The cheapest study that would settle this: nine assays and a hand radiograph, on children already being followed
also called menin endocrine surveillance study · paediatric menin inhibitor endocrine panel
Add a short list of blood tests and one hand X-ray to the follow-up that children treated with menin inhibitors already receive, at fixed times during and after treatment. It uses blood that is already being drawn and would answer within two years whether the drug leaves a hormonal mark.
The same thing, for a scientist
A prospective observational add-on to existing paediatric menin-inhibitor treatment and survivorship follow-up, collecting a nine-analyte endocrine and mineral panel plus auxology and a bone age at baseline, monthly on treatment, and - critically - at 3, 6 and 12 months **after** the last dose. The design is deliberately the minimum that discriminates the three live explanations for the labelled 34% intact-PTH elevation [E1]: on-target parathyroid effect, renal/phosphate secondary hyperparathyroidism, and vitamin D deficiency. The off-treatment arm is the part that does not exist anywhere in the current evidence base and is the part FDA's postmarketing requirement 4730-1 does not compel, because that requirement concerns patients on treatment for at least three years [E2].
The panel2 deeper
| Assay | Discriminates | Cost tier |
|---|---|---|
| Intact PTH | the index finding | low |
| Corrected and ionised calcium | primary (high Ca) versus secondary (low Ca) hyperparathyroidism | low |
| Phosphate | renal retention driving PTH | low |
| 25-OH vitamin D | the commonest confounder in inpatients | low |
| Creatinine / eGFR | renal contribution | low, already collected |
| Bone-specific alkaline phosphatase | whether the 33% total-ALP rise is bone or liver | low |
| IGF-1 | GH-axis versus chondrocyte-autonomous growth effect | low |
| Prolactin | the pituitary axis MEN1 also hits | low |
| Fasting glucose and insulin | the islet axis | low |
| Height, weight, height velocity Z score | the outcome that matters most in a child | free |
| Left-hand bone age radiograph, baseline and annual | the juvenile-rat growth-plate finding, translated | low |
The label already requires monthly electrolytes and liver enzymes, so the blood is being drawn. Six of the nine assays are add-ons to an existing tube.
Schedule and why the off-treatment arm is the whole point
Baseline; day 28 of cycles 1, 2, 3; every 3 months thereafter while on treatment; then 3, 6 and 12 months after the last dose.
Every published endocrine number for this class is an on-treatment number. Not one off-treatment value has been reported for any menin inhibitor. Yet reversibility is the entire crux of the risk (risk-menin-inhibitor-exposure-duration-vs-men1-latency): the argument that months of reversible occupancy is harmless stands or falls on whether the biochemical changes reverse, and nobody has looked. Three post-treatment draws in a child who is already attending survivorship clinic would answer it.
Analysis and what each result would mean
Primary: proportion with intact PTH above the age-specific reference range at any on-treatment timepoint, and the proportion of those whose PTH has normalised by 6 months off treatment.
- PTH rises, calcium rises with it, both normalise off treatment -> reversible on-target parathyroid effect. Surveillance, not alarm.
- PTH rises, calcium falls, tracks creatinine and phosphate -> secondary hyperparathyroidism. The endocrine hypothesis loses its clinical anchor and this beat should shrink.
- PTH stays elevated at 12 months off treatment -> the finding that would change practice, and the one nobody can currently exclude.
- Height velocity Z score falls and bone age advances relative to chronological age -> translation of the juvenile-rat growth-plate finding, which the label already asks clinicians to watch for without telling them how.
Controls: contemporaneous children on the same protocols not receiving a menin inhibitor, for the renal and nutritional background; and each child's own pre-treatment baseline.
Added 2026-09-19 - the covariates without which this analysis is uninterpretable. Grade >= 3 differentiation syndrome occurred in 16.0% and febrile neutropenia in 37.2% of the AUGMENT-101 phase 2 population [counter-evidence to E1]. Differentiation syndrome is managed with systemic corticosteroids, which independently raise fasting glucose, suppress IGF-1, and lower bone formation and bone-specific alkaline phosphatase - four of this panel's readouts, plus height velocity. Cumulative corticosteroid dose, differentiation-syndrome occurrence and grade, sepsis episodes, concomitant strong CYP3A4 inhibitor and transplant/conditioning exposure must be recorded at every timepoint; they are now listed in data.covariates_required. Whether these can be separated at realistic paediatric sample sizes should be settled by simulation before the study is proposed.
Feasibility
AUGMENT-101 has enrolled from 30 days of age and the revumenib label's safety population already contains 25 infants under 2, 78 children aged 2 to under 12 and 29 adolescents 1. cAMeLot-1 enrols from 2 years and has paediatric-specific renal criteria, so paediatric protocol adaptation is already routine in this class 3. The COG and the Princess Maxima Center both run infant-leukaemia survivorship follow-up into which this panel drops without a new visit.
