Edoxaban treatment in atrial fibrillation in routine clinical care: One‐year outcomes of the prospective observational ETNA‐AF study in South Korean patients

Abstract Background The real‐world outcomes of edoxaban treatment in patients with atrial fibrillation (AF) were analyzed in the ETNA‐AF (Edoxaban Treatment in Routine Clinical Practice) study involving data from multiple regional registries. This report addresses effectiveness and safety of edoxaban in the Korean ETNA‐AF population. Methods One‐year data from 1887 Korean ETNA‐AF participants were analyzed according to edoxaban dose and patient age and compared with results of other ETNA‐AF registries. Results Approximately 70% of patients received the recommended doses of edoxaban (60 mg/30 mg); non‐recommended 60 mg and 30 mg doses were prescribed to 9.6% and 19.8% of the patients, respectively. The proportions of reference age (<65 years), youngest‐old (65–74 years) and middle‐old/oldest‐old (≥75 years) groups were 21.4%, 40.2%, and 38.4%, respectively. Incidence of major or clinically relevant nonmajor bleeding was similar within dose (0.57%–1.71%) and age subgroups (1.26%–1.63%). Incidence of net clinical outcome, a composite of stroke, systemic embolic event, major bleeding, and all‐cause mortality, was also comparable among dose subgroups (1.14%–3.10%) and age subgroups (2.28%–2.78%). The percentage of Korean patients receiving non‐recommended 30 mg (19.8%) was over twice that of the European population (8.4%). However, the clinical outcomes were generally similar among different populations included in the ETNA‐AF study. Conclusions The outcomes in the Korean ETNA‐AF population are like those in the global ETNA‐AF population, with overall low event rates of stroke, major bleeding and all‐cause mortality across age and dose subgroups. Edoxaban can be used effectively and safely in specific populations of Korean AF patients, including the elderly.


| INTRODUC TI ON
Atrial fibrillation (AF) is the most common cardiac arrhythmia, with 1%-4% of the general population being affected. [1][2][3][4] The prevalence of AF in South Korea has been increasing over the last two decades, with a 1.7-fold increase in the incidence observed between 2008 and 2015. 5 AF poses multiple risks, with the two most graves being a two-fold increase in mortality risk and a five-fold increase in the risk of ischemic stroke when compared with persons without AF. 6,7 Further, hospitalizations due to AF and its complications, such as aggravation of heart failure, thromboembolic complications, and acute arrhythmia management, constitute a significant burden for healthcare systems, and this form of arrhythmia has a detrimental effect on the quality of life and exercise tolerance of the patients. 8,9 Many of the risks mentioned above, particularly stroke and venous thromboembolism, can be effectively prevented with anticoagulation therapy. Until 2009, the only oral anticoagulants available for patients with AF were warfarin and vitamin K antagonists (VKAs). 10 While effective, those drugs have a serious drawback in the form of a narrow therapeutic index which necessitates constant monitoring and dose adjustment. Due to resultant inconvenience, patient adherence to warfarin/VKA treatment tends to be poor, which rather contributes to increasing the risk of bleeding and thromboembolic events. [11][12][13][14] Most of the drawbacks mentioned above were overcome by new non-VKA oral anticoagulants (NOACs), including edoxaban. Unlike VKAs, NOACs have a rapid onset and offset of action and, due to their predictable anticoagulant effects, independent of dietary intake of vitamin K, can be administered at fixed doses with no need for routine coagulation monitoring. [15][16][17][18][19] NOACs are being used more frequently to prevent stroke in patients with AF because of their ease of administration and comparative efficacy compared with warfarin in reducing thromboembolism and major bleeding. 20 Edoxaban, one of the NOACs, was approved in South Korea and many other countries for the prevention of stroke and systemic embolic event (SEE) in patients with nonvalvular AF. It was also approved to treat deep vein thrombosis and pulmonary embolism. 21 The approval was based on the results of the ENGAGE AF-TIMI 48 (Effective Anticoagulation with Factor Xa Next Generation in Atrial Fibrillation-Thrombolysis in Myocardial Infarction 48) and

