By Michael H. Crawford, MD, Editor
A subgroup analysis of the ENGAGE-AF TIMI 48 study of edoxaban 60 mg/day vs. 30 mg/day compared to warfarin in patients with atrial fibrillation of the group 80 years of age or older has shown that 30 mg/day results in less major bleeding without a concomitant increase in stroke risk compared to 60 mg/day or warfarin.
Zimerman A, Braunwald G, Steffel, J, et al. Dose reduction of edoxaban in patients 80 years and older with atrial fibrillation: Post hoc analysis of the ENGAGE AF-TIMI 48 randomized clinical trial. JAMA Cardiol; Jul 10. doi:10.1001/jamacardio.2024.1793. [Online ahead of print].
Effective Anticoagulation With Factor Xa Next Generation in Atrial Fibrillation-Thrombolysis in Myocardial Infarction 48 (ENGAGE AF-TIMI 48) was a parallel-design, double-blind, randomized trial of either 60 mg/day or 30 mg/day of edoxaban (or 15 mg/day if dosage reduction criteria were met) vs. warfarin in patients with atrial fibrillation (AF). The high-dose edoxaban (HDE) was safe and effective, but the low-dose edoxaban (LDE) resulted in a stroke risk that was 41% higher than that with warfarin, despite a significant 33% lower risk of gastrointestinal (GI) bleeding. For this reason, the LDE was not approved for clinical use. However, clinical interest in lower-dose oral anticoagulants persists because they could be useful for patients at high bleeding risk.
Subgroup analyses of ENGAGE suggested that patients ≥ 80 years of age taking LDE were less likely to experience the net clinical outcome of stroke, major bleeding, or death. Hence, this post hoc analysis of ENGAGE AF-TIMI 48 studied the outcomes of the patients ≥ 80 years of age randomized to edoxaban 60 mg/day vs. 30 mg/day who did not meet reduced-dose criteria and edoxaban 30 mg/day irrespective of dose-reduction criteria vs. warfarin.
ENGAGE was conducted in 46 countries, between 2008-2010 in more than 21,000 patients, of whom 2,966 were ≥ 80 years of age. Reduced-dose criteria were creatinine clearance ≤ 50 mL/min, weight ≤ 60 kg, or if the patient was taking another drug that strongly interfered with edoxaban. Inclusion criteria included a CHADS2 score ≥ 2 and AF within the prior year. Exclusion criteria included creatinine clearance < 30 mL/min, mitral stenosis, prosthetic heart valve, high risk for bleeding, or need to remain on dual anti-platelet therapy. The primary endpoint was stroke or systemic embolism. The safety endpoint was major bleeding. Secondary endpoints were ischemic stroke, intracranial hemorrhage, all-cause death, and the net clinical outcome of stroke, major bleeding, or death. There was an assessment of plasma factor Xa activity. The mean age of the study group was 83 years; 87% were white and 56% were men.
In the 1,138 patients without a dose-reduction criterion, those who received 60 mg/day vs. 30 mg/day of edoxaban had more major bleeding (hazard ratio [HR], 1.57; 95% confidence interval [CI], 1.04-2.38; P = 0.03) and GI hemorrhage (HR, 2.24; 95% CI, 1.29-3.90; P = 0.004) without any difference in stroke incidence. Factor Xa inhibition was similar in older patients taking 30 mg/day of edoxaban and younger patients taking 60 mg/day from the original trial. Also, factor Xa inhibition in older patients taking 60 mg/day of edoxaban was 35% higher than in younger patients taking 60 mg/day in the original trial.
In the 2,406 patients who were randomized to 30 mg/day regardless of dose-reduction criteria vs. warfarin, net clinical outcomes were reduced significantly (HR, 0.78; 95% CI, 0.68-0.91; P = 0.001). In addition, major bleeding also was less frequent (HR, 0.59; 95% CI, 0.45-0.77; P < .0001) as was all-cause death (HR, 0.83; 95% CI, 0.70-1.00; P = 0.046). Rates of ischemic stroke were not different, but intracranial hemorrhage was lower on 30 mg/day vs. warfarin (HR, 0.49; 95% CI, 0.28-0.87; P = 0.02).
The authors concluded that major bleeding rates were lower with edoxaban 30 mg/day compared to either edoxaban 60 mg/day or warfarin without an increase in ischemic stroke in AF patients age 80 years of age or older.
COMMENTARY
The concept of using lower doses of direct-acting oral anticoagulants (DOACs) in older patients who are believed to generally be at higher risk for bleeding has been controversial because observational studies have shown increased stroke and death rates with this practice. Hence, this prespecified subgroup analysis of ENGAGE-AF TIMI 48 is of interest.
The study has shown that in older AF patients who do not meet dosage reduction criteria, 30 mg/day vs. 60 mg/day of edoxaban significantly reduced the rate of major bleeding with no increase in the risk of stroke. This finding was supported by factor Xa plasma activity, which demonstrated that in older patients the anticoagulant effect of 30 mg/day of edoxaban was similar to 60 mg/day in younger patients from the original trial. Similar results were seen in the older patients on 30 mg/day regardless of dose-reduction criteria, suggesting that all patients 80 years of age or older should be taking the lower dose regardless of lower-dose criteria.
The authors pointed out that this is especially relevant to transcutaneous aortic valve replacement (TAVR) patients with AF, since they often are 80 years or older in age and have risk factors for increased GI bleeding, such as acquired von Willebrand disease and Heyde syndrome. Also, many such patients are placed on antiplatelet therapy after TAVR.
The major limitation of this subgroup analysis of ENGAGE is that it is post hoc and not adjusted for multiple comparisons. Also, trial participants often are more compliant with medications. In fact, in this study, the warfarin-treated patients were maintained in the therapeutic range a median of 70% of the time. In addition, age 80 years is an arbitrary cut point, and some estimate of biologic age or other criteria may be more appropriate. Finally, in my area, edoxaban rarely is used as compared to apixaban or rivaroxaban, so we need data on all the DOACs if we are going to apply these dosage concepts for older patients.
Those who disagree with this dose-lowering concept in older patients without regard to dose-lowering criteria would point out that in ENGAGE, the lower-dose-for-all concept resulted in nine fewer major bleeds at the cost of one stroke or death. Since most major bleeds are GI and, as such, are rather easy to manage successfully, the cost is too high. Thus, we need randomized trials of this concept for the DOACs to confidently apply the lower dose for all older patients hypothesis.
