THE P2Y12 BLOCKERS PLAY A CRITICAL ROLE IN PREventing coronary thrombosis and are now widely used along with aspirin in patients undergoing coronary stent placement. Despite proven clinical efficacy, the antiplatelet effect of clopidogrel, the most widely used P2Y12 inhibitor, is highly variable as determined by ex vivo platelet function assays. Additionally, studies have demonstrated that high on-treatment platelet reactivity is a major cardiovascular risk factor for subsequent thrombotic events in patients treated with stents. This has led some to propose that patients who have received stents and were treated with clopidogrel should undergo on-treatment platelet function testing. Yet the degree to which such testing should guide therapy is not yet known. In this issue of JAMA, Price and colleagues report the results of the Gauging Responsiveness with A VerifyNow assay—Impact on Thrombosis And Safety (GRAVITAS) trial, the first large randomized, prospective trial to determine whether personalization of antiplatelet therapy based on platelet function testing actually improves 6-month clinical outcomes among patients treated with stents. In the GRAVITAS trial, patients treated with clopidogrel were screened for high on-treatment platelet reactivity 12 to 24 hours after receiving drug-eluting stents and those with high on-treatment reactivity were randomly assigned to receive either high-dose (600 mg initial dose and 150 mg/d thereafter) or standard-dose clopidogrel (no additional load, 75 mg/d). In addition, a random sample of patients without high on-treatment reactivity were observed while being treated with standard-dose clopidogrel. The major findings of the GRAVITAS trial were that the rate of ischemic events (cardiovascular death, nonfatal myocardial infarction, or stent thrombosis) was low (2.3%) in randomized patients treated with highand standard-dose clopidogrel; that bleeding risks did not differ by clopidogrel dose; that high-dose clopidogrel therapy was partially effective in eliminating high on-treatment platelet reactivity (40% prevalence at 30 days); and that in the secondary observational comparison, patients with high on-treatment reactivity treated with standard-dose clopidogrel had increased event rates compared with those who did not have high on-treatment platelet reactivity. There are several possible reasons personalizing therapy using high-dose clopidogrel was not effective. Among these are that high-dose clopidogrel had a limited effect in overcoming high platelet reactivity. The overall trial event rates were low, and the cutpoint for defining high on-treatment platelet reactivity may have been too high. The low event rate in GRAVITAS may, in part, be explained by the absence of periprocedural myocardial infarction in the primary end point as has been used in recent observational studies with higher event rates. Given the low event rates in GRAVITAS, only an extremely effective treatment would have had any chance to significantly influence outcomes. In addition, several important questions cannot be answered based on the data presented in the report by Price et al, including whether the platelet reactivity before the ischemic event was higher than the platelet reactivity in patients without ischemic events. And, whether ischemic events were clustered among patients whose platelet reactivity was lower than the cutpoint used to define high on-treatment reactivity. Answers to these questions will be useful in addressing the hypothesis that ischemic event occurrence increases markedly above a critical cutpoint. However, the possibility also exists that ischemic events occurred in the presence of low platelet reactivity in GRAVITAS. The latter finding would suggest that high on-treatment platelet reactivity identified early after stenting is only a marker of risk
Read full abstract