- Email: [email protected]
Rationale and design of the clinical evaluation of the Resolute Zotarolimus-Eluting Coronary Stent System in the treatment of de novo lesions in native coronary arteries (the RESOLUTE US clinical trial)
Trial Design
Rationale and design of the clinical evaluation of the Resolute Zotarolimus-Eluting Coronary Stent System in the treatment of de novo lesions in native coronary arteries (the RESOLUTE US clinical trial) Laura Mauri, MD, MSc, a,b Martin B. Leon, MD, c Alan C. Yeung, MD, d Manuela Negoita, MD, e Michelle J. Keyes, PhD, f and Joseph M. Massaro, PhD f Boston, MA; New York, NY; and Stanford and Santa Rosa, CA
Background Drug-eluting stents (DES) are commonly used to treat obstructive coronary disease and avoid restenosis. Newer DES have been developed to improve effectiveness and safety. We describe a clinical trial to evaluate a DES with a novel polymer that may improve the antirestenosis effectiveness while maintaining the safety standards of currently Food and Drug Administration–approved DES. Methods The RESOLUTE US Trial is a multicenter, nonrandomized trial prospectively designed to compare the Resolute zotarolimus-eluting stent (R-ZES) to the Food and Drug Administration–approved Endeavor ZES using patient-level historical control data, adjusting for baseline covariates through propensity score. The stents differ primarily in the polymer, which, in the R-ZES, is designed to elute zotarolimus over a longer period. The study will enroll up to 1,574 patients with ischemic heart disease due to de novo native coronary lesions suitable for 1- or 2-vessel treatment with stents from 2.25 to 4.0 mm in diameter. The primary end point is target lesion failure at 12 months postprocedure, defined as the composite of cardiac death, targetvessel myocardial infarction (MI), and clinically driven target lesion revascularization by percutaneous or surgical methods. Secondary end points include device, lesion and procedural success, death, MI, cardiac death and MI, composites of these clinical events, and stent thrombosis at each follow-up assessment up to 5 years postprocedure. Conclusions The RESOLUTE US Trial (ClinicalTrials.gov #NCT00726453) is a prospective, multicenter, observational study with a patient-level historical control designed to assess the safety and efficacy of the R-ZES for the treatment of de novo lesions in native coronary arteries. (Am Heart J 2011;161:807-14.)
Background Drug-eluting stents (DES) that allow local delivery of the antiproliferative drugs from the surface of a stent have been shown to dramatically reduce vascular neointimal hyperplasia and subsequent renarrowing of the coronary stent and the requirement for subsequent procedures.1-3 However, repeat procedures within the first year are still required in about 5% of DES patients and in a greater proportion of higher risk patients or lesions.4 Further-
From the aBrigham and Women's Hospital, Boston, MA, bHarvard Medical School, Boston, MA, cColumbia University Medical Center/New York Presbyterian Hospital, New York, NY, dStanford University School of Medicine, Stanford, CA, eMedtronic, Santa Rosa, CA, and fHarvard Clinical Research Institute, Boston, MA. Submitted December 30, 2010; accepted March 9, 2011. Reprint requests: Laura Mauri, MD, MSc, Division of Cardiovascular Medicine, Department of Medicine, Brigham and Women's Hospital, 75 Francis Street, Boston, MA 02115. E-mail: [email protected] 0002-8703/$ - see front matter © 2011, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2011.03.015
more, the long-term safety of DES remains a matter of clinical investigation.5 Currently approved DES use various different medications, polymers, and stent designs. The Resolute zotarolimus-eluting coronary stent (R-ZES; Medtronic, Santa Rosa, CA) uses the same cobalt chromium stent platform and drug (zotarolimus) as the Food and Drug Administration–approved Endeavor zotarolimus-eluting coronary stent (E-ZES; Medtronic). The primary difference between these 2 stents is that the R-ZES uses a new polymer that allows for the extended release of zotarolimus over a period of approximately 180 days.6
Study design The primary objective of the RESOLUTE US Trial is to assess the safety and efficacy of the R-ZES for the treatment of de novo lesions in native coronary arteries with a reference vessel diameter of 2.25 to 4.2 mm and up to 35 mm in length. The RESOLUTE US Trial (ClinicalTrials.gov #NCT00726453) is a prospective,
American Heart Journal May 2011
808 Mauri et al
Figure 1 Reference Vessel Diameter Lesion length ≤27 mm (≤35 mm for 38 mm Length Sub-study)
Study Designation Per stent size
Sites Participating Two lesion treatment permitted during index procedure (1 lesion per vessel up to 2 vessels) Enrollment Enrollment Breakdown by Stent Size
3.0 mm – 4.2 mm
2.25 mm – 4.2 mm
38 mm Length Group
*
2.25 mm – 3.5 mm Angio/IVUS Cohort
4.0 mm Stent Group
All sites
Limited number of sites
All sites
Limited number of sites
Yes
Yes
Yes
Yes
100
58
110 - 175
2.25 mm – 3.5 mm Clinical Cohort
1241 129 (2.25 mm) 1112 (2.5 mm – 3.5 mm)
20 (2.25 mm) 80 (2.5 mm – 3.5 mm)
58 (4.0 mm)
110 - 175 (38 mm)
Angiographic Follow-up
No
Eight (8) Month
Eight (8) Month
No
IVUS Follow-up
No
Eight (8) Month
No
No
Twelve (12) Month Patient Contact
Nine (9) Month Clinic Contact
Nine (9) Month Clinic Contact
Twelve (12) Month Patient Contact
Primary Clinical Follow-up
up to 1574
RESOLUTE US Clinical Trial Design.
multicenter, observational study with a patient-level historical control (Figure 1). The study is designed to reproduce the inclusion and exclusion characteristics of the E-ZES clinical trials, with uniform definition and adjudication, to allow direct comparison of patient-level historical clinical data for the primary comparison.
Population Patients will be enrolled at up to 116 investigational sites in the United States, either with prospective, clinical follow-up only (clinical cohort, n = 1,351-1,416) or with protocol-specified angiographic follow-up (angiographic cohort, n = 158). Table I lists the general and angiographic inclusion and exclusion criteria. A patient's inclusion in a given subanalysis cohort is dependent on the size (diameter) or length of the stent(s) the patient receives (Figure 1). General inclusion and exclusion criteria are similar for all of the cohorts (Table I). Clinical cohort The clinical cohort includes patients enrolled in the 2.25- to 3.5-mm main clinical cohort and 38-mm length stent group. The main clinical cohort will enroll patients with ischemic heart disease due to de novo native coronary lesions between 2.25 and 3.5 mm and lesion lengths of ≤27 mm, amenable to percutaneous treatment with a stent. Patients may receive treatment of up to 2 lesions, provided that the lesions are located in separate target vessels and treated during a single index procedure. Of the 1,241 patients planned to be enrolled in the main study, 1,112 will receive only 2.5- to 3.5-mm stents, and
129 will receive at least one 2.25-mm stent. The 2.25-mm stent group will be analyzed separately from the 2.5- to 3.5-mm stent clinical cohort. For the 38-mm length substudy, between 110 and 175 patients with lesion lengths ≤35 mm will be enrolled.
