NCT05631574 (COVALENT-102): the third solid-tumour menin-inhibitor trial, missing from this graph because its drug is called BMF-219 - terminated after 13 patients for a strategic reason, explicitly NOT for efficacy or safety
also called NCT05631574 · COVALENT-102 · BMF-219 solid tumour · icovamenib solid tumour
A covalent menin-blocking drug was tested in lung, pancreatic and bowel cancers driven by a KRAS mutation. The company closed the study after thirteen patients because it decided to stop working on cancer with this drug altogether, and it stated that neither safety nor lack of benefit caused the closure.
The same thing, for a scientist
Phase 1 dose-finding of BMF-219 (icovamenib), an **oral covalent** menin inhibitor, in unresectable/locally advanced/metastatic NSCLC, pancreatic ductal adenocarcinoma and colorectal cancer, enriched for KRAS-mutant disease. Sponsor Biomea Fusion. Started 2023-01-12, completed 2025-01-15, enrollment **13 actual**, status TERMINATED. The registry's `whyStopped` is unusually explicit and is the reason this trial matters to this graph's reasoning: *"Biomea Fusion, Inc., is no longer pursuing oncology indications for BMF-219. No safety concerns or efficacy observations led to this study closure."* That is a **portfolio decision, not a negative result** - and it is the opposite of the other terminated solid-tumour menin trial, NCT05731947 (revumenib in colorectal and other solid tumours, n=41), which was stopped *"after Phase 1b as pre-defined signals of efficacy were not met"*. This graph previously held that only two menin-inhibitor trials had ever opened in solid tumours and that the terminated one failed on efficacy. There are three, and the two terminations have different causes.
Why it was invisible here
The bounded absence that surveyed this space queried ClinicalTrials.gov query.intr over thirteen strings: revumenib, ziftomenib, bleximenib, enzomenib, VTP-50469, VTP50469, MI-503, MI-3454, SNDX-5613, KO-539, and three phrasings of menin inhibitor. BMF-219, icovamenib, DS-1594 and BN104 were not among them. query.intr=BMF-219 returns 5 studies and icovamenib returns 3; the union of fifteen names returns 69 trials against the original list's smaller set 1. The same measurement exposes a second name-list defect worth carrying: enzomenib returns 1 trial and its development code DSP-5336 returns 2, so the INN and the code are not interchangeable even for a drug both names are current for.
What it does and does not change
The headline absence survives. No menin inhibitor has enrolled a KMT2A-rearranged solid tumour, and none has enrolled a soft-tissue sarcoma. Across all 69 trials in the expanded union, the non-haematological, non-diabetes ones are exactly four: NCT05731947 (colorectal/solid, terminated), NCT06655246 (GIST plus imatinib, recruiting, n=157 estimated), this trial, and two healthy-volunteer food-effect studies of BN104 12.
One supporting statement does change. The claim only two have ever been opened in solid tumours is now three, and the summary that the solid-tumour experience is a straightforward efficacy failure is only half true: one termination was for efficacy, one was strategic and the sponsor said so in the registry.
And there is a near-miss term trap. Searching menin trials for sarcoma now returns NCT05735184, whose condition list includes Myeloid Sarcoma - an extramedullary mass of leukaemic myeloblasts, not a soft-tissue sarcoma. A sarcoma absence scored by string match on the condition field would read that as a hit 2.
