PII-102 - CLINICAL DRUG-DRUG INTERACTION STUDY INTEGRATED WITH PBPK MODELLING PROVIDES INSIGHTS INTO THE INDUCTION POTENTIAL OF MAVACAMTEN ACROSS CYP2C19 PHENOTYPES.

S. Merali, C. Sychterz, V. Perera, L. Gaohua, V. Florea, B. Murthy; Bristol Myers Squibb.

Background: Mavacamten is a potential inducer of cytochrome P450 (CYP) 3A4, and as such could reduce the effectiveness of concomitant drugs that are metabolized by this enzyme, such as midazolam. The aim of this study was to determine if repeat doses of mavacamten affect the exposure to midazolam.

Methods: This was a single-center open-label study in healthy participants who could be normal, rapid, or ultrarapid CYP2C19 metabolizers. Period 1: participants received midazolam 5 mg on day 1. Period 2: participants received mavacamten 25 mg on Days 2–3, and 15 mg on Days 4‒16. Period 3: participants received mavacamten 15 mg and midazolam 5 mg on Day 17. Plasma concentrations of mavacamten, midazolam and its metabolite, 1′-hydroxymidazolam, were measured. A physiologically based pharmacokinetic (PBPK) model was used to simulate mavacamten-mediated CYP3A4 induction with midazolam by CYP2C19 phenotypes.

Results: In total, 13 participants were enrolled (46.2% female; mean [SD] age: 34.0 [9.0] years; mean [SD] body mass index: 24.5 [2.8]); eight participants (61.5%) were CYP2C19 normal metabolizers and 5 (38.5%) were rapid metabolizers. Coadministration of mavacamten with midazolam resulted in reduction in the maximum serum concentration of midazolam (Cmax), the area under the curve (AUC) extrapolated to infinite time (AUC0-inf), and AUC from time 0 to last measurable concentration (AUC0-last) by 7%, 13%, and 24%, respectively (Figure). For 1′-hydroxymidazolam, AUC0-inf increased by 20% and AUC0-last by 11%. A total of 10 participants experienced adverse events, the majority of which were mild in severity. The PBPK model demonstrated that all simulated midazolam and mavacamten PK parameters were within 2-fold of the observed values. The PBPK model predicted reductions in midazolam Cmax and AUC over 24-hour dosing interval (AUC0–24) which varied by CYP2C19 phenotype, with greater reductions in poor metabolizers compared with ultrarapid metabolizers (Figure). The overall impact of mavacamten on midazolam Cmax and AUC0–24 was predicted to be weak regardless of CYP2C19 phenotype.

Conclusion: At clinically relevant exposures, mavacamten had a limited impact on the potential for CYP3A4 induction. PBPK modelling confirmed an overall weak impact of mavacamten on midazolam exposure regardless of CYP2C19 phenotype.