Persistence to edoxaban treatment in patients with atrial fibrillation: Analysis from the Global ETNA-AF program

Abstract

Abstract Background/Introduction Direct oral anticoagulants (DOACs) are now the standard treatment for stroke prevention in most patients with atrial fibrillation (AF). However, many studies have reported suboptimal adherence and persistence to DOAC treatment. Data collected in routine care, such as in the Global Edoxaban Treatment in routiNe clinical prActice (ETNA)-AF program, may provide insight into factors associated with treatment persistence. Purpose Evaluate the persistence to edoxaban treatment in a large, unselected, real-world population using 2-year data from the Global ETNA-AF study. Methods The Global ETNA-AF program is a predefined integration of prospective, observational, non-interventional regional studies from Europe, Japan, and other Asian countries to evaluate the safety and effectiveness of edoxaban in patients with AF. Demographic and clinical characteristics were collected at baseline. In this analysis, patients were stratified by persistence to edoxaban treatment during a 2-year follow-up period (ie, persistence [PERS], suspension [SUS], permanent discontinuation [PERD], and suspension with later permanent discontinuation [SUS+PERD]). Results Overall persistence was high, with 87.1% of patients under edoxaban treatment at 2 years considering both groups that discontinued edoxaban at any time (PERD and SUS+PERD). Mean patient age, creatinine clearance, CHA2DS2-VASc, and HAS-BLED score were similar between the PERS (n = 20,053), SUS (n = 3088), PERD (n = 2936), and SUS+PERD (n = 503) groups (Table). In the PERS group, there were slightly fewer patients who received previous anticoagulation (vitamin K antagonist or DOAC) and were concomitantly using anti-platelet drugs at baseline. A larger proportion of SUS+PERD vs all other patients had a history of ischaemic stroke and/or major bleeding. Overall mortality was lower in patients in the PERS and SUS groups (0.79% [95% confidence interval {CI} 0.70–0.89] and 1.85% [95% CI 1.53–2.24], respectively) compared with patients who discontinued the treatment (PERD: 22.50% [95% CI 21.12–23.98]; SUS+PERD: 13.05% [95% CI 10.73–15.88]). Notably, persistence in the Global ETNA-AF program was generally higher than reported in GLORIA-AF at 2 years (70.9%) and XANTUS at 1 year (77.4%), despite a more compromised health profile especially considering the advanced patient age. Conclusions Treatment with edoxaban in routine clinical practice is associated with high persistence (87%) at 2 years in elderly patients with AF and multiple comorbidities. Persistence to edoxaban treatment was associated with better survival during the 2-year follow-up period. Further investigation is ongoing to determine the specific causes and consequences of edoxaban suspension and discontinuation to improve personalization of thromboembolic prophylaxis in patients with AF.Table 1