Cardiology · Medical Literature
Total Clinical Event Burden With Edoxaban Monotherapy Versus Dual Antithrombotic Therapy in Atrial Fibrillation and Stable Coronary Artery Disease: Insights From the EPIC-CAD Trial.
Abstract
Oral anticoagulation is essential for stroke prevention in atrial fibrillation, whereas antiplatelet therapy remains standard for coronary artery disease (CAD), raising concerns about bleeding when both are combined. The effect of these regimens on recurrent events, reflecting cumulative disease burden, has not been fully evaluated. This post hoc analysis of the EPIC-CAD (Edoxaban Versus Edoxaban With Antiplatelet Agent in Patients With Atrial Fibrillation and Chronic Stable Coronary Artery Disease) trial included 1040 patients with atrial fibrillation and stable CAD randomized to edoxaban monotherapy or edoxaban plus a single antiplatelet agent. The primary end point was the total number of net adverse clinical events at 12 months. Over a median follow-up of 12.1 months, 155 total events occurred (123 first, 32 subsequent). Event rates were lower with edoxaban monotherapy than with dual therapy (9.4 versus 19.8 per 100 person-years; incidence rate ratio, 0.47 [95% CI, 0.34-0.67]). Monotherapy reduced first events (incidence rate ratio, 0.46 [95% CI, 0.32-0.68]) and showed a lower cumulative hazard of total events (hazard ratio, 0.48 [95% CI, 0.32-0.71]; P<0.001). Bleeding was the predominant event type in both groups but occurred more often with dual therapy (85.7% versus 55.3%; P<0.001). Rates of ischemic events and death were similar between groups. Edoxaban monotherapy significantly reduced the overall burden of clinical events compared with dual antithrombotic therapy in patients with atrial fibrillation and stable CAD. Bleeding complications were more frequent than ischemic events, emphasizing the need to minimize bleeding risk in this population.
Clinical relevance
In patients with atrial fibrillation and stable coronary artery disease, does edoxaban monotherapy reduce the total (first and recurrent) net adverse clinical event burden compared with dual antithrombotic therapy (edoxaban plus a single antiplatelet agent) over 12 months? __SNAPSHOT__ {"studyDesign":"Post hoc analysis of a multicenter, open-label, adjudicator-masked randomized controlled trial (EPIC-CAD trial), using a recurrent-events analytical framework to assess total clinical event burden","setting":"18 high-volume cardiac centers in South Korea","sampleSize":"n=1,040 (edoxaban monotherapy: 524; dual antithrombotic therapy: 516)","population":"Adults with high-risk atrial fibrillation (CHA₂DS₂-VASc score ≥2) and stable CAD (defined as revascularization for chronic coronary syndrome ≥6 months prior, revascularization for ACS ≥12 months prior, or anatomically confirmed CAD managed medically); mean age 72.1 years; ~23% women; mean CHA₂DS₂-VASc score 4.3","intervention":"Edoxaban monotherapy: 60 mg once daily (or 30 mg once daily per dose-reduction criteria)","comparator":"Dual antithrombotic therapy: edoxaban at equivalent dose plus a single antiplatelet agent (aspirin or a P2Y12 inhibitor, per physician choice)","primaryOutcome":"Total number of net adverse clinical events (NACE) — comprising all-cause death, myocardial infarction, stroke, systemic embolism, unplanned urgent revascularization, major bleeding, and clinically relevant nonmajor bleeding — at 12 months, capturing both first and subsequent (recurrent) events","followUp":"Median 12.1 months","fundingSource":"CardioVascular Research Foundation (South Korea), Daiichi Sankyo, and Daewoong Pharmaceutical (mixed: non-profit/industry)"}
Key findings
Over a median follow-up of 12.1 months, 155 total NACE events occurred across both arms: 123 first events and 32 subsequent events, demonstrating that conventional first-event analyses would have underestimated cumulative disease burden by approximately 21% (calculated). Total event rates were substantially lower with edoxaban monotherapy versus dual antithrombotic therapy (9.4 vs. 19.8 per 100 person-years; incidence rate ratio [IRR] 0.47, 95% CI 0.34–0.67), representing a 53% relative rate reduction. Analysis restricted to first events yielded a nearly identical treatment effect (IRR 0.46, 95% CI 0.32–0.68), and the cumulative hazard of total events likewise favored monotherapy (HR 0.48, 95% CI 0.32–0.71; P<0.001). These findings are consistent with the primary EPIC-CAD trial result, in which the time-to-first NACE event occurred in 6.8% of the monotherapy group vs. 16.2% of the dual therapy group (HR 0.44, 95% CI 0.30–0.65; P<0.001), yielding an absolute risk reduction of 9.4 percentage points and an NNT of approximately 10.6 (as reported). The primary EPIC-CAD trial separately demonstrated that bleeding drove the composite benefit: major or clinically relevant nonmajor bleeding occurred in 4.7% (monotherapy) vs. 14.2% (dual therapy) (HR 0.34, 95% CI 0.22–0.53), while major ischemic events were similar (1.6% vs. 1.8%; HR 1.23, 95% CI 0.48–3.10), and all-cause mortality was not significantly different (0.6% vs. 0.7%; HR 1.29, 95% CI 0.29–5.76). The post hoc analysis (PMID 42024702) extends these findings by demonstrating that the monotherapy advantage persists and is not materially amplified when recurrent events are counted — bleeding remained the predominant driver of total event burden reduction. Specific event-type breakdown of recurrent events and NNH values were not reported in the abstract.