Angiographic cohort The angiographic cohort includes 100 patients to be enrolled in the 2.25- to 3.5-mm stent angiography and intravascular ultrasound (angio/IVUS) cohort, of which 80 will receive only 2.5- to 3.5-mm stents and 20 will receive at least one 2.25-mm stent, and 58 patients are to be enrolled in the 4.0-mm stent group with at least 1 lesion amenable to treatment with a 4.0-mm stent. Medications All patients will receive aspirin (a minimum of 75 mg daily) and a loading dose of ≥300 mg clopidogrel within 24 hours before the procedure or within 30 minutes postprocedure. Bivalrudin or heparin, to maintain an activated clotting time ≥250 seconds, will be administered at the time of the procedure. Administration of 50 to 200 μg of intracoronary nitroglycerin is required before stenting and before postintervention angiograms. Postprocedure, patients will receive a minimum of 75 mg of aspirin daily indefinitely and 75 mg daily of clopidogrel for a minimum of 6 months in all patients and up to 12 months in patients who are not at high risk of bleeding by investigator determination, per American College of Cardiology/American Heart Association/Society for Cardiac Angiography and Interventions guidelines.7 Dual-
American Heart Journal Volume 161, Number 5
Mauri et al 809
Table I. RESOLUTE US Inclusion and Exclusion Criteria General inclusion criteria (subjects must meet all of the following criteria for eligibility for treatment in the trial) • N18 y of age • Acceptable candidate for PCI stenting and emergent CABG surgery • Clinical evidence of ischemic heart disease, stable or unstable angina, silent ischemia, and/or a positive functional study • Negative pregnancy test within 7 d before the trial procedure if female of childbearing potential • Provided written informed consent as approved by the institutional review board/ethics committee of the respective investigational site • Agrees to comply with specified follow-up evaluations and to return to the same investigational site where the procedure was performed Angiographic inclusion criteria (the subject and each target lesion/vessel must meet all of the following angiographic criteria for the patient to be considered for inclusion in the trial) • Single target lesion, or 2 target lesions located in separate target vessels • Target lesion(s) is(are) de novo lesion(s) in native coronary artery(ies) • Target lesion(s) is(are) ≤27 mm in length (or ≤35 mm for a lesion to be treated with a 38-mm length stent) • Target lesion(s) is(are) a stenosis of ≥50% and b100% • Target-vessel(s) reference vessel diameter 2.25-4.2 mm (or 3.0-4.2 mm for it to be treated with a 38-mm length stent) • Target-vessel(s) thrombolysis in MI flow ≥2 General exclusion criteria (subjects will be excluded from the trial if any of the following criteria are met) • Known hypersensitivity or contraindication to aspirin, heparin and bivalirudin, clopidogrel and ticlopidine, cobalt, nickel, chromium, molybdenum, polymer coatings (eg, BioLinx), or a sensitivity to contrast media, which cannot be adequately premedicated • History of an allergic reaction or significant sensitivity to drugs such as zotarolimus, rapamycin, tacrolimus, everolimus, or any other analog or derivative • Platelet count b100,000 or N700,000 cells/mm3, or a white blood cell count b3,000 cells/mm3 within 7 d before index procedure • Serum creatinine level N2.5 mg/dL within 7 d before index procedure • Evidence of an acute MI within 72 h of the intended trial procedure: (a) Q wave MI or (b) Any elevation of CK-MB isoenzyme above the laboratory upper limit of normal and has not returned to normal at the time of the index procedure • Previous PCI of the target vessel(s) within 9 mo before the procedure • Planned PCI of any vessel within 30 d postindex procedure and/or planned PCI of the target vessel(s) within 12 mo postprocedure • During the index procedure, the target lesion(s) requires treatment with a device other than PTCA before stent placement (including, but not limited to, cutting balloon, atherectomy, laser, thrombectomy, etc) • History of a stroke or transient ischemic attack within the prior 6 mo • Active peptic ulcer or upper gastrointestinal bleeding within the prior 6 mo • History of bleeding diathesis or coagulopathy or will refuse blood transfusions • Concurrent medical condition with a life expectancy of less than 12 mo • Any previous treatment of the target vessel(s) for restenosis, including brachytherapy • Currently participating in an investigational drug or another device trial that has not completed the primary end point or that clinically interferes with the current trial end points, or requires coronary angiography, IVUS, or other coronary artery imaging procedures • Documented left ventricular ejection fraction b30% at the most recent evaluation • Inability to comply with the required trial antiplatelet regimen Angiographic exclusion criteria (the subject will be excluded from the trial if any of the following criteria are met (for patients with 2 target lesions, both target lesions/vessels must not meet any of the criteria below)) • Target lesion(s) is(are) located in native vessel(s) distal to anastomosis with a bypass graft (including but not limited to saphenous vein graft or a left/right internal mammary artery) with N40% diameter stenosis anywhere within the graft • Previous stenting in the target vessel(s) unless the following conditions are met: (a) It has been at least 9 mo since the previous stenting and (b) Target lesion(s) is(are) at least 15 mm away from the previously placed stent • Target vessel(s) has/have other lesions with N40% diameter stenosis based on visual estimate or online quantitative coronary angiography • The target vessel(s) has(have) evidence of thrombus • The target vessel(s) is(are) excessively tortuous (2 bends N90° to reach the target lesion) • The target lesion(s): (a) Location is aorto-ostial, an unprotected left main lesion, or within 5 mm of the origin of the left anterior descending or left circumflex. (b) It involves a side branch N2.0 mm in diameter. (c) It is at or distal to a N45° bend in the vessel. (d) It is severely calcified. • Unprotected left main coronary artery disease (an obstruction N50% in the left main coronary artery)
antiplatelet therapy use will be assessed at 1, 6, and 12 months and annually through 5-year follow-up.
Device description The R-ZES has been previously described in detail elsewhere.8,9 The study stent is available in diameters of 2.25, 2.5, 2.75, 3.0, 3.5, and 4.0 mm and lengths of 12, 18, 24, 30, and 38 mm for each diameter. Resolute zotarolimus-eluting stent 8 mm long with diameters of
2.25, 2.5, and 2.75 mm and 9 mm long with diameters of 3.0, 3.5, and 4.0 mm are available for use only as a secondary stent in cases of insufficient lesion coverage or procedure-related edge dissections.
Treatment/study procedures All patients must provide institutional review board– approved informed consent before study enrollment. After baseline angiography, target lesions must be
American Heart Journal May 2011
810 Mauri et al
predilated with standard percutaneous transluminal balloon angioplasty techniques. Postdilatation of the stent segments is at the discretion of the operator. Trial stents ≥3mm longer than the lesion length will be selected to treat each target lesions. For patients with planned treatment of 2 lesions, the first lesion must be treated successfully, and the patient must be clinically stable before treatment of the second lesion is attempted. Successful treatment of the first lesion requires that a b10% residual diameter stenosis result is achieved (visual assessment), that thrombolysis in myocardial infarction (MI) 3 flow is present posttreatment, and that there is no evidence of dissection, thrombus, or distal embolization at the first study lesion site posttreatment. Final angiography will be performed upon procedure completion. Treatment failure and patients not treated with the study stent postenrollment will be followed for safety purposes and included in the intention-to-treat (ITT) analyses described below. Predischarge, creatine kinase (CK) and CK-MB isoenzymes will be measured at 6 to 8, 12 to 16, and 20 to 24 hours postprocedure. Although not required for comparison with the historical data, troponin levels will be sampled to allow for comparison with the Academic Research Consortium (ARC)/ Universal MI definition.10,11
Follow-up schedule Demographic, clinical, and procedural information at the time of enrollment will be captured as well as subsequent clinical end points, serious adverse events, and concomitant medications. Clinical follow-up assessments at the same investigational site where the index procedure was performed will be performed for all patients at 30 days and at 6, 9, 12, and 18 months postprocedure and annually through 5 years postprocedure. In addition, clinical, angiographic, and IVUS assessment will be performed at 8 months in the 2.25- to 3.5-mm angio/IVUS cohort and the 4.0-mm stent group. Figure 2 represents a flowchart of follow-up assessments.
Figure 2 All Patients Thirty (30) Day Clinic Visit Six (6) Month Patient Contact
Patients in 4.0 mm Stent Group
Patients in 2.25 mm – 3.5 mm Clinical Cohort and 38 mm Length Group
Patients in 2.25 mm – 3.5 mm Angio/IVUS Cohort Eight (8) Month Clinic Visit with Angiography and IVUS
Eight (8) Month Clinic Visit with Angiography All Patients
Nine (9) Month Clinic Visit
Twelve (12) Month Patient Contact
Eighteen (18) Month Patient Contact
Two-Five (2-5) Year Patient Contact
RESOLUTE US Clinical Trial patient follow-up flowchart.
End points and definitions
For the angio/IVUS cohort and 4.0-mm stent group, the primary end point is 8-month late lumen loss (LLL), defined as the difference between the postprocedure minimal lumen diameter and the follow-up angiography minimal lumen diameter as measured by quantitative coronary angiography. A full listing of the study end points can be found in Table II.
