Oncofusion

start / graph / concept-why-bcr-abl1-became-curable

Explainerestablished confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.Q11-comparative

Why BCR::ABL1 became treatable — and what KMT2A::AFF1 does not share with it

also called imatinib lesson · kinase paradigm · CML paradigm

Chronic myeloid leukaemia became a manageable disease because its fusion protein is an enzyme with a deep chemical slot that a pill can block. The same pill barely helped when it was given alone to patients whose disease had already relapsed - but given early, alongside chemotherapy, to children whose acute leukaemia carried the same fusion, it more than doubled cure. KMT2A::AFF1 has no such slot, so the drug cannot be copied; the lesson that does copy is about when in the illness you give it.

The same thing, for a scientist

BCR::ABL1 is the archetype of targeted therapy, but its tractability rests on a single structural property: the fusion retains an intact tyrosine kinase catalytic domain whose ATP-binding cleft is deep, enclosed, hydrophobic and conformationally exploitable. Imatinib occupies that canonical ATP site and traps a specific inactive activation-loop conformation, giving both potency and selectivity. The clinical result is long-term survival approaching that of the general population in chronic-phase CML, though 'cure' is imprecise: most patients require indefinite therapy, resistance arises by point mutation at the drug-binding surface, and quiescent leukaemic stem cells persist. Two qualifiers belong in the headline. First, every impressive SINGLE-AGENT number is a chronic-phase number: against the same fusion in myeloid blast crisis imatinib gave median survival 6.9 months, and in relapsed Philadelphia-positive ACUTE LYMPHOBLASTIC leukaemia 4.9 months. Those are monotherapy-at-relapse numbers. Given frontline with intensive chemotherapy to newly diagnosed children with Philadelphia-positive ALL, the same drug against the same fusion produced a 3-year event-free survival of 80% against 35% in historical controls, and a 5-year disease-free survival of 70% that matched allogeneic transplant (counter-evidence on [E7]). The qualifier is line of therapy and combination, not disease acuity. Second, what rescued the field from interface resistance was not more chemotypes at the ATP site but an orthogonal allosteric site, the ABL1 myristoyl pocket, now registered as asciminib. KMT2A::AFF1 resembles the acute case, and has no candidate orthogonal site. KMT2A::AFF1 contains no catalytic domain — the KMT2A SET methyltransferase is always lost at the breakpoint — so no analogous active site exists on the fusion. What does transfer is the *method*: define the single non-negotiable molecular function, find a structured surface that executes it, and design against that surface.

Go deeper
Feature 1 — the oncogenic function is catalysis in an enclosed pocket1 deeper

BCR contributes a coiled-coil that forces ABL1 kinase domains together; they trans-autophosphorylate without a ligand and signal continuously. Everything downstream flows from catalysis, so inhibiting catalysis inhibits the disease completely. Crystal structures show imatinib bound in the canonical ATP-binding site of the ABL1 kinase domain, capturing a specific inactive conformation of the activation loop, while a chemically distinct inhibitor (PD173955) binds an active-like conformation of the same site 1. Two conformations, two chemotypes, one pocket — that is the definition of a rich drug target.

The clinical consequence was immediate and enormous: in the phase 1 trial, 53 of 54 interferon-refractory chronic-phase CML patients dosed at 300 mg or above achieved complete haematologic response, mostly within four weeks, with minimal toxicity and no maximum tolerated dose reached 2. At a median 10.9 years of follow-up in the randomised IRIS trial, estimated overall survival for imatinib-assigned patients was 83.3% with 82.8% achieving complete cytogenetic response 3.

