The safety and efficacy of the XIENCE V everolimus-eluting stent (EES) compared to the Taxus Express paclitaxel-eluting stent (PES) has been demonstrated through 2 years in the SPIRIT II and III randomized clinical trials, but limited longer-term data have been reported. In the SPIRIT III trial, 1,002 patients with up to 2 lesions in 2 coronary arteries were randomized 2:1 to EESs versus PESs at 65 United States sites. At completion of 3-year follow-up, treatment with EES compared to PES resulted in a significant 30% decrease in the primary clinical end point of target vessel failure (cardiac death, myocardial infarction, or ischemic-driven target vessel revascularization, 13.5% vs 19.2%, hazard ratio 0.70, 95% confidence interval 0.50 to 0.96, p = 0.03) and a 43% decrease in major adverse cardiovascular events, cardiac death, myocardial infarction, or ischemic-driven target lesion revascularization (9.1% vs 15.7%, hazard ratio 0.57, 95% confidence interval 0.39 to 0.83, p = 0.003). In a landmark analysis, major adverse cardiovascular events were decreased to a similar extent with EES compared to PES 0 through 1 year and 1 year through 3 years (hazard ratio 0.56, 95% confidence interval 0.35 to 0.90; hazard ratio 0.59, 95% confidence interval 0.31 to 1.11, respectively). In conclusion, patients treated with EES rather than PES in the SPIRIT III trial had significantly improved event-free survival at 3 years. From 1 year to 3 years hazard curves continued to diverge in favor of EES, consistent with an improving long-term safety and efficacy profile of EES compared to PES, with no evidence of late catchup.
The Clinical Trial of the XIENCE V(r) Everolimus Eluting Coronary Stent System (SPIRIT II) randomized trial demonstrated angiographic and clinical efficacy and safety of the everolimus-eluting stent (EES) compared to the first-generation Taxus Express paclitaxel-eluting stent (PES; Boston Scientific, Natick, Massachusetts) through 3 years of follow-up and at years 1 through 3, but the modest sample size (n = 300) precluded definitive conclusions. The SPIRIT III trial was therefore performed to evaluate the relative performance of EESs and PESs in a larger cohort of patients with symptomatic coronary artery disease. In the SPIRIT III EES compared to PES demonstrated a significant decrease in the primary angiographic end point of in-segment late loss at 8 months and noninferiority for the primary clinical end point of target vessel failure (TVF) at 9 months. At 2 years, patients treated with EES versus PES showed significant decreases in TVF and major adverse cardiac events (MACEs) because of fewer myocardial infarctions and target lesion revascularization (TLR) procedures. Longer-term follow-up is necessary to characterize whether these results are sustained or increase over time. This is especially important because a small substudy from the SPIRIT II trial suggested that late angiographic catchup between the 2 stents is possible. We therefore report a 3-year follow-up from the SPIRIT III trial using landmark analysis to investigate the possibility of late clinical catchup.
Methods
The SPIRIT III trial design has been described previously. In brief, SPIRIT III was a multicenter prospective single-blinded randomized controlled trial that evaluated the safety and efficacy of EESs compared to PESs in patients with up to 2 de novo coronary artery lesions with a reference vessel diameter of 2.5 to 3.75 mm and lesion length ≤28 mm. In total 1,002 patients were randomized 2:1 to EES versus PES at 65 United States sites.
Patients ≥18 years of age with stable or unstable angina or inducible ischemia undergoing percutaneous coronary intervention were considered for enrollment. Key inclusion criteria included a maximum of 2 de novo native coronary artery lesions, each in a different epicardial vessel, with visually assessed diameter stenoses ≥50% and <100%, reference vessel diameter of 2.5 to 3.75 mm, and lesion length ≤28 mm. Major clinical exclusion criteria included percutaneous coronary intervention before or planned within 9 months of the index procedure, acute or recent myocardial infarction, left ventricular ejection fraction <30%, long-term use of warfarin or immunosuppressive therapy, known autoimmune disease, renal insufficiency, recent major bleed, hemorrhagic diathesis or objection to blood transfusions, contraindications or allergy to any study medication, components of study stents, or iodinated contrast that could not be premedicated, elective surgery planned within 9 months after the procedure necessitating antiplatelet agent discontinuation, platelet count <100,000 or >700,000 cells/mm 3 , white blood cell count <3,000 cells/mm 3 , serum creatinine >2.5 mg/dl or dialysis or liver disease, stroke or transient ischemic attack within 6 months, co-morbid conditions limiting life expectancy to <1 year or that could affect protocol compliance, and participation in another investigational study that had not yet reached its primary end point. Key angiographic exclusion criteria included aorto-ostial or left main coronary artery location, excessive tortuosity or angulation, heavy calcification, thrombus, or other significant lesions (>40% diameter stenosis) in the target vessel or side branch. If 2 target lesions were treated, each lesion had to meet all angiographic inclusion/exclusion criteria.
