(185) Predicting Findings of Two On-Going Phase IV Randomized Trials of Direct Oral Anticoagulants: An RCT-DUPLICATE Study.

Background: Direct oral anticoagulants (DOACs) such as apixaban and rivaroxaban, have transformed the management of thromboembolic disorders. Evidence from network meta-analyses of randomized clinical trials (RCTs) and observational studies have consistently found a lower risk of major bleeding for apixaban versus rivaroxaban. Citing clinical equipoise given that head-to-head RCTs had never been conducted, investigators launched RCTs in 2017 (COBRRA-VTE) and 2021 (COBRRA-AF) to compare apixaban and rivaroxaban on risk of bleeding in patients with venous thromboembolism (VTE) and atrial fibrillation (AF).

Objectives: We aimed to predict the results of two ongoing trials (COBRRA-AF and COBRRA-VTE) using database studies designed to emulate the corresponding RCT designs.

Methods: For each trial emulation, we developed and implemented a new initiator cohort design within three US health insurance claims databases (Medicare, Optum Clinformatics, Merative Marketscan; 2003-2019). We identified >100 pre-exposure patient characteristics to balance via propensity score matching (PSM), with a caliper of 1%. Pre-specified data and feasibility checks included evaluation of expected power and covariate balance. Trial emulation protocols were registered on clinicaltrials.gov (COBRRA-AF NCT05256797, COBRRA-VTE NCT05264168) before conducting inferential analyses with the primary outcome of major and clinically relevant non-major bleeding. Our primary analyses used on-treatment follow up to mimic the high adherence typically found in trials. Sensitivity analyses included intention-to-treat analyses and evaluation of two control outcomes. Results across databases were pooled using fixed effects meta-analyses.

Results: In the primary analysis for our COBRRA-AF emulation (177,866 PSM pairs), we found that in patients with AF, apixaban was associated with a 31% decreased risk of major and clinically relevant non-major bleeding [HR= 0.69 (0.66, 0.71)] compared to rivaroxaban. In the COBRRA-VTE emulation (14,202 PSM pairs), the primary analysis indicated that in patients with VTE, apixaban was associated with a 33% decreased risk of major and clinically relevant non-major bleeding compared to rivaroxaban [HR= 0.67 (0.58, 0.78)]. These results are consistent with previously published network meta-analyses and database studies.

Conclusions: Results from database studies that closely emulated the design of the ongoing COBRRA-AF and COBRRA-VTE trials indicate that apixaban has an approximate one-third lower risk of major and clinically relevant non-major bleeding compared to rivaroxaban in patients with AF and VTE. The ongoing RCTs are expected to be completed in Dec 2024 and 2026.