Oncofusion

start / graph / risk-menin-inhibitor-class-multiorgan-hyperplasia-nonclinical

Riskmoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q-endocrine-liability

Both approved menin inhibitors cause multi-organ hyperplasia in animals, and one of them in the pancreas irreversibly

also called menin inhibitor hyperplasia · class nonclinical proliferative finding · revumenib pancreatic hyperplasia

In animal safety studies, both approved menin-blocking drugs caused tissues in several organs to overgrow. For revumenib this included the pancreas, and in that organ the overgrowth did not go away after the drug was stopped.

The same thing, for a scientist

The revumenib label reports hyperplasia in multiple organs including the testes, mammary gland, uterus, pancreas and kidney in rats at 75 mg/kg/day or above for up to 13 weeks, **irreversible in the testes, mammary gland and pancreas**, at approximately 2x the human AUC [E1]. The ziftomenib label independently reports that in 13-week studies in mice and dogs, *"incidences of hyperplasia in a variety of organs occurred in several animals from both species"*, in mice at exposures as low as **0.3x** and in dogs as low as **0.05x** the human AUC at the recommended dose [E2]. A proliferative phenotype in multiple tissues at sub-therapeutic exposure, in three species, across two chemically distinct compounds, is a class signal rather than a compound idiosyncrasy. **Correction, 2026-09-19:** this entry previously continued *"and hyperplasia is precisely what loss of a tumour suppressor produces"*. That inference is withdrawn as stated. Endocrine hyperplasia is equally what **adaptive compensation** produces: menin falls physiologically in pregnancy and under hyperglycaemia to permit islet expansion [E4], acute Men1 excision reverses established hyperglycaemia in three diabetic models by expanding beta-cell mass [E5], and 12 weeks of a menin inhibitor at 70 mg/kg/day increased beta-cell proliferation and mass with **improved** fasting glucose and glucose tolerance [E6]. Hyperplasia is a tissue response, not a diagnosis, and the class signal here is on-target proliferation whose sign depends on tissue, duration and reversibility. The pancreatic finding is the one that lines up with MEN1 biology, where beta-cell Men1 ablation gives hyperplastic islets in 2 months [E3].

Go deeper
What each label says1 deeper

Revumenib, section 13.2 1:

"Hyperplasia was observed in multiple organs including the testes, mammary gland, uterus, pancreas, and kidney in rats treated at >= 75 mg/kg/day for up to 13 weeks. The findings were irreversible in the testes, mammary gland, and pancreas. Revumenib exposures at 75 mg/kg/day in rats are approximately 2 times the human exposure (AUC) at the recommended dose."

Ziftomenib, section 13.2 2:

"In 13-week repeat-dose toxicity studies in mice and dogs, incidences of hyperplasia in a variety of organs occurred in several animals from both species. The findings in the mouse study were observed at >= 12 mg/kg/day at exposures as low as 0.3 times the AUC at the recommended dose, and the findings in the dog study were observed at >= 6 mg/kg/day at exposures as low as 0.05 times the AUC at the recommended dose."

Both labels also record, in section 13.1, that lymphoma was observed in multiple organs in one animal in a 13-week repeat-dose study - rat for revumenib, mouse for ziftomenib 12. Neither drug has had a carcinogenicity study conducted.

Why the exposure multiples are the striking part

The ziftomenib dog finding at 0.05x human AUC is twenty-fold below clinical exposure. That is unusual for a proliferative finding and is hard to dismiss as a high-dose artefact. If the mechanism is on-target - relief of menin-dependent transcriptional repression in proliferating epithelia - then low-exposure sensitivity is what one would predict, because partial occupancy would suffice.

The caveat that must be stated with equal force: the ziftomenib label does not name the organs, does not give incidences, does not state severity or reversibility, and 'hyperplasia in several animals' in a 13-week rodent or dog study is a common, often background, finding that toxicologic pathologists frequently consider non-adverse. The revumenib label at least names organs and states reversibility. The two entries are not of comparable evidentiary weight.

The MEN1 alignment, stated carefully

Menin is a tumour suppressor whose loss causes hyperplasia before it causes tumours - in MEN1 the parathyroid lesion is multiglandular hyperplasia, and beta-cell Men1 ablation produces hyperplastic islets at 2 months before insulinomas at 6 months 3. So 'multi-organ hyperplasia' is the exact nonclinical phenotype that on-target menin inhibition predicts.

But the organs only partly match. MEN1 targets parathyroid, enteropancreatic and pituitary. The revumenib rat findings were testes, mammary gland, uterus, pancreas and kidney - overlapping MEN1 in the pancreas only, and including three reproductive organs that MEN1 does not typically involve. Rats are notoriously prone to spontaneous proliferative lesions in mammary gland, testis and kidney. So the honest reading is: one of five organs matches the MEN1 prediction, and that one was irreversible. Whether that pancreatic hyperplasia was islet or exocrine is not stated on the label and is the most consequential unresolved detail in this beat.

Correction, 2026-09-19: hyperplasia is not a diagnosis

This entry read the nonclinical hyperplasia findings as a tumour-suppressor-loss signature. That reading is one of two available and the entry did not acknowledge the other.