After confirmation of angiographic eligibility, telephone randomization was performed using an automated voice response system, stratified by presence of diabetes, planned 2-vessel treatment, and study site. Although operators were by necessity unblinded during stent implantation, the patient and staff involved in follow-up assessments remained blinded through the follow-up period, with a standardized follow-up interview script used to decrease bias.
Predilatation of the target lesion with standard balloon angioplasty was mandatory. The stent was implanted to cover a minimum of 3 mm of near normal vessel on either side of the lesion. Postdilatation was left to the discretion of the investigator. If an additional stent was required for bailout purposes, a stent from the same treatment arm was used.
Subjects who were not on long-term antiplatelet or aspirin therapy were required to receive a loading dose of aspirin ≥300 mg before the procedure and clopidogrel bisulfate ≥300 mg no later than 1 hour after the procedure. All patients were to be maintained on clopidogrel bisulfate 75 mg/day for a minimum of 6 months and aspirin ≥80 mg/day throughout the length of the trial (5 years) after the index procedure. Other medications were prescribed according to standard of care.
Clinical follow-up was prespecified and scheduled at 30 ± 7, 180 ± 14, 240 ± 28, 270 ± 14, and 365 ± 28 days with subsequent telephone follow-up yearly (± 28 days) thereafter through 5 years. Angiographic follow-up was planned in a subset of 564 patients at 8 months as previously described.
The primary clinical end point of the SPIRIT III trial was TVF, consisting of the composite of cardiac death, myocardial infarction, or ischemic-driven target vessel revascularization by percutaneous coronary intervention or bypass graft surgery. Secondary end points included MACEs defined as the composite of cardiac death, myocardial infarction, or ischemia-driven TLR, and individual components of TVF and MACE, and stent thrombosis. Target vessel (or lesion) revascularization was considered ischemia driven if associated with a positive functional study, a target vessel (or lesion) diameter stenosis ≥50% by core laboratory quantitative analysis with ischemic symptoms, or a target vessel (or lesion) diameter stenosis ≥70% with or without documented ischemia. Myocardial infarction was defined as development of new pathologic Q waves ≥0.4 second in duration in ≥2 contiguous leads or an increase of creatine phosphokinase levels to >2.0 times normal with positive creatine phosphokinase-MB. Stent thrombosis was prospectively defined by protocol as an acute coronary syndrome with angiographic evidence of thrombus within or adjacent to a previously treated target lesion or, in the absence of angiography, any unexplained death or acute myocardial infarction with ST-segment elevation or new Q waves in the distribution of the target lesion occurring within 30 days after the procedure. Definite or probable stent thrombosis was also adjudicated in post hoc analysis using Academic Research Consortium definitions.
Independent study monitors verified 100% of case-report form data onsite. An independent committee blinded to treatment allocation adjudicated all deaths, myocardial infarctions, revascularizations, and protocol stent thromboses after review of original source documentation. A second clinical events committee blinded to randomization performed post hoc adjudication of stent thrombosis using Academic Research Consortium definitions. Independent core angiographic and intravascular ultrasound laboratory analyses were performed by technicians blinded to treatment assignment and clinical outcomes using validated methods as previously described.
All analyses were performed on an intention-to-treat population consisting of patients randomized in the study regardless of the treatment actually received. However, patients lost to follow-up in whom no event had occurred before follow-up windows were not included in the denominator for calculations of binary end points. Categorical variables were compared by Fisher’s exact test. Continuous variables are presented as mean ± 1 SD and were compared by t test. Time-to-event hazard curves were constructed using Kaplan–Meier estimates and compared by log-rank test. Landmark analyses were created to assess clinical event rates after different intervals. For these analyses, patients who had events between the index procedure and the landmark interval were excluded from Kaplan–Meier estimates after the landmark. Logistic regression analysis with interaction testing was performed to examine whether relative decreases in MACEs for EESs compared to PESs were consistent across important subgroups at 3 years. A 2-sided alpha values equal to 0.05 was used for all statistical tests to define significance. All statistical analyses were performed by SAS 9.1.3 (SAS Institute, Cary, North Carolina).
