Rivaroxaban versus low-molecular-weight heparin for venous thromboembolism in gastrointestinal and pancreatobiliary cancer

Abstract

Introduction and Objectives: Although low-molecular-weight heparin (LMWH) remains the standard treatment for venous thromboembolism (VTE) in patients with active cancer, a factor Xa inhibitor such as rivaroxaban is increasingly used without clinical evidence. We compared the incidence of bleeding and other treatment outcomes using rivaroxaban and LMWH for the treatment of VTE in patients with gastrointestinal and pancreatobiliary cancer (GI cancer). Methods: This single-center retrospective analysis included patients with VTE associated with GI cancer, who were treated with either rivaroxaban or LMWH. The primary end-point was the incidence of clinically relevant bleeding. Secondary outcomes included the incidence of major bleeding, recurrent VTE, and mortality. Results: Of 281 patients, 78 received rivaroxaban and 203 received LMWH. Clinically relevant bleeding occurred in 20 patients (26%) in the rivaroxaban group and 31 (15%) in the LMWH group (P=0.043). There was no statistically significant difference in the VTE recurrence rate (4% with rivaroxaban vs. 4% with LMWH, P>0.999) or incidence of major bleeding (5% with rivaroxaban vs. 9% with LMWH, P=0.296). Multivariate Cox proportional hazards analysis for cancer type, stage, chemotherapy history, and Eastern Cooperative Oncology Group performance status showed a 1.904-fold higher risk of bleeding with rivaroxaban (1.031 to 3.516, P=0.040). Rivaroxaban use was not associated with a higher hazard ratio than LMWH use for all-cause mortality (HR 1.00, P=0.999). Conclusions: Rivaroxaban use was associated with more bleeding than LMWH use in GI cancer patients.