Safety and effectiveness of apixaban compared with warfarin among clinically-relevant subgroups of venous thromboembolism patients in the United States Medicare population

Abstract Background The AMPLIFY trial found significantly lower major bleeding (MB) and similar recurrent venous thromboembolism (VTE) risks associated with apixaban vs warfarin among patients with VTE. Objectives To compare MB, clinically-relevant non-major (CRNM) bleeding, and recurrent VTE risks among clinically-relevant subgroups of newly diagnosed elderly patients with VTE prescribed apixaban vs warfarin. Methods US Medicare patients prescribed apixaban or warfarin within 30 days post-VTE encounter were identified. Propensity score matching (PSM) was used to control for patient characteristics. Cox models were used to assess MB, CRNM bleeding, and recurrent VTE. Subgroup analyses were conducted for index VTE encounter type, index VTE diagnosis type, index VTE etiology, sex, and frailty. Results Post-PSM, 11,363 matched pairs of patients prescribed apixaban or warfarin were identified. Apixaban had lower MB (Hazard Ratio [HR]:0.76; 95% CI:0.64–0.91) and similar recurrent VTE risks (HR:1.04; 95% CI:0.75–1.43) vs warfarin. No significant interactions were observed between treatment and index VTE encounter type, index VTE diagnosis type, or sex for risk of MB, CRNM bleeding, or recurrent VTE. Significant interactions: frail patients prescribed apixaban had a 15% lower, while non-frail patients prescribed apixaban had 32% lower CRNM bleeding risk vs those prescribed warfarin. Patients with provoked VTE prescribed apixaban trended toward a higher, while those with unprovoked VTE trended toward a lower risk of recurrent VTE vs patients prescribed warfarin. Conclusions Apixaban was associated with significantly lower risks of MB and CRNM bleeding, and similar risk of recurrent VTE as compared with warfarin across the overall population and most subgroups.