Post-Marketing Safety of the Menin Inhibitor Revumenib: A Bayesian Disproportionality Analysis of a National Database.

Abstract

Introduction: Revumenib is a first-in-class oral inhibitor of the menin-KMT2A protein-protein interaction, approved by the United States Food and Drug Administration in November 2024 for relapsed or refractory acute leukemia with a KMT2A translocation and in October 2025 for relapsed or refractory acute myeloid leukemia with a susceptible NPM1 mutation. The registrational program identified differentiation syndrome and (QT)-interval prolongation as adverse events of special interest. The post-marketing adverse event experience has not been systematically characterized.

Methods: A retrospective disproportionality analysis of the FDA Adverse Event Reporting System (FAERS, 14,104,743 reports) was performed, restricted to reports in which revumenib was named as a primary or secondary suspect product (n = 283). Frequentist (proportional reporting ratio with Pearson χ2; reporting odds ratio with 95% CI) and Bayesian (information component with lower 2.5% bound; empirical Bayes geometric mean [EBGM] with lower 5% bound) metrics were calculated for every adverse event reported in association with the drug. A signal required more than three reports together with concurrent fulfilment of all four disproportionality criteria.

Results: Twenty-eight adverse events were evaluated; nineteen met the signal definition. Differentiation syndrome produced the strongest disproportionality (EBGM 15.38, EBGM₀₅ 9.17). Cytopenias of every lineage, electrocardiogram QT prolonged (EBGM₀₅ 3.67), and lower magnitude constitutional and gastrointestinal events were also signaled. Nine events did not constitute signals, principally because of the case-count threshold.

Conclusion: Post-marketing reporting corroborates the safety profile established in AUGMENT-101 and identifies no novel toxicity category.