Every one of those numbers is a chronic-phase number, and that qualifier is load-bearing. Run the same drug against the same fusion in acute disease and it nearly vanishes. In myeloid blast crisis, imatinib gave sustained haematologic response in 31%, major cytogenetic response in 16%, and median survival 6.9 months 6. In relapsed/refractory Ph-positive acute lymphoblastic leukaemia -- the disease class KMT2A::AFF1 actually belongs to -- complete responses were 29% but sustained in only 6%, median time to progression 2.2 months, median overall survival 4.9 months 7. So the correct reading of the CML precedent is not 'a deep pocket cures a fusion leukaemia'. It is: a deep pocket controls an indolent, single-driver, chronic-phase myeloproliferative disease, and does much less against an acute leukaemia carrying the same fusion. Infant KMT2A-r ALL is an acute leukaemia.

Is it really cure? Be precise, because the precision matters strategically

CML is controlled, and for a subset genuinely cured: treatment-free remission is achievable in a minority after prolonged deep molecular response. But the honest framing is that imatinib converted a fatal disease into one requiring lifelong therapy, with resistance and stem-cell persistence as the two unsolved problems 4. That distinction matters here because APL is the fusion that is cured by removing the oncoprotein, whereas CML is the fusion that is controlled by occupying the oncoprotein. The KMT2A programme is currently on the CML side of that line — menin inhibitors occupy — and the APL comparison (see concept-why-pml-rara-became-curable) argues that is not enough.

Feature 2 — the pocket is also the escape hatch

Because drug and substrate share the site, single amino-acid substitutions abolish drug binding while preserving catalysis. Clinical resistance to imatinib was traced within a couple of years to BCR-ABL kinase-domain point mutations and gene amplification 4. The field survived this only because an enzyme offers more than one site: second-generation ATP-competitive inhibitors, then a third-generation agent for T315I, and finally an allosteric myristoyl-pocket drug. That last step is the one worth naming precisely, because it is the manoeuvre KMT2A cannot copy. ABL001/asciminib does not bind the ATP site at all; it occupies the myristoyl pocket, the autoinhibitory site that BCR-ABL1's fusion architecture disables, and combining a myristoyl-pocket agent with an ATP-site agent suppresses resistance that neither controls alone 9. In ASCEMBL, asciminib beat bosutinib in third-line chronic-phase CML (96-week major molecular response 37.6% vs 15.8%) 9. CML was rescued by a second, orthogonal, functionally distinct pocket on the same oncoprotein. Menin has one.

Compare directly with menin inhibition. There, resistance also arose at the drug-binding surface — MEN1 mutations that block revumenib but preserve the menin-MLL1 interaction 5. The mechanism of escape is structurally identical to the CML story. The difference is the depth of the bench. BCR::ABL1 had five approved chemotypes to fall back on; menin has one validated pocket and a handful of chemically related molecules, all of which are likely vulnerable to overlapping interface mutations. That asymmetry, not the biology, is the thing that should worry the lab most.

Does KMT2A::AFF1 share anything? A line-by-line answer
Feature that made CML treatablePresent in KMT2A::AFF1?
Retained catalytic domain with an ATP pocketNo. KMT2A contributes AT-hooks, CXXC and the menin/LEDGF docking motifs; the SET methyltransferase is lost. AFF1 is a scaffold with no catalytic activity.
Single non-redundant biochemical function that the whole phenotype depends onApproximately yes — recruitment of an elongation platform to HOXA/MEIS1 — but it is executed by an assembly, not by one active site.
Multiple chemotypes and multiple sites on the same targetNo. One validated pocket (menin), one chemical series. CML's decisive rescue was not a new chemotype at the old site but an orthogonal allosteric site -- the ABL1 myristoyl pocket, now a registered drug 9. No allosteric second site on menin has been described.
Partner-supplied forced oligomerisation as the activating eventUnclear. BCR, NPM1 and ETV6 all activate their kinases by oligomerisation; whether AFF1 does anything analogous is untested and is written up as a hypothesis.
Resistance by point mutation at the drug-binding surfaceYes, already observed. This is the one feature KMT2A-r has fully inherited, and it arrived faster than it did in CML.
An enzyme in the complex rather than in the fusionYes — CDK9 in the SEC, and DOT1L. This is the nearest available kinase-grade target, and both have been drugged with disappointing selectivity windows.

