Loading
Every cancer drug eventually meets resistance. For 15 major classes, the 52 known escape routes, sorted into eight kinds, how often they occur where that is known, and the countermeasures, linked to the products, targets, and ideas in the map.
15 drug classes · 52 documented escape routes · darker cells mean more routes of that kind. Click a row to jump to it, a column to follow one kind of escape across the page.
Lung cancers on osimertinib escape by mutating the drug's binding site, switching on a bypass receptor (MET), or changing cell type entirely.
Mutation of the cysteine that osimertinib binds covalently; abolishes drug binding while EGFR stays active.
Amplified MET signals to PI3K/MAPK independently of EGFR.
Conversion to small-cell lung cancer (RB1/TP53 co-loss) or squamous histology; EGFR mutation persists but the cell no longer depends on it.
Small subclones that already carry a resistance route expand under single-agent TKI; adding chemotherapy up front kills them before they take over.
Each ALK drug generation was beaten by a new mutation in the kinase; lorlatinib covers nearly all of them, so resistance now runs through other pathways.
Steric clash blocks first- and second-generation inhibitors; compound mutations (G1202R + L1196M etc.) emerge after lorlatinib.
Alternative receptors or downstream mutations re-activate MAPK/PI3K.
Resistance can be to the address (antigen) or to the poison (payload). Payload resistance is shared across every TOP1 ADC regardless of target, which is why a second one often fails.
TOP1 mutations (e.g., E418K) or reduced expression prevent trapping of the cleavage complex.
Schlafen-11 is required for replication-stress-induced death; its epigenetic silencing confers resistance to TOP1 (and platinum) agents.
SN-38 is an ABCG2 substrate; DXd and MMAE are ABCB1 substrates; mesenchymal states upregulate both.
Reduced HER2 or TROP2 surface expression after treatment; less frequent than payload resistance for HER2-low disease.
Defective endocytosis or lysosomal cathepsin activity limits payload release.
Most patients never respond (primary resistance) and some responders relapse (acquired). The routes are loss of antigen presentation, no T cells in the tumour, and a suppressive microenvironment.
Exhausted T cells co-express other inhibitory receptors.
Immunoediting removes the clones that carried immunogenic mutations.
Mutations in B2M or HLA class I stop tumour cells displaying antigen; JAK1/2 loss removes interferon responsiveness (and PD-L1 induction).
No pre-existing T-cell infiltrate (cold tumour) or T cells held at the margin by TGF-β and stroma.
MDSCs, M2 macrophages, and VEGF suppress T-cell function and dendritic-cell maturation.
Hormone-positive breast cancer escapes either by mutating the oestrogen receptor so it no longer needs oestrogen, or by rewiring the cell-cycle engine (RB loss, cyclin E) so CDK4/6 no longer matters.
Y537S/D538G render ER constitutively active; arise under aromatase-inhibitor pressure, detectable in ctDNA.
Without RB, CDK4/6 inhibition cannot arrest the cell cycle.
PIK3CA mutation, PTEN loss, or AKT1 E17K sustain growth independent of ER.
Tumours that lost BRCA can regain repair by re-mutating BRCA back into working order, or by finding another way to protect their DNA.
Secondary mutations restore the open reading frame and homologous recombination; also confers platinum resistance.
Loss of end-protection factors lets BRCA1-deficient cells resect DNA ends and repair by HR.
Olaparib and rucaparib are P-gp substrates.
Blocking one RAS mutant makes the cell turn up every upstream receptor and often mutate KRAS again; that is why responses are short and why combinations and pan-RAS drugs followed.
Alter the switch-II pocket or overwhelm the drug.
Relief of ERK-mediated negative feedback re-activates receptors within hours, producing new wild-type KRAS-GTP the drug cannot bind.
Alternative MAPK activation.
Myeloma escapes BCMA drugs by deleting or mutating the target, or by exhausting the T cells that were supposed to do the killing.
Deletion or mutation removes or alters the epitope; more common after bispecifics than CAR-T.
