Oncofusion

start / graph / resistance-ras-flt3-wt1-comutation-menin

Resistancemoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q6b-therapeutics2

Signalling co-mutations (RAS pathway, FLT3-ITD, WT1) under menin-inhibitor pressure

also called RAS pathway resistance · FLT3-ITD menin resistance · WT1 menin resistance · bypass signalling

Even if a menin inhibitor successfully switches off the fusion's gene program, a leukemia cell that also carries a mutation in a growth-signalling gene like KRAS or FLT3 may keep growing anyway. These mutations are common in KMT2A-rearranged leukemia and they are the most obvious reason the drug can work and the patient still relapse.

The same thing, for a scientist

RAS pathway mutations (KRAS, NRAS, NF1, PTPN11) are the dominant co-mutation class in KMT2A-rearranged infant leukemia, and FLT3 wild-type overexpression is a defining feature of the genotype. In adults with relapsed/refractory AML failing menin inhibition, FLT3-ITD, WT1 and MEN1 mutations were the alterations associated with resistance, and no patient with FLT3-ITD responded to subsequent gilteritinib (0 of 6 gilteritinib-naive). In patient-derived xenografts of multiply pretreated KMT2A::AFF1 ALL, RAS pathway and TP53 co-mutations emerged alongside menin-inhibitor resistance but neither was sufficient to induce resistance in vitro, so the relationship is associative rather than proven causal. Dual menin plus FLT3 inhibition is already in phase 1 (NCT06222580, NCT06769490).

Go deeper
The clinical association

In 84 adults with relapsed/refractory AML after menin-inhibitor failure (63% KMT2A-rearranged), FLT3-ITD, WT1 and MEN1 mutations were associated with resistance 1. The outcomes after failure were grim: 40% received supportive care only, CR/CRi with non-trial therapy 19%, overall response 32%, and median overall survival from the start of next therapy 4.4 months (15.4 months if a CR/CRi was achieved, versus 3.4 months if not, p=0.048).

One specific and clinically important negative: no patient with FLT3-ITD responded to gilteritinib after menin-inhibitor failure (0 of 6 gilteritinib-naive) 1. That is a small number, but it directly contradicts the intuitive salvage plan of treating a FLT3-ITD-associated menin failure with a FLT3 inhibitor, and it argues that FLT3-ITD here is a marker of a resistant disease state rather than a targetable driver of the resistance.

The preclinical caution: association is not mechanism1 deeper

In serially sampled patient-derived xenografts from two patients with multiply relapsed KMT2A::AFF1 ALL, the resistant later samples carried emergent RAS pathway and TP53 mutations — but neither was sufficient to induce menin-inhibitor resistance when tested in vitro 2. The resistant cells also showed sustained on-target drug activity against canonical menin targets.

So the honest picture is: RAS pathway mutations travel with menin-inhibitor resistance without, on current evidence, causing it. They may be a marker of the subclonal diversity and prior treatment exposure that produce resistance by other routes. This distinction matters, because it determines whether adding a MEK inhibitor to a menin inhibitor is a rational combination or a toxicity for nothing.

Why this is more relevant in infants than the adult data suggest

RAS pathway mutations are the recurrent co-lesion in KMT2A-rearranged infant ALL, present in a substantial fraction at diagnosis and enriched at relapse; this graph covers them in detail under comut-ras-pathway-kmt2a-r, comut-kras-kmt2a-r, comut-nras-kmt2a-r, comut-nf1-kmt2a-r and comut-ptpn11-kmt2a-r. FLT3 is overexpressed as wild type in this genotype rather than ITD-mutated, which is a different pharmacology and means the adult FLT3-ITD observations do not transfer cleanly.

The practical prediction for the lab: if RAS co-mutation is a genuine cause of menin-inhibitor failure, infants should respond less well than the adult trials suggest, and the RAS-mutant subclone should expand under menin-inhibitor pressure. Both are directly measurable in the paediatric trials now enrolling (NCT05761171, NCT06376162, NCT05848687), and neither has been reported.

What is being done about it

Two phase 1 trials are testing dual menin plus FLT3 inhibition in FLT3-mutant AML with concurrent KMT2A rearrangement or NPM1 mutation: NCT06222580 (revumenib + gilteritinib, recruiting, n=30) and NCT06769490 (ziftomenib + quizartinib, recruiting, n=44). The ziftomenib combination study NCT06001788 also includes a gilteritinib arm 3.

