015 policy brief read in raw text and the paper's abstract read through Europe PMC.) · Both phrases now follow the sources, on the page and in the HITAP and paper records.
- 2026-11-17 · Thailand: civil servants' scheme: The plain text said the scheme's OCPA system pays for cancer and blood cancer medicines and gave its 2018 rules in the present tense. The source speaks of cancer and haematology patients and is a statement from July 2018. (Second accuracy pass; Thai Post report of 17 July 2018 read in raw text.) · Reworded to medicines for cancer and blood diseases, and dated to the department's 2018 description.
- 2026-11-17 · Thailand: bowel screening: The card called the faecal immunochemical test a home stool test. The National Health Security Office's statement does not say where the test is done. (Second accuracy pass; PPTV report of 6 July 2026 read in raw text.) · The word home was removed.
- 2026-11-17 · Thailand: a targeted cancer medicine made in Thailand: The card gave four cancers as the medicine's listed uses and called them the indications of imatinib. The report gives the four as examples and does not mention imatinib. (Second accuracy pass; Thai PBS report of 4 June 2026 read in raw text.) · Now says the report names these among its uses, and the comparison was removed.
- 2026-11-17 · Thailand: screening paid for by one fund: The card said a single fund gives one national dataset of who was tested and who was followed up. The office's reports give follow-up rates for gold-card members only and make no such statement. (Second accuracy pass; PPTV reports of 6 July and 7 June 2026 read in raw text.) · The sentence was removed and the follow-up rates are described as covering gold-card members only.
- 2026-11-17 · Thailand: what could not be read: Two gap cards said the National Health Security Office's sites could not be reached and that current scheme membership could not be read, and a third said the National Drug Policy Division's live site returned an error. On 17 November 2026 the office's English site answered and gives 47.026 million people entitled to the Universal Coverage Scheme in fiscal year 2025, and the division's page of notices answered. (Second accuracy pass; eng.nhso.go.th and ndp.fda.moph.go.th fetched again as raw text on 17 November 2026.) · The gap cards now say which sites answer and which still do not, give the 2025 figure with its source, and keep the gap for the other two schemes.
- 2026-11-17 · Iran: page description and opening paragraph: The page said that public insurers, since 2014, aim to cap a patient's share of a hospital bill at 10%. The source describes the cap as part of the 2014 Health Transformation Plan, in a programme managed directly by the health ministry, and it applies to stays in public hospitals. (Second accuracy pass; the open-access full text of Doshmangir et al. (Int J Equity Health 2021) read as raw text from Europe PMC.) · Reworded to a 2014 reform, run by the health ministry, that aimed to cap what a patient pays for a stay in a public hospital at a tenth of the bill.
- 2026-11-17 · Iran: What insurance pays for in breast cancer, and what families still pay: The household spending figures were given as a survey of 138 women with breast cancer without saying that all were interviewed in one city, Urmia. (Second accuracy pass; the full text of Ahmadi et al. (BMC Health Serv Res 2021) read as raw text from Europe PMC.) · The card now says the survey was done in one city, Urmia.
- 2026-11-17 · Iran: Cancer care under sanctions: The card said that without insurance only the richest tenth of households could afford the children's cancer regimens studied in Urmia. That is the paper's summary, but its results table differs by regimen: no income group could afford the leukaemia regimen without insurance, and every group could afford the brain tumour regimen. (Second accuracy pass; the full text and Table 5 of Baghban et al. (BMC Pediatr 2025) read as raw text from Europe PMC.) · The card now gives both the paper's summary and what its results table shows.
- 2026-11-17 · Iran: Ministry of Health and Medical Education record: The record said the ministry delivers care, public health and training through universities of medical sciences in each province. None of the sources it cites says that. (Second accuracy pass; the archived ministry page, the 2016 registry paper and Doshmangir et al. (Int J Equity Health 2021) read as raw text.) · Reworded to what the policy analysis says: the ministry was established in 1985, replacing the earlier health ministry, in part to answer a lack of integration between medical schools and the community. The source is now linked from the record.
- 2026-11-17 · Iran: Iran Health Insurance Organization record: The one-line description said the insurer pays for cancer care through its fund for special and hard-to-treat diseases. Its news item says only that cancer is among the commonest markers and among the conditions with the highest spending in that fund. (Second accuracy pass; the insurer's news item of 16 April 2023 read in Persian as raw text from its Internet Archive copy of 11 June 2023.) · Reworded to say the insurer has a fund for special and hard-to-treat diseases in which cancer is among the costliest conditions.
- 2026-11-17 · Iran: AryoGen Pharmed record: The record presented two aims, meeting the medicine needs of patients with hard-to-treat diseases and localising every stage of production, as part of the company's mission statement. On the company's page they are in the research and development paragraph, not the mission line. (Second accuracy pass; the company's about page read in Persian as raw text from its Internet Archive copy of 17 May 2025.) · The record now attributes each statement to the part of the page it comes from.
- 2026-11-17 · Ukraine: Outpatient medicines: hormone therapy for breast cancer is reimbursed: The card said no other cancer medicine was on the 2026 reimbursement list and that targeted therapies were not on it, and it named the Affordable Medicines scheme, pharmacies and electronic prescriptions. The list also holds methotrexate, everolimus, morphine and fentanyl under other headings, and the resolution cited does not name the scheme or say how medicines are collected. The card's title also said pharmacy medicines. (Second accuracy pass; section III of Cabinet resolution No. 1808 read in raw text from the Internet Archive copy of zakon.rada.gov.ua.) · The card and the National Health Service of Ukraine record now say that three medicines are listed under the heading of cancer, name the other four and their headings, and say what the resolution does not cover. The title now says outpatient medicines.
- 2026-11-17 · Ukraine: The registry measured how much later cancers are found since 2020: The rise in late diagnosis from 26.1% to 30.9% in men and from 14.6% to 16.8% in women was given for solid cancers other than skin cancer. The registry's group leaves out only non-melanoma skin cancer; melanoma is included. (Second accuracy pass; the thematic article in the National Cancer Registry of Ukraine's Bulletin No. 27, read in raw text from the PDF.) · The sentence now says solid cancers other than non-melanoma skin cancer.
- 2026-11-17 · Ukraine: Expensive cancer medicines are bought centrally and given to patients free: The title said most cancer medicines are bought centrally and given free in hospital, and the text said they are bought for cancer centres. No source read gives a share of cancer medicines or those two details. (Second accuracy pass; the Ministry of Health announcement of 22 October 2025 and the Interfax-Ukraine report of 4 February 2026, read in raw text.) · The title and the sentence now say only what the Ministry says: expensive cancer medicines are bought centrally and patients receive them free.
- 2026-11-17 · Ukraine: The National Health Service of Ukraine buys cancer care in packages: The blood cancer package was given as 74,429 hryvnias per patient per year. The resolution sets that figure as a rate to which weights are applied, as it does for chemotherapy. (Second accuracy pass; chapter 21 of Cabinet resolution No. 1808 read in raw text from the Internet Archive copy of zakon.rada.gov.ua.) · The card and the National Health Service of Ukraine record now say the package is built from that rate with its own weights.
- 2026-11-17 · Ukraine: Patients evacuated for treatment abroad, and equipment rebuilt at home: MedEvac was called a Ministry programme, the National Cancer Institute's third linear accelerator was said to have been bought with the Ministry's and the procurement agency's support, and the Zaporizhzhia pilot was placed close to the front line. The paper's abstract says only that the data came from the Ministry's database, the Institute credits the two bodies' support without saying who bought the machine, and the second paper calls the region war-affected. (Second accuracy pass; the two abstracts from Europe PMC and the National Cancer Institute's announcement of 28 July 2026, read in raw text.) · Each sentence now uses the source's own description.
- 2026-11-17 · Ukraine: A population registry in every region, built on compulsory reporting since 1953: The card said the registry's linkage with the Chornobyl register has been used in studies of the accident's consequences. The registry's history says the linkage matched and filled in both registers and was used in joint studies with two named institutes. (Second accuracy pass; the registry's thirtieth anniversary history in Bulletin No. 27, read in raw text from the PDF.) · The sentence now says what the history says.
- 2026-11-17 · Ukraine: 124,801 cancers registered in 2024, and prostate cancer now the commonest in men: The card said the registry attributes the 35% fall in childhood cancers since 2021 partly to hospitals not reporting. The registry says the fall may be a consequence of that and that migration cannot be ruled out. (Second accuracy pass; the analytical review in Bulletin No. 27, read in raw text from the PDF.) · The sentence now gives it as a possibility, as the registry does.
- 2026-11-17 · Nigeria: No fully established national programmes yet for breast, prostate or bowel cancer; HPV vaccine since 2023: The card and a gap entry said no published national HPV vaccination coverage figure had been found. The World Health Organization's immunisation data holds national figures for Nigeria. (Second accuracy pass; the WHO immunisation coverage feed for Nigeria was queried for every HPV antigen and read as raw JSON.) · The card now gives WHO's figures for girls: first-dose programme coverage of 27% in 2023, 60% in 2024 and 60% in 2025, coverage by age 15 of 28% in 2024 and 67% in 2025, and the administrative first-dose figure of 84% for 2025. The gap entry no longer lists HPV coverage as missing.
- 2026-11-17 · Nigeria: Cancer has been notifiable since 2024: The lede joined two facts with "so", saying national figures are IARC estimates because cancer is counted in only a few cities. The registry card and the registry record said as fact that no national count exists, although the page also reports NICRAT describing a national burden report that OnCo could not find. The registry record stated that no Nigerian registry has contributed to Cancer Incidence in Five Continents beside a sentence saying Ibadan's data were in its first three volumes. (Second accuracy pass; IARC's GLOBOCAN 2024 fact sheet, the imPACT Review summary of September 2025 and the full text of Jedy-Agba and colleagues (2015) were read in raw text.) · The lede states the two facts separately. The card and record now say OnCo could not find a published national count. The record attributes the Cancer Incidence in Five Continents statement to the imPACT Review and notes that it does not match the 2015 paper.
- 2026-11-17 · Nigeria: Public funds and price deals for patients who cannot pay: The card said the access partnership sells medicines at half price or less and that the national plan says it cuts prices by 50 to 60%. The plan's wording is "at drastically reduced rates of 50-60% to the patient", and another passage says the cost falls "by as much as 50%". (Second accuracy pass; the National Cancer Control Plan 2026 to 2030 was read as PDF text with OCR of its image-only pages.) · The card now quotes the plan's two passages as written and no longer says "or less".
- 2026-11-17 · Nigeria: For a medicine approved by a stringent regulator, NAFDAC commits to a decision in about 90 days: The card said NAFDAC's own page dates its official establishment to October 1992. The page says only that NAFDAC was established by Decree No. 15 of 1993 as amended by Decree No. 19 of 1999. (Second accuracy pass; NAFDAC's "About NAFDAC" page was read in raw text.) · The 1992 clause was removed and the sentence now follows NAFDAC's page.
- 2026-11-17 · Nigeria: A national health technology assessment system, started in 2025: The card and the NAFDAC and NHIA records said the national steering committee was inaugurated in September 2026, and the NHIA record said the NHIA hosts it. The NHIA's article is dated 22 September 2026, gives no inauguration date and lists the NHIA's assessment division as the committee's secretariat. (Second accuracy pass; the NHIA's article of 22 September 2026 was read in raw text.) · The three texts now say the NHIA reported the inauguration on 22 September 2026, and the NHIA record says its assessment division is the secretariat.
- 2026-11-17 · Nigeria: Radiotherapy through public and private partnerships: The card gave only the count made by international reviewers in October 2024 (15 working machines) and did not compare it with the IAEA's directory of radiotherapy centres. (Second accuracy pass; the directory's public data feed, fetched as raw JSON on 16 November 2026, was read row by row for Nigeria.) · The card now also gives the directory's listing: 15 centres with 16 linear accelerators and 2 cobalt-60 units, 10 centres with brachytherapy, entries last updated in 2025 or 2026, with a note that the two counts are not the same measure.
- 2026-11-17 · Nigeria: Five year cancer incidence in Calabar (paper record): The record described Calabar's registry as being in the south-east. The paper says only that Calabar is the capital of Cross River State. (Second accuracy pass; the paper's abstract was read through Europe PMC after resolving its DOI.) · The record now says Calabar in Cross River State.
- 2026-11-17 · Nigeria: sources on NICRAT's website: Claims taken from NICRAT's 2023 to 2027 plan, its cervical screening guidelines, its January to August 2026 activity report and its news pages could not be checked again, because NICRAT's website answered every request with a bot check. (Second accuracy pass; each NICRAT address was requested as raw text and returned a bot verification page, which was not bypassed.) · The registry, funds, cervical screening, other screening, workforce, radiotherapy and research cards, two gap entries and the NICRAT and registry records now say these sources could not be re-read on 17 November 2026.
- 2026-11-17 · The Philippines: Abridged and verification reviews for medicines already approved by reference regulators: The card, the gap card and the checking note said the FDA's web pages refused requests, that the base Circular 2022-004 could not be found and that its review times were known only from a US Commercial Service summary. (Second accuracy pass; the FDA's own page for Circular 2022-004 answered on 17 November 2026 and was read in raw text.) · The page now says the circular was read on the FDA's page and gives not more than 45 working days for an abridged review and 30 for a verification review, the same as the US summary; the gap card no longer lists the FDA's pages as unreadable.
- 2026-11-17 · The Philippines: trials with a site in the Philippines: The card counted 69 trials, 4 of them phase 2/3 and 15 recruiting, and said one was a bladder-function study in children. (Second accuracy pass; the ClinicalTrials.gov participation data and trial records in OnCo's own corpus were recounted.) · The count is now 68 trials, 61 phase 3, 4 phase 2 and 3 phase 2/3, with 14 recruiting, dated 17 November 2026; the sentence about a children's bladder study, which matches no trial in the set, is removed.
- 2026-11-17 · The Philippines: Figures looked for and not found: The gap card said HPV vaccination coverage after January 2025 was not found from a primary source. (Second accuracy pass; the World Health Organization's immunisation coverage data for the Philippines was read as raw JSON.) · The card now gives WHO's administrative coverage in girls: 72.28 per cent for a first dose and 6.73 per cent for a final dose in 2024, and 24.36 per cent for a first dose in 2025, with the 2025 final-dose rows blank.
- 2026-11-17 · The Philippines: Cervical screening and HPV vaccination: The card said the two HPV doses are given six months apart and wrote P4,000 a dose as if it were part of the free offer. (Second accuracy pass; the GMA News report of 18 January 2025 and the BusinessWorld report of 7 April 2025 were read in raw text.) · The interval, which neither report gives, is removed, and P4,000 a dose is stated as the cost to the government quoted from the Secretary of Health.
- 2026-11-17 · The Philippines: Mammograms, low-dose CT, liver tests and colonoscopy: The start date of 14 August 2025 was given as if it came from PhilHealth Circular 2025-0014. (Second accuracy pass; the circular (pdftotext) and the PTV News report of 8 August 2025 were read in raw text. The circular takes effect fifteen days after publication and gives no date.) · The card and the table caption now say PhilHealth announced on 8 August 2025 that the cover would start on 14 August, as PTV News reported.
- 2026-11-17 · The Philippines: Every Filipino is a PhilHealth member: The card said the premium rose to 5 per cent of monthly income in 2024 and 2025 as a fact about what was charged. (Second accuracy pass; Republic Act 11223 was read in raw text from Lawphil. Section 10 is a schedule of rates.) · The card now says the Act's schedule sets the premium at 5 per cent for 2024 and 2025.
- 2026-11-17 · The Philippines: People with cancer and survivors have the legal rights of persons with disabilities: The card said the cancer law bans discrimination against people with cancer. (Second accuracy pass; Section 27 of Republic Act 11215 was read in raw text from Lawphil.) · The card now says the law requires government agencies to ensure that people living with cancer and survivors are free from discrimination at school, at work and in the community.
- 2026-11-17 · The Philippines: A health technology assessment council: The card said the council must make its deliberations public. (Second accuracy pass; Section 34 of Republic Act 11223 was read in raw text from Lawphil.) · The card now says the council must make the result of its deliberations public.
- 2026-11-17 · The Philippines: how to check this page: The description gave 72,855 registered cancer deaths without saying the count is provisional and is of neoplasms, and the checking note said IARC's estimate is built from Metro Manila registry rates. (Second accuracy pass; the Philippine Statistics Authority release of 26 September 2024 (Internet Archive copy) and IARC's GLOBOCAN 2024 fact sheet were read with pdftotext. The fact sheet says only that the data source is a subnational registry.) · The description now says registered deaths from neoplasms, a provisional count, and the checking note says subnational registry rates.
- 2026-11-17 · Food and Drug Administration, Philippines: The record said the text of FDA Circular 2022-004 was not read. (Second accuracy pass; the FDA's own page for Circular 2022-004 was read in raw text on 17 November 2026.) · The record now says the FDA's page gives the same review times as the US Commercial Service summary, and links to it.
- 2026-11-17 · Philippine Cancer Center: The one-line description said construction of the 20-storey building began in March 2024. (Second accuracy pass; the GMA News report of 8 March 2024 was read in raw text. It reports a groundbreaking ceremony.) · The description now says ground was broken in March 2024.
- 2026-11-17 · Ting et al., colorectal Z package at the pilot site: The one-line description said 70 per cent of patients were alive at three years. (Second accuracy pass; the abstract was read through Europe PMC. It reports a three-year survival probability of 70 per cent and mortality of 19 per cent.) · The description now gives the three-year survival probability.
- 2026-11-17 · Argentina: 445 cancer trials in OnCo's records have a site in Argentina, almost all run by companies: The page counted 446 trials with a site in Argentina, gave 42 phase 2 trials and 94 completed, and named MOUNTAINEER as a five-country trial that included Argentina. One registry study, MOUNTAINEER-03 (NCT05253651), had been counted twice because OnCo's data attach it to two records, and the five-country MOUNTAINEER trial (NCT03043313) lists Belgium, France, Italy, Spain and the United States, not Argentina. (Second accuracy pass; the ClinicalTrials.gov participation data held in OnCo (src/data/trial-participation.ts, retrieved 7 October 2026) were recounted by registry number.) · The count is now 445 studies, 41 phase 2, 93 completed and 1 not yet recruiting; the example is replaced by two studies that do list Argentina among four and five countries, and the page's trial list now links MOUNTAINEER-03 instead of MOUNTAINEER.
- 2026-11-17 · Argentina: A law that gives every child with cancer full cover, a monthly allowance and leave for a parent: The card said the Ministry of Health's page lists the parent's leave as paid by the social security administration in place of the salary. The Ministry's page lists the leave but does not say who pays. (Second accuracy pass; the Ministry's page on Law 27.674 and the text of the law on argentina.gob.ar/normativa were read as raw text.) · The payment is now attributed to article 13 of the law, which makes the leave unpaid by the employer while the social security administration pays a sum equal to the salary, and the law's text is linked.
- 2026-11-17 · Argentina: Six in ten people have social insurance or a prepaid plan; the rest rely on public hospitals: The card gave the census health cover figures as shares of the whole population counted, and said retired people are mostly covered by PAMI. The census volume reports cover for the population in private dwellings, and PAMI's page says only that it covers 5 million retired people, dependants, pensioners and veterans. (Second accuracy pass; the 2022 census volume on health and social security (PDF, pdftotext) and PAMI's history page were read as raw text.) · The figures are now described as the population in private dwellings, the same qualifier is added where the registries card compares registry coverage with the census, and the word mostly is removed.
- 2026-11-17 · Argentina: PAMI pays all of the cost of cancer medicines; a public drug bank supplies the uninsured: Two cards said the Banco de Drogas Especiales supplies medicines through public hospitals. Its page requires that the patient be treated in a public hospital and that documents go to the province, but does not say the medicine is delivered through the hospital. (Second accuracy pass; the Ministry of Health's Banco de Drogas page was read as raw text.) · Both cards now say the bank supplies people treated in a public hospital.
- 2026-11-17 · Argentina: Two in three women had a recent mammogram in 2018; one in three people aged 50 to 75 had ever had a bowel test: The survey figures were given without their scope, and the screening card called the bowel test a home test, which the programme's page does not say. (Second accuracy pass; the 4th National Risk Factor Survey (PDF, pdftotext) and the colorectal programme's page were read as raw text.) · The card now says the survey covers localities of 5,000 or more inhabitants, and the word home is removed.
- 2026-11-17 · Argentina: Childhood cancer: about 1,358 children a year, and 70% alive five years later: The card and the Hospital Garrahan record dated to 2013 to 2022 the finding that 45% of cases reported by public hospitals were treated in two reference centres. The registry's book gives that sentence no period. (Second accuracy pass; the childhood cancer registry's 2025 book (PDF, pdftotext) was read as raw text.) · The period is removed from that sentence in the card and in the record.
- 2026-11-17 · Argentina: Sources and figures that could not be read on 4 November 2026: The gap card said the Ministry's legislation pages and the live Boletín Oficial did not answer, and two sentences stated as fact that ANMAT and other official bodies publish no approval times. The two sites answered on 17 November 2026, and only the pages read had been searched for approval times. (Second accuracy pass; argentina.gob.ar/normativa and boletinoficial.gob.ar were fetched as raw text on 17 November 2026.) · The card now says both sites answered and what was read there, and the two sentences now say the figures were not found on the pages read.
- 2026-11-17 · Egypt: Universal health insurance: no fees for people with cancer, rolled out governorate by governorate: The card gave the law's exemption of people with chronic diseases and tumours from co-payments without its condition, and gave the caps of 1,000, 750 and 300 pounds as if they were current amounts. (Second accuracy pass; the Andersen Egypt English translation of Law 2 of 2018 read in raw text, where article 40 makes the exemption subject to a decision of the Prime Minister and a note to the tables raises the fixed amounts by 7% each year, and the Universal Health Insurance Authority's post of 29 December 2025 read in Arabic.) · The card now states the condition, cites the authority's own reference to Prime Minister's decree 2273 of 2021, and says the caps are the amounts as written in the law, which rise by 7% a year.
- 2026-11-17 · Egypt: Old health insurance, treatment at state expense, and a new universal scheme: The card said students, young children and farmers were covered by the Health Insurance Organization and that people without cover can apply for treatment at state expense; the page description said state expense covers everyone else. None of the sources read says who runs the separate schemes or who qualifies for state-expense treatment. (Second accuracy pass; the Andersen Egypt translation of Law 2 of 2018 and the Al Mal News report of the Health Ministry's statement of 18 January 2026 read in raw text.) · The card now says a 1964 decree set up the organisation for government and public-sector employees, that later laws set up insurance for the other groups, and that state-expense treatment is granted by decision without the sources setting out who qualifies. The description was reworded to match.
- 2026-11-17 · Egypt: A free breast risk assessment for every woman from 18, with treatment paid for: The page description, opening paragraph and card title said the presidential initiative offers a breast check to every woman from 18, and the paper's record said Egypt chose clinical breast examination for all adult women. The evaluation says all women are invited to a risk questionnaire and only those classified as at risk receive the examination. The record also said the paper was written by the programme's leaders, which the paper does not state. (Second accuracy pass; the full text of Abdelghaffar et al., Archives of Public Health 2026, read from Europe PMC in raw text.) · The description, opening paragraph, card title and the paper's record now say a free breast risk assessment, with examination for women found at risk, and the caveat now says the authors are largely staff of the Health Ministry and of the initiative, with publication funded by Roche Egypt.
- 2026-11-17 · Egypt: No national cancer incidence found for any year after 2011: The page said as fact that Egypt's registry has not published national incidence after 2011 and that OnCo could find no registry website. No source says nothing was published, and the 2014 report itself gives a website address for the programme. The card also said a 2021 regional survey mentions Egypt's registry only to note an unanswered question, when the survey also names Egypt among the 19 countries that sent data. (Second accuracy pass; the full text of Ibrahim et al., Journal of Cancer Epidemiology 2014, and of Abdul-Sater et al., JCO Global Oncology 2021, read from Europe PMC in raw text; the address cancerregistry.gov.eg was requested on 17 November 2026 and did not answer.) · The gap card, the liver cancer card and the registry's record now say that OnCo found no later national report, name the website the 2014 report gives and say it did not answer on 17 November 2026, and describe the survey as it reads.
- 2026-11-17 · Egypt: Hepatitis C: more than 60 million people tested and over 4 million treated: The card and the list of unreachable sources said WHO's Eastern Mediterranean page returned a server error and could be read only from an archived copy. The live page answers. The record of the 2018 liver cancer forecast also dated the national screening campaign to 2018 to 2019, dates WHO's page does not give. (Second accuracy pass; WHO's Eastern Mediterranean announcement of 9 October 2023 read in raw text from the live site on 17 November 2026 and compared with the archived copy.) · Both notes now say the live page answered on 17 November 2026 with the same figures, and the paper's caveat now says only that it was published in 2018, before WHO reported the campaign's reach in 2023.
- 2026-11-17 · Netherlands: Two clocks: European approval, then Dutch reimbursement: The card said patients in the daraxonrasib trial lived "a median 12.4 months" and that Dutch reimbursement "can take longer" than European approval. The regulator's notice says "on average" and does not mention medians, and no source gives the two durations. (Second accuracy pass; the CBG news item of 15 October 2026 read in raw text.) · The figures are now given as the CBG words them, with the patient group and trial size it states, and the card says only that the reimbursement step starts after approval.
- 2026-11-17 · Netherlands: Off-label use as a national trial: DRUP and the Drug Access Protocol: The card said successful cohorts "can make the case for payment" and that the fresh biopsy is for whole-genome sequencing. The trial registry says next-generation sequencing, and neither it nor the paper's abstract mentions payment. (Second accuracy pass; the ClinicalTrials.gov record NCT02925234 and the Nature 2019 abstract read in raw text.) · Reworded to say results are recorded and shared to inform later decisions, and that the biopsy is taken for sequencing.
- 2026-11-17 · Netherlands: Whole-genome sequencing of metastatic cancer, pooled in one national database: The card said DRUP takes its biopsies for the same sequencing as the Hartwig database, which none of the cited sources states. (Second accuracy pass; the Hartwig Medical Foundation database page, the DRUP registry record and both Nature abstracts read in raw text.) · Replaced with what the paper record shows: Hartwig Medical Foundation is among the affiliations of the DRUP report's authors.
- 2026-11-17 · Netherlands: Concentrating care: every child in one hospital, and minimum volumes for adults: SONCOS was described as a foundation of the specialist societies. Version 14 of its report describes it as a platform that is part of the Federatie Medisch Specialisten. (Second accuracy pass; the SONCOS standards report version 14 (2026) read as pdftotext output.) · Description corrected; the volume norms were re-read and stand.
- 2026-11-17 · Netherlands: Three national screening programmes, with a home test for cervical screening: The sentence gave the 10 to 15 years that cervical cancer takes to develop as the reason for the home test. RIVM gives it as the reason more frequent testing is not needed. The statement that there is no programme for prostate or lung cancer had no source. (Second accuracy pass; RIVM's cervical screening page and its overview of population screening read in raw text.) · Sentence corrected, and the three-programme statement now cites RIVM's overview page.
- 2026-11-17 · Netherlands: What could not be sourced: The card said the central government's pages on insurance could not be read by a script. The page on the deductible answered in full. (Second accuracy pass; the Rijksoverheid page on the compulsory deductible fetched as raw text on 17 November 2026.) · The card now says the page was read and gives the same €385 as the Zorginstituut.
