Effect of Platelet Inhibition by Cangrelor Among Obese Patients Undergoing Coronary Stenting: Insights From CHAMPION

Background: In randomized trials, cangrelor reduced periprocedural ischemic events related to percutaneous coronary intervention without increasing GUSTO severe bleeding. However, some antiplatelet agents have shown a differential treatment effect by body mass index (BMI). Methods: Patients from the 3 CHAMPION trials (Cangrelor Versus Standard Therapy to Achieve Optimal Management of Platelet Inhibition) who were randomized to cangrelor versus clopidogrel during percutaneous coronary intervention were stratified by BMI. The primary efficacy end point was a composite of death, myocardial infarction, ischemia-driven revascularization, or stent thrombosis within 48 hours. The principal safety outcome was GUSTO moderate or severe bleeding at 48 hours, although more sensitive bleeding measures such as Thrombolysis in Myocardial Infarction major bleeding were also assessed. We examined obese patients (defined as BMI≥30) versus nonobese patients. Results: There were 24 893 patients, with 8979 (36.1%) having BMI of ≥30. There was no significant difference in the primary efficacy end point among obese versus nonobese patients (4.3% versus 4.2%; rate ratio, 1.01 [95% CI, 0.89–1.15]; P=0.82). There was a consistent benefit in the primary efficacy end point in patients who received cangrelor versus placebo who were obese (3.9% versus 4.7%, rate ratio, 0.83 [95% CI, 0.68–1.02]; P=0.07) and not obese (3.8% versus 4.7%; rate ratio, 0.81 [95% CI, 0.69–0.94]; P=0.0053); interaction P=0.77. There was no difference in GUSTO moderate or severe bleeding among patients who received cangrelor versus placebo who were obese (0.6% versus 0.6%; rate ratio, 0.99 [95% CI, 0.58–1.67]; P=0.96). Conclusions: Cangrelor at the time of percutaneous coronary intervention is effective and safe in obese and nonobese patients. There was no difference in short-term efficacy between obese and nonobese patients. Periprocedural cangrelor is an effective and safe antiplatelet agent, irrespective of BMI. Registration: URL: https://www.clinicaltrials.gov; Unique identifier: NCT01156571, NCT00385138, NCT00305162.

factors, affecting obese patients at a significantly younger age than their peers. [2][3][4][5] It has been postulated, however, that there is a different prothrombotic milieu among obese versus nonobese patients with coronary artery disease, possibly suggesting that a differential treatment response to antiplatelet therapy may exist. 6 Obesity and insulin resistance are known to cause increased platelet count, platelet volume, and oxidative stress. 7 The optimal use of antithrombotic therapy in the obese at the time of percutaneous coronary intervention (PCI) has also been a subject of debate, given the differential pharmacodynamics and physiology associated with obesity. For example, different disciplines may use lean versus actual body mass for dosing of heparin. 8 Very obese patients taking edoxaban versus warfarin may have a higher risk of bleeding, while obese patients taking apixaban versus warfarin had lower all-cause mortality and cardiovascular mortality. 9,10 In 1 study, when 402 patients were loaded with 600 mg of clopidogrel before PCI, obese patients exhibited significantly higher adenosine diphosphate-induced platelet aggregation than nonobese patients. 11 In another study, 1542 consecutive patients who underwent stenting were loaded with clopidogrel versus prasugrel. In both thienopyridine groups, platelet reactivity indices were higher among obese patients with the metabolic syndrome but not obese patients without the metabolic syndrome. 12 Then, in a follow-up study, 186 patients undergoing PCI were assigned to prasugrel or ticagrelor; obese patients taking prasugrel had higher platelet reactivity indices, but obese patients taking ticagrelor did not. 13 Cangrelor is frequently given in addition to the abovementioned agents during PCI, especially during treatment of ST-segment-elevation myocardial infarction and thrombotic lesions. Large-scale randomized data, however, have not yet detected a difference in the response to cangrelor or other P2Y 12 inhibition among obese versus nonobese patients. 14 Although cangrelor has 100% bioavailability and universal dosing, it is not known whether there may be differential treatment effect among the obese, as with the novel oral anticoagulants and other P2Y 12 inhibitors-or whether the addition of cangrelor to these agents with variable pharmacological effects in the obese may lead to a differential treatment effect or bleeding risk.
The 3 phase-3 CHAMPION trials (Cangrelor Versus Standard Therapy to Achieve Optimal Management of Platelet Inhibition) were conducted to test the safety and efficacy of cangrelor infusion versus clopidogrel at the time of PCI. In CHAMPION PLAT-FORM and CHAMPION PCI, there was no difference in the composite efficacy or bleeding safety end points among patients who received cangrelor versus clopidogrel at the time of PCI, whereas, in the larger CHAMPION PHOENIX trial, there was a significant reduction in the combined efficacy end point (death, myocardial infarction, ischemia-driven revascularization, or stent thrombosis at 48 hours) without an increase in bleeding. [15][16][17] This analysis of the combined CHAMPION trials aims to assess whether there is a differential treatment response with respect to major adverse cardiovascular events (MACE) or bleeding among obese versus nonobese patients who were treated with cangrelor versus placebo at the time of PCI.

