eAppendix. Centers Participating in the CHAMPION Trials
eTable 1. Baseline Characteristics in the Propensity Score–Matched Cohorts of the CHAMPION PCI Trial
eTable 2. Efficacy and Safety Endpoints at 48 Hours After Randomization in Original and Propensity Score–Matched Cohorts of the CHAMPION PCI Trial
Customize your JAMA Network experience by selecting one or more topics from the list below.
Vaduganathan M, Harrington RA, Stone GW, et al. Evaluation of Ischemic and Bleeding Risks Associated With 2 Parenteral Antiplatelet Strategies Comparing Cangrelor With Glycoprotein IIb/IIIa Inhibitors: An Exploratory Analysis From the CHAMPION Trials. JAMA Cardiol. 2017;2(2):127–135. doi:10.1001/jamacardio.2016.4556
What are the relative ischemic and bleeding risks associated with glycoprotein IIb/IIIa inhibitors and more potent P2Y12 antagonists, such as cangrelor, in patients undergoing percutaneous coronary intervention?
In this exploratory analysis of pooled patient-level data from the 3 Cangrelor vs Standard Therapy to Achieve Optimal Management of Platelet Inhibition (CHAMPION) trials including 12 140 patients, cangrelor alone was associated with similar ischemic risk and lower risk-adjusted severe bleeding risk compared with clopidogrel given with glycoprotein IIb/IIIa inhibitors.
Cangrelor may offer an alternative potent, parenteral strategy to glycoprotein IIb/IIIa inhibitors with a favorable bleeding profile in percutaneous coronary intervention.
In the context of contemporary pharmacotherapy, optimal antiplatelet management with percutaneous coronary intervention (PCI) has not been well established.
To compare the ischemic and bleeding risks associated with glycoprotein IIb/IIIa inhibitors (GPIs) and a potent P2Y12 antagonist, cangrelor, in patients undergoing PCI.
Design, Setting, and Participants
An exploratory analysis of pooled patient-level data from the 3 phase 3 Cangrelor vs Standard Therapy to Achieve Optimal Management of Platelet Inhibition (CHAMPION PCI, CHAMPION PLATFORM, and CHAMPION PHOENIX) trials of patients undergoing elective or nonelective PCI. The participants included 10 929 patients assigned to cangrelor but not receiving GPIs (cangrelor alone) and 1211 patients assigned to clopidogrel (or placebo) and receiving routine GPIs (clopidogrel-GPI). Patients requiring bailout or rescue GPI therapy were excluded. To account for risk imbalances, 1:1 propensity score matching based on 16 baseline clinical variables yielded 1021 unique matched pairs. The present study’s data analysis was conducted from October 28, 2015, to August 6, 2016.
Main Outcomes and Measures
The primary efficacy end point was the composite of all-cause mortality, myocardial infarction, ischemia-driven revascularization, or stent thrombosis at 48 hours. Safety was assessed by 3 validated bleeding scales (Global Use of Strategies to Open Occluded Coronary Arteries [GUSTO], Thrombolysis in Myocardial Infarction [TIMI], and Acute Catheterization and Urgent Intervention Triage) and requirement for blood transfusions.
Of the 12 140 patients included in the analysis, 8779 were men (72.3%), and the mean (SD) age was 63.2 (11.3) years. Patients in the clopidogrel-GPI group were more likely to be male (75.6% vs 71.9%), younger (median, 60 [range, 23-91] years vs 64 [range, 26-95] years), enrolled from the United States (77.9% vs 40.0%), and present with an acute coronary syndrome, but they had lower comorbid disease burden and were less likely to receive bivalirudin (8.8% vs 27.3%). In the matched cohorts, the rates of the primary efficacy end point were not significantly different between the cangrelor alone and clopidogrel-GPI groups (2.6% vs 3.3%; odds ratio [OR], 0.79; 95% CI, 0.48-1.32). There was a nonsignificant trend toward lower rates of GUSTO-defined severe/life-threatening bleeding with cangrelor alone compared with clopidogrel-GPI (0.3% vs 0.7%; OR, 0.43; 95% CI, 0.11-1.66). Rates of TIMI-defined major or minor bleeding were significantly lower in patients treated with cangrelor alone (0.7% vs 2.4%; OR, 0.29; 95% CI, 0.13-0.68).