- 2026-11-17 · Netherlands: Three counting choices to know before comparing Dutch figures with anyone else's: The card called basal cell carcinoma the commonest skin cancer of all, which IKNL's page does not say, and said GLOBOCAN was close to the registry's counts when only the case count is close: GLOBOCAN estimates 51,051 deaths for 2024 against IKNL's "just over 46,000". (Second accuracy pass; IKNL's guide to registry figures and the IARC fact sheet read in raw text.) · Reworded to IKNL's own description, and the comparison is limited to new cases.
- 2026-11-17 · Netherlands: Bowel cancer is falling, which IKNL links to screening: The title said the fall was what screening "was expected to do". IKNL links the fall to screening but states no such expectation. (Second accuracy pass; IKNL's release of 4 February 2026 read in raw text.) · Title reworded.
- 2026-11-17 · Zorginstituut Nederland: The record read as if a 60 per cent price cut for durvalumab had followed on 29 August 2026. The Zorginstituut says only that the minister reached an agreement and the medicine has been reimbursed since that date. (Second accuracy pass; the Zorginstituut's package advice on durvalumab read in raw text.) · Reworded to say what followed was reimbursement after an agreement.
- 2026-11-17 · DRUP (Drug Rediscovery Protocol): The record said the biopsy is for whole-genome sequencing and that the drugs that work "can be paid for". The registry says next-generation sequencing and does not mention payment. (Second accuracy pass; the ClinicalTrials.gov record NCT02925234 read as JSON.) · Both statements corrected to the registry's wording.
- 2026-11-17 · Colombia: Every new medicine is graded for therapeutic value against the best treatment used in Colombia: The card said that since 2018 a company registering a new medicine must also file a comparison and its intended price. Decreto 433 de 2018 does not say the duty began in 2018: its final article makes the decree applicable three months after the assessment manuals and the price method are issued, and not before 5 December 2018. (Second accuracy pass; Decreto 433 de 2018 read in raw text from INVIMA's legal compilation.) · Reworded to "under a 2018 decree" and the decree's own applicability clause added.
- 2026-11-17 · Colombia: How Colombia counts: The card said Colombia has no single registry that counts every cancer. The source says only that Colombia has seven population-based registries covering small areas, which are cities. (Second accuracy pass; Pardo-Ramos and Cendales-Duarte, Revista Colombiana de Cancerología 2024, read in raw text from the journal's PDF.) · Reworded to say that the population-based registries cover cities, not the whole country.
- 2026-11-17 · Colombia: What cancer looks like in Colombia: The card gave rising deaths from bowel cancer and falling deaths from breast, prostate, stomach and cervical cancer as a general finding. The cancer plan makes that statement only about the cancers the Ministry prioritises. (Second accuracy pass; Resolución 1233 de 2026, technical annex section 4.5.2, read in raw text from the PDF.) · The qualifier was added.
- 2026-11-17 · Colombia: A backlog of more than 14,000 procedures in 2025: The card said INVIMA adopted its contingency plan in March 2025 to clear more than 14,000 procedures. INVIMA's release of 20 March 2025 announces the plan without giving the date it was adopted, and says the aim is to speed the procedures up. (Second accuracy pass; INVIMA press release of 20 March 2025 read in raw text.) · Reworded to say INVIMA announced the plan in March 2025 to speed up more than 14,000 procedures.
- 2026-11-17 · INVIMA: The record repeated two statements corrected on the Colombia page: that the comparative dossier has been required since Decreto 433 de 2018, and that the contingency plan was adopted in March 2025 to clear pending procedures. (Second accuracy pass; Decreto 433 de 2018 and INVIMA's release of 20 March 2025 read in raw text.) · Reworded to "under Decreto 433 de 2018" and to "announced" and "speed up".
- 2026-11-17 · Instituto de Evaluación Tecnológica en Salud: The record said the institute has classified the therapeutic value of new medicines since 2018. The decree fixes no start in 2018, only that it does not apply before 5 December 2018 and depends on manuals being issued. (Second accuracy pass; Decreto 433 de 2018 read in raw text from INVIMA's legal compilation.) · Reworded to "under a 2018 decree".
- 2026-11-17 · Ethiopia: HPV vaccine for girls since 2018, and a national push for earlier breast cancer diagnosis: The card called WHO's own HPV coverage estimates (58% in 2024, 51% in 2025) rows that Ethiopia had reported, and said the national plan's 80% baseline differed from WHO's data, without the administrative and official coverage rows WHO also holds for Ethiopia (90% to 95% final dose for 2018 to 2023, first dose 88% in 2024 and 96% in 2025). (Second accuracy pass; WHO's immunisation data API read as raw JSON with the antigen filter widened from the programme rows to every HPV row.) · The card and the gap note now give both sets of figures, each named for what it is, and the link opens the wider query.
- 2026-11-17 · Ethiopia: lede, GLOBOCAN note and what to verify: The page said national figures are IARC estimates built on Addis Ababa's registry because cancer is counted only in the capital. IARC's fact sheet says its source is a subnational registry and does not name it or give a reason. (Second accuracy pass; IARC's GLOBOCAN 2024 fact sheet for Ethiopia read as PDF text.) · The description, lede, note, verify line and two records now state IARC's method in its own terms and say separately that the Addis Ababa registry's papers call it the country's only population-based registry.
- 2026-11-17 · Ethiopia: How many radiotherapy centres are treating patients now is not confirmed: The gap card set out three conflicting counts and did not give the IAEA directory's listing. (Second accuracy pass; the IAEA Directory of Radiotherapy Centres public data feed, read as raw JSON on 16 November 2026.) · The card now says the directory lists two centres, Tikur Anbessa in Addis Ababa (one linear accelerator, one cobalt-60 unit, brachytherapy, entry of 2021) and Jimma (one linear accelerator, entry of 2022), and that these entries are old.
- 2026-11-17 · Ethiopia: Breast cancer chemotherapy taken out to regional hospitals: The card said Ethiopia trained teams in regional hospitals, that many women no longer have to travel to Addis Ababa for every cycle, and that oncology-trained doctors and nurses are the staff task-sharing relies on. The national plan says none of these. (Second accuracy pass; the Integrated National Cancer Control Plan 2025 to 2029 read as PDF text, section 1.7.2 and appendices 5 and 6.) · The card now says what the plan says: a national initiative to decentralise breast cancer chemotherapy, 30 hospitals equipped, and the staff counts in its workforce table.
- 2026-11-17 · Ethiopia: Where treatment is: The card said Ethiopia once had one cancer treatment centre at Tikur Anbessa and that the Jimma centre opened on 20 February 2022. The plan does not name the one earlier centre, and the Ministry page is dated 20 February 2022 without giving the opening day. (Second accuracy pass; the national plan read as PDF text and the Ministry of Health news page read as raw HTML.) · Reworded to what each source says: radiation treatment was only at the Black Lion hospital until Jimma opened, as reported on 20 February 2022.
- 2026-11-17 · Ethiopia: Community-based health insurance covers about 65 million people: The card stated as fact that a consultant's figure of 80% of citizens covered describes the share the scheme is meant for. The insurer's page does not say what the 80% measures. (Second accuracy pass; the Ethiopian Health Insurance Service news pages of June and August 2025 read as raw HTML.) · The reading is now given as a possibility, with the arithmetic (65 million of more than 115 million is about 57%) and a note that the page does not say.
- 2026-11-17 · Ethiopia: registry papers of 2018: The caveats on two paper records said case finding was still incomplete in the registry's first years and that regional cases were not from population-based registries. Neither is in the abstracts; the first abstract suggests the opposite. (Second accuracy pass; the abstracts of Timotewos et al. 2018 and Memirie et al. 2018 read through Europe PMC.) · Both caveats rewritten to what the abstracts say.
- 2026-11-17 · Romania: 325 cancer trials in OnCo's records have a site in Romania, all led from abroad: The card gave 326 trials with a Romanian site, 22 of them phase 2/3 and 107 recruiting. OnCo's records now hold 325, with 21 phase 2/3 and 106 recruiting. (Second accuracy pass; the ClinicalTrials.gov participation data in OnCo's own corpus was counted again.) · Title, summary and detail now give 325, 21 and 106, with the date of the recount.
- 2026-11-17 · Romania: A registry decided by government in 2023, with a 30-day reporting duty: The page said as present fact, in six places, that Romania has no working national cancer registry and no national count of new cancers. Its source is the OECD and European Commission profile of 2025, which said the registry was under development and expected to be fully working by the end of 2025. (Second accuracy pass; the OECD and European Commission Country Cancer Profile Romania 2025 read in raw text.) · The description, opening paragraph, GLOBOCAN note, deaths table note, registry card, gap card and closing note now say what the 2025 profile said and that no national count was found, instead of asserting that none exists.
- 2026-11-17 · Romania: Travelling for treatment: 24 of 42 counties without radiotherapy: The summary dated the 24 counties without radiotherapy equipment to 2022. The OECD profile gives no year for that figure. The card also gave no count of centres or machines. (Second accuracy pass; the OECD profile read in raw text, and the public data feed of the IAEA Directory of Radiotherapy Centres read on 16 November 2026.) · The year is removed from that figure and the card now adds what the IAEA directory lists: 55 centres with 97 linear accelerators, no cobalt-60 units and 2 gamma knife units, 16 centres with brachytherapy, every entry last updated in 2026.
- 2026-11-17 · Romania: Breast and bowel screening: EU-funded pilots, and 9% mammography coverage in 2019: The summary said Romania has no national breast or bowel screening programme yet. The OECD profile says organised programmes are in their initiation phases and that national methods were adopted in 2024; it does not say none exists. (Second accuracy pass; the OECD and European Commission Country Cancer Profile Romania 2025 read in raw text.) · The summary now says what the profile describes: pilots, national methods adopted in 2024, and organised programmes in their first phases.
- 2026-11-17 · Romania: Paid leave for carers, psychological support and a right to be forgotten: The card credited counselling sessions, paid transport and carers' leave to a 2023 regulation. The OECD profile says the 2023 regulation concerns psychology facilities for cancer patients and reports the new reimbursed services separately. (Second accuracy pass; the OECD and European Commission Country Cancer Profile Romania 2025 read in raw text.) · The sentence now keeps the two statements apart, as the profile does.
- 2026-11-17 · Romania: Cost-volume contracts, negotiated within a yearly sum: The summary said many new cancer medicines enter the reimbursed list on condition. Neither source counts them. (Second accuracy pass; Ministry of Health order 861/2014 and the Health Insurance House's cost-volume page, both read in raw text.) · The summary now says a new medicine can enter on condition and that cancer medicines are among those negotiated this way.
- 2026-11-17 · Romania: Sources that could not be read directly on 4 November 2026: The gap card said the European Cancer Inequalities Registry's site rejected scripts. On 17 November 2026 it served the OECD profile directly. (Second accuracy pass; the profile was fetched from the registry's own site and compared with the Internet Archive copy, which is identical.) · The card now gives both dates and says which official sites still refused automated reading on 17 November 2026.
- 2026-11-17 · South Africa: Current cervical screening coverage, a South African count of radiotherapy machines and two court outcomes were not found: The gap card said that no count of radiotherapy machines had been read and that no national HPV vaccination coverage was found, and spoke of a national switch to one dose that no source read for the page describes. (Second accuracy pass; the IAEA radiotherapy directory feed and the WHO immunisation coverage rows for South Africa read as raw JSON.) · The card now gives what the IAEA directory lists (59 centres, 99 linear accelerators, 3 cobalt-60 units, 19 centres with brachytherapy, entries last updated 2015 to 2022, one possible double entry) and WHO's first-dose HPV coverage of 79% for 2024, each with its date, and no longer mentions a national switch to one dose.
- 2026-11-17 · South Africa: HPV vaccination at school since 2014: The card title said the vaccine cut infection with the cancer-causing types by 83%, when the study measured the two types the vaccine targets and found no significant fall in most other types; the card also said no current national coverage figure had been found. (Second accuracy pass; the abstract of Machalek and colleagues read through Europe PMC and the WHO immunisation coverage rows read as raw JSON.) · The title now says the two virus types the vaccine targets, and the card gives WHO's first-dose coverage among girls of 79% for 2024 beside the 88.57% South Africa reported.
- 2026-11-17 · South Africa: IARC estimates and the registry's count: The comparison of the registry's laboratory count with IARC's estimates said the gap was largest for lung and prostate cancer because those cancers are often diagnosed by scan or examination alone; no source read gives a reason cancer by cancer. (Second accuracy pass; the National Cancer Registry's 2024 tables, its Ekurhuleni 2024 report and IARC's fact sheet read as raw text.) · The note and the table note now give the figures and say only what the registry's reports say: a laboratory count leaves out cancers diagnosed clinically, radiologically or at death, and IARC's figures are modelled estimates.
- 2026-11-17 · Ruffieux and colleagues, age and cancer incidence in people with HIV: A caveat on the paper's record said the data ended before universal HIV treatment was fully rolled out, which the abstract does not say. (Second accuracy pass; the abstract read through Europe PMC.) · The caveat now says only that the data end in 2014.
- 2026-11-17 · Algeria: lede, registry table and registry network record: The page said Algeria counts cancer through a registry in each wilaya and labelled the national totals as covering all wilayas. The registry report says only that wilaya registries were created in 2015 and grouped into three regional networks, and a press report of the network's October 2026 meeting has the minister naming the extension of coverage to the whole country as a priority still ahead. (Second accuracy pass; the INSP Registre des Tumeurs d'Alger 2022 (PDF text, Internet Archive copy) and the Algérie360 report of 30 October 2026 read in raw text.) · The lede, the two table rows and the record now say wilaya registries grouped into three regional networks, and the table note reports the minister's stated priority. The statement that deaths are not part of the count now says only that the reports read give new cases, not deaths.
- 2026-11-17 · Algeria: Specialised anti-cancer centres: The card said a patients' federation counted 22 anti-cancer centres in 2026. The figure is in the radio reporter's own narration, not in the federation's words. The card also gave only a minister's 2025 count of linear accelerators. (Second accuracy pass; the Radio Algérienne report of 5 May 2026 read in raw text, and the IAEA DIRAC public feed read as raw JSON on 16 November 2026.) · The 22 is attributed to the Radio Algérienne report. The card now also gives what the IAEA directory lists (23 centres, 64 linear accelerators, no cobalt-60 units, nine centres with brachytherapy, all entries last updated in 2026) beside the minister's figure, each with its date.
- 2026-11-17 · Algeria: Free care in public hospitals: The card said free care covers diagnosis, surgery, chemotherapy and radiotherapy. Article 13 of the 2018 health law speaks of the means of diagnosis, treatment and hospitalisation in all public health structures and lists no treatments. (Second accuracy pass; Law 18-11 in the Journal officiel No. 46 of 2018 read as PDF text.) · The sentence now says the law does not list treatments one by one.
- 2026-11-17 · Algeria: A decision within 150 days: The card said exceptions to the country-of-origin rule were set by a 2021 order. The decree says only that rules for products registered but not marketed in their country of origin are left to a ministerial order; it gives no year and no such order was read. (Second accuracy pass; executive decree 20-325 in the Journal officiel No. 69 of 2020 read as PDF text.) · Reworded to what article 19 says.
- 2026-11-17 · Algeria: Breast screening is offered: The card said colorectal screening was piloted from 2017 and stated as its own finding that no national programme exists. The source says the pilots were run in 2017, and the absence of an organised programme is that report's statement. (Second accuracy pass; the IHE cancer dashboard for Algeria (commissioned by MSD) read as PDF text.) · Changed to in 2017, and the absence of an organised national programme is attributed to the report.
- 2026-11-17 · Sweden: What OnCo could not read or verify: The card said Cancerfonden's annual research spending was not found on its site and that the figure in OnCo's country table was not verified. The annual report answers the question. (Second accuracy pass; Cancerfonden's annual report for 2024 read as raw PDF text, which gives 1,002 million kronor for research funding in 2024 and 966 million in 2023.) · The gap statement was replaced with the figures from the annual report, which is now linked.
- 2026-11-17 · Sweden: A new national cancer strategy in January 2026: The title and text said the 2026 budget bill added 500 million kronor, giving 1.5 billion kronor for 2026. The source is the government's proposal of October 2025, not a record of money granted. (Second accuracy pass; the government's article of 8 October 2025 read as raw text, which says the government proposes a further 500 million kronor in the budget bill for 2026.) · Title and text now say the sums were proposed in the budget bill.
- 2026-11-17 · Sweden: Standardised care pathways: The page said Sweden sets a maximum wait for every suspected cancer and has a pathway for each common suspected cancer, and that part of the regions' state money depends on meeting the inclusion goal. The sources describe time limits to aim for in 27 pathways, and make the regions' yearly report the condition for the second half of the money. (Second accuracy pass; the regional cancer centres' pathway pages and their 2025 annual report read as raw text.) · The lede, the section title and the card now speak of a time limit in each of 27 pathways, and the funding sentence states only the reporting condition.
- 2026-11-17 · Sweden: What a patient pays: The card and the table note said hospital cancer medicines are bought and paid for by the region directly. The 1177 pages cited do not say this. (Second accuracy pass; the three 1177 pages and the regions' page on ordered introduction read as raw text.) · The sentences now say what the sources say: the cap is described for medicines on prescription, and medicines given by health-care staff go through the regions' joint introduction process. The regions' page is linked.
- 2026-11-17 · Sweden: The NT council: The card said most new cancer medicines are infusions given in hospital, which the regions buy themselves. No source cited says most, or that the regions buy them. (Second accuracy pass; the NT council pages and TLV's page on hospital medicines read as raw text.) · Reworded to the sources: medicines given by health-care staff are called clinic medicines, and TLV notes that more cancer medicines are expected to be approved.
- 2026-11-17 · Sweden: TLV: The card and the gap card said TLV publishes no fixed price limit or threshold. That states an absence as fact on the strength of one page. (Second accuracy pass; TLV's health-economics page read as raw text, which gives no limit and says each case is judged on its own.) · Both now say that TLV's health-economics page gives no limit and that each case is judged on its own.
- 2026-11-17 · Sweden: area differences in cancer: The card title said where a person lives changes which cancer they get and whether it kills them. The fact sheet reports differences between types of area, not a cause. (Second accuracy pass; Socialstyrelsen's fact sheet on new cancer cases in 2024 read as raw PDF text.) · The title now says that which cancers people get, and die of, differs with the kind of area they live in.
- 2026-11-17 · Sweden: survival: The card title said about three in four people are alive five years after diagnosis. The figures are relative survival, which the source defines as not having died of the cancer. (Second accuracy pass; Cancer i siffror 2023 read as raw PDF text.) · The title now says relative five-year survival of about 77 per cent.
- 2026-11-17 · Sweden: What the register misses: The shortfall of 12.3 per cent among people diagnosed at 80 or older, against 2.7 per cent below 80, was given without its period. (Second accuracy pass; Socialstyrelsen's 2024 report on missing cases read as raw PDF text, which gives these figures for 2014 to 2022.) · The period was added, on the card and in the register's record.
- 2026-11-17 · Sweden: approval through the EU: The card said the time between approval and treatment for a Swedish patient is set mainly by the national steps. No source read measures this. (Second accuracy pass; the European Medicines Agency's leaflet on the centralised procedure read as raw PDF text.) · The sentence now only describes which steps are European and which are national.
- 2026-11-17 · Sweden: Prostate, organised testing: The card gave an MRI scan or a risk score as the second test Socialstyrelsen had in mind. Its recommendation names no test. (Second accuracy pass; Socialstyrelsen's prostate screening recommendation read as raw text.) · The examples were removed.
- 2026-11-17 · Sweden: Registers that can be linked: The card said Socialstyrelsen credits the quality registers with improving the cancer register's completeness. The report says only that this must be assumed. (Second accuracy pass; Socialstyrelsen's 2024 report on missing cases read as raw PDF text.) · Reworded to say that Socialstyrelsen says it must be assumed they have improved reporting.
- 2026-11-17 · Sweden: Twenty-one regions deciding together: The card said the regions publish the recommendations so that a patient can see what has been decided and why. The pages cited do not say this. (Second accuracy pass; the NT council's pages read as raw text.) · Replaced with a statement that the regions' site has a page for patients and the public explaining how the council decides.
- 2026-11-17 · Sweden: smaller rewordings: Five phrases went beyond their sources: the regions' own income tax, lung cancer as a men's disease for most of the register's history, GLOBOCAN covering C00-97, a Cancerfonden report dated 14 October 2026, and quality register reports as among the most detailed outcome data in Europe. (Second accuracy pass; the SKR page, Socialstyrelsen's 2024 fact sheet, IARC's fact sheet for Sweden and Cancerfonden's article read as raw text.) · Each was reworded to the source or, for the last, removed.
- 2026-11-17 · Dental and Pharmaceutical Benefits Agency (TLV): The record said TLV publishes no fixed threshold and that hospital drugs are bought by the regions rather than the pharmacy benefit. Neither is stated on the pages cited. (Second accuracy pass; TLV's pages on health economics and on hospital medicines read as raw text.) · Reworded: the health-economics page gives no threshold and judges each case, and TLV assesses selected hospital drugs for the NT council without deciding on benefits.
- 2026-11-17 · NT Council (Rådet för nya terapier): The record said the council covers most cancer infusions and that the regions pay for clinic medicines directly. The council's pages say neither. (Second accuracy pass; the NT council's pages read as raw text.) · Reworded to the council's own description: medicines most often used in hospital, given by health-care staff.
- 2026-11-17 · Switzerland: Why GLOBOCAN's death estimate is higher than the Swiss count: The gap card and the note under the GLOBOCAN table said neither source explained why GLOBOCAN's 19,408 deaths exceed the Swiss count of 17,220, and that GLOBOCAN's method for Switzerland was not described. IARC's fact sheet does give it: national rates from WHO data projected to 2024. (Second accuracy pass; the IARC GLOBOCAN 2024 fact sheet for Switzerland was read as raw text (PDF) on 17 November 2026.) · Both now state the method IARC gives, say that one figure is a projection for 2024 and the other a count for 2022, and say only that neither source compares the two.
- 2026-11-17 · Switzerland: About 50,800 new cancers and 17,200 cancer deaths in 2022: The card said breast cancer is the commonest cause of cancer death in women without a period, and that the Federal Statistical Office's next release was scheduled for 16 November 2026. The release has appeared, and the office now says lung cancer is slightly ahead of breast cancer among women in the latest period. (Second accuracy pass; the office's cancer page (updated 16 November 2026), its key figures workbook and its open data on stats.swiss were read as raw text and data.) · The breast and lung figures are now dated 2018-2022 (about 1,370 against 1,364 deaths a year), the office's newer statement is given, and the card and the checking note say the 1984-2023 release was published on 16 November 2026 and that this page still shows the release before it.
- 2026-11-17 · Switzerland: Why the national figures are still partly an estimate: The card gave about 252 new childhood cancers a year in 2018-2022 as a figure for under-20s. The office's page that carried it has been replaced and the figure could not be re-read; the office's methods note says the childhood figures it receives are for ages 0 to 14. The same sentence was in the Childhood Cancer Registry record. (Second accuracy pass; the Federal Statistical Office's cancer page and its methods note (version of 4 August 2026) were read as raw text; one Wayback Machine lookup for the old page found no copy.) · Both now give the office's current figure, about 247 a year in 2019-2023, with the age range the methods note states.
- 2026-11-17 · Switzerland: What is changing: The card credited percentage changes in lung, bowel, melanoma, breast and prostate cancer rates to a statement by the Federal Statistical Office. The percentages are correct but are arithmetic on the office's published rates; its page gives the trends in words only. The stage IV comparison from the early-onset bowel cancer study was given without its period. (Second accuracy pass; the rates were recomputed from the office's open data (stats.swiss, 1976 European standard), and the study's abstract was read through Europe PMC.) · The figures are now attributed to the open data with their periods and labelled as OnCo's arithmetic, the data source is linked, and the stage comparison is dated 2010 to 2021.
- 2026-11-17 · Switzerland: The commonest cancers, 2018-2022 annual average: The table listed kidney cancer (about 1,170 new cases a year) among the commonest cancers but left out cancers of the mouth and throat, which the same data put slightly higher at about 1,180. (Second accuracy pass; the Federal Statistical Office's open data on stats.swiss were downloaded and ranked.) · The table note now says so.
- 2026-11-17 · Switzerland: What the registry now shows about care: One sentence read as if both the breast and the prostate care-quality reports were published with Swiss Urology in September 2026. The breast summary is dated October 2025. (Second accuracy pass; both NKRS summaries were read as raw text (PDF).) · The two dates are given separately.
- 2026-11-17 · Switzerland: Compared with the EMA and FDA, and why availability is late anyway: The card said a late Swiss approval is mostly a late Swiss application. Swissmedic says only that availability also depends substantially on when companies apply. (Second accuracy pass; Swissmedic's note of 6 July 2026 on the CIRS study was read as raw text.) · The sentence now gives Swissmedic's own reading.
- 2026-11-17 · Switzerland: One national academic trials group: The statement that the renaming from SAKK left management and contracts unchanged was credited to the press release, which does not say it. (Second accuracy pass; the press release of 1 July 2025 and the organisation's questions and answers on the renaming were read as raw text (PDF).) · The statement is now attributed to the questions and answers, which say it, and that document is linked.
- 2026-11-17 · Switzerland: Free advice by phone, chat or WhatsApp: The card said Swiss Cancer Research's projects are chosen by a scientific committee and that the Swiss Cancer League runs smoking-cessation support. The foundation's board decides on the committee's recommendation, and the cited League page does not mention smoking cessation. The same wording was in the two organisations' records. (Second accuracy pass; the about pages of Krebsforschung Schweiz and Krebsliga Schweiz and the League's KrebsInfo page were read as raw text.) · The card and records now say the board decides on the committee's recommendation and mention prevention only; the page that states the advice service is free is linked.
- 2026-11-17 · Malaysia: Rates differ by ethnic group, and the registry reports every cancer by group: The card said nasopharyngeal cancer was most common among Other Bumiputera, and gave a doubling of liver cancer cases without naming the right periods. The registry report gives the highest nasopharyngeal risk to Chinese and Other Bumiputera men alike (1 in 100), and compares 8,256 liver cancers in 2017 to 2021 with 4,033 in 2012 to 2016. (Second accuracy pass; the Malaysia National Cancer Registry Report 2017-2021 read in raw text (sections 5.6 and 5.7, Table 32).) · Both statements reworded to what the report says; the same nasopharyngeal wording corrected in the card on the nationwide registry.
- 2026-11-17 · Malaysia: Measured: a median of 515 days in 2017 and 399 in 2019, for all new medicines: The card treated the 2020 study's separate medians as parts of one total: 135 days of queuing "of" the 515, and 166 and 131 days "of" the 279-day assessment. The abstract reports each as its own median, and 166 and 131 do not add up to 279. (Second accuracy pass; the abstract of Sani et al. 2020 read through Europe PMC (PMID 32865804).) · The figures are now given as separate medians, here and in the paper's own record.
- 2026-11-17 · Malaysia: trials in OnCo's records with a site in Malaysia: The page gave 171 trials with a Malaysian site, 13 of them phase 2/3 and 63 recruiting. OnCo's records on 17 November 2026 hold 170, with 12 phase 2/3 and 62 recruiting. (Second accuracy pass; the count re-run against the trial records in the repository.) · Count, breakdown and count date corrected in the title, description, card and the Clinical Research Malaysia record.
