Patients in the cangrelor group had a significantly lower rate of the composite of all-cause death, myocardial infarction (MI), ischemia-driven revascularization, and stent thrombosis at 48 hours compared with clopidogrel (4.7% versus 5.9%), according Bhatt and colleagues, who reported the findings online in the New England Journal of Medicine.
Researchers also found a significant 38% reduction in stent thrombosis at 48 hours (OR 0.62, 95% CI 0.43 to 0.90, P=0.01).
The rate of major bleeding tended to be higher with cangrelor, though not significantly so, at 48 hours versus clopidogrel: 0.16% and 0.11% (OR 1.50, 95% CI 0.53 to 4.22, P=0.44).
With respect to the individual endpoints, those on cangrelor had fewer MIs and stent thrombosis, which drove the significance of primary endpoint, researchers noted.
In two previous phase 3 trials — CHAMPION PCI and CHAMPION PLATFORM — cangrelor reduced the secondary endpoints of all-cause death and stent thrombosis, but not MIs, which caused the trials to miss their primary endpoints.
Major bleeding was significantly greater among cangrelor patients in the CHAMPION PLATFORM trial, but in CHAMPION PCI, the rate of major bleeding was numerically higher, but it did not reach statistical significance.
At the time of the reporting of the two prior randomized controlled trials in 2009, many interventional cardiologist saw a glimmer of hope in the results (particularly the reduction in stent thrombosis), but also felt the definition of periprocedural MI had hampered the chance of the primary endpoint reaching significance.
“[T]he definition of periprocedural myocardial infarction in those studies did not allow discrimination of reinfarction in patients presenting for PCI [percutaneous coronary intervention] soon after admission with a biomarker-positive acute coronary syndrome,” Bhatt and colleagues wrote.
The median time from hospital admission to randomization was 4 hours, which reflects the high-volume centers in the study, but it also makes it difficult to have serial biomarker readings.
In CHAMPION PHOENIX, baseline biomarker status was carefully assessed to determine periprocedural MIs, he added.
Unlike clopidogrel (and similar oral antiplatelet drugs) cangrelor has a quick onset and quick offset, a plasma half-life of 3 to 6 minutes, and returns patients to normal platelet function in 60 minutes, a feature that makes the agent attractive for use in stent patients.
“Where does cangrelor fit in the armamentarium of dual antiplatelet therapy?” asked Richard A. Lange, MD, and L. David Hillis, MD, of the University of Texas Health Sciences Center at San Antonio, in an accompanying editorial.
“Unfortunately, the study by Bhatt et al. does not answer this question definitively” because a quarter of those in the clopidogrel group received a 300-mg loading dose, “which is inferior to a dose of 600 mg in achieving platelet inhibition and preventing periprocedural ischemic events,” Lange and Hillis contended.
Bhatt and colleagues addressed this concern in the paper. The odds ratio for the primary endpoint with cangrelor was 0.77 (95% CI 0.63 to 0.94) with a loading dose of 600 mg clopidogrel and 0.84 (95% CI 0.62 to 1.14) with the 300 mg loading dose (P=0.62 for interaction), they wrote.
Lange and Hillis also frowned upon the fact that slightly more than one-third of the clopidogrel group were given the drug during or after PCI, essentially resulting in “suboptimal” antiplatelet effects.
Bhatt and colleagues noted this fact but said the results were similar when they adjusted for the timing of the loading dose (P=0.99 for interaction):
- Patients who received clopidogrel immediately before PCI — OR 0.80 (95% CI 0.64 to 0.98)
- Patients who received clopidogrel during and after PCI — OR 0.79 (95% CI 95% CI 0.59 to 1.06)
The investigators wrote that although longer duration of clopidogrel might have “attenuated” the results, pretreatment with clopidogrel “necessitates treatment before delineation of the coronary anatomy, which might then be problematic if emergency cardiac surgery” is warranted.
With respect to the two-thirds of patients who received clopidogrel before PCI, the editorialists questioned why Bhatt and colleagues did not include information about how much time elapsed between clopidogrel administration and PCI, which “makes it difficult to ascertain whether the antiplatelet effects of the drug were maximal at the time of PCI.”
