In a recent issue of The American Journal of Cardiology ( AJC ), an editorial appears that comments on mortality in the Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition (TRITON)–Thrombolysis In Myocardial Infarction (TIMI) 38. Complete overall mortality statistics were published in the primary TRITON–TIMI 38 report and reported publicly in the final integrated United States Food and Drug Administration (FDA) review with identical results. There was no significant effect of prasugrel on mortality in acute coronary syndromes, nor was there an associated excess. In an editorial about the prasugrel review in The New England Journal of Medicine , FDA deputy director Dr. Ellis Unger noted, “There was no significant difference in overall mortality, although there was a trend in favor of prasugrel.”
The author of this AJC editorial quotes an exploratory analysis (contained within the original FDA review) by the FDA’s secondary reviewer, in which site-reported “first-event” deaths are listed. As described, in this analysis, it appears that if a subject first had a nonfatal myocardial infarction (MI) or stroke (the additional components of the primary composite end point), the death of that subject would not be counted. Because prasugrel reduced nonfatal MI and subsequent death after MI, a first-event death analysis, compared to counting all deaths, systematically must favor the clopidogrel arm, in which there were more nonfatal events. If a subject had a nonfatal MI, he or she would no longer be eligible to have a first-event death.
Indeed, 82 subjects who died during TRITON–TIMI 38 (26 randomized to prasugrel and 56 randomized to clopidogrel) are not included in the analysis forming the basis of the AJC editorial. As would be expected, the clinical events committee–adjudicated death rates are identical to the site-reported death rates when all deaths reported are considered.
The complete accounting of deaths in TRITON–TIMI 38 is listed in Table 1 .
