Diffuse midline glioma, H3 K27-altered (including DIPG)
Diffuse midline glioma grows through the brainstem and cannot be removed surgically. A single change in a histone protein (H3 K27M) rewires how the tumour reads its DNA. Radiotherapy was long the only help; in 2025 the first drug aimed at this tumour, dordaviprone (ONC201), was approved after durable shrinkage in some patients, and GD2 CAR-T cells have produced striking early responses.
Overview
Diffuse midline glioma (DMG), H3 K27-altered, is the WHO 2021 name for the tumour long called diffuse intrinsic pontine glioma when it arises in the pons; it also occurs in the thalamus and spinal cord. Most carry a lysine-to-methionine substitution at position 27 of histone H3 (H3.3 K27M or H3.1 K27M), or an equivalent EZHIP overexpression, which inhibits the PRC2 complex, causes global loss of H3K27 trimethylation and locks cells in a stem-like state. Co-alterations in TP53, ACVR1 (H3.1 tumours), PDGFRA and PIK3CA define subgroups. The tumour infiltrates rather than displaces, so it cannot be resected; stereotactic biopsy is now standard because the molecular diagnosis guides trials and prognosis.
Focal radiotherapy remains the only treatment with a proven effect, temporarily restoring function; re-irradiation at progression is now supported by prospective data. Decades of chemotherapy and targeted-agent trials added nothing. The first change came from an unexpected direction: dordaviprone (ONC201), an imipridone that antagonises the dopamine receptor D2 and activates the mitochondrial protease ClpP, produced durable objective responses in a minority of recurrent H3 K27M tumours in pooled phase 2 data (JCO 2024) and received FDA accelerated approval on 6 August 2025 for progressive H3 K27M-mutant DMG in patients aged one year and older, the first systemic therapy approved for this disease. The phase 3 ACTION trial (NCT05580562) tests it after radiotherapy in newly diagnosed patients. In parallel, GD2-directed CAR-T cells delivered intravenously and into the ventricles (Stanford; Nature 2022 and 2024) have produced radiographic and clinical improvement, including a sustained complete response, alongside a manageable but serious inflammatory swelling syndrome. Convection-enhanced delivery, focused-ultrasound opening of the blood-brain barrier, and combination epigenetic strategies (PRC2 and HDAC axis) are in early trials.
Open problems are the ones that define brain-tumour drug development: getting drugs across the blood-brain barrier into an infiltrating tumour, measuring response in a region where swelling is dangerous, and building trials fast enough for a disease that progresses within months. Groups such as the Pacific Pediatric Neuro-Oncology Consortium, CONNECT and SIOPE HGG, with tissue donation programmes, have turned DIPG from the least studied to one of the best characterised childhood cancers.
State of the art today
- H3 K27M defines the disease molecularly; biopsy is safe and standard, and cerebrospinal-fluid tumour DNA is emerging as a way to track it.
- Dordaviprone (ONC201) is the first drug approved for DMG (FDA accelerated approval 2025), a mitochondrial-stress and DRD2 mechanism found by screening rather than design; ACTION tests it in newly diagnosed disease.
- GD2 CAR-T cells given into the blood and the ventricles have produced meaningful neurological improvement and at least one durable complete response in early trials, showing the tumour is immunologically reachable.
- Radiotherapy remains the backbone; re-irradiation extends benefit at progression.
Where it starts and where it drains
Gliomas infiltrate along white matter and can cross the corpus callosum, medulloblastoma sits in the cerebellum, and CNS lymphoma favours deep periventricular tissue; none spread through lymph nodes.