| Chronic, indolent disease biology giving a long therapeutic window | No. All of the CML numbers quoted here are chronic-phase. Against the same fusion in acute disease, imatinib gave median survival 6.9 months (myeloid blast crisis 6) and 4.9 months (relapsed Ph+ ALL 7). |

The honest conclusion: KMT2A::AFF1 has inherited CML's resistance mechanism without inheriting CML's druggable pocket, without CML's orthogonal second site, and without CML's chronic-phase therapeutic window. That is the worst of both worlds and it argues for (i) building chemical diversity against the menin interface before resistance becomes routine, and (ii) taking seriously the non-occupancy modalities — degradation, immunotherapy against the junction, and interruption of the assembly at a second point.

Correction (cycle 33): three things this entry originally let the analogy get away with

  1. The phase qualifier was missing from the headline numbers. 'BCR::ABL1 became treatable' is true of chronic-phase CML. In myeloid blast crisis the same drug gave median survival 6.9 months 6; in relapsed/refractory Ph-positive acute lymphoblastic leukaemia, 4.9 months 7. The disease KMT2A::AFF1 resembles is the second one, and the precedent there is bad. The entry's summary and data now say so.

  2. 'Depth of the bench' understated the asymmetry. The rescue was not more chemotypes at the ATP site; it was an orthogonal allosteric site (the ABL1 myristoyl pocket) that made dual targeting possible and is now registered 9. Menin has one validated site and no described allosteric analogue. The right research question is not 'build chemical diversity at the menin interface' but 'is there a second site anywhere in the KMT2A-fusion assembly that is orthogonal to the menin pocket' -- a strictly harder question.

  3. 'Cure' was already hedged, and should be hedged further. Among 755 patients pre-selected as the best molecular responders, half had lost major molecular response 24 months after stopping 8. Treatment-free remission is a minority outcome within a minority.

What survives intact: the transferable lesson is the method (find the one non-negotiable function and design against a structured surface that executes it), and the observation that KMT2A-r inherited the resistance mechanism without the pocket.

Correction 2026-09-19 — the acuity framing was wrong, and the mistake matters for the lab's own trial design

This entry previously said, in its lay summary, that "the same pill barely helped patients whose disease had turned acute", and in its bite-back that "Matched to disease acuity, the honest CML comparator for infant KMT2A-r ALL is relapsed Ph-positive ALL on imatinib monotherapy: 6% sustained responses and 4.9 months median survival." Both are withdrawn as stated.

They matched on acuity and silently also matched on line of therapy and on monotherapy. When only acuity is matched - newly diagnosed paediatric Philadelphia-positive ALL, imatinib added to intensive chemotherapy - the result was a 3-year event-free survival of 80% +/- 11% against 35% +/- 4% in historical controls treated with chemotherapy alone, with no appreciable added toxicity; at five years the disease-free survival was 70% +/- 12%, indistinguishable from sibling-donor (65%) or unrelated-donor (59%) transplant, and transplant lost its advantage. Ph-positive ALL went from the worst-prognosis paediatric ALL subgroup to one treated without transplant, by adding a targeted agent to frontline chemotherapy.

Why this is the most useful correction in this entry. The comparison the lab needs is not 'chronic disease drug versus acute paediatric leukaemia'. It is 'targeted agent plus chemotherapy, frontline, in newly diagnosed acute paediatric leukaemia' - and in the one case where that has been run to maturity it doubled cure. Menin inhibitors are currently tested almost entirely in the relapsed/refractory setting, which is precisely the setting in which imatinib looked ineffective in acute disease for the best part of a decade. Recorded here so the analogy is not quoted again in its withdrawn form.