For the main clinical cohort, the 2.25-mm stent group, and the 38-mm length group, the primary end point is target lesion failure (TLF) at 12 months, defined as cardiac death, target-vessel MI (Q wave and non–Q wave), or clinically-driven target lesion revascularization (TLR) by percutaneous or surgical methods. Secondary end points include death, MI, cardiac death and MI, major adverse cardiac events, target-vessel failure (TVF) composite and individual components, and stent thrombosis (ST) at each follow-up assessment. Definitions are listed below.
Definitions For the RESOLUTE US Clinical Trial, all deaths are considered cardiac unless an unequivocal noncardiac cause can be established. Major adverse cardiac event is defined as death, MI (Q wave and non–Q wave), emergent coronary bypass surgery, or clinically-driven TLR by percutaneous or surgical methods. Target-vessel failure components include cardiac death, target-vessel MI, and clinically-driven target vessel revascularization (TVR).
American Heart Journal Volume 161, Number 5
Table II. RESOLUTE US Clinical, angiographic, and IVUS end points Clinical end points: (at 30 d and at 6, 9, 12, and 18 mo postprocedure and annually thereafter through year 5) • Device success • Lesion success • Procedure success • Death • MI, Q wave, and non–Q wave • Cardiac death and MI reported at 30 d and at 6, 9, 12, and 18 mo postprocedure and annually thereafter through year 5 • Major adverse cardiac events composite end point (death, MI, Q wave, and non–Q wave), emergent CABG MI, or clinically driven repeat TLR by percutaneous or surgical methods) and each individual component • TLF composite end point (cardiac death, target-vessel MI, or clinically driven TLR) and each individual component • TVF composite end point (cardiac death, target-vessel MI, or clinically driven target-vessel revascularization by percutaneous or surgical methods) and each individual component • Stent thrombosis (Medtronic historic protocol definitions and ARC definitions) Angiographic end points (at 8 mo) • In-segment LLL at 8 mo • In-stent and in-segment binary angiographic restenosis rate (defined as N50% diameter stenosis) at 8 mo postprocedure • In-stent and in-segment minimum lumen diameter at 8 mo postprocedure • In-stent and in-segment percent diameter stenosis at 8 mo postprocedure IVUS end points (at 8 mo) • Rates of incomplete stent apposition at 8 mo postprocedure, categorized as persistent or late • Neointimal hyperplastic volume and percent volume obstruction (%VO) at 8 mo postprocedure
Q wave MI requires a history of chest pain or other acute symptoms consistent with myocardial ischemia together with new pathological Q waves in 2 or more contiguous electrocardiogram leads or new pathological Q waves and elevated cardiac enzymes. Non–Q wave MI is defined by elevated CK two or more times the upper limit of normal with the presence of an elevated CK-MB in the absence of new pathological Q waves. TLR is defined as clinicallydriven repeat percutaneous coronary intervention (PCI) of the target lesion or coronary artery bypass grafting (CABG) including the target vessel. TVR is defined as clinically-driven repeat PCI or CABG of the target vessel. ST is defined as in the prior E-ZES studies, and all ST events will be adjudicated according to ARC criteria.11
Statistical analysis Main clinical cohort analysis The 2.5- to 3.5-mm stent clinical cohort analysis is designed to assess the noninferiority of the 2.5-, 2.75-, 3.0-, and 3.5-mm diameter R-ZES compared with patientlevel historical control data from the ENDEAVOR Clinical Program. Historical E-ZES patients with diameters 2.5- to 3.5 mm who were part of a clinical follow-up cohort (no
Mauri et al 811
planned angiographic follow-up) were identified from those E-ZES patients with study clinical follow-up only (ENDEAVOR II, ENDEAVOR IV, ENDEAVOR PK, and ENDEAVOR II Continued Access trials).12-14 The 12month TLF rate for this group of 1,076 patients was 6.5% (calculated from Medtronic data on file). Because the historical control arm only includes single-lesion patients, the 2.5- to 3.5-mm stent clinical cohort will be analyzed for comparing the single-lesion subset of the R-ZES arm and historical control arm, using a noninferiority margin (delta) of 3.3%. A total of 800 analyzable patients will yield 87% power to assess the noninferiority of the R-ZES compared with EZES using a 1-sided .05 level of significance. Up to 890 single-lesion patients will be enrolled to account for loss to follow-up (expected to be less than 10%). Anticipating a rate of dual-vessel treatment of 20%, a total of 1,112 patients will be enrolled with either single- or dual-vessel treatment. The primary analysis cohort will be the ITT main clinical cohort defined as patients who met the study entry criteria, signed the written informed consent, and were enrolled in the trial. In addition, as a secondary analysis, the noninferiority assessment will be performed on the per-protocol analysis cohort, which is defined as the ITT sample excluding patients who are deregistered or do not meet certain key entry criteria. Because R-ZES and the E-ZES patients are from different studies, there is a possibility of treatment imbalance on clinically relevant baseline characteristics. If there are baseline imbalances between the 2 groups (at a 2-sided .05 level of significance), a propensity score for group (Resolute, Endeavor) membership will be calculated using logistic regression and then categorized into quintiles based on this propensity score. To assess noninferiority of R-ZES to E-ZES at the 1-sided .05 level of significance, a 1-sided upper 95% CI of the treatment difference (Resolute minus Endeavor) in TLF rates adjusted for propensity score quintile will be calculated according to Koch and Amar15 and compared with the noninferiority margin of 3.3%. This noninferiority analysis is the primary analysis of the study. If noninferiority is met at the 1-sided .05 level of significance, only then will analysis on prespecified subsets proceed as described below.
2.25-mm stent group analysis The 2.25-mm stent group is designed to assess the effectiveness and safety of the 2.25-mm diameter R-ZES compared with a performance goal derived from a logistic regression model based on E-ZES patients. The analysis population consists of all 129 patients receiving a 2.25-mm stent in the clinical cohort and 20 patients in the angio/IVUS cohort (149 total patients). Based on historical E-ZES data and the planned proportion of subjects with scheduled angiographic follow-up, the expected
American Heart Journal May 2011
812 Mauri et al
TLF rate is 12.87% at 12 months (calculated from Medtronic data on file). The derived performance goal for TLF at 12 months was set at 20%. If the 12-month TLF rate of the 2.25-mm stent group of the pooled clinical and angio/IVUS cohorts is shown to be significantly less than 20% at a 1-sided .05 level of significance, then the analysis of this group will be considered to have met its primary objective. At a 1-sided .05 level of significance, a sample size of 140 evaluable subjects yields 80% power to claim the true TLF rate is b20% under the assumption that the true TLF rate is in actuality 12.87%; 149 patients will be enrolled to account for up to 6% premature withdrawal by the 12-month analysis time point. Generally, with regard to clinical outcomes analyses, in patients receiving treatment of both lesions with stents of the same size group (2.25, 4.0, or 38 mm), the lesion to be included in the primary analysis, the “analysis lesion,” will be randomly selected at the time of analysis. For patients receiving treatment of 2 lesions, 1 with a 2.25-mm and 1 with a 4.0-mm stent, the primary analysis will include both lesions, and the patient will be counted twice, once in each study. A secondary analysis will be conducted for all target lesions.
Angiographic cohort analysis The 2.25- to 3.5-mm angio/IVUS cohort analysis will compare mean 8-month in-stent LLL in the RESOLUTE 2.25- to 3.5-mm imaging cohort, compared with mean 8month in-stent LLL of the historical control subset of the ENDEAVOR II trial, using a noninferiority margin (delta) of 0.16 mm. Assuming a true equivalence of mean LLL between the 2 cohorts and with a common standard deviation of 0.46, an approximate 20% nonevaluable rate, and 264 subjects in the historical control, 100 angio/IVUS patients (80 evaluable) achieves 86% power to detect noninferiority at a 1-sided significance level of .05. 4.0-mm stent group analysis The 4.0-mm stent group analysis will compare the mean 8-month in-segment LLL of the 4.0-mm diameter RZES to that of a historical control derived from the Medtronic Driver stent data, for superiority at a 1-sided .05 level of significance. Assuming at least 0.15-mm mean LLL benefit of R-ZES over Driver, standard deviations of 0.62 and 0.46 in the Driver and R-ZES arms and 894 Driver patients, 46 R-ZES patients yield 90% power to claim superiority of R-ZES to Driver; 58 RZES patients will be enrolled to account for an expected 20% loss to follow-up. 38-mm length group analysis The 38-mm length stent group analysis will compare the 12-month TLF rate to a performance goal of 19%. Additional data on the 38-mm R-ZES will be used from other RESOLUTE worldwide clinical trials in this com-
parison. Assuming the true 12-month TLF rate is 12.8% and allowing for 10% premature withdrawal, 221 patients are required to achieve 80% power to claim that the 12-month TLF rate is significantly less than 19%.