Study Population
The data that support these findings of the study are available from the corresponding author upon reasonable request. The study designs of the 3 prospective, double-blind, doubledummy CHAMPION trials have been published previously. 18

WHAT THE STUDY ADDS
• In one of the largest prospective studies of antithrombotic therapy in patients undergoing percutaneous coronary intervention, we show that there is a consistent benefit to using cangrelor versus clopidogrel at the time of percutaneous coronary intervention, without any excess of bleeding complications, among obese patients (body mass index ≥30). • Patients with lower body mass index had a higher risk of bleeding regardless of antiplatelet strategy compared with patients with higher body mass index.
minute infusion for at least 2 hours, or the duration of PCI, whichever was longer) or clopidogrel. Timing of oral clopidogrel load (300 or 600 mg) was different in the 3 trials: at the beginning of PCI in CHAMPION PCI, at the end of PCI in CHAMPION PLATFORM, and according to site-specific standard of care in CHAMPION PHOENIX. Periprocedural anticoagulation, stent selection, and sheath management were left to the discretion of the individual clinicians. Subsequent to PCI, patients were treated with dual antiplatelet therapy with aspirin and clopidogrel at the direction of site investigators. Exclusion criteria related to recent P2Y 12 inhibitor use, fibrinolytic, and glycoprotein IIb/IIIa inhibitor therapy are published in the individual trials. [15][16][17] All patients provided written and informed consent before enrollment. The trial protocols and all subsequent amendments were approved by national regulatory agencies in participating countries and by institutional review boards or ethics committees at all participating sites. Independent data safety and monitoring boards provided regular oversight of patient data at regular intervals.

End Points and Follow-Up
As in the CHAMPION PHOENIX trial, the primary end point for this analysis is the cumulative incidence of a composite of death, myocardial infarction, ischemia-driven revascularization, and stent thrombosis at 48 hours. Incidence of the same composite end point at 30 days as well as the individual components were used as secondary end points. For the pooled analysis, the second universal definition of myocardial infarction was used and stent thrombosis was defined by the academic research consortium. 20,21 As specified in the trial protocol, end points were adjudicated according to a prespecified modified intention to treat population, consisting of patients who received the study drug and underwent PCI. The primary safety end point was cumulative incidence of GUSTO (Global Use of Strategies to Open Occluded Arteries) defined noncoronary artery bypass graft moderate or severe bleeding. Secondary bleeding end points included the need for any blood transfusion within 48 hours or Thrombolysis in Myocardial Infarction defined bleeding.

Statistical Analysis
For the primary analysis, patients were divided into 2 groups according to body mass index (BMI); obese patients (BMI≥30), and nonobese patients (BMI<30). Baseline characteristics were compared between the 2 groups; continuous variables expressed as mean (SD) or median (interquartile range) and categorical variables expressed as n (%). Between-group comparisons were made using T-tests for continuous variables and χ 2 for categorical variables.
Efficacy and safety end points were compared between BMI categories and by cangrelor versus clopidogrel arms within BMI categories using Poisson regression analysis to estimate event rates ratio (RR) and 95% CIs. Kaplan-Meier curves were constructed for the primary efficacy end point and compared using the log-rank test. Between-group comparisons of treatment effect were performed using Poisson regression, adjusting for age, sex, and diabetes. In a separate analysis, BMI was also considered as a continuous variable against the efficacy and safety end points, adjusted for age, sex, and diabetes.
In an exploratory analysis, efficacy and safety end points were compared by cangrelor versus clopidogrel within different BMI categories: (BMI, <25, BMI, 25 to <30, BMI 30 to <35, and BMI ≥35). There was no statistical adjustment for multiple comparisons. All statistical analyses were performed using SAS 9.4 (SAS institute, Cary, NC).

RESULTS
Of the 25 384 patients randomized in the CHAMPION trials, 474 (1.9%) did not receive the study drug or did not have a PCI and were not included in the modified intention-to-treat analysis. An additional 17 patients did not have a BMI available for calculation. Thus, there were 24 893 patients included for this analysis; 8979 (36.1%) with a BMI≥30 and 15 914 (63.9%) with a BMI<30.

