Effectiveness and safety of apixaban vs warfarin among venous thromboembolism patients using five US databases: a subgroup analysis of chronic liver disease

Abstract

Abstract Background Venous thromboembolism (VTE) is a serious disease in the United States affecting approximately 1 in 1000 patients each year. Patients with chronic liver disease (CLD) are at an increased risk of VTE and major bleeding (MB). Currently, insufficient clinical and real-world evidence exists on the efficacy/effectiveness and safety of apixaban or warfarin in VTE patients with CLD. Purpose To evaluate the risk of recurrent VTE, MB, and clinically relevant non-major (CRNM) bleeding among VTE patients initiating apixaban or warfarin stratified by CLD status. Methods VTE patients ≥18 years of age (≥65 years for Medicare) initiating apixaban or warfarin were identified from CMS Medicare and four commercial claims databases. To balance the characteristics between apixaban and warfarin patients, stabilized inverse probability treatment weighting (IPTW) was conducted. Post-IPTW, subgroup interaction analysis was conducted to evaluate whether treatment effects of apixaban vs. warfarin were consistent across patients with and without a diagnosis of CLD. Cox proportional hazard models were used to evaluate the interaction of the treatment (apixaban vs. warfarin) and CLD on recurrent VTE, MB, and CRNM bleeding. The statistical significance of the interaction was set to p-value <0.10. Results A total of 60,786 apixaban and 94,333 warfarin patients with VTE were eligible for analysis. Post-IPTW, all patient characteristics were balanced between the apixaban and warfarin treatment cohorts. Apixaban treated patients had significantly lower risk of recurrent VTE, MB, and CRNM bleeding compared to warfarin patients (Figure). In the IPTW weighted population, 4,766 (7.8%) apixaban patients and 6,320 (6.7%) warfarin patients had a diagnosis of CLD. For the apixaban or warfarin patients, those with a diagnosis of CLD were generally younger (mean 64.0–65.2 vs 66.9 years), had higher Charlson comorbidity index scores (mean 3.8–3.9 vs 2.1) and were more likely to have an inpatient VTE event (67.8–69.5% vs 53.0–53.2%) or provoked VTE events (66.2–67.8% vs 55.4–55.5%) compared to patients without a diagnosis of CLD. The incidence rate of recurrent VTE, MB, and CRNM bleeding was higher among VTE patients with CLD than without CLD and was also higher for patients treated with warfarin compared to those treated with apixaban regardless of CLD status (Figure). There were no significant interactions observed between treatment and CLD status for recurrent VTE, MB or CRNM (Figure). Conclusion Treatment with apixaban had a lower risk of recurrent VTE, MB, and CRNM bleeding compared to treatment with warfarin. The benefits of apixaban were consistently observed among subgroups of VTE patients with and without CLD. Additional studies are needed to evaluate VTE patients with CLD. Funding Acknowledgement Type of funding sources: Private company. Main funding source(s): Bristol-Myers Squibb Company and Pfizer, Inc.