A Better Antiplatelet Drug?
Abstract & Commentary
By Jonathan Abrams, MD, Professor of Medicine, Division of Cardiology, University of New Mexico, Albuquerque. Dr. Abrams serves on the speaker's bureau for Merck, Pfizer, and Parke-Davis
Sources: Wiviott SD, et al. Prasugrel versus clopidogrel in patients with acute coronary syndromes. N Engl J Med. 2007;357;2001-2015; Bhatt DL. Intensifying platelet inhibition navigating between Scylla and Charybdis. N Engl J Med. 2007;357:2078-2081.
Treatment of cardiovascular disease continues to emphasize pharmacologic anti-platelet therapy. The platelet is the target of many medications, as thrombosis is an extremely common problem, resulting in acute coronary syndromes, cerebral vascular events, and conditions such as atrial fibrillation. Many efforts are underway to develop reliable and effective antiplatelet agents that have a minimal risk for bleeding. Prasugrel (Pras), not yet approved by the FDA, is a potent thienopyridine that acts as a prodrug similar to clopidogrel; both require conversion to an active metabolite before binding to the platelet receptor P2Y12. The TRITON-TIMI 38 trial is a phase III study in ACS patients undergoing scheduled percutaneous coronary intervention (PCI); almost all had an angioplasty; a small percentage had bypass surgery. Pras was compared to clopidogrel in this large randomized study, testing the hypothesis whether inhibition of adenosine diphosphate induced platelet aggregation with Pras would be more beneficial and/or safer than clopidogrel. The study also was designed to establish dose response relationships between the two drugs, particularly in terms of variability of effectiveness.
Wiviott and colleagues enrolled 13,608 patients with ACS who were scheduled to receive PCI or had undergone cardiac catheterization. Of the enrolled patients, three-quarters had non-ST segment elevation MI or moderate-to-high risk unstable angina; 25% had an ST elevation MI (3500 patients). Study drug therapy included a loading dose of Pras 60 mg, or 300 mg of clopidogrel in those who had a PCI. Decisions about cardiac catheterization and angioplasty were left to the discretion of the physician. Patients who had PCI received maintenance dosing with Pras or clopidogrel, 10 and 75 mg/day, respectively. All patients were on aspirin. Mean study interval was 6-15 months. The primary efficacy end point was a composite of death from CV causes, nonfatal MI, or nonfatal stroke during follow-up. Secondary end points at one month and 90 days included the primary composite, as well as death from CV causes, nonfatal MI, urgent target vessel revascularization, stent thrombosis, and a composite of multiple cardiovascular events. TIMI bleeding rates were assessed in all patients; major bleeding had to be unrelated to coronary bypass surgery. Study centers included 707 sites in 30 countries, with treatment between November 2004 and January 2006; the mean duration of therapy was 14.5 months. Half the patients received at least one drug- eluting stent, and virtually all patients had a PCI at the time of randomization.
Results: There was a major reduction in risk of the primary end point in the unstable angina/NSTEMI group in the Pras cohort of 18% (P = 0.002). In the ST elevation cohort, there was a 21% reduction in risk in Pras patients. Overall, 12% of the clopidogrel patients had a primary end point compared to 9.9% in the Pras group, resulting in a 19% reduction, P <0.001. Similar findings were noted at 3 days and persisted throughout follow-up. From day three to the completion of the study, the primary end point had occurred in 6.9% of clopidogrel patients vs 5.6% of Pras subjects, hazard ratio 0.80, P = 0.003. The major component of benefit was acute myocardial infarction, reduced by 2.3% in Pras Subjects (7.4%) vs clopidogrel (9.7%). There was no difference in the rate of stroke or cardiovascular death. Also, Pras showed superior efficacy in major prespecified subgroups, without significant interactions. Benefit with Pras, with regard to the primary end point, was found both with and without glycoprotein IIB-3A receptor antagonist therapy (21% reduction in risk). Diabetics tended to have a greater benefit with Pras, 12.2% patients vs 17% in clopidogrel patients. In 10,500 patients without diabetes, the primary end point was greater in the clopidogrel group by 14%, P = 0.02. Significant reductions with Pras were seen for all the prespecified secondary end points, including early target vessel revascularization. Pras was beneficial for cardiovascular death, non-fatal MI, non-fatal stroke, rehospitalization for ischemia, and stent thrombosis. Bare metal stent subjects had a 52% reduction in hazard ratio P < 0.001 compared to patients who had at least one DES, hazard ratio of 0.43, P < 0.001.
The safety profile of Pras was not as good as clopidogrel, with greater bleeding risk. Major bleeding was 1.8% with clopidogrel vs 2.4% Pras. Life-threatening bleeds in Pras patients were also higher, 1.4% vs 0.9%, and fatal bleeding was greater in Pras patients, 0.4% vs 0.1%. Instrumentation-related bleeding and non-fatal, life-threatening bleeding were greater with Pras, as were all major and minor hemorrhages. Nevertheless, Wiviott et al concluded that there was a net clinical benefit favoring Pras (13.9% vs 12.2%), with a hazard ratio of 0.87, P = 0.004. Intracranial hemorrhage, or procedural bleeds, was greater in clopidogrel patients, 2.4 oz vs 2.2%. Wiviott et al state "as a result of the discordance between the efficacy result and the safety results, extensive subgroup analyses were carried out." Three groups of patients were noted to have a different risk profile than the main cohort; these include previous stroke or TIA; patients greater than 75years; and patients weighing less than 60 Kg. None of these groups had benefit from Pras, and clopidogrel patients had lower rates of complications. There was "... significant harm from Pras among patients with a history of cardiovascular events." However, in individuals who had none of these three risk factors, the risk gradient favored Pras by 26%, P < 0.001. Thus, Wiviott et al found a "substantially favorable net clinical benefit from the use of Pras."
