Dr Elliott Antman

The trial showed significantly reduced ischemic events compared with clopidogrel, but at the expense of an increase in major (and fatal) bleeding in ACS patients scheduled for PCI. Dr Elliott Antman (Brigham and Women's Hospital, Boston, MA), who presented the results at a late-breaking clinical-trial session, said: "We have shown that greater inhibition of platelet aggregation than that achieved with clopidogrel is associated with a greater reduction in ischemic events. That is an important principle. Yes, there is an increase in bleeding, but we believe that, with exclusion of patients with prior stroke/[transient ischemic attack] TIA and dose reduction in the elderly and those with low body weight, prasugrel will have an important role."

Dr Sidney Smith (University of North Carolina, Chapel Hill), chair of the ACC/AHA taskforce for practice guidelines, was positive about the study. He commented to heartwire: "This is a very interesting trial. I am impressed with the improvement in outcomes. Yes, there is a trade-off with bleeding, but we have to look to see whether we can firmly identify a group that gains significant benefit without so much risk. If that can be done, I am positive about the potential for prasugrel in patient care."

Dr Raymond Gibbons (Mayo Clinic, Rochester, MN) had a similar view: "The data are impressive. I don't think the bleeding should put people off this drug. At the end of the day, for every 150 patients treated with prasugrel, there were three MIs prevented at the cost of one major bleed. I think that's worth it. But we always look for refinements to reduce risk further, and there appear to be areas here where this can be done. There is always an evolution in finding the right dose of new drugs for all patient groups. When we first used unfractionated heparin we tended to overdose people. You have to learn from the data."

Dr Robert Harrington (Duke University, Durham, NC) was a little more cautious: "TRITON is a big advance, in that we now know that better inhibition of ADP results in better ischemic outcomes. But the bleeding is a concern and will require more review of the data," he said. Dr Michael Lincoff (Cleveland Clinic, OH) commented: "I think there could well be role for prasugrel in clinical practice, but it won't be for everyone with ACS. In selected patients, it will be a valuable addition to our antithrombotic armamentarium."

But whether or not the data from TRITON-TIMI 38 alone will actually be sufficient for approval of prasugrel appears uncertain, with the identification of factors that will minimize the bleeding risk dependent to a large extent on post hoc subgroup analyses.

Many experts suggested that approval of prasugrel may depend on whether the FDA feels that the subgroups that may not benefit or may be harmed by the drug can be clearly defined. Dr Gordon Tomaselli (Johns Hopkins University, Baltimore, MD), chair of the AHA Scientific Sessions program committee, said: "I cannot speak for the FDA, but I would probably want some more assurance that these post hoc subgroup analyses are true. I suspect there will have to be more trials, even if they are just small confirmatory studies."

Other concerns about the trial highlighted by various experts include the fact that the TRITON trial did not use an optimum loading dose of clopidogrel, the fact that most patients had already had their coronary anatomy defined before treatment, which is not in line with clinical practice, and the suggestion that most of the benefit of prasugrel appeared to occur early on but the bleeding accumulated over time, giving rise to the idea that prasugrel maybe more suited to acute rather than maintenance therapy.

In his presentation, Antman drew attention to a prespecified analysis of net clinical benefit encompassing all-cause mortality, MI, stroke, and major bleeding, which showed a significant benefit for prasugrel.

Antman noted that in exploratory post hoc analyses, certain patient groups could be identified in which prasugrel may not be recommended or in whom a reduced dose would be appropriate, and the remaining trial population, which encompasses the vast majority of patients, gained strong benefit from prasugrel.

Antman commented: "In patients with a history of stroke/TIA, who made up 4% of this study, the bleeding risk appears to outweigh the benefits of prasugrel vs clopidogrel. In patients older than 75 or less than 60 kg in weight, who accounted for 16% of the population enrolled, the bleeding risks of prasugrel balance the benefits, and it may well be that a reduction in the maintenance dose may be required for these patients. In the remaining 80% of the population enrolled in the study, there was a strong net clinical benefit of prasugrel. So we believe that 96% of the population enrolled in this study should be able to benefit from this drug, albeit with some dosage reductions in a minority."

He continued: "There were 24 excess cardiovascular deaths in the clopidogrel group and 16 excess bleeding deaths in the prasugrel group. There was one fewer noncardiovascular death in the prasugrel group, so overall there were nine fewer deaths with prasugrel than there were with clopidogrel. This was not significant, but there is a suggestion of a trend toward benefit in terms of mortality. Then, when you factor in the other benefits in terms of reduced MI and revascularizations, and when you exclude the subgroups who may need dose reduction or who should not take the drug, there is a substantial overall benefit in a very large proportion of patients treated."

Antman noted that there is a pharmacokinetic substudy under way to investigate whether a reduced maintenance dose would be appropriate for older or smaller patients and what that dose might be. This study is estimating the concentration of active metabolite in around 1200 to 1500 patients with a broad range of ages and body weights and predicting what maintenance dose would be reasonable in various patient groups. "These results are not yet available, but they will be shortly, and I feel confident that they will provide information on the 16% of patients who may need a modified dose. I don't know if these data will be enough for approval of the drug, but the FDA specifically requested that this substudy be incorporated into the trial, so I would think it likely that they would take notice of its results," he added.

