Irreversible femoral growth-plate closure in juvenile rats: the label already tells paediatricians to watch growth
also called revumenib growth plate · physeal closure menin inhibitor · REVUFORJ bone growth warning
In young rats given revumenib for three months, the growth plates in the thigh bone closed early and did not recover. Because of this the approved label instructs doctors to monitor bone growth and development in children taking the drug.
The same thing, for a scientist
Section 8.4 of the REVUFORJ label reports that in 6-7 week old rats treated at 75, 150 or 300 mg/kg/day for 13 weeks, an **irreversible** increase in femur growth-plate closure was observed at exposures approximately 2 times the human AUC at the recommended dose, and concludes: *"Based on the findings in animals, monitor bone growth and development in pediatric patients"* [E1]. This is a formal, label-level paediatric growth warning for a menin inhibitor, and it is the most concrete late-effect risk in the class. It sits alongside raised alkaline phosphatase in 33% of treated patients and a mineral axis that moves in most of them [E1], and it lands on a population - infant-leukaemia survivors - in whom growth is already the commonest late effect, affecting 66% in the only infant-specific cohort.
What was observed1 deeper
| Element | Value |
|---|---|
| Species and age | Rat, 6-7 weeks at start (juvenile, growing) |
| Doses | 75, 150, 300 mg/kg/day |
| Duration | 13 weeks |
| Finding | Increase in femur growth-plate closure |
| Reversibility | Irreversible |
| Exposure at lowest dose with the finding | approximately 2x the human AUC at the recommended dose |
| Label consequence | Monitor bone growth and development in pediatric patients (section 8.4) |
Source: REVUFORJ full prescribing information, section 8.4 Animal Data, read verbatim 1.
The drug is approved for patients 1 year and older, and the safety population contained 25 infants under 2, 78 children aged 2 to under 12 and 29 adolescents 1. Every one of those patients is in the window where the physis is open and growth is ongoing. The rat study used animals at the equivalent life stage deliberately.
How strong is the exposure margin?
Two-fold is a small margin. For comparison, the same 13-week rat study at the same 75 mg/kg/day dose produced the multi-organ hyperplasia described in risk-menin-inhibitor-class-multiorgan-hyperplasia-nonclinical, and the dog study's irreversible nerve-fibre degeneration occurred at 1.3x human AUC 1. A 2x margin means the finding is not remote from clinical exposure, particularly in a child whose exposure is raised by a strong CYP3A4 inhibitor - which is how revumenib is actually dosed in the approved regimen, at 163 mg equivalent with a CYP3A4 inhibitor.
What would detect it in children, and what it would cost
Everything needed is standard paediatric endocrinology and mostly free:
- Height and height velocity at every visit, plotted as a Z score. Costs nothing.
- A left-hand bone age at baseline and annually. Costs one plain radiograph.
- IGF-1 to distinguish a chondrocyte-autonomous growth-plate effect (IGF-1 normal or high) from a pituitary one (IGF-1 low).
- Bone-specific alkaline phosphatase to interpret the 33% total-ALP rise.
None of these appears as an outcome measure in AUGMENT-101, KOMET-001 or cAMeLot-1 (see concept-endocrine-endpoint-absence-menin-trial-registries), even though AUGMENT-101 enrols from 30 days of age and cAMeLot-1 from 2 years.
| kind | essential-function |
|---|---|
| severity | irreversible in the animal model; unmeasured in humans |
| species | rat |
| age at dosing weeks | 6-7 |
| duration weeks | 13 |
| finding | increase in femur growth plate closure, irreversible |
| exposure multiple of human auc | 2 |
| label action | monitor bone growth and development in pediatric patients (section 8.4) |
| mitigation | serial height velocity Z score, annual left-hand bone age, IGF-1, bone-specific alkaline phosphatase |
| currently in any trial protocol | false |
affects
[ "drug-revumenib" ]
doses mg kg day
[ 75, 150, 300 ]
exposed paediatric population on label — the numbers behind the chart
{
"infants_under_2": 25,
"children_2_to_under_12": 78,
"adolescents_12_to_under_17": 29
}- The counter-argument at full strength: rats do not normally fuse their epiphyses, so accelerated physeal closure in a rat is a species-specific readout whose human translation is genuinely uncertain, and the same package's irreversible ocular and neural findings clearly overstated human risk.