- Frontal lobe (glioblastoma commonest)EGFR-mutant diffuse midline glioma (bithalamic)
- Temporal lobe
- Corpus callosum (butterfly glioma)
- Lower-grade IDH-mutant glioma
- Cerebellum (medulloblastoma)
- Brainstem (diffuse midline glioma)H3.3 K27M (most common; pons, thalamus, spine) · H3.1 K27M (younger children, ACVR1 mutations, longer natural history) · EGFR-mutant diffuse midline glioma (bithalamic) · Adult H3 K27M diffuse midline glioma (thalamic and spinal)
- Ventricles and ependymal lining (ependymoma)
- Sella and pituitary (pituitary tumours, craniopharyngioma)
- Deep periventricular tissue (CNS lymphoma)
No conventional lymphatics: gliomas spread along white matter tracts and, rarely, through cerebrospinal fluid; medulloblastoma can seed the spine.
Same organ: Glioma & glioblastoma, Primary CNS lymphoma, Medulloblastoma, Paediatric low-grade glioma, Atypical teratoid/rhabdoid tumour (ATRT), Ependymoma, Craniopharyngioma, Pituitary tumours (pituitary neuroendocrine tumours) and pituitary carcinoma
Show survival figures (1)
Averages across everyone diagnosed, often years ago. Your stage, subtype, age, fitness and the treatment you receive matter more than the average, and the numbers are improving quickly.
- A few hundred children a year in the United States, mostly aged 5 to 10; the leading cause of brain-tumour death in children (NCI PDQ).
- DNA methylation profilingEstablished
- AI in radiologyEstablished
- cfDNA fragmentomicsEstablished
- Colorectal cancer screening (colonoscopy, FIT, stool DNA, blood)Standard of care
- HCC surveillance in cirrhosis (ultrasound + AFP)Standard of care
- High-risk pancreatic surveillance (CAPS / PRECEDE)Established
- A 28-day national pathway for people with a positive multi-cancer blood test
- A breath test to rule out cancer in people with vague symptoms
- A cancer blood test for older people arriving at A&E with unexplained symptoms
- A legislated, publicly reported 28-day standard from urgent referral to diagnosis
- A live national dashboard of stage at diagnosis as the scorecard for early detection
- A single 'cancer check at 60' appointment bundling all screening tests
Background: Alpha-fetoprotein (AFP), Barrett's oesophagus, CA 19-9, Early detection, Faecal immunochemical test (FIT). Also on OnCo: Symptoms and red flags · Early detection roadmap.
Where the cases are
No country-level case numbers. This cancer is not mapped to a GLOBOCAN site.
Stereotactic biopsy for molecular diagnosis and trial eligibility, then focal radiotherapy (about six weeks; hypofractionated schedules are non-inferior); steroids for symptoms. Enrolment in a trial such as ACTION (dordaviprone after radiotherapy) is recommended.
Dordaviprone (ONC201), FDA accelerated approval August 2025 for patients aged one year and older with progressive disease; re-irradiation is an alternative or addition.
GD2 CAR-T (phase 1, Stanford and others), convection-enhanced delivery, epigenetic and combination trials through consortium networks.
Early palliative care, steroid-sparing strategies, and support for the family; tissue donation at autopsy has been central to research progress.
Subtypes & biomarkers
top- H3.3 K27M (most common; pons, thalamus, spine)
- H3.1 K27M (younger children, ACVR1 mutations, longer natural history)
- H3-wild-type with EZHIP overexpression
- EGFR-mutant diffuse midline glioma (bithalamic)
- Adult H3 K27M diffuse midline glioma (thalamic and spinal)
- H3 K27M by immunohistochemistry or sequencing
- Loss of H3K27me3
- TP53, ACVR1, PDGFRA, PIK3CA co-alterations
- EGFR alteration (bithalamic subtype)
- Cerebrospinal-fluid cell-free tumour DNA for H3 K27M (in development for monitoring)
- MRI with diffusion and perfusion for response assessment
Target prevalence in this cancer
- 1990Two decades of negative chemotherapy trials begin
Radiation-only remains standard as more than 200 agents fail to add benefit.
- 2012H3 K27M mutation discovered
Schwartzentruber and Wu (Nature Genetics) find recurrent histone H3 mutations in DIPG and paediatric glioblastoma.
- 2016WHO recognises diffuse midline glioma, H3 K27M-mutant
First histone-defined tumour entity.