Conclusions and Relevance
Based on a pooled analysis from the 3 phase 3 CHAMPION trials, cangrelor alone was associated with similar ischemic risk and lower risk-adjusted bleeding risk compared with clopidogrel-GPIs.
clinicaltrials.gov Identifiers: NCT00305162, NCT00385138, and NCT01156571
Glycoprotein IIb/IIIa inhibitors (GPIs) robustly and consistently reduce major adverse cardiovascular events, driven primarily by periprocedural myocardial infarction (MI) compared with heparin alone in patients undergoing elective1 or urgent percutaneous coronary intervention (PCI).2 Unfortunately, these ischemic benefits conferred by routine use of GPIs are partially offset by a significant excess in major bleeding.3 Interval advancements in stenting, increased upfront utilization of oral P2Y12 receptor antagonists, and introduction of more potent antiplatelet and anticoagulants (including bivalirudin) have attenuated the overall benefits of GPIs observed in prior studies.4-6 Furthermore, selective GPI utilization in high-risk patients, coupled with other contemporary means of reducing bleeding risk (including greater use of transradial access), have improved the safety and tolerability of these agents.7 As such, a reappraisal of the role of GPIs in the background of currently available pharmacotherapies is required.8
Cangrelor is a novel, fast-onset, fast-offset, potent, intravenous P2Y12 antagonist.9-12 The Cangrelor vs Standard Therapy to Achieve Optimal Management of Platelet Inhibition (CHAMPION) program13,14 collectively demonstrated that cangrelor reduces ischemic complications up to 30 days following PCI compared with clopidogrel or placebo without increasing Global Use of Strategies to Open Occluded Coronary Arteries (GUSTO)–defined severe/life-threatening bleeding15 or requirement for blood transfusions. These data formed the basis for the recent approval of cangrelor for use in elective or urgent PCI in the United States and Europe. To our knowledge, no study to date has evaluated the comparative efficacy and safety of more potent P2Y12 inhibitors, such as cangrelor, with GPIs. In this exploratory pooled analysis of individual patient-level data from the 3 phase 3 CHAMPION trials (CHAMPION PHOENIX,14 CHAMPION PLATFORM,16 and CHAMPION PCI17), we compared ischemic and bleeding risks in patients assigned to cangrelor but not receiving GPIs with patients assigned to clopidogrel and receiving planned routine GPIs.
The study designs and primary results14,16,17 of each of the CHAMPION trials have been reported previously. The present study’s data analysis was conducted from October 28, 2015, to August 6, 2016. In brief, patients 18 years or older who required PCI for stable angina, non–ST-segment elevation acute coronary syndrome (NSTE-ACS), or ST-segment elevation MI (STEMI) were eligible for enrollment. Patients who had received P2Y12 antagonists or abciximab within 5 to 7 days or eptifibatide, tirofiban hydrochloride, or fibrinolytic therapy within 12 hours before randomization were excluded (except in CHAMPION PCI,17 in which prestudy use of clopidogrel at doses ≤75 mg/d was permitted).
Trial protocols were approved by the institutional review boards or ethics committees at each participating center (eAppendix in the Supplement); the present study was included within that approval. All patients provided written informed consent for participation.