- 2026-11-17 · Malaysia: Breast, bowel, cervical and oral cancer screening, and no national lung programme in the 2021 plan: The card stated in the present tense that Malaysia has no national lung screening programme. The source is the cancer plan published in 2021, which says none had yet been established. (Second accuracy pass; the National Strategic Plan for Cancer Control Programme 2021-2025 read in raw text (section 10.2).) · The statement is now dated and attributed to the plan, in the card title, the card, the MaHTAS card and the MaHTAS record.
- 2026-11-17 · Malaysia: Six Ministry of Health oncology centres, with more planned: The plain summary gave a newspaper's account of a written parliamentary reply as a statement by the ministry, and the report can no longer be read in full: all but its two opening paragraphs are behind the newspaper's paywall. (Second accuracy pass; The Star's page of 6 November 2024 fetched as raw HTML.) · The summary now attributes the plan to the newspaper's report, and the card says the details could not be re-read on 17 November 2026.
- 2026-11-17 · Malaysia: RM1 at a clinic, nothing for a first specialist visit on referral, and a cap of RM500 a stay: The card said a first specialist visit is free on referral from a government medical officer or specialist. The schedule names only referral from a government medical officer. (Second accuracy pass; the Ministry of Health schedule of charges under the Fees (Medical) Order 1982 read in raw text (Jadual A, Bahagian 2).) · The words "or specialist" removed.
- 2026-11-17 · Malaysia: 245 working days for a new medicine, 90 when NPRA relies on a reference regulator, 60 in a 2026 pilot: The card said the 245-working-day timeline includes consultation with local clinical specialists. The guidance lists the consultation as a requirement for a new chemical entity and gives the timeline separately. (Second accuracy pass; NPRA's Drug Registration Guidance Document, Appendix 3 (July 2026), read in raw text.) · Reworded to say the consultation is required.
- 2026-11-17 · Malaysia: record of Mohd Sani et al. 2025 on NPRA approval times: The record said the study predates reliance routes introduced from 2024 and called it the only measured approval time found. NPRA's reliance pathway began in March 2019 and was revised from 1 January 2024, and the 2020 baseline study is also a measured time. (Second accuracy pass; NPRA's Directive No. 13 of 2023 read in raw text, and both abstracts read through Europe PMC.) · Both statements reworded.
- 2026-11-17 · Israel: What a cancer patient pays turns on a code: The card said travel to treatment was refunded on the same terms at Maccabi and Leumit. The terms it gave were Maccabi's alone. Leumit's schedule has no charge for public transport, a 50 percent patient share for an ambulance or taxi approved in advance by the fund, and a litre of fuel per 10 kilometres by car. (Second accuracy pass; Maccabi's co-payment PDF (October 2026) and Leumit's co-payment schedule (current at 1 July 2026) read in raw text.) · Each fund's terms are now given separately, and the plain text says the code pays for or towards the journey.
- 2026-11-17 · Israel: What we could not establish: The card listed an official HPV vaccination coverage figure and national lung cancer screening as things that could not be sourced. Both are answered by sources within reach: WHO's immunisation data hold official final-dose coverage for Israel (53.92 percent of girls and 49.07 percent of boys for 2022), and the ynet report already cited says lung screening was in the basket for smokers aged 65 to 74 and was extended in 2026 to those aged 60 to 64. (Second accuracy pass; WHO immunisation API queried for every HPV antigen row for Israel and read as JSON, the Chodick and Strassberg paper read in full text, and the ynet article read in raw HTML.) · The two items are removed from the list of gaps and the card now states what each source says, with the lung screening detail marked as a press report.
- 2026-11-17 · Israel: Outside the top ten on papers per head, fourth in the world on early-phase cancer trials per head: The card called Israel mid-pack and unremarkable on oncology papers per head, and the page text said eleventh, while the panel the page computes from the corpus printed twelfth of 119 countries. (Second accuracy pass; rank recomputed from public/openalex/countries.json and the population rows in the worktree, and the ClinicalTrials.gov counts re-run through the API.) · Title, plain text and detail now say outside the top ten; the page lede and panel text no longer state a fixed rank that the computed figure contradicts.
- 2026-11-17 · Israel: Cancer takes the largest share, and the share swings hard from year to year: The 2025 and 2026 oncology shares of the basket supplement, and the Minister's account of the 2026 budget in the card beside it, were written as official facts. They come from Calcalist and ynet reports of the committee's announcements. (Second accuracy pass; the archived Calcalist article and the ynet article read in raw HTML.) · Both cards, the page lede and the note on how the page is built now attribute these figures to the press reports.
- 2026-11-17 · Israel: What happens to a drug that does not get in: The card said life-extending medicines were removed from supplementary plans in three tranches up to 2010 in exchange for basket additions. None of the linked sources says this. (Second accuracy pass; section 10 of the National Health Insurance Law (Hebrew consolidated text), the State Comptroller's May 2022 report, the Adva Center paper and the Israel Medical Association paper read in raw text.) · The clause is removed; the statutory ban in section 10(b)(4) stays.
- 2026-11-17 · Israel: A national founder-variant test with no family history needed, which Israeli clinicians describe as pioneering: The title and the page lede called the programme the first in any country to test a whole population. The source says Israel pioneered a national programme, and the test is for women of Ashkenazi origin. The card also described a Clalit reminder-message experiment that none of its sources reports, and gave pancreatic surveillance from 50 for BRCA2 carriers as settled where the protocol says to consider it. (Second accuracy pass; Ministry circular 3/2020 (Hebrew PDF) and the full texts of Armon 2026, Michaelson-Cohen 2026 and Greenberg 2023 read through Europe PMC.) · Title and lede reworded to the source's claim, the experiment sentence removed, and the pancreatic wording corrected.
- 2026-11-17 · Israel: What a founder variant is, and what it is not: The card said the BRCA1 founder variant had been found on the Ashkenazi haplotype in families with no reported Jewish ancestry, and that BRCA1 c.5266dupC is a recurrent pan-Slavic allele. Neither statement is in the linked papers. (Second accuracy pass; Laitman 2013 read in full text, and the abstracts of Neuhausen 1996, Bar-Sade 1998 and Ahmad 2023 read through Europe PMC.) · Replaced with what Laitman and colleagues report: independent origin at least twice in non-Jewish families and spread of the founder haplotype into a Hispanic population about 650 years ago. The second sentence is removed.
- 2026-11-17 · Israel: A small professional society, and a charity that started the screening: The title and plain text gave the oncology society about a hundred members from an undated union page, while a 2021 survey cited elsewhere on the page contacted about 250 oncologist members. (Second accuracy pass; the Israel Medical Association union page read in raw HTML and Bashkin 2021 (Current Oncology) read in full text.) · Both figures are now given with their sources and the title no longer states a number.
- 2026-11-17 · Israel: Four health funds, and a bill the state pays more of than the health tax does: The card said a fund that treats more cancer than its capitation share carries the difference, which is an inference no source states, and cited no source for the five illnesses paid by patient numbers. (Second accuracy pass; the State Comptroller's May 2022 report and the National Insurance Institute's 2024 collection chapter read in raw text.) · The five illnesses are now named from the Comptroller's report, which is linked, and the sentence says only that cancer money is divided by the formula.
- 2026-11-17 · Israel: Screening programmes a charity started, and the gaps they have not closed: The card said the colonoscopy follow-up indicator counts colonoscopy within 180 days; the cited slides give no time window. The plain text of the card beside it said bowel cancer arrives about eight years earlier in Arab patients, where the registry gives six years for men and ten for women. (Second accuracy pass; the quality indicators slide file unpacked and its chart data read, and the registry's colorectal report (Hebrew PDF) read in raw text.) · The time window is removed and the two ages at diagnosis are given as the registry states them.
- 2026-11-17 · Israel: Two hospitals that make their own cell therapies: The card said Michel Sadelain wrote the editorial on the Hadassah CAR-T paper and gave no source. The editorial has two authors. It also described Doxil in the card beside it as the first nano-drug approved anywhere, where the source says the first approved by the FDA. (Second accuracy pass; the editorial resolved through Europe PMC, and the Barenholz 2012 abstract read.) · Now says co-wrote, with the editorial linked; Doxil is described as the source describes it.
- 2026-11-17 · Taiwan: Figures looked for and not found: The page said no population-based cancer survival figure for Taiwan could be found and that the registry report gives none. The report points readers to the Taiwan Cancer Registry Center, which publishes five-year survival. (Second accuracy pass; the Cancer Registry Annual Report 2023 and the Center's survival table for 2019 to 2023 diagnoses, both read as raw PDF text.) · The gap now gives five-year relative survival for cancers diagnosed from 2019 to 2023 (62.3% for all invasive cancers, with five named sites) and the stage-specific liver cancer figures in the Ministry's hepatitis plan, each with its source.
- 2026-11-17 · Taiwan: New invasive cancers and cancer deaths in Taiwan, 2023, by site: The table left the incidence rates for pancreatic and oesophageal cancer empty and said the report gives both-sex rates only for its ten leading sites. The report's table of incidence by site gives both. (Second accuracy pass; the Cancer Registry Annual Report 2023 read as raw PDF text, table of incidence by site.) · The two rates are filled in (7.53 and 7.18 per 100,000) and the note now says where they come from and why the thyroid death rate is still empty.
- 2026-11-17 · Taiwan: 810 trials in OnCo's records have a site in Taiwan, and 36 are run in Taiwan alone: The card gave 812 trials. That is the number of ClinicalTrials.gov registrations; two OnCo records each hold two, so 810 trial records list Taiwan, which is what the card's own phase figures add up to. The list of commonest cancers also skipped the second largest group. (Second accuracy pass; recount of the ClinicalTrials.gov participation data in OnCo's records on 17 November 2026.) · The count is 810 in the title and text, with the 812 registrations explained, and the 65 trials filed under advanced cancer of any type are now named.
- 2026-11-17 · Taiwan: Since 2025, a separate fund pays for new cancer medicines while their value is reassessed: The card called the immunotherapy decision the fund's first big use, gave atezolizumab a first-line lung cancer indication, and called six-month listing a target. The releases give pembrolizumab as an example, do not state atezolizumab's new indication, and give six months as an estimate. (Second accuracy pass; the National Health Insurance Administration's releases of 4 December 2023, 24 April 2025 and 28 May 2025, read as raw HTML.) · The three statements now say only what the releases say.
- 2026-11-17 · Taiwan: A licence is the first step; insurance listing is the second: The card said that for most patients a cancer medicine becomes usable when the insurance lists it, which no source read states, and called six-month listing the insurer's target. (Second accuracy pass; the National Health Insurance Act and the insurer's release of 4 December 2023, read as raw HTML.) · The opening now says that a licence does not by itself mean the insurance pays, and the six months is given as the insurer's estimate when it announced the route.
- 2026-11-17 · Taiwan: IARC's GLOBOCAN has no separate figure for Taiwan, so this page uses Taiwan's own registry: The page description promised to say why IARC gives no figure for Taiwan, and the card said the 1976 standard is the one older IARC tables use. Neither the page's sources nor IARC say either thing. (Second accuracy pass; IARC's list of populations for GLOBOCAN 2024 read as raw JSON, and the Cancer Registry Annual Report 2023 as raw PDF text.) · The description now says only that IARC's GLOBOCAN has no separate entry for Taiwan, and the clause about older IARC tables is removed.
- 2026-11-17 · Taiwan: People screened in 2023 under the HPA's programmes: The table note said a positive smear is LSIL or worse. The dashboard's note defines its positive rate that way, but its chart of positive results also includes ASCUS, so the count's definition is not certain. (Second accuracy pass; the Health Promotion Administration's 2023 screening dashboard read as raw PDF text.) · The note now reports both statements and says the definition behind the count of 46,642 is not certain.
- 2026-11-17 · Taiwan: Hepatitis B vaccination of every newborn since 1986, to prevent liver cancer: The card called the hepatitis and liver cancer plan for 2026 to 2030 the Health Promotion Administration's. Its cover names the Ministry of Health and Welfare, October 2025. (Second accuracy pass; the plan read as raw PDF text.) · The plan is now attributed to the Ministry, as published by the administration.
- 2026-11-17 · Taiwan: A cancer law that puts the Premier in charge and names a research institute: The duty to consult patients and families was given in a sentence about Article 8 of the Cancer Control Act. It is in Article 7. (Second accuracy pass; the Cancer Control Act on the Ministry of Justice's law database, read as raw HTML.) · The sentence now cites Article 7 and gives its wording: other experts, industry, and representatives of cancer patients and their families.
- 2026-11-12 · Vein-to-vein time: The summary gave unscoped regional turnaround averages and attributed cross-trial outcome differences to waiting time, without the population and comparison needed for those claims (Raw FDA January 2024 CAR T-cell guidance, sections VI.B.4 and VI.B.5, plus its distinct manufacturing, eligibility and failure-to-treat considerations) · Defines the interval and its stages, separates product availability from clinical readiness, and retains participant-flow examples with their actual denominators
- 2026-11-12 · EGFR: A retained lung note assigned sensitivity or resistance to every inhibitor generation by mutation class, including resistance of exon 20 insertions to all inhibitors; its testing sentence did not distinguish protein from DNA (Lynch 2004 and the raw FDA July 2025 sunvozertinib approval notification) · Keeps the cohort frequencies, replaces the universal drug-response claims with variant-specific evidence and a dated, population-qualified accelerated-approval example, and distinguishes immunostaining from identification of a DNA alteration
- 2026-11-12 · BRD4: The account referred to an unavailable lesson figure and could make a fixed drawn acetyl mark sound like a measured unchanged state (Fresh views of both pinned figure themes and resolved JQ1 and Shu primary evidence) · Explains pocket occupation, protein depletion and the fixed-mark drawing convention in self-contained prose
- 2026-11-12 · Other-cause mortality: The introduction inferred dependence from mutually exclusive events and conflated cumulative-incidence estimation with Fine-Gray regression (Austin's method definitions and the Therneau, Crowson and Atkinson vignette) · Separates Aalen-Johansen probabilities, cause-specific rates and subdistribution hazards; replaces the single-cause explanation of disagreement between prostate trials without recertifying the retained prostate-cohort figures
- 2026-11-12 · Bayesian design and response-adaptive randomisation: Leads implied a model-free probability that a treatment works and guaranteed efficiency or better assignments (Original I-SPY2 methods, complete primary abstracts, FDA's 2019 final guidance and January 2026 draft) · States model, prior, endpoint, future-trial success criterion and operating-characteristic conditions; distinguishes predictive graduation, exploratory survival and a person's response-guided treatment path
- 2026-11-12 · Non-inferiority and margins: Broad claims obscured margin justification, interval level and analysis-population limits (FDA's 2016 guidance and named Z0011, PERSEPHONE and EA1131 reports) · Keeps M1, M2, clinical scope and confidence level explicit; distinguishes non-inferiority, equivalence, superiority and an inconclusive result
- 2026-11-12 · Power and sample size: Underpowering was called the commonest reason useful treatments fail and blinded re-estimation was said to have no statistical cost (ICH E9, adaptive-design guidance, CALOR methods and power-method sources) · Replaces universal claims with specified effects, event information, planning assumptions and analysis requirements; independently checks the complete fictional calculations
- 2026-11-12 · Estimands, master protocols and seamless designs: Overviews conflated analysis sets with target effects and implied general patient-number or speed advantages (Targeted ICH E9(R1) and FDA guidance passages) · Separates estimand, estimator and sensitivity analysis; makes efficiency conditional. Unverified clinical catalogues are withdrawn from these overviews with complete before/after fields retained in the packet and existing links preserved
- 2026-11-12 · Breast follow-up: The record said no later ASCO guideline was found and presented the 2013 calendar as current (Resolved 2026 ASCO abstract and official ASCO/Guideline Central practice summary) · Adds risk-based intensity and the age, risk, anatomy and time conditions for mammography; preserves older quotations explicitly as history and names the original-journal access limit
- 2026-11-12 · MRI, ultrasound and mammography: Broad leads and brief summaries did not carry the measured quantity, complete examples or detection limits (Primary MRI abstracts and targeted methods, raw official modality descriptions and complete source figures) · Separates contrast, morphology, length, selected-voxel volume, acoustic images and X-ray reconstructions from viable-cell amount and sampled tissue; withdraws unsupported general rankings, screening percentages and cost/deployment catalogue claims from these leads
- 2026-11-12 · PET, FDG PET and PET/CT: Signal was described as tumour activity, localisation as exact and availability as universal (Technical EANM 2015 passages and official PET descriptions) · Names tracer, background, coverage, time, uptake quantity and partial-volume limits; keeps the inaccessible 2025 revision distinct from the technical version actually read
- 2026-11-12 · Immuno-PET and granzyme B PET: The antibody account included a non-antibody tracer and implied any binder could provide complete target imaging (Resolved primary feasibility studies and targeted CD8/granzyme passages) · Keeps antibody-derived imaging separate from peptide/peptidomimetic imaging; preserves study populations, competition, normal-organ signal and exploratory outcome scope without a scan-guided treatment claim
- 2026-11-12 · Theranostics and dosimetry: The short account lacked the complete product-bridging and unit exercises (Named PSMA guideline abstracts, targeted FDA 2019 nonclinical guidance and the source diagram) · Adds vector, activity, mass, time and dose distinctions; a bright diagnostic scan or shared target does not establish a different therapeutic product's distribution, absorbed dose or clinical benefit
- 2026-11-12 · Cell-free DNA and tumour-informed assays: Leads gave an unqualified tumour-fraction range, product-category claims and an unnecessarily narrow tumour-reference description (Complete primary abstracts, targeted assay passages and FDA's November 2024 drug-development guidance) · States the specimen mixture, reference quality, target list and intended use; keeps monitoring, profiling and screening questions separate without a universal sensitivity ranking
- 2026-11-12 · Residual-disease testing and ctDNA units: The lead and retained catalogue encouraged cross-method comparisons and exhaustive subclone-detection claims without their required assumptions (Specimen and assay comparisons, analytical-limit definitions and the complete fictional sampling exercises) · Keeps quantities, calibration, input and calling rules together; separates pooled signal detection from validated clone measurement and below-range detection from an exact fold change
- 2026-11-12 · Blood and imaging discordance: The short account did not carry the complete measurement comparison or worked residual-tissue course (Primary MRI methods, raw modality definitions and NRG-BR005's actual complete or near-complete imaging eligibility) · Explains regional contrast-dependent features, finite blood sampling, all four interpretation patterns and why favourable measurements do not establish tissue pCR or a useful treatment change
- 2026-11-12 · Clonal haematopoiesis: The plasma-origin sentence treated a method category as sufficient to resolve false positives (Shah's named tumour cohort and the Bolton and Razavi studies at their recorded reading depth) · Requires origin assessment within tumour and blood-cell coverage and sensitivity; preserves ambiguous assignments and distinguishes compartments from a gene name alone
- 2026-11-12 · Magbanua's I-SPY2 ctDNA analysis: The account blurred the 84-person cohort with the 60-person joint post-treatment analysis and implied equivalence from similar observed outcomes (Primary abstract and targeted methods and analysis passages) · Retains 291 available plasma samples, the 17 pCR and 43 non-pCR groups, both hazard-ratio intervals and the absence of a randomised blood-guided strategy; the exact denominator behind the reported final-draw 9% remains an explicit audit gap
- 2026-11-12 · Lines of therapy and neoadjuvant/adjuvant: The leads restricted first-line to advanced cancer and made timing labels a simple before/after rule; retained prose generalised treatment ranking and line counting (Raw NCI definitions, the KEYNOTE-355 exposure variables and comparison of the complete treatment sequence) · States the disease setting and counting convention, separates treatment role from calendar position and withdraws unsupported universal ranking rules
- 2026-11-12 · De-escalation and chemotherapy regimen: The account conflated a response-guided research rule with proven safe omission, and described KEYNOTE-522 as residual-only escalation (I-SPY2.2 abstract, KEYNOTE-522 reports and label, and the complete fictional rule and calendar exercises) · Makes the input, threshold, action, outcome and safety boundary explicit; preserves the whole perioperative assignment and the distinct jobs, doses, intervals and support in a regimen
- 2026-11-12 · Residual TNBC, c-TRAK TN and BRE12-158: A ctDNA percentage used the wrong denominator; the c-TRAK lead inferred that detection was too late to prevent relapse, and residual burden sounded sufficient to choose a medicine (Radovich's 142-person ctDNA cohort, Turner's detection/staging/treatment groups and the randomised BRE12-158 report) · Keeps assay populations, imaging findings and the small treated subset separate; distinguishes prognostic association from useful treatment and retains eligibility, local-care and delivery questions
- 2026-11-12 · EA1131 and KEYNOTE-355: EA1131 was described as unindexed, its result became a universal treatment rule, and KEYNOTE-355's registry total was explained using an incorrect 32-person run-in (Live publication identities, primary abstracts and registry participant-flow tables) · Separates EA1131's 410 randomised participants from 415 registry enrolments; KEYNOTE-355 lists 35 part-1 starters and 847 part-2 randomisations, matching 882 enrolments; neither trial establishes a universal rule outside its question
- 2026-11-12 · OlympiA, CREATE-X and neoadjuvant versus adjuvant breast treatment: Leads overgeneralised eligibility, subgroup ranking or timing equivalence; the B-51 account turned a nonsignificant comparison into proof of no benefit (Primary trial abstracts, targeted OlympiA eligibility text, the historical chemotherapy meta-analysis and B-51 report) · Retains studied populations, confidence intervals and follow-up; separates historical same-chemotherapy timing from modern strategies and local-treatment questions
- 2026-11-12 · pCR, perioperative immune window, ASCENT-05 and TROPION-Breast03: The retained CTNeoBC associations lacked intervals, the review timeframe was omitted and two registry update dates were stale (Resolved CTNeoBC and Horowitz abstracts and current registry records) · Adds the reported intervals and review-specific days-to-weeks framing; updates registry dates without presenting unposted results as treatment evidence
- 2026-11-12 · Intention to treat: The account overstated real-world representativeness, defined modified and per-protocol sets too narrowly, treated evaluable analyses as necessarily inflated and asserted a site-wide default without verification (ICH E9 analysis-set and safety sections and the E9(R1) estimand framework) · Preserves assignment while qualifying exclusions, missing outcomes, design and sensitivity analysis; withdraws the unsupported site-wide default and uses an explicit fictional denominator example
- 2026-11-12 · Endpoint: Meeting a primary endpoint was defined only as beating a comparator, and secondary outcomes were categorically described as not powered (ICH trial-design guidance, CONSORT reporting and FDA's December 2018 endpoint guidance) · Refers to the actual planned question and testing rule; separates event definitions, clocks, composite components and context-dependent surrogate evidence
- 2026-11-12 · Data maturity: Unreached medians were conflated with unreliable treatment comparisons, censoring was reduced to event-free cutoff observations and regulatory follow-up requirements were generalised (Primary KEYNOTE-522 abstracts, registry analyses, NIST methods and ICH guidance) · Distinguishes median outcome, follow-up, event information and precision; removes the universal regulatory assertion and retains the named trial's planned analyses
- 2026-11-12 · KEYNOTE-522: The lead made an unsupported survival superlative; response cohorts and dates were compressed, and exploratory residual-disease strata were described without their uncertainty (Primary reports, the label, sponsor-posted registry and the authors' 2026 conference abstract; Pusztai's 2024 exploratory abstract) · States the whole assigned strategy, separates first-602, first-1,002 and all-1,174 response analyses, keeps each survival cutoff and preserves the seven-year figures with a primary conference citation; qualifies post-treatment subgroup estimates
- 2026-11-12 · OptimICE-pCR: The summary described a further year of postoperative immunotherapy and implied that pCR meant all cancer had disappeared (The current registry specifies the postoperative randomisation, response requirement and 27-week comparison) · States the actual continuation-versus-observation question, non-inferiority endpoint, duration and recruiting state without presenting safe omission as established
- 2026-11-12 · Trial protocol: A categorical sentence treated lack of a traceable prespecified analysis as enough to classify every result as merely a hypothesis (ICH and CONSORT distinguish evidence role, planning, relevant outcome access and reporting) · Requests the dated plan and access history, distinguishing confirmatory questions, planned exploration and later discoveries rather than classifying a result from its p-value alone
- 2026-11-12 · Analytical and clinical validation: The lead limited clinical validity to predicting a patient outcome, while the summary implied one universal validation ladder (MedlinePlus Genetics separates detecting a variant, its relationship to disease presence or risk, and useful clinical information) · States three intended-use questions and distinguishes reliable measurement, clinical meaning and consequences of a decision
- 2026-11-12 · External validation: The lead defined external validation through a different hospital, country or platform and called it the only evidence that a result travels; the summary favoured other investigators/platforms and dismissed internal evidence as no result (TRIPOD-Cluster explanation, Box 1, distinguishes apparent, internal and external evaluation and includes later-period data from the same investigators) · States the evaluation question and its limits, preserves the value of internal resampling and removes a universal pass/fail implication
- 2026-11-12 · External validation, retained benchmark assertions: Routine TCGA-to-METABRIC testing and a typical concordance-index decline were asserted without primary verification here (Source and field comparison) · Intentionally withdraws those generalisations rather than calling them false; retains the separately sourced Michiels seven-study result at its historical scope
- 2026-11-12 · mRNA manufacturing: The lead, process description and cards generalised modified nucleotides, a cell-free whole chain, production in weeks, process invariance and fixed storage temperatures (TNBC-MERIT primary methods specify ordinary uridine and report a 34 to 125 day sample-to-release range; Rojas reports another uridine lipoplex, with product-specific formulation and protocol) · Replaces generalisations with a specified product, manufacturing clock and separate quality/handling evidence; preserves the earlier scoped mechanism paragraph and source links
- 2026-11-12 · mRNA manufacturing, retained commercial narrative: Facility locations, supply/patent rankings and an industrialisation prediction had no primary verification in this assessment (Field-by-field comparison with the cited study and manufacturing evidence) · Deliberately withdraws these assertions pending named source follow-up, without calling them false; the lipid-nanoparticle owner retains formulation teaching
- 2026-11-11 · Agretopicity: The lead and opening defined agretopicity itself as the mutant-versus-normal comparison (Duan's original definition and Ghorani's distinct predicted-affinity calculation, checked in their primary text) · Distinguishes peptide-MHC interaction from the differential index, retains both formula conventions and the scoped model/cohort findings, and clarifies Simple English
- 2026-11-11 · Neoantigen: Novel tumour sequence was said to guarantee an immune target that could be attacked without harming normal cells (TESLA primary prediction/validation methods and the separate presentation, recognition and safety questions) · Removes the blanket safety claim, adds a candidate-evidence framework, withdraws eight obsolete unreviewed translations and corrects Simple English