There is no trial anywhere combining a menin inhibitor with a MEK or other RAS-pathway inhibitor in KMT2A-rearranged leukemia that we identified in this sweep. Given that RAS pathway mutation is the dominant co-lesion in the paediatric population, that is a conspicuous gap.

Facts and figures
kindresistance
mechanism classbypass signalling / co-mutation
seen in patientstrue
seen in modelstrue
causalityassociative in patients; RAS pathway and TP53 mutations were NOT sufficient to induce resistance in vitro, so a causal role is unproven
severityhigh in adults; predicted higher in infants because RAS pathway mutation is the dominant co-lesion in KMT2A-rearranged infant ALL
key negativeno patient with FLT3-ITD responded to gilteritinib after menin-inhibitor failure (0/6 gilteritinib-naive)
affects
[
  "drug-revumenib",
  "drug-ziftomenib",
  "drug-bleximenib",
  "drug-enzomenib",
  "drug-icovamenib"
]
genes
[
  "KRAS",
  "NRAS",
  "NF1",
  "PTPN11",
  "FLT3",
  "WT1",
  "TP53"
]
pre emption strategies
[
  "dual menin plus FLT3 inhibition (NCT06222580 revumenib+gilteritinib; NCT06769490 ziftomenib+quizartinib; NCT06001788 includes a gilteritinib arm)",
  "menin plus MEK or SHP2 inhibition — mechanistically obvious for the paediatric population and NOT currently in any trial we identified",
  "venetoclax, which is agnostic to upstream signalling and produced the only responses after menin failure in the retrospective cohort",
  "immunotherapy, which is entirely indifferent to RAS pathway status"
]
gaps
[
  "no menin + RAS-pathway-inhibitor trial identified in KMT2A-rearranged leukemia",
  "no paediatric data on co-mutation and menin response"
]
How this could bite us back
  • Bypass: a signalling mutation lets the cell grow even when the fusion program is successfully switched off, so on-target success and clinical failure coexist.
  • Combination toxicity: adding a FLT3 or MEK inhibitor to a menin inhibitor stacks QTc prolongation, myelosuppression and differentiation syndrome in an infant, with no paediatric dose-finding.
  • Essential-function toxicity: RAS pathway and FLT3 signalling are required for normal progenitor expansion; blocking them in an infant targets developmental haematopoiesis, and MEK inhibitors carry growth, cardiac and ocular toxicity in children.
  • Wrong-target risk: the preclinical data say RAS and TP53 mutations are not sufficient to cause resistance, so a menin plus MEK trial could deliver real toxicity against a non-causal association.
  • Salvage failure: the 0-of-6 gilteritinib result means the obvious rescue plan for FLT3-associated menin failure does not work, and there is currently no evidence-based alternative except venetoclax-containing regimens.
Still open
  1. Does the RAS-mutant subclone expand under menin-inhibitor pressure in patients? Serial variant-allele-frequency tracking on the paediatric menin trials would answer this directly and nobody has reported it.
  2. Should a MEK or SHP2 inhibitor be combined with a menin inhibitor in KMT2A-rearranged infant ALL? RAS pathway mutation is the dominant co-lesion in this population and no such trial exists.
  3. Why did no FLT3-ITD patient respond to gilteritinib after menin failure? If FLT3-ITD marks a globally resistant state rather than a driver, then the dual menin-FLT3 trials now running are testing the wrong hypothesis.
  4. Does wild-type FLT3 overexpression, the infant lesion, behave like FLT3-ITD with respect to menin-inhibitor resistance? The two are different pharmacological problems and the adult data should not be assumed to transfer.
Every source behind this entry (3)
  1. E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    In 84 adults with relapsed/refractory AML after menin-inhibitor failure, FLT3-ITD, WT1 and MEN1 mutations were associated with resistance; of 50 treated after failure the CR/CRi rate for non-trial therapies was 19% and overall response 32%, no patient with FLT3-ITD responded to gilteritinib (0 of 6 gilteritinib-naive), and median overall survival from the start of next therapy was 4.4 months, 15.4 months for those achieving CR/CRi versus 3.4 months otherwise (p=0.048).