The binding constraint is blood volume in infants, which is why the panel is nine analytes and not thirty, and why bone age is annual rather than six-monthly.
| design | prospective observational add-on to existing paediatric menin-inhibitor treatment and survivorship follow-up; each child serves as their own baseline control, with contemporaneous non-menin-inhibitor controls on the same protocols for renal and nutritional background |
|---|---|
| readout | proportion with intact PTH above the age-specific reference range on treatment, and the proportion of those normalising by 6 months off treatment; secondary readouts height velocity Z score, bone age advance, IGF-1, prolactin, fasting glucose and insulin |
| schedule | baseline; day 28 of cycles 1-3; every 3 months on treatment; and 3, 6 and 12 months after the last dose |
| controls | each child's own pre-treatment baseline plus contemporaneous children on the same protocols not receiving a menin inhibitor; and, added 2026-09-19, adjustment for the covariates in data.covariates_required, without which the corticosteroid and sepsis burden of this population will dominate at least four of the nine analytes |
| cost tier | low |
| timeline | first interpretable readout at 18-24 months; definitive off-treatment answer at 3 years |
| novelty | the off-treatment arm; no off-treatment endocrine value has been published for any menin inhibitor |
| constraint | blood volume in infants, which caps the panel at nine analytes |
panel
[ "intact PTH", "corrected calcium", "ionised calcium", "phosphate", "25-OH vitamin D", "creatinine/eGFR", "bone-specific alkaline phosphatase", "IGF-1", "prolactin", "fasting glucose and insulin" ]
auxology
[ "height", "weight", "height velocity Z score", "left-hand bone age radiograph at baseline and annually" ]
covariates required
[ "cumulative systemic corticosteroid dose (differentiation-syndrome management perturbs glucose, IGF-1, bone ALP and height velocity)", "differentiation syndrome occurrence and grade (grade >= 3 in 16.0% of the AUGMENT-101 phase 2 population, PMID 39121437)", "intercurrent sepsis or febrile neutropenia episodes (grade >= 3 febrile neutropenia in 37.2%, PMID 39121437)", "concomitant strong CYP3A4 inhibitor (azole antifungals), which sets revumenib exposure and is near-universal on this protocol", "transplant status and conditioning exposure, which independently damage every axis in the panel" ]
- The panel's biggest weakness is not overdiagnosis, it is confounding. Differentiation syndrome occurs at grade >= 3 in 16.0% of treated patients and is managed with systemic corticosteroids, which move glucose, IGF-1, bone-specific alkaline phosphatase and height velocity - four of this panel's readouts - harder and faster than any plausible on-target menin effect. Without corticosteroid dose as a covariate the study will produce an uninterpretable positive.
- Adding assays generates abnormal numbers, and abnormal numbers in children generate referrals, imaging and anxiety. A protocol for this study needs an explicit action threshold, or it will do net harm through overdiagnosis.
- If the result is a reversible grade 1 biochemical drift, the study will have spent two years to confirm something clinically unimportant - which is still the right outcome to know, but it should be expected.
- Blood volume in infants is a genuine constraint and the panel competes with pharmacokinetic and disease-monitoring sampling that has more immediate value to the child in front of you.
- The 'blood is already being drawn' economy is partly illusory at the design stage: labelling data from a second approved menin inhibitor carries none of these analytes, so nothing can be borrowed from existing programmes and every value has to be collected de novo under a new protocol amendment.
- What are the age-specific reference ranges for intact PTH in infants under 2? These are less well established than in adults and would need to be fixed prospectively rather than assumed.
- Can the off-treatment arm be attached to an existing survivorship cohort rather than to the trials, which would avoid protocol amendment entirely?
- Should the panel include fasting gastrin and chromogranin A for the enteropancreatic axis? They are cheap but less interpretable in children and were left out to protect blood volume.
- Can a menin-inhibitor-attributable endocrine effect be separated from the corticosteroid and sepsis burden of this population at all? With grade >= 3 differentiation syndrome in 16.0% and febrile neutropenia in 37.2% (PMID 39121437), the covariate structure may be too collinear at realistic paediatric sample sizes, and that should be established by simulation before the study is proposed.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
The revumenib label reports intact PTH increased in 34%, corrected calcium increased in 15%, phosphate increased in 51% and decreased in 25%, alkaline phosphatase increased in 33% and creatinine increased in 38% of 241 patients; lists hyperparathyroidism under Endocrine disorders; instructs clinicians to monitor bone growth and development in paediatric patients on the basis of irreversible juvenile-rat femoral growth-plate closure; and describes a paediatric 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 DailyMed on 2026-09-18. The label reports on-treatment laboratory abnormalities only; no off-treatment value, no age-stratified rate and no growth measurement appears anywhere in it.
