A-CLOSE: clopidogrel monotherapy versus extended DAPT after drug-eluting stents

Summary

A-CLOSE randomized 3203 patients at high ischemic risk to clopidogrel monotherapy or extended DAPT. Randomization took place 12 months after drug-eluting stent implantation, with 24 months of follow-up.

The primary endpoint combined death from any cause, MI, stent thrombosis, stroke and BARC 2, 3 or 5 bleeding. There were 80 events (5.0%) with monotherapy and 81 (5.1%) with extended DAPT. Risk difference −0.1 percentage points (90% CI −1.3 to 1.2), against a prespecified margin of 2.3 points. P = 0.001 for noninferiority.

The two key secondary endpoints, tested in a prespecified hierarchy, were significant in opposite directions: ischemic composite HR 2.33 (1.47 to 3.69) and bleeding HR 0.43 (0.27 to 0.67).

Strengths

• Randomized, with a prespecified noninferiority margin.
• Prespecified testing hierarchy for secondary endpoints.
• Enough power to detect the ischemic and bleeding effects separately.

Methodological issues

• Equal weights in the composite. BARC 2 is nonmajor, medically actionable bleeding. In the composite it counts the same as death.
• Components in opposite directions. Monotherapy adds 14 deaths, 17 MIs and 9 stent thromboses, and avoids 20 major and 17 nonmajor bleeds. Fatal bleeding (BARC 5) was 3 vs 2, so it does not explain the mortality difference. The null primary result reflects cancellation, not low power.
• Noninferiority at the edge. With BARC 3 or 5 only (post hoc), the composite was 4.3% vs 3.2%, difference 1.1 points (95% CI −0.3 to 2.4), HR 1.33 (0.93 to 1.92). Converted to a 90% CI, the upper bound is about 2.23 points against a margin of 2.3. Noninferiority holds by about 0.07 percentage points.
• Low fragility indices. From the published counts: 18 for the ischemic composite, 20 for bleeding, 8 for MI, 5 for all-cause death, 4 for cardiovascular death and 2 for stent thrombosis. All-cause death sits outside the testing hierarchy: a signal to track, not a basis for action.
• Broad entry criteria. 79.3% presented with acute coronary syndrome and 68.6% had a diffuse long lesion (total stent length ≥ 28 mm). Most contemporary PCI patients meet one criterion. Without contrast inside the cohort, the trial cannot identify who benefits. PRECISE-DAPT and DAPT score subgroups point the same way in both strata.
• Women underrepresented. 589 patients (18.4%), ischemic HR 2.35 (0.91 to 6.05) and bleeding HR 0.50 (0.20 to 1.24). Both intervals cross 1 for lack of events.
• Exclusions that limit generalization. Patients over 80 (the age band where women are best represented in PCI cohorts), oral anticoagulation and major bleeding within 3 months. Single country, homogeneous population, mean age 62.6 years.
• No CYP2C19 genotyping or platelet function testing. The ischemic excess cannot be separated from clopidogrel hyporesponsiveness.