M. Green,1 T. Leil,2 C. Frost,2 X. Wang,2 R. Wada1; 1Quantitative Solutions, Menlo Park, CA, 2Bristol-Myers Squibb, Princeton, NJ

BACKGROUND: Apixaban and rivaroxaban are both direct, reversible Factor Xa inhibitors used for prevention of thromboembolism in patients with nonvalvular atrial fibrillation (NVAF). Apixaban is administered as a 5 mg twice daily (BID) regimen, while rivaroxaban is given as a 20 mg once daily (QD) regimen. In this study, the adherence rates for QD and BID dosing regimens, in conjunction with the population PK (PPK) models for apixaban and rivaroxaban, were used to predict whether differences in adherence would have meaningful effects on exposures in NVAF patients.
METHODS: Apixaban and rivaroxaban NVAF population exposures were simulated using the published PPK models, with population characteristics sampled from the apixaban ARISTOTLE study. Adherence levels and patterns were incorporated into the simulations based published studies of BID and QD adherence (Coleman et al. mean adherence of 93.1% for QD and 86.2 % for BID; Laliberte et al. mean adherence of 86% for QD; 84 % for BID). The impact of adherence on exposure was primarily determined by evaluating the fraction of the 50 day treatment period that subjects are within 50% of the population median concentration.
RESULTS: The fraction of time that exposure was within 50% of the median was higher for apixaban (81% for apixaban vs. 41% for rivaroxaban) using the Coleman et al. adherence estimates. The results were very similar when using the adherence estimates from Laliberte et al. (79% for apixaban vs. 36% for rivaroxaban). These results largely reflect the greater variability in exposure for rivaroxaban.
CONCLUSION: PPK simulations in the NVAF population indicate that despite lower adherence for the BID regimen, exposures for apixaban are expected to be maintained within 50% of the median more consistently than the QD regimen of rivaroxaban.