Practice implications
This post hoc recurrent-events analysis confirms and extends the primary EPIC-CAD trial finding: in patients with high-risk AF and stable CAD, edoxaban monotherapy significantly reduces total net adverse clinical event burden compared with edoxaban plus a single antiplatelet agent. The treatment benefit was consistent whether analyzed as time-to-first event or total recurrent events, indicating robustness of the overall signal. Critically, however, the entire composite advantage was attributable to bleeding reduction — ischemic event rates were statistically similar between arms and the trial was not powered to detect a difference in ischemic outcomes alone. Clinicians should therefore interpret this as strong evidence for bleeding safety of monotherapy, not as evidence that antiplatelet therapy is unnecessary for ischemic protection in this population — rather, the data suggest that any incremental ischemic benefit of adding antiplatelet therapy is outweighed by the increased bleeding risk in patients with stable (not acute) CAD. The population studied was exclusively Asian (South Korean), elderly (mean age 72), with a high mean CHA₂DS₂-VASc score of 4.3 and relatively low absolute ischemic event rates, limiting generalizability to non-Asian populations, younger patients, those with more recent ACS or complex PCI, or those with lower baseline stroke risk. This is a post hoc analysis, which introduces the risk of unplanned multiple comparisons and should be considered hypothesis-confirming rather than hypothesis-generating. Only the abstract was available for review; the full statistical methodology for recurrent-event modeling (e.g., Lin-Wei-Yang-Ying, negative binomial, or joint frailty model) could not be independently verified. __LIMITATIONS__ Post hoc analysis — not a pre-specified primary analysis; recurrent-event framework was applied after the trial completion, raising the possibility of analytical flexibility bias.,Open-label design — patients and treating physicians were not blinded, which may have influenced reporting of clinically relevant nonmajor bleeding (a subjective endpoint) and clinical management decisions.,Adjudicator-masked outcomes mitigate but do not eliminate ascertainment bias.,Trial conducted exclusively at 18 sites in South Korea; all participants were Asian, limiting generalizability to non-Asian populations who may have different bleeding and ischemic risk profiles.,Not powered to detect differences in ischemic outcomes (MI, stroke, death) individually; the similarity in ischemic rates must be interpreted as non-inferiority by observation, not formal statistical non-inferiority testing.,The composite NACE endpoint is heavily weighted toward bleeding events given their higher incidence, which structurally favors any strategy that reduces bleeding, potentially masking a clinically important ischemic risk.,Short follow-up of 12 months; longer-term effects of withdrawing antiplatelet therapy in stable CAD patients remain uncertain.,Only 32 recurrent events were observed, limiting statistical precision of the recurrent-events analysis.,Antiplatelet agent choice (aspirin vs. clopidogrel) was at physician discretion, introducing heterogeneity in the dual therapy comparator.,Mixed industry and non-profit funding (Daiichi Sankyo manufactures edoxaban) carries potential for sponsorship bias.,Exclusion of patients with recent ACS (<12 months) or recent PCI (<6 months) restricts applicability to the stable, long-term post-revascularization setting only. __BOTTOM_LINE_CATEGORY__ confirmatory __BOTTOM_LINE__ In patients with high-risk AF and stable CAD, edoxaban monotherapy reduces the total burden of net adverse clinical events — including both first and recurrent events — by approximately 53% compared with dual antithrombotic therapy, driven entirely by fewer bleeding events with no detectable increase in ischemic risk. These findings, consistent across first-event and recurrent-event analytical frameworks, support edoxaban monotherapy as the preferred long-term antithrombotic strategy in this population, aligning with the primary EPIC-CAD trial results. Clinicians should note the exclusively Asian study population and short 12-month follow-up when applying these findings to broader practice, and should exercise caution in patients with recent ACS or complex PCI who were not represented in this trial.