EPIDAURUS Concludes Potent P2Y12 Inhibitors Could Raise Bleeding in AFib, Myocardial Infarction

The EPIDAURUS trial was stopped early after potent P2Y12 inhibitors combined with a direct oral anticoagulant (DOAC) increased bleeding without reducing ischemic events in patients with atrial fibrillation (AF) and a recent myocardial infarction (MI), according to results presented at the European Society of Cardiology (ESC) Congress 2026 and published simultaneously in Nature Medicine.1

Patients with AF typically receive DOACs to prevent stroke, while patients with a recent MI typically receive dual antiplatelet therapy, often aspirin plus a potent P2Y12 inhibitor, to prevent recurrent ischemic events. Little evidence exists on combining potent P2Y12 inhibitors, prasugrel or ticagrelor, with a DOAC in patients with both conditions, which EPIDAURUS examined.

“In the first trial of its kind, we have answered an important clinical question,” said Konstantinos Rizas, MD, principal investigator and head of the Heart Failure Outpatient Clinic at Ludwig Maximilians University, Munich, Germany. “These findings do not support the routine use of potent P2Y12 inhibitors in combination with direct oral anticoagulants in patients with AF and MI.”

EPIDAURUS trial design and early termination

EPIDAURUS was a prospective, investigator-initiated, open-label, randomized trial conducted at 37 sites in Germany and Austria, enrolling patients with an indication for oral anticoagulation due to AF who had undergone successful PCI within 5 days for a biomarker-positive MI, with or without ST-segment elevation1. Patients were randomized 1:1 to a DOAC plus prasugrel or ticagrelor for 4 weeks followed by a DOAC plus clopidogrel (experimental group), or a DOAC plus clopidogrel with in-hospital aspirin (control group). Edoxaban, apixaban, and rivaroxaban were the permitted DOACs. The trial planned to enroll 1,474 patients but was stopped after 602 patients (302 experimental, 300 control) following a recommendation from the independent Data and Safety Monitoring Board (DSMB), which cited a low likelihood of demonstrating superiority alongside concordant safety signals including higher numbers of deaths, ischemic strokes, and bleeding events in the experimental group1. Median patient age was 78 years, and 26% were female.

For the primary efficacy endpoint, a hierarchic composite of death, stent thrombosis, MI, ischemic stroke, systemic thromboembolism, and urgent revascularization evaluated 6 weeks after randomization using a win-loss ratio (WLR) analysis, the experimental group showed no significant advantage over control (WLR, 1.19; 95% CI, 0.61-2.32; P = .610 for superiority)1. For the primary safety endpoint, a hierarchic composite of death and Bleeding Academic Research Consortium (BARC) bleeding categories 2 through 3c, the WLR numerically favored the control group without reaching statistical significance for superiority and without meeting the prespecified non-inferiority margin (WLR, 0.66; 95% CI, 0.39-1.11; P = .115 for superiority; P = .713 for non-inferiority).

Bleeding rates and secondary safety findings drove the DSMB decision

In the 6 weeks after randomization, secondary bleeding endpoints ran consistently higher in the experimental group, with BARC grade 3 or greater bleeding occurring more than 3-fold more often than in the control group (hazard ratio, 3.54; 95% CI, 1.15-10.85; P = .028). None of the primary or secondary efficacy endpoints differed significantly between groups at 6 weeks or 6 months across ischemic outcomes including cardiovascular death, MI, and stroke, per the trial’s exploratory secondary analyses.1,2

A landmark analysis using conditional power extrapolated to the originally planned sample size additionally suggested a higher rate of ischemic complications, including ischemic stroke, in the experimental group between weeks 6 and 6 months (hazard ratio, 1.49; 95% CI, 1.08-2.07; P = .018), though investigators cautioned this stroke signal did not follow the bleeding-then-ischemia pattern seen in prior studies and requires confirmation in future pooled analyses.1,2

Exploratory subgroup analyses found a nominally significant interaction between sex and ischemic benefit, with no ischemic events among 87 women treated with prasugrel or ticagrelor versus 6 events among 67 women treated with clopidogrel and in-hospital aspirin, alongside a signal for higher bleeding risk among the small number of patients already on a P2Y12 inhibitor at baseline who were escalated to intensified therapy.1,2

Investigators emphasized these subgroup findings were prespecified but exploratory, generated from a prematurely stopped trial with low overall event counts, and should be considered hypothesis-generating rather than confirmatory.

“Many patients who have AF receive direct oral anticoagulants to reduce their risk of stroke. Patients who have recently had a myocardial infarction receive antiplatelet therapy, in particular the combination of aspirin and a potent P2Y12 inhibitor, to reduce their risk of another ischaemic event,” said Rizas.2 “But the optimal regimen for patients suffering from both AF and MI remains unclear.”

The trial authors concluded EPIDAURUS reinforces current guideline preference for clopidogrel and in-hospital aspirin in most patients with AF and acute coronary syndrome requiring oral anticoagulation, and does not support escalation to prasugrel or ticagrelor in this setting given early termination and limited statistical power.1

A separate ongoing trial, ADONIS-PCI, is evaluating a related strategy using dabigatran and ticagrelor over a longer treatment duration and may offer complementary evidence.

References

1. Rizas KD, Mourouzis K, Rath D, et al. Dual antithrombotic therapy using potent antiplatelet inhibitors in atrial fibrillation and acute coronary syndrome: a randomized controlled trial. Nat Med. Published online August 29, 2026. doi:10.1038/s41591-026-04629-7

2. European Society of Cardiology. Intensified antiplatelet therapy plus direct oral anticoagulants increased bleeding in patients with atrial fibrillation after a heart attack. Published August 29, 2026. Accessed August 29, 2026.