Each of the CHAMPION clinical trials14,16,17 was prospective, double-blind, double-dummy, and randomized; patients were randomized to receive cangrelor or clopidogrel. Cangrelor or matching placebo was administered as a 30-μg/kg intravenous bolus followed by a 4-μg/kg/min infusion for at least 2 hours or for the duration of PCI, whichever was longer. Clopidogrel (600 mg) was administered to all patients in the cangrelor arm at the end of the infusion. In the comparator arm, clopidogrel (300 or 600 mg) was given to patients at the beginning of PCI (CHAMPION PCI),17 at the end of PCI (CHAMPION PLATFORM),16 or at the start or end of PCI based on the study site standard of care (CHAMPION PHOENIX).14 Aspirin (75-325 mg) and clopidogrel (75 mg/d) were administered to all patients during the first 48 hours. Dual antiplatelet therapy after 48 hours, periprocedural anticoagulation, choice of stent and access site, and sheath management protocol were left to the discretion of individual site investigators.
All efficacy end points were assessed in the modified intention-to-treat (mITT) population, which included all patients who underwent index PCI and received the study drug; safety end points were assessed in patients who underwent randomization and received the study drug. For the purposes of this post hoc pooled analysis, the prespecified efficacy end points in the CHAMPION PHOENIX trial were used.14 The primary efficacy end point was the composite of all-cause mortality, MI, ischemia-driven revascularization, or stent thrombosis at 48 hours. The key secondary efficacy end point was the occurrence of stent thrombosis at 48 hours. Periprocedural MI was defined according to the universal definition, as used in the CHAMPION PHOENIX trial14 and in the overall pooled patient-level analysis.13 This definition for MI was retrospectively applied to all adjudicated events in the CHAMPION PLATFORM16 and CHAMPION PCI trials.17 Stent thrombosis was categorized as Academic Research Consortium–defined stent thrombosis (if it occurred after PCI)18 or intraprocedural (if it occurred during PCI).19 Intraprocedural stent thrombosis was specifically adjudicated by a blinded angiographic core laboratory (Cardiovascular Research Foundation) in the CHAMPION PHOENIX trial.14 Stent thrombosis in the other 2 trials was assessed only in patients requiring ischemia-driven revascularization. The primary and secondary efficacy end points were specifically adjudicated by an independent clinical events committee in each of the trials. Major safety end points at 48 hours included requirement for blood transfusion and non-coronary artery bypass graft surgery–related severe bleeding assessed by GUSTO, Thrombolysis in Myocardial Infarction (TIMI),20 and Acute Catheterization and Urgent Intervention Triage Strategy (ACUITY)21 criteria. Bleeding indices were not independently adjudicated.
All 3 CHAMPION trials collected information prospectively on the use and indications of GPIs. Only rescue GPI therapy was permitted in CHAMPION PHOENIX14 and CHAMPION PLATFORM16 for the management of PCI-related thrombotic complications; planned procedural GPI therapy was allowed in CHAMPION PCI17 at the discretion of the individual operator. To minimize indication bias and confounding, patients requiring bailout or rescue GPI therapy were excluded from the present analysis. In this post hoc analysis, patients randomized to cangrelor who did not receive GPIs (cangrelor alone group) were compared with patients randomized to clopidogrel (or placebo) who received routine GPIs (clopidogrel-GPI group). All patients in the clopidogrel-GPI group received GPIs as a planned strategy at the discretion of the site investigator.
