Icovamenib (BMF-219) — the covalent menin inhibitor that left oncology for diabetes
also called BMF-219 · BMF219
Icovamenib is a menin blocker that latches on permanently (covalently) rather than reversibly. Its leukaemia trials have been stopped, and the company has redirected the drug to type 1 and type 2 diabetes — because blocking menin makes insulin-producing cells grow, which is exactly the kind of side effect that should worry anyone planning years of menin blockade in a child.
The same thing, for a scientist
Icovamenib (BMF-219, Biomea Fusion) is an orally administered irreversible/covalent inhibitor of menin. Its oncology programmes are terminated: COVALENT-101 in acute leukaemia/DLBCL/MM/CLL (NCT05153330, n=55, terminated, primary completion 2025-02-13) and the solid-tumour COVALENT-103 study in KRAS-mutant NSCLC/PDAC/CRC (NCT05631574, n=13, terminated). A published case report documented a complete remission in NUP98-fusion-positive AML. The active development programme is now entirely metabolic: completed phase 1/2 and phase 2 studies in type 2 and type 1 diabetes and two ongoing phase 2 T2D trials (NCT07502495, NCT07502508). For this lab the drug's chief value is now as evidence about the mechanism, not as a therapeutic candidate: menin inhibition expands pancreatic beta cells, which is the on-target endocrine consequence of drugging a tumour suppressor.
Covalent mechanism — why it was attractive and why it may not have mattered
A covalent inhibitor forms a permanent bond with its target, so target engagement outlives plasma exposure. The hope in the menin field was that this would give deeper, more durable pathway suppression than reversible competitors, and possibly retain activity where a reversible drug's affinity is eroded by a pocket mutation. The programme did produce a documented complete remission in NUP98-fusion-positive AML — a menin-dependent subtype outside both approved indications 1 — which is real mechanistic proof of concept.
But the oncology trials were terminated 23 while the metabolic programme advanced. No peer-reviewed dataset establishes whether covalency solved the resistance problem. That question remains open and, in the lab's framing, important: if a covalent inhibitor can overcome MEN1 M327I/G331D, the menin-interface hypothesis survives resistance intact; if it cannot, the class needs a genuinely different mechanism.
The diabetes pivot is the most informative safety datum in the whole class
Biomea's active icovamenib programme is now in type 2 and type 1 diabetes: a completed phase 1/2 ascending-dose study in healthy adults and T2D (NCT05731544, n=443), a completed phase 2 in T1D (NCT06152042, n=37), and two phase 2 T2D trials as of 2026 (NCT07502495, ACTIVE_NOT_RECRUITING; NCT07502508, RECRUITING) 4.
Read as a toxicology signal, this is striking. MEN1 is a tumour suppressor; germline loss causes multiple endocrine neoplasia type 1, with parathyroid, pancreatic islet and pituitary tumours. A drug that blocks menin is therefore expected to promote endocrine cell proliferation — and a company has built a diabetes franchise on exactly that effect. Meanwhile the revumenib label's most common adverse reactions include increased phosphate and increased intact parathyroid hormone [see drug-revumenib E3].
The implication for a paediatric programme is concrete: a child cured of infant KMT2A-r ALL who received months or years of menin inhibition should be followed for parathyroid, islet and pituitary endpoints, and that surveillance should be built into any trial the lab helps design. See risk-menin-on-target-endocrine-proliferation.
| target | gene-men1 |
|---|---|
| target interaction | covalent (irreversible) engagement of menin |
| modality | small molecule (oral, covalent inhibitor) |
| status | Oncology programmes TERMINATED (NCT05153330 acute leukaemia/lymphoid; NCT05631574 solid tumours). Active development is in type 1 and type 2 diabetes. Not approved for any indication. |
| sponsor | Biomea Fusion |
| notable | the repurposing to diabetes is itself evidence that menin inhibition drives endocrine (beta-cell) proliferation — the on-target consequence of drugging a tumour suppressor |
trials
[ "NCT05153330", "NCT05631574", "NCT05731544", "NCT06152042", "NCT07502495", "NCT07502508", "NCT07254286" ]
efficacy
[
{
"population": "NUP98-fusion AML case report",
"n": 1,
"outcome": "complete remission"
}
]key ae
[ "not reported in a peer-reviewed leukaemia dataset" ]
resistance
[ "untested against MEN1 pocket mutants in any published dataset" ]
- The clearest signal in this node is an on-target toxicity signal, not an efficacy one: the same mechanism that kills KMT2A-r blasts makes endocrine cells proliferate, and a whole diabetes programme is built on it.