Wiviott et al conclude that a 60 mg loading dose reduced multiple cardiovascular outcomes in patients with ACS, with a "significant 2.2% absolute reduction in the rate of the primary efficacy end point of fatal and nonfatal cardiovascular causes and nonfatal MI and stroke. Rates of ischemic events were reduced by 24%, with an absolute reduction of 2.3%.
The Triton-Timi 38 data are comparable to a number of studies indicating a significant reduction in ischemic events with aspirin, thienopyridines, and glycoprotein 2B-3A receptor antagonists, all of which have also demonstrated increased bleeding rates. In the CURE trial, the odds of major bleeding was increased by 38%, and in the Antithrombotic Trials Collaboration, aspirin increased major bleeding risk by 60%. In the present study, major hemorrhage was increased by 32% with Pras and there was "an increase in the rate of life threatening bleeding with Pras, including an increase in fatal major hemorrhage." Subgroup analyses of stroke patients, the elderly, and individuals weighing less than 60 Kg are of interest and require further exploration. It is noteworthy that patients with a prior stroke clearly did worse and had more frequent intracranial bleeding. Nevertheless, the "large majority of patients without any of these risk factors had a significant net clinical benefit with the Pras regiment study," with a hazard ratio of 0.80, P < 0.001.
Wiviott et al underscore prior data that with Pras, comparable dosage has been shown to "generate higher and more consistent levels of active metabolite than treatment with approved doses of clopidogrel," resulting in higher levels of mean inhibition of platelet aggregation, lower inter-patient variability, and fewer patients considered to have hyporesponsiveness." Wiviott et al note a "more rapid onset of an intra-antiplatelet effect with Pras" and a continued benefit of platelet inhibition with a higher than standard loading dose of clopidogrel. Pras 60 mg "has been shown to result in greater inhibition of platelet aggregation than use of clopidogrel (600/75 mg) in patients with chronic CAD." The data confirm a significantly greater inhibition of platelet aggregation with Pras.
The final conclusion is that the hypothesis of enhanced oral P2Y12 inhibition has been supported by doses used in this study, however, with an increase in major bleeding. Wiviott et al do not provide an algorithm for safe use of Pras or Clopidogrel and suggest that "clinicians need to weigh the benefits and risk of intensive inhibition of platelet aggregation."
An editorial by Bhatt supports the belief that TRITON-TIMI38 confirms the concept that increased platelet inhibition is associated with greater suppression of clinical events in ischemic heart disease patients. He points out that this study had more fatal events from anti-platelet therapy, more than has been seen in the other large trials. He asks, "how then, is the clinician to decide which agent to use? There is not one easy answer to this question."
This very large trial appears to be a role model of outstanding clinical research, with careful description of patients, close patient follow-up, and consistent results between the two antiplatelet drugs. At the time of this writing, it is not known whether Pras will be approved for marketing; but if so, it clearly provides the cardiologist with another arrow in the quiver for treatment of acute coronary syndrome. It is clear that patients with known or suspected cerebral vascular disease should not be given Pras, and individuals with a small body mass also should be guarded against receiving standard doses of either clopidogrel or Pras. Diabetics did well in this study, but elderly patients did not. Wiviott et al's suggestion that patients should be carefully selected before choosing a thienopyridine seems a bit unrealistic, as Pras is not available, and that sufficient discussion needs to take place regarding guidelines for use of more potent antiplatelet therapy. Certainly one consideration should be the dose of Pras. If one looks at the Kaplan-Meier curve, it is immediately clear that the major events occurred extremely early in the trial, with a very rapid upslope for both Clopidogrel and Pras. It does seem appropriate, then, that studies with Pras at a lower loading dose than 60 mg and a lower daily dose should be tested. While it is clear that Pras is more potent than clopidogrel, these are rather arbitrary doses, including the 60 mg Pras loading dose which has not been tested adequately for safety and efficacy, other than in this study. The issue is whether lowering the dose of Pras will maintain its outstanding vascular protective actions, with less bleeding and no loss of efficacy. The loading dose of clopidogrel of 300 mg is also a potential culprit in that many events were very early and presumably occurred in and around the time of peak concentrations of the drug. The same can be said about Pras, although we have no prior clinical data to help us analyze this phenomenon.
Finally, it should be recognized that the results of this trial are very robust, with respect to reduction of hard cardiovascular end points when the more potent thienopyridine is used. This is not an academic matter; the benefits from Pras are consistent, but bleeding is a big problem. It, therefore, behooves clinicians and investigators to untangle the dosing knot as rapidly as possible.