- 2018GD2 identified as a target on H3 K27M gliomas
Mount and Monje (Nature Medicine) show GD2 CAR-T clears tumours in mouse models.
- 2022First GD2 CAR-T responses in patients
Majzner and colleagues (Nature) report clinical and radiographic improvement with intravenous then intracerebroventricular dosing.
- 2024Pooled ONC201 phase 2 data
Durable objective responses in recurrent H3 K27M DMG reported in JCO; ACTION phase 3 open.
- 2025Dordaviprone approved
FDA accelerated approval on 6 August 2025 for progressive H3 K27M-mutant diffuse midline glioma, the first systemic therapy for the disease.
Open problems, and what is being done about each
Most tumours still progress; dordaviprone helps a minority and the phase 3 ACTION trial will show whether earlier use extends survival.
Drug delivery across the blood-brain barrier into infiltrating tumour: convection-enhanced delivery, focused ultrasound and intraventricular cell therapy are being tested.
Response assessment: swelling and pseudo-progression confound MRI in the brainstem; cerebrospinal-fluid tumour DNA is being validated as a marker.
Trial speed: a disease measured in months needs platform trials and regulatory paths built for rarity, which the RACE for Children Act and paediatric consortia are meant to provide.
and how the field plans to fix it →What is being done about thisRare cancers and small trialsAvailable now- AI trial matching & clinical decision supportEstablished
- Comprehensive genomic profilingStandard of care
- Dinutuximab (ch14.18) / dinutuximab betaApproved
- Eflornithine (DFMO)Approved
- LarotrectinibApproved
- Limb-salvage surgery and endoprosthetic reconstructionStandard of care
In trials- ACTIONRecruiting
- NCI-COG Pediatric MATCH (APEC1621)Active
- COG AALL1731Positive
- COG ANBL0032Positive
- DeFiPositive
- Euro Ewing 2012Positive
Ideas and roadmaps- A digital second-opinion network answering community oncologists within 72 hours
- A DRUP-style protocol for off-label generic targeted drugs in rare tumours
- A funded expert second opinion for every new high-stakes or rare cancer diagnosis
- A global first-in-human network for academic cancer trials with single ethics review
- A global open trials operating system any hospital can plug into
- A live 'seats available' feed for trial slots, like airline inventory
Background: Basket, umbrella, and platform trials, Centralisation and high-volume centres, FNCLCC grade (soft-tissue sarcoma), Histotype-tailored therapy, INRG staging and risk groups. Also on OnCo: Find a trial · Expert centres.
Trials
topRecruiting now (live from ClinicalTrials.gov)
Country and place are remembered in this browser only. A postcode is sent to OpenStreetMap's Nominatim service to find coordinates when you press the button; nothing else leaves your device.
Landmark trials in OnCo
Expert centres
topCentres linked to this cancer in OnCo
- via CAR-T for glioma (IL13Rα2, GD2, EGFRvIII, multi-target)
- via this cancer, CAR-T for glioma (IL13Rα2, GD2, EGFRvIII, multi-target)
- Children's Oncology Group (COG)Monrovia, CA, USvia this cancer, NCI-COG Pediatric MATCH (APEC1621)
- Innovative Therapies for Children with Cancer (ITCC)Villejuif, FRvia this cancer, NCI-COG Pediatric MATCH (APEC1621)
- via IMRT / IGRT (modern external beam), GD2 (disialoganglioside)
- Aarhus University HospitalAarhus, DKvia IMRT / IGRT (modern external beam)
- via CAR-T for glioma (IL13Rα2, GD2, EGFRvIII, multi-target)
- ACCELERATEBrussels, BEvia this cancer
- American Society for Radiation OncologyArlington, VA, USvia IMRT / IGRT (modern external beam)
- Centre Antoine LacassagneNice, FRvia IMRT / IGRT (modern external beam)