Baseline characteristics and efficacy and safety end points at 48 hours were compared between the cangrelor alone and clopidogrel-GPI groups. To account for differences in baseline risk profiles and confounding by indication, a propensity score (PS)–matched analysis was undertaken. Logistic regression modeling calculated the PS for each patient, representing the probability of receiving a GPI conditional on the following covariate set: age, sex, weight, region (United States vs non–United States), indication for PCI (stable angina, NSTE-ACS, and STEMI), abnormal levels of cardiac biomarkers, stent type, prior thienopyridine exposure, prior stroke or transient ischemic attack, medical history (diabetes, heart failure, and peripheral artery disease), planned clopidogrel loading dose (300 or 600 mg), and periprocedural heparin or bivalirudin. Variables chosen for PS determination were those with a significance level of P < .15 for baseline comparisons. Patients in the cangrelor alone and the clopidogrel-GPI groups were matched 1:1 using nearest-neighbor matching without replacement, with caliper width of 0.2 times the SD.22 Adequacy of matching was tested using the standard differences method, and values less than 10% reflect more balanced matched groups. Primary and secondary efficacy and safety outcomes were compared between the 2 matched cohorts using conditional logistic regression. Ischemic and bleeding end points were also combined to determine net clinical benefit of each antiplatelet approach. Because CHAMPION PCI was the only trial that allowed for GPI utilization for any indication, additional sensitivity analyses were performed to test the comparability of the CHAMPION PCI sample to the overall results of the pooled analysis. Patients in the cangrelor alone and the clopidogrel-GPI groups were rematched based on calculated PS within the mITT population of the CHAMPION PCI alone. Baseline characteristics and clinical outcomes were compared between the new matched pairs in CHAMPION PCI.
Continuous variables are presented as mean (SD) or as median (interquartile range) and were compared using t tests or Wilcoxon rank sum tests, as appropriate. Categorical variables are presented as number (percentage) and were compared using χ2 testing or Fisher exact tests, as appropriate. Logistic regression analyses were used to calculate effect sizes in both the original and PS-matched cohorts; the effect sizes are expressed as odds ratios (ORs) and associated 95% CIs. All statistical tests were 2-tailed with a level of significance of P < .05, and statistical analyses were performed using SAS, version 9.3 (SAS Institute Inc).
Across the mITT populations of the 3 CHAMPION trials (N = 24 902), 3173 patients (12.7%) received a GPI during hospitalization, of whom 745 (23.5%) received GPIs for bailout or rescue purposes and thus were excluded from the present analysis. A total of 12 140 patients were included in the analysis; 8779 were men (72.3%), and the mean (SD) age was 63.2 (11.3) years. Of the total mITT population, 10 929 patients were assigned to the cangrelor arm and did not receive a GPI (cangrelor alone group), and 1211 patients were assigned to clopidogrel (or placebo) and received routine GPI (clopidogrel-GPI group). Most patients in the clopidogrel-GPI cohort received eptifibatide (941 [77.7%]), 176 (14.5%) received abciximab, and 91 (7.5%) were given tirofiban; the specific GPI used was not reported in 3 patients.
There were baseline discrepancies in patient profiles between the 2 selected cohorts (Table 1). Patients in the clopidogrel-GPI group were more likely to be male, younger, and enrolled from the United States compared with those in the cangrelor alone group. The clopidogrel-GPI cohort also presented for PCI more frequently for NSTE-ACS and STEMI and were more likely to receive drug-eluting stents. However, patients in the clopidogrel-GPI cohort had a lower comorbid disease burden (diabetes, heart failure, and peripheral artery disease) and were more likely to receive a planned clopidogrel loading dose of 600 mg. Patients in the cangrelor alone group were more likely to receive bivalirudin compared with those in the clopidogrel-GPI group.
To account for these baseline risk imbalances, 1:1 PS matching yielded 1021 unique matched pairs. Most (828 [81.1%]) of the patients included in the clopidogrel-GPI group in the final PS-matched analysis were participants in CHAMPION PCI, which allowed GPI use for any indication (Table 2). No residual significant differences in baseline characteristics existed between the PS-matched cangrelor alone and clopidogrel-GPI cohorts (Table 1).
In the original cohort analysis, rates of the primary and secondary efficacy end points did not differ significantly between the cangrelor alone and clopidogrel-GPI groups. However, patients in the cangrelor alone cohort experienced consistently lower crude rates of safety end points, assessed by all bleeding indices and blood transfusion requirement (Table 3).