Facts and figures
comparison axiswhat made a fusion oncoprotein directly druggable
fusionfusion-bcr-abl1
cure mechanismindefinite occupancy of the fusion active site; genuine treatment-free remission in only a minority
transferable lessonidentify the one non-negotiable molecular function and build chemical diversity against it BEFORE resistance becomes routine
non transferablethe ATP pocket itself — KMT2A fusions retain no catalytic domain
implication for kmt2aKMT2A-r has inherited the resistance mechanism without the pocket, without an orthogonal second site, and without the chronic-phase window: the CML numbers that make the analogy attractive are all chronic-phase, and against acute leukaemia carrying the same fusion imatinib gave median survival of 4.9-6.9 months. Chemical diversity at the menin interface is necessary but the CML rescue actually came from a second, functionally orthogonal site, which KMT2A has no candidate for.
tractability features
[
  {
    "feature": "retained catalytic domain with an enclosed ATP pocket",
    "present": true,
    "drug": "imatinib",
    "shared_with_kmt2a_aff1": false
  },
  {
    "feature": "multiple chemotypes and multiple exploitable sites on one target",
    "present": true,
    "drug": "dasatinib, nilotinib, ponatinib, asciminib",
    "shared_with_kmt2a_aff1": false
  },
  {
    "feature": "partner-supplied forced oligomerisation as the activating event",
    "present": true,
    "drug": "none directly",
    "shared_with_kmt2a_aff1": null
  },
  {
    "feature": "single non-redundant biochemical function driving the whole phenotype",
    "present": true,
    "drug": "n/a",
    "shared_with_kmt2a_aff1": true
  },
  {
    "feature": "resistance by point mutation at the drug-binding surface",
    "present": true,
    "drug": "n/a",
    "shared_with_kmt2a_aff1": true
  },
  {
    "feature": "chronic-phase disease biology giving a long therapeutic window",
    "present": true,
    "drug": "n/a",
    "shared_with_kmt2a_aff1": false
  },
  {
    "feature": "an orthogonal allosteric site on the same oncoprotein",
    "present": true,
    "drug": "asciminib",
    "shared_with_kmt2a_aff1": false
  }
]
nearest enzyme in the complex
[
  "gene-cdk9",
  "gene-dot1l"
]
phase dependence of the precedent
{
  "chronic_phase_10y_os_pct": 83.3,
  "myeloid_blast_crisis_median_survival_months": 6.9,
  "myeloid_blast_crisis_major_cytogenetic_response_pct": 16,
  "relapsed_ph_positive_all_median_os_months": 4.9,
  "relapsed_ph_positive_all_sustained_response_pct": 6,
  "pmids": [
    "28273028",
    "11986204",
    "12200353"
  ],
  "implication": "the CML precedent is a precedent about indolent chronic-phase disease; against acute leukaemia carrying the same fusion the same drug is close to palliative"
}
treatment free remission is a minority within a minority
{
  "cohort": "EURO-SKI, 755 chronic-phase patients pre-selected for >=3 y TKI and >=1 y deep molecular response",
  "molecular_relapse_free_survival_6m_pct": 61,
  "molecular_relapse_free_survival_24m_pct": 50,
  "pmid": "29735299",
  "nct": "NCT01596114",
  "title": "Discontinuation of tyrosine kinase inhibitor therapy in chronic myeloid leukaemia (EURO-SKI): a prespecified interim analysis of a prospective, multicentre, non-randomised, trial.",
  "authors": "Saussele S, Richter J, Guilhot J, et al.",
  "venue": "Lancet Oncol",
  "year": 2018,
  "title_provenance": "Title/authors/venue backfilled from PubMed esummary on 2026-09-20 because this citation stored a bare PMID with no title, which made it unscoreable by verify-citations.py. BACKFILLING IS NOT VERIFICATION: it confirms the PMID resolves to a real paper, not that this is the right paper for the claim. That still requires reading."