  • Menin is physiologically down-regulated in the beta cell - by pregnancy/prolactin and by glucose - precisely in order to permit proliferation, and forcing it to stay high causes gestational diabetes 4.
  • Acute inducible Men1 excision reverses established hyperglycaemia in streptozotocin, db/db and high-fat-diet mice by expanding endogenous beta-cell mass 5.
  • Twelve weeks of the menin inhibitor MI-463 at 70 mg/kg/day in diabetic mice raised BrdU-positive beta cells and beta-cell mass and lowered fasting glucose, with improved glucose tolerance and no change in body weight or food intake 6.

So pancreatic hyperplasia under a menin inhibitor has a published beneficial counterpart in the same organ. What the entry's argument actually needs, and still lacks, is the irreversibility datum - and that it has: the revumenib label reports the pancreatic, testicular and mammary hyperplasia as irreversible 1. The irreversibility, not the hyperplasia, is the finding that should carry the weight. Note also that Ma 2021 performed no pancreatic histopathology for neoplasia, so its silence on tumours is absence of measurement, not a negative result.

See mechanism-in-vivo-islet-phenotype-under-pharmacological-menin-inhibition and concept-four-exposure-classes-of-menin-loss-are-not-interchangeable.

Facts and figures
kindessential-function
severitynonclinical proliferative findings, adversity not stated by either label
carcinogenicity studies conductedfalse
men1 alignmentpancreas only, of five named organs; hyperplasia is the first phenotype of Men1 loss in endocrine tissue
decisive missing factwhether the revumenib rat pancreatic hyperplasia was islet or exocrine
affects
[
  "drug-revumenib",
  "drug-ziftomenib"
]
revumenib
{
  "species": "rat",
  "organs": [
    "testes",
    "mammary gland",
    "uterus",
    "pancreas",
    "kidney"
  ],
  "irreversible_in": [
    "testes",
    "mammary gland",
    "pancreas"
  ],
  "duration_weeks": 13,
  "lowest_exposure_multiple_of_human_auc": 2
}
ziftomenib
{
  "species": [
    "mouse",
    "dog"
  ],
  "organs": "not named on label",
  "duration_weeks": 13,
  "lowest_exposure_multiple_of_human_auc_mouse": 0.3,
  "lowest_exposure_multiple_of_human_auc_dog": 0.05
}
How this could bite us back
  • The counter-argument at full strength: 13-week hyperplasia in rodents and dogs is a common, often non-adverse finding; neither label characterises it as adverse; regulators with access to the full study reports approved both drugs, one of them for one-year-olds, without a warning derived from it.
  • Four of the five named revumenib organs - testes, mammary gland, uterus, kidney - are not MEN1 target tissues and are classic sites of spontaneous rat proliferative lesions. Reading the set as a MEN1 signal requires ignoring most of it.
  • No carcinogenicity study has been done for either drug, so the correct statement is that the long-term proliferative consequence is unknown, not that it is bad. Unknown and bad are different claims and the distinction matters for children who need the drug.
  • Reading every proliferative finding as tumour-suppressor loss makes the entry unable to distinguish an adaptive response from a pre-neoplastic one, which is the distinction the whole beat turns on.
Still open
  1. Islet or exocrine? The FDA multidisciplinary pharmacology review for NDA 218944 should state which pancreatic compartment was hyperplastic in the rat, and that single fact would move this entry substantially in one direction or the other.
  2. Which organs were hyperplastic in the ziftomenib mouse and dog studies? The label does not say, and a parathyroid or islet entry would make this a class effect in MEN1 target tissue.
  3. Why does a proliferative finding appear at 0.05x human AUC in dogs? Either the finding is background, or partial menin occupancy is sufficient to relieve repression - and those two possibilities have opposite implications for paediatric use.
  4. Has any menin inhibitor been given to an animal for longer than 13 weeks? Every nonclinical study cited on both labels is 13 weeks, which is shorter than the FDA postmarketing requirement's own 3-year treatment specification for humans.
  5. Is the pancreatic hyperplasia in the revumenib rat study islet or exocrine, and if islet, is it beta-cell or alpha-cell? The label does not say, and the answer decides whether it is the beneficial phenotype or the MEN1 one.
Every source behind this entry (6)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    The REVUFORJ label reports hyperplasia in multiple organs including the testes, mammary gland, uterus, pancreas and kidney in rats treated at 75 mg/kg/day or above for up to 13 weeks, irreversible in the testes, mammary gland and pancreas, at exposures approximately 2 times the human AUC at the recommended dose; carcinogenicity studies have not been conducted and lymphoma was observed in multiple organs in one rat in a 13-week study.

    REVUFORJ (revumenib) tablets, for oral use - US full prescribing information (NDA 218944) · Syndax Pharmaceuticals / US FDA · DailyMed / Drugs@FDA (NDA 218944) 2025
    What this evidence cannot show

    Read verbatim from the DailyMed full prescribing information, sections 13.1 and 13.2, on 2026-09-18. No incidences, no severity grades, no statement of whether the pancreatic hyperplasia was islet or exocrine, and no recovery-period duration for the rat study. Rats develop spontaneous proliferative lesions in mammary gland, testis and kidney at appreciable background rates, and a 13-week study has no concurrent carcinogenicity context.