In the final PS-matched end point analysis (Table 3), rates of the primary efficacy end point were not significantly different between the cangrelor alone and clopidogrel-GPI groups (2.6% vs 3.3%; OR, 0.79; 95% CI, 0.48-1.32). Similarly, rates of the key secondary efficacy end point, stent thrombosis, were low in both cohorts and did not differ significantly by intravenous antiplatelet group (0.1% vs 0.6%; OR, 0.17; 95% CI, 0.02-1.38).
There were numerically (but not statistically) lower rates of GUSTO-defined severe/life-threatening bleeding (0.3% vs 0.7%; OR, 0.43; 95% CI, 0.11-1.66) and GUSTO-defined moderate or severe/life-threatening bleeding (1.2% vs 2.3%; OR, 0.50; 95% CI, 0.24-1.03) with cangrelor alone compared with clopidogrel-GPI. Rates of TIMI-defined major or minor bleeding were significantly lower in patients receiving cangrelor alone compared with patients receiving clopidogrel-GPI (0.7% vs 2.4%; OR, 0.29; 95% CI, 0.13-0.68). Similarly, the cangrelor alone cohort experienced lower rates of ACUITY-defined major bleeding compared with the clopidogrel-GPI cohort (3.6% vs 5.8%; OR, 0.61; 95% CI, 0.40-0.94). Furthermore, blood transfusion requirement was lower in the cangrelor alone group compared with the clopidogrel-GPI group (1.0% vs 2.1%; OR, 0.45; 95% CI, 0.20-0.99). In the analysis of net clinical benefit, the cangrelor alone group experienced numerically lower rates of combined primary ischemic and bleeding end points compared with the clopidogrel-GPI group (2.9% vs 3.8%; OR, 0.77; 95% CI, 0.48-1.24) (Table 3), but this comparison did not reach statistical significance.
Dedicated sensitivity analysis was performed using patients included in the mITT population of the CHAMPION PCI trial alone. Subsequent PS matching yielded 665 unique matched pairs of patients enrolled in CHAMPION PCI. Major baseline characteristics were well balanced between the PS-matched cangrelor alone and clopidogrel-GPI groups in the CHAMPION PCI trial (eTable 1 in the Supplement). Similar patterns and trends were observed in efficacy and safety end points between the 2 selected groups in the CHAMPION PCI trial as in the overall pooled analysis of the 3 CHAMPION trials (eTable 2 in the Supplement).
In this exploratory analysis of the CHAMPION program, patients randomized to cangrelor who did not receive GPIs experienced similar periprocedural ischemic risks compared with patients randomized to clopidogrel who received routine GPIs (as a planned strategy). There were lower risk-adjusted bleeding risks in patients treated with cangrelor alone compared with those in the clopidogrel-GPI group, even after PS matching to account for baseline discrepancies in patient profiles.
There is a continued unmet clinical need for potent but safe antithrombotic agents in contemporary PCI practice,23 and the optimal approach to intensification of antiplatelet and antithrombotic therapy in high-risk patient scenarios remains to be determined.24,25 Abciximab was approved for use in the United States by the Food and Drug Administration in December 1994.26 Since their introduction as a class, GPIs continue to be a guideline-recommended strategy to attenuate PCI-related thrombotic risk27 and maintain robust utilization in contemporary PCI in the United States, with rates exceeding 30% in the setting of ACS.28,29 A series of landmark randomized clinical trials has shown that GPIs lower the rates of major cardiovascular events compared with heparin-based regimens in PCI.2 More recently, real-world data from almost 1 million unselected patients with ACS included in the National Cardiovascular Data Registry have corroborated these randomized experiences.28
Unfortunately, the antiplatelet effects of GPIs last long after cessation of infusion, contributing to an excess in significant bleeding,3 an end point that has consistently been linked to adverse clinical course and excess costs.30-32 These risks have been partially attenuated by several bleeding-reduction strategies, including limiting the dosage and duration of GPI infusion33; restricting use to carefully selected, high-risk patients; augmenting radial access approaches; and improving sheath management protocols.7 In parallel, there have been significant advancements in the available antithrombotic armamentarium (namely, bivalirudin and newer-generation P2Y12 antagonists), practice patterns, and processes of care since the initial GPI studies, prompting reappraisal of GPI use in contemporary PCI.