}
the rescue was an orthogonal site not a new chemotype
{
  "site": "ABL1 myristoyl pocket (allosteric, autoinhibitory) -- asciminib/ABL001",
  "evidence": "dual targeting of myristoyl and ATP sites suppresses resistance neither controls alone; ASCEMBL 96-week MMR 37.6% vs bosutinib 15.8%",
  "pmids": [
    "28329763",
    "36717654"
  ],
  "kmt2a_analogue": "none described; menin has one validated pocket and no reported allosteric second site"
}
How this could bite us back
  • The KMT2A programme has inherited CML-style interface resistance without CML-style chemical redundancy, so the first resistant clones may exhaust the entire menin-inhibitor class at once.
  • Occupancy-based control implies indefinite therapy in infants and young children, with unknown effects on growth, endocrine function and the developing marrow.
  • The CML analogy encourages the assumption that a single molecular driver explains the disease; in KMT2A-r infant ALL the cooperating RAS-pathway lesions and the fetal cell-of-origin context may set a ceiling no fusion-directed drug can pass.
  • The chronic-phase framing is the single most flattering thing in this analogy, and the relapsed-monotherapy framing is the mirror-image error. Matched on acuity AND line of therapy, the CML comparator splits: 6% sustained responses and 4.9 months median survival for single-agent imatinib in relapsed Ph-positive ALL, and 3-year EFS of 80% for imatinib added to frontline chemotherapy in newly diagnosed paediatric Ph-positive ALL. Which number is honest depends on the setting being compared, and the second is the one that describes how a menin inhibitor would actually be deployed if it worked. (Corrected 2026-09-19 against PMID 19805687; this item previously quoted only the relapsed-monotherapy figures.)
  • Treatment-free remission, the nearest thing to cure in this story, has no paediatric evidence base: EURO-SKI required age 18 or older, at least three years of TKI and a confirmed deep molecular response, and excluded transplant. Any claim that a KMT2A-r child might one day stop a menin inhibitor is extrapolating from a trial that would not have enrolled them.
Still open
  1. Is there a second, chemically distinct or allosteric site on menin -- an analogue of the ABL1 myristoyl pocket -- that would give the KMT2A field the orthogonal target that actually rescued the CML field? No such site has been described.
  2. Does the KMT2A fusion depend on any obligate self-association analogous to the BCR coiled-coil, which would be a mutation-resistant target because mutating it would inactivate the oncoprotein?
  3. Given that paediatric Ph-positive ALL is far less well controlled by kinase inhibition than adult chronic-phase CML, how much of the CML success is about the drug and how much about the indolent, single-driver biology of chronic phase?
  4. Should the lab pre-emptively build a resistance-mutation map of the menin pocket (base-editor or deep mutational scan) before resistance is widespread, as the CML field wishes it had done?
  5. The kinase precedent in acute paediatric leukaemia is a frontline targeted-agent-plus-chemotherapy combination, not salvage monotherapy. What is the earliest line in which a menin inhibitor has been tested in newly diagnosed KMT2A-rearranged infant ALL, and is the field repeating the relapsed-monotherapy design that made imatinib look ineffective in acute disease for a decade?
Every source behind this entry (9)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    Imatinib binds the canonical ATP-binding site of the ABL1 kinase domain and captures a specific inactive activation-loop conformation, while a distinct inhibitor binds an active-like conformation of the same site.