Subset analysis Several subset analyses were prespecified in the trial protocol, including diabetic patients, long lesions ≥20 mm, overlapped stents, left anterior descending coronary artery vs other vessels, and dual-lesion patients. The intention of these analyses is to assess consistency of results within subgroups; no formal statistical testing of hypotheses is used within these subgroups. All statistical analyses will be performed by Harvard Clinical Research Institute (Boston, MA) using PC SAS for Windows version 9.1 (SAS Institute, Cary, NC).
Study management The trial is entirely funded by Medtronic. The authors are solely responsible for the study design, conduct, and analyses as well as the drafting, editing, and final content of the paper. Harvard Clinical Research Institute is contracted by Medtronic to maintain the complete study database and perform all key analyses including the primary safety and efficacy analyses. The data safety monitoring board evaluates safety data, including serious and other significant adverse events, on an ongoing basis. An independent clinical events committee, composed of interventional and noninterventional cardiologists who are not participants in the trial, will adjudicate all clinical end point events for which the minimum data are available.
Discussion The RESOLUTE US Trial is a prospective, multicenter observational study with a patient-level historical control, designed to evaluate the safety and efficacy of the R-ZES in a US population as part of the global RESOLUTE clinical program. The R-ZES was designed to deliver zotarolimus to the coronary arterial tissue over a longer period of time than other currently approved DES. This may help minimize the risk of restenosis as well as the risk of occurrence of adverse clinical events. The trial design is unique in that it evaluates a novel device in a nonrandomized fashion. Although early trials of DES required randomization to ensure high-quality long-term comparative data, this trial benefits from the large body of detailed, pooled patient-level data on DES already acquired with consistent methodology. By using the same inclusion and exclusion criteria and using propensity score adjustment, it is possible to compare patient-level information between the 2 ZES. Propensity matching, for example, defining a subset of E-ZES
American Heart Journal Volume 161, Number 5
patients that match a subset of R-ZES patients on baseline characteristics (through propensity-matching techniques) and then performing the primary analyses on this matched sample would be desirable if baseline characteristic outliers were expected. However, because the experimental and the control data are derived from uniformly conducted studies with the same inclusion criteria, this design avoids the occurrence of such outliers and allows for the valid use of propensity score adjustment to resolve differences in the distribution of characteristics that could occur in known confounders and has the additional benefit of analyzing the entire population treated under the study protocol. Although a single-arm study may be valid and efficient, it depends on the assumption of no unmeasured confounders. Given this limitation, randomized studies remain the criterion standard for evaluation. The nature of the current study, where only a single aspect of the stent is altered in reference to the comparator (only the polymer differs substantially), and ability to ensure uniform data collection, definition, and adjudication allowed us to consider the single-arm study as a reliable method in this specific circumstance of new device evaluation. Despite inclusion of subjects with small vessels and longer lesions than previous studies designed to evaluate novel DES, the study population represents a restricted sample compared with what is observed in populationbased studies.16 Although this study design is necessary to allow the valid and direct comparison of efficacy to prior benchmarks, more inclusive studies evaluating the R-ZES will help to extend the observations beyond the population being evaluated for possible Food and Drug Administration–approved indication. These include prospective randomized and registry designs17,18 and will likely also include independent population-based studies, which will complement these data with increased power and generalizability. The design of this single study will not be sufficient to determine ST rates with complete certainty, given the low expected rates (b1.0% at 1 year and 2%-3% at 5 years) relative to the sample size of this single study.19,20 However, important aspects of this study design that will allow ongoing monitoring of stent thrombosis risks are (1) follow-up in both the RESOLUTE and ENDEAVOR data sets to 5 years and (2) uniform case definition and adjudication across pooled studies in both clinical trials programs.9,11 Combined analysis from this and other studies will be performed to increase the power to define ST rates at and beyond 1 year.
Disclosures Dr. Mauri receives institutional research support from Cordis, Medtronic, Abbott Vascular, Boston Scientific, Eli Lilly, Daiichi-Sankyo, Bristol Myers Squib, and sanofi-
Mauri et al 813
aventis. Drs. Leon and Yeung are on a scientific advisory board for Medtronic. Dr. Negoita is an employee of Medtronic. Drs. Keyes and Massaro are paid consultants for Harvard Clinical Research Institute and, through relationship with Harvard Clinical Research Institute, participated in and received salary for analysis on the R-ZES discussed in the paper.
References 1. Moses JW, et al. Sirolimus-eluting stents versus standard stents in patients with stenosis in a native coronary artery. N Engl J Med 2003; 349:1315-23. 2. Stone GW, et al. A polymer-based, paclitaxel-eluting stent in patients with coronary artery disease. N Engl J Med 2004;350:221-31. 3. Babapulle MN, et al. A hierarchical Bayesian meta-analysis of randomised clinical trials of drug-eluting stents. Lancet 2004;364: 583-91. 4. Stolker JM, et al. Predicting restenosis of drug-eluting stents placed in real-world clinical practice: derivation and validation of a risk model from the EVENT registry. Circ Cardiovasc Interv 2010;3: 327-34. 5. Stone GW, et al. Safety and efficacy of sirolimus- and paclitaxeleluting coronary stents. N Engl J Med 2007;356:998-1008. 6. Udipi K, et al. Development of a novel biocompatible polymer system for extended drug release in a next-generation drug-eluting stent. J Biomed Mater Res A 2008;85:1064-71. 7. Smith Jr SC, et al. ACC/AHA/SCAI 2005 Guideline Update for Percutaneous Coronary Intervention—summary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/SCAI Writing Committee to Update the 2001 Guidelines for Percutaneous Coronary Intervention). Circulation 2006;113:156-75. 8. Meredith IT, et al. Clinical and angiographic results with the nextgeneration resolute stent system: a prospective, multicenter, first-inhuman trial. JACC Cardiovasc Interv 2009;2:977-85. 9. Meredith IT, et al. Long-term clinical outcomes with the nextgeneration Resolute Stent System: a report of the two-year follow-up from the RESOLUTE clinical trial. EuroIntervention 2010;5:692-7. 10. Thygesen K, et al. Universal definition of myocardial infarction. Circulation 2007;116:2634-53. 11. Cutlip DE, et al. Clinical end points in coronary stent trials: a case for standardized definitions. Circulation 2007;115:2344-51. 12. Fajadet J, et al. Randomized, double-blind, multicenter study of the Endeavor zotarolimus-eluting phosphorylcholine-encapsulated stent for treatment of native coronary artery lesions: clinical and angiographic results of the ENDEAVOR II trial. Circulation 2006; 114:798-806. 13. Leon MB, et al. A randomized comparison of the ENDEAVOR zotarolimus-eluting stent versus the TAXUS paclitaxel-eluting stent in de novo native coronary lesions 12-month outcomes from the ENDEAVOR IV trial. J Am Coll Cardiol 2010;55:543-54. 14. Schultheiss HP, et al. Safety of direct stenting with the Endeavor stent: results of the Endeavor II continued access registry. EuroIntervention 2007;3:76-81. 15. Koch G, Amar I. Categorical data analysis. New York: Marcel Dekker; 1989. 16. Mauri L, et al. Long-term clinical outcomes after drug-eluting and bare-metal stenting in Massachusetts. Circulation 2008;118: 1817-27.
814 Mauri et al
17. Serruys PW, et al. Comparison of zotarolimus-eluting and everolimus-eluting coronary stents. N Engl J Med 2010;363:136-46. 18. Widimsky P, Belardi J, Neumann FJ. TCT-282: 1 year outcomes of patients with Resolute zotarolimus-eluting stent: results of the RESOLUTE International Registry. J Am Coll Cardiol 2010;56:B65.
American Heart Journal May 2011
19. Mauri L, et al. Stent thrombosis in randomized clinical trials of drugeluting stents. N Engl J Med 2007;356:1020-9. 20. Mauri L, et al. Long-term clinical outcomes with zotarolimus-eluting versus bare-metal coronary stents. JACC Cardiovasc Interv 2010;3: 1240-9.
