September 1st, 2013

Inside the ACCOAST Trial of Pretreatment Prasugrel

, , and

CardioExchange’s John Ryan, Rick Lange, and David Hillis interview Gilles Montalescot, lead investigator of the ACCOAST trial, about why prasugrel pretreatment did not benefit patients with non–ST-segment elevation acute coronary syndromes. The study, published in The New England Journal of Medicine, was presented at the European Society of Cardiology conference in Amsterdam.


In a study funded by the manufacturer of prasugrel, researchers randomized 4033 patients with NSTE acute coronary syndromes and a positive troponin level who were scheduled to undergo catheterization to receive either placebo or a 30-mg pre-angiography loading dose of prasugrel. The two groups did not differ significantly in the primary composite endpoint of death from cardiovascular causes, myocardial infarction, stroke, urgent revascularization, or glycoprotein IIb/IIIa inhibitor rescue therapy through day 7. Major bleeding was significantly more common with pretreatment than without it. The findings persisted at 30 days and in prespecified subgroups.


CardioExchange: In the patients pretreated with prasugrel, did the low (i.e., 30-mg) loading dose provide adequate antiplatelet activity to influence ischemic events before cardiac catheterization?

Montalescot: Yes! See the pharmacodynamic substudy (graph in the article’s appendix) with significant antiplatelet effect at the time of catheterization in the pretreatment group. Moreover, if this group had more bleeding, it is likely that pretreament with 30 mg was effective.

CardioExchange: Pretreatment with aspirin and a P2Y12 antagonist is common practice for NSTE patients. Your study shows that pretreatment with prasugrel did not limit ischemic events in these patients. Under what circumstances, if any, should dual antiplatelet therapy be administered to NSTE patients before coronary angiography?

Montalescot: The value of pretreatment with clopidogrel is controversial with several negative randomized trials (CREDO, PRAGUE 8, ARMYDA5) and a meta-analysis showing no effect on mortality; our study shows no benefit of prasugrel pretreatment. Ticagrelor has not been tested as pretreatment. Everyone must now draw his own conclusions.

CardioExchange: Your study’s mean time from randomization to cardiac catheterization was very short (median, 4 hours), even though most patients were not considered “high risk” (only one fourth had a GRACE score >140). Does this reflect real-world experience? What antiplatelet regimen would you recommend for centers that perform angiography 24 to 48 hours after patient presentation?

Montalescot: Median time to catheterization in ACCOAST was similar to that in CURRENT, ACUITY, PLATO-NSTEACS PHOENIX, and TRACER — all had delays <5 hours. One fourth of patients with a GRACE score >140 is also similar to other studies such as TIMACS. All our patients had a positive troponin a marker of poorer prognosis (most previous studies did not require positive markers for inclusion). Our population is an intermediate- to high-risk cohort. ACCOAST allowed catheterization up to 48 hours, and the results are the same when examined by quartile of timing. So our results are also applicable to centers that perform angiography 24 to 48 hours after patient presentation. We do not know beyond 48 hours.

CardioExchange: The study was stopped early, in November 2012, because of increased harm (i.e., excessive bleeding) in the patients pretreated with prasugrel without public disclosure. Was the sponsor obliged to divulge and disseminate the results as soon as it had a reasonably reliable estimate of the harm?

Montalescot: You should ask the sponsor of the study about the necessity of a press release in this situation. Investigators, authorities, and the clinical-trials website all were informed. The harm was not for the on-label use of the drug, as determined by the TRITON study.


Are you surprised by the findings from the ACCOAST trial? Why or why not?

Journal Watch Cardiology‘s coverage of this study

Comments are closed.