August 26th, 2012
TRILOGY at ESC: No Advantage for Prasugrel Over Clopidogrel in Medical ACS Patients
Larry Husten, PHD
The newer antiplatelet agent prasugrel was no better than the old standby clopidogrel for treating patients with acute coronary syndrome (ACS) who are not undergoing revascularization. The results of the TRILOGY ACS (Targeted Platelet Inhibition to Clarify the Optimal Strategy to Medically Manage Acute Coronary Syndromes) trial were presented by Matt Roe at the European Society of Cardiology meeting in Munich and published simultaneously in the New England Journal of Medicine.
In the primary analysis of the trial, prasugrel (10 mg daily) was compared with clopidogrel (75 mg daily) in 7243 ACS patients younger than age 75. At 17 months, the rate of CV death, MI, or stroke did not differ significantly between the two groups:
- 13.9% in the prasugrel group versus 16.0% in the clopidogrel group (hazard ratio [HR] for prasugrel, 0.91; 95% CI, 0.79-1.05, P=0.21)
The results were similar in a secondary analysis, which also included 2083 patients who were age 75 or older and who received either low-dose prasugrel (5 mg daily) or the standard clopidogrel dose.
In a separate prespecified analysis, prausgrel patients under age 75 had a decreased risk for multiple recurrent ischemic events (HR, o.85; 95% CI, 0.72-1.00, P=0.04). According to the TRILOGY investigators, this finding is consistent with results from the earlier TRITON trial. TRITON compared prasugrel with clopidogrel in ACS patients undergoing revascularization; recurrent ischemic events in the prasugrel group were reduced by 30%, with most of the reduction occurring later in the trial. “Although this observation is exploratory,” the TRILOGY investigators wrote, ” it raises the question of whether investigation of the multiplicity of ischemic events is warranted in future secondary-prevention trials, rather than solely analyzing the time to the first event, as has been traditional” in previous studies.
The prasugrel and clopidogrel groups did not differ significantly in the rate of severe and intracranial bleeding. However, within the younger subgroup (taking the higher prasugrel dose), prasugrel recipients had more major and minor bleeding complications than clopidogrel recipients did — a reflection, the authors wrote, of “the more intense platelet inhibition with prasugrel.”
The prasugrel and clopidogrel groups did not differ significantly in the rate of severe and intracranial bleeding. However, within the younger subgroup (taking the higher prasugrel dose), prasugrel recipients had more major and minor bleeding complications than clopidogrel recipients did — a reflection, the authors wrote, of “the more intense platelet inhibition with prasugrel.”
An exploratory analysis found that in the main group of patients under the age of 75, the Kaplan-Meier curves for the primary endpoint and for the individual components of the endpoint were not different for the first year but diverged afterward in favor of prasugrel. “The reasons for this finding remain uncertain,” the investigators wrote.
A subgroup analysis suggested the possibility that prasugrel might be more effective than clopidogrel in current or recent smokers, in patients who underwent angiography prior to randomization, and in patients taking a proton-pump inhibitor.
[ADDITIONAL COVERAGE: See Harlan Krumholz’s take on TRILOGY in Journal Watch Cardiology]
Agree with Harlan’s take on Trilogy that “greater platelet inhibition with prasugrel than with clopidogrel did not translate into a discernible benefit in patients with acute coronary syndromes who did not undergo revascularization. The lower-cost agent was as good as the higher-cost one”.
I find the TRILOGY investigators’ focus on time-dependent treatment effect (greater benefit after 12 months of treatment exposure) and recurrent events, both of which favor prasugrel, an attempt at cherry picking the data that portray prasugrel in the best light. They need to be cautious in not overinterpreting these observations as well as the results of subgroup analyses.
The time to first event is the most robust statistical approach to survival analysis and is generally accepted as the regulatory standard for approval. Although accounting for total events (including recurrent events) can be informative and add to trial efficiency, there are statistical issues that need to be considered. For example, are the recurrent events independent of or related to the first event? This cannot be gleaned from the NEJM main paper or the supplement.
It is important to note that the TRILOGY findings are relevant to the low- to moderate-risk patients with USAP/NSTEMI that do not undergo revascularization. Less than 11% of patients were at high-risk as reflected in a GRACE score >140.
Would the study have been “positive” if it had been powered to detect a more “realistic” relative risk reduction (RRR) instead of the 22% RRR? I would be interested in hearing others’ perspectives.