- Centre Oscar LambretLille, FRvia IMRT / IGRT (modern external beam)
- via IMRT / IGRT (modern external beam)
- via IMRT / IGRT (modern external beam)
- Christian Medical College, VelloreVellore, INvia Early integrated palliative care
- via CAR-T for glioma (IL13Rα2, GD2, EGFRvIII, multi-target)
- Comprehensive Cancer Center Freiburg (CCCF)Freiburg im Breisgau, DEvia IMRT / IGRT (modern external beam)
- Dan L Duncan Comprehensive Cancer Center, Baylor College of MedicineHouston, TX, USNCI comprehensivevia GD2 (disialoganglioside)
- European Society for Radiotherapy and OncologyBrussels, BEvia IMRT / IGRT (modern external beam)
- Geneva University Hospitals (HUG)Geneva, CHvia IMRT / IGRT (modern external beam)
- German Cancer Research Center (DKFZ)Heidelberg, DEvia this cancer
- Groote Schuur Hospital / University of Cape TownCape Town, ZAvia IMRT / IGRT (modern external beam)
- via IMRT / IGRT (modern external beam)
- Hacettepe University Cancer InstituteAnkara, TRvia IMRT / IGRT (modern external beam)
- via IMRT / IGRT (modern external beam)
- Ho Chi Minh City Oncology HospitalHo Chi Minh City, VNvia IMRT / IGRT (modern external beam)
- Hokkaido University HospitalSapporo, JPvia IMRT / IGRT (modern external beam)
- Hunan Cancer HospitalChangsha, CNvia IMRT / IGRT (modern external beam)
- Institut BergoniéBordeaux, FRvia IMRT / IGRT (modern external beam)
- Institut National d'Oncologie, RabatRabat, MAvia IMRT / IGRT (modern external beam)
- Institut Salah AzaïezTunis, TNvia IMRT / IGRT (modern external beam)
- Institute of Oncology LjubljanaLjubljana, SIvia IMRT / IGRT (modern external beam)
- via IMRT / IGRT (modern external beam)
- via IMRT / IGRT (modern external beam)
- International Extranodal Lymphoma Study GroupBellinzona, CHvia IMRT / IGRT (modern external beam)
- IRCCS Humanitas Research HospitalRozzano (Milan), ITvia IMRT / IGRT (modern external beam)
- Istanbul University Institute of OncologyIstanbul, TRvia IMRT / IGRT (modern external beam)
- via IMRT / IGRT (modern external beam)
- Juravinski Cancer Centre / Escarpment Cancer Research InstituteHamilton, ON, CAvia IMRT / IGRT (modern external beam)
- Kenyatta National HospitalNairobi, KEvia IMRT / IGRT (modern external beam)
- Korle Bu Teaching HospitalAccra, GHvia IMRT / IGRT (modern external beam)
- Lagos University Teaching HospitalLagos, NGvia IMRT / IGRT (modern external beam)
- via IMRT / IGRT (modern external beam)
- Li Ka Shing FoundationHong Kong, HKvia Early integrated palliative care
- via IMRT / IGRT (modern external beam)
- via IMRT / IGRT (modern external beam)
- National Cancer Institute (NIH)Bethesda, MD, USvia NCI-COG Pediatric MATCH (APEC1621)
- via IMRT / IGRT (modern external beam)
- National Institute of Oncology, HungaryBudapest, HUvia IMRT / IGRT (modern external beam)
- via IMRT / IGRT (modern external beam)
- Ocean Road Cancer InstituteDar es Salaam, TZvia IMRT / IGRT (modern external beam)
- via IMRT / IGRT (modern external beam)
- Rajiv Gandhi Cancer Institute and Research CentreNew Delhi, INvia IMRT / IGRT (modern external beam)
- Rambam Health Care CampusHaifa, ILvia IMRT / IGRT (modern external beam)
- Rigshospitalet – Copenhagen University HospitalCopenhagen, DKvia IMRT / IGRT (modern external beam)
- Robert H. Lurie Comprehensive Cancer Center of Northwestern UniversityChicago, IL, USNCI comprehensivevia Focused-ultrasound blood-brain barrier opening
- Royal Adelaide HospitalAdelaide, AUvia IMRT / IGRT (modern external beam)
- via IMRT / IGRT (modern external beam)
- via this cancer
- Siriraj Hospital, Mahidol UniversityBangkok, THvia IMRT / IGRT (modern external beam)
- Sunnybrook Odette Cancer CentreToronto, ON, CAvia Focused-ultrasound blood-brain barrier opening