Edoxaban Treatment in Routine Clinical Practice for Patients
With Nonvalvular Atrial Fibrillation (ETNA-AF) was designed as a noninterventional study to collect data from edoxaban-treated patients with AF from multiple regional registries. [23][24][25] The results for Korean patients participating in the study were presented elsewhere. 26,27 Although the safety and effectiveness of edoxaban in routine clinical practice were reported for Korean and Taiwanese AF patients through the ETNA-AF registry, the data associated with the Korean special populations is limited. The aim of the present report, also based on the results of the ETNA-AF study, is to address some specific aspects of edoxaban treatment in Asians, especially the South Korean population, such as the effectiveness and safety of the drug according to its dose and patient age. Additionally, the outcomes of edoxaban treatment and baseline characteristics of Korean patients were compared with those of the other regional countries in the ETNA-AF registry to observe the ethnic difference.

| Study design
This multicenter, prospective, noninterventional study (ETNA-AF-KOR-TWN; NCT02951039) was conducted in patients treated with edoxaban for AF both in Korea and Taiwan as a regional study under the global scheme of the ETNA program. The ETNA program was designed to integrate the safety and effectiveness data of edoxaban indicated for AF or VTE in routine practice conditions across Europe, Japan, and East and Southeast Asia. The detailed study design and rationale of the Global ETNA program were previously published, 23,24 and the overall plan of the ETNA-AF-KOR-TWN of the European population (8.4%). However, the clinical outcomes were generally similar among different populations included in the ETNA-AF study.

Conclusions:
The outcomes in the Korean ETNA-AF population are like those in the global ETNA-AF population, with overall low event rates of stroke, major bleeding and all-cause mortality across age and dose subgroups. Edoxaban can be used effectively and safely in specific populations of Korean AF patients, including the elderly.

K E Y W O R D S
major bleeding, non-vitamin K antagonist oral anticoagulants, real-world, registry, stroke prevention stayed the same. Here, we present the study focused on the Korean registry. Two-thousand patients were to be enrolled under different care settings (primary or secondary care or other specialties) in Korea, and each patient was to be followed up for 2 years. Here, we report a 1-year follow-up. To adhere to the noninterventional nature of the study, all clinical procedures and decisions, including modification of edoxaban dose, continuation or discontinuation of edoxaban treatment, and concomitant treatment, were solely based on the physician's clinical judgement for the best interest of the patient.

| Target patients
Unselected patients with nonvalvular AF who were treated with edoxaban for a reduction in the risk of stroke and systemic embolism according to the approved local label were eligible to participate in the study if they provided written informed consent. The only exclusion criterion was simultaneously participating in an interventional study. Patients meeting the eligibility criteria were to be consecutively enrolled in each center to the extent possible.

| Study variables and outcomes
At baseline, patient demographics and clinical characteristics, laboratory parameters, medical history, and details of edoxaban therapy were collected. Follow-up data were collected approximately 12 and 24 months after baseline. If a patient's edoxaban treatment was permanently discontinued, the patient was followed up for one more year after the discontinuation of edoxaban or until the end of the 2-year follow-up period, whichever came first. As safety outcomes, bleeding events such as major and clinically relevant nonmajor (CRNM) bleeding are presented. For effectiveness outcomes, strokes, systemic embolic events, myocardial infarction, CV mortality, and all-cause mortality are included. Net clinical outcome is a composite of ischemic or hemorrhagic stroke, SEE, major bleeding, and all-cause mortality.

| Statistical analysis
All analyses were exploratory and descriptive. Only events with a start date later than the date of baseline and before the date of baseline +365 days were included. Continuous variables were summarized as mean ± standard deviation (SD) and categorical variables as frequencies (percentages). The main safety and effectiveness outcomes were presented as annualized event rates (% per year) with 95% confidence intervals (CIs). All results were presented for subgroups based on age and edoxaban dose. Age subgroups were divided into reference (<65 years), youngest-old (65-74 years), and middle/oldest-old (≥75 years) groups. For major outcomes, HR of the two older groups relative to the reference age group were presented with 95% CI. Dose subgroups were divided into patients who met none of the dose-reduction criteria(recommended 60 mg, non-recommended 30 mg) and patients who met ≥1 of the dose-reduction criteria (nonrecommended 60 mg, recommended 30 mg) groups. For major outcomes, HR between recommended groups and non-recommended groups were presented with 95% CI. Event rates were presented alongside those reported in other registries of the ETNA-AF program and ENGAGE AF-TIMI 48 study.