Rationale and design of the clinical evaluation of the Resolute Zotarolimus-Eluting Coronary Stent System in the treatment of de novo lesions in native coronary arteries (the RESOLUTE US clinical trial) Laura Mauri, MD, MSc, a,b Martin B. Leon, MD, c Alan C. Yeung, MD, d Manuela Negoita, MD, e Michelle J. Keyes, PhD, f and Joseph M. Massaro, PhD f Boston, MA; New York, NY; and Stanford and Santa Rosa, CA
Background Drug-eluting stents (DES) are commonly used to treat obstructive coronary disease and avoid restenosis. Newer DES have been developed to improve effectiveness and safety. We describe a clinical trial to evaluate a DES with a novel polymer that may improve the antirestenosis effectiveness while maintaining the safety standards of currently Food and Drug Administration–approved DES. Methods The RESOLUTE US Trial is a multicenter, nonrandomized trial prospectively designed to compare the Resolute zotarolimus-eluting stent (R-ZES) to the Food and Drug Administration–approved Endeavor ZES using patient-level historical control data, adjusting for baseline covariates through propensity score. The stents differ primarily in the polymer, which, in the R-ZES, is designed to elute zotarolimus over a longer period. The study will enroll up to 1,574 patients with ischemic heart disease due to de novo native coronary lesions suitable for 1- or 2-vessel treatment with stents from 2.25 to 4.0 mm in diameter. The primary end point is target lesion failure at 12 months postprocedure, defined as the composite of cardiac death, targetvessel myocardial infarction (MI), and clinically driven target lesion revascularization by percutaneous or surgical methods. Secondary end points include device, lesion and procedural success, death, MI, cardiac death and MI, composites of these clinical events, and stent thrombosis at each follow-up assessment up to 5 years postprocedure. Conclusions The RESOLUTE US Trial (ClinicalTrials.gov #NCT00726453) is a prospective, multicenter, observational study with a patient-level historical control designed to assess the safety and efficacy of the R-ZES for the treatment of de novo lesions in native coronary arteries. (Am Heart J 2011;161:807-14.)
Background Drug-eluting stents (DES) that allow local delivery of the antiproliferative drugs from the surface of a stent have been shown to dramatically reduce vascular neointimal hyperplasia and subsequent renarrowing of the coronary stent and the requirement for subsequent procedures.1-3 However, repeat procedures within the first year are still required in about 5% of DES patients and in a greater proportion of higher risk patients or lesions.4 Further-
From the aBrigham and Women's Hospital, Boston, MA, bHarvard Medical School, Boston, MA, cColumbia University Medical Center/New York Presbyterian Hospital, New York, NY, dStanford University School of Medicine, Stanford, CA, eMedtronic, Santa Rosa, CA, and fHarvard Clinical Research Institute, Boston, MA. Submitted December 30, 2010; accepted March 9, 2011. Reprint requests: Laura Mauri, MD, MSc, Division of Cardiovascular Medicine, Department of Medicine, Brigham and Women's Hospital, 75 Francis Street, Boston, MA 02115. E-mail: [email protected] 0002-8703/$ - see front matter © 2011, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2011.03.015
more, the long-term safety of DES remains a matter of clinical investigation.5 Currently approved DES use various different medications, polymers, and stent designs. The Resolute zotarolimus-eluting coronary stent (R-ZES; Medtronic, Santa Rosa, CA) uses the same cobalt chromium stent platform and drug (zotarolimus) as the Food and Drug Administration–approved Endeavor zotarolimus-eluting coronary stent (E-ZES; Medtronic). The primary difference between these 2 stents is that the R-ZES uses a new polymer that allows for the extended release of zotarolimus over a period of approximately 180 days.6
Study design The primary objective of the RESOLUTE US Trial is to assess the safety and efficacy of the R-ZES for the treatment of de novo lesions in native coronary arteries with a reference vessel diameter of 2.25 to 4.2 mm and up to 35 mm in length. The RESOLUTE US Trial (ClinicalTrials.gov #NCT00726453) is a prospective,
American Heart Journal May 2011
808 Mauri et al
Figure 1 Reference Vessel Diameter Lesion length ≤27 mm (≤35 mm for 38 mm Length Sub-study)
Study Designation Per stent size
Sites Participating Two lesion treatment permitted during index procedure (1 lesion per vessel up to 2 vessels) Enrollment Enrollment Breakdown by Stent Size
3.0 mm – 4.2 mm
2.25 mm – 4.2 mm
38 mm Length Group
*
2.25 mm – 3.5 mm Angio/IVUS Cohort
4.0 mm Stent Group
All sites
Limited number of sites
All sites
Limited number of sites
Yes
Yes
Yes
Yes
100
58
110 - 175
2.25 mm – 3.5 mm Clinical Cohort
1241 129 (2.25 mm) 1112 (2.5 mm – 3.5 mm)
20 (2.25 mm) 80 (2.5 mm – 3.5 mm)
58 (4.0 mm)
110 - 175 (38 mm)
Angiographic Follow-up
No
Eight (8) Month
Eight (8) Month
No
IVUS Follow-up
No
Eight (8) Month
No
No
Twelve (12) Month Patient Contact
Nine (9) Month Clinic Contact
Nine (9) Month Clinic Contact
Twelve (12) Month Patient Contact
Primary Clinical Follow-up
up to 1574
RESOLUTE US Clinical Trial Design.
multicenter, observational study with a patient-level historical control (Figure 1). The study is designed to reproduce the inclusion and exclusion characteristics of the E-ZES clinical trials, with uniform definition and adjudication, to allow direct comparison of patient-level historical clinical data for the primary comparison.
Population Patients will be enrolled at up to 116 investigational sites in the United States, either with prospective, clinical follow-up only (clinical cohort, n = 1,351-1,416) or with protocol-specified angiographic follow-up (angiographic cohort, n = 158). Table I lists the general and angiographic inclusion and exclusion criteria. A patient's inclusion in a given subanalysis cohort is dependent on the size (diameter) or length of the stent(s) the patient receives (Figure 1). General inclusion and exclusion criteria are similar for all of the cohorts (Table I). Clinical cohort The clinical cohort includes patients enrolled in the 2.25- to 3.5-mm main clinical cohort and 38-mm length stent group. The main clinical cohort will enroll patients with ischemic heart disease due to de novo native coronary lesions between 2.25 and 3.5 mm and lesion lengths of ≤27 mm, amenable to percutaneous treatment with a stent. Patients may receive treatment of up to 2 lesions, provided that the lesions are located in separate target vessels and treated during a single index procedure. Of the 1,241 patients planned to be enrolled in the main study, 1,112 will receive only 2.5- to 3.5-mm stents, and
129 will receive at least one 2.25-mm stent. The 2.25-mm stent group will be analyzed separately from the 2.5- to 3.5-mm stent clinical cohort. For the 38-mm length substudy, between 110 and 175 patients with lesion lengths ≤35 mm will be enrolled.