- 2026-11-11 · Cross-presentation: The opening confused tumour-cell class-I display with naive CD8 priming and described a single intracellular route as universal (Cross-presentation review abstract, RNA-delivery study and T-cell activation teaching) · Separates display from priming and retains the previously scoped model findings and direct-presentation explanation
- 2026-11-11 · Exome and genome sequencing: The lead implied perfect coverage of all genes or all DNA, with a universal depth/pairing rule and unqualified bias/depth comparisons (MedlinePlus definitions and Jones's targeted/exome methods) · Names the intended territory and workflow limits, removes universal depth/pairing claims, withdraws eight obsolete unreviewed translations and corrects Simple English
- 2026-11-11 · Personalised mRNA vaccines: Programme phases and ongoing status were collapsed; low toxicity, target count and manufacturing time were stated as platform-wide properties (Company announcement, named ClinicalTrials.gov records and Rojas's phase 1 report) · Preserves the August 2026 phase 3 report with company attribution, corrects individual programme stages/status, and makes antigen counts, combinations, toxicity and timing product-specific
- 2026-11-11 · BRCA reversion mutations: The uncited lead and summary supplied pooled frequencies, guaranteed cross-drug non-response and treatment-switch instructions, including a negative-reversion rechallenge rule (Resolved Sakai 2008 and 2009 abstracts and targeted 2009 cell-line, protein-restoration and siRNA passages) · States the possible mechanism at its experimental scope, separates sequence prediction from function and clinical response, and withdraws one stale unreviewed Chinese summary
- 2026-11-11 · DNA and genetic code: The genetic-code alias led to the DNA molecule rather than the rules for reading codons (Canonical-owner census, official code tables and NHGRI definitions) · Moves the alias to a dedicated code/reading-frame owner and links the existing molecule and variant-class records; their teaching fields remain unchanged
- 2026-11-11 · Grade, stage and receptors: The lead promised receptor-defined drug response; the summary made grade and stage determine treatment intensity and imposed a fixed reporting order ending in surgery (NCI pathology/staging text, resolved ER/PR and HER2 guideline abstracts, and targeted CAP/Ki67/TIL passages) · Separates measured features, staging contexts, treatment evidence and individual benefit; replaces the unreviewed old summary and retains its source links
- 2026-11-11 · Biopsy: The generic lead called biopsy the only way to confirm any cancer diagnosis (NCI's specimen-specific pathology explanation and comparison of current summary consumers) · Defines collection and specimen limits, corrects Simple English and withdraws eight stale unreviewed translations
- 2026-11-11 · Olaparib: Notes made germline BRCA alone sound sufficient and treated an 82% TNBC trial population as an effectively TNBC-only indication (Raw Lynparza sections 1.4/14.4 and targeted OlympiA eligibility/discussion text) · Retains high-risk, HER2-negative, prior-treatment and qualifying-variant criteria; includes the hormone receptor-positive population and distinguishes trial criteria from the label and individual suitability
- 2026-11-11 · Placebo: The lead required every cancer-trial placebo to accompany standard treatment, while the summary ruled out blinding whenever adverse effects reveal treatment (Complete text of FDA's August 2019 cancer-trial placebo and blinding guidance, checked with NCI's trial explanation) · Retains selected justified placebo-only settings and standard-care add-on designs; describes possible loss of blinding and qualified unblinding recommendations; corrects Simple and withdraws eight stale unreviewed machine summaries
- 2026-11-11 · Expanded access: A retained note made prospective IRB review an absolute rule for every subsequent institutional use (FDA's April 2026 expanded-access guidance, Q11) · States the normal prospective-review rule and FDA's qualified discretion for subsequent emergency use when prior review is infeasible, with the five-working-day reporting requirement
- 2026-11-11 · Grade: Generic leads equated grades with fixed growth speeds and generalised chemotherapy response (NCI breast grading/staging and the source-first introductory assessment) · Defines cancer-specific microscopic grading separately from growth rate, extent and adverse-event severity
- 2026-11-11 · Stage: The lead ranked stage as the single strongest guide and the summary imposed broad treatment and state-change rules (Raw NCI staging explanation and current TNM owner comparison) · Keeps system, edition, evidence and treatment moment; distinguishes anatomical and prognostic groups without a universal treatment rule
- 2026-11-11 · Residual TNBC: The opening made residual burden decide the medicine; exploratory RCB subgroup effects lacked their uncertainty (Resolved CREATE-X, OlympiA and KEYNOTE-522 RCB abstracts) · Separates eligibility and prior care from score alone and retains the subgroup confidence intervals
- 2026-11-11 · PD-1 and pembrolizumab: Short and mechanism wording implied automatic restored function, killing and persistent memory after blockade (Raw KEYTRUDA mechanism/warning sections and the clinical-comparison evidence) · States receptor/ligand blockade with separate recognition, access, function, benefit and harm questions; withdraws stale short translations
- 2026-11-11 · Nucleus: The short definition required the nucleus to be at a cell's centre (Source/field comparison and the raw NCI definition) · Defines the membrane-enclosed chromosome compartment without requiring a central location; corrects simple English and withdraws stale short translations
- 2026-11-11 · Remission: The summary generalised blood-cancer response criteria and a five-year cure proxy across cancers (Source/field comparison, the NCI definition and NCI prognosis explanation) · Defines partial and complete remission as observed changes without a universal cure timetable or eradication guarantee
- 2026-11-11 · ER, PR, hormone receptor status and TNBC: Leads equated ER-low with negativity or inferred treatment eligibility, benefit or personal risk directly from staining (Coordinator audit; CAP/ASCO guidance, RCPath passages and the source receipts) · Keeps the validated categories, separate ER/PR observations, low-positive caveats and prognosis-versus-benefit distinction
- 2026-11-11 · HER2-positive: A dual-probe group was described as uniformly negative under US rules; test status, RNA subtype and treatment eligibility were blurred (Raw CAP testing rules and RCPath dataset passages) · Retains concurrent IHC and indicated recounting, the attributed UK comparison and the separate medicine-evidence question; removes an unsupported treatment ranking
- 2026-11-11 · Germline testing, gBRCA and germline BRCA readout: Origin, family risk, tumour repair function and medicine access were conflated, with universal inheritance or sensitivity wording (NCI genetic-testing explanations, classification guidance and the source/field audit) · Separates constitutional origin, family implications, assay coverage, current tumour function and indication-specific evidence
- 2026-11-11 · HRD in breast cancer: Selected-study findings were generalised as prevalence or evidence of added platinum benefit (Resolved studies and the scoped assay/source comparison) · Keeps assay-specific findings and study populations separate from treatment-benefit inference
- 2026-11-11 · Hereditary cancer syndromes: MEN1 and RET-associated MEN2 were conflated, and an MRI-alone causal mortality figure lacked support (NCI MEN table and selected raw explanatory passages) · Separates the syndrome/gene relationships and withdraws the unsupported MRI-alone figure; other catalogue assertions retain their earlier source scope
- 2026-11-11 · Oligometastatic disease: A small detected lesion count implied curability by treating each site (Consensus definitions and the scoped trial evidence) · Separates detected burden, prior history, biological hypothesis and comparative treatment outcomes
- 2026-11-11 · Recurrence: Timing implied deterministic biology and ctDNA monitoring implied a universal lead time (Historical cohort reports, selected methods/results and NCI explanations) · Retains the population, event, clock, follow-up and conditional denominator instead of a personal deadline or universal monitoring interval
- 2026-11-11 · Curative intent: Intent was inferred from treatment intensity, modality or a universal disease-setting rule (NCI prognosis, chemotherapy, stage-based and palliative-care explanations) · Separates the documented goal from observed response, medicine identity and eventual outcome
- 2026-11-11 · Chemotherapy: Leads reduced the family to DNA injury in fast-dividing cells (Raw product mechanisms and NCI definitions/purposes) · Distinguishes DNA, RNA, metabolism and microtubule effects, exposure, recovery and clinical evidence
- 2026-11-11 · Alkylating agents and antimetabolites: Aliases and shorthand conflated fluoropyrimidines, vinca drugs and alkyl transfer (Current labels, terminology and canonical-owner census) · Detailed fluoropyrimidine content stays on its drug owners; vinca aliases move to tubulin; the combined overview names the family boundary
- 2026-11-11 · Capecitabine and fluorouracil: Tablet-form shorthand, dated testing language and an unwarranted CREATE-X germline restriction (Current SPL/FDA text and resolved trial abstract) · Distinguishes prodrug, host handling, complete/partial deficiency and the trial population; CAPOX/FOLFOX components remain separate
- 2026-11-11 · Anthracyclines, taxanes and TOP1 inhibitors: Mechanism shorthand suggested inevitable death, fixed cardiac thresholds or comparative safety (Raw label sections, abstracts and targeted mechanistic passages) · Keeps enzyme intermediates, interphase effects, exposure, source populations and uncertainty explicit
- 2026-11-11 · Antibody tumour penetration: No broad owner carried the transport and measurement teaching (Independent source inventory and named mouse/patient studies) · Registers the parent without duplicating the narrower binding-site-barrier concept; universal distance and mass-ratio claims are not adopted
- 2026-11-16 · Shasqi: Said SQ3370 had completed its first-in-human study and that the company continues clinical development (ClinicalTrials.gov NCT04106492 (terminated, results posted 14 November 2025), the investigators' medRxiv preprint and the company's June 2025 description of itself as preclinical) · The record now says the trial was stopped because its prespecified response criterion was not met, gives the pipeline as the company publishes it, and carries sourced financing and grants
- 2026-11-11 · Immune-stimulating conjugates: Innate activation and delivery were too easy to read as guaranteed priming or benefit (Resolved papers, targeted sensor-pathway passages and eight dated registries) · Separates sensing, uptake, target identity, functional immune evidence and clinical endpoints; [receipt](research/therapy-conjugate-source-followup-2026-11-11/README.md)
- 2026-11-11 · TOP1 payloads: Compressed redesign notes omitted experimental alternatives and the difference between chemistry and cell response (Raw abstracts and named model comparisons) · Retains five design questions, Takegawa alternatives, model scope and clinical-evidence boundaries
- 2026-11-11 · Transcription-inhibitor payloads: Gene dosage, corrected figures, poisoning evidence and sponsor denominators needed distinct scopes (Raw abstracts, correction notices, sponsor clinical paragraphs and registry fields) · Distinguishes copy loss from TP53 mutation, models from patients, and 13/31 from 2/5; earlier notes and links preserved
- 2026-11-16 · FDA oncology approval notifications (
fda-oncology-approvals,fda-oce-events): The fetcher keyed a notice by its address, so FDA's move of the notices from /drugs/resources-information-approved-drugs/ to /drugs/cancer-approvals/ would have counted 183 notices twice (364 became 548) (Review backlog; FDA's notifications page fetched) · A notice is now its date and title, kept at the newer address; the refreshed list has 367 notices, four of them new - 2026-11-16 · Pimicotinib: Recorded as a phase 2 agent with no named kinase; FDA approved it on 10 November 2026 (FDA notice read; Abbisko's Hong Kong exchange filing of 22 December 2025 read) · Approved, with the FDA's indication, MANEUVER figures and dose quoted, and the China approval of December 2025
- 2026-11-16 · Amivantamab, durvalumab, giredestrant: Lacked the FDA approvals of 5, 13 and 16 November 2026; giredestrant was still phase 3 (FDA notices read) · Approval rows added, each quoting the FDA page; "Rybrevant Faspro" added as a name so the subcutaneous notice reaches the record
- 2026-11-16 · Apmonia Therapeutics, Loki Therapeutics, Cansera, Japan Psycho-Oncology Society: Showed a web host's stock icon as a logo: the WordPress W (two), a grey block, the HostGator mascot (Review backlog; the files compared by md5 with icons already removed from unrelated records, and looked at. The fourth was found by the new check) · Logos removed; the fetcher refuses a file that matches a known generic icon
- 2026-11-16 · NCT05491525 (KANGUROO): A trial of vibegron for neurogenic detrusor overactivity in children was linked to childhood cancers, and to the CIC gene because the registry abbreviates clean intermittent catheterisation as CIC (ClinicalTrials.gov API v2) · Record and both links removed; it is not a cancer trial
- 2026-11-16 · LINC-4 and NCT01331239: Described as trials "in Pituitary tumours" and said the registry lists pituitary tumours as a condition. The registry's condition is Cushing's disease and the trials measured cortisol control (ClinicalTrials.gov API v2; NCI PDQ pituitary tumours treatment summary) · Reworded. The link to pituitary tumours stays, with its reason stated: Cushing's disease is caused by an ACTH-producing pituitary tumour
- 2026-11-16 · PILDI and its second registration: Held twice, as NCT06923787 and NCT07432399; ClinicalTrials.gov lists the second as an alias of the first (ClinicalTrials.gov API v2 (
nctIdAliases)) · Not yet merged: the merge adds a redirect tovercel.json, which waits for the owner. Both records now say they are one study, and the first carries its full title - 2026-11-16 · Electrochemotherapy for vulvar cancer: Phase 2, recruiting, 13 participants, under NCT07443475, which ClinicalTrials.gov now serves as an alias of NCT05916690: no phase, status unknown since January 2024, 20 participants, started March 2021 (ClinicalTrials.gov API v2) · Record carries NCT05916690's entry and says what changed
- 2026-11-16 · ESPAC-4; Greece, Hungary: ESPAC-4 showed the sites, sponsor and status of NCT00058201, which is ESPAC-3; the two country pages listed ESPAC-4 as a trial with a site there and counted it (ISRCTN API (ISRCTN96397434, which itself lists NCT00058201 as a secondary number) and ClinicalTrials.gov API v2 (NCT00058201, secondary id RLUH-NCRI-ESPAC-3V2)) · Snapshot removed and the cross-reference refused in the fetcher; the country pages name ESPAC-3 and their counts drop by one
- 2026-11-16 · SARAH and SIRveNIB: Carried only SARAH's registration and result (ClinicalTrials.gov API v2 (NCT01135056); Chow et al., J Clin Oncol 2018, doi 10.1200/JCO.2017.76.0892, abstract read through Europe PMC) · SIRveNIB's registration, publication and overall survival result added
- 2026-11-16 · Instituto Angolano de Controlo do Cancro: Website field held a paper's DOI (Review backlog; no site of the institute's own found) · Website removed; the paper stays in the links
- 2026-11-16 · University of Puerto Rico Comprehensive Cancer Center: Country US while every other Puerto Rico record is PR, so it was missing from Puerto Rico's own list (Review backlog; the country page reads the country code) · Country PR
- 2026-11-16 · King Hussein Cancer Center, King Faisal Specialist Hospital and Research Centre: Technology tags with no source on the record (The hospitals' own pages read) · KHCC: PET/CT, radioligand therapy and germline testing each now cite a department page. KFSHRC: CAR-T and radioligand therapy now cite the hospital's pages; germline testing removed, because no page of the hospital's describes testing for inherited cancer risk
- 2026-11-16 · Madagascar, Angola, Somalia: Said the IAEA's radiotherapy directory could not be read, or was not read; Somalia's page said no source on medicines regulation was found (DIRAC public country table (JSON); the Somali Federal Ministry of Health's page on its medicines regulatory authority) · Radiotherapy counts added for Madagascar (2 centres, 3 machines, 2022) and Angola (4 centres, 7 machines, 2026); Somalia has no DIRAC entry, and its ministry says the country has no functioning medicines regulator yet
- 2026-11-16 · IPO Lisboa: Said it hosts Portugal's southern regional cancer registry (Lei n.º 53/2017 read (Diário da República PDF): the regional registries' data were integrated into the Registo Oncológico Nacional from 1 January 2018) · Reworded and linked to the national registry
- 2026-11-16 · Triple-negative breast cancer: Gave a mean age of 44.8 for 507 women in Nigeria and Senegal; the figure is for the first cohort of 378 (Huo et al., J Clin Oncol 2009, doi 10.1200/jco.2008.19.6873, abstract read through Europe PMC) · Wording corrected; the second cohort of 129 named
- 2026-11-16 · Guinea country page: The page said IARC's national estimate was built on the Conakry registry ("The subnational registry is the Registre des cancers de Guinée", "the country's cancer figures rest on one city registry", card title "One registry"). IARC's fact sheet says only "Registry (subnational)" and does not name it. (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet for Guinea (PDF text) and the African Cancer Registry Network page) · Reworded in the description, lede, estimate note, registry card title and registry record: the fact sheet does not name the registry; the Conakry registry is the one OnCo has read about. The unsourced phrase "most of it rural" was removed.
- 2026-11-16 · Guinea country page: The screening card said a query of WHO's immunisation data API for Guinea's HPV coverage "returned no rows". The query filtered on an antigen prefix (PRHPV) that matches nothing. (Second accuracy pass; WHO immunisation data API (JSON), queried for Guinea with any HPV antigen) · The query and its link were corrected: WHO holds 44 rows for 2018 to 2025, every one with no doses and no coverage recorded. The card also now cites the health ministry's World Cancer Day 2025 item, which mentions the introduction of vaccines against liver and cervical cancer without dates.
- 2026-11-16 · Guinea country page: The radiotherapy card said IAEA's DIRAC database "needs an account for most queries" and that OnCo found no more recent source. (Second accuracy pass; IAEA DIRAC public equipment data (JSON, 8,627 rows, 157 countries)) · DIRAC answers without an account and lists no radiotherapy centre in Guinea (Senegal has three). The card, lede, description and gap card now say so, as what the directory lists and not as proof that none exists.
- 2026-11-16 · Guinea country page: The regulator card said OnCo had not read anything on Guinea's medicines regulator and stated nothing about it. A readable source exists. (Second accuracy pass; NIH ClinRegs profile of Guinea (HTML text, section reviewed 8 January 2026)) · The card now gives ClinRegs' account: the health ministry is the authority for marketing authorisation and trials, CNERS approves trial applications, the DNPM registers, authorises and imports medicines, and its website has been unreachable since April 2025. The Guinean texts themselves remain unread, and the card says so.
- 2026-11-16 · Guinea country page: The registry card, radiotherapy card and both institution records said the registry's network page "dates from the registry's early years", and the registry record put its radiotherapy statement in the past tense. The page is undated, reports results for 2006 to 2010 and uses the present tense. The radiotherapy card also said specialised oncology "is limited to" one department where the page says there is little apart from it. (Second accuracy pass; African Cancer Registry Network page for the Registre de Cancer de Guinée (HTML text)) · Reworded to "not dated", present tense restored, and "little specialised oncology apart from" used. The registry card now also cites the ministry's 4 February 2025 item, in which the cancer programme's coordinator reports 840 new cases and 595 deaths in the registry's data, and notes that the item's cervical cancer figures differ from IARC's.
- 2026-11-16 · Honduras country page: The page said twice that the social security institute's website (ihss.hn) refused a script, answering 403, which was true on 11 November and no longer is (Second accuracy pass; https://www.ihss.hn/ fetched as raw HTML on 16 November 2026) · Both passages now say the site answered on 16 November 2026, that its home page does not mention cancer, and that OnCo has not read further into it
- 2026-11-16 · Honduras country page: The gap card said OnCo had not read a count of radiotherapy machines or the cancer centres in Tegucigalpa and San Pedro Sula, and that IAEA's DIRAC database needs an account for most queries; OnCo's own records for Honduran hospitals and cancer centres cite DIRAC, and DIRAC's home page says no such thing (Second accuracy pass; https://dirac.iaea.org/ as raw HTML, WHO's Cancer Country Profile 2020 for Honduras as PDF text, and OnCo's institution records) · The card now says what this page does not describe, drops the claim about DIRAC accounts, and gives WHO's 2019 figure of 7.0 external beam radiotherapy machines per 10,000 cancer patients, a rate and not a count
- 2026-11-16 · Honduras country page: The screening card listed WHO's 2019 answers of no cervical or breast screening programme and left out that the same profile answers yes for cervical early detection guidelines and a defined referral system (Second accuracy pass; WHO's Cancer Country Profile 2020 for Honduras, PDF text) · The card now gives both answers for cervical cancer and names the four cancers the early detection row covers
- 2026-11-16 · Armenia country page: The gap card said IAEA's DIRAC radiotherapy database "needs an account for most queries", so the page gave no radiotherapy count; the database's countries table is public (Second accuracy pass; dirac.iaea.org/Query/Countries and its data feed read as raw JSON) · The card now says the public table, read on 16 November 2026, lists three radiotherapy centres in Armenia, and that the equipment detail was not read
- 2026-11-16 · Armenia country page: The note on IARC's estimate said "IARC worked backwards from Armenia's recorded deaths", an inference the fact sheet does not state (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet for Armenia, read with pdftotext) · Reworded to the fact sheet's method: incidence is modelled from mortality estimates, not counted from registered cases, and mortality comes from trends in national rates projected to 2024
- 2026-11-16 · Somalia country page: The research card said OnCo had not yet read hospital-based studies of cancer in Somalia, and the gap card said no source was found on where cancer is treated, although OnCo already held records for two Mogadishu hospitals, one resting on a published pathology study. (Second accuracy pass; Europe PMC abstract of PubMed 33061565 (DOI 10.2147/CMAR.S277202, resolved through Crossref) and the raw text of The Star's report of 17 August 2024) · Added a card on the study: 1,306 confirmed cancers from 5,238 pathology reports at one Mogadishu hospital, 2017 to 2020, with oesophageal cancer commonest (284, 21.7%), then liver (7.6%), breast (7.3%) and cervical (6.7%), stated as one hospital's records and not population rates. Reworded the gap card, labelled the Oasis Cancer Hospital account as a press report, and linked both hospital records.
- 2026-11-16 · Somalia country page: The description, lede and a card title said WHO's 2020 profile recorded "no radiotherapy". The profile's row is the number of external beam radiotherapy machines, 0.0 per 10,000 cancer patients, and says nothing about other forms of radiotherapy. (Second accuracy pass; WHO Cancer Country Profile 2020 for Somalia, PDF text) · Now reads "no external beam radiotherapy machines" in the description, lede, card title and card text.
- 2026-11-16 · Somalia country page: The screening card said HPV vaccination coverage was "marked not available for 2018". The profile prints "n/a" and does not say whether that means not available or not applicable. (Second accuracy pass; WHO Cancer Country Profile 2020 for Somalia, PDF text) · Now quotes the profile's "n/a" and says the profile does not define it.
- 2026-11-16 · Somalia country page: The method note said no Somali national cancer source was found, which overstated the absence: a Somali hospital study exists and is cited in OnCo. (Second accuracy pass; Europe PMC abstract of PubMed 33061565) · Now says no Somali national cancer plan or registry report was found, and names the hospital study.
- 2026-11-16 · El Salvador country page: The gap card said IAEA's DIRAC radiotherapy database "needs an account for most queries", and so gave no radiotherapy figure for El Salvador (Second accuracy pass; DIRAC's public countries table (dirac.iaea.org, JSON read as raw text on 16 November 2026) answers without an account) · The card now says the public table lists five radiotherapy centres in El Salvador, and that OnCo has not read each centre's entry for this page
- 2026-11-16 · El Salvador country page: The gap card named "El Salvador's cancer law, its national cancer registry or plan" as sources not read, which presents each as existing although no source for any of them was read (Second accuracy pass; no source on the page supports their existence) · Reworded to "any cancer law, national cancer registry report or cancer plan that El Salvador may have", with a sentence saying OnCo does not know from its own reading whether each exists
- 2026-11-16 · El Salvador country page: Two cards said the Ministry of Health's home page was read and carried nothing on cancer; on re-reading, salud.gob.sv answered HTTP 403 with an automated browser check (Second accuracy pass; curl of salud.gob.sv and isss.gob.sv on 16 November 2026) · The Ministry of Health statement is labelled as not re-verified on 16 November 2026; the social security institute statement is kept, dated, as "no text on cancer services"
- 2026-11-16 · Jordan country page: The page said 13 trials in OnCo's records list a site in Jordan (11 phase 3, 1 phase 2, 1 phase 2/3; 2 recruiting). The thirteenth was a trial of a bladder medicine in children that was not a cancer trial and has since been removed from the records. (Second accuracy pass; OnCo's ClinicalTrials.gov participation data recounted, and NCT03088540 read from the registry API) · Now 12 trials: 11 phase 3 and 1 phase 2; 1 recruiting, 10 active and not recruiting, 1 completed, as the registry stood on 7 October 2026.
- 2026-11-16 · Jordan country page: The page said WHO's 2020 profile came "from Jordan's answers to WHO surveys" and that HPV vaccination coverage was "marked not available". The profile's text names no survey and prints "n/a" without defining it. (Second accuracy pass; WHO Cancer Country Profile 2020 for Jordan, PDF text) · The survey attribution is removed and the HPV entry is quoted as "n/a".
- 2026-11-16 · Jordan country page: The page said the Jordan Breast Cancer Program is "run by the King Hussein Cancer Foundation", citing a holding page that says only that the website is under construction. (Second accuracy pass; jbcp.jo holding page and the King Hussein Cancer Center's accreditations page, raw HTML) · Now says the centre's accreditations page describes the programme as led by the foundation and the centre, and cites that page.
- 2026-11-16 · Jordan country page: The regulator card reasoned that because trials list sites in Jordan, "trials are authorised there", and stated the Jordan Food and Drug Administration's role without a readable source. (Second accuracy pass; jfda.jo did not respond on 16 November 2026) · Now says only that cancer trials do run there, and that the regulator's role could not be checked against its own pages on 16 November 2026.
- 2026-11-16 · Rwanda country page: The page description, the 'Paying' section title and the insurance card title said community health insurance 'adds cancer medicines from July 2025' as a fact (Second accuracy pass; the National Cancer Control Plan 2025 to 2029 (PDF text, section 2.2.2.4) and RSSB's scheme page, which still lists no cancer medicines) · All three now say the cancer plan says insurance would add cancer medicines from July 2025; the gap card on whether it took effect stays
- 2026-11-16 · Rwanda country page: The registry note said the plan compared 'its 2022 count of 5,548' with GLOBOCAN 2022 (Second accuracy pass; the plan (section 1.3 and its text on figure 3) and the African Cancer Registry Network's page on the registry) · The plan does not date 5,548. The page now says 'a registry count of 5,548', and the table note adds that the plan also gives 5,395 for 2022 and that the registry's network page gives 5,548 as the 2023 count
- 2026-11-16 · Rwanda country page: The cost card said the plan 'does not say how much of the cost is already funded' (Second accuracy pass; the plan's costing chapter (PDF text, figures 20 and 21), whose charts carry an 'Available Financing' label) · Now says the charts carry a label for available financing but OnCo could not read a funded amount from the PDF text
- 2026-11-16 · Rwanda country page: The screening card gave 521,143 clinical breast examinations 'from 2019 to 2024' (Second accuracy pass; the plan's text on table 3) · Now 521,143 people screened by clinical breast examination from 2019 to September 2024
- 2026-11-16 · Rwanda country page: The Butaro card called it a rural district hospital 'where the government and Partners In Health run a cancer centre', citing the national plan, which does not say so (Second accuracy pass; the plan (sections 2.2.2.4 and 3.3.4); the Partners In Health page refused the request on 16 November 2026) · Now 'a district hospital in the north with a cancer centre of excellence', which is what the plan supports
- 2026-11-16 · Rwanda country page: The Butaro card said the plan 'proposes a referral route modelled on the United Kingdom's two-week rule' and that its objectives include 'cancer services in public insurance', and tied the medicines budget to the insurance change (Second accuracy pass; the plan (section 2.2.2.4 and pillar 4 actions)) · Now: the plan suggests a time-bound route like the UK's 14-day rule for seeing a specialist; its actions include adding cancer medicines to community insurance; the inferred link is removed
- 2026-11-16 · Moldova country page: The page said the new in the 2026 to 2030 programme included a National Cancer Registry run to IARC and ENCR standards. The explanatory note calls only the governance objective and the childhood cancer sub-objective new; the registry is an expected result, and the 2024 action plan already had an objective on it. (Second accuracy pass; the Ministry of Health's explanatory note and the draft action plan for 2024 and 2025, PDF text in Romanian) · Reworded to say what the note names as new and that the registry is an expected result.