- Tata Medical Center, KolkataKolkata, INvia IMRT / IGRT (modern external beam)
- Tawam HospitalAl Ain, AEvia IMRT / IGRT (modern external beam)
- Tel Aviv Sourasky Medical CenterTel Aviv, ILvia IMRT / IGRT (modern external beam)
- Texas Children's Cancer and Hematology CenterHouston, TX, USvia GD2 (disialoganglioside)
- via IMRT / IGRT (modern external beam)
- TROG Cancer ResearchNewcastle, NSW, AUvia IMRT / IGRT (modern external beam)
- via GD2 (disialoganglioside)
- UMC Utrecht Cancer CenterUtrecht, NLvia IMRT / IGRT (modern external beam)
- University of Malaya Medical CentreKuala Lumpur, MYvia IMRT / IGRT (modern external beam)
- via Focused-ultrasound blood-brain barrier opening
- Velindre Cancer CentreCardiff, GBvia IMRT / IGRT (modern external beam)
- via IMRT / IGRT (modern external beam)
- Zhejiang Cancer HospitalHangzhou, CNvia IMRT / IGRT (modern external beam)
Questions to ask
topQuestions to ask your oncologist about Diffuse midline glioma, H3 K27-altered
Newly diagnosed
- What is my exact diagnosis, stage, and grade, and which tests established them?Why: Everything else follows from an accurate stage and subtype.
- Which biomarkers have been tested on my tumour (for example H3 K27M by immunohistochemistry or sequencing, Loss of H3K27me3, TP53, ACVR1, PDGFRA, PIK3CA co-alterations, EGFR alteration, Cerebrospinal-fluid cell-free tumour DNA for H3 K27M), and what were the results?Why: These results decide eligibility for targeted therapy, immunotherapy, and trials.
- Which subtype is my cancer, and does that change the recommended treatment?Why: Recognised subtypes for this cancer include H3.3 K27M, H3.1 K27M, H3-wild-type with EZHIP overexpression.
- Is germline (inherited) genetic testing recommended for me or my family?Why: Inherited variants can change treatment and matter for relatives.
- For my situation (newly diagnosed), which of the standard options do you recommend and why?Why: Guideline options include: Stereotactic biopsy for molecular diagnosis and trial eligibility, then focal radiotherapy (about six weeks; hypofractionated schedules are non-inferior); steroids for symptoms. Enrolment in a trial such as ACTION (dordaviprone after radiotherapy) is recommended.
- Am I a candidate for Dordaviprone, and what side effects should I expect?Why: Knowing the expected toxicities helps you plan work, family, and supportive care.
- How do the results of ACTION apply to someone like me?Why: Trial populations differ from individual patients; ask how closely you match.
Progressive after radiotherapy, H3 K27M-mutant
- For my situation (progressive after radiotherapy, h3 k27m-mutant), which of the standard options do you recommend and why?Why: Guideline options include: Dordaviprone (ONC201), FDA accelerated approval August 2025 for patients aged one year and older with progressive disease; re-irradiation is an alternative or addition.
- Am I a candidate for Dordaviprone, and what side effects should I expect?Why: Knowing the expected toxicities helps you plan work, family, and supportive care.
Recurrent, trial-eligible
- For my situation (recurrent, trial-eligible), which of the standard options do you recommend and why?Why: Guideline options include: GD2 CAR-T (phase 1, Stanford and others), convection-enhanced delivery, epigenetic and combination trials through consortium networks.
All stages
- For my situation (all stages), which of the standard options do you recommend and why?Why: Guideline options include: Early palliative care, steroid-sparing strategies, and support for the family; tissue donation at autopsy has been central to research progress.