| nct | NCT05631574 |
|---|---|
| phase | Phase 1 |
| status | TERMINATED (completed 2025-01-15) |
| n | 13 |
| readout | safety, tolerability, dose finding; no results posted |
| sponsor | Biomea Fusion Inc. |
| start date | 2023-01-12 |
| why stopped verbatim | Biomea Fusion, Inc., is no longer pursuing oncology indications for BMF-219. No safety concerns or efficacy observations led to this study closure. |
| why it is in this graph | it is the third solid-tumour menin-inhibitor trial and the bounded absence that surveyed this space knew of only two, because its drug name list did not contain BMF-219 or icovamenib |
| kmt2a selected | false |
| sarcoma | false |
| shown for fusion | NO - no KMT2A rearrangement in the eligibility criteria and no fusion-selected cohort |
arms
[ "BMF-219 (icovamenib), oral covalent menin inhibitor, dose finding" ]
conditions
[ "NSCLC", "pancreatic ductal adenocarcinoma", "colorectal cancer", "KRAS mutation-related tumours" ]
- A covalent menin inhibitor commits irreversibly to a target that normal HSCs and pancreatic beta cells both need; Biomea's own pivot to diabetes with the same molecule is evidence that the on-target pharmacology reaches the endocrine compartment, which is a toxicity argument in a paediatric leukaemia setting rather than a reassurance.
- Adding names to a drug-class query makes every previously published count in that class stale. The correction here is small, but the same expansion applied to every menin-inhibitor count in this corpus would move numbers that other entries quote.
- Were any of the 13 patients profiled for a KMT2A rearrangement or a NUP98 fusion? The record does not say, and a covalent menin inhibitor's first solid-tumour cohort is the only place such a patient could have been treated to date.
- Biomea has moved BMF-219 to diabetes. Does a covalent menin inhibitor's safety database in a non-oncology population change the risk calculus for using one in infants, where the oncology exposure is tiny?
- How many other drug-class absences in this corpus are bounded by a name list that predates the newest agent in the class? This is the first one re-run and it was short by four names.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.trial_registry
NCT05631574 is a Phase 1 trial of the covalent menin inhibitor BMF-219 in KRAS-driven NSCLC, pancreatic and colorectal cancer, terminated after 13 patients for a stated non-efficacy, non-safety reason.
A Phase 1/1b Dose Finding Study of BMF-219, an Oral Covalent Menin Inhibitor, in Adult Patients With Unresectable, Locally Advanced, or Metastatic Non-small Cell Lung Cancer (NSCLC), Pancreatic Cancer and Colorectal Cancer · ClinicalTrials.gov 2026What this evidence cannot show
Fetched from the v2 API on 2026-09-22. No results are posted; the trial has no publication linked in the record, so nothing is known about what the 13 patients experienced beyond the sponsor's stated reason for stopping. A sponsor's own `whyStopped` text is a self-report.
Evidence that argues against this (1)
- A company exiting an entire indication can be a polite way of saying the data were unpromising. The registry text distinguishes the two and should be read as the sponsor's framing rather than as an independent finding; the honest statement is that no efficacy data from this trial are public either way. ClinicalTrials.gov 2026
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.trial_registry
The other terminated solid-tumour menin trial was stopped for a pre-specified efficacy signal that was not met, which is a different fact from this one.
A Phase 1/2 Study to Investigate the Safety, Tolerability, Pharmacokinetics, Pharmacodynamics, and Efficacy of SNDX-5613 in Patients With Colorectal Cancer and Other Solid Tumors · ClinicalTrials.gov 2026What this evidence cannot show
n=41 actual, Phase 1/2, terminated 2025-06-30. The colorectal setting is menin-dependent for a different reason than KMT2A-rearranged leukaemia is, so this negative does not transfer to a KMT2A-rearranged solid tumour if one were ever enrolled.
Evidence that argues against this (1)
- Neither terminated trial selected patients for a KMT2A rearrangement, so neither is a test of the mechanism this project cares about. Reading them as evidence against menin inhibition in KMT2A-rearranged solid tumours would be exactly the partner-imported-evidence error the corpus keeps catching. ClinicalTrials.gov 2026
the supporting count only: that entry says two menin-inhibitor trials have ever opened in solid tumours; there are three and this is the third. Its headline - no KMT2A-rearranged solid tumour has ever been enrolled - survives
the registry-query class re-run: a drug name list short by four names, and an INN that retrieves fewer trials than its own development code