In the Effective Anticoagulation with Factor Xa Next Generation in Atrial Fibrillation (ENGAGE-AF) trial , patients with atrial fibrillation and a CHADS2 score of > 2 were randomized in a double-blind, double-dummy fashion to either high-dose edoxaban (60 mg daily), low-dose edoxaban (30 mg daily), or warfarin titrated to an INR of 2-3. Patients with an estimated creatinine clearance (CrCl) < 30 mL/min and those taking dual antiplatelet therapy were excluded. The dose of edoxaban was halved if patients were < 60 kg, had a CrCl of 30-50 mL/min, or if a strong p-glycoprotein inhibitor (i.e., dronedarone, quinidine, verapamil) was added. Notable baseline characteristics include median age of 72, history of stroke in 28.3% and heart failure in 57.4%; over three-fourths of patients had a CHADS2 score < 3. The median time in therapeutic range (TTR) for those on warfarin was 68.4%.
In terms of the primary endpoint of stroke or systemic embolism, both doses of edoxaban were non-inferior to warfarin (1.18% with high-dose edoxaban and 1.61% with low-dose edoxaban vs. 1.50% with warfarin [p < 0.001 and p = 0.005, respectively]). While a trend favoring edoxaban was observed in the superiority analysis, it did not reach statistical significance. Rates of major bleeding were lower with both doses of edoxaban (2.75% and 1.61% with high and low-dose edoxaban vs. 3.34% with warfarin, both p < 0.001), as were rates of intracranial hemorrhage. Gastrointestinal bleeding (GI) was higher with edoxaban. Improvements in several key secondary endpoints were also observed with edoxaban, including death from cardiovascular causes. Patients receiving the lower dose of edoxaban had a higher rate of ischemic strokes, while rates were similar among those on warfarin and high-dose edoxaban.
With a few exceptions, the findings of ENGAGE-AF are almost identical to those observed in the comparison of rivaroxaban, another factor Xa inhibitor, and warfarin in the ROCKET-AF trial . Most experts argue that is not possible to compare the new oral anticoagulants with each other because none were studied head-to-head. While this is true from the standpoint of academic purity, it does little to guide the clinician when faced with the challenge of selecting an agent in an individual patient. For this latter case, a discussion of the similarities and differences are important.
Similarities between ENGAGE-AF and ROCKET-AF include:
- Trial design: both were randomized, double-blind, double-dummy trials
- Inclusions/exclusions: similar thresholds for renal function, high-risk exclusions (e.g., patients on concomitant dual antiplatelet therapy were excluded)
- Drug dosing: once daily dosing vs. warfarin titrated to an INR of 2-3
- Efficacy: both shown to be non-inferior to warfarin
- Safety: both safer than warfarin in the severest of safety endpoints (fatal bleeding, intracranial hemorrhage), but higher rates of GI bleeding
A few key differences:
- Study population: on average, the patients enrolled in ENGAGE-AF were healthier than those in ROCKET-AF, as demonstrated by lower median CHADS2 scores and fewer patients with a history of stroke or heart failure
- Warfarin management: warfarin was more optimally managed in ENGAGE-AF based on a TTR of 68.4% compared to 55% in ROCKET-AF; that being said, INR control is closely related to overall health status, so the fact that the INR was less problematic in the healthier population of ENGAGE-AF is not altogether surprising
- Safety endpoints: although both drugs reduced the incidence of severe bleeding, rivaroxaban was similar to warfarin in the primary safety endpoint of major bleeding, whereas edoxaban was safer in ENGAGE-AF
- Transition at study termination: the investigators of ENGAGE-AF should be applauded for the lessons learned from ROCKET-AF, where rebound thrombotic events were observed among patients being transitioned from rivaroxaban to open-label warfarin at conclusion of the trial. At the end of ENGAGE-AF, a carefully monitored transition from edoxaban to warfarin was performed, resulting in no differences in rebound thrombotic events
- Differences in secondary endpoints: edoxaban showed improvements in some secondary endpoints (e.g., cardiovascular death, other composites)
So now that edoxaban will represent the fourth alternative to warfarin, what are clinicians to do? To be honest, it is my personal opinion that edoxaban offers few if any clinical advantages to rivaroxaban. Despite being studied in a healthier population (i.e., where differences in drug metabolism and clearance are less likely to complicate management), edoxaban was still only non-inferior to warfarin. While it was safer than warfarin in terms of major bleeding, this could again be attributed to the healthier nature of the patient population. Both rivaroxaban and edoxaban reduced the incidence of the severest safety endpoints – fatal bleeding and intracranial hemorrhage.
Compared to the other new oral anticoagulants apixaban and dabigatran, my feelings on edoxaban are similar to rivaroxaban, which I wrote about in this entry in November 2011 and again in October 2012. I still tend to favor apixaban as my first-line alternative to warfarin based on it having the most comparative advantages, although I would still consider dabigatran in younger patients with normal renal function. The whole notion that the once daily dosing made possible by rivaroxaban, and now edoxaban, is more ideal for less compliant patients is still a dangerous proposition (see my note at the end of the selection tool posted in October 2012). For patients likely to miss doses, a once daily drug that only lasts half a day actually offers less protection from stroke and systemic embolism than one taken twice daily.
Edoxaban is non-inferior to warfarin for preventing stroke and systemic embolism in patients with atrial fibrillation while reducing the risk of major bleeding.
- Giugliano RP, Ruff CT, Braunwald E, Murphy SA, Wiviott SD, Halperin JL, et al. Edoxaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2013 Nov 28;369(22):2093–104.
- Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med. 2011 Sep 8;365(10):883–91.