- A terminated leukaemia programme with no peer-reviewed cohort report means the field lost the covalent-versus-reversible resistance comparison it most needed.
- Commercial gravity moved a promising menin chemotype out of paediatric leukaemia entirely — a reminder that the lab cannot rely on industry to answer the infant KMT2A-r questions.
- The endocrine worry in this entry rests on mechanism plus a diabetes programme, not on an observed signal in a treated child - and the trials were not designed to produce one. Stating it as a risk is defensible; stating it as a finding would not be.
- Does covalent menin engagement overcome the MEN1 M327I/G331D mutations that defeat both approved reversible inhibitors? This is answerable in a week with isogenic cell lines and would materially change the resistance outlook for the whole class.
- Why were the oncology programmes terminated — futility, toxicity, or capital allocation? The registry does not say, and the answer determines whether covalency is a dead end or an unexplored option.
- If menin inhibition reliably expands beta cells in adults with diabetes, what is the long-term endocrine phenotype of a child treated for months during infancy, and should the lab push for a survivorship endocrine registry now rather than in ten years?
- E1weak confidence 2 of 5: Preliminary, small, indirect, or not yet peer reviewed.paper
Icovamenib (BMF-219), a covalent menin inhibitor, produced a complete remission in a patient with NUP98-fusion-positive acute myeloid leukaemia.
Complete remission of NUP98 fusion-positive acute myeloid leukemia with the covalent menin inhibitor BMF-219, icovamenib · Carraway HE, Nakitandwe J, Cacovean A, Ma Y, Munneke B, Waghmare G, Mandap C, Ahmed U, Kowalczyk N, Butler T, Morris SW · Haematologica 2024What this evidence cannot show
Single patient; most authors are employees of the sponsor; n=1 cannot establish response rate or durability, and the programme was subsequently terminated.
Evidence that argues against this (1)
- The one reported remission was in a NUP98-fusion patient, not a KMT2A-rearranged one, so it speaks only indirectly to the disease this graph is about. And it does not establish that covalency contributed anything: reversible menin inhibitors already produce responses across menin-dependent genotypes, with 63.2% overall response in the AUGMENT-101 KMT2A-rearranged cohort. A single covalent response is not evidence that covalency is the reason for it. Journal of Clinical Oncology 2025
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.trial_registry
COVALENT-101, the first-in-human phase 1 study of BMF-219 in acute leukaemia, DLBCL, multiple myeloma and CLL/SLL (n=55, 41 sites), is registered with status TERMINATED and primary completion 2025-02-13.
A Phase 1 First-in-human Dose-escalation and Dose-expansion Study of BMF-219, an Oral Covalent Menin Inhibitor, in Adult Patients With Acute Leukemia (AL), DLBCL, Multiple Myeloma and CLL/SLL · Biomea Fusion Inc. · ClinicalTrials.gov 2022What this evidence cannot show
The registry status 'TERMINATED' does not state the reason; termination may reflect portfolio prioritisation rather than a safety or futility finding, and no peer-reviewed full report of the leukaemia cohort has appeared.
Evidence that argues against this (2)
- Reading TERMINATED as a verdict on covalent menin chemistry is not supported by the record. The registry states the status and not the reason, the solid-tumour study stopped at 13 enrolled, and the sponsor simultaneously expanded a large diabetes programme - a pattern at least as consistent with capital allocation toward a bigger indication as with futility or toxicity. Nothing here licenses a mechanistic inference. ClinicalTrials.gov 2025
- Covalency was hoped to matter because it might overcome MEN1 interface mutations, but that gap is being closed without it: reversible ziftomenib retains antileukaemic activity against T349M and low-nanomolar potency against G331R, revumenib-selected mutants. The strategic loss from terminating a covalent programme is therefore smaller than this entry implies. Blood 2026
- E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.trial_registry
The solid-tumour BMF-219 study in KRAS-mutant NSCLC, pancreatic and colorectal cancer (n=13) is also registered as TERMINATED.