Any stage
- Are there clinical trials I could join, for example of Dordaviprone, ACTION, CAR-T for glioma (IL13Rα2, GD2, EGFRvIII, multi-target), Focused-ultrasound blood-brain barrier opening?Why: Trials are how the next standard of care is set; asking early keeps options open.
- Would a second opinion at a high-volume centre change anything, and can you help arrange it?Why: Rare or high-stakes decisions benefit from a centre that treats many similar patients.
- What supportive care (symptom control, nutrition, exercise, mental health, financial help) is available from the start?Why: Supportive care improves quality of life and helps patients complete treatment.
- I read that “Most tumours still progress; dordaviprone helps a minority and the phase 3 ACTION trial will show whether earlier use extends survival”. How does that affect my plan?Why: Open problems are where trials and second opinions matter most.
- I read that “Drug delivery across the blood-brain barrier into infiltrating tumour: convection-enhanced delivery, focused ultrasound and intraventricular cell therapy are being tested”. How does that affect my plan?Why: Open problems are where trials and second opinions matter most.
Print this page for your appointment (your browser's print command). These prompts are for discussion; your clinical team knows your case.
Direct links plus the targets, companies, and technologies of this cancer's products.
technologies
7targets
2drugs
2companies
1institutions
6pathways
2terms
4trials
2bottlenecks
2Latest papers
topQuery for this cancer: (TITLE:"Diffuse midline glioma, H3 K27-altered" OR ABSTRACT:"Diffuse midline glioma, H3 K27-altered" OR TITLE:"including DIPG" OR ABSTRACT:"including DIPG" OR TITLE:"DIPG" OR ABSTRACT:"DIPG" OR TITLE:"Diffuse intrinsic pontine glioma" OR ABSTRACT:"Diffuse intrinsic pontine glioma" OR TITLE:"DMG" OR ABSTRACT:"DMG" OR TITLE:"H3 K27M glioma" OR ABSTRACT:"H3 K27M glioma") AND (treatment OR therapy OR trial OR survival OR diagnosis). Results are unfiltered search hits about Diffuse midline glioma, H3 K27-altered (including DIPG), not a curated reading list.
Pages like this
not linked directly; found by shared links- CancerEpendymoma
Shares NCI-COG Pediatric MATCH (APEC1621), Late effects and survivorship toxicity, SIOP Europe – European Society for Paediatric Oncology, Paediatric low-grade glioma and the tags nci-coverage, paediatric, cns.
- CancerAtypical teratoid/rhabdoid tumour (ATRT)
Shares NCI-COG Pediatric MATCH (APEC1621), Late effects and survivorship toxicity, SIOP Europe – European Society for Paediatric Oncology, Early integrated palliative care and the tags nci-coverage, paediatric, cns.
- CancerCraniopharyngioma
Shares Late effects and survivorship toxicity, Paediatric low-grade glioma, Children's Oncology Group (COG), Rare and paediatric cancers without markets and the tags nci-coverage, paediatric, cns.
- TrialFIREFLY-1
Shares RACE for Children Act, Paediatric low-grade glioma and the tags nci-coverage, paediatric, cns.
- TrialTADPOLE (CDRB436G2201)
Shares RACE for Children Act, Paediatric low-grade glioma and the tags nci-coverage, paediatric, cns.
- TrialCOG ACNS0331
Shares Late effects and survivorship toxicity, Children's Oncology Group (COG), IMRT / IGRT (modern external beam) and the tags nci-coverage, paediatric, cns.
- InstitutionHistiocyte Society
Shares Rare and paediatric cancers without markets and the tags nci-coverage, paediatric.
- CancerRare cancers of childhood (NCI PDQ umbrella)
Shares RACE for Children Act, NCI-COG Pediatric MATCH (APEC1621), SIOP Europe – European Society for Paediatric Oncology, Children's Oncology Group (COG) and the tags nci-coverage, paediatric.