- 2026-11-16 · Moldova country page: The gap card said OnCo could not reach the Oncological Institute's website. The address tried, oncology.md, is not the institute's; its site is at onco.md and answers. The card also linked cnam.gov.md, which does not resolve, and said IAEA's DIRAC database needs an account for most queries, which its home page does not say. (Second accuracy pass; onco.md and dirac.iaea.org home pages read as raw HTML, DNS lookups for oncology.md and cnam.gov.md) · Gap card now says the institute's site is at onco.md and has not been read for figures, drops the dead link and the claim about DIRAC accounts.
- 2026-11-16 · Moldova country page: The regulator card concluded from 22 registry trials that cancer trials are authorised in Moldova, an inference no source read states; the HPV card compared girls' coverage with a target the plan sets for children and called the draft plan a plan. (Second accuracy pass; AMDM home page as raw HTML and the draft action plan for 2024 and 2025, PDF text) · Inference removed and replaced with what the home page shows; the HPV card now says the draft plan's target is for children vaccinated.
- 2026-11-16 · Latvia country page: The research card counted 34 trials listing Latvia (31 phase 3, 2 phase 2, 1 phase 2/3; 6 recruiting) and named four lead sponsors (Second accuracy pass; OnCo's own trial participation data, recounted after a non-cancer trial (NCT05491525) was removed from the corpus) · Now 33 trials, 31 phase 3 and 2 phase 2, 5 recruiting, with the registry read date (7 October 2026) stated and AbbVie and Bayer added to the sponsors leading three trials each
- 2026-11-16 · Latvia country page: The plans card said Latvia funded more oncology residencies after a quarter of medical oncologists neared retirement in 2022 (Second accuracy pass; EU Country Cancer Profile Latvia 2025, section 5.1, PDF text) · Now says Latvia has funded additional oncology residencies since 2019, as the profile states
- 2026-11-16 · Latvia country page: Prostate screening uptake of 80% in 2023 was given without the profile's warning, on the page and in the National Health Service record (Second accuracy pass; EU Country Cancer Profile Latvia 2025, section 4, PDF text) · Both now say the figure should be read with caution because it counts all PSA tests in men aged 50 to 75 except those done for a cancer diagnosis
- 2026-11-16 · Latvia country page: The Green Corridor card title said an audit found it 'did not speed diagnosis' (Second accuracy pass; EU Country Cancer Profile Latvia 2025, section 5.1, PDF text, citing the State Audit Office (2023)) · Title now says the audit found it did not improve early detection, which is the finding the profile reports
- 2026-11-16 · Latvia country page: The approval card said the profile 'finds the national step is where the delay lies' and 'attributes' low coverage to prices and late filing (Second accuracy pass; EU Country Cancer Profile Latvia 2025, section 5.1, PDF text) · Now says limited coverage is probably related to high prices, and that the wait of over 1,100 days includes the 528 days companies took to apply for reimbursement
- 2026-11-16 · Latvia country page: The registry card title and the SPKC record presented the Cancer Registry as dating from 2009 (Second accuracy pass; EU Country Cancer Profile Latvia 2025, section 5.2, PDF text) · Now says SPKC has managed the registry since 2009 according to the profile, without dating the registry itself
- 2026-11-16 · Latvia country page: The plans card title listed a fourth national cancer plan for 2025 to 2027 as if it existed (Second accuracy pass; EU Country Cancer Profile Latvia 2025, section 2, PDF text) · Title now says a fourth plan is in development, as the profile reports
- 2026-11-16 · Latvia country page: The gap card said the State Agency of Medicines' site returned a server error, the European Commission's copy of the profile refused a script and IAEA's DIRAC needs an account for most queries (Second accuracy pass; zva.gov.lv/en, the European Cancer Inequalities Registry PDF and dirac.iaea.org, each fetched on 16 November 2026) · Gap card now says both sites answered on 16 November 2026 and that the Commission's file matches the archived copy; the DIRAC account claim is removed
- 2026-11-16 · Latvia country page: The HPV card gave vaccination ages as 12 to 18 for girls as settled fact (Second accuracy pass; SPKC World Cancer Day notice of 4 February 2025 (raw HTML) and the EU Country Cancer Profile Latvia 2025, section 3) · Now names both: the profile's 12 to 18 for girls and 12 to 17 for boys, and SPKC's 12 to 17 for both
- 2026-11-16 · Latvia country page: The GLOBOCAN note said the registry count is close to IARC's 'because IARC built it from the registry' (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet for Latvia, PDF text) · Now says the fact sheet states the estimate rests on national registry data
- 2026-11-16 · Panama country page: The lede, description and registry card title said the Instituto Oncológico Nacional "saw 5,273 cancer patients in 2024", with no qualifier. The registry counts cases, and the bulletin marks every 2024 figure as preliminary. (Second accuracy pass; the institute's 2024 hospital registry bulletin, PDF text in Spanish ("Cifras preliminares", and the note on its cut-off date)) · Reworded to 5,273 cases recorded by the hospital registry in 2024, marked as a preliminary count in the lede, the card, the table caption and the institute's record.
- 2026-11-16 · Panama country page: The registry card said many provinces "have diagnosis but lack specialised treatment". (Second accuracy pass; the bulletin's conclusions, which say they lack some options ("carentes de algunas opciones")) · Now reads that many provinces have access to diagnosis but lack some options for specialised cancer treatment.
- 2026-11-16 · Panama country page: The registry card named three bodies that trace deaths: the statistics institute, the Comptroller General and the Electoral Tribunal. (Second accuracy pass; the bulletin's introduction: the statistics institute (INEC) is part of the Comptroller General's office) · Now names two sources: INEC's national mortality data and the Electoral Tribunal's system.
- 2026-11-16 · Panama country page: The registry card stated as fact that "the registry does not yet record stage at diagnosis". (Second accuracy pass; the bulletin's conclusions, which say only that the registry hopes in future to hold clinical stage) · Reworded to what the bulletin says, adding that it gives no figures by stage.
- 2026-11-16 · Panama country page: A card title said "half of the institute's patients" come through the Caja de Seguro Social; the figure is 48% of cases. (Second accuracy pass; graph 3 of the bulletin (2,552 of 5,273)) · Title now says nearly half of the institute's cases.
- 2026-11-16 · Panama country page: The regulator card said the institute publishes a "list of research topics". (Second accuracy pass; the institute's home page, which carries only a menu entry for research protocols) · Now says the website has menu entries for a medicines commission and research protocols, and that OnCo has read neither.
- 2026-11-16 · Panama country page: The gap card said IAEA's radiotherapy directory "needs an account for most queries" and that no radiotherapy count had been read, though WHO's profile, already cited, gives one. (Second accuracy pass; dirac.iaea.org (answers without an account, in an embedded viewer) and WHO's Cancer Country Profile 2020, PDF text) · Gap narrowed to a count by centre; the card now gives WHO's 9.7 external beam machines per 10,000 cancer patients in 2019 and says the directory could not be read as text on 16 November 2026.
- 2026-11-16 · Singapore country page: The page said MediShield Life claim limits on the Cancer Drug List run from $200 to $9,600 a month. (Second accuracy pass; the Ministry of Health's Cancer Drug List spreadsheet dated 1 September 2026 (sheet XML read as text) and its MediShield Life benefits page (raw HTML)) · The range is $200 to $12,800 a month: the list has four rows at $12,800 and the benefits page gives the same range. The page now says so and notes that the tier table printed on the list page stops at $9,600.
- 2026-11-16 · Singapore country page: The page said MediShield Life pays for cancer drug services up to $1,200 a year, the 2022 launch figure, as if current. (Second accuracy pass; the Ministry of Health's MediShield Life benefits page, last updated 3 June 2026 (raw HTML)) · Now $4,500 a year for a patient treated for one primary cancer and $9,000 for more than one, with the $1,200 figure kept as the 2022 brochure's.
- 2026-11-16 · Singapore country page: The HPV card said the rise in coverage from 2018 "matches a move to vaccination in schools", and a gap card referred to a Ministry announcement of it, with no source read for either. (Second accuracy pass; WHO Global Health Observatory API (raw JSON), which gives the coverage series and no reason) · The explanation is removed; the card and the gap card now say no Singapore source explaining the rise was read.
- 2026-11-16 · Singapore country page: The page and the Health Sciences Authority record said registration "is the first test" for the Cancer Drug List, an inference credited to no source. (Second accuracy pass; the Ministry of Health's FAQs on cancer treatment financing (raw HTML)) · Reworded to what the FAQs say: a drug generally counts as clinically proven if registered for the indication. The FAQs are now cited on the card and the record.
- 2026-11-16 · Singapore country page: MediSave was called "the compulsory savings accounts", and the Ministry of Health record said the Ministry "runs Singapore's public healthcare institutions" and listed Healthier SG Screening among its programmes; none of the sources read says so. (Second accuracy pass; the Ministry's MediSave and MediShield Life pages and HealthHub's screening page (raw HTML)) · MediSave is now "the national medical savings scheme", the Ministry's own words; the record describes MediShield Life as the Ministry does (administered by the Central Provident Fund Board); the two unsupported items are removed.
- 2026-11-16 · Singapore country page: A gap card said the Agency for Care Effectiveness's "about us" page returned page-not-found and that its founding was described from the Ministry's FAQs. (Second accuracy pass; ace-hta.gov.sg home, about-us and who-we-are pages (raw HTML), all answering) · The card now says the site answered on 16 November 2026, gives the agency's own description, and says no founding year was found.
- 2026-11-16 · Singapore country page: The lede said Singapore "has counted every cancer in its residents since 1968", a card title gave "about 19,000 new cancers a year", and the subsidy card left out that the income-based subsidy table is for Singaporeans and Medication Assistance Fund drugs. (Second accuracy pass; Singapore Cancer Registry Annual Report 2023 and the Ministry's 2022 brochure (PDF text)) · The lede says the registry has collected information on cancers since 1968; the title gives about 19,300 in 2023 (19,321; the five-year mean is 18,315); the qualifier is added.
- 2026-11-16 · New Zealand country page: The trials card, its title and the page description said 229 trials in OnCo's records have a site in New Zealand, 217 of them also with sites in Australia, out of 5,899 trials. (Second accuracy pass; recounted from OnCo's own ClinicalTrials.gov participation data (src/data/trial-participation.ts)) · 230 trials, 218 of them also in Australia, out of 5,897; the 55 recruiting is unchanged.
- 2026-11-16 · New Zealand country page: The screening card and the Health New Zealand record said bowel screening starts at 58 from October 2025, as if nationwide. (Second accuracy pass; Te Aho o Te Kahu, The State of Cancer in New Zealand 2025, full report PDF text) · The age is being lowered from 60 to 58: the change began in two regions in October 2025, with the other two due to follow in March 2026.
- 2026-11-16 · New Zealand country page: The Medsafe card and record said Medsafe's approval allows a medicine to be sold. (Second accuracy pass; Medsafe's 'about' page read as raw text) · A new medicine cannot be marketed without the consent of the Minister of Health, given after Medsafe's evaluation; the page also says Medsafe is not involved in funding.
- 2026-11-16 · New Zealand country page: The Pharmac card and record credited the statement that the law holds Pharmac to its budget to Pharmac's 'about' page, which does not say it. (Second accuracy pass; Pharmac's 'about' and 'How Pharmac works' pages read as raw text) · The statement is now cited to 'How Pharmac works', which says the Healthy Futures (Pae Ora) Act requires Pharmac to stay within its budget.
- 2026-11-16 · New Zealand country page: The card on the cancer agency called its State of Cancer report five-yearly and said its monitoring reports and medicines comparisons each break results down by ethnicity; neither was read in a source. (Second accuracy pass; Te Aho o Te Kahu, The State of Cancer in New Zealand 2025, full report and media release) · The card now says the agency reported in 2020 and 2025 and that the State of Cancer report breaks results down by ethnicity.
- 2026-11-16 · Croatia country page: The research card said 9 of the 45 trials with a site in Croatia were recruiting, that they were mostly sponsored by drug companies, and that OnCo's participation data cover 5,899 trials (Second accuracy pass; recount of OnCo's ClinicalTrials.gov participation data in the repository) · 8 were recruiting, all 45 are sponsored by companies, and the data cover 5,897 trials
- 2026-11-16 · Croatia country page: The screening card title read 'Breast screening since 2006, bowel since 2007, cervical since 2010', but the source gives the years the Government adopted each programme, on a page last changed in 2017 (Second accuracy pass; the Croatian Institute of Public Health's page on early detection programmes, raw text) · Title now gives the years of adoption, and the card says the page dates from September 2017 and does not describe how the programmes run today
- 2026-11-16 · Croatia country page: The page told readers 'Screening invitations come by post', which no source read for the page says (Second accuracy pass; the Croatian Institute of Public Health's page on early detection programmes, raw text) · Removed; readers are told to ask their family doctor how the programmes invite people
- 2026-11-16 · Croatia country page: The gap card said the tables of the registry report do not extract as text and that survival by cancer could not be quoted; the registry table left the 2022 counts for men and women blank (Second accuracy pass; the report Rak u Hrvatskoj 2023./2024. (PDF text) and the institute's 2022 incidence page) · Five-year net survival for breast (84.5% to 87.2%), prostate (89.8% to 92.8%), bowel (54.6% to 57.3%) and lung cancer (13% to 18.1%) and the prevalence count of 187,118 are now quoted; 2022 shows 13,865 men and 12,240 women; stage at diagnosis remains a named gap
- 2026-11-16 · Croatia country page: HALMED's role (authorisation, inspection, safety monitoring) was credited to its 'about' page, which shows only a menu; the HPV coverage series left out 55% in 2022 and 48% in 2023 (Second accuracy pass; HALMED's page listing its activities and WHO's Global Health Observatory API, raw text) · HALMED's role is now cited to its activities page; the HPV series gives 2022 and 2023 and names the cohort as girls aged 9 to 14
- 2026-11-16 · Slovakia country page: The page counted 110 trials with a site in Slovakia and 24 recruiting, and said the participation data cover 5,899 trials. Two OnCo records share one registration (NCT05253651), so both counts were one too high, and the total was stale (Second accuracy pass; src/data/trial-participation.ts recounted by registry number) · Now 109 trials and 23 recruiting, from 5,897 trial records, in the description, the card title and the card
- 2026-11-16 · Slovakia country page: The card title said a medicine 'is paid for once the Ministry categorises it', and the text that the Ministry decides what insurers pay for and that a company 'must submit' a pharmacoeconomic analysis 'required under' Decree 422/2011. The Ministry page cited publishes the list and the forms and states none of this; a further line said an insurer 'can approve case by case' a medicine that is not categorised, with no source read (Second accuracy pass; health.gov.sk categorisation page and the Ministry's portal kategorizacia.mzsr.sk, raw HTML) · Reworded to what the pages say: the Ministry categorises medicines under the law on what public health insurance pays for and publishes the template for the analysis; the unsourced line on exceptions is removed. The Ministry of Health record is reworded the same way
- 2026-11-16 · Slovakia country page: The page said Slovakia set up the National Oncology Institute 'to coordinate' the National Oncology Programme, titled the card 'to run the cancer programme', and said it publishes 'yearly accounts' of the action plans. Its mission page says it is a platform for cooperation on activities that help fulfil the programme, and the programme page shows accounts for 2019 and 2020 with an archive (Second accuracy pass; noisk.sk mission and programme pages, raw HTML) · Card, title and the National Cancer Institute record now say it was set up to support the programme, and name the years of the accounts
- 2026-11-16 · Slovakia country page: The page said the registry's counts sit in an infographic and a PDF on data.nczisk.sk. The infographics are Power BI pages that open only in a browser; only the written reports are on that server, which still did not answer on 16 November 2026 (Second accuracy pass; nczisk.sk registry and expected-count pages, raw HTML, and a repeat request to data.nczisk.sk) · Table note, registry card and gap card corrected, with the second date
- 2026-11-16 · Slovakia country page: The page said ŠÚKL's pages showed only menus, so its role was not described. Its home page now states its role (Second accuracy pass; sukl.sk home page, raw HTML) · The card now gives ŠÚKL's own description (safety, efficacy and quality of medicines and medical devices) and the gap card is narrowed
- 2026-11-16 · Slovakia country page: The registry card described a quality measure as cases 'known only from a death certificate' and the insurers' data as sent 'under the health insurance law'. The source says cases supplemented from the death certificate, and cites the law on health insurers (Second accuracy pass; nczisk.sk pages, raw HTML) · Both phrases reworded to match
- 2026-11-16 · Mozambique country page: The page said OnCo's participation data cover 5,899 trials, fetched on 7 October 2026. The total was stale after a non-cancer trial was removed, and not all rows were fetched that day. (Second accuracy pass; recount of src/data/trial-participation.ts) · Now 5,897 trials, 5,578 of which list their countries, almost all fetched on 7 and 8 October 2026; still none with a site in Mozambique.
- 2026-11-16 · Mozambique country page: The regulator ANARME was described as "rated at WHO maturity level 3", and its record dated the Minister's remark to 10 September 2026. ANARME's own news item reports the level and does not name WHO; 10 September is the item's dateline, and the remark was made at the opening of the council. (Second accuracy pass; ANARME's news item and home page, raw HTML in Portuguese) · The page and the ANARME record now say ANARME reports reaching the level, that the item does not name WHO, and that no WHO announcement was read; the date is given as the news item's.
- 2026-11-16 · Mozambique country page: The estimate note said the national IARC figure "applies city rates to the whole country" because the Maputo and Beira registries are the network's members, and the description called them "the registries behind the estimates". IARC's fact sheet names only a sub-national registry. (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet for Mozambique, PDF text) · Reworded as a likely reading that the fact sheet does not itself state.
- 2026-11-16 · Mozambique country page: The page said "Mozambique has no national cancer registry in the sources read" and that it "has no national cancer plan". WHO's 2021 country capacity survey records a population-based registry and an operational cancer plan, and the registry network's Maputo page names a director of the National Cancer Control Plan. (Second accuracy pass; WHO Global Health Observatory API (NCD_CCS_PBCR, NCD_CCS_CancerPlan) and the African Cancer Registry Network's Maputo page) · Now says the registries described cover two cities, and that a cancer plan is recorded but OnCo has not read it.
- 2026-11-16 · Mozambique country page: The Beira registry was described as "kept going" by an IARC grant that "has paid" its costs, and the Maputo registries as merged "to share staff time". The network pages are undated (latest year named is 2013 to 2015), and the Maputo page says only that staff time was reorganised. (Second accuracy pass; African Cancer Registry Network pages for Beira and Maputo, raw HTML) · Reworded to what the pages say, with a note that they are undated and funding may have changed.
- 2026-11-16 · Mozambique country page: A gap card said the IAEA's directory of radiotherapy centres "works only in a browser", and ANARME was said to license "manufacturers, importers and pharmacies". (Second accuracy pass; dirac.iaea.org/Data/CountriesLight and ANARME's home page, raw HTML) · The gap card now says the directory publishes country tables and that Mozambique's row was not read on 16 November 2026; ANARME's licensing is given as on its site: making, importing, exporting, distributing and selling medicines.
- 2026-11-16 · Malawi country page: A gap card said the IAEA's directory of radiotherapy centres 'works only in a browser' and that radiotherapy was not counted (Second accuracy pass; IAEA DIRAC country data read as JSON on 16 November 2026) · The card now reports what DIRAC lists: two centres, Kamuzu Central Hospital's cancer centre in Lilongwe (two linear accelerators, one cobalt-60 unit) and the International Blantyre Cancer Center (one linear accelerator), both entries updated in 2026, with the 2016 paper's statement that radiotherapy was then not available
- 2026-11-16 · Malawi country page: The screening card described screen-and-treat without saying most eligible women were not treated (Second accuracy pass; Msyamboza et al. 2016 abstract through Europe PMC) · Added: only 1,001 of 2,311 women eligible for cryotherapy were treated (43.3%) and 266 of 836 referred women (31.8%) received care; also added to the paper record
- 2026-11-16 · Malawi country page: The lede said the Blantyre registry 'has shown for decades how HIV shapes cancer'; the cited paper covers 2008 to 2010 and its abstract does not mention HIV (Second accuracy pass; Chasimpha et al. 2017 abstract through Europe PMC; IARC fact sheet PDF text) · Reworded to the dated finding (Kaposi sarcoma commonest in men in 2008 to 2010) and IARC's 2024 rank for men
- 2026-11-16 · Malawi country page: The page said IARC's national figure 'applies city rates to the whole country' and is 'built on one city's registry', though the fact sheet names no registry (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet for Malawi, PDF text) · Reworded: rates from a sub-national registry, one city's only if the registry is Blantyre's
- 2026-11-16 · Malawi country page: A card said the Ministry tied cancer work to HIV services 'because so much cancer in Malawi is linked to HIV', and a paper record said the 2012 count 'shows how much cancer is linked to HIV and HPV'; neither source says so (Second accuracy pass; Msyamboza et al. 2016 and 2012 abstracts through Europe PMC) · Motive removed; the record now says cervical cancer and Kaposi sarcoma made up most cancers registered
- 2026-11-16 · Malawi country page: The gap card said no national cancer control plan was found, without noting that WHO records one as existing (Second accuracy pass; WHO Global Health Observatory API, NCD_CCS_CancerPlan, Malawi 2021) · Added that WHO's 2021 survey records an operational cancer policy, strategy or action plan; the document itself was not found
- 2026-11-16 · Malawi country page: The registry card said each record is checked in CanReg, the linkage study was said to use 'the Blantyre registry', and a table note said low histology was why national rates were lower (Second accuracy pass; Chasimpha 2017, Horner 2019 and Msyamboza 2012 abstracts through Europe PMC) · Reworded to the abstracts' terms: standard checks in CanReg 4; the population-based cancer registry; causal 'so' removed
- 2026-11-16 · Madagascar country page: The lede, description and two cards said flatly that Madagascar "has no cancer registry" and that "there is no Malagasy measure" of cancer, and that IARC "therefore" estimates from neighbours (Second accuracy pass; WHO Global Health Observatory API (NCD_CCS_PBCR 2021, NCD_CCS_CancerRegNational 2013 to 2019), IARC GLOBOCAN 2024 fact sheet PDF text, and the ClinicalTrials.gov API record NCT03803735) · Now says Madagascar told WHO it has no population-based cancer registry, that IARC builds its estimate from neighbouring countries' data, and names the hospital-based childhood cancer registry of the French Africa Pediatric Oncology Group with a site in Madagascar
- 2026-11-16 · Madagascar country page: The research card said OnCo's participation data cover 5,899 trials fetched on 7 October 2026, and did not say what ClinicalTrials.gov itself lists for Madagascar (Second accuracy pass; recount of src/data/trial-participation.ts and the ClinicalTrials.gov API (cancer studies with a site in Madagascar)) · 5,897 trials, almost all fetched on 7 and 8 October 2026; the card now names the six cancer studies ClinicalTrials.gov lists with a site in Madagascar
- 2026-11-16 · Madagascar country page: The capacity table's note defined "not generally available" as "not available to most people in the public sector", which no source read says (Second accuracy pass; WHO Global Health Observatory API (NCD_CCS_CANCER_DEPT and NCD_CCS_cancer_surgery, Madagascar, 2021)) · Definition removed; the note now gives Madagascar's own Yes answers for tertiary cancer centres or departments and for cancer surgery in the public sector
- 2026-11-16 · Madagascar country page: The same-day card said non-attendance was "highest" among rural women and women without education, and that the published work "was led with partners abroad" (Second accuracy pass; Europe PMC abstracts and affiliations for PMID 31408474, 28643659 and 37749705) · "higher", as the abstract says; the papers are described as having authors in Madagascar and abroad
- 2026-11-16 · Madagascar country page: The Dumont 2017 paper record said it was "Published in French as" an English title in quotation marks, and that women who need hospital care "often do not get it" (Second accuracy pass; Europe PMC record for PMID 28643659 (language French, translated title, abstract)) · The title is given as the indexed English title, and the meaning line now carries the authors' own conclusion (16 of 35 referred women lost to follow-up)
- 2026-11-16 · Madagascar country page: The schistosomiasis study was described as "a 2023 study of 302 women"; the radiotherapy card said DIRAC "holds what each country and centre has reported to the IAEA" (Second accuracy pass; Europe PMC abstract for PMID 37749705; IAEA DIRAC country table JSON) · 500 women were enrolled in 2021 and 302 had complete results, now stated; the unsourced sentence about DIRAC is removed (its counts, 2 centres and 3 machines, last update 2022, were confirmed)
- 2026-11-16 · Dominican Republic country page: The gap card said the IAEA's directory of radiotherapy centres "works only in a browser, so no machine count is given" (Second accuracy pass; IAEA DIRAC equipment report, read as JSON) · The card now gives DIRAC's count: 12 centres with 20 linear accelerators and one Gamma Knife, brachytherapy at six, INCART with three linear accelerators, entries last updated between 2016 and 2025
- 2026-11-16 · Dominican Republic country page: The gap card said SENASA did not answer, the Ministry of Public Health's site was almost empty and the medicines regulator's address was not found, and two cards said OnCo could not read them (Second accuracy pass; the three sites' home pages, which answered on 16 November 2026 at arssenasa.gob.do, msp.gob.do/web/ and digemaps.gob.do) · The cards now say the sites answer at those addresses and that only their home pages have been read; the regulator is named as DIGEMAPS, and SENASA is no longer called "the public insurer", which no cited source said
- 2026-11-16 · Dominican Republic country page: The gap card said the page "has no national cancer plan", which read as if none exists (Second accuracy pass; WHO Global Health Observatory API, NCD_CCS_CancerPlan for DOM) · Reworded to say the page does not describe the plan, and that the country told WHO in 2021 it has an operational cancer policy, strategy or action plan
- 2026-11-16 · Dominican Republic country page: INCART's registry counts of 1,115 cases for 2017 and 1,411 for 2018 were given without saying the report's own tables disagree (Second accuracy pass; INCART's Indicadores Estadísticos y Epidemiológicos 2017-2021, PDF text) · The page and the INCART record now say these come from the table by age and that the tables by province and stage and the annex of rates give 1,261 and 1,468
- 2026-11-16 · Dominican Republic country page: The page and the SISALRIL record said anyone can request a health technology assessment through SISALRIL's online office, and the record called EVTESA a unit (Second accuracy pass; SISALRIL's EVTESA page, HTML text) · Now says requests are made during a call for requests that the page says is open to all interested parties, and describes EVTESA as a process
- 2026-11-16 · Dominican Republic country page: The research card said OnCo's participation data cover 5,899 trials, fetched on 7 October 2026 (Second accuracy pass; src/data/trial-participation.ts, recounted) · Now 5,897 trials, almost all fetched on 7 and 8 October 2026; still none with a site in the Dominican Republic
- 2026-11-16 · Guatemala country page: The summary line and the opening sentence said flatly that Guatemala "has no population-based cancer registry" and that IARC's estimate had "no population-based registry behind it". The sources say less: IARC's fact sheet says it had no incidence data for Guatemala, and the registry statement is Guatemala's own answer to WHO's survey, which was no in 2019 and 2021 but yes in 2015 and 2017. (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet for Guatemala (PDF text) and WHO Global Health Observatory API, indicators NCD_CCS_PBCR and NCD_CCS_CancerRegNational) · The page now says the estimate was made without incidence data from Guatemala and that Guatemala told WHO in 2019 and 2021 it has no population-based registry, after answering yes in 2015 and 2017.
- 2026-11-16 · Guatemala country page: The opening sentence said IARC's figures "are modelled from death records". Only the number of new cancers is modelled from mortality; the deaths are WHO national data projected to 2024. The registry card also said the ratios were based on survival "in other registries", which the fact sheet does not say. (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet for Guatemala (PDF text), data source and methods) · The sentence now says IARC modelled the number of new cancers from death records, and the card says the ratios are based on registry survival data, as the fact sheet words it.
