Author + information
- Shinya Goto, MD, PhD∗ ( and )
- Shinichi Goto, MD, PhD
- ↵∗Address for correspondence:
Dr. Shinya Goto, Department of Medicine (Cardiology), Tokai University School of Medicine, 143 Shimokasuya, Isehara 259-1193, Japan.
In this issue of the Journal, Lee et al. (1) reported an observational analysis of the large Korean National Health Insurance Service database showing the difference between edoxaban-treated and warfarin-treated patients with nonvalvular atrial fibrillation (AF). The noninferior efficacy of edoxaban for prevention of stroke/systemic embolism and its better safety for serious bleeding/mortality from cardiovascular causes, as compared with warfarin, were shown by the global ENGAGE AF-TIMI 48 (Effective Anticoagulation with Factor Xa Next Generation in Atrial Fibrillation–Thrombolysis In Myocardial Infarction 48) trial (2). However, the ENGAGE AF-TIMI 48 trial only included a small number of East Asian subjects. Lee et al.’s data provide an important opportunity to validate the global trial results in a regional “real-world” setting. This type of regional validation of global trials in East Asia is important due to the perception that East Asian patients differ from other patient populations with regard to the use of anticoagulants. This strong perception—even influencing direct oral anticoagulant (DOAC) development in Japan—also affected the separate analysis in elderly subjects more than 70 years of age in the J-ROCKET-AF (Efficacy and Safety of Rivaroxaban for the Prevention of Stroke in Subjects With Non-Valvular Atrial Fibrillation; NCT00494871) study (3) treated with a reduced dose of rivaroxaban versus warfarin with a target prothrombin time–international normalized ratio of 1.6 to 2.0, which was outside of the ROCKET-AF (An Efficacy and Safety Study of Rivaroxaban With Warfarin for the Prevention of Stroke and Non-Central Nervous System Systemic Embolism in Patients With Non-Valvular Atrial Fibrillation; NCT00403767) trial (4). Regional randomized trials with small sample sizes did not provide strong scientific evidence for changing in the standard of care around the globe. Regional validation of global trial results with observatory data, as conducted by Lee et al. (1), is one way to overcome potential heterogeneity within the worldwide population (5).
The huge effort by Lee et al. (1) to analyze the comparison between edoxaban-treated and warfarin-treated patients should be commended. Their data should reflect “real-world” practice in Korea due to >97% population representation. However, their analysis contains several important limitations with regard to the comparison of edoxaban and warfarin. Lee et al. (1) started with a large population of 648,560 patients with prevalent AF. The vast majority of patients were excluded to achieve the final comparison of 35,765 new users of either warfarin or edoxaban in a primary prevention cohort. Although there are previous publications that had suggested better efficacy and safety of DOACs using the same Korean national database (6), it is of note that edoxaban was the least prescribed DOAC (edoxaban 4,200; dabigatran 14,375; rivaroxaban 22,968; apixaban 12,657). Physicians’ choices were not based on commonly used risk stratification tools such as the CHA2DS2-VASc (congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, prior stroke, transient ischemic attack, or thromboembolism, vascular disease, age 65–74 years, sex category [female]) or the HAS-BLED (hypertension, abnormal renal and liver function, stroke, bleeding, labile INR, elderly, drugs or alcohol) scores, and, therefore, were hard to adjust for, even by propensity score matching. The lower ischemic/bleeding/mortality events in the edoxaban group, as compared with the warfarin group, may just indicate the presence of strong drug selection bias. The short and heterogenous follow-up periods of 0.3 years for edoxaban and 0.9 years for warfarin also should be considered as limitations of this analysis. The slight statistical difference between edoxaban and warfarin (hazard ratio: 0.693; 95% confidence interval: 0.487 to 0.959; p = 0.033) may be refuted by a longer term of follow-up, even with the same dataset.