Evidence that argues against this (2)
- The design has a confounder large enough to swallow most of its readouts and does not control for it. In AUGMENT-101 phase 2 (N = 94 treated), grade >= 3 differentiation syndrome occurred in 16.0% and febrile neutropenia in 37.2%. Differentiation syndrome is managed with systemic corticosteroids, and corticosteroids raise fasting glucose, suppress IGF-1 and bone formation, and lower bone-specific alkaline phosphatase - which is four of the eleven analytes in this panel, plus height velocity. Any child who has had differentiation syndrome is effectively in a different study. The panel needs corticosteroid cumulative dose, differentiation-syndrome occurrence and intercurrent sepsis recorded as covariates at every timepoint, and the current specification records none of them. Issa GC et al. 2025
- The cross-drug comparison this design implicitly assumes is not available from labelling and probably never will be. Ziftomenib (KOMZIFTI, NDA 220305, label revised 11/2025) is a second approved menin inhibitor whose prescribing information contains no mention of parathyroid, parathyroid hormone or hyperparathyroidism, no Endocrine disorders heading, and no calcium, phosphate or PTH row in its selected laboratory-abnormality table (N approximately 108-110; creatinine increased 45%, alkaline phosphatase increased 41%). Sponsors select which analytes appear in a label, so absence there is uninformative about whether the analyte was measured. The practical consequence for this experiment is that it cannot be answered by mining existing programmes and must collect its own data prospectively - which raises its cost above the 'add-on to blood already being drawn' framing in the summary. Kura Oncology / Kyowa Kirin / US FDA 2025
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
FDA postmarketing requirement 4730-1 requires a clinical trial to determine whether long-term revumenib use is associated with serious potential risk of endocrine-related disorders due to chronic menin inhibition, with all patients to have at least 3 years of treatment or to have discontinued earlier, reporting in May 2028.
NDA 218944 approval letter for REVUFORJ (revumenib), including postmarketing requirements 4730-1 to 4730-6 · US Food and Drug Administration, Center for Drug Evaluation and Research · Drugs@FDA, NDA 218944 approval letter 2024What this evidence cannot show
Read verbatim from the NDA 218944 approval-letter PDF on 2026-09-18. The requirement names no paediatric endpoint, no analyte panel and no off-treatment follow-up, which is precisely the gap this experiment fills; the trial protocol is not public so those omissions are inferred from the requirement text rather than from the design.
Evidence that argues against this (1)
- The postmarketing requirement is the strongest justification this experiment has and it also limits it: the regulator has already mandated a trial addressing long-term risk, so an investigator-initiated surveillance panel must define what it adds beyond that trial rather than duplicating it. The obvious answer is paediatric and pre-pubertal endpoints, which the entry should state explicitly. FDA Drugs@FDA 2024
- E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.trial_registry
cAMeLot-1 (NCT04811560) enrols a paediatric cohort from 2 years of age with paediatric-specific performance-status and renal-function criteria, and lists no endocrine outcome measure among its primary or secondary outcomes.
cAMeLot-1: A Phase 1/2 Study of Bleximenib in Participants With Acute Leukemia · Janssen Research & Development, LLC · ClinicalTrials.gov 2026What this evidence cannot show
Read in full via the ClinicalTrials.gov API on 2026-09-18. A registry record summarises rather than reproduces the protocol, so laboratory collection not registered as an outcome measure cannot be excluded.
Evidence that argues against this (1)
- Enrolling from age 2 with paediatric-specific criteria establishes that children are being treated, not that growth, puberty or bone-mineral endpoints are being captured - none of which appears in the outcome list. A surveillance panel justified by this record is justified by an inference from enrolment criteria. ClinicalTrials.gov 2026
the minimum panel that discriminates on-target parathyroid effect from renal or vitamin-D secondary hyperparathyroidism
the off-treatment arm at 3, 6 and 12 months is the only proposed source of reversibility data for this class
- related to →A documented absence: not one endocrine outcome measure across the three lead menin-inhibitor trials
fills the endpoint gap the registry audit documents
operationalises the label's instruction to monitor bone growth in paediatric patients with height velocity and bone age
Design counter (Issa 2025, PMID 39121437): differentiation syndrome at grade >=3 in 16.0% is managed with systemic corticosteroids, which move glucose, IGF-1, bone ALP and height velocity - four of this panel's readouts. Covariate list added to data.covariates_required.
Feasibility counter (KOMZIFTI label): a second approved menin inhibitor reports none of this panel's analytes, so nothing can be borrowed from existing programmes and every value must be collected prospectively.
- ← supportsWhat the labels measure in a treated child: bone growth, in one sentence, and nothing else endocrine
Itemises exactly what the proposed surveillance panel would add, and confirms that none of it is currently required: no height velocity, bone age, Tanner staging, IGF-1, thyroid, gonadotrophin, glucose or DXA
P3: PTH recovers off drug, height velocity does not; the off-treatment arm this hypothesis needs is the one that experiment already specifies and FDA's PMR does not compel
Companion substudy: that panel covers the mineral and growth axes and deliberately excludes the glycaemic one; both should run on the same draws