Effect of 15-mg Edoxaban on Clinical Outcomes in 3 Age Strata in Older Patients With Atrial Fibrillation A Prespecified Subanalysis of the ELDERCARE-AF Randomized Clinical Trial

IMPORTANCE Long-term use of oral anticoagulants (OACs) is necessary for stroke prevention in patients with atrial fibrillation (AF). The effectiveness and safety of OACs in extremely older patients (ie, aged 80 years or older) with AF and at high risk of bleeding needs to be elucidated. OBJECTIVE To examine the effects of very low-dose edoxaban (15 mg) vs placebo across 3 age strata (80-84 years, 85-89 years, and >= 90 years) among patients with AF who were a part of the Edoxaban Low-Dose for Elder Care Atrial Fibrillation Patients (ELDERCARE-AF) trial. DESIGN, SETTING, AND PARTICIPANTS This prespecified subanalysis of a phase 3, randomized, double-blind, placebo-controlled trial was conducted from August 5, 2016, to December 27, 2019. Patients with AF aged 80 years or older who were not considered candidates for standard-dose OACs were included in the study; reasons these patients could not take standard-dose OACs included low creatinine clearance (<30 mL per minute), low body weight (<= 45 kg), history of bleeding from critical organs, continuous use of nonsteroidal anti-inflammatory drugs, or concomitant use of antiplatelet drugs. Eligible patients were recruited randomly from 164 hospitals in Japan and were randomly assigned 1:1 to edoxaban or placebo. INTERVENTIONS Edoxaban (15 mg once daily) or placebo. MAIN OUTCOMES AND MEASURES The primary efficacy end point was the composite of stroke or systemic embolism. The primary safety end point was International Society on Thrombosis and Hemostasis-defined major bleeding. RESULTS A total of 984 patients (mean [SD] age: age group 80-84 years, 82.2 [1.4] years; age group 85-89 years, 86.8 [1.4] years; age group >= 90 years, 92.3 [2.1] years; 565 women [57.4%]) were included in this study. In the placebo group, estimated (SE) event rates for stroke or systemic embolism increased with age and were 3.9% (1.2%) per patient-year in the group aged 80 to 84 years (n = 181), 7.3% (1.7%) per patient-year in the group aged 85 to 89 years (n =184), and 10.1% (2.5%) per patient-year in the group aged 90 years or older (n = 127). A 15-mg dose of edoxaban consistently decreased the event rates for stroke or systemic embolism with no interaction with age (80-84 years, hazard ratio [HR], 0.41; 95% CI, 0.13-1.31; P = .13; 85-89 years, HR, 0.42; 95% CI, 0.17-0.99; P = .05; >= 90 years, HR, 0.23; 95% CI, 0.08-0.68; P = .008; interaction P = .65). Major bleeding and major or clinically relevant nonmajor bleeding events were numerically higher with edoxaban, but the differences did not reach statistical significance, and there was no interaction with age. There was no difference in the event rate for all-cause death between the edoxaban and placebo groups in all age strata. CONCLUSIONS AND RELEVANCE Results of this subanalysis of the ELDERCARE-AF randomized clinical trial revealed that among Japanese patients aged 80 years or older with AF who were not considered candidates for standard OACs, a once-daily 15-mg dose of edoxaban was superior to placebo in preventing stroke or systemic embolism consistently across all 3 age strata, including those aged 90 years or older, albeit with a higher but nonstatistically significant incidence of bleeding.