Lee et al. (1) clearly explain their reasons for excluding patients with a previous history of stroke, intracranial hemorrhage, and gastrointestinal bleeding from their analysis. However, the exclusion of 22,931 patients with a prior stroke, 2,120 with intracranial hemorrhage, and 2,745 with gastrointestinal bleeding, is a limitation for the validation of the ENGAGE AF-TIMI 48 trial by the Lee et al. (1) data. Indeed, >28% of the ENGAGE AF-TIMI 48 participants had a previous history of stroke. The numerically higher risk of all-cause death (5.59 per 100 patient-years [PY] for edoxaban and 6.63 PY for warfarin) than ischemic stroke (3.22 PY for edoxaban and 3.89 PY for warfarin) shown by Lee et al.’s analysis suggest higher event rates in “real-world” complicated patients, even with anticoagulation and selection of a primary prevention cohort.
In the majority of clinical trials, including the ENGAGE TIMI-AF 48 trial, the efficacy comparisons were conducted as intention-to-treat analyses. However, the comparison of events in Lee et al.’s study was conducted using an on-treatment basis. Without randomized assignment of warfarin/edoxaban, on-treatment analysis may be fine, but, the difference between the Lee et al. (1) method and that of the ENGAGE-AT TIMI 48 trial should be noted. Further randomized trials are necessary to clarify whether or not certain doses of edoxaban might provide better efficacy and safety over warfarin in East Asian patients.
The Lee et al. (1) data comparing the higher and lower doses tested in the ENGAGE TIMI 48 trial are interesting for hypothesis generation. As expected, a higher proportion of patients were prescribed the lower dose (n = 2,371) rather than the higher one (n = 1,835). More than one-half of those prescribed the 30-mg dose were more than 75 years of age, whereas the majority of those prescribed the 60-mg dose were between 65 and 74 years of age. It is interesting to note that CHA2DS2-VASc scores in patients prescribed the 30-mg dose were higher than in patients prescribed the 60-mg dose, whereas stroke prevention effects were proven only for the 60-mg dose in the ENGAGE TIMI-48 trial (2). Lee et al.’s data may support the need for a clinical trial to test the efficacy and safety of the lower dose of edoxaban in elderly East Asian patients. The results of an ongoing phase III trial comparing 15 mg daily edoxaban (DU-176b) and warfarin in elderly patients (80 years of age or older) are awaited (Study of DU-176b Aged 80 Years or Older; NCT02801669).
Edoxaban was developed as DU-176b by the Japanese pharmaceutical company Daiichi-Sankyo (7) and thus is of East Asian origin (Figure 1). Initial development of DU-176b follow the historical tradition for development of enzyme inhibitors represented by argatroban (8), tranexamic acid, and so on (9). The global clinical trial led by the TIMI study group (2) further developed edoxaban by showing its efficacy and safety in stroke prevention in patients with AF. Here, Lee et al. (1) provide East Asian clinical data to validate the results of the ENGAGE-AF TIMI 48 global trial. Moreover, support was provided for the attractive hypothesis that edoxaban has better efficacy and safety compared with warfarin at certain doses that are suitable for East Asians. Further hypothesis-testing clinical trials will provide more data on the value of edoxaban.
The long journey of edoxaban started in East Asia. After its efficacy and safety were established for stroke patients around the globe, edoxaban returned to East Asia to produce “real-world” data. The long journey of edoxaban is just beginning!
↵∗ Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.
Dr. Shinya Goto has received a Grant-in-Aid for Scientific Research in Japan (17K19669). Dr. Shinya Goto has received a contracted research grant from Ono and independent research grants from Sanofi/Bristol-Myers Squibb/Pfizer/Astellas/Daiichi-Sankyo; has served as a member of the data safety committee for the ENGAGE-AF TIMI 48 trial; and is an associate editor for Circulation. Dr. Shinichi Goto has reported that he has no relationships relevant to the contents of this paper to disclose.
- 2018 American College of Cardiology Foundation
- Lee S.-R.,
- Choi E.-K.,
- Han K.-D.,
- Jung J.-H.,
- Oh S.,
- Lip G.Y.H.
- Yabushita H.,
- Goto S.
- Cha M.J.,
- Choi E.K.,
- Han K.D.,
- et al.
- Watts G.