Abstract 9030: Effects of Edoxaban on Clinical Outcomes in Extremely Elderly Patients (aged ≥ 90 Years) With Atrial Fibrillation: A Subanalysis of Data from the ELDERCARE-AF Trial

Introduction: Atrial fibrillation (AF) affects about 10% of the very elderly (≥ 80 years), and its prevalence increases exponentially with age. Data are lacking on the efficacy and safety of anticoagulants in extremely elderly AF patients (≥ 90 years). We tested the hypotheses that such patients are at increased risk of stroke or systemic embolism (SSE) and could benefit from low-dose edoxaban. Methods: This was a subanalysis of ELDERCARE-AF, a multicenter, double-blind, phase 3 study in which very elderly Japanese patients with non-valvular AF (NVAF) considered ineligible for standard anticoagulants were randomly allocated to low-dose edoxaban (15 mg once daily) or placebo. Outcomes were compared between the three prespecified subgroups of age at baseline 80-84, 85-89, and ≥ 90 years. Primary efficacy and safety endpoints were incidence of SSE and ISTH-defined major bleeding, respectively (% per patient-year). Results: A total of 984 patients were randomized. Mean (± SD) age and CHADS 2 score of the patients aged ≥ 90 years were 92.3 (2.1) years and 3.1 (1.1), respectively. SSE occurred in 14/354 (2.8%), 25/374 (5.0%), and 20/256 (6.1%) of patients aged 80-84, 85-89, and ≥ 90 years, respectively; its incidence was significantly higher in the ≥ 90 years vs 80-84 years group (HR = 2.16 [95% CI, 1.09-4.27], P = 0.03). Major bleeding occurred in 6/353 (1.3%), 12/373 (2.6%), and 13/256 (4.3%) of patients aged 80-84, 85-89, and ≥ 90 years, respectively; its incidence was significantly higher in the ≥ 90 years vs 80-84 years (HR = 3.42 [1.30-9.01], P = 0.01). In this patient group ≥ 90 years with higher SSE and major bleeding incidences, edoxaban showed a significantly lower SSE incidence than placebo (4/129 [2.4%] vs 16/127 [10.1%]; HR = 0.23 [0.08-0.68], P = 0.008); however, edoxaban showed a greater tendency toward major bleeding than placebo (10/129 [6.5%] vs 3/127 [2.1%]; HR = 3.02 [0.82-11.21], P = 0.10), and significantly higher incidence of gastrointestinal bleeding (9/129 [5.9%] vs 1/127 [0.7%]; HR = 8.37 [1.04-67.07], P < 0.05). Conclusions: In extremely elderly Japanese NVAF patients, low-dose edoxaban may be an option for preventing SSE. The prophylactic benefit must be balanced against the increased risk of major bleeding, particularly gastrointestinal bleeding.