A Phase 1/1b Dose Finding Study of BMF-219, an Oral Covalent Menin Inhibitor, in Adult Patients With Unresectable, Locally Advanced, or Metastatic NSCLC, Pancreatic Cancer and Colorectal Cancer · Biomea Fusion Inc. · ClinicalTrials.gov 2023What this evidence cannot show
Very small enrolment (13) before termination; no efficacy conclusion can be drawn.
Evidence that argues against this (1)
- 'TERMINATED' here describes an indication, not the molecule, and reading it as a signal about menin biology is a mistake the registry itself forecloses. The raw ClinicalTrials.gov v2 whyStopped field on both terminated oncology studies carries the identical sponsor statement: 'Biomea Fusion, Inc., is no longer pursuing oncology indications for BMF-219. No safety concerns or efficacy observations led to this study closure.' Meanwhile the same compound, under the name icovamenib, is in ACTIVE phase 2 development in type 2 diabetes — NCT07502495 (phase 2, n=64, ACTIVE_NOT_RECRUITING) and NCT07502508 (phase 2, n=60, RECRUITING) — plus a completed healthy-volunteer food-effect study (NCT07254286, phase 1, n=60). All retrieved 2026-09-19. A terminated n=13 solid-tumour cohort is therefore not evidence that covalent menin engagement failed; it is evidence that a company changed therapeutic area, and the continuing diabetes programme is the on-target endocrine consequence of drugging a tumour suppressor, not an abandonment of the target. Biomea Fusion Inc. 2025
- E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.trial_registry
Icovamenib's active development programme is in diabetes, with a completed phase 1/2 study in healthy adults and type 2 diabetes (n=443), a completed phase 2 in type 1 diabetes (n=37), and two phase 2 type 2 diabetes trials registered for 2026.
Phase 2 Randomized, Double-blind, Placebo-Controlled Trial Assessing the Efficacy and Safety of Icovamenib in Participants With Type 2 Diabetes Not Achieving Glycemic Targets Despite GLP-1-Based Therapy · Biomea Fusion Inc. · ClinicalTrials.gov 2026What this evidence cannot show
Registry records establish the existence and design of the diabetes programme but not that beta-cell proliferation is the operative mechanism in humans; the inference from MEN1 tumour-suppressor biology to clinical endocrine risk is mechanistic reasoning, not a measured outcome in leukaemia survivors.
Evidence that argues against this (1)
- The inference from a diabetes programme to paediatric endocrine risk has not been borne out in observed toxicity - but the reason is that nobody looked. The pivotal AUGMENT-101 cohort included 23 children at median age 4.0 years and prespecified its adverse events of special interest as differentiation syndrome, QTc prolongation and peripheral neuropathy, with NO glycaemic or endocrine endpoint. Median treatment duration was 10 weeks. So the absence of a reported endocrine signal in treated children is absence of measurement over a short exposure, and is evidence neither for the worry nor against it. This entry should not be read as showing harm, and the registries should not be read as showing safety. Journal of Clinical Oncology 2025
covalent menin binder, development moved to diabetes
- ← related toBiomea Fusion — icovamenib (BMF-219): the covalent menin inhibitor that left oncology for diabetes
sponsor of icovamenib (BMF-219), whose development has moved from oncology to metabolic disease
500 mg daily, COVALENT-101
- related to →Counted absence: no menin-inhibitor trial has ever reported a response RATE in NUP98-rearranged patients
COVALENT-101 terminated with 55 enrolled and no posted results
registry: its AML/ALL phase 1 was terminated for commercial reasons with safety and efficacy explicitly disclaimed
Counter-evidence (ClinicalTrials.gov v2 API, NCT05153330/NCT05631574, retrieved 2026-09-19): both icovamenib oncology studies were TERMINATED with the identical whyStopped string stating no safety or efficacy observation drove closure, while bleximenib opened two phase 3 registration studies -- a mid-2025 six-inhibitor enumeration cannot support a present-tense claim about the class