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

Abstract

Importance: In older patients with atrial fibrillation who take anticoagulants for stroke prevention, bleeding is increased compared with younger patients, thus, clinicians frequently prescribe lower than recommended doses in older patients despite limited randomized data.

Objective: To evaluate ischemic and bleeding outcomes in patients 80 years and older with atrial fibrillation receiving edoxaban, 60 mg vs 30 mg, and edoxaban, 30 mg vs warfarin.

Design, setting, and participants: The ENGAGE AF-TIMI 48 trial (Effective Anticoagulation With Factor Xa Next Generation in Atrial Fibrillation-Thrombolysis in Myocardial Infarction 48) was a parallel-design, double-blind, global clinical trial that randomized patients with atrial fibrillation to either one of 2 edoxaban dosing regimens or warfarin. This secondary analysis focused on patients 80 years or older without dose-reduction criteria receiving edoxaban, 60 mg vs 30 mg, as well as patients with or without dose-reduction criteria receiving edoxaban, 30 mg, vs warfarin. Study data were analyzed between October 2022 and December 2023.

Interventions: Oral edoxaban, 30 mg once daily; edoxaban, 60 mg once daily; or warfarin.

Main outcomes and measures: Primary net clinical outcome of death, stroke or systemic embolism, and major bleeding and each individual component.

Results: The current analysis included 2966 patients 80 years and older (mean [SD] age, 83 [2.7] years; 1671 male [56%]). Among 1138 patients 80 years and older without dose-reduction criteria, those receiving edoxaban, 60 mg vs 30 mg, had more major bleeding events (hazard ratio [HR], 1.57; 95% CI, 1.04-2.38; P = .03), particularly gastrointestinal hemorrhage (HR, 2.24; 95% CI, 1.29-3.90; P = .004), with no significant difference in efficacy end points. Findings were supported by analyses of endogenous factor Xa inhibition, a marker of anticoagulant effect, which was comparable between younger patients receiving edoxaban, 60 mg, and older patients receiving edoxaban, 30 mg. In 2406 patients 80 years and older with or without dose-reduction criteria, patients receiving edoxaban, 30 mg, vs warfarin had lower rates of the primary net clinical outcome (HR, 0.78; 95% CI, 0.68-0.91; P = .001), major bleeding (HR, 0.59; 95% CI, 0.45-0.77; P < .001), and death (HR, 0.83; 95% CI, 0.70-1.00; P = .046), whereas rates of stroke or systemic embolism were comparable.

Conclusions and relevance: In this post hoc analysis of the ENGAGE AF-TIMI 48 randomized clinical trial, in patients 80 years and older with atrial fibrillation, major bleeding events were lower in patients randomized to receive edoxaban, 30 mg per day, compared with either edoxaban, 60 mg per day (in patients without dose-reduction criteria), or warfarin (irrespective of dose-reduction status), without an offsetting increase in ischemic events. These data support the concept that lower-dose anticoagulants, such as edoxaban, 30 mg, may be considered in older patients with atrial fibrillation even in the absence of dose-reduction criteria.

Trial registration: ClinicalTrials.gov Identifier: NCT00781391.

Figure 1.. Patient Flowchart

A total of 2966 patients 80 years and older were randomized to edoxaban, 30 mg once daily; edoxaban, 60 mg once daily; or warfarin and included in this analysis. Comparisons between edoxaban, 30 mg (n = 1201), vs warfarin (n = 1205) included all patients randomized to the respective dosing arm regardless of the presence of dose-reduction criteria. Comparisons between edoxaban, 30 mg (n = 578), vs edoxaban, 60 mg (n = 560), vs warfarin (n = 562) included patients without dose-reduction criteria. The edoxaban dose was reduced by 50% in patients meeting 1 or more of the following dose reduction criteria: creatinine clearance of 50 mL/min or less, body weight of 60 kg or less, or use of concomitant strong P-glycoprotein (permeability glycoprotein) inhibitors. ENGAGE AF-TIMI 48 indicates Effective Anticoagulation With Factor Xa Next Generation in Atrial Fibrillation–Thrombolysis in Myocardial Infarction 48; HDER, higher-dose edoxaban regimen (60/30 mg); LDER, lower-dose edoxaban regimen (30/15 mg).

Figure 2.. Efficacy and Safety Outcomes of Edoxaban, 60 mg vs 30 mg, in Patients 80 Years and Older With No Dose-Reduction Criteria

Cumulative event rates are shown for stroke or systemic embolism (A), major bleeding (B), the primary net clinical outcome of stroke, systemic embolic event, major bleeding, or death from any cause (C), death from any cause (D), ischemic stroke (E), and intracranial hemorrhage (F). HR indicates hazard ratio.

Figure 3.. Pharmacodynamic and Clinical Outcomes in Patients with No Dose-Reduction Criteria

A, Absolute difference in ischemic stroke and major gastrointestinal bleeding at 3 years for patients 80 years and older without dose-reduction criteria randomized to edoxaban, 60 mg (n = 560) vs 30 mg (n = 578). Event rates were calculated using Kaplan-Meier estimates. Patients 80 years and older randomized to edoxaban, 60 mg, had higher rates of major gastrointestinal bleeding (hazard ratio [HR], 2.24; 95% CI, 1.29-3.90; P = .004) with no significant differences in ischemic stroke (HR, 0.84; 95% CI, 0.48-1.49; P = .56) compared with edoxaban, 30 mg. B, Endogenous factor Xa (FXa) activity at trough by age. Lower values indicate greater anticoagulant effect. Line graph indicates the predicted endogenous FXa activity at trough by age in patients without dose-reduction criteria randomized to edoxaban, 60 mg (n = 5251) vs 30 mg (n = 5249). The predicted FXa activity at trough declined with age and was 8.0% to 11.4% lower for edoxaban, 60 mg (vs edoxaban, 30 mg) across the age range. In patients without dose-reduction criteria, the predicted FXa activity of 71% at trough in a 67-year-old patient taking edoxaban, 60 mg, was similar to that of an 82-year-old patient taking 30 mg. GI indicates gastrointestinal.