- 2026-11-16 · Guatemala country page: The gap card said IAEA's directory of radiotherapy centres "works only in a browser", so no radiotherapy figure was given. The directory answered a plain request, and its public table of countries lists five radiotherapy centres in Guatemala. (Second accuracy pass; dirac.iaea.org front page and its public country table data, read as raw text on 16 November 2026) · The card now says DIRAC answered and lists five radiotherapy centres in Guatemala, and that OnCo has not read each centre's entry for this page, so machine numbers are not given.
- 2026-11-16 · Guatemala country page: The summary line said aflatoxin exposure was found "in every adult tested" without saying this was one study of 461 adults in five departments. (Second accuracy pass; abstract of Smith et al., PLoS One 2017 (doi 10.1371/journal.pone.0189255) through Europe PMC) · The line now reads "in all 461 adults tested in one study".
- 2026-11-16 · Guatemala country page: The research card said OnCo's participation data cover 5,899 trials. A non-cancer trial has since been removed and the count was stale. (Second accuracy pass; recount of src/data/trial-participation.ts) · Now 5,897 trial records. The 36 trials with a site in Guatemala, 7 recruiting, all 7 sponsored by Merck Sharp & Dohme, were recounted and are unchanged.
- 2026-11-16 · Côte d'Ivoire country page: The page said the national health insurance fund's site (ipscnam.ci) redirected to a login page, so nothing could be said about cover. The site answers a plain request. (Second accuracy pass; CNAM's page of covered care (ipscnam.ci/soins-couverts/) read as raw text on 16 November 2026) · The financing card and the gap card now report what the page says: the acts universal health coverage covers, a 30% patient share with 70% paid by the fund, eleven named specialties, and no mention of cancer; what it covers for cancer is still a named gap.
- 2026-11-16 · Côte d'Ivoire country page: The gap card said the IAEA's directory of radiotherapy centres "works only in a browser, so no machine count is given". Its public countries page names a data service that answers a plain request. (Second accuracy pass; IAEA DIRAC countries and regions table (JSON) read on 16 November 2026) · The gap card now gives DIRAC's row for Côte d'Ivoire (last update 2026): 2 radiotherapy centres, both with radiotherapy, 4 high-energy photon and electron beam teletherapy units and no brachytherapy.
- 2026-11-16 · Côte d'Ivoire country page: The gap card and the HPV card said the Ministry of Health's website "showed only an entry page". The entry page links to a full home page that can be read. (Second accuracy pass; sante.gouv.ci and sante.gouv.ci/accueil read as raw text on 16 November 2026) · Reworded: the site opens on an entry page, the home page behind it carries news and documents and names no national cancer plan, and the document library was not searched.
- 2026-11-16 · Côte d'Ivoire country page: The research card said OnCo's participation data cover 5,899 trials. (Second accuracy pass; recount of src/data/trial-participation.ts after a non-cancer trial was removed) · 5,897 trials; the count of trials with a site in Côte d'Ivoire is unchanged at one, and its two Abidjan sites are now named from ClinicalTrials.gov.
- 2026-11-16 · Guinea country page: The radiotherapy card's summary and the gap card said IAEA's directory lists no radiotherapy centre in Guinea without giving the date it was read (the card's detail already gave it). (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · Both now say the directory, read on 16 November 2026, lists no centre in Guinea. Re-checked against the feed: 8,627 rows, 157 countries, none for Guinea by ISO code or name; three centres in Senegal (two in Dakar, one in Touba). Still says a directory can lag and that this is not proof nothing exists.
- 2026-11-16 · Honduras country page: The gap card said radiotherapy centres are listed in IAEA's DIRAC database, "which this page has not drawn on", and that the page does not describe the number of radiotherapy machines; its link went to the directory's front page. (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · The gap card now gives the directory's listing: five centres with six linear accelerators and four cobalt-60 units, three in Tegucigalpa and two in San Pedro Sula, each named with its machines, two listed with brachytherapy, entries last updated 2019 to 2024, with a note that the older entries may be out of date. WHO's 2019 rate of 7.0 machines per 10,000 cancer patients is kept beside it. The link goes to the directory's countries data.
- 2026-11-16 · Armenia country page: The gap card said the directory's country table lists three radiotherapy centres in Armenia and that OnCo "has not read the equipment detail behind that count". The national centre's card repeated the centre's own description of its brachytherapy department as the first and only one in Armenia with nothing set beside it. (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · The gap card now names the three centres, all in Yerevan, with seven linear accelerators, no cobalt-60 unit, two listed with brachytherapy, all entries last updated 2026. The national centre's card and the gap card now say the directory lists brachytherapy at two centres (the national centre and Erebuni Radiotherapy Centre), that this differs from the centre's own description, and that OnCo cannot say which is current.
- 2026-11-16 · Somalia country page: The radiotherapy card and gap card cited the directory's country table ("rows for 155 countries and territories and none for Somalia") and linked a JSON address of its data service; the directory's per-centre equipment data had not been read. (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · Now says the directory, read on 16 November 2026, lists no centre in Somalia: 8,627 rows, one per centre, covering 157 countries, none for Somalia by ISO code or name. Still says this is not proof that no machine exists. WHO's 2019 figure of 0.0 external beam machines per 10,000 cancer patients is kept. The link goes to the directory's countries data.
- 2026-11-16 · El Salvador country page: The gap card said the directory's table of countries lists five radiotherapy centres in El Salvador and that OnCo "has not read each centre's entry for this page"; its link went to the directory's front page. (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · Now names the five centres, all in San Salvador, with ten linear accelerators, three cobalt-60 units and one Gamma Knife, two listed with brachytherapy, entries last updated 2023 to 2026. The link goes to the directory's countries data.
- 2026-11-16 · Jordan country page: The description, a card title and the King Hussein Cancer Center card gave the centre's six linear accelerators, from its own facilities page, as the only figure, and the gap card listed "IAEA figures for radiotherapy outside KHCC" as something a future pass should read. (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · The directory lists five centres in Jordan, all in Amman, with 21 linear accelerators, no cobalt-60 unit and one Gamma Knife, one centre listed with brachytherapy, entries last updated 2019 to 2026; the gap card now names them. The directory lists 12 linear accelerators at the King Hussein Cancer Center (entry last updated 2025) against six on the centre's own page (read 11 November 2026). The description and the card now give both figures, attributed and dated, and say OnCo cannot say which is current; the card title no longer states a number.
- 2026-11-16 · Rwanda country page: The gap card said "IAEA's DIRAC database needs an account for most queries" and listed IAEA radiotherapy figures as unread. (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · The radiotherapy card and gap card now say the directory lists one centre, Rwanda Military Hospital, Kigali City, with two linear accelerators, no cobalt-60 unit and no brachytherapy, in an entry last updated in 2023. That matches the national cancer plan's one centre and two machines; the card notes the entry is older than the plan and does not show whether brachytherapy has started since.
- 2026-11-16 · Moldova country page: The gap card said OnCo "has not read Moldova's entry" in IAEA's DIRAC database and that the page did not read radiotherapy figures; its link went to the directory's front page. (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · Now says the directory lists one centre, National Oncology Institute, Chisinau, with three linear accelerators, two cobalt-60 units and brachytherapy, in an entry last updated in 2025. The link goes to the directory's countries data.
- 2026-11-16 · Latvia country page: The gap card said radiotherapy capacity is published in IAEA's DIRAC database, "which was not read for this page", and that the page has no count of radiotherapy machines in Latvia. (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · Now gives the directory's listing: six entries (Pauls Stradins Clinical University Hospital appears twice) in Riga, Daugavpils, Liepaja and Sigulda, with 15 linear accelerators, two X-ray generators and no cobalt-60 unit, one entry listed with brachytherapy, all last updated in 2023, with a note that they may be out of date.
- 2026-11-16 · Panama country page: The gap card said the directory's tables "load in an embedded viewer that could not be read as text on 16 November 2026" and that OnCo did not read a count of radiotherapy machines by centre. (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · Now names the four centres the directory lists, each with the city given as Panama, with 12 linear accelerators and no cobalt-60 unit, two listed with brachytherapy, all entries last updated in 2026. WHO's 2019 rate of 9.7 machines per 10,000 cancer patients is kept beside it.
- 2026-11-16 · Singapore country page: The gap card said IAEA's directory of radiotherapy centres "works only in a browser" and is "a Tableau application that does not return its figures to a plain request, so no machine count is given" (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · The gap card states what the directory lists: 11 centres, 27 linear accelerators, no cobalt-60 unit, one Gamma Knife (Mount Elizabeth Hospital), proton or ion therapy at three centres, brachytherapy at four; entries last updated 2018 to 2024, six of them in 2018, so the listing may be out of date. It is given as IAEA's listing, not a Singapore count
- 2026-11-16 · New Zealand country page: The gap card listed "the IAEA directory of radiotherapy centres" among sources OnCo did not read (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · The gap card states what the directory lists: 10 centres, 35 linear accelerators, no cobalt-60 unit, an X-ray generator at three centres, brachytherapy at four; entries last updated 2017 to 2024, four of them in 2017 or 2018, so the listing may be out of date
- 2026-11-16 · Croatia country page: The gap card said "The IAEA's directory of radiotherapy centres was not queried for this page" and that no count of radiotherapy machines was read (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · The gap card states what the directory lists: 10 centres, 37 linear accelerators, no cobalt-60 unit, one Gamma Knife (University Hospital Centre Zagreb), brachytherapy at four; every entry last updated 2026. It notes that two entries carry the name Sestre milosrdnice and that no official Croatian count was found
- 2026-11-16 · Slovakia country page: The gap card said "The IAEA's directory of radiotherapy centres works only in a browser" and that the page gave "no national radiotherapy figures" (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · The gap card states what the directory lists: 13 centres, 21 linear accelerators, four radionuclide teletherapy units of which three are cobalt-60, an X-ray generator at nine centres, brachytherapy at six; ten of 13 entries last updated 2018, the others 2020, 2021 and 2026, so the listing may be out of date. The page still has no radiotherapy figures from the National Health Information Centre
- 2026-11-16 · Mozambique country page: The radiotherapy gap card said OnCo "did not read Mozambique's row" in IAEA's directory and found no source giving the number of radiotherapy machines "or where they are" (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · The card states what the directory lists: one centre, Maputo Central Hospital in Maputo, with one linear accelerator, no cobalt-60 unit and no brachytherapy, entry last updated 2025; beside WHO's 2021 survey answer that radiotherapy is generally available in the public sector
- 2026-11-16 · Malawi country page: The radiotherapy card gave the directory's machines (two linear accelerators and one cobalt-60 unit in Lilongwe, one linear accelerator in Blantyre, both entries 2026) but not brachytherapy, and said "The directory relies on what centres report", which the feed does not state (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · Figures confirmed against the feed. The card adds that the Lilongwe centre is listed with brachytherapy, and the unsupported sentence about how the directory is filled is removed
- 2026-11-16 · Madagascar country page: The radiotherapy card said the directory lists two centres with "three high-energy treatment machines" and "The entry was last updated in 2022", and linked the directory's country-table data address (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · The card names the two Antananarivo centres: CHU-HJRA with one cobalt-60 unit and brachytherapy (entry last updated 2022) and Polyclinic ILAFY with two linear accelerators (entry last updated 2019). The gap card and method text match, and the link goes to the directory's public countries page
- 2026-11-16 · Dominican Republic country page: The gap card gave the directory's figures (12 centres, 20 linear accelerators, one Gamma Knife, brachytherapy at six, entries 2016 to 2025) and said "DIRAC depends on centres reporting", which the feed does not state (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · Figures confirmed against the feed and unchanged. The unsupported clause is replaced by a plain note that a directory can lag
- 2026-11-16 · Guatemala country page: The gap card said IAEA's directory lists five radiotherapy centres but "OnCo has not read each centre's entry for this page, so the number of radiotherapy machines is not given"; a registry card said Guatemala has no population-based registry "so there is no national count" (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · The gap card names the five centres (four in Guatemala, one in Quetzaltenango) with eight linear accelerators, no cobalt-60 unit, one X-ray generator and brachytherapy at four; entries last updated 2023 to 2026. The registry card now gives WHO's 2019 and 2021 survey answers and says OnCo found no national count, without the causal link
- 2026-11-16 · Côte d'Ivoire country page: The gap card said the directory gives "2 radiotherapy centres", "4 high-energy photon and electron beam teletherapy units", "(last update 2026)" and "The table does not name the centres" (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · The gap card names both entries, each with two linear accelerators, no cobalt-60 unit and no brachytherapy: National Center for Medical Oncology and Radiotherapy (NARC), Abidjan, last updated 2024, and CENTRE NATIONAL D'ONCOLOGIE MEDICALE ET DE RADIOTHERAPIE (CNRAO), ABBIDJAN, last updated 2026. It says the second name is the first in French, so the two may be one centre (two linear accelerators, not four), and sets WHO's 2021 answer that radiotherapy was not generally available in the public sector beside it
- 2026-11-16 · Somalia country page and Jordan country page: Said a query of WHO's immunisation data for HPV coverage returned no rows; the query used a filter (antigen names beginning PRHPV) under which neither country is filed (Second accuracy pass; WHO immunisation data API queried for every HPV antigen and read as raw JSON on 16 November 2026) · WHO holds 44 rows for Somalia and 40 for Jordan for 2018 to 2025, none with doses or a coverage figure; the pages say so and link the broader query
- 2026-11-16 · Chad country page: The research card said that of 5,916 ClinicalTrials.gov participation records in OnCo's graph, 5,595 list their countries, counted on 11 November 2026. (Second accuracy pass; recount of src/data/trial-participation.ts after a non-cancer trial was removed) · Now 5,897 trials with participation data, of which 5,578 list their countries, counted on 16 November 2026; still none with a site in Chad.
- 2026-11-16 · Chad country page: The research card said the papers on the page were written with partners in France, the Central African Republic and elsewhere. (Second accuracy pass; author affiliations in the Europe PMC records for PMID 31158254 and 27299910) · The partner countries are France, Gabon and the Democratic Republic of the Congo; no author is affiliated in the Central African Republic.
- 2026-11-16 · Chad country page: The gap card said the IAEA's DIRAC database of radiotherapy centres was not read, and that the page could not say whether Chad has a working radiotherapy machine. (Second accuracy pass; dirac.iaea.org and its public country list, read as raw text on 16 November 2026) · DIRAC answers a plain request: its list holds 155 countries, each with at least one radiotherapy centre, and Chad is not among them. The card says so and notes that this is an absence from a list, not a count.
- 2026-11-16 · Chad country page: The page and the Ministry record called the Centre National d'Excellence en Cancérologie a "first public cancer centre" that was "under construction since June 2025", and a section title spoke of "the first pathology service". (Second accuracy pass; Tchadinfos (15 June 2025), Times Kuwait (24 June 2025) and Alwihda Info (23 September 2024) read as raw text) · The sources report a first-stone ceremony on 15 June 2025 and the opening of a pathology department; neither "first" is stated. The titles, description and record now say the first stone was laid in June 2025 and that a pathology service opened.
- 2026-11-16 · Chad country page: The page said Chad "had no public cancer treatment centre in WHO's 2021 records", and the Hôpital Général de Référence Nationale record said WHO "recorded no cancer centre or cancer department at tertiary level". (Second accuracy pass; WHO cervical cancer country profile 2021 for Chad, PDF text) · WHO's entry is that a cancer centre or cancer department at tertiary level was not generally available (2021); both texts now carry that qualifier.
- 2026-11-16 · Chad country page: The pathology card and the CHU La Renaissance record said the department was "set up with a team from India", that one member "was to stay", and that it opened "so that cancer diagnoses could be made in the country". (Second accuracy pass; Alwihda Info report of 23 September 2024, raw text) · The report says the department opened in the presence of a team from the Pondicherry Institute of Medical Sciences, one of whom was to train Chadian staff, and that some tests previously done abroad could be done locally. The wording now matches.
- 2026-11-16 · Chad country page: The registry card stated as fact that "there is no registry for the rest of the country, and deaths are not recorded by cause nationally". (Second accuracy pass; WHO cancer country profile 2020 for Chad, PDF text) · Reworded: OnCo found no registry for the rest of the country, and WHO's 2020 profile recorded no coverage for the registration of deaths by cause.
- 2026-11-16 · Chad country page: The estimates note said WHO's 2020 profile was "built on GLOBOCAN 2018" and "not based on any registry"; the page and the Ministry record said the 2019 coverage law "began with" the scheme for the poorest; the who-pays card said the Ministry describes cancer medicines as costly and often out of stock. (Second accuracy pass; WHO cancer country profile 2020, WHO Africa's page on the coverage law and the Ministry's 2017 to 2021 plan, raw text) · The profile does not name GLOBOCAN and says its estimates are not based on a population-based cancer registry; WHO Africa records a decision to start with that scheme; the plan's statement is about medicines for non-communicable diseases. All three are reworded to the source.
- 2026-11-16 · Chad country page: The registry's name on the African Cancer Registry Network list was given in quotation marks as "Chad, Registre de Cancer de N'Djamena", on the page and in the registry record. (Second accuracy pass; afcrn.org member page, raw text) · The quotation now reads as the source does: "Chad - Registre de Cancer de N'Djamena".
- 2026-11-16 · Lebanon country page: The research card said KontRASt-02 (NCT05132075) was still active, one of six trials "active but no longer recruiting" (Second accuracy pass; ClinicalTrials.gov API record for NCT05132075 read on 16 November 2026) · The card now dates its counts to the October 2026 registry read and says ClinicalTrials.gov lists KontRASt-02 as terminated
- 2026-11-16 · Lebanon country page: Two source links pointed at unrelated 2015 Ministry news items: the MediTrack link (/en/Pages/6/2020/) and the list of authorised review boards (/en/Pages/4/1729/) (Second accuracy pass; both addresses fetched as raw HTML from moph.gov.lb) · Links replaced with the pages that carry the content, /en/Pages/6/15089/ and /en/Pages/4/11561/
- 2026-11-16 · Lebanon country page: The radiotherapy gap card said the IAEA directory "loads its data by script and returned no readable figures for Lebanon" and that there was no current count of radiotherapy machines (Second accuracy pass; the directory's public data feed (dirac.iaea.org, EquipmentReport) read as raw JSON on 16 November 2026) · The cards now give the listing: 12 centres named with their cities, 16 linear accelerators, no cobalt-60 units, two centres with brachytherapy, all entries last updated in 2026; the gap is narrowed to specialists, working status and waiting times
- 2026-11-16 · Lebanon country page: The Children's Cancer Center of Lebanon was described as "founded in 2002", citing a page that does not carry the founding agreement (Second accuracy pass; the centre's home, history and FAQ pages read as raw HTML) · Now "inaugurated in 2002" (established in 1998 by its own history page), with the home page that states the agreement linked; same fix on the centre's record
- 2026-11-16 · Lebanon country page: Hepatitis B vaccine was said to have been "in the schedule since at least 2006", which the cited WHO coverage series does not state (Second accuracy pass; WHO immunisation coverage API read as JSON) · Reworded to what the series shows: coverage estimates for Lebanon begin in 1999
- 2026-11-16 · Lebanon country page: Four statements went beyond their sources: the share of breast cancers found late "rose" (the speaker said the number of cases rose); radiotherapy patients moved "to centres elsewhere" (the paper says to its authors' hospital); short-course radiotherapy as a share "of treatments" (of breast treatments); and a present-tense title "No organised cervical screening" resting on a 2017 paper (Second accuracy pass; Ministry page of 8 October 2026 (Arabic text), Frontiers in Oncology full text and Vaccine abstract through Europe PMC) · Each reworded to what its source says; the title now dates the statement to the 2017 study
- 2026-11-16 · Lebanon country page: Arafat Tfayli's record said in the present tense that he "heads" the Hematology/Oncology Division at AUB, from a magazine article (Second accuracy pass; Ministry page of 8 October 2026, Europe PMC record for 10.1200/go-26-00203 and the Syndicate of Hospitals PDF) · The short description now gives his affiliation from his 2026 paper and the role the Ministry names; the article's description stays attributed in the summary
- 2026-11-16 · Bolivia country page: The research card gave 5,899 trials in OnCo's participation data and 79 studies of any condition with a Bolivian site on ClinicalTrials.gov (Second accuracy pass; the worktree's src/data/trial-participation.ts recounted (5,897 records, none in Bolivia, 143 in Peru) and the ClinicalTrials.gov API queried on 16 November 2026) · Now 5,897 trials, and 78 studies of any condition with a Bolivian site on 16 November 2026; the seven cancer studies are unchanged
- 2026-11-16 · Bolivia country page: A gap card said the IAEA's directory of radiotherapy centres returned no readable figures to a plain request, so radiotherapy outside the nuclear agency's three centres was not counted, and the page called those "the country's three" centres (Second accuracy pass; the directory's public feed behind dirac.iaea.org/Data/CountriesLight, raw JSON fetched 16 November 2026) · The card now gives what the directory lists: 11 named centres with 12 linear accelerators and 4 cobalt-60 units, nine listed with brachytherapy, entries last updated 2018 to 2025, and where it differs from ABEN's own notes; "the country's" is dropped
- 2026-11-16 · Bolivia country page: AGEMED's list of clinical studies was described as holding "seven observed studies of COVID-19 and leishmaniasis" (Second accuracy pass; AGEMED's health technology page through its content API, raw text, and the Norma para Estudios Clínicos (PDF text)) · The seven are under the heading "observados", meaning the regulator raised observations on the application: six of COVID-19, four of them of chlorine dioxide, and one of leishmaniasis
- 2026-11-16 · Bolivia country page: AGEMED's creation on 21 September 2016 by Decreto Supremo 2905 was credited to AGEMED's own pages, which do not state it (Second accuracy pass; Decreto Supremo 2905 as reproduced by Lexivox, raw text) · Re-cited to the decree, with a link; the 1937 origin stays credited to AGEMED's own history
- 2026-11-16 · Bolivia country page: The page said the Ministry "opened" Consultorios Rosa in October 2026 and that they would operate "first in the main cities and then in all nine departments" (Second accuracy pass; the Ministry's note of 16 October 2026, raw text) · Now "launched", starting in the country's central axis and then the rest, with a pilot network planned for all nine departments
- 2026-11-16 · Bolivia country page: Two card titles stated absences as fact: the registry's reports "are not published" and "no published national cancer plan" (Second accuracy pass; Ley 1223, Decreto Supremo 4881, PAHO's note and the Ministry's note, none of which says so) · Both titles now say OnCo found no published registry report and no published national cancer plan
- 2026-11-16 · Bolivia country page: The La Paz registry paper of 1981 was called "the first report" and "the first population-based picture of cancer in Bolivia" (Second accuracy pass; the abstract through Europe PMC (PMID 7319674)) · Now "a report" and "an early population-based picture"; the abstract does not claim to be first
- 2026-11-16 · Bolivia country page: The ABEN record said the network offers brachytherapy and chemotherapy among its services, and the National Cancer Commission was said to have places for "oncologists and patient groups" and to be "set up" by Decreto Supremo 4881 (Second accuracy pass; ABEN's note of 4 February 2026 and Decreto Supremo 4881 and Ley 1223 (Lexivox), raw text) · Services now follow ABEN's own list (no chemotherapy); the Commission is created by the law, the decree sets its membership: one scientific society representative and two from organised civil society
- 2026-11-16 · Bolivia country page: A gap card said the Gaceta Oficial de Bolivia did not answer, and that Decreto Supremo 4881 asks the national programme to approve the registry's manuals (Second accuracy pass; a plain request to the Gaceta's home page on 16 November 2026, and Decreto Supremo 4881 article 5) · The card now says the Gaceta's home page answered on 16 November but the laws were not re-read there; the programme writes the manuals and the Ministry approves them
- 2026-11-16 · Zambia country page: The registry table's note said IARC's 2024 rates (cervix 74.8, prostate 50.3, breast 24.5, Kaposi sarcoma 8.7) "are for both sexes". (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet for Zambia, PDF text) · The note now says the cervix and breast rates are among women, the prostate rate among men, and only the Kaposi sarcoma rate is for both sexes (12.4 in men, 5.4 in women).
- 2026-11-16 · Zambia country page: The page said Zambia has had "compulsory national health insurance since 2018". The insurer's page dates the Act to 2018 and its board to 15 March 2019; it does not say the scheme ran from 2018. (Second accuracy pass; NHIMA About Us page, raw HTML) · Reworded to "compulsory national health insurance under a 2018 Act" in the description and the card title.
- 2026-11-16 · Zambia country page: The registry card's title said the Zambia National Cancer Registry counts one district "completely", and the registry record said it "counts every cancer" in Lusaka district; one gap card and the Shumba paper record said IARC uses the registry's Lusaka rates. The network page gives complete coverage as an aim, and IARC's fact sheet names no registry. (Second accuracy pass; African Cancer Registry Network page (raw HTML) and IARC fact sheet (PDF text)) · Reworded to "aiming for a complete count of one district" and "aims to record every cancer"; the IARC link is now given as most likely, with the note that the fact sheet does not name the registry.
- 2026-11-16 · Zambia country page: The registry card said the Ministry of Health "is developing" registries for Ndola and Livingstone, in the present tense, on the strength of a 2022 plan. (Second accuracy pass; National Cancer Control Strategic Plan 2022 to 2026, PDF text) · Now "The Ministry of Health's 2022 plan said registries were being developed for Ndola and Livingstone."
- 2026-11-16 · Zambia country page: The regulator card said ZAMRA's guidelines page publishes "a reliance guideline listing the authorities whose decisions it may rely on", and dated guideline MA/GUD/016 to 2026. The page lists guideline titles only, and no date is legible on the scan. (Second accuracy pass; ZAMRA guidelines page (raw HTML) and MA/GUD/016 (scanned PDF, optical character recognition)) · Now says the page lists guidelines on reliance and recognition, which were not read, and that the guideline was posted on ZAMRA's site in October 2026 with no date readable on the scan.
- 2026-11-16 · Zambia country page: The insurance card said the plan's finding on childhood cancers "puts children at risk of treatment abandonment"; the plan says the cost falls on the patient's family and puts the patient at risk of late presentation and treatment abandonment. The decentralisation card said the plan "is titled" its cover theme. (Second accuracy pass; National Cancer Control Strategic Plan 2022 to 2026, PDF text) · Both reworded to the plan's own terms: patients at risk of late presentation and abandonment; the plan "carries the theme".
- 2026-11-16 · Zambia country page: The screening card and the Wu paper record said 5,775 women were "treated at" the Cancer Diseases Hospital and that the authors "attribute" the reversal to low uptake; the Fuady paper record called the study "evidence behind Zambia's move from cryotherapy towards thermal ablation". The abstracts say "seen at", "likely due to", and report no such move. (Second accuracy pass; Europe PMC abstracts for 10.1007/s10552-025-02111-2 and 10.2471/BLT.24.292792) · Reworded to "seen at" and "likely due to"; the Fuady record now says what the study measured.
- 2026-11-16 · Zambia country page: Figures credited to two Lusaka Times reports (the March 2024 equipment contract of about 30.8 million US dollars, the backlog of 1,600 patients, more than 800 waiting to go abroad, the Ndola centre 85% complete, debts to Indian hospitals cleared) stood without a note that they could not be re-read; the site answered with a bot challenge. The gap card on IAEA's DIRAC was vague about why it was not read. (Second accuracy pass; lusakatimes.com (HTTP 403, not worked around), Zambia Monitor (raw HTML), dirac.iaea.org (raw HTML)) · Each card carrying those figures now says the Lusaka Times reports could not be re-read on 16 November 2026; the DIRAC card says its pages open to a plain request without an account but the country table is filled by script and was not read.