    PMID 12154025
    What this evidence cannot show

    Structures of the isolated ABL1 kinase domain rather than the intact BCR::ABL1 fusion, so conformational consequences of BCR-driven oligomerisation are not represented.

    Evidence that argues against this (2)
    • The structural story is usually told as the explanation for imatinib's selectivity, and NMR plus fast-kinetics measurement says it is not. Abl and Src adopt almost identical structures when bound to imatinib but have vastly different affinities; the origin of Abl's high affinity lies predominantly in a conformational change AFTER binding, an induced fit, rather than in selection of the inactive activation-loop conformation seen in the crystal. Every conformational-selection model of the drug's specificity was tested against this and failed. For a 'why this one became curable' narrative that matters: the crystallographic fit is the part of the story that is easiest to tell and is not the part doing the work. Agafonov RV et al. 2014
    • What is in the structure is less than the narrative implies. PDB 1IEP, the primary deposition of the cited paper, is a 2.1 A X-ray structure of a SINGLE polymer chain of 293 residues, and SIFTS maps it to UniProt P00520 - ABL1_MOUSE, mouse Abl1, 1123 residues - starting at reference residue 229. No BCR moiety, no SH3-SH2 regulatory cap, no myristoylated C-lobe, and not human. The curability narrative rests on a structure of roughly a quarter of the murine partner protein, which is fine for what the paper claimed and is worth naming in an entry that uses it to explain a clinical outcome. Checked directly against the RCSB entry record on 2026-09-19. Nagar B et al. 2002
  2. E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    In the first-in-human imatinib trial, 53 of 54 interferon-refractory chronic-phase CML patients treated at 300 mg per day or more achieved complete haematologic response, typically within four weeks, with no maximum tolerated dose identified.

    Efficacy and safety of a specific inhibitor of the BCR-ABL tyrosine kinase in chronic myeloid leukemia · Druker BJ, Talpaz M, Resta DJ, et al. · New England Journal of Medicine 2001
    PMID 11287972DOI
    What this evidence cannot show

    Phase 1 dose escalation, n=83, no control arm, short follow-up, and only 7 complete cytogenetic remissions at that point; haematologic response is a shallow endpoint.

    Evidence that argues against this (2)
    • The imatinib result is a result about CHRONIC PHASE, and the entry's framing borrows its authority for acute disease. In the parallel phase 2 in myeloid blast crisis (n=229 confirmed CML-BC), imatinib gave haematologic responses in 52% but sustained responses in only 31%, complete haematologic response in 8%, major cytogenetic response in 16%, and MEDIAN SURVIVAL OF 6.9 MONTHS. Same fusion, same drug, acute biology: the pocket did not rescue it. Sawyers CL et al. 2002
    • Closer still to the paediatric target disease: in relapsed/refractory Philadelphia-positive ACUTE LYMPHOID leukaemia, imatinib gave complete haematologic or complete marrow responses in 29% of ALL patients but sustained for at least four weeks in only 6%, with median time to progression 2.2 months and MEDIAN OVERALL SURVIVAL 4.9 MONTHS. BCR::ABL1 in an acute lymphoblastic context behaves nothing like BCR::ABL1 in chronic phase, and it is the lymphoblastic case that KMT2A::AFF1 resembles. Ottmann OG et al. 2002
  3. E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    At a median follow-up of 10.9 years, estimated overall survival among imatinib-assigned patients in the IRIS trial was 83.3%, with 82.8% achieving complete cytogenetic response and no unacceptable late toxicity.

    Long-term outcomes of imatinib treatment for chronic myeloid leukemia · Hochhaus A, Larson RA, Guilhot F, et al. · New England Journal of Medicine 2017
    PMID 28273028DOI
    What this evidence cannot show

    65.6% of the comparator arm crossed over so the randomised comparison is effectively lost; only 48.3% of imatinib-assigned patients completed study treatment on imatinib, and this is adult chronic-phase CML rather than paediatric Ph-positive ALL.

    Evidence that argues against this (1)
    • 'Cure' should be discounted further. In EURO-SKI, 755 chronic-phase patients pre-selected for at least three years of TKI and at least one year of confirmed deep molecular response -- a small and favourable subset of all CML patients -- had molecular relapse-free survival of 61% at 6 months and only 50% at 24 months after stopping. Half of the best-responding half relapse molecularly on withdrawal, which is what indefinite occupancy looks like. Saussele S et al. 2018
  4. E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    Clinical resistance to imatinib arises from BCR-ABL kinase-domain point mutation or gene amplification, i.e. escape at the drug-binding surface with the oncogenic mechanism intact.