Lange and Hillis also pointed out that ticagrelor or prasugrel would be used today in about 44% of the patients who presented with an acute coronary syndrome, but these cangrelor results cannot be generalizable to those other antiplatelet agents because studies comparing them are “lacking.”
They concluded that “the routine use of [cangrelor] for all patients undergoing PCI is not yet justified.”
Bhatt said that although the CHAMPION PHOENIX was a trial about a drug, it was also a trial about a strategy, which involved some sites not pretreating patients with clopidogrel.
“This aspect might have made the data a bit more challenging to interpret. We realize not everyone practices that way, but for those who do, we showed cangrelor reduces events,” he said.
The researchers randomized 11,145 patients from 145 sites from September 2010 to October 2012. The total modified intention-to-treat population was 10,942, randomized nearly equally.
Patients were randomized to double-blind, double-dummy treatment with a bolus and infusion of cangrelor (30 μg/kg then 4 μg/kg/minute) or a loading dose of 600 mg or 300 mg of clopidogrel.
The median age was 64, two-thirds were men, most were Caucasian, and 37% were from the U.S.
The diagnosis at presentation was stable angina in about 56%, non-ST-segment elevation MI (non-STEMI) in about 25%, and STEMI in about 18%. The reduction in the primary efficacy endpoint was similar among the three types of diagnoses.
The usual cardiovascular risk factors were present in typical proportions: diabetes (28%), smokers (28%), hypertension (80%), and high cholesterol (70%). Many also had prior stroke, MI, PCI, CABG, heart failure, and peripheral arterial disease.
Drug-eluting stents were implanted in 55% of patients, bare-metal stents in 42%, and balloon angioplasty in the rest. There was no difference in the effect of cangrelor between the two stent types (OR 0.80 versus 0.76).
Bhatt and colleagues reported that the number needed to treat with cangrelor to prevent one primary endpoint was 84 (95% CI 49 to 285).
The reduction in the primary endpoint for cangrelor persisted out to 30 days (OR 0.85, P=0.03), as did the reduction in stent thrombosis (OR 0.68, P=0.01).
Other favorable outcomes for cangrelor included significantly lower:
- Intraprocedural stent thrombosis (OR 0.65)
- Use of rescue therapy with a glycoprotein IIb/IIIa inhibitor (OR 0.65)
- Rate of procedural complications (OR 0.74)
- Net rate of adverse clinical events from a post hoc analysis (OR 0.008)
“Besides its role in ischemic complications of PCI, cangrelor may be useful in clinical situations in which ADP-receptor blockade is needed by a short-acting intravenous agent would be preferred,” Bhatt and colleagues concluded.
The trial was funded by The Medicines Company.
Bhatt reported grant funds to his institution from the Medicines Company; board membership with Medscape Cardiology, Boston VA Research Institute, and the Society of Chest Pain Centers; and payment for developing educational presentations from WebMD. He also reported grants to his institution from Amarin, AstraZeneca, Bristol-Myers Squibb, Eisai, Ethicon, Medtronic, and Sanofi-Aventis.
His co-authors reported relationships with The Medicines Company, Bristol-Myers Squibb, Sanofi-Aventis, Daiichi Sankyo, Eli Lilly, AstraZeneca, Merck, Evolva, Abbott Vascular, Plx Pharma, GlasxSmithKline, Otsuka, J&J, St. Jude Medical, Sunovion, Baxter Healthcare, Bayer, Boston Clinical Research Institute, Cardiovascular Research Foundation, Consensus Medical Communication, CSL Behring, Cytori Therapeutics, Exeter Group, Ischemix, Janssen Pharmaceuticals, Ortho McNeil, Regado Biosciences, Angel Medical, Atrium Medical Systems, Genentech, Ikaria, Lantheus Medical Imaging, Portola Pharmaceuticals, Roche Diagnostics, Stealth Peptides, Volcano, Walk Vascular, Correvio, Boehringer Ingelheim, Gilead, Novartis, Polymedix, Schering-Plough, Medtronic, Iverson Genetics, Amarin, Reva Medical, Boston Scientific, Osprey, Abbott Vascular, Inspire MD, Therox, InfraReDx, Miracor, MPP Group, Lutonix, Velomedix, Cardiovascular Systems, AGA Medical, and Thoratec.
Lange and Hillis had nothing to disclose.