157: Pharmacokinetics & Exposure-Response Analyses of Revumenib Support Dosing in a Broad Population of Patients With Relapsed/Refractory Acute Leukemias

Statement of Purpose, Innovation or Hypothesis: Revumenib, a selective, orally bioavailable small molecule inhibitor of the menin-lysine methyltransferase 2A (KMT2A) interaction, is being investigated for treatment of KMT2A-rearranged (KMT2Ar) or NPM1-mutated (NPM1m) relapsed/refractory (R/R) acute leukemias. This analysis sought to determine the pharmacokinetic (PK) and exposure-response (ER) relationship of revumenib for the treatment of patients with R/R acute leukemias.

Description of Methods & Materials: The AUGMENT-101 Phase 1/2 [NCT04065399] dose-escalation and expansion arms evaluated revumenib in adult and pediatric participants with R/R KMT2Ar or NPM1m acute leukemias. Revumenib was administered orally in 28-day cycles as a capsule, tablet, or oral-solution and administered ± CYP3A4 inhibitors (CYP3A4i). All patients (N=251) who had evaluable PK profiles were included in this analysis. Physiologically-based-PK (PBPK), population PK (PopPK), ER-efficacy and -safety models were developed to determine or justify doses and assess drug-drug interactions. Monte-Carlo simulations (N=1000) were conducted to support the selection of proposed registrational doses.

Data & Results: Revumenib was metabolized primarily by CYP3A4. PBPK analysis showed co-administration of strong and moderate CYP3A4-inducers decreased revumenib exposures ( >2.5-fold). PopPK analysis showed no relevant change in revumenib exposures (AUC and C_max) with moderate or weak CYP3A4i; and a ~2-fold increase in exposures with strong CYP3A4i. PK of revumenib and M1, the only circulating metabolite exceeding 10% of total drug-related material, was dose-proportional across the therapeutic dose-range of 113–276mg (or BSA equivalent dose) q12h when administered ± strong CYP3A4i. PopPK analysis found no clinically meaningful effect of formulation (tablet [160 or 270mg] vs. capsules [163 or 276mg] and oral solution) or food effect (low fat meal vs. fasted) for revumenib and M1. No clinically relevant differences in the PK of revumenib were observed in patients with body weights between 40–146 kg, age ≥6-months, race, gender, mild-to-moderate renal and hepatic impairment. For patients < 40 kg, ~30–58% increase in revumenib exposure was observed. In pediatric patients < 6-months, 2.3-fold increase in revumenib exposures were observed. Mild-to-moderate renal or hepatic impairment did not have a significant impact on PK. Safety and PK of revumenib in patients with severe renal impairment, end-stage renal disease, or severe hepatic impairment have not been studied (no recommended dose has been established). All dose levels (113mg to 339mg q12h) in AUGMENT-101 were within an efficacious range, leading to statistically significant downregulation of leukemogenic genes, or upregulation of myeloid differentiation genes, with no clear correlations between exposures and expression. 270mg BID without strong CYP3A4i or 160mg with strong CYP3A4i represents the dose necessary to maintain target-exposures above the preclinical-target in ≥90% of patients (Figure). There were no statistically significant ER-efficacy relationships for key efficacy endpoints (response-rates or duration-of-response) within the dose-ranges tested; however, higher exposures were associated with an increase in adverse events, which were manageable with dose reductions.

Interpretation, Conclusion or Significance: Revumenib is a well-tolerated first-in-class menin inhibitor with acceptable PK for oral daily dosing via flat- or BSA-based dosing, administered ± a strong CYP3A4i and as fasted or with a low-fat meal. The proposed 270mg BID dose without a strong CYP3A4i and 160mg BID dose with a strong CYP3A4i, were confirmed as efficacious with a manageable safety profile.

Disclosures: Syndax: YWL, SSm consultants, SSu employee; EC, JS, HVN, RN, NM employees; ATW, GCI trial investigators.

Citations/References:

Additional Information/Authors: YuWei Lin1, Steven Smith2, Enoch Cobbina2, Joshua Stewart2, Austin T. Wilson3, Huy Van Nguyen2, Ghayas C. Issa3, Rob Nordness2, Nicole McNeer2, Soujanya Sunkaraneni2