Quantitative Model-Informed Dose Selection for a Milvexian Phase III Study in Patients With Atrial Fibrillation.

Abstract

Milvexian, an activated factor XI inhibitor, is being evaluated in the LIBREXIA phase III program for the prevention of thrombotic events, including in patients with atrial fibrillation (LIBREXIA AF). To guide the selection of a phase III dose regimen for LIBREXIA AF, two main approaches were used: (1) population pharmacokinetics and exposure-response modeling of milvexian phase I and phase II (AXIOMATIC-TKR) study data was performed to characterize the relationship between milvexian population pharmacokinetics and clinical efficacy, safety, and coagulation biomarkers; and (2) a model-based meta-analysis was developed using published clinical trial data from comparator anticoagulants in total hip or knee replacement patients. When used to simulate exposures for an atrial fibrillation population, the milvexian 100 mg twice-daily dose regimen had a concentration maximum peak-to-trough ratio of 1.35. At this regimen, the simulated median venous thromboembolism rate was 9.4% (lower than enoxaparin at ~21%). No relationship between bleeding and milvexian exposure was observed. Simulations using the model-based meta-analysis estimated that milvexian 100 mg twice-daily would lead to a systemic embolism odds ratio of 0.84 (95% CI: 0.65, 1.12) compared with apixaban 5 mg twice-daily for patients with atrial fibrillation. This integrated model-informed dose-selection approach supported 100 mg twice-daily as a regimen of milvexian to be evaluated in the phase III milvexian LIBREXIA AF study.