Derivation of the non-inferiority margin for the evaluation of direct oral anticoagulants in the treatment of venous thromboembolism

BACKGROUND Direct oral anticoagulants that target a single coagulation factor have been developed as an alternative to standard therapies with heparin and/or vitamin K antagonists. The purpose of this study was to derive non-inferiority margins suitable for randomised clinical studies designed to evaluate these agents for the treatment of venous thromboembolism (VTE). METHODS We performed a systematic review to derive non-inferiority margins suitable for use in studies evaluating direct oral anticoagulants for the treatment of VTE. A PubMed search identified publications that evaluated current standard treatment versus placebo, 'no treatment' or 'less intensive treatment' in patients with symptomatic deep vein thrombosis (DVT) and/or pulmonary embolism (PE). Publications were eligible if they had a randomised study design, included patients with symptomatic DVT and/or PE, used objective diagnostic methods to document the index event and reported objectively confirmed symptomatic recurrent VTE. RESULTS Fourteen publications were included in the analysis. Recurrent VTE occurred in 25 (1.5%) out of 1715 patients who received current standard of care and in 157 (9.2%) out of 1711 patients who received placebo, 'no treatment' or 'less intensive treatment', for an odds ratio of 0.18 (95% confidence interval, 0.14-0.25; test for heterogeneity, p=0.87). In order to preserve 50% or 75% of the established treatment effect using a linear scale, the corresponding thresholds for non-inferiority equalled 2.50 and 1.75, respectively. CONCLUSIONS This systematic review and statistical approach determined non-inferiority margins suitable for use in studies of direct oral anticoagulants for the treatment of DVT and/or PE.