Six months of dual antiplatelet therapy (DAT; aspirin and clopidogrel) was noninferior to 12 months of DAT after percutaneous coronary intervention (PCI) with implantation of a drug-eluting stent (DES). Hyeon-Cheol Gwon, MD, Sungkyunkwan University School of Medicine, Seoul, Korea, presented the findings of the EXCELLENT trial.

The Randomized Comparison of 6-Month versus 12-Month Duration of Dual Antiplatelet Therapy after Implantation of Drug-Eluting Stent: From Comparison of Everolimus- versus Sirolimus-Eluting Stents for Coronary Revascularization (EXCELLENT; NCT00698607) Trial was designed to prospectively test the hypothesis that 6 months of DAT after DES is as safe and effective as the current guideline recommendations of at least 12 months.

This open-label trial of 1443 South Korean patients with myocardial ischemia and at least 50% coronary stenosis who were undergoing PCI randomly assigned patients in a 2 × 2 factorial design to a type of DES (3:1, everolimus-eluting:sirolimus-eluting) and 1:1 to 6 versus 12 months of DAT. Subjects were stratified by the presence of diabetes and the index coronary lesion length. The primary results of the comparison of the two types of stents for in-segment late luminal loss at 9 months were presented at TCT in 2010 and showed that everolimus-eluting stents were noninferior to sirolimus-eluting stents (p for non-inferiority=0.023).

The primary endpoint for the comparison of duration of DAT was the incidence of target vessel failure (TVF) at 12 months, defined as a composite of cardiac death, myocardial infarction (MI), or target vessel revascularization (TVR). The noninferiority margin (one-sided 97.5% confidence limit) was set at a 4% absolute difference. Several secondary endpoints were evaluated, including individual components of the primary outcome, stroke, stent thrombosis, TIMI major bleeding, and a composite safety endpoint. The trial was powered, based on an estimated rate of the primary endpoint occurring in the 12-month DAT group of 10%.

The mean subject age was 63 years, one-third were women, about 40% were diabetic, and there was an even split of presentation with either stable or acute coronary syndrome. Approximately half of the patients had single-vessel coronary artery disease, predominantly involving the LAD territory. The background cardiovascular therapy included aspirin and clopidogrel (>99%) and statins (>80%), over half were on a beta-blocker, and about one-third received an ACE inhibitor.

The primary endpoint of TVF at 12 months did not differ significantly between the two groups (4.7% for 6-month therapy vs 4.4% for 12-month therapy; risk difference +0.3%, 97.5% CI +3.6%; p=0.43 for superiority; p=0.0031 for noninferiority).

However, there was a statistically significant interaction (p < 0.001) with the presence of diabetes, with diabetics having less favorable results with 6 months of DAT (HR, 3.15; p=0.005) and nondiabetics having better results with 6 months of DAT (HR, 0.42; p=0.022).

There was no evidence of interaction according to stent type.

There were no differences between 12- and 6-month DAT in the primary safety composite (3.1% vs 3.4%; p=0.76), major bleeding (0.6% vs 0.7%; p=0.42), or the secondary endpoint of stent thrombosis (0.4% vs 0.8%; p=0.33), which is of particular concern with shorter durations of DAT.

Although the primary endpoint of TVF at 12 months satisfied the study’s noninferiority criteria, the trial had low power due to the unexpectedly low event rate (10% predicted vs 4.4% observed) and a wide noninferiority margin (+4% absolute difference). With such a low event rate (4.4%) in the group that was randomized to 12 months of DAT, the upper bound of the event rate in the 6-month DAT group would needed to have been nearly doubled (>8.4%) for it to have been declared inferior. The data are limited further by a very low rate of hard clinical endpoints and a statistically significant interaction that was dependent upon diabetic status. Although the findings suggest that treatment with a thienopyridine (in addition to aspirin) may be discontinued at 6 months for nondiabetic patients, more data are needed from a larger, appropriately powered trial that assesses hard clinical endpoints before clinical practice should change