    PMID 11423618DOI
    What this evidence cannot show

    A small early series of relapsed patients; the relative weight of kinase-domain mutation versus BCR-ABL-independent resistance and quiescent stem-cell persistence emerged only in later work.

    Evidence that argues against this (1)
    • The rescue that saved the CML field from interface resistance was not another ATP-site chemotype but a structurally different site: ABL001/asciminib binds the ABL1 myristoyl pocket, an allosteric autoinhibitory site, and combining it with an ATP-site inhibitor suppressed resistance that either agent alone could not. The strategic lesson is therefore narrower and worse for KMT2A than the entry implied -- CML was rescued by a SECOND, ORTHOGONAL SITE ON THE SAME PROTEIN, and the KMT2A field has no candidate for such a site. Wylie AA et al. 2017
  5. E5strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Resistance to menin inhibition in KMT2A-rearranged leukaemia also arises from mutations at the drug-target interface that spare the biological menin-MLL1 interaction, structurally mirroring the BCR-ABL kinase-domain mutation paradigm.

    MEN1 mutations mediate clinical resistance to menin inhibition · Perner F, Stein EM, Wenge DV, et al. · Nature 2023
    PMID 36922589DOI
    What this evidence cannot show

    Few patients with paired pre- and post-treatment samples; whether interface mutation is the dominant clinical resistance route or one of several is not yet established.

    Evidence that argues against this (1)
    • And asciminib is now a registered standard of care in third line (ASCEMBL, 96-week MMR 37.6% vs 15.8% for bosutinib), which is the concrete measure of how much a second orthogonal site is worth. Menin has one validated pocket and no allosteric analogue, so the KMT2A programme cannot execute the manoeuvre that actually rescued CML. Hochhaus A et al. 2023
  6. E6established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    In CML myeloid blast crisis, imatinib gave sustained haematologic responses in 31%, major cytogenetic responses in 16% and a median survival of 6.9 months.

    PMID 11986204DOI
    What this evidence cannot show

    Abstract only. Single-arm phase 2, 229 confirmed blast-crisis patients, 2002, no comparator and no modern transplant/second-generation-TKI context. The point stands regardless of era: the same drug against the same fusion loses most of its effect when the disease is acute.

    Evidence that argues against this (1)
    • The blast-crisis figure is monotherapy in patients who had already failed everything available in 2002, and the entry uses it to support a general statement about acute disease. Against the same fusion in acute disease, given the way targeted agents are actually deployed - frontline, added to intensive chemotherapy, in newly diagnosed patients - imatinib produced a 3-year event-free survival of 80% +/- 11% in children and adolescents with Philadelphia-positive ALL against 35% +/- 4% in historical controls (p < .0001), with no significant additional toxicity. The correct generalisation is about line of therapy and combination, not about acuity. Schultz KR et al. 2009
  7. E7established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    In relapsed or refractory Philadelphia-positive acute lymphoblastic leukaemia, imatinib monotherapy gave complete haematologic or marrow responses in 29% of patients but sustained responses in only 6%, with median time to progression 2.2 months and median overall survival 4.9 months.

    PMID 12200353DOI
    What this evidence cannot show

    Abstract only. n=48 Ph+ ALL plus 8 lymphoid blast crisis, adults, relapsed/refractory, monotherapy, 2002; modern Ph+ ALL outcomes with TKI plus chemotherapy or blinatumomab are far better. But that is the point: the pocket alone was not sufficient in acute lymphoblastic disease, which is the disease class the KMT2A analogy actually needs.