Prasugrel34 and ticagrelor35 have been demonstrated to reduce periprocedural ischemic events compared with clopidogrel, even in the presence of provisional or planned GPI utilization. It is conceivable that the relative benefit of cangrelor would have been attenuated had the comparator been prasugrel or ticagrelor (instead of clopidogrel). However, exploring the transition from parenteral agents to oral clopidogrel remains clinically relevant because clopidogrel continues to be the most commonly used P2Y12 inhibitor worldwide.29,36 Furthermore, platelet function studies have revealed that cangrelor provides faster-acting, more consistent, and more complete platelet inhibition compared with either prasugrel37 or ticagrelor.38 The dedicated CHAMPION platelet function substudy validated cangrelor’s potent and fast-acting antiplatelet properties and demonstrated no significant pharmacodynamic interaction in the transition from cangrelor to clopidogrel therapy.9 Furthermore, all oral P2Y12 inhibitors are limited by variable gastrointestinal absorption and drug bioavailability, especially in the context of ACS in which there is potential for impaired perfusion.39,40 As such, the pharmacologic profile of cangrelor (intravenous, complete, and robust periprocedural antiplatelet efficacy as well as rapidly reversible effects)10-12 may provide a more favorable index of safety compared with GPIs and can be more broadly applied to high-risk PCI patients compared with available potent, oral P2Y12 antagonists.41 Future studies will be needed to examine the cost-effectiveness of different approaches toward platelet inhibition.
This pooled, individual patient-level data analysis from the 3 landmark CHAMPION trials represents what we believe to be the largest clinical experience of cangrelor available to date. Based on our PS-matched analysis, cangrelor may offer an alternative potent, parenteral strategy to GPIs with a favorable bleeding profile in PCI. Most patients included in the present analysis presented with an ACS and, as such, our data primarily reflect this experience and inform, to a lesser extent, stable coronary artery disease. Emerging data have suggested that bivalirudin may complement the antiplatelet efficacy of cangrelor42 while limiting attendant bleeding hazards in contemporary PCI. However, further data are required to validate this approach.
To our knowledge, there are no randomized data available directly comparing these therapies. In the absence of randomized data, matched comparisons have been used as a validated alternative33; however, as with all nonrandomized data, we may not be able to identify all relevant factors to facilitate adequate matching regardless of the methodology used. Thus, analysis and inference may be subject to residual bias and confounding due to unmeasured or unknown covariates.33 We acknowledge that PS matching was incomplete in our study, failing to account for certain variables attributable to lack of consistent data availability across the 3 CHAMPION trials, including specific access site, composite bleeding risk scores, and angiographic characteristics. Although we accounted for regional differences (US sites vs non-US sites) in the PS calculation, we did not pursue further country-specific matching since several countries enrolled few patients,43 which would have substantially limited the power of this exploratory analysis. Despite excluding patients requiring bailout GPI therapy, inherent to the study design, the results are subject to indication bias (eg, sicker patients requiring GPI therapy administered routinely may be at lower perceived bleeding risk). The overall CHAMPION program was not designed or powered to compare cangrelor with GPIs.
