Apixaban is noninferior to dalteparin for venous thromboembolism associated with cancer

Reference

Agnelli G, Becattini C, Meyer G, et al, for the Caravaggio Investigators. Apixaban for the treatment of venous thromboembolism associated with cancer. N Engl J Med 2020;382(17):1599-1607.

Synopsis

Apixaban is a direct oral anticoagulant that in previous trials has had lower rates of major bleeding than other similar agents. In this study, 1155 patients with active cancer (diagnosed within 6 months, currently being treated, or recurrent cancer) who experienced an episode of deep vein thrombosis (DVT) or pulmonary embolism (PE) were identified. They were randomized to receive apixaban (10 mg twice daily for 1 week, then 5 mg twice daily) or dalteparin (200 IU/kg once daily subcutaneously for 1 month, then 150 IU/kg [maximum dose 18,000 IU]) for 6 months. This was an open-label, industry-funded, noninferiority trial. Allocation was concealed, and adjudication of cause of death was masked, but it is not clear that the diagnosis of PE or DVT was given by persons masked to treatment assignment. Groups were balanced at the start of the study and analysis was by modified intention to treat for all patients who received at least one dose of the study drug. The patients' mean age was 67 years, approximately half had DVT alone as the index event, and approximately half had PE with or without DVT as the index event. The study was designed as a noninferiority trial and the upper noninferiority margin for the hazard ratio was set to 2.0. That meant apixaban could be up 1.99 times worse and still be considered noninferior. The primary outcome of PE or DVT was noninferior for apixaban (5.6% for apixaban, 7.9% for dalteparin; hazard ratio [HR] 0.63; 95% CI 0.37 - 1.07; P = .09 for superiority; P < .001 for noninferiority). Most of this apparent benefit was due to fewer PEs. There was no difference between groups in major bleeding (3.8% vs 4.0%). There were numerically fewer deaths in the apixaban group, but this difference was not statistically significant (135 vs 153; HR 0.82; 0.62 - 1.09). Patients with brain tumors, cerebral metastases, and acute leukemia were excluded, so we can't extrapolate the results to those populations.