Prior lines, high tumour burden, and continuous bispecific dosing exhaust T cells; CAR-T products from heavily pretreated patients expand poorly.
Leukaemia and lymphoma relapse after CD19 CAR-T either without CD19 (the target is gone) or with it (the CAR-T cells are gone or exhausted).
Alternative splicing, mutation, or lineage switch (to myeloid) removes the CD19 epitope.
PD-1 upregulation, TGF-β, and myeloid suppression in lymphoma.
Limited expansion or early loss of CAR-T cells; 4-1BB products persist longer than CD28.
Prostate cancer keeps the androgen receptor working without hormones (amplification, splice variants), or abandons it and becomes a neuroendocrine cancer.
More receptor, or mutations (F877L, T878A) that turn antagonists into agonists.
Truncated receptor lacking the ligand-binding domain is constitutively active and invisible to enzalutamide.
Reciprocal feedback between AR and PI3K pathways.
GR drives an AR-like transcriptional programme under enzalutamide.
RB1/TP53 loss enables transdifferentiation; AR-indifferent, DLL3-positive, PSMA-negative.
B-cell lymphomas escape rituximab either by stopping making CD20, which a report will only show if the relapse is rebiopsied, or by exhausting the complement and effector cells the antibody has to borrow in order to kill.
Transcriptional down-regulation of MS4A1 rather than deletion: CD20 messenger RNA is lower in the negative cells than in the positive cells from the same patient.
Macrophages strip antibody-CD20 complexes off a living cell, lowering surface antigen without any change to the gene.
The antibody kills by borrowing complement, natural killer cells and macrophages; repeated dosing depletes complement locally and leaves the effectors refractory, so retreatment works less well even where the antigen is intact.
Covalent BTK inhibitors are usually defeated by a change at the single cysteine they grip. A non-covalent inhibitor answers that, and is in turn defeated by changes elsewhere in the pocket; a mutation below the kinase is beyond all of them.
Substituting the cysteine the drug binds covalently leaves the kinase active and makes inhibition reversible, so the drug no longer holds.
The kinase-dead L528W substitution and the gatekeeper T474I change the pocket rather than the covalent cysteine, so they defeat reversible inhibitors too.
R665W and L845F make B-cell receptor signalling autonomous below the kinase, so blocking BTK above them achieves nothing.
Venetoclax is escaped either by a change in the groove it has to fit into, which can be seen in blood months before the disease grows again, or by the cell switching its dependence to BCL-2's relatives.
The valine substitution at glycine 101 reduces BCL-2 affinity for venetoclax about 180-fold, so the drug can no longer displace the pro-apoptotic proteins while BCL-2 keeps working.
A cell that needed BCL-2 can be made to need its relatives instead, which is the main reason venetoclax has not repeated its chronic lymphocytic leukaemia result in follicular and large B-cell lymphoma.
Hodgkin lymphoma answers PD-1 blockade because it has genetically amplified the brake. When it stops answering, the usual explanation is not the brake but what the Reed-Sternberg cell presents to a T cell.
Hodgkin lymphoma responds despite deficient MHC class I. Across the CheckMate 205 biopsies, class I and beta-2-microglobulin did not predict complete remission but class II did, which points at a CD4 T cell as the effector and makes class II loss the escape route.
Fusions of the master transactivator of MHC class II lower class II on the tumour cell and in the same event place the PD-1 ligands under new promoters, so one rearrangement both hides the tumour and raises the brake.
The alteration is near-universal but graded, and higher-level copy gain predicted longer progression-free survival, so the patients with least of it do least well.
A bispecific needs the target on the tumour and a working T cell in the lymph node, so it is exposed to both failures at once: the antigen can go, and the patient's T cells can be worn out by the treatments that came before.
The CD20 arm fails for the same reasons rituximab does: transcriptional down-regulation and shaving.
Suppressive myeloid cells, PD-L1 on the tumour and the stroma, and fibrosis can switch off a synapse the drug has successfully formed.
Patients who have had several lines of treatment, and especially those who have had CD19 CAR-T, bring a depleted and exhausted T-cell compartment to a drug that is entirely dependent on it.