- 2026-11-16 · Puerto Rico country page: The radiotherapy gap card said there was no authoritative public count of radiotherapy machines or centres and that the IAEA's Directory of Radiotherapy Centres "serves its figures through an interactive application that a plain request could not read" (Second accuracy pass; the directory's public data feed, read as raw JSON on 16 November 2026) · The card now reports what the directory lists: 13 radiotherapy centres, named, with 19 linear accelerators, one radionuclide teletherapy unit and one Gamma Knife, one centre with brachytherapy, and entries last updated between 1996 and 2021, so possibly out of date
- 2026-11-16 · Puerto Rico country page: The Plan Vital card said a patient in special coverage can get "chemotherapy, radiotherapy, tests and medicines without a referral" (Second accuracy pass; the Triple-S Plan Vital beneficiary manual (PDF text) and ASES's Plan Vital page, which list "chemotherapy, radiology, and related services" and do not name radiotherapy) · Now says medicines, tests and other special coverage services need no referral, and that the listed services include chemotherapy, radiology and related services
- 2026-11-16 · Puerto Rico country page: The medicines register card was titled as "starting with cancer medicines" and the page description called it a "register of cancer medicines" (Second accuracy pass; the Department of Health's November 2024 and March 2025 notices, read by OCR, which cover a register of medicines generally and a 2024 authorisation for distributors of oncology medicines, vaccines and antibiotics) · Title now reads "cancer medicines among them" and the description "register of medicines"
- 2026-11-16 · Puerto Rico country page: The page and the ASES record said Law 72 of 1993 created ASES, and the ASES record said ASES contracts the insurers for Medicare Platino (Second accuracy pass; MACPAC's February 2021 fact sheet (PDF text), which says the Act established the Government Health Plan and describes Platino without naming ASES as its contractor; ASES's page shows only a Medicare Platino section) · Both passages now say only what MACPAC and ASES's page say
- 2026-11-16 · Puerto Rico country page: The registry card said "hospitals and laboratories" must report every case; the NCORP record described the programme's trial types from a page that does not; the spending gap said no copy of Medicaid.gov's pages "was held by the Internet Archive"; the method note said the university cancer centre's record uses the country code US (Second accuracy pass; the registry's 2018-2022 report (PDF text), NCI's NCORP site list, and the corpus record itself) · Now "public and private medical institutions"; the NCORP record describes only what the list shows; "OnCo did not find an archived copy"; the method note says the centre is filed under PR
- 2026-11-16 · Uruguay country page: The tobacco card said Uruguay "banned smoking in enclosed public places and all tobacco advertising in 2008". (Second accuracy pass; the ICSID award in Philip Morris v. Uruguay (PDF text, paragraphs on the 2005 decrees and Law 18,256) and Ley 18.256 as published by IMPO) · Enclosed public places were made smoke-free by Presidential Decree 268/005 of 5 September 2005 and the 2008 law reaffirmed it; the 2008 law limited advertising to the point of sale, and the total ban in article 7 is the wording given by Ley 19.244 of 25 July 2014. The dissent is now described as partial.
- 2026-11-16 · Uruguay country page: The radiotherapy gap card said the IAEA's DIRAC directory returned an error and that no current list of centres and machines could be read. (Second accuracy pass; the IAEA Directory of Radiotherapy Centres' public data, read as raw JSON on 16 November 2026) · The directory lists 12 centres in Uruguay with 25 linear accelerators, no cobalt-60 units and brachytherapy at 5 centres, in entries last updated 2023 to 2026; the centres are now named, beside WHO's 2019 rate, each dated.
- 2026-11-16 · Uruguay country page: The fund card and the Fondo Nacional de Recursos record said cetuximab and daratumumab were added to cover "in 2023". (Second accuracy pass; the fund's 2023 annual report (PDF text)) · The report prints the year as "20223", so the page now says the year is 2022 or 2023 and cannot be read; the 4,980 treatments authorised in 2023 stand.
- 2026-11-16 · Uruguay country page: The registry card gave, without qualification, the registry's statement that there is no sanction for not reporting a cancer. (Second accuracy pass; Ley 16.097 as published by IMPO, articles 13 and 14) · The card and the registry record now also say that the law makes every confirmed cancer diagnosis notifiable to the registry and treats a failure to report as professional misconduct, and give the registry's statement as its own.
- 2026-11-16 · Uruguay country page: The trends card said lung cancer in women was "rising by about 3.5% a year", a figure the registry's report does not give; the trials card gave 110 studies of any condition in Uruguay on ClinicalTrials.gov. (Second accuracy pass; the Registro Nacional de Cáncer's November 2025 report (PDF text) and the ClinicalTrials.gov API on 16 November 2026) · New lung cancers in women are rising by 3.65% a year and deaths by 3.43%; ClinicalTrials.gov returns 108 studies of any condition (33 cancer studies, 7 recruiting, unchanged).
- 2026-11-16 · Uruguay country page: The regulator card credited El País for the statement that two earlier attempts at a medicines agency failed; the research card said the cancer commission lists a tumour bank among its work. (Second accuracy pass; the Salud y Fármacos bulletin, Ley 16.097 and the commission's pages, read as raw text) · The failed attempts are credited to Salud35 and El País as the bulletin cites them; the tumour bank claim, not found on the commission's pages, is removed from the card and from the commission record's programmes.
- 2026-11-16 · Lithuania country page: The trials card, its title and the page description counted 53 trials with a Lithuanian site, 47 of them phase 3 or phase 2/3 and 15 recruiting (Second accuracy pass; recount of src/data/trial-participation.ts after the non-cancer trial NCT05491525 was removed) · 52 trials, 46 of them phase 3 or phase 2/3, 14 recruiting at their last capture on 7 October 2026
- 2026-11-16 · Lithuania country page: The leading causes of cancer death were listed as lung (1,208), colorectum (926), prostate (558) and breast (550), skipping pancreatic cancer (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet for Lithuania, PDF text) · Pancreas (556 deaths, fourth) added before breast (550, fifth)
- 2026-11-16 · Lithuania country page: The registry card said the Cancer Registry was rebuilt in 2025 and moved to fully electronic reporting, and the section title said it had been kept since 1978 (Second accuracy pass; the National Cancer Institute's Cancer Registry page and its news page of 17 December 2025, in Lithuanian) · The registry began rebuilding its information system in 2025; reports still arrive on paper forms as well as electronically; the title now says records from 1978
- 2026-11-16 · Lithuania country page: The institutes card title said the National Cancer Institute became a research institute in 2025 (Second accuracy pass; the institute's activity report for 2025, PDF text, which says its status did not change) · Title now says it passed its hospital work to Santaros Klinikos in 2025 and remains a research institute
- 2026-11-16 · Lithuania country page: The screening card, and the VLK record, said two hospitals became co-ordination centres and began sending invitations in 2025 (Second accuracy pass; EU Country Cancer Profile Lithuania 2025, PDF text, published 3 February 2025, which describes this as planned) · Reworded as the profile's statement that they were to become co-ordination centres, with EUR 340,000 allocated; the page says the profile does not report how that has gone
- 2026-11-16 · Lithuania country page: The radiotherapy card gave 3.8 linear accelerators per million as a plain fact and did not use the IAEA directory (Second accuracy pass; the Baltic census full text on Europe PMC and the IAEA DIRAC public feed saved on 16 November 2026) · The 3.8 figure is attributed to the 2026 census; the directory's listing is added (five centres, 15 linear accelerators, one Gamma Knife, 4 X-ray generators, brachytherapy at four centres, entries updated 2023 and 2024), with the difference between sources stated
- 2026-11-16 · Lithuania country page: The institutes card credited the names of the two children's cancer hospitals to SIOP Europe, and the PSA card and the Patasius paper record credited that study to the Cancer Registry (Second accuracy pass; SIOP Europe's country profile (PDF), the Country Cancer Profile and the Europe PMC author affiliations) · SIOP Europe is credited with two institutions in Vilnius and Kaunas and the Country Cancer Profile with their names; the study is credited to the National Cancer Institute's cancer epidemiology laboratory
- 2026-11-16 · Niger country page: The radiotherapy gap card said the IAEA's radiotherapy directory "needs a browser" and returned no figures, so the number of machines could not be read (Second accuracy pass; the IAEA directory's public data feed, read as raw JSON on 16 November 2026) · The page now states what the directory lists: one centre, the Centre National de Lutte contre le Cancer in Niamey, with one cobalt-60 unit, no linear accelerator and no brachytherapy, in an entry last updated in 2021 that may be out of date. Corrected in the treatment card, the gap card and the centre's record
- 2026-11-16 · Niger country page: The research card counted "5,916 ClinicalTrials.gov participation records, 5,595" listing countries (Second accuracy pass; recount of OnCo's trial participation data on 16 November 2026) · Now 5,897 trials, 5,578 listing their countries, almost all fetched on 7 and 8 October 2026; still none with a site in Niger
- 2026-11-16 · Niger country page: The opening line said Niger "has no national count of cancer, so its figures are IARC estimates", and three places called IARC's figure "mostly" or "mainly" a regional estimate; the registry card said deaths are modelled because WHO rates death registration as having no coverage (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet for Niger and WHO cancer country profile 2020, PDF text) · The causal wording is gone. The page gives IARC's own method (rates scaled to regional all-cancer rates; deaths from incidence and modelled ratios) and WHO's dated rating separately
- 2026-11-16 · Niger country page: Cervical screening figures (8% in five years, 10% ever) were attributed to "a 2019 survey" (Second accuracy pass; WHO cervical cancer country profile for Niger, 2021, PDF text) · The figures are given as WHO's profile gives them, for 2019, without calling them a survey
- 2026-11-16 · Niger country page: WHO's 2014 profile was said to have "recorded no radiotherapy machines" (Second accuracy pass; WHO cancer country profile for Niger, 2014, PDF text) · The profile gives a rate, 0.0 high-energy teletherapy units per million people, and no count of centres; the page now quotes the rate, in two places
- 2026-11-16 · Niger country page: The registry was called "pathology-based" in a section title and its record said it registers "from pathology and hospital records"; a table note said its cases were those "that reached the laboratory or hospitals" (Second accuracy pass; the Ministry of Public Health's 2012 plan (PDF text, French) and the abstract of Garba et al. 2013 through Europe PMC) · Neither source says what records the registry draws on, only that it is kept at a pathology laboratory in Niamey; the wording now says that, and the Garba paper's record credits the location to the Ministry's plan
- 2026-11-16 · Niger country page: The card on the 2007 decree said it "reaches most of the cancers women there get", and the Ministry's record said the Ministry itself made care free by decree and had the cancer centre created by law (Second accuracy pass; the Ministry of Public Health's 2012 plan (PDF text, French) and the IARC fact sheet) · The card now gives IARC's share (breast, cervical and ovarian cancer are six in ten cancers in women) without a claim about the decree's reach; the record says a 2007 decree and a 2011 law did these things
- 2026-11-16 · Niger country page: A gap card dated the survey of African medicines regulators to 2023; the breast cancer awareness paper's record said most women "had not heard of the disease"; the vaccines card said hepatitis B was found "in most people with liver cancer or cirrhosis" (Second accuracy pass; full text of Ncube et al. 2023 and abstracts of Abdou et al. 2020 and Cenac et al. 1995 through Europe PMC) · The survey ran in October and November 2021 and was published in 2023; 41.2% of women were aware of breast cancer; the 1995 study's 73% covers patients with chronic hepatitis, cirrhosis or liver cancer together
- 2026-11-16 · Cameroon country page: The gap card said the IAEA radiotherapy directory "loads its country data through a script and could not be read", so it was not used. (Second accuracy pass; the directory's public data feed, read as raw JSON on 16 November 2026) · The radiotherapy card and gap card now give what the directory lists: two centres, both in Douala, General Hospital of Douala with one cobalt-60 unit (entry last updated 2022) and Cameroon Oncology Centre with one linear accelerator (entry last updated 2023), neither with brachytherapy, with a note that entries that old can be out of date.
- 2026-11-16 · Cameroon country page: The page said Mbingo Baptist Hospital had "trained radiation oncologists, dosimetrists, radiation therapists and physicists, and built a bunker for a linear accelerator", and its record said the hospital described "installing a linear accelerator". (Second accuracy pass; full text of Akundoh Tontu et al. 2024 (doi 10.1016/j.tipsro.2024.100293) through Europe PMC) · Corrected to the paper: bunker construction began in January 2023 with completion expected in early 2025, the donated linear accelerator had not been commissioned, one doctor had qualified as an oncologist and a medical physicist was still needed.
- 2026-11-16 · Cameroon country page: The Douala General Hospital record said the Ministry's April 2026 profile "lists 320 beds". (Second accuracy pass; the Ministry of Public Health's page of 11 April 2026, read as raw HTML) · The page gives no bed count (320 is an image height in its markup); the figure was removed. The record also now says the oncology figures are given without a period, not that they were reported for April 2026.
- 2026-11-16 · Cameroon country page: The medicines card said "doctors, pharmacists and regulators" were interviewed for the 2026 study of cancer medicine quality. (Second accuracy pass; full text of Saidu et al. 2026 (doi 10.1371/journal.pgph.0005370) through Europe PMC) · Corrected to the paper's list: 25 pharmacists, 14 medical oncologists, 12 oncology nurses, five drug distributors and one regulator, interviewed in Yaoundé and Douala in October 2023.
- 2026-11-16 · Cameroon country page: The opening said Cameroon "has no working population-based cancer registry, so its national figures are IARC estimates", the description and checks said the estimates were "made without" a Cameroonian registry, and a gap card said they were "built mostly on other countries' registries". (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet, WHO cancer profile 2020 and cervical cancer profile 2021, and the national plan, all as PDF text) · Each source is now stated as it reads and dated: the plan says the Yaoundé registry was interrupted in 2012, WHO's 2021 profile records no population-based registry, WHO's 2020 profile entered "Registration activity" for 2019, and IARC names no country-specific source and gives its method as national or subnational rates scaled to regional rates. The causal wording was removed.
- 2026-11-16 · Cameroon country page: The trials card counted "5,899 trials" in OnCo's records, "5,580" of them listing countries. (Second accuracy pass; recount of OnCo's ClinicalTrials.gov participation data on 16 November 2026) · Corrected to 5,897 and 5,578; still none with a site in Cameroon. The ClinicalTrials.gov search was re-run the same day (169 studies, 17 matching cancer), and the Wilms tumour project is now credited to its Amsterdam sponsor.
- 2026-11-16 · Cameroon country page: The page said the 10,533 deaths a year in the national plan were "a figure its summary attributes to WHO", called an unnamed hospital in the plan's chemotherapy list "the gynaeco-obstetric and paediatric hospital", and said Cameroon "has no independent medicines agency". (Second accuracy pass; the National Strategic Plan 2020 to 2024 (PDF text), WHO's 2020 profile and the Ministry's page of 13 March 2025) · The plan gives the deaths without attribution and the figure matches WHO's 2020 profile for 2018; the hospital is listed only as HGOPED; and the regulator card now says regulation sits inside the Ministry while a separate agency is being prepared.
- 2026-11-16 · Cameroon country page: The screening card's title said "No national screening programme yet" and gave WHO's 5% and 6% screening figures without their year. (Second accuracy pass; WHO cervical cancer country profile 2021 (PDF text)) · The title now says no programme was recorded up to 2021, and the figures are dated 2019 as the profile dates them.
- 2026-11-16 · Georgia country page: The page said 77 of 5,899 trials in OnCo's records list a site in Georgia, 28 of them recruiting, about 33 in lung cancer, and its card title said most were industry trials in lung cancer. (Second accuracy pass; recount from src/data/trial-participation.ts and the trial records on 16 November 2026) · Now 76 of 5,897 trial records (5,578 list countries): 41 active but not recruiting, 27 recruiting, 5 completed, 3 terminated; 32 involve lung cancer, about two in five, and the title says so.
- 2026-11-16 · Georgia country page: A gap card said the IAEA Directory of Radiotherapy Centres could not be read by a plain request, that WHO's 2020 figure was the only radiotherapy figure, and that no current count of machines and centres could be found. (Second accuracy pass; the directory's public feed, read as raw JSON on 16 November 2026 (rows for the country Georgia, ISO GE/GEO)) · The treatment card and the gap card now give the listing: nine centres (six in Tbilisi, two in Batumi, one in Kutaisi), each named, with 18 linear accelerators, one cobalt-60 unit, one Gamma Knife and one X-ray generator, three centres with brachytherapy, entries last updated 2025 and 2026.
- 2026-11-16 · Georgia country page: A gap card said the regulator's website did not resolve and that the legal herald, Matsne, refused a search. (Second accuracy pass; one plain request to each site on 16 November 2026) · The regulator's site now resolves and answers with a Cloudflare challenge, which OnCo does not bypass; Matsne's English search answered. The card and the regulator's record say so, and say that the resolutions on the cancer package have still not been read.
- 2026-11-16 · Georgia country page: The paper record for Batavani et al. (Cancer Epidemiology 2022) said most stage I thyroid cancers were small papillary tumours. (Second accuracy pass; the abstract through Europe PMC) · Now "at least 42% of stage I thyroid cancers in women were very small papillary tumours", as the abstract gives it (42.4% in women, 37.4% in men).
- 2026-11-16 · Georgia country page: The page said IARC "does not use" Georgia's registry and that the registry "is not the basis" of IARC's incidence figure, and a record said the two differ because of this. (Second accuracy pass; IARC's GLOBOCAN 2024 fact sheet for Georgia, PDF text) · Reworded to what the fact sheet states: it lists no country-specific data source for incidence and gives its method; the causal wording is removed in the note, the registry card, the gap title, the verify line and the registry record.
- 2026-11-16 · Georgia country page: The lede said Georgia "has counted every new cancer in the country" since 2015, "about 10,400 a year". (Second accuracy pass; NCDC's "Cancer in Georgia 2015-2021", PDF text (foreword and table 1)) · Now "has run a national cancer registry since 1 January 2015. It recorded 10,432 new cancers in 2021"; the report itself says coverage is still improving, and yearly counts ran from 10,073 to 11,380.
- 2026-11-16 · Georgia country page: The HPV card gave only WHO's programme coverage rows, skipped the lower years, and linked a query that hides WHO's other HPV rows for Georgia. (Second accuracy pass; WHO immunisation data API queried with the broad HPV antigen filter on 16 November 2026) · First-dose programme coverage for girls now includes 34% in 2021 and 36% in 2023; the card adds that WHO's other rows are lower (2025: 38% of girls with a first dose by age 15 and 28% with the last; official first-dose rows 16% for girls and 10% for boys), and the link uses the broad filter.
- 2026-11-16 · Georgia country page: The stage card said "just over a third" of new cancers were found at stage I or II in 2021 with no qualifier, and a gap card said the NCDC report was "published in November 2022". (Second accuracy pass; NCDC's "Cancer in Georgia 2015-2021", PDF text and file date) · Now "about 38%" (38.2%), with the report's qualifier that the shares leave out in situ tumours and blood and lymphatic cancers; the report is described as a file dated 17 November 2022, since it prints no publication date.
- 2026-11-16 · Georgia country page: The page said City Cancer Challenge names Tbilisi as one of four pilot cities for WHO's breast cancer framework, a record said the Ministry signed the 2025 agreement, and the Ministry's record said it "runs" the Universal Health Care Programme. (Second accuracy pass; City Cancer Challenge's articles of 21 February and 2 October 2025 and WHO Europe's Health Systems in Action 2024, raw text) · The four-city statement is attributed to the signed article on C/Can's site; the agreement is described as signed by the government, with the Minister quoted; the Ministry's summary line no longer says it runs the programme, which WHO describes as purchased by the National Health Agency.
- 2026-11-16 · Georgia country page: The page said two WHO out-of-pocket figures "cannot both be right" and that cervical screening "has halved since 2016". (Second accuracy pass; World Bank API, WHO Europe's 2024 profile and the BMC Public Health 2025 abstract) · Now "the two figures do not agree" (51.9% and 31.2% for 2021, both confirmed), and "about halved between 2016 and 2022", the years the paper covers.
- 2026-11-16 · Azerbaijan country page: The radiotherapy card said the IAEA directory lists "3 CT simulators" in Azerbaijan. (Second accuracy pass; IAEA DIRAC public equipment feed, raw JSON fetched 16 November 2026) · The feed's CT column shows 2 for Baku, 1 for Ganja and none for Nakhchivan, and its simulators column shows none, with no definition of either; the card now says exactly that, names the three centres as the feed spells them, and adds the one X-ray generator listed in Baku.
- 2026-11-16 · Azerbaijan country page: The registry card said the National Oncology Centre asks hospitals, dispensaries and laboratories to report "every new diagnosis on one form", and credited counts for 2008 and 2013 to "the registry". (Second accuracy pass; the Centre's cancer registry page, raw HTML text) · The card now says what the page says: a registry depends on doctors, hospitals and pathology laboratories sending their cases, and completed first-diagnosis forms (TQS No. 025/2-O) are asked for by email; the 2008 and 2013 figures are given as national counts whose source the page does not name. The National Oncology Centre record says "asks for" rather than "collects".
- 2026-11-16 · Azerbaijan country page: The 2014 treatment building at the National Oncology Centre was said to hold a "day chemotherapy" department. (Second accuracy pass; the Centre's history page, raw HTML text) · The source lists endoscopy, invasive diagnostics, chemotherapy, outpatient chemotherapy and onco-haematology; the card and the Centre's record now match it.
- 2026-11-16 · Azerbaijan country page: The insurance card stated "Outpatient medicines are not yet covered" as the agency's position. (Second accuracy pass; the insurance agency's release of 22 October 2026 on the pharmacists' conference, raw HTML text) · The release describes work under way to bring outpatient medicines into the insurance (a positive list, e-prescribing); the card now says that, and that medicines for high blood pressure are planned to be covered on prescription in 2027.
- 2026-11-16 · Azerbaijan country page: A gap card said completeness checks and microscopic confirmation are what "IARC uses to judge a registry", with no source for it. (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet for Azerbaijan (PDF text) and State Statistical Committee table 1.2.7 (Excel cells)) · The unsourced clause is removed; the card states only what the national tables do and do not give.
- 2026-11-16 · Azerbaijan country page: The HPV card linked WHO's coverage data through a narrow filter (programme-coverage antigens only) and said only that the rows "carry no values". (Second accuracy pass; WHO immunisation data, all HPV antigens for AZE, raw JSON fetched 16 November 2026) · The link now uses the broad HPV filter and the card states what is there: 172 rows for 2010 to 2025, for girls and boys, first and last dose, none with doses or coverage.
- 2026-11-16 · Bosnia and Herzegovina country page: The radiotherapy card, its title and the page description gave 20 radiotherapy machines, 19 of them linear accelerators, with one at Cantonal Hospital Zenica. (Second accuracy pass; the IAEA Directory of Radiotherapy Centres public feed, read as raw JSON on 16 November 2026) · The directory lists Zenica with 2 linear accelerators, so the page now gives 21 machines in six centres: 20 linear accelerators and 1 cobalt-60 unit, with brachytherapy at three centres (4 units).
- 2026-11-16 · Bosnia and Herzegovina country page: The page said Brčko District's health insurance fund website could not be reached. (Second accuracy pass; the fund's home page at fzobrcko.ba, fetched as raw HTML on 16 November 2026) · The site now answers. The gap card says the fund's pages have not been read for cancer cover, and links to the fund.
- 2026-11-16 · Bosnia and Herzegovina country page: The HPV card said fewer than one girl in ten "had completed vaccination in 2024". The 9% is WHO's programme coverage of that year's target group for the last dose, not the share of girls vaccinated. The link also used a narrow filter that hid WHO's other HPV rows. (Second accuracy pass; WHO immunisation data API rows for BIH with the broad HPV filter, read as raw JSON) · Reworded to last-dose programme coverage of 9% of the girls targeted in 2024; WHO's coverage by age 15 added (7% first dose, 5% last dose, girls, 2024); link changed to the broad filter.
- 2026-11-16 · Bosnia and Herzegovina country page: The registry card stated as fact that "there is no single national cancer registry". No source read says so. (Second accuracy pass; the Federation's 2021 health report, Republika Srpska's 2024 report and the European Network of Cancer Registries member list, read as raw text) · Reworded: cancer is registered entity by entity, and OnCo found no single national registry.
- 2026-11-16 · Bosnia and Herzegovina country page: The clinical trials card said the medicines agency can inspect trial sites. (Second accuracy pass; the Agency's clinical trials page, read as raw HTML) · Reworded to what the page says: a trial's proposer may ask the Agency to check a trial where it is run, and the Agency may stop or ban a trial it is supervising.
- 2026-11-16 · Bosnia and Herzegovina country page: The shortages card dated earlier Sarajevo cytostatics notices 26 October and 2 November 2026 and called the Federation insurer's list to 6 November the latest; it gave ifosfamide without saying the notice names the brand; and Republika Srpska's medical aids were given as including "breast prostheses". (Second accuracy pass; the Sarajevo clinical centre's notice list, the Federation insurer's medicines page and the Republika Srpska fund's guide, read as raw text) · Notices now dated 2, 9 and 16 November 2026; a supply interruption list dated 13 November 2026 is noted; ifosfamide is marked as listed under the brand Holoksan; the aids read "wigs, bras and prostheses".
- 2026-11-16 · Chad country page: The gap card said the directory's "public list of countries ... holds 155 countries" and that Chad is not among them (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · The gap card says the directory's public data, read on 16 November 2026, holds 8,627 entries from 157 countries and territories and lists no centre in Chad, by name or country code; it says this is an absence from a directory, not proof that Chad has no radiotherapy, and keeps WHO's count of no machines for 2019
- 2026-11-16 · Zambia country page: The gap card said the directory's "country table is filled by script from a data service that was not read"; the page's only machine count was the Ministry's 2022 plan (one linear accelerator, two cobalt-60 units, two brachytherapy units) (Second accuracy pass; IAEA DIRAC public equipment feed read as raw JSON on 16 November 2026) · The treatment card and the gap card state what the directory lists: one centre, Cancer Diseases Hospital in Lusaka, with four linear accelerators, no cobalt-60 unit and two high dose rate iridium-192 brachytherapy units, entry last updated 2026. The Ministry's 2022 count is kept beside it and the difference is stated
- 2026-11-16 · Haiti country page: The HPV card, its link label and a gap card said WHO's immunisation data service "returned no rows for Haiti", and the card title and text said Haiti had reported no HPV coverage to WHO. The first query used a narrow antigen filter. (Second accuracy pass; WHO immunisation data API read as raw JSON with the broad HPV filter on 16 November 2026) · States what is there: 44 rows for 2018 to 2025 (final dose for girls and boys, first dose for 2023 to 2025, administrative and official coverage), none with doses, a target or a coverage value. The title now reads "no coverage figure in WHO's data" and the link uses the broad filter.
- 2026-11-16 · Haiti country page: The lede, description, table note, registry section and the closing note said as fact that "Haiti has no cancer registry, so its national figures are IARC models". WHO's 2020 profile records "No information" on a population-based registry for 2019, and IARC's fact sheet gives "No data" as its incidence source. (Second accuracy pass; WHO cancer country profile 2020 and IARC GLOBOCAN 2024 fact sheet, PDF text) · Each source is now stated as it reads: WHO had no information on a population-based registry, and IARC lists no Haitian data source for incidence and models the figures.
