DataNon-Hodgkin lymphoma (all types)

Direct links plus the targets, companies, and technologies of this cancer's products.

• Living with non-Hodgkin lymphoma, the decisions and the people who make them: This is orientation from public patient pages, guidelines and the trials themselves, not advice for your case: your own team's instructions and the 24-hour number they gave you come first, and no figure here is a prediction about you. The decision records, the question sets, the first 60 days checklist and the red cards on this page were written from the NHS, Lymphoma Action, Macmillan and Cancer Research UK patient pages, from NICE NG52, NG47, CG151 and NG234, from the NHS Breast Screening Programme protocols and the Green Book, and from the trial reports named beside each figure, all read on 1 October 2026.
• The biology in detail, 1 of 5. The antigens. Ten surface targets carry the lymphoma medicines, and nine of the ten are normal proteins on normal cells. CD20 and CD19 are B-cell lineage antigens, so anti-CD20 and anti-CD19 treatments empty the B-cell compartment and leave hypogammaglobulinaemia, more infection, a blunted vaccine response and a risk of hepatitis B reactivation; after CAR-T the aplasia is deeper and can last years. CD79b and CD22 are also B-cell antigens, chosen for conjugates because they internalise. CD30 is the nearest to a restricted antigen, carried by activated lymphocytes and monocytes but not by resting ones, which is why brentuximab vedotin's limiting toxicity is the auristatin payload rather than the target. CD38 is mostly a myeloma antigen with no lymphoma approval and one laboratory consequence worth knowing, a positive indirect antiglobulin test that masks alloantibodies. CD47 is on every red cell, so blocking it causes anaemia by design and the dosing had to be redesigned around that. CCR4 is on regulatory T cells as well as on cutaneous and adult T-cell lymphoma, so depleting it causes rash and raises transplant risk. CD52 is on B cells, T cells, monocytes and dendritic cells at very high density, which makes alemtuzumab efficient and leaves CD4 counts low for a year. CD3 is the one target not on the tumour at all in B-cell disease: it is the handle a bispecific antibody grips, and cytokine release syndrome is that mechanism working.
• The biology in detail, 2 of 5. The lesions. Chronic active B-cell receptor signalling defines the activated B-cell-like group and was shown by killing those cells, and only those cells, with knockdown of IgM, CD79A, CD79B or BTK; CD79B ITAM mutations were present in 18% of them and activating CARD11 mutations in about 10%. MYD88 L265P feeds the same NF-kB hub from the toll-like receptor side in 29% of activated B-cell-like cases, 9% of MALT lymphomas and 91% of lymphoplasmacytic lymphoma. BCL2 arrives by a recombinase error at the pre-B-cell stage, t(14;18), present in most follicular lymphomas and 13.5% of unselected large B-cell lymphomas. MYC arrives the same way in Burkitt lymphoma, where it needs a partner that supplies survival: TCF3 or ID3 mutation in 70% of sporadic cases and CCND3 mutation in 38%. BCL6 is rearranged in 28.7% of large B-cell lymphomas. EZH2 Tyr646 substitutions, 7.2% of follicular and 21.7% of germinal-centre large B-cell lymphoma, are gain-of-function and absent from the activated subtype. CREBBP or EP300 inactivation affects about 39% and 41%, and KMT2D 32% and 89%. TP53 loss is the commonest route out of chemotherapy sensitivity, 11% in mantle cell lymphoma with median overall survival of 1.8 years against 12.7, and 47.1% in Richter transformation.
• The biology in detail, 3 of 5. The immune lesions and the viruses. Classical Hodgkin lymphoma and primary mediastinal B-cell lymphoma amplify 9p24.1, which carries both PD-1 ligand genes and JAK2; 97% of 108 classical Hodgkin lymphomas had concordant alterations of the two ligand loci, and CIITA, the master regulator of MHC class II, is rearranged in 38% of mediastinal and 15% of Hodgkin cases. That is why PD-1 blockade produced responses in 20 of 23 heavily pre-treated Hodgkin patients and does very little elsewhere in B-cell lymphoma. The Hodgkin node is mostly not cancer: the Reed-Sternberg cells are a small minority, their clonal B-cell identity had to be proved by picking single cells off a slide with a micromanipulator, and the infiltrate carries prognostic information of its own, with more CD68-positive macrophages predicting worse outcome. Two viruses matter. Epstein-Barr virus persists in almost everyone and causes different lymphomas according to which latency programme the infected cell runs: EBNA1 alone in Burkitt lymphoma, LMP1 added in Hodgkin and NK/T-cell lymphoma, and the full set where T-cell surveillance has gone. HTLV-1 starts adult T-cell leukaemia/lymphoma with Tax and HBZ, and the 426-case genomic analysis showed the acquired mutations cluster in the pathways Tax itself touches.
• The biology in detail, 4 of 5. Resistance. Anti-CD20 fails by transcriptional down-regulation of CD20, found in 5 of 19 rebiopsied relapses, and by trogocytosis, in which macrophages shave the antigen off a living cell. Covalent BTK inhibitors fail on-target through BTK C481S, found in five of six patients with acquired resistance, which is answered by a non-covalent inhibitor giving 73.3% responses after a covalent one; they also fail through gain-of-function PLCG2 mutations below the kinase, which no BTK inhibitor reaches. Non-covalent inhibitors are in turn defeated by L528W and T474I, and L528W is enriched after zanubrutinib specifically, 7 of 13 progressing patients against 1 of 24 after ibrutinib. Venetoclax fails through BCL2 G101V in its own binding groove, which lowers affinity about 180-fold, appeared in 7 of 15 paired patients and is detectable months before progression. CD19 CAR-T fails through antigen loss by deletion, exon 2 mutation or exon 2 skipping, through the fitness of the collected T cells, where memory signatures predict response and exhaustion signatures predict failure, and through the node itself. Checkpoint blockade in Hodgkin lymphoma fails with loss of MHC class II, not class I, which is what points at a CD4 T cell as the effector.
• The biology in detail, 5 of 5. The tests. An excisional biopsy rather than a core where it is possible, because architecture decides several of these diagnoses and because in Hodgkin lymphoma the malignant cells are rare enough to be missed. An immunohistochemistry panel of 10 to 12 stains, from which the Hans algorithm reads cell of origin. FISH for MYC, with BCL2 and BCL6 where MYC is rearranged or MYC protein is high, for CCND1 in a suspected mantle cell lymphoma, for t(11;18) in a gastric MALT lymphoma before relying on Helicobacter eradication, and for 9p24.1 where a research question needs it. Flow cytometry where cells can be put in suspension, which is how Sezary cells are counted and marrow involvement measured. Immunoglobulin and T-cell receptor clonality where the morphology is equivocal, remembering that clonality is not malignancy. A sequencing panel where TP53, MYD88, CXCR4, EZH2, BTK, PLCG2, BCL2 or the T-follicular-helper genes would change the plan. And FDG-PET, scored 1 to 5 against the mediastinal blood pool and the liver, which is the measurement lymphoma treatment is actually adapted to, with the warning that a score of 3 means different things in different protocols.

The same record for scripts and assistants; checked 2026-09-29. Data CC BY-NC 4.0, attribute “Data from OnCo (onco.cc)”.