The intention of the present analysis was to examine GPIs when used routinely as a standard of care. As such, more than 80% of the PS-matched clopidogrel-GPI group of the analyzed mITT population was enrolled from CHAMPION PCI, a trial that permitted GPI utilization for any indication. The use of GPIs by operators in CHAMPION PLATFORM and CHAMPION PHOENIX was intended to be for bailout or rescue purposes alone in the management of PCI-related thrombotic complications. However, despite these protocol specifications, a small proportion of operators continued to use GPIs as a part of routine care in these trials. Despite the variation in GPI utilization across the 3 CHAMPION trials, dedicated sensitivity analysis demonstrated the comparability of the results in the large CHAMPION PCI subset with the overall pooled analysis. Worldwide, GPI use appears to be declining, especially for routine indications; however, uptake in the United States remains robust with important regional variation.28,29 Furthermore, clinicians are varying the dose, duration, and timing of GPI therapy to attenuate the attendant bleeding risks; however, data regarding these factors were not routinely collected or accounted for in our study. Similar to patients in other contemporary antithrombotic clinical trials, patients with a history of major bleeding or at high risk for bleeding were largely excluded from the CHAMPION experience. Bleeding events in the CHAMPION trials were site reported by blinded investigators and were not independently adjudicated by a clinical events committee. However, independent post hoc adjudication of bleeding events in the CHAMPION PHOENIX trial by the sponsor and the US Food and Drug Administration did not appear to qualitatively change the safety results when compared with the original prespecified, investigator-reported analyses.44
Based on a pooled analysis from the 3 phase 3 CHAMPION trials, cangrelor was associated with similar ischemic risk and reduced bleeding risk compared with clopidogrel plus routinely used GPIs. Cangrelor, coupled with augmented use of bleeding-mitigating approaches, may reduce periprocedural thrombotic events with a favorable safety profile. Comparative effectiveness and cost-effectiveness data on fast-acting, potent antiplatelet regimens are required.
Corresponding Author: Deepak L. Bhatt, MD, MPH, Brigham and Women’s Hospital Heart & Vascular Center and Harvard Medical School, 75 Francis St, Boston, MA 02115 (email@example.com).
Accepted for Publication: September 30, 2016.
Published Online: November 30, 2016. doi:10.1001/jamacardio.2016.4556
Author Contributions: Drs Harrington and Bhatt had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Harrington, Stone, Steg, Hamm, Mahaffey, White, Bhatt.
Acquisition, analysis, or interpretation of data: Vaduganathan, Stone, Deliargyris, Steg, Gibson, Hamm, Price, Menozzi, Prats, Elkin, Bhatt.
Drafting of the manuscript: Vaduganathan, Bhatt.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Prats, Elkin.
Administrative, technical, or material support: Harrington, Bhatt.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Harrington serves on advisory boards with honoraria from Amgen, Gilead Sciences, Merck, MyoKardia, The Medicines Company, and WebMD; has received research funding from AstraZeneca, CSL Behring, GlaxoSmithKline, Merck, Portola, Regado, Sanofi, The Medicines Company, and Jansen; has ownership in SignalPath, Scanadu, MyoKardia, and Element Science; is a member of the board of directors of the American Heart Association and Stanford Healthcare, with no financial compensation. Dr Stone has received honoraria from Boston Scientific, InspireMD, Atriaum, Eli Lilly-Daiichi Sankyo partnership, and AstraZeneca. Dr Deliargyris is employed by The Medicines Company. Dr Steg has received funding (provided to Institut National de la Santé et de la Recherche Médicale U1148–Laboratory for Vascular Translational Science) from Sanofi and Servier; has received speaking or consultant fees from Amarin, AstraZeneca, Bayer, Boehringer-Ingelheim, Bristol-Myers Squibb, CSL-Behring, Daiichi-Sankyo, GlaxoSmithKline, Janssen, Eli Lilly, Novartis, Pfizer, Regeneron, Roche, Sanofi, Servier, and The Medicines Company; and owns stock in Aterovax. Dr Gibson has received honoraria from The Medicines Company. Dr Hamm has received honoraria from AstraZeneca, Sanofi, and Lilly and research funding from Astra Zeneca and The Medicines Company. Dr Price