Angiographic cohort The angiographic cohort includes 100 patients to be enrolled in the 2.25- to 3.5-mm stent angiography and intravascular ultrasound (angio/IVUS) cohort, of which 80 will receive only 2.5- to 3.5-mm stents and 20 will receive at least one 2.25-mm stent, and 58 patients are to be enrolled in the 4.0-mm stent group with at least 1 lesion amenable to treatment with a 4.0-mm stent. Medications All patients will receive aspirin (a minimum of 75 mg daily) and a loading dose of ≥300 mg clopidogrel within 24 hours before the procedure or within 30 minutes postprocedure. Bivalrudin or heparin, to maintain an activated clotting time ≥250 seconds, will be administered at the time of the procedure. Administration of 50 to 200 μg of intracoronary nitroglycerin is required before stenting and before postintervention angiograms. Postprocedure, patients will receive a minimum of 75 mg of aspirin daily indefinitely and 75 mg daily of clopidogrel for a minimum of 6 months in all patients and up to 12 months in patients who are not at high risk of bleeding by investigator determination, per American College of Cardiology/American Heart Association/Society for Cardiac Angiography and Interventions guidelines.7 Dual-
American Heart Journal Volume 161, Number 5
Mauri et al 809
Table I. RESOLUTE US Inclusion and Exclusion Criteria General inclusion criteria (subjects must meet all of the following criteria for eligibility for treatment in the trial) • N18 y of age • Acceptable candidate for PCI stenting and emergent CABG surgery • Clinical evidence of ischemic heart disease, stable or unstable angina, silent ischemia, and/or a positive functional study • Negative pregnancy test within 7 d before the trial procedure if female of childbearing potential • Provided written informed consent as approved by the institutional review board/ethics committee of the respective investigational site • Agrees to comply with specified follow-up evaluations and to return to the same investigational site where the procedure was performed Angiographic inclusion criteria (the subject and each target lesion/vessel must meet all of the following angiographic criteria for the patient to be considered for inclusion in the trial) • Single target lesion, or 2 target lesions located in separate target vessels • Target lesion(s) is(are) de novo lesion(s) in native coronary artery(ies) • Target lesion(s) is(are) ≤27 mm in length (or ≤35 mm for a lesion to be treated with a 38-mm length stent) • Target lesion(s) is(are) a stenosis of ≥50% and b100% • Target-vessel(s) reference vessel diameter 2.25-4.2 mm (or 3.0-4.2 mm for it to be treated with a 38-mm length stent) • Target-vessel(s) thrombolysis in MI flow ≥2 General exclusion criteria (subjects will be excluded from the trial if any of the following criteria are met) • Known hypersensitivity or contraindication to aspirin, heparin and bivalirudin, clopidogrel and ticlopidine, cobalt, nickel, chromium, molybdenum, polymer coatings (eg, BioLinx), or a sensitivity to contrast media, which cannot be adequately premedicated • History of an allergic reaction or significant sensitivity to drugs such as zotarolimus, rapamycin, tacrolimus, everolimus, or any other analog or derivative • Platelet count b100,000 or N700,000 cells/mm3, or a white blood cell count b3,000 cells/mm3 within 7 d before index procedure • Serum creatinine level N2.5 mg/dL within 7 d before index procedure • Evidence of an acute MI within 72 h of the intended trial procedure: (a) Q wave MI or (b) Any elevation of CK-MB isoenzyme above the laboratory upper limit of normal and has not returned to normal at the time of the index procedure • Previous PCI of the target vessel(s) within 9 mo before the procedure • Planned PCI of any vessel within 30 d postindex procedure and/or planned PCI of the target vessel(s) within 12 mo postprocedure • During the index procedure, the target lesion(s) requires treatment with a device other than PTCA before stent placement (including, but not limited to, cutting balloon, atherectomy, laser, thrombectomy, etc) • History of a stroke or transient ischemic attack within the prior 6 mo • Active peptic ulcer or upper gastrointestinal bleeding within the prior 6 mo • History of bleeding diathesis or coagulopathy or will refuse blood transfusions • Concurrent medical condition with a life expectancy of less than 12 mo • Any previous treatment of the target vessel(s) for restenosis, including brachytherapy • Currently participating in an investigational drug or another device trial that has not completed the primary end point or that clinically interferes with the current trial end points, or requires coronary angiography, IVUS, or other coronary artery imaging procedures • Documented left ventricular ejection fraction b30% at the most recent evaluation • Inability to comply with the required trial antiplatelet regimen Angiographic exclusion criteria (the subject will be excluded from the trial if any of the following criteria are met (for patients with 2 target lesions, both target lesions/vessels must not meet any of the criteria below)) • Target lesion(s) is(are) located in native vessel(s) distal to anastomosis with a bypass graft (including but not limited to saphenous vein graft or a left/right internal mammary artery) with N40% diameter stenosis anywhere within the graft • Previous stenting in the target vessel(s) unless the following conditions are met: (a) It has been at least 9 mo since the previous stenting and (b) Target lesion(s) is(are) at least 15 mm away from the previously placed stent • Target vessel(s) has/have other lesions with N40% diameter stenosis based on visual estimate or online quantitative coronary angiography • The target vessel(s) has(have) evidence of thrombus • The target vessel(s) is(are) excessively tortuous (2 bends N90° to reach the target lesion) • The target lesion(s): (a) Location is aorto-ostial, an unprotected left main lesion, or within 5 mm of the origin of the left anterior descending or left circumflex. (b) It involves a side branch N2.0 mm in diameter. (c) It is at or distal to a N45° bend in the vessel. (d) It is severely calcified. • Unprotected left main coronary artery disease (an obstruction N50% in the left main coronary artery)
antiplatelet therapy use will be assessed at 1, 6, and 12 months and annually through 5-year follow-up.
Device description The R-ZES has been previously described in detail elsewhere.8,9 The study stent is available in diameters of 2.25, 2.5, 2.75, 3.0, 3.5, and 4.0 mm and lengths of 12, 18, 24, 30, and 38 mm for each diameter. Resolute zotarolimus-eluting stent 8 mm long with diameters of
2.25, 2.5, and 2.75 mm and 9 mm long with diameters of 3.0, 3.5, and 4.0 mm are available for use only as a secondary stent in cases of insufficient lesion coverage or procedure-related edge dissections.
Treatment/study procedures All patients must provide institutional review board– approved informed consent before study enrollment. After baseline angiography, target lesions must be
American Heart Journal May 2011
810 Mauri et al
predilated with standard percutaneous transluminal balloon angioplasty techniques. Postdilatation of the stent segments is at the discretion of the operator. Trial stents ≥3mm longer than the lesion length will be selected to treat each target lesions. For patients with planned treatment of 2 lesions, the first lesion must be treated successfully, and the patient must be clinically stable before treatment of the second lesion is attempted. Successful treatment of the first lesion requires that a b10% residual diameter stenosis result is achieved (visual assessment), that thrombolysis in myocardial infarction (MI) 3 flow is present posttreatment, and that there is no evidence of dissection, thrombus, or distal embolization at the first study lesion site posttreatment. Final angiography will be performed upon procedure completion. Treatment failure and patients not treated with the study stent postenrollment will be followed for safety purposes and included in the intention-to-treat (ITT) analyses described below. Predischarge, creatine kinase (CK) and CK-MB isoenzymes will be measured at 6 to 8, 12 to 16, and 20 to 24 hours postprocedure. Although not required for comparison with the historical data, troponin levels will be sampled to allow for comparison with the Academic Research Consortium (ARC)/ Universal MI definition.10,11
Follow-up schedule Demographic, clinical, and procedural information at the time of enrollment will be captured as well as subsequent clinical end points, serious adverse events, and concomitant medications. Clinical follow-up assessments at the same investigational site where the index procedure was performed will be performed for all patients at 30 days and at 6, 9, 12, and 18 months postprocedure and annually through 5 years postprocedure. In addition, clinical, angiographic, and IVUS assessment will be performed at 8 months in the 2.25- to 3.5-mm angio/IVUS cohort and the 4.0-mm stent group. Figure 2 represents a flowchart of follow-up assessments.
Figure 2 All Patients Thirty (30) Day Clinic Visit Six (6) Month Patient Contact
Patients in 4.0 mm Stent Group
Patients in 2.25 mm – 3.5 mm Clinical Cohort and 38 mm Length Group
Patients in 2.25 mm – 3.5 mm Angio/IVUS Cohort Eight (8) Month Clinic Visit with Angiography and IVUS
Eight (8) Month Clinic Visit with Angiography All Patients
Nine (9) Month Clinic Visit
Twelve (12) Month Patient Contact
Eighteen (18) Month Patient Contact
Two-Five (2-5) Year Patient Contact
RESOLUTE US Clinical Trial patient follow-up flowchart.
End points and definitions
For the angio/IVUS cohort and 4.0-mm stent group, the primary end point is 8-month late lumen loss (LLL), defined as the difference between the postprocedure minimal lumen diameter and the follow-up angiography minimal lumen diameter as measured by quantitative coronary angiography. A full listing of the study end points can be found in Table II.
For the main clinical cohort, the 2.25-mm stent group, and the 38-mm length group, the primary end point is target lesion failure (TLF) at 12 months, defined as cardiac death, target-vessel MI (Q wave and non–Q wave), or clinically-driven target lesion revascularization (TLR) by percutaneous or surgical methods. Secondary end points include death, MI, cardiac death and MI, major adverse cardiac events, target-vessel failure (TVF) composite and individual components, and stent thrombosis (ST) at each follow-up assessment. Definitions are listed below.