    Evidence that argues against this (2)
    • This is the same disease in which the frontline result exists, which makes the contrast decisive rather than analogical. Relapsed or refractory Ph-positive ALL on imatinib monotherapy gave 6% sustained responses; newly diagnosed paediatric Ph-positive ALL on imatinib plus intensive chemotherapy gave 3-year EFS of 80%. On longer follow-up the same cohort had 5-year disease-free survival of 70% +/- 12%, statistically indistinguishable from sibling-donor (65% +/- 11%) and unrelated-donor (59% +/- 15%) transplant, and the authors concluded transplant offered no advantage. Quoting the 6% without the 80% makes the kinase precedent look far less relevant to paediatric acute leukaemia than it is. Schultz KR et al. 2009
    • The frontline benefit did not decay with follow-up, which removes the standard objection that an early EFS figure is an artefact of timepoint. Five-year disease-free survival was 70% +/- 12% for chemotherapy plus imatinib, transplant conferred no advantage, and patients with additional cytogenetic abnormalities did worse (p = 0.05) - so the residual failures are traceable to co-lesions rather than to the fusion escaping the drug. Schultz KR et al. 2014
  8. E8established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    In 755 highly selected chronic-phase CML patients who stopped TKI after at least 3 years of therapy and at least 1 year of deep molecular response, molecular relapse-free survival was 61% at 6 months and 50% at 24 months.

    PMID 29735299DOINCT01596114
    What this evidence cannot show

    Abstract only. Prospective non-randomised (NCT01596114), prespecified interim analysis, eligibility already restricted to the best responders, and loss of major molecular response is a surrogate rather than relapse to overt disease -- most patients regain response on re-treatment.

    Evidence that argues against this (1)
    • Treatment-free remission is the closest thing this narrative has to cure, and its denominator is the thing to look at. The EURO-SKI registry record gives actual enrolment of 868 and an inclusion list that is only partly biological: chronic phase, at least three years of TKI, confirmed MR4 through a EUTOS-CMR laboratory, no previous or planned allogeneic transplant - and also age 18 or older and health insurance coverage. Two consequences for a paediatric-leukaemia graph reasoning from this precedent. First, the 50% figure applies to a population defined by having already done well for years on a drug, which is a survivorship-selected group, not the CML population. Second, the trial excludes children by protocol, so there is no paediatric TFR evidence here at all. Checked against the ClinicalTrials.gov v2 API on 2026-09-19. EURO-SKI investigators 2012
  9. E9established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    The escape from ATP-site resistance in CML came from a structurally distinct allosteric site: ABL001/asciminib binds the ABL1 myristoyl pocket, and dual targeting of the myristoyl and ATP sites suppresses resistance that neither achieves alone; asciminib is now registered and outperformed bosutinib in third-line chronic-phase CML (96-week MMR 37.6% vs 15.8%).

    The allosteric inhibitor ABL001 enables dual targeting of BCR-ABL1. · Wylie AA, Schoepfer J, Jahnke W, et al. · Nature 2017
    PMID 28329763DOI
    What this evidence cannot show

    Abstract only for both sources. Industry-sponsored discovery paper plus an open-label randomised trial with a weak comparator; the clinical claim rests on ASCEMBL (PMID 36717654), a molecular-response endpoint rather than survival. The structural claim -- a second, orthogonal, autoinhibitory pocket on the same oncoprotein -- is not in dispute.

    Evidence that argues against this (2)
    • The allosteric site moved the problem rather than solving it, and the randomised data give its size. In ASCEMBL, after two or more prior TKIs, asciminib's major molecular response rate at week 96 was 37.6% against bosutinib's 15.8% - better, and still a minority of patients - with grade >=3 adverse events in 56.4%. A second site bought about 22 percentage points in a heavily pretreated population. Hochhaus A et al. 2023
    • And the myristoyl pocket is itself a drug-binding surface that mutates. Clinical asciminib resistance runs through mutations in the pocket (A337V/T, I502L), through kinase-domain mutations that break the autoinhibited conformation asciminib requires (F359C/I/V at the kinase-SH2 interface, M244V in the N-lobe), and through compound mutations involving T315I. Cited as a map of the reported resistance mutations rather than as primary evidence for any one of them. Costa A et al. 2025
Connected to