    Outcomes of relapsed or refractory acute myeloid leukemia after menin inhibition failure · See PubMed record PMID 41351513 for the full author list (not transcribed here) · Blood Advances 2026
    PMID 41351513DOI
    What this evidence cannot show

    Retrospective, four US centres, adults only, heavily pretreated (86% prior intensive chemotherapy, 77% prior venetoclax). Mutation-resistance associations in a retrospective cohort are not causal and are confounded by the mutations' independent prognostic effect. The gilteritinib observation rests on 6 patients. No paediatric data, and FLT3-ITD is not the FLT3 lesion that characterises infant KMT2A-rearranged ALL.

    Evidence that argues against this (1)
    • Association in a retrospective post-failure cohort is not causation, and the direct test goes the other way. In serially sampled patient-derived xenografts from multiply relapsed KMT2A::AFF1 ALL, RAS-pathway and TP53 co-mutations emerged alongside reduced menin-inhibitor sensitivity, but neither alteration was SUFFICIENT to induce menin-inhibitor resistance when tested in vitro. Signalling co-mutations may therefore be markers of a heavily pretreated, genomically complex disease state rather than the mechanism of menin-inhibitor failure - which matters, because it predicts that adding a FLT3 or RAS-pathway inhibitor will not restore menin sensitivity. Mahdavi L et al. 2025
  2. E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    In serially sampled patient-derived xenografts from patients with multiply relapsed KMT2A::AFF1 ALL, later samples showed reduced menin-inhibitor sensitivity with emergence of RAS pathway and TP53 co-mutations, but neither alteration was sufficient to induce menin-inhibitor resistance in vitro.

    Upfront menin-inhibitor resistance in multiply pretreated leukemias · See PubMed record PMID 41027490 for the full author list (not transcribed here) · Experimental Hematology 2025
    PMID 41027490DOI
    What this evidence cannot show

    Serial samples from only two patients. A negative in vitro sufficiency test in a small number of engineered contexts does not exclude a contributory role in vivo, where microenvironmental ligands feed RAS signalling. The authors state that the drivers of upfront resistance need broader clarification.

    Evidence that argues against this (1)
    • 'Neither alteration was sufficient in vitro' is a statement about two ALL xenograft lines, and the clinical picture in myeloid disease is the opposite. In 1,750 patients enrolled on Children's Oncology Group trials, RAS-pathway mutations were found in 43% of AML cases, and in KMT2A-rearranged AML their presence was associated with a LOWER complete remission rate, poor event-free and overall survival, and early relapses. RAS-pathway co-mutation is therefore a clinically validated adverse modifier in KMT2A-r AML whether or not it is sufficient to confer menin-inhibitor resistance in a dish, and the in vitro insufficiency result should not be read as evidence that RAS status does not matter. Scheidegger N et al. 2026
  3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.trial_registry

    Dual menin plus FLT3 inhibition is being tested in recruiting phase 1 trials: revumenib plus gilteritinib in relapsed/refractory FLT3-mutated AML with concurrent KMT2A rearrangement or NPM1 mutation (NCT06222580, n=30), and ziftomenib plus quizartinib (NCT06769490, n=44).

    NCT06222580
    What this evidence cannot show

    Registry records only; both trials are adult, both are phase 1, neither has reported. FLT3-mutant AML is not the infant KMT2A-rearranged ALL setting, where FLT3 is overexpressed wild type rather than ITD-mutated.

    Evidence that argues against this (1)
    • Both dual-inhibition studies are small adult safety trials with no efficacy primary endpoint, so 'is being tested' does not mean 'will be answered'. Raw ClinicalTrials.gov v2 API, retrieved 2026-09-19: NCT06222580 (Ohio State) is phase 1, ESTIMATED n=30, minimum age 18 years, with primary outcomes of haematologic adverse-event incidence, non-haematologic adverse-event incidence and the recommended phase 2 dose, primary completion 2027-01-31. NCT06769490 (MD Anderson) is phase 1, ESTIMATED n=44, minimum age 18 years, with the SOLE primary outcome 'Safety and Adverse Events', primary completion 2028-09-15. Neither enrols children, and neither prespecifies response. The parallel preclinical strategy for the RAS half is also weaker than it looks: menin plus MEK inhibition reduced leukaemia burden in KMT2A-r RAS-mutant paediatric patient-derived xenografts but did NOT improve overall survival compared with menin inhibition alone. Scheidegger N et al. 2026
Connected to