Baseline Characteristics
There were notable and expected differences among patients who were obese versus nonobese included in this analysis. Compared with nonobese patients, obese patients were younger (61 versus 64 years), more likely to be female (29.6% versus 26.7%), from the United States (54.3% versus 37.5%), and less likely to be smokers than nonobese patients (26.6% versus 30.0%). Obese patients were more likely to have comorbid conditions than nonobese patients: diabetes (40.2% versus 23.6%), hypertension (82.4% versus 71.6%), hyperlipidemia (67.7% versus 54.9%), prior stroke (5.3% versus 5.0%), prior MI (24.4% versus 22.0%). Obese patients were more likely than nonobese patients to present with stable angina (35.0% versus 28.8%) and less likely to present with elevated cardiac biomarker (46.6% versus 52.0%), non-STsegment-elevation myocardial infarction (55.1% versus 58.7%), or ST-segment-elevation myocardial infarction (9.9% versus 12.5%), and more likely than nonobese patients to be treated with a drug-eluting stent (56.7% versus 51.2%). Among patients for whom arterial access site was known, there were similar rates of radial access among obese versus nonobese patients (26.8% versus 25.7%; Table 1). While there were important demographic differences between obese and nonobese patients, baseline characteristics were evenly balanced among patients who were randomized to receive cangrelor versus clopidogrel in both obese and nonobese patients (Table 1).
There were some notable differences in antithrombotic therapies given to obese versus nonobese patients. Obese patients were somewhat more likely to be loaded with 600 mg of clopidogrel and less likely to be loaded with 300 mg of clopidogrel. Obese patients were also more likely to receive bivalirudin and less likely to received unfractionated heparin or low molecular weight heparin (Table 1). After adjusting for geographic region, the differences in clopidogrel loading dose and unfractionated heparin were similar, but there was no difference in frequency of bivalirudin or low molecular weight heparin use. Among obese patients, there was no difference in any of the bleeding end points in patients who received cangrelor versus clopidogrel. However, among nonobese patients,

DISCUSSION
This large-scale analysis of the 3 CHAMPION trials reveals that cangrelor has a consistent benefit over clopidogrel with respect to ischemic events and stent thrombosis at 48 hours and at 30 days in obese and nonobese patients. There was no difference in any of the bleeding end points examined among obese patients who received cangrelor versus clopidogrel. The increasing global rates of obesity and concomitant complications of obesity such as coronary artery disease highlight the importance of these findings.
Understanding the subtleties of pharmacodynamics among obese and very obese patients has become increasingly relevant-in addition to the technical obstacles of performing PCI in this challenging patient population. It is interesting and encouraging to note that there were no differences in procedure length, rates of radial versus femoral arterial access or of successful PCI. Also, the rates of receiving a full 600 mg clopidogrel load were similar among obese and nonobese patients.
The distribution and characteristics of the patients in this study mimicked the so-called obesity paradox, in which obese patients have lower mortality from myocardial infarction, likely because of younger age and more stable presentation of coronary disease. As expected, there was lower unadjusted all-cause mortality among the obese patients when compared with the nonobese patients. As evidenced in the adjusted models, the lower all-cause mortality of the obese group is likely in large measure due to comparing obese patients with older, frailer patients in the lowest BMI categories. 4 Despite these differences, there was a similar benefit to receiving cangrelor versus clopidogrel in the obese and nonobese patients, suggesting against a differential treatment response among obese versus nonobese patients. The rates of both efficacy and bleeding among obese patients were very similar to those of the combined primary trials. There was an increased risk of GUSTO moderate to severe bleeding among nonobese patients who received cangrelor versus clopidogrel, and an increase with decreasing BMI analyzed continuously. Some other measures of serious bleeding trended in this direction among the nonobese group, but none reached significance. This trend toward an excess of bleeding seems to be driven by those in the lowest weight categories, as evidenced in the exploratory analysis and as has been described previously in other antiplatelet agents. 22,23 After adjustment for age, sex, and geographic location, the increased bleeding was seen for all patients with a BMI≤25, irrespective of treatment assignment.
Limitations of this study include the fact that this was not a prespecified analysis of the CHAMPION trials, and thus all P should be considered exploratory. Thus, we do not adjust for multiple comparisons. The 3 trials occurred over an extended time period, in which there were significant advances in clinical practice and stent technology as well as in prevalent pharmacological practices associated with PCI, potentially limiting the generalizability of the data from the earlier trials to current practice. However, when stratified by clinical trial, the benefit of cangrelor in obese patients was the most prominent in the most recent of the 3 trials, CHAMPION PHOENIX, which would be the most generalizable in terms of recent stent technology, medical therapy practices, and patient characteristics. Furthermore, the treatment allocation was randomized, and there appears to be even distribution of potential confounders among both obese and nonobese patients randomized to cangrelor versus clopidogrel. In addition, the very large sample size likely allows for the introduction of these new comparisons without introducing additional confounders.

CONCLUSIONS
There was a consistent benefit of cangrelor versus clopidogrel in obese and nonobese patients undergoing PCI, with respect to short-term efficacy. Also, there was no significant increase in periprocedural bleeding among obese patients undergoing PCI. Periprocedural cangrelor is a safe and effective antiplatelet option in obese patients.