| Ethical standards
The study was conducted in compliance with the Declaration of Helsinki and Guidelines for Good Pharmacoepidemiology Practice.
The study was approved by the institutional review boards for all participating centers before study initiation. All patients provided written informed consent.
F I G U R E 1 Study cohort diagram. AF, atrial fibrillation; CRF, case report form; NVAF, nonvalvular atrial fibrillation.

| Patient characteristics
During the period between February 2017 and April 2018, a total of 1928 patients were registered in the study from 37 clinics and hospitals in Korea. For the 1-year outcome analysis, the data were cut off on 26 October 2020, and 1887 patients were included in the analysis ( Figure 1). Of these, 60.7% were male, and the mean ± SD age was 70.9 ± 9.0 years. The proportions of reference age (<65 years), youngest-old (65-74 years) and middle-old/oldest-old (≥75 years) groups were 21.4%, 40

| One-year outcome by age group
In terms of effectiveness and safety outcomes, no statistically significant trend was observed across the age subgroups ( Figure 2; TA B L E 1 Patient demographics and baseline characteristics in overall population.   Figure S1).

| One-year outcome by dose subgroup
The annual incidence rates of major clinical outcomes are summarized in Table S4. The annual incidence of any stroke among the dose subgroups ranged from 1.05% (recommended 60 mg) to 1.55% (rec-

| Comparison with the data from other regions
Patient demographics and other baseline characteristics are presented in Notably, however, on a drug-specific analysis, non-recommended dosing of edoxaban was not associated with an increase in stroke/ embolism (adjusted HR = 1.43, 95% CI: 0.53, 3.89). 30 Similarly, the analysis of the European participants of the ETNA-AF study did not identify differences in the rates of stroke/SEE and ischemic stroke in patients receiving a non-recommended 30-mg dose of edoxaban and those treated with the recommended 60 mg. 39 Also, the results of the present analysis showed that the non-recommended dosing of edoxaban did not affect much on either the safety or effectiveness of the drug. The annual incidence of any stroke varied from 1.05% in the recommended 60-mg group to 1.55% in the recommended 30mg group, 1.38% in the non-recommended 30-mg group and 1.14% in the non-recommended 60-mg group without a marked difference between the recommended and non-recommended dose groups.
Also, the dose groups did not differ in terms of the safety outcomes and net clinical outcome. Although major clinical outcomes except for any stroke in the non-recommended 30-mg group and all clinical outcomes in the non-recommended 60-mg group seem favorable, the statistical power of this study is inadequate in that sample sizes in the non-recommended dosage groups were small and the event rates were extremely low in real-world clinical practice ( Figure 4).
These findings imply that edoxaban represents a safe and effective treatment option also in patients with AF; furthermore, this obser- was less likely to contribute to a higher incidence of stroke in the Korean registry. It cannot be excluded that the regional differences in the occurrence of stroke, the statistical significance of which was in fact not verified, were instead associated with the disparity of the sample sizes. The overall number of European patients participating in the ETNA-AF study (n = 13133) was substantially higher than the number of Korean participants (n = 1887), and hence, the percentages of stroke in the latter group might be well overestimated.
The present study was not free from potential limitations. First, the outcomes of edoxaban treatment were not verified against a comparator, whether a NOAC or a VKA. Second, the study participants were followed for a relatively short time, and notably, the available follow-up for Korean patients was shorter than for those from Europe. While these limitations should be considered during the interpretation of the results, the strengths of the study related to its real-world character should also be highlighted; namely, the large sample size and the fact that all therapeutic decisions were solely at physicians' discretion.

| CON CLUS IONS
ETNA-AF Korea is the first prospective, noninterventional study to investigate the effectiveness and safety of all edoxaban doses in a Korean AF population. The annual incidence rates of stroke, major bleeding and all-cause mortality were low overall across age and dose subgroups and comparable to those in the Global ETNA population. Our observation confirmed that edoxaban could be used safely and effectively even in the elderly patients and those with ineligible for standard anticoagulation therapy and requiring dose adjustment in routine care in Korea.

FU N D I N G I N FO R M ATI O N
This study was funded by Daiichi Sankyo, Inc.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data underlying this article cannot be shared publicly, as the Global ETNA-AF program is currently ongoing.