Low-Dose Edoxaban Safe, Effective in Older Patients With AFib, High Bleeding Risk

Edoxaban, an oral anticoagulant and factor Xa inhibitor, was shown to be safe and effective (vs placebo) among patients 80 years of older—regardless of frailty status—at reducing their risk of stroke or systemic embolism, according to a new subanalysis of data from the Edoxaban Low-Dose for Elder Care Atrial Fibrillation Patients (ELDERCARE-AF) trial.

These multicenter randomized double-blind placebo-controlled phase 3 study findings from Japan were published online today in JAMA Network Open.

Patients living with atrial fibrillation (AFib) ineligible for standard-dose oral anticoagulant treatment were randomized 1:1 to treatment with 15-mg daily edoxaban or placebo—if placebo, anticoagulants were not provided during the study—in the study that ran from August 5, 2016, to November 5, 2019. Final patient follow-up was December 27, 2019. Five frailty parameters, each worth 1 point, were used to assess patient status: weight loss, grip strength, walking speed, exhaustion, and activity level. The higher the score, the higher the degree of frailty (0 = robust, 1 or 2 = prefrail, 3 or higher = frail). The patients in each cohort were also stratified by CHADS₂ score, with higher scores again indicating greater risk. The overall patient cohort was 984, and of that group, data from 944 were included in this analysis. Of this group, 57.4% were not considered frail, 57.3% were women, and the mean (SD) age was 86.6 years.

“On the basis of the results from the ELDERCARE-AF trial, 15-mg edoxaban has been approved for use in elderly patients with AF at high risk of bleeding in Japan,” the authors wrote, “although questions remain regarding the outcomes and safety of edoxaban in frail patients.”

Overall, frailty status was established in 96.3% (n = 474) of the edoxaban group and 95.5% (n = 470) of the placebo group. More patients in the placebo group were designated as frail, at 46.2% vs 39.0%. The most common frailty status was prefrail status in 51.0%.

Mean (SD) creatinine clearance rate was lowest in the placebo frail cohort and highest in the placebo nonfrail cohort, at 32.5 (12.4) and 38.9 (15.2) mL/min compared with 34.4 (15.5) and 37.7 (13.5) mL/min in the edoxaban cohorts, respectively. Total incidence of each of vascular disease, dementia, and fall history for the prior year was higher in the placebo group vs the edoxaban group.

The mean (SD) CHADS₂ scores were equivalent across the patients, at 3.0 (1.1) and 3.0 (1.0) in the edoxaban frail and nonfrail groups and 2.9 (1.1) and 3.2 (1.2) in the placebo frail and nonfrail cohorts. Total numbers of patients with scores of 2 and scores of 3 and above were close to equal in the 2 patient cohorts. The most common frailty assessments were low grip strength in the edoxaban nonfrail and both placebo groups, followed by slow walking speed in the edoxaban frail group.

Rates of stroke and systemic embolism (SSE) for the placebo group were 7.1% (1.6%) per patient-year in the frail cohort and 6.1% (1.3%) in the nonfrail cohort, with the authors determining this risk did not differ between the groups (adjusted HR [aHR], 1.04; 95% CI, 0.56-1.94). In the edoxaban group, these totals were 2.5% (1.0%) and 1.54% (0.6%), respectively. Overall, the highest risk of SSE was from lower grip strength (unadjusted HR, 2.91; 95% CI, 1.04-8.16). Further, major bleeding risk was lower in both placebo groups vs both edoxaban groups, but neither frailty status nor its parameters influenced these outcomes, according to the investigators.

For the net composite outcome (stroke, systemic embolism, major bleeding) and all-cause death, 61% (aHR, 1.61; 95% CI, 1.06-2.45) and 144% (aHR, 2.44; 95% CI, 1.42-4.18) higher risks, respectively, were seen in the participants who were considered frail. Again, lower grip strength was associated with the highest risks of both outcomes, from among the 5 frailty assessments, at 462% (aHR, 5.62; 95% CI, 2.28-13.83) and 1526% (aHR, 16.26; 95% CI, 2.28-121.00) higher risks, respectively.

The risk of other bleeding events—major or clinically relevant nonmajor—was 100% greater (aHR, 2.00; 95% CI, 1.13-3.53) in the frail patients vs the nonfrail patients, and exhaustion had the greatest association with higher risk, at a 97% higher chance of this outcome (aHR, 1.97; 95% CI, 1.12-3.47). Frailty status and its assessment parameters had no interaction with this outcome.

Because only Japanese patients were included in this study, using other frailty evaluation criteria may have led to different results, and event rates in the subgroups were low, the authors note their findings may not be generalizable to other patient groups or ethnic populations.

However, the trial’s design and that older patients with AFib and a high bleeding risk do not have a standard-of-care therapy mean that “the present subanalysis provided important information showing that when elderly, high-risk, frail patients remain untreated with any OACs, they have a substantially higher risk of SSE,” the authors concluded. “Edoxaban may be a suitable treatment option for these patients.”

Reference

Akashi S, Oguri M, Ikeno E, et al. Outcomes and safety of very-low-dose edoxaban in frail patients with atrial fibrillation in the ELDERCARE-AF randomized clinical trial. JAMA Netw Open. Published online August 23, 2022. doi:10.1001/jamanetworkopen.2022.28500