Results
From June 22, 2005 through March 15, 2006, 1,002 patients were enrolled at 65 United States sites and randomized to receive EES (n = 669) or PES (n = 332). Baseline demographics and clinical and angiographic characteristics of the 2 treatment groups have been published previously and were well matched in the study population ( Table 1 ).
Variable | EES (n = 669) | PES (n = 332) | p Value |
---|---|---|---|
Age (years) | 63.2 ± 10.5 | 62.8 ± 10.2 | 0.54 |
Men | 70.1% (469/669) | 65.7% (218/332) | 0.17 |
Diabetes mellitus requiring | 29.6% (198/669) | 27.9% (92/330) | 0.60 |
Insulin | 7.8% (52/669) | 5.5% (18/330) | 0.19 |
Hypertension ⁎ | 76.2% (510/669) | 74.0% (245/331) | 0.48 |
Hypercholesterolemia ⁎ | 74.2% (489/659) | 71.5% (233/326) | 0.36 |
Current smoker | 23.4% (154/659) | 22.5% (73/324) | 0.81 |
Previous myocardial infarction | 19.9% (130/652) | 18.0% (59/327) | 0.49 |
Unstable angina pectoris | 18.7% (123/657) | 25.1% (82/327) | 0.02 |
Lesions treated | 1.2 ± 0.4 | 1.2 ± 0.4 | 0.99 |
Target coronary artery | |||
Patients | 772 | 383 | |
Left anterior descending | 41.3% (317/768) | 42.9% (164/382) | 0.61 |
Left circumflex | 27.6% (212/768) | 28.3% (108/382) | 0.83 |
Right | 31.0% (238/768) | 28.5% (109/382) | 0.41 |
Left main (protected) | 0.1% (1/768) | 0.3% (1/382) | 0.55 |
Target lesion | |||
Patients | 772 | 383 | |
Reference vessel diameter (mm) | 2.77 ± 0.45 | 2.76 ± 0.46 | 0.87 |
Minimal luminal diameter (mm) | 0.82 ± 0.41 | 0.83 ± 0.40 | 0.79 |
Diameter stenosis (%) | 70.0 ± 13.3 | 69.4 ± 13.6 | 0.54 |
Lesion length (mm) | 14.7 ± 5.6 | 14.7 ± 5.7 | 0.92 |
Aspirin use | |||
At 6 months | 97.4% (646/663) | 97.2% (318/327) | 0.84 |
At 1 year | 97.7% (631/646) | 97.1% (306/315) | 0.66 |
At 2 years | 96.1% (598/622) | 95.1% (289/304) | 0.49 |
At 3 years | 93.7% (577/616) | 95.3% (284/298) | 0.37 |
Thienopyridine use | |||
At 6 months | 94.4% (627/664) | 94.2% (308/327) | 0.88 |
At 1 year | 71.9% (464/645) | 70.8% (223/315) | 0.76 |
At 2 years | 57.3% (356/621) | 60.5% (184/304) | 0.36 |
At 3 years | 52.4% (323/616) | 52.0% (155/298) | 0.94 |
Three-year clinical follow-up was available in 948 patients (94.6%) including 636 with EESs and 312 with PESs. Aspirin use was high in the EES and PES arms throughout the 3-year follow-up period, whereas thienopyridine use decreased progressively over time ( Table 1 ).
Figure 1 shows binomial rates of major clinical end points at 3 years. Figure 2 displays Kaplan–Meier hazard event curves through 3 years. By the end of the 3-year follow-up period, treatment with EES compared to PES resulted in a significant 30% decrease in rate of TVF and a significant 43% decrease in rate of MACEs. These differences were driven by a significant 39% decrease in ischemia-driven TLR and a trend toward a 38% decrease in the composite end point of cardiac death or myocardial infarction. Rates of stent thrombosis by the protocol and Academic Research Consortium definitions were not significantly different between the 2 groups.