Effectiveness and safety of edoxaban in 27,333 patients from ETNA-AF with and without a history of intracranial haemorrhage after 2 years of treatment

Abstract

Abstract Funding Acknowledgements Type of funding sources: Private company. Main funding source(s): This study was sponsored by Daiichi Sankyo, Inc. Medical writing and editorial support were provided by Atreju Lackey, PhD of AlphaBioCom, LLC, and funded by Daiichi Sankyo, Inc. Background/Introduction Once-daily edoxaban significantly reduced the risk of intracranial haemorrhage (ICH) compared with well-managed warfarin in atrial fibrillation (AF) patients in the ENGAGE AF-TIMI 48 trial. The effectiveness and safety of edoxaban in patients with prior ICH is unknown. Purpose To compare the effectiveness and safety of edoxaban in AF patients with or without a history of ICH. Methods The Global ETNA-AF programme is composed of and, thus, integrates data from multiple prospective, observational, and noninterventional regional studies collecting data of AF patients treated with edoxaban for stroke prevention. This snapshot analysis presents global and regional baseline characteristics with medical history and 2-year annualised rates of all-cause mortality, stroke (haemorrhagic, ischaemic, any), and bleeding (major bleeding [MB] including ICH, clinically relevant nonmajor bleeding [CRNMB], any bleeding), in patients with or without ICH history. Results Overall, 27,333 patients from Europe, Japan, South Korea, and Taiwan were analysed, including 367 with prior ICH and 26,966 without prior ICH. There were proportionally fewer patients with a history of ICH in the European population. Patients with a history of ICH were older (P=0.006), had a lower body mass index (P<0.0001), had a lower creatinine clearance (P=0.0001), and had more comorbidities, with a higher percentage of patients with a history of stroke, transient ischaemic attack (TIA), or MB (Table 1); the higher level of comorbidities noted in patients with a history of ICH was also reflected by higher baseline CHA2DS2-VASc and HAS-BLED scores (Table 1). Patients with a history of ICH were more likely receiving 30 mg edoxaban at baseline, whereas patients without ICH history were more often on 60 mg edoxaban (each P<0.0001). In patients with vs without ICH history, all-cause mortality (5.10% vs 3.14%; P=0.01), ischaemic stroke (1.79% vs 0.73%; P=0.006), and any stroke rates (3.25% vs 0.95%; P<0.0001) were higher (Table 2). Patients with vs without ICH history had higher annualised rates of MB (2.50% vs 1.00%; P=0.001), ICH (1.42% vs 0.27%; P<0.0001), haemorrhagic stroke (1.42% vs 0.20%; P<0.0001), CRNMB (2.49% vs 1.40%; P=0.04), and any bleeding (7.57% vs 4.27%; P=0.001), but these rates were low compared to other high-risk populations. ICH was not selected as a predictor of cardiovascular outcomes on multivariate prediction modelling. ICH had non-significant effects in predicting all-cause death (HR 1.22), ischemic stroke (HR 1.14), and major bleeding (HF 1.37) and repeat ICH (HR 1.94). Conclusions Patients with a history of ICH are a small, elderly, multimorbid subgroup of patients with AF. Treatment with the non-vitamin K antagonist oral anticoagulant edoxaban resulted in relatively low rates of major events.