Definitions For the RESOLUTE US Clinical Trial, all deaths are considered cardiac unless an unequivocal noncardiac cause can be established. Major adverse cardiac event is defined as death, MI (Q wave and non–Q wave), emergent coronary bypass surgery, or clinically-driven TLR by percutaneous or surgical methods. Target-vessel failure components include cardiac death, target-vessel MI, and clinically-driven target vessel revascularization (TVR).
American Heart Journal Volume 161, Number 5
Table II. RESOLUTE US Clinical, angiographic, and IVUS end points Clinical end points: (at 30 d and at 6, 9, 12, and 18 mo postprocedure and annually thereafter through year 5) • Device success • Lesion success • Procedure success • Death • MI, Q wave, and non–Q wave • Cardiac death and MI reported at 30 d and at 6, 9, 12, and 18 mo postprocedure and annually thereafter through year 5 • Major adverse cardiac events composite end point (death, MI, Q wave, and non–Q wave), emergent CABG MI, or clinically driven repeat TLR by percutaneous or surgical methods) and each individual component • TLF composite end point (cardiac death, target-vessel MI, or clinically driven TLR) and each individual component • TVF composite end point (cardiac death, target-vessel MI, or clinically driven target-vessel revascularization by percutaneous or surgical methods) and each individual component • Stent thrombosis (Medtronic historic protocol definitions and ARC definitions) Angiographic end points (at 8 mo) • In-segment LLL at 8 mo • In-stent and in-segment binary angiographic restenosis rate (defined as N50% diameter stenosis) at 8 mo postprocedure • In-stent and in-segment minimum lumen diameter at 8 mo postprocedure • In-stent and in-segment percent diameter stenosis at 8 mo postprocedure IVUS end points (at 8 mo) • Rates of incomplete stent apposition at 8 mo postprocedure, categorized as persistent or late • Neointimal hyperplastic volume and percent volume obstruction (%VO) at 8 mo postprocedure
Q wave MI requires a history of chest pain or other acute symptoms consistent with myocardial ischemia together with new pathological Q waves in 2 or more contiguous electrocardiogram leads or new pathological Q waves and elevated cardiac enzymes. Non–Q wave MI is defined by elevated CK two or more times the upper limit of normal with the presence of an elevated CK-MB in the absence of new pathological Q waves. TLR is defined as clinicallydriven repeat percutaneous coronary intervention (PCI) of the target lesion or coronary artery bypass grafting (CABG) including the target vessel. TVR is defined as clinically-driven repeat PCI or CABG of the target vessel. ST is defined as in the prior E-ZES studies, and all ST events will be adjudicated according to ARC criteria.11
Statistical analysis Main clinical cohort analysis The 2.5- to 3.5-mm stent clinical cohort analysis is designed to assess the noninferiority of the 2.5-, 2.75-, 3.0-, and 3.5-mm diameter R-ZES compared with patientlevel historical control data from the ENDEAVOR Clinical Program. Historical E-ZES patients with diameters 2.5- to 3.5 mm who were part of a clinical follow-up cohort (no
Mauri et al 811
planned angiographic follow-up) were identified from those E-ZES patients with study clinical follow-up only (ENDEAVOR II, ENDEAVOR IV, ENDEAVOR PK, and ENDEAVOR II Continued Access trials).12-14 The 12month TLF rate for this group of 1,076 patients was 6.5% (calculated from Medtronic data on file). Because the historical control arm only includes single-lesion patients, the 2.5- to 3.5-mm stent clinical cohort will be analyzed for comparing the single-lesion subset of the R-ZES arm and historical control arm, using a noninferiority margin (delta) of 3.3%. A total of 800 analyzable patients will yield 87% power to assess the noninferiority of the R-ZES compared with EZES using a 1-sided .05 level of significance. Up to 890 single-lesion patients will be enrolled to account for loss to follow-up (expected to be less than 10%). Anticipating a rate of dual-vessel treatment of 20%, a total of 1,112 patients will be enrolled with either single- or dual-vessel treatment. The primary analysis cohort will be the ITT main clinical cohort defined as patients who met the study entry criteria, signed the written informed consent, and were enrolled in the trial. In addition, as a secondary analysis, the noninferiority assessment will be performed on the per-protocol analysis cohort, which is defined as the ITT sample excluding patients who are deregistered or do not meet certain key entry criteria. Because R-ZES and the E-ZES patients are from different studies, there is a possibility of treatment imbalance on clinically relevant baseline characteristics. If there are baseline imbalances between the 2 groups (at a 2-sided .05 level of significance), a propensity score for group (Resolute, Endeavor) membership will be calculated using logistic regression and then categorized into quintiles based on this propensity score. To assess noninferiority of R-ZES to E-ZES at the 1-sided .05 level of significance, a 1-sided upper 95% CI of the treatment difference (Resolute minus Endeavor) in TLF rates adjusted for propensity score quintile will be calculated according to Koch and Amar15 and compared with the noninferiority margin of 3.3%. This noninferiority analysis is the primary analysis of the study. If noninferiority is met at the 1-sided .05 level of significance, only then will analysis on prespecified subsets proceed as described below.
2.25-mm stent group analysis The 2.25-mm stent group is designed to assess the effectiveness and safety of the 2.25-mm diameter R-ZES compared with a performance goal derived from a logistic regression model based on E-ZES patients. The analysis population consists of all 129 patients receiving a 2.25-mm stent in the clinical cohort and 20 patients in the angio/IVUS cohort (149 total patients). Based on historical E-ZES data and the planned proportion of subjects with scheduled angiographic follow-up, the expected
American Heart Journal May 2011
812 Mauri et al
TLF rate is 12.87% at 12 months (calculated from Medtronic data on file). The derived performance goal for TLF at 12 months was set at 20%. If the 12-month TLF rate of the 2.25-mm stent group of the pooled clinical and angio/IVUS cohorts is shown to be significantly less than 20% at a 1-sided .05 level of significance, then the analysis of this group will be considered to have met its primary objective. At a 1-sided .05 level of significance, a sample size of 140 evaluable subjects yields 80% power to claim the true TLF rate is b20% under the assumption that the true TLF rate is in actuality 12.87%; 149 patients will be enrolled to account for up to 6% premature withdrawal by the 12-month analysis time point. Generally, with regard to clinical outcomes analyses, in patients receiving treatment of both lesions with stents of the same size group (2.25, 4.0, or 38 mm), the lesion to be included in the primary analysis, the “analysis lesion,” will be randomly selected at the time of analysis. For patients receiving treatment of 2 lesions, 1 with a 2.25-mm and 1 with a 4.0-mm stent, the primary analysis will include both lesions, and the patient will be counted twice, once in each study. A secondary analysis will be conducted for all target lesions.
Angiographic cohort analysis The 2.25- to 3.5-mm angio/IVUS cohort analysis will compare mean 8-month in-stent LLL in the RESOLUTE 2.25- to 3.5-mm imaging cohort, compared with mean 8month in-stent LLL of the historical control subset of the ENDEAVOR II trial, using a noninferiority margin (delta) of 0.16 mm. Assuming a true equivalence of mean LLL between the 2 cohorts and with a common standard deviation of 0.46, an approximate 20% nonevaluable rate, and 264 subjects in the historical control, 100 angio/IVUS patients (80 evaluable) achieves 86% power to detect noninferiority at a 1-sided significance level of .05. 4.0-mm stent group analysis The 4.0-mm stent group analysis will compare the mean 8-month in-segment LLL of the 4.0-mm diameter RZES to that of a historical control derived from the Medtronic Driver stent data, for superiority at a 1-sided .05 level of significance. Assuming at least 0.15-mm mean LLL benefit of R-ZES over Driver, standard deviations of 0.62 and 0.46 in the Driver and R-ZES arms and 894 Driver patients, 46 R-ZES patients yield 90% power to claim superiority of R-ZES to Driver; 58 RZES patients will be enrolled to account for an expected 20% loss to follow-up. 38-mm length group analysis The 38-mm length stent group analysis will compare the 12-month TLF rate to a performance goal of 19%. Additional data on the 38-mm R-ZES will be used from other RESOLUTE worldwide clinical trials in this com-
parison. Assuming the true 12-month TLF rate is 12.8% and allowing for 10% premature withdrawal, 221 patients are required to achieve 80% power to claim that the 12-month TLF rate is significantly less than 19%.
