3- or 1-Month DAPT in Patients at High Bleeding Risk Undergoing Everolimus-Eluting Stent Implantation

OBJECTIVES The aim of this study was to evaluate 2 abbreviated dual-antiplatelet therapy (DAPT) regimens in patients at high bleeding risk (HBR) undergoing percutaneous coronary intervention (PCI). BACKGROUND Current-generation drug-eluting stents are preferred over bare-metal stents for HBR patients, but their optimal DAPT management remains unknown. METHODS The XIENCE Short DAPT program included 3 prospective, multicenter, single-arm studies enrolling HBR patients who underwent successful PCI with a cobalt-chromium everolimus-eluting stent. After 1 month (XIENCE 28 USA and XIENCE 28 Global) or 3 months (XIENCE 90) of DAPT, event-free patients discontinued the P2Y 12 inhibitor. The postmarketing approval XIENCE V USA study was used as historical control in a propensity score – strati ﬁ ed analysis. RESULTS A total of 3,652 patients were enrolled. The propensity-adjusted

D ual-antiplatelet therapy (DAPT) with aspirin and a P2Y 12 receptor inhibitor is generally indicated for 6 to 12 months to prevent ischemic events, including stent thrombosis, after percutaneous coronary intervention (PCI) (1,2).
However, DAPT is also associated with an increased risk in bleeding, which is proportional to the duration and intensity of treatment (3). Bleeding complications affect patient morbidity and mortality (4)(5)(6)(7), and as many as 40% of patients undergoing PCI have high bleeding risk (HBR) conditions that make prolonging the duration of DAPT clinically unattractive (8). Shortening the duration of DAPT by discontinuing P2Y 12 -inhibiting therapy has therefore been proposed among such patients, but data evaluating their optimal DAPT management remain scarce.   Patients were eligible to discontinue dual-antiplatelet therapy (DAPT) at 1 or 3 month if they had been adherent to treatment (ie, without interruption of either aspirin and/or P2Y 12 inhibitor (inh.) for more than 7 consecutive d) and free from myocardial infarction, repeat coronary revascularization, stroke, or stent thrombosis.
*Reasons for ineligibility to discontinue DAPT at 1 or 3 months are not mutually exclusive. **Patients who missed the 3-months visit but completed the 12-months visit were included in the ineligible group.
Mehran et al. program and who were adherent to treatment and free from ischemic events at the time points defined previously.
The primary endpoint was assessed for the noninferiority of a short DAPT regimen to the historical control. For XIENCE 90, we assumed an event rate of 6.1% between 3 and 12 months in both arms, with noninferiority declared if the upper limit of the 1sided 97.5% confidence interval of the propensity score-stratified difference was <2.8%. We estimated that 2,000 patients were required in XIENCE 90 to achieve 87% power after allowing an attrition rate of 15%. In XIENCE 28 (combining the global and US studies), the rate of the primary endpoint was assumed to be 4.3% between 1 and 6 months.
Considering an attrition rate of 10%, with a 2.5% noninferiority margin and a 1-sided alpha level of 0.025, about 1,600 patients were required to achieve 90% power (12).
The key secondary endpoint of BARC type 2 to 5 bleeding was tested for the superiority of a short DAPT regimen over the historical control. We estimated the event rate in the historical control group to be 6.0% between 3 and 12 months and 4.6% between 1 and 6 months. The sample sizes of XIENCE 90 and XIENCE 28 provided each study with at least 90% power to detect a 50% reduction in bleeding rates with 3-month DAPT and 1-month DAPT, respectively, using a 1-sided alpha level of 0.025 (12). Finally, in XIENCE 90, the key secondary endpoint of definite or probable stent thrombosis was evaluated against a prespecified performance goal of 1.2%. An assumed 0.5% rate of stent thrombosis between 3 and 12 months in 2,000 patients receiving 3-month DAPT provided 85% power using the exact test with a 1sided alpha level of 0.025 (12).  (Supplemental Tables S7 to   S16, S20, and S24).  The propensity score-stratified mean rate of death or myocardial infarction was 3.5% in patients who discontinued DAPT at 1 month and 4.3% in those who continued DAPT up to 6 months, with a 1-sided upper  Tables S18 and S22.   Values are n/N (%) or mean AE SD, unless otherwise specified. *Anemia was defined as hemoglobin <11 g/dL. †Renal insufficiency was defined as creatinine $2 mg/dL or maintenance dialysis. ‡This category includes any systemic conditions associated with an increased bleeding risk (eg, hematologic disorders, including a history of or current thrombocytopenia defined as a platelet count <100,000/mm 3 , or any known coagulation disorder associated with increased bleeding risk). §Chronic kidney disease was defined as an estimated glomerular filtration rate <60 mL/min.