    Evidence that argues against this (2)
    • Hyperplasia across testes, mammary gland, uterus, pancreas and kidney in rats is a real class signal and also the pattern rodent toxicology most often produces at high exposure multiples without human correlate. The entry does not carry the dose multiples, the duration or whether the changes reversed on recovery, and those three numbers decide how much weight it can bear. Reported human endocrine findings so far are laboratory, not structural. FDA Drugs@FDA 2024
    • The comparator that matters is not 'no targeted therapy' but the endocrine late effects of the targeted and immune-based agents paediatric oncology already uses, which are themselves substantial and increasingly documented. A class signal in rats is a reason to measure, and it does not by itself make menin inhibitors unusual among modern paediatric oncology drugs. Cells 2026
  2. E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    The KOMZIFTI (ziftomenib) label reports that in 13-week repeat-dose toxicity studies in mice and dogs, incidences of hyperplasia in a variety of organs occurred in several animals from both species, at exposures as low as 0.3 times the human AUC in mice and 0.05 times the human AUC in dogs; carcinogenicity studies have not been conducted and lymphoma was observed in multiple organs in one mouse in a 13-week study.

    KOMZIFTI (ziftomenib) capsules, for oral use - US full prescribing information · Kura Oncology / US FDA · DailyMed / Drugs@FDA 2025
    What this evidence cannot show

    Read verbatim from the DailyMed full prescribing information, sections 13.1 and 13.2, on 2026-09-18. The label names no organs, gives no incidence, no severity and no reversibility, which makes it much weaker evidence than the revumenib entry despite the striking exposure multiples. 'Several animals' is not quantified.

    Evidence that argues against this (1)
    • Hyperplasia in a 13-week rodent or dog toxicity study is frequently a background or adaptive finding and is often classified as non-adverse by toxicologic pathologists. Neither label calls these findings adverse, neither drug's label carries a warning derived from them, and FDA approved both drugs including revumenib for children aged 1 year and older with only a bone-growth monitoring instruction. Regulators who saw the full study reports did not treat these findings as a barrier. US Food and Drug Administration et al. 2024
  3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Mice with Men1 disrupted specifically in pancreatic beta cells developed hyperplastic islets as early as 2 months and insulinomas at 6 months with full penetrance, showing that hyperplasia is the first phenotype of menin loss in an endocrine tissue.

    PMID 12941803
    What this evidence cannot show

    Read as abstract. Complete biallelic ablation from development, not pharmacological occupancy; the relevance here is only that hyperplasia is the expected early phenotype of menin loss, which makes a hyperplastic nonclinical signal mechanistically coherent rather than surprising.

    Evidence that argues against this (1)
    • The beta-cell model shows what CONSTITUTIVE BIALLELIC loss from development does to one tissue, whereas the rat finding is a pharmacological, reversible, multi-organ observation. Placing them in one entry invites the reader to treat the mouse as the mechanism of the rat finding, which is not established and which the tissue distributions do not match. Cancer Research 2003
  4. E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Menin is physiologically down-regulated in beta cells to permit adaptive proliferation; forcing it high causes gestational diabetes in mice.

    Menin controls growth of pancreatic beta-cells in pregnant mice and promotes gestational diabetes mellitus. · Karnik SK, Chen H, McLean GW, Heit JJ, Gu X, Zhang AY, et al. · Science 2007
    PMID 17975067DOI
    What this evidence cannot show

    Mouse pregnancy; transgenic gain-of-function.

    Evidence that argues against this (1)
  5. E5strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    Acute inducible Men1 excision reverses established hyperglycaemia in three diabetic mouse models by expanding beta-cell mass.

    Reversal of preexisting hyperglycemia in diabetic mice by acute deletion of the Men1 gene. · Yang Y, Gurung B, Wu T, Wang H, Stoffers DA, Hua X · Proceedings of the National Academy of Sciences USA 2010
    PMID 21059956DOI
    What this evidence cannot show

    Genetic excision, short follow-up, no tumour incidence reported.

    Evidence that argues against this (1)
    • Longer follow-up of beta-cell Men1 loss yields insulinoma and glucagon-expressing tumours. Li F et al. 2015
  6. E6moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    MI-463 at 70 mg/kg/day for 12 weeks increased beta-cell proliferation and mass and lowered fasting glucose in diabetic mice.

    Menin-regulated Pbk controls high fat diet-induced compensatory beta cell proliferation. · Ma J, Xing B, Cao Y, He X, et al. · EMBO Molecular Medicine 2021
    PMID 33821572DOI
    What this evidence cannot show

    n=4-5, tool compound, adult mice, and critically no pancreatic histopathology for neoplasia and no off-drug follow-up.

    Evidence that argues against this (1)
    • The revumenib label reports the rat pancreatic hyperplasia as irreversible at ~2x human AUC, which no beta-cell benefit study has addressed. Syndax Pharmaceuticals et al. 2026
Connected to