Antman also presented data showing that in diabetic patients, a well-known group at high risk of ischemic events, prasugrel appeared particularly beneficial. "In the 3100 diabetics in TRITON, there was a 30% reduction in the primary ischemic end point, and we did not see a difference in bleeding risk in this group. So in this extremely high-risk group who are known to have very sticky platelets, there was a very strong benefit."

Among the concerns voiced to heartwire by various experts was the fact that the TRITON trial did not use an optimum loading dose of clopidogrel. TRITON-TIMI 38 used a 300-mg loading dose for clopidogrel, whereas several studies have suggested a 600-mg dose may be better.

Addressing this, Antman noted that a laboratory study recently completed but not yet published (PRINCIPLE-TIMI 44) has evaluated the antiplatelet effects of the prasugrel 60-mg loading dose used in TRITON vs a 600-mg loading dose of clopidogrel in patients undergoing elective PCI. "This showed higher levels of platelet inhibition with prasugrel 60 mg after 30 minutes than with clopidogrel 600 mg even after six hours, so these data suggest that reduced ischemic outcomes will still be seen with prasugrel even if the clopidogrel loading dose had been doubled." Antman said.

Another point raised by some has been the fact that patients were not pretreated in this study and that could have disadvantaged clopidogrel, which takes time to reach optimum levels. On this issue, Dr Paul Armstrong (University of Alberta, Edmonton) told heartwire: "The study event rate was driven by periprocedural MI, which was ensured by lack of pretreatment with clopidogrel." He added: "The relevance of this study to current practice is marginal, in my view. TRITON for the moment is a bridge too far for me to cross, but I remain open to its possible application with better understanding of right dosage and duration of therapy."

Dr Keith Fox (University of Edinburgh, Scotland) also focused on this issue. "By design, the study drug was only given after angio, and this is a major issue for clopidogrel because of the delay to convert to the active metabolite. Hence, prasugrel was effectively going against negligible concentrations of the active component of clopidogrel at the time of the PCI. Note the very early event difference (in the first few hours) in favor of prasugrel, and the very early difference in stent thrombosis," he commented to heartwire.

Official discussant of the trial at the late-breaking trial session, Dr Eric Topol (Scripps Translational Science Institute, La Jolla, CA), said the requirement to know the coronary anatomy before starting the drugs made the design of the TRITON study "somewhat artificial" in terms of clinical practice. Antman responded that as this was a double-blind trial, both drugs had to be given at the same time. "In practice, the fact that pretreatment is not necessary with prasugrel is a big advantage. It means we can wait until we are in the cath lab to give the drug. And then the patients found to need surgery will not be given the drug at all." He noted, however, that this trial did not address the issue of upstream treatment in lower-risk medically managed patients.

This issue is being investigated in another trial, TRILOGY ACS, which is just about to start. Dr Magnus Ohman (Duke Clinical research Institute), who is involved in TRILOGY, said that this trial addresses some of the issues raised by TRITON in that it will use a reduced dose in certain subgroups. While acknowledging that there is a problem with using prasugrel in patients who end up needing surgery, Ohman said these patients just have to wait five days for the drug to clear.

Dr Shamir Mehta (McMaster University, Hamilton, ON) also highlighted the fact that the large proportion of ACS patients who did not have PCI were not studied in TRITON. He commented to heartwire: "We know from CURE that early administration of clopidogrel benefits all types of ACS patients, including those who receive a PCI and those who do not." Mehta suggested that an alternative strategy for better platelet inhibition could be to use a higher loading and maintenance dose of clopidogrel, which probably will not have the same bleeding issues as prasugrel. This is being tested in the CURRENT trial, where 14 000 ACS patients managed with an early invasive strategy receive 600 mg of clopidogrel followed by a maintenance dose of 150 mg daily for one week and then 75 mg daily for the duration of the study.

Several observers pointed out that the ischemic-outcome curves in TRITON appeared to diverge early, suggesting most of the benefit of prasugrel occurs early on. Dr James Ferguson (Texas Heart Institute, Houston) commented: "What did we see in TRITON? Primarily early clinical benefit (with no substantial increase in early bleeding) and primarily late bleeding risk (with underwhelming divergence of the clinical-event curves from 30 days onward). The better pharmacodynamics of prasugrel seem to win acutely, but the greater potency gives you more long-term bleeding complications and little apparent extended clinical benefit."

Another to voice this opinion was Dr Steven Steinhubl (University of Kentucky, Lexington). He commented to heartwire: "A huge proportion of the benefit seems early. Maybe a better future treatment option will be a short-acting P2Y12 inhibitor like cangrelor (The Medicines Company/AstraZeneca) in the hospital to prevent periprocedural events followed by clopidogrel (especially when it is generic)."