DISCUSSION
The XIENCE Short DAPT program was designed to assess the safety and efficacy of 2 abbreviated DAPT regimens in a large cohort of HBR patients who underwent successful PCI with a cobalt-chromium everolimus-eluting stent. Among patients who had been adherent to treatment and free from ischemic events while on DAPT, the discontinuation of the Importantly, XIENCE V USA did not mandate the collection of less severe bleeding corresponding to BARC type 2, which may have led to underreporting of these events and the subsequent low rates observed in our historical control compared with other similar studies (18,19). Conversely, the propensity score-stratified analysis for BARC types 3 to 5 bleeding, although post hoc, is likely to provide more reliable treatment effect estimates on a clinically relevant bleeding endpoint. In aggregate, these results suggest that among HBR patients who underwent PCI with a cobalt-chromium everolimuseluting stent, a short course of DAPT for 1 or 3 months provides similar ischemic risk and may be associated with lower occurrences of major bleeding (Central

Illustration).
Two previous studies showed that discontinuing the P2Y 12 inhibitor after 3 months of DAPT in patients undergoing PCI with a zotarolimus-eluting stent was noninferior to a standard regimen of 12 months with respect to a composite endpoint of cardiovascular and bleeding events (20,21). Contrary to our study, however, both excluded HBR patients, as shown by the very low incidence of major bleeding (<1% per year) and the absence of a significant bleeding-related benefit with the experimental arm. Furthermore, aggregate outcomes from these 2 studies suggested an increased risk for myocardial infarction or stent thrombosis with DAPT for 3 months after an acute coronary syndrome (3), with similar findings reported in another study of patients undergoing PCI with a sirolimus-eluting stent (22). More recently, a singlearm study suggested that HBR patients receiving 3month DAPT after placement of a bioabsorbable- The evidence on the effects of discontinuing the P2Y 12 inhibitor after 1 month of DAPT is more limited.
To date, such a strategy has been used only as background therapy in trials comparing different stent platforms, thus failing to provide any comparative data on the overall safety and risks of this approach (18,19,24,25). The present study is the first specifically aimed at evaluating the treatment effects of 2 different short DAPT regimens in HBR patients.
There are several risk scores and consensus-based definitions developed to assist clinicians in identifying patients at high risk for bleeding after PCI (26)(27)(28). However, commonly encountered conditions such as advanced age, long-term oral anticoagulation, anemia, and chronic kidney disease, among others, are all contemplated by the available risk assessment tools and, accordingly, were considered as inclusion criteria in our study. Furthermore, a risk for BARC type 3 or 5 bleeding of at least 4% at 1 year after PCI has been proposed as a threshold to objectively recognize HBR patients (28). In keeping with this, we observed an incidence of BARC types 3 to 5 bleeding of 6.4% between index PCI and 12 months among all patients enrolled in XIENCE 90 and of 4.6% at 6 months among those enrolled in XIENCE 28, thereby confirming the HBR nature of our study On the left, rates of the primary endpoint expressed as propensity score-stratified mean across quintiles in patients discontinuing dualantiplatelet therapy (DAPT) at 3 months (A) or at 1 month (B) after index procedure and in their respective historical controls. Error bars indicate the 95% CI of the mean. On the right, the propensity score-stratified difference between the test and the control group, with the upper limit of the 1-sided 97.5% CI (error bar) evaluated for noninferiority to the prespecified margin (dotted line) using the stratified Farrington-Manning method. UCL ¼ 1-sided upper 97.5% confidence limit.    (18,19).
Current European and American guidelines provide cautious recommendations on the use DAPT regimens of 3 months or less after drug-eluting stent implantation, acknowledging the limited available evidence, derived mostly from trials on low-risk populations (1,2,29). Our findings build on this prior evidence by demonstrating that discontinuing DAPT as early as 1 or 3 months after PCI with a new-generation everolimus-eluting stent is a safe approach and may be considered in HBR patients undergoing noncomplex procedures. Of note, prior studies investigating other drug-eluting stent platforms reported non-negligible rates of stent thrombosis with 1 month of DAPT (18,19). Such findings were attributed partly to the enhanced thrombotic risk associated with HBR conditions, which render the overall risk profile of these patients particularly elevated. It was therefore reassuring that stent thrombosis was rare (<1%) in both XIENCE 28 and XIENCE 90, even after including events occurring in the early post-PCI phase before discontinuing the P2Y 12 inhibitor. Corroborating these findings, 2 Japanese studies showed very few instances of definite or probable stent thrombosis among lower risk patients undergoing PCI with the same cobalt-chromium everolimus-eluting stent system, with no events reported among those receiving 3 months of DAPT followed by aspirin (30) and only 4 events (0.27%) among those receiving 1 month of DAPT followed by clopidogrel monotherapy (31). Whether these findings apply to other contemporary drug-eluting stents remains unclear (32). Future investigations should explore the effects of monotherapy with a P2Y 12 inhibitor compared with aspirin after a short course of DAPT among patients with HBR undergoing PCI (33).

CONCLUSIONS
Among HBR patients undergoing PCI with a cobaltchromium everolimus-eluting stent, a DAPT regimen of 1 or 3 months compared with DAPT for 6 or 12 months resulted in noninferior ischemic outcomes and a low incidence of stent thrombosis and may be associated with lower major bleeding occurrences.