- 2026-11-16 · Haiti country page: The Mirebalais card said a study of 312 women found "nearly half had delays of more than 12 weeks in surgery and in starting chemotherapy", as if of all 312. (Second accuracy pass; abstract of Fadelu et al. 2022 (doi 10.1007/s10549-022-06582-8) read through Europe PMC) · Now gives the denominators: 80% completed surgery; 43% of the 223 with assessable surgery dates had a delay, and 47% of the 141 who had adjuvant chemotherapy started it late.
- 2026-11-16 · Haiti country page: The research card said the linked ClinicalTrials.gov search "shows cancer studies registered with a Haitian site", described OnCo's trials as "mostly large industry and academic trials of new medicines", and said published research from Haiti "comes mostly from partnerships", none of it sourced. (Second accuracy pass; ClinicalTrials.gov API read on 16 November 2026) · States the registry's count: two studies, one withdrawn and one terminated. The unsourced descriptions were removed or narrowed to the research cited on the page.
- 2026-11-16 · Haiti country page: The radiotherapy card title said Haiti had "few specialists", which WHO's profile does not say (it gives no radiation oncologists but does count surgeons and radiologists), and the directory reading was dated 11 November. (Second accuracy pass; WHO cancer country profile 2020, PDF text; IAEA directory feed read on 16 November 2026) · Title now says "no radiation oncologist in WHO's 2019 figures"; the directory is described as listing no entry for Haiti on 16 November 2026, which is not proof that none exists.
- 2026-11-16 · Haiti country page: OFATMA was said to have reported "just over half a million" insured people in 2015 (page and the OFATMA record); the HPV card said the minister announced a campaign "working with PAHO/WHO"; a gap card said the essential medicines list returned "page not found". (Second accuracy pass; OFATMA annual report 2014 to 2015 (PDF text), Prensa Latina report of 6 November 2025, and the Ministry's PDF fetched on 16 November 2026) · Now "more than half a million", as the report says; the Prensa Latina wording is followed (the launch of a campaign, and separately that PAHO/WHO works with the health authorities); the gap card notes the list loaded again on 16 November 2026.
- 2026-11-16 · Senegal country page: The HPV card said "WHO gives no coverage figures for boys". WHO's data holds one: a country-reported final dose for boys in 2021, 49,974 doses and 24.13%. The page had queried only the programme-estimate rows. (Second accuracy pass; WHO immunisation data for SEN read as raw JSON with the broad HPV antigen filter) · The card now gives the 2021 figure for boys, says every other row for boys is empty, labels the girls' series as WHO's programme coverage estimates, notes that Senegal's own reported 2025 figures differ (100% first dose, 62% final dose), and links the broad query.
- 2026-11-16 · Senegal country page: The lede and a card said Senegal "has no population-based cancer registry that IARC can use, so" its figures are borrowed, and the card title said "No population-based registry". WHO's 2020 profile records "No information" on a registry for 2019 and IARC's fact sheet only leaves the country-specific source blank. (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet and WHO 2020 profile, PDF text) · Reworded to what each source says: IARC names no Senegalese incidence source and builds its rates from neighbouring countries' data; the causal wording is gone.
- 2026-11-16 · Senegal country page: The lede stated as fact that the state has paid for chemotherapy since 2019 "and from 2022 for all anticancer medicines", and a card title said "then for all cancers". The source is a review citing a press report, and it says "three years later" and "all anticancer treatments". (Second accuracy pass; Gbenonsi et al. 2024 full text through Europe PMC, and the AlloDocteurs Africa report it cites) · The lede now attributes the claim to the review and uses its wording; the title reads "then for all anticancer treatments"; the press report is linked on the gap card and its hospital list is credited to the reporter.
- 2026-11-16 · Senegal country page: The radiotherapy card said IAEA's directory lists "one CT simulator". The feed's column is "CT" (and "TPS"), with no definition. (Second accuracy pass; IAEA DIRAC equipment feed, copy of 16 November 2026) · Now "one CT and one TPS also listed" at Dalal Jamm; the card states three centres, four linear accelerators, no cobalt-60 unit and brachytherapy at one centre, read on 11 and 16 November 2026.
- 2026-11-16 · Senegal country page: The Ministry of Health record said it "oversees the public hospitals, the health coverage agency and the medicines regulator"; no source read says which ministry supervises SEN-CSU, whose site names both the family and the health ministries. The SEN-CSU record said free care for "children under five"; the agency says 0 to 5 years. (Second accuracy pass; agencecmu.sn script bundle and arp.sn pages, raw text) · The ministry record now says only that SEN-CSU and ARP name it on their sites; the SEN-CSU record says children aged 0 to 5 and people aged 60 and over.
- 2026-11-16 · Senegal country page: Liver cancer was said to follow hepatitis B "mostly" as a statement about Senegal, the screening title said one woman in ten "has ever been screened" without "of reproductive age", and LYMPHODAK was "registered nationally". The sources say west Africa, women of reproductive age, and give only the number SEN18/11. (Second accuracy pass; abstracts of Ramírez Mena 2024, Adam and Sougou 2025 and Diop 2025 through Europe PMC) · Qualifiers restored in the lede, two card titles and the paper record; "nationally" removed.
- 2026-11-16 · Slovenia country page: The research card said 26 trials in OnCo's records have a site in Slovenia. (Second accuracy pass; the trial participation data recounted in raw form, and ClinicalTrials.gov's API for the two redirected registration numbers) · The count is 27: 28 records list Slovenia, one of them a duplicate registration, and ROMANA 1 had been left out. Title, text and trial list corrected.
- 2026-11-16 · Slovenia country page: The radiotherapy card and the UKC Maribor record said IAEA's directory lists the Institute of Oncology Ljubljana with "5 brachytherapy units", from an ambiguous column, and gave the reading date as 11 November 2026. (Second accuracy pass; IAEA DIRAC's public feed, saved as raw JSON on 16 November 2026) · Now states what the directory plainly lists: two centres with entries last updated in 2026, Ljubljana with 7 linear accelerators and brachytherapy including one high-dose-rate iridium-192 unit, Maribor with 5 linear accelerators and no brachytherapy, no cobalt-60 units.
- 2026-11-16 · Slovenia country page: The HPV card gave 42% of girls and 26% of boys "vaccinated in 2025" without saying which WHO series, and linked a narrow query. (Second accuracy pass; WHO's immunisation API queried with contains(ANTIGEN,'HPV'), raw JSON) · The card now names the programme-coverage rows it quotes and adds WHO's by-age-15 rows for 2025, 59% of girls and 35% of boys with a first dose; the link uses the broad filter.
- 2026-11-16 · Slovenia country page: A gap card said the registry's full 2023 annual report could not be read because its link answered with a web page. (Second accuracy pass; the SLORA portal's links and the full report's PDF text, fetched on 16 November 2026) · The card now says the full report answered on 16 November 2026, links it, and adds its count of 99 cervical cancers in 2023 beside ZORA's 100.
- 2026-11-16 · Slovenia country page: The spending card called 2023 "the year before complementary insurance ended". (Second accuracy pass; ZZZS's page on the compulsory health contribution, raw text) · Complementary insurance ended on 31 December 2023, so the card now says 2023 was its last year.
- 2026-11-16 · Slovenia country page: The registry card gave 62% five-year net survival for all adults with cancer. (Second accuracy pass; the registry's 2023 summary report, PDF text) · The figure leaves out non-melanoma skin cancer; the qualifier is now stated, as it is for the 60% and 64% figures and the 406.6 rate.
- 2026-11-16 · Slovenia country page: The profile card said IARC projects registry trends "rather than modelling from deaths", unlike many countries. (Second accuracy pass; IARC's GLOBOCAN 2024 fact sheet for Slovenia, PDF text) · Replaced with what the fact sheet states: source the national registry, method trends in national rates projected to 2024.
- 2026-11-16 · Slovenia country page: The JAZMP card and record said the agency authorises clinical trials and runs pharmacovigilance "under the Medicinal Products Act". (Second accuracy pass; JAZMP's agency profile, mission, prices and HTA pages, raw text) · Reworded to what those pages say: its fields of work, its site sections on clinical trials and pharmacovigilance, and the Act as the basis for its price-setting.
- 2026-11-16 · Slovenia country page: The record for the Irzaldy 2025 paper described the vaccination programme as a school programme and the comparison as the first five years of screening (Second accuracy pass; Europe PMC record for DOI 10.1016/j.lanepe.2024.101203) · The summary now uses the abstract's wording: a preadolescent programme, compared in cytology screening
- 2026-11-16 · Slovenia country page: The right-to-be-forgotten gap card said the period after a diagnosis before 18 is 5 years in every listed jurisdiction. (Second accuracy pass; the ESMO Open review's table in Europe PMC full text) · Added the review's own qualifier: before 18, or 21 in Italy and the Netherlands.
- 2026-11-16 · Mali country page: The page said WHO's HPV coverage rows for Mali "begin in 2024" and that there were no rows with values for boys. It had queried only the programme-coverage antigens. (Second accuracy pass; WHO immunisation data (MT_AD_COV_LONG, Mali, all HPV antigens), raw JSON fetched 16 November 2026) · Now states the programme coverage figures (15% in 2024, 64% in 2025) beside Mali's administrative reports (100.93% of a target of 50,096 in 2024; 72.63% of 308,494 in 2025), an earlier row of 10,321 final doses to girls in 2018 against a target of 10,324, and zero rows for girls and boys in 2019. The link now uses the broad filter and the card title no longer says vaccination "began in 2024".
- 2026-11-16 · Mali country page: The leading-cancers table, the registry card, the lede and the "verify" note said IARC's estimates are built on the Bamako registry's rates. The fact sheet says only "Registry (subnational)" and names none. (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet for Mali, PDF text) · Reworded to "subnational registry rates", with a sentence that the fact sheet does not name the registry and that the registry on the African Cancer Registry Network's list for Mali is Bamako's.
- 2026-11-16 · Mali country page: The GLOBOCAN note said "Mortality has no national source". The fact sheet gives SURVCAN as the country-specific mortality source. (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet for Mali, PDF text) · Now says the fact sheet gives SURVCAN as the country-specific source, with deaths estimated from incidence using modelled mortality-to-incidence ratios.
- 2026-11-16 · Mali country page: The table note said "Mali has no national cancer registry and no national death data on cancer", an absence no cited source states. (Second accuracy pass; WHO cancer country profile 2020 for Mali, PDF text) · Replaced with what WHO's profile says: mortality registration "No coverage" for 2007 to 2016, and mortality estimates not based on any national data on deaths from noncommunicable diseases.
- 2026-11-16 · Mali country page: The description, lede, a card title and the registry record said the Bamako registry had "counted cancers since 1986" or been "running since 1986". The registry's page says it was created in 1986; the registry paper dates its data from 1987. (Second accuracy pass; African Cancer Registry Network member page (HTML text) and Amadou et al. 2022 abstract through Europe PMC) · Reworded to "created in 1986", with data from 1987.
- 2026-11-16 · Mali country page: The radiotherapy card's title and the lede stated as fact that Mali has one radiotherapy centre, and dated the IAEA directory reading 11 November 2026. (Second accuracy pass; IAEA DIRAC public equipment feed, raw JSON fetched 16 November 2026) · Now a dated listing: the directory, read on 16 November 2026, lists one centre, the Hôpital du Mali Radiotherapy Center in Bamako (entry last updated 2026), with two linear accelerators, no cobalt-60 unit and no brachytherapy. The per-population figure is marked as OnCo's arithmetic.
- 2026-11-16 · Mali country page: The free chemotherapy card said the DPM "validates" the hospital's needs and "the Ministry orders", and that "women interviewed in 2024" described free care as incomplete. The paper says the DPM passes the validated needs to the Ministry's finance directorate; 2024 is the interview study's publication year. (Second accuracy pass; Ballo et al., Health Sciences and Disease 2022, PDF text; Schantz et al. 2024 abstract through Europe PMC) · Supply chain reworded to the paper's steps; the sentence now reads "An interview study published in 2024".
- 2026-11-16 · Mali country page: The record for the 2026 cervical cancer cohort at Gabriel Touré carried a shortened title and concluded that "most women with cervical cancer in Mali are diagnosed too late for cure" from one hospital's cohort. The liver cancer paper's "first such decrease in Africa" was stated as fact. (Second accuracy pass; abstracts of DOI 10.1155/ogi/9966827 and 10.4254/wjh.v14.i9.1767 through Europe PMC) · Published title restored; the conclusion is limited to the referral hospital; the "first" is attributed to the authors.
- 2026-11-16 · Mali country page: The Ministry's record called it the ministry "for health and social protection", listed free anticancer medicines as supplied "to public hospitals", and CHU du Point G's record said it hosts "the national cancer registry". (Second accuracy pass; Ministry of Health and Social Development news pages, the Health Sciences and Disease paper and the African Cancer Registry Network page, raw text) · Now "health and social development"; the supply is described as documented at CHU du Point G; the registry is called the Bamako cancer registry.
- 2026-11-16 · Papua New Guinea country page: The HPV card said WHO's coverage data "held no HPV rows for Papua New Guinea ... so there is no coverage figure yet". The query had used a narrow antigen filter. (Second accuracy pass; WHO immunisation coverage rows for PNG fetched as raw JSON with the broad HPV filter on 16 November 2026) · Now says the data holds 44 HPV rows for 2018 to 2025 (final-dose rows for girls and boys from 2018, first-dose rows from 2023), none with a dose count or coverage value, and has no figure yet for the programme begun in 2026; the causal "so" is gone and the link uses the broad filter.
- 2026-11-16 · Papua New Guinea country page: The regulator card said the memorandum with Australia ran to 31 March 2021 and that the TGA offered "up to five" samples a year. (Second accuracy pass; the scanned memorandum on the National Department of Health site, read by OCR) · Now says the memorandum ran until 1 June 2021 and that the TGA undertook to test five samples in each Australian financial year until 31 March 2021.
- 2026-11-16 · Papua New Guinea country page: The lede said IARC has no national incidence data "so" the figures are a model, called the registry "set up in the 1980s", and said radiotherapy "is given in one place" without a source or date. (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet (PDF text), National Cancer Control Policy 2015 (PDF text), IAEA directory feed saved 16 November 2026) · Each statement is now given as its source gives it: the fact sheet lists no country-specific incidence data and models the figures; the policy says a registry was in place in the 1980s; IAEA's directory lists one centre.
- 2026-11-16 · Papua New Guinea country page: The HPV card said the 2015 policy "had only proposed exploring the vaccine's feasibility", and that vaccination was "starting" in two provinces in November 2026. (Second accuracy pass; National Cancer Control Policy 2015 (PDF text) and WHO's release of 8 September 2026 (page text)) · Now says the policy listed introducing HPV vaccination among its objectives with a strategy to explore feasibility, and that the first phase was due to begin in November 2026.
- 2026-11-16 · Papua New Guinea country page: The radiotherapy card gave 2015 statements as current (Port Moresby General Hospital "gives chemotherapy but not radiotherapy"), said the policy "proposed" a second centre, and described the IAEA entry with "one X-ray therapy unit", read on 11 November. (Second accuracy pass; National Cancer Control Policy 2015 (PDF text) and IAEA's directory feed saved 16 November 2026) · The 2015 statements are attributed to the policy, which said it would consider a second centre; the IAEA listing is restated from the feed (two cobalt-60 units, no linear accelerator, one low-dose-rate remote afterloading brachytherapy unit, one X-ray generator, entry last updated 2024) with a note that a directory can lag.
- 2026-11-16 · Papua New Guinea country page: The admissions table and the registry card credited the 28%, 15%, 9% and 9% shares to the National Health Information System's admissions, and said a childhood cancer registry "has been developed" at Port Moresby General Hospital. (Second accuracy pass; National Cancer Control Policy 2015, section 3.1 (PDF text) and the abstract of Raj et al. 2025 through Europe PMC) · The shares are given as the policy's own recalculation, with the section's credit to the information system noted; the registry sentence now says the partnership with SIOP Oceania has helped develop a registry.
- 2026-11-16 · Papua New Guinea country page: The Port Moresby General Hospital record said the 2015 policy describes it as the national referral hospital; the survival gap card said there is no national survival estimate because there is no working registry, and predicted when WHO coverage figures would come. (Second accuracy pass; National Cancer Control Policy 2015 (PDF text)) · The record no longer credits that description to the policy; the gap card says OnCo found no survival estimate and that the policy said population-based data did not exist, and the prediction is removed.
- 2026-11-16 · Burkina Faso country page: The lede said "Burkina Faso has no cancer registry whose figures IARC can use, so its national cancer figures are estimates", the description said the estimate was "made without a national incidence source", and a card was titled "No registry incidence figures that IARC can use". WHO's 2020 cancer country profile records a population-based cancer registry for 2019, and WHO's survey rows answer yes for 2019 and 2021; IARC's fact sheet only leaves its country-specific incidence source blank and gives no cause. (Second accuracy pass; WHO cancer country profile 2020 (PDF text), WHO Global Health Observatory rows NCD_CCS_PBCR and NCD_CCS_CancerRegNational (API JSON), IARC GLOBOCAN 2024 fact sheet (PDF text)) · Reworded to what each source says: IARC names no Burkinabè incidence source; WHO recorded a population-based registry in 2019 and 2021 (no in 2013, 2015 and 2017) without naming it; OnCo found no published figures from one. The causal wording is gone.
- 2026-11-16 · Burkina Faso country page: A gap card said WHO's 2020 cancer country profile for Burkina Faso "could not be reached" and that its indicators were therefore missing. The profile is published at a different file name (cancer-bfa-2020.pdf) and answers. (Second accuracy pass; WHO cancer country profile 2020, Burkina Faso (PDF text)) · The gap card now says what the profile records (a population-based registry for 2019, no coverage for mortality registration in 2007 to 2016, no external beam radiotherapy in 2019, visual inspection screening, staff only as rates per 10,000 cancer patients) and that its indicators predate the Bogodogo radiotherapy centre; the link points to the working file.
- 2026-11-16 · Burkina Faso country page: The HPV card said WHO's coverage rows "begin in 2022" and linked a narrow query; it also called the 2014 and 2015 vaccination in two districts "pilot campaigns", a word the Ministry's page does not use. (Second accuracy pass; WHO immunisation data queried with the broad HPV filter (API JSON), Ministry of Health page of 10 February 2021 (Internet Archive capture, raw HTML)) · The card now says the rows run from 2010 with no values before 2022, keeps WHO's estimates (first dose 70%, 96%, 99%, 95% for 2022 to 2025), adds the country's own administrative reports (95.78% in 2023, 103.81% in 2024, 149.59% in 2025), and says "a vaccination campaign in two health districts". The link uses the broad filter.
- 2026-11-16 · Burkina Faso country page: The screening card named the organiser of the 2023 breast screening campaign as the "Coalition Burkinabè Contre le Cancer" and said the campaign would assess capacity for "breast ultrasound and mammography"; the Ministry's page says "coalition burkinabè de lutte contre le cancer (COBUCAN)" and "écho-mammographies". (Second accuracy pass; Ministry of Health page of 24 April 2023 (Internet Archive capture, raw HTML)) · Reworded to "the Burkinabè coalition against cancer (COBUCAN)" and "breast imaging (écho-mammographies in the Ministry's report)".
- 2026-11-16 · Burkina Faso country page: The radiotherapy card said IAEA's directory recorded "no CT simulator" and described "megavoltage machines"; the directory's columns for CT, simulators and TPS are zero and undefined. The Faso FDA card said "patients can report a side effect through the Med Safety app", where the regulator's page only links to the app. The CHU de Bogodogo record said IAEA gave "support for licensing"; the 2019 statement says assistance for institutional capacity building. (Second accuracy pass; IAEA DIRAC public feed (JSON saved 16 November 2026), Faso FDA presentation page (raw HTML), Burkina Faso statement to the IAEA General Conference 2019 (PDF text)) · The directory is now quoted by its plain columns, with the centre name as the feed spells it, the read date and the 2025 entry year; the Faso FDA card says its site links to Med Safety; the Bogodogo record says IAEA assistance for institutional capacity building.
- 2026-11-16 · Burkina Faso country page: The trials card said only that a ClinicalTrials.gov search "may find" Burkinabè studies and gave no size for OnCo's own data. (Second accuracy pass; src/data/trial-participation.ts recount and the ClinicalTrials.gov API (JSON)) · The card now gives the denominators (5,897 trial records, 5,578 listing countries, none listing Burkina Faso) and says a ClinicalTrials.gov search on 16 November 2026 returned seven studies, one of them a malaria study.
- 2026-11-16 · Costa Rica country page: The lede said IARC "builds its estimate" on the national registry kept since 1976, and a table note said IARC's figure is above the registry's counts because it projects them forward. IARC's fact sheet names no registry and gives no such cause. (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet for Costa Rica, PDF text) · The lede now says IARC's estimate is projected from national registry and mortality data, and the note states IARC's method and the registry's lower, preliminary counts separately.
- 2026-11-16 · Costa Rica country page: The page said WHO records 57% of boys vaccinated against HPV in 2025, with no qualifier, and linked WHO's data with a narrow filter that hides the rows the country reported. (Second accuracy pass; WHO immunisation data API, raw JSON, queried with contains(ANTIGEN,'HPV')) · The card now gives both rows: WHO's programme coverage of 57% first dose and 57% last dose for boys in 2025, and the administrative row of 96.4% first dose (35,553 doses for 36,881 boys), and says the first-dose figure is uncertain. The link uses the broad filter. The title no longer attributes the 2021 fall to the pandemic.
- 2026-11-16 · Costa Rica country page: The radiotherapy card called two hospitals public and three centres private and dated its reading 11 November 2026; the IAEA directory has no ownership field. (Second accuracy pass; IAEA DIRAC public feed, raw JSON saved 16 November 2026 (Costa Rica's five rows are filed under country code CS)) · The card states what the directory lists, read on 16 November 2026: five centres, 12 linear accelerators, no cobalt-60 unit, brachytherapy only at Hospital México (one iridium-192 unit), entries last updated 2021 to 2026, with a note that the 2021 entry may be out of date. The public and private labels are removed.
- 2026-11-16 · Costa Rica country page: The description and a card title said the Guanacaste trials showed one HPV vaccine dose "is enough", and the card placed two decades of trials in Guanacaste. The ESCUDDO abstract reports noninferiority against HPV 16 and 18 infection, and the Costa Rica Vaccine Trial and PRISMA abstracts say only Costa Rica. (Second accuracy pass; Europe PMC abstracts for PubMed 41337735, 33271093 and DOI 10.1016/j.vaccine.2026.129236, and the ClinicalTrials.gov record NCT03180034) · The text now says ESCUDDO, based in Liberia in Guanacaste, found one dose protects against HPV 16 and 18 infection as well as two, and that the trials have run in Costa Rica since 2004.
- 2026-11-16 · Costa Rica country page: The trials card said OnCo's data list Costa Rican sites in 23 cancer trials, "almost all" company trials for six cancers. The data list Costa Rica in 24 records, one of them a Cushing's disease trial, and all are company-sponsored. (Second accuracy pass; recount of src/data/trial-participation.ts) · The card now says 24 of 5,897 trial records list Costa Rica, 23 of them cancer trials, all sponsored by international companies, and that they also cover bladder cancer, lymphoma and three extension studies.
- 2026-11-16 · Costa Rica country page: Several statements went beyond the national cancer plan: the workforce table was credited to a 2022 follow-up of the 2014 imPACT review, the screening register was said to be planned because results are not collected in one system, HPV coverage was said to come from the Ministry's own data, and 4,296 people were said to be seen by multidisciplinary teams "in 2021". (Second accuracy pass; Plan Nacional para el Control del Cáncer 2024-2030, PDF text) · The workforce table is described as a 2021 count sourced to a Ministry report of 2022; the register as a new project with no baseline; the data-source phrase is removed; the multidisciplinary figure reads "by 2021"; the EBAIS and health-area counts are attributed to the 2015 CCSS plan the document cites.
- 2026-11-16 · Costa Rica country page: The courts card and the Loaiza paper record said "fewer rulings" granted experimental medicines after 2014; the study reports that the percentage of such rulings declined, not their number. The ACIB-FUNIN record said staff were lent "in 2023", a year the Ministry's release does not give. (Second accuracy pass; Europe PMC full text of PMC6039739 and the Ministry of Health release of 15 December 2023) · Both now say "a smaller share of rulings", and the ACIB-FUNIN record reports what the Ministry said in December 2023.
- 2026-11-16 · Paraguay country page: The lede and the registry card said Paraguay "has no population-based cancer registry that IARC can use, so" its incidence is modelled, and that it "has no cancer registry whose figures IARC uses". No source gives a cause, and WHO's latest row says a registry exists. (Second accuracy pass; IARC GLOBOCAN 2024 fact sheet PDF text ("Country-specific data source: No data") and WHO Global Health Observatory JSON rows NCD_CCS_CancerRegNational and NCD_CCS_PBCR for PRY) · Now states IARC's "No data" without a cause, and WHO's rows for "Existence of population-based cancer registry": "No" for 2013, 2015, 2017 and 2019, "Yes" for 2021, with both rows linked
- 2026-11-16 · Paraguay country page: The HPV card, its title and the page description said coverage for girls "fell from about 80% to 42%" without naming the measure; the link queried only WHO's programme-coverage rows (Second accuracy pass; WHO immunisation data JSON for PRY with every HPV antigen (174 rows)) · Named as WHO's yearly programme coverage; added WHO's by-age-15 first-dose estimate (98% in 2017, 78% in 2020, 77% in 2025), the 24% last-dose figure for 2023 and Paraguay's own reported 2025 figures (41.76% girls, 38.53% boys); link widened to all HPV rows
- 2026-11-16 · Paraguay country page: The radiotherapy card called the directory's count "11 radiotherapy machines" and "megavoltage machines", said "most" other centres were in Asunción, and dated the reading 11 November (Second accuracy pass; IAEA DIRAC public equipment feed, raw JSON fetched 16 November 2026, six rows for Paraguay) · Now "11 linear accelerators in six centres", no cobalt-60 unit, three of six centres with brachytherapy (two high-dose-rate iridium-192 units, one high-dose-rate cobalt-60), centre names as the directory spells them, three of the other five in Asunción, entries last updated 2026
- 2026-11-16 · Paraguay country page: The cancer fund card said "a quarter of a tobacco tax" goes to the fund and gave "40% of a health allocation from the FONACIDE public investment fund" (also in the Ministry of Health record) (Second accuracy pass; Law 6266 of 2018, article 23, full text on the Congress library site) · Now a quarter of a share of the tobacco tax set by an earlier law (25% of the participation rate in article 21(b) of Law 5538/15), and 40% of the contribution that FONACIDE and the fund for excellence in education and research make to the national health fund
- 2026-11-16 · Paraguay country page: The gap card said "Paraguay screens for cervical cancer" with no source, and described the April 2026 imPACT team as "IAEA, PAHO and WHO reviewers" coordinated by PRONAC (Second accuracy pass; INCAN guidelines page and PAHO's release of 25 April 2026, raw HTML text) · Screening sentence reworded to what the sources show (an HPV and Pap study, a 2022 national guideline, no coverage figure); the review is now an IAEA-led team with PAHO/WHO technical support, coordinated by the Ministry with PRONAC's director