has received honoraria from AstraZeneca, Merck & Co, Accriva Diagnostics, and The Medicines Company. Dr Menozzi has received honoraria from Astra Zeneca and Correvio. Dr Prats and Mr Elkin are employed by The Medicines Company. Dr Mahaffey has received research funding from Amgen, Daiichi, Johnson & Johnson, Medtronic, Merck, St Jude, and Tenax; has received consultant fees from the American College of Cardiology, AstraZeneca, BAROnova, Bayer, Boehringer Ingelheim, Bio2 Medical, Bristol-Myers Squibb, Cubist, Eli Lilly, Elsevier, Epson, Forest, GlaxoSmithKline, Johnson & Johnson, Medtronic, Merck, Mt. Sinai, MyoKardia, Omthera, Portola, Purdue, Springer, The Medicines Company, Theravance, Vindico, and WebMD; and owns stock in BioPrint Fitness. Dr White has received honoraria from AstraZeneca and research funding from Sanofi, Eli Lilly, National Health Institutes, GlaxoSmithKline, Merck Sharpe & Dohme, and AstraZeneca. Dr Bhatt is a member of the advisory boards of Cardax, Elsevier Practice Update Cardiology, Medscape Cardiology, and Regado Biosciences; serves on the board of directors of Boston Veterans Affairs (VA) Research Institute and the Society of Cardiovascular Patient Care; is chair of the American Heart Association Quality Oversight Committee; serves on data monitoring committees of Duke Clinical Research Institute, Harvard Clinical Research Institute, Mayo Clinic, and Population Health Research Institute; has received honoraria from the American College of Cardiology (senior associate editor, Clinical Trials and News, ACC.org), Belvoir Publications (editor in chief, Harvard Heart Letter), Duke Clinical Research Institute (clinical trial steering committees), Harvard Clinical Research Institute (clinical trial steering committee), HMP Communications (editor in chief, Journal of Invasive Cardiology), Journal of the American College of Cardiology (guest editor and associate editor), Population Health Research Institute (clinical trial steering committee), Slack Publications (chief medical editor, Cardiology Today’s Intervention), Society of Cardiovascular Patient Care (secretary/treasurer), WebMD (continuing medical education steering committees); Clinical Cardiology (deputy editor), National Cardiovascular Data Registry–Acute Coronary Treatment and Intervention Outcomes Network Registry Steering Committee (chair), and VA Clinical Assessment, Reporting, and Tracking Program Research and Publications Committee (chair); has received research funding from Amarin, Amgen, AstraZeneca, Bristol-Myers Squibb, Eisai, Ethicon, Forest Laboratories, Ischemix, Medtronic, Pfizer, Roche, Sanofi, The Medicines Company (including for his role as cochair of the 3 CHAMPION trials); has received royalties from Elsevier (editor, Cardiovascular Intervention: A Companion to Braunwald’s Heart Disease); has been a site co-investigator for Biotronik, Boston Scientific, and St. Jude Medical; is a trustee of the American College of Cardiology with no financial compensation; and performed unfunded research for FlowCo, PLx Pharma, and Takeda. There were no other disclosures.
Funding/Support: The CHAMPION program was supported by The Medicines Company.
Role of the Funder/Sponsor: Each of the CHAMPION trials was designed by an academic executive committee and the sponsor (The Medicines Company). The primary data for each trial were collected by the sponsor. Statistical analyses for the present study were performed by the sponsor and independently verified by the Harvard Clinical Research Institute. The results were subsequently compared with the results obtained by the sponsor, and discrepancies were resolved collaboratively. The sponsor had the right to review but not approve the final manuscript and did not influence the decision to submit the manuscript for publication.
Disclaimer: Dr Harrington is a deputy editor of JAMA Cardiology but was not involved in the editorial review or the decision to accept the manuscript for publication.
Additional Information: A full list of the CHAMPION investigators can be found in Bhatt DL et al14,16 and Harrington et al.17
Additional Contributions: Debra Bernstein, PhD (The Medicines Company), provided statistical support; there was no additional financial compensation. Yuyin Liu, MS, and Lanyu Lei, MS (Harvard Clinical Research Institute), performed independent verification of the analyses. Harvard Clinical Research Institute received funding from The Medicines Company for these analyses.