Design: Phase 3, randomised, placebo-controlled; stopped for futility
Population: 14,194 patients within 7 days of an acute coronary syndrome
Primary outcome: Cardiovascular death, myocardial infarction or ischaemic stroke
Effect: 5.4% vs 5.1%; HR 1.05 (95% CI 0.91-1.21)

LIBREXIA ACS randomised 14,194 patients within 7 days of an acute coronary syndrome to milvexian 25 mg twice daily, an oral factor XIa inhibitor, or placebo, on top of standard antiplatelet therapy. Published in the New England Journal of Medicine on 29 August 2026.

The trial was stopped at a planned interim analysis. After a median of 12.2 months, cardiovascular death, myocardial infarction or ischaemic stroke occurred in 5.4% on milvexian and 5.1% on placebo (HR 1.05, 95% CI 0.91 to 1.21). Intracranial, vision-threatening or fatal bleeding was 0.3% in both groups.

Factor XIa inhibitors were developed in the hope of preventing thrombosis without much bleeding. The bleeding half of that held here, but there was no ischaemic benefit after ACS at this dose.

Nothing changes: dual antiplatelet therapy remains the standard after ACS, and adding an anticoagulant is reserved for people with a separate indication such as atrial fibrillation.

Continue standard dual antiplatelet therapy after ACS; factor XIa inhibition added nothing here
Patients asking about 'safer blood thinners' after a heart attack can be told this trial was negative
Factor XIa inhibitors are still being tested in atrial fibrillation and stroke, so this result is specific to ACS
Add an oral anticoagulant after ACS only for a separate indication

Why it matters

Low bleeding is only valuable if the drug also prevents events, and after ACS this one did not.

The statistics, in plain English

A hazard ratio of 1.05 with a confidence interval from 0.91 to 1.21 includes no effect and rules out anything more than a small benefit. Stopping for futility means the interim data made a positive final result very unlikely.