Subset analysis Several subset analyses were prespecified in the trial protocol, including diabetic patients, long lesions ≥20 mm, overlapped stents, left anterior descending coronary artery vs other vessels, and dual-lesion patients. The intention of these analyses is to assess consistency of results within subgroups; no formal statistical testing of hypotheses is used within these subgroups. All statistical analyses will be performed by Harvard Clinical Research Institute (Boston, MA) using PC SAS for Windows version 9.1 (SAS Institute, Cary, NC).
Study management The trial is entirely funded by Medtronic. The authors are solely responsible for the study design, conduct, and analyses as well as the drafting, editing, and final content of the paper. Harvard Clinical Research Institute is contracted by Medtronic to maintain the complete study database and perform all key analyses including the primary safety and efficacy analyses. The data safety monitoring board evaluates safety data, including serious and other significant adverse events, on an ongoing basis. An independent clinical events committee, composed of interventional and noninterventional cardiologists who are not participants in the trial, will adjudicate all clinical end point events for which the minimum data are available.
Discussion The RESOLUTE US Trial is a prospective, multicenter observational study with a patient-level historical control, designed to evaluate the safety and efficacy of the R-ZES in a US population as part of the global RESOLUTE clinical program. The R-ZES was designed to deliver zotarolimus to the coronary arterial tissue over a longer period of time than other currently approved DES. This may help minimize the risk of restenosis as well as the risk of occurrence of adverse clinical events. The trial design is unique in that it evaluates a novel device in a nonrandomized fashion. Although early trials of DES required randomization to ensure high-quality long-term comparative data, this trial benefits from the large body of detailed, pooled patient-level data on DES already acquired with consistent methodology. By using the same inclusion and exclusion criteria and using propensity score adjustment, it is possible to compare patient-level information between the 2 ZES. Propensity matching, for example, defining a subset of E-ZES
American Heart Journal Volume 161, Number 5
patients that match a subset of R-ZES patients on baseline characteristics (through propensity-matching techniques) and then performing the primary analyses on this matched sample would be desirable if baseline characteristic outliers were expected. However, because the experimental and the control data are derived from uniformly conducted studies with the same inclusion criteria, this design avoids the occurrence of such outliers and allows for the valid use of propensity score adjustment to resolve differences in the distribution of characteristics that could occur in known confounders and has the additional benefit of analyzing the entire population treated under the study protocol. Although a single-arm study may be valid and efficient, it depends on the assumption of no unmeasured confounders. Given this limitation, randomized studies remain the criterion standard for evaluation. The nature of the current study, where only a single aspect of the stent is altered in reference to the comparator (only the polymer differs substantially), and ability to ensure uniform data collection, definition, and adjudication allowed us to consider the single-arm study as a reliable method in this specific circumstance of new device evaluation. Despite inclusion of subjects with small vessels and longer lesions than previous studies designed to evaluate novel DES, the study population represents a restricted sample compared with what is observed in populationbased studies.16 Although this study design is necessary to allow the valid and direct comparison of efficacy to prior benchmarks, more inclusive studies evaluating the R-ZES will help to extend the observations beyond the population being evaluated for possible Food and Drug Administration–approved indication. These include prospective randomized and registry designs17,18 and will likely also include independent population-based studies, which will complement these data with increased power and generalizability. The design of this single study will not be sufficient to determine ST rates with complete certainty, given the low expected rates (b1.0% at 1 year and 2%-3% at 5 years) relative to the sample size of this single study.19,20 However, important aspects of this study design that will allow ongoing monitoring of stent thrombosis risks are (1) follow-up in both the RESOLUTE and ENDEAVOR data sets to 5 years and (2) uniform case definition and adjudication across pooled studies in both clinical trials programs.9,11 Combined analysis from this and other studies will be performed to increase the power to define ST rates at and beyond 1 year.
Disclosures Dr. Mauri receives institutional research support from Cordis, Medtronic, Abbott Vascular, Boston Scientific, Eli Lilly, Daiichi-Sankyo, Bristol Myers Squib, and sanofi-
Mauri et al 813
aventis. Drs. Leon and Yeung are on a scientific advisory board for Medtronic. Dr. Negoita is an employee of Medtronic. Drs. Keyes and Massaro are paid consultants for Harvard Clinical Research Institute and, through relationship with Harvard Clinical Research Institute, participated in and received salary for analysis on the R-ZES discussed in the paper.
References 1. Moses JW, et al. Sirolimus-eluting stents versus standard stents in patients with stenosis in a native coronary artery. N Engl J Med 2003; 349:1315-23. 2. Stone GW, et al. A polymer-based, paclitaxel-eluting stent in patients with coronary artery disease. N Engl J Med 2004;350:221-31. 3. Babapulle MN, et al. A hierarchical Bayesian meta-analysis of randomised clinical trials of drug-eluting stents. Lancet 2004;364: 583-91. 4. Stolker JM, et al. Predicting restenosis of drug-eluting stents placed in real-world clinical practice: derivation and validation of a risk model from the EVENT registry. Circ Cardiovasc Interv 2010;3: 327-34. 5. Stone GW, et al. Safety and efficacy of sirolimus- and paclitaxeleluting coronary stents. N Engl J Med 2007;356:998-1008. 6. Udipi K, et al. Development of a novel biocompatible polymer system for extended drug release in a next-generation drug-eluting stent. J Biomed Mater Res A 2008;85:1064-71. 7. Smith Jr SC, et al. ACC/AHA/SCAI 2005 Guideline Update for Percutaneous Coronary Intervention—summary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/SCAI Writing Committee to Update the 2001 Guidelines for Percutaneous Coronary Intervention). Circulation 2006;113:156-75. 8. Meredith IT, et al. Clinical and angiographic results with the nextgeneration resolute stent system: a prospective, multicenter, first-inhuman trial. JACC Cardiovasc Interv 2009;2:977-85. 9. Meredith IT, et al. Long-term clinical outcomes with the nextgeneration Resolute Stent System: a report of the two-year follow-up from the RESOLUTE clinical trial. EuroIntervention 2010;5:692-7. 10. Thygesen K, et al. Universal definition of myocardial infarction. Circulation 2007;116:2634-53. 11. Cutlip DE, et al. Clinical end points in coronary stent trials: a case for standardized definitions. Circulation 2007;115:2344-51. 12. Fajadet J, et al. Randomized, double-blind, multicenter study of the Endeavor zotarolimus-eluting phosphorylcholine-encapsulated stent for treatment of native coronary artery lesions: clinical and angiographic results of the ENDEAVOR II trial. Circulation 2006; 114:798-806. 13. Leon MB, et al. A randomized comparison of the ENDEAVOR zotarolimus-eluting stent versus the TAXUS paclitaxel-eluting stent in de novo native coronary lesions 12-month outcomes from the ENDEAVOR IV trial. J Am Coll Cardiol 2010;55:543-54. 14. Schultheiss HP, et al. Safety of direct stenting with the Endeavor stent: results of the Endeavor II continued access registry. EuroIntervention 2007;3:76-81. 15. Koch G, Amar I. Categorical data analysis. New York: Marcel Dekker; 1989. 16. Mauri L, et al. Long-term clinical outcomes after drug-eluting and bare-metal stenting in Massachusetts. Circulation 2008;118: 1817-27.
814 Mauri et al
17. Serruys PW, et al. Comparison of zotarolimus-eluting and everolimus-eluting coronary stents. N Engl J Med 2010;363:136-46. 18. Widimsky P, Belardi J, Neumann FJ. TCT-282: 1 year outcomes of patients with Resolute zotarolimus-eluting stent: results of the RESOLUTE International Registry. J Am Coll Cardiol 2010;56:B65.
American Heart Journal May 2011
19. Mauri L, et al. Stent thrombosis in randomized clinical trials of drugeluting stents. N Engl J Med 2007;356:1020-9. 20. Mauri L, et al. Long-term clinical outcomes with zotarolimus-eluting versus bare-metal coronary stents. JACC Cardiovasc Interv 2010;3: 1240-9.