Its diabetic patients who received prasugrel, which remains investigational on both sides of the Atlantic, showed a superior reduction in the rate of the composite end point relative to those who received clopidogrel, but comparable rates of major or minor bleeding and of major bleeding, compared with nondiabetics, report the authors, led by Dr Stephen D Wiviott (Brigham and Women's Hospital, Boston, MA). Their results were published online August 31, 2008 by Circulation, and Wiviott presented the findings the same day at the European Society of Cardiology Congress 2008 [2].

Considered together, according to the researchers, their ischemic-event and bleeding results point to a significantly increased overall "net clinical benefit" from prasugrel relative to clopidogrel among the trial's diabetics compared with its nondiabetics. That's different from what was seen in the primary analysis of the overall population.

As previously published [3] and covered then by heartwire, prasugrel was associated with a 19% lower rate of CV death, nonfatal MI, or nonfatal stroke that was somewhat countered by a 32% rise in risk of major bleeding (excluding CABG-related bleeding per trial definition), compared with outcomes with clopidogrel, over a median follow-up of 14.5 months.

Coauthor Dr Elliott M Antman (Brigham and Women's Hospital) told heartwire that with the diabetes subanalysis as well as a separate one that compared the two antiplatelet agents solely in patients with ST-segment-elevation MI (STEMI), who also showed an increased net benefit from prasugrel [4], the message from TRITON-TIMI 38 remains as it was when the primary analysis was first released: the drug's potential for reduced ischemic risk is big compared with the small absolute risk of bleeding. "The relative benefit compared with the relative risk is what clinicians need to think about, and there is a profound treatment benefit with prasugrel."

Dr Robert A Harrington (Duke University, Durham, NC), who wasn't associated with TRITON-TIMI 38, said the overall trial suggests that in the clinical setting that was studied, prasugrel offers better efficacy than clopidogrel. "But that better efficacy comes at a price. What has to be figured out now is how to take advantage of that efficacy while having tolerable bleeding," he said.

"To me, the next series of analyses really needs to figure out, with the TRITON data, what is the risk for both ischemia and bleeding at any one time," Harrington said. "And then, can we construct a way of thinking about prasugrel, assuming it's approved, that will allow us to use it? And I think this [analysis in diabetics] adds to that story."

Of the trial's 13 608 randomized PCI-eligible patients with moderate- to high-risk STEMI, unstable angina, or non-ST-segment-elevation MI (NSTEMI), 3146 had diabetes; about one-fourth of those were on insulin. Prasugrel was given as a 60-mg loading dose plus maintenance at 10 mg/day, and clopidogrel as 300 mg initially plus 75 mg/day maintenance. Everyone received aspirin.

CV death or nonfatal MI or stroke occurred in 9.9% of the 10 462 patients without diabetes, 13.4% of the 2370 diabetics not on insulin, and 18.3% of the 776 insulin-treated diabetics (p for trend <0.0001). The adjusted hazard ratio (HR) for the composite end point for prasugrel vs clopidogrel was 0.74 for diabetics not on insulin and 0.63 among diabetics on insulin (p=0.009 for both subgroups).

In addition, prasugrel was associated with significantly reduced risks of the composite end point and of MI in both nondiabetics (p=0.02 and p=0.006, respectively) and diabetics (p<0.001 for both end points), with significantly increased risks of major hemorrhage (p=0.02) and of major or minor hemorrhage (p=0.006) in nondiabetics and with comparable risks of those two bleeding end points in diabetics, all compared with clopidogrel. Taken together, according to the authors, the findings point to an increased net benefit—that is, a significant reduction in the composite of all-cause mortality or nonfatal MI, stroke, or major bleeding—for prasugrel among the diabetics (p=0.001) but not nondiabetics.

Antman said the results of the two TRITON-TIMI 38 subanalyses are consistent with a state of increased platelet activation in diabetics and patients experiencing STEMI. Greater platelet aggregation means a lower bleed risk, "so if you're using a more powerful antiplatelet regimen, such as prasugrel compared with clopidogrel, in a condition where the platelets are more aggregable to begin with, the signal of increased bleeding with prasugrel vs clopidogrel is less likely to be observed."