- A 13-week continuous rat exposure at 2x human AUC is not the same as a median 2.5-month human exposure at 1x, and most treated children go to transplant rather than staying on drug.
- The label already mandates monitoring, so the regulatory system has responded proportionately; the gap is that no trial protocol operationalises it with a defined measurement schedule.
- What was the incidence and dose-response of growth-plate closure in the rat study? The label gives neither, and the FDA pharmacology review would.
- Has any child on a menin inhibitor had a bone age or a height-velocity Z score recorded prospectively? If the answer is no across three programmes and several hundred paediatric exposures, that is itself the finding.
- Is the effect chondrocyte-autonomous or GH-axis-mediated? One IGF-1 measurement per patient would decide it.
- Does menin inhibition affect the growth plate through the same MBM pocket, or through a menin function the drug does not touch? No growth-plate chromatin experiment under menin inhibition has been published.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
The REVUFORJ prescribing information reports that in 6-7 week old rats treated with revumenib at 75, 150 or 300 mg/kg/day for 13 weeks an irreversible increase in femur growth plate closure occurred at exposures approximately 2 times the human exposure at the recommended dose, and instructs clinicians to monitor bone growth and development in paediatric patients; the label also records irreversible nerve fibre degeneration in dogs at 1.3 times human exposure, irreversible lens opacities in rats, raised alkaline phosphatase in 33% of patients, and a paediatric safety population of 25 infants under 2 years, 78 children aged 2 to under 12 and 29 adolescents.
REVUFORJ (revumenib) tablets, for oral use - US full prescribing information (NDA 218944) · Syndax Pharmaceuticals / US FDA · DailyMed / Drugs@FDA (NDA 218944) 2025What this evidence cannot show
Read verbatim from the DailyMed full prescribing information, sections 6.1, 8.4 and 13.2, on 2026-09-18. The label does not give the incidence of growth-plate closure, the number of animals affected, the recovery-period length for the rat study, or any measurement of body length or IGF-1. Rat physeal biology differs from human: rats do not normally undergo full epiphyseal fusion, which makes the finding conspicuous in that species but complicates extrapolation in both directions.
Evidence that argues against this (1)
- The same nonclinical package contains irreversible findings that did not translate proportionately into human harm - lens opacities that progressed in rats and did not reverse, against cataract listed only among human reactions under 20%, and nerve fibre degeneration in dogs that did not reverse, against peripheral neuropathy in 3.2% of patients in the pivotal publication, all below grade 2. That is direct, same-drug evidence that irreversible animal findings here can overstate clinical risk. Issa GC et al. 2024
label section 8.4 mandates monitoring bone growth in paediatric patients on the basis of irreversible rat physeal closure at 2x human AUC
lands on a population in which growth is already the commonest late effect (66% in the only infant-specific cohort)
- supports →The same pocket suppresses JUND: menin inhibition's unmeasured endocrine liability in infants
the most concrete paediatric late-effect signal in the class, and it is already on the label
- ← supportsA documented absence: not one endocrine outcome measure across the three lead menin-inhibitor trials
the label mandates monitoring bone growth in children and no trial operationalises it
- ← related toThe reversibility question: months of partial occupancy against a syndrome built from decades of loss
the internal inconsistency: 13 weeks of dosing already produced irreversible findings at 2x human AUC
operationalises the label's instruction to monitor bone growth in paediatric patients with height velocity and bone age
Offers the first mechanistic hypothesis for the label's growth-plate finding - accelerated terminal osteoblast maturation plus depleted osteoprogenitor commitment - while noting the chondrocyte route it would actually require has a negative result behind it
Verified absence: 2 records for menin/Men1 with growth plate/chondrocyte/epiphysis (both acromegaly), 0 for any menin inhibitor with skeletal terms, against 118 for 'menin inhibitor' as a control
- ← supportsWhat the labels measure in a treated child: bone growth, in one sentence, and nothing else endocrine
The growth-plate finding produced exactly one label sentence, with no named measurement, interval or action threshold