Prasugrel in UA/NSTEMI Patients: The Implications of TRILOGY ACS
How did non-revascularized ACS patients do on prasugrel versus clopidogrel? Some surprising and hypothesis-generating results.
Performed at 966 sites in 52 countries, TRILOGY ACS (TaRgeted platelet Inhibition to cLarify the Optimal strateGy to medicallY manage Acute Coronary Syndromes) was a double-blind, randomized trial in which the effect of prasugrel was compared with that of clopidogrel for up to 30 months of treatment in acute coronary syndrome patients with unstable angina or non-ST elevation myocardial infarction (non-STEMI) managed without revascularization. This patient population has not been exclusively studied before in a randomized trial.
All study subjects took aspirin, the clopidogrel dose was 75 mg daily for all patients, and the typical prasugrel dose of 10 mg daily was reduced to 5 mg daily for patients < 75 years of age weighing <60 kilograms and those ≥75 years of age. Rates of TIMI major bleeding events did not differ significantly between the prasugrel plus aspirin and clopidogrel plus aspirin treatment groups in patients <75 years of age (n=7,243) or in the overall study population (including an additional group of patients ≥75 years of age, n=2,083).
The primary end point of the trial was cardiovascular death, myocardial infarction, or stroke. At 30 months, 13.9% of prasugrel patients vs. 16.0% of clopidogrel patients <75 years of age, the primary analysis population, experienced the composite primary endpoint. This outcome was not statistically significant (P=0.21).
A post-hoc exploratory analysis observed a trend for a lower risk in MI, stroke and death among patients treated with prasugrel beyond one year.
Dr. Roe, why did TRILOGY ACS look at only non-revascularized acute coronary syndrome (ACS) patients?
First, ACS research has traditionally been focused on the cardiac catheterization laboratory. Prior ACS studies evaluating patients with UA/NSTEMI have primarily evaluated medications used before or during PCI in the cath lab. However, it has been shown that UA/NSTEMI patients either do not undergo a cath (more likely in certain parts of the world) or they receive a cath, but not go on to subsequent revascularization. We wanted to evaluate those patients because they have never been studied directly before and they are somewhat of an unknown quantity in this field. The second reason was that the drug being tested, prasugrel, had previously been shown to benefit ACS patients getting a percutaneous coronary intervention (PCI), but the non-revascularized group had never really been tested before. TRILOGY ACS was a novel study in a novel group of patients.
Can you tell us more about the patient population?
There has been some criticism of the trial that these were low-risk patients. We know that they were not low risk — these patients were medium to high risk, based on the trial inclusion criteria, baseline characteristics of the patients, and their GRACE (Global Registry of Acute Coronary Events) risk scores. The challenge is that in real-world practice, despite what most people think, the highest risk ACS patients do not get angiography and/or revascularization, because there is concern about the risk of these procedures for causing contrast-induced nephropathy, bleeding, or other problems. There also is a group of patients that may have a cath done, but not proceed to revascularization, because technically revascularization is not possible or there may be other reasons why revascularization is not performed. We wanted to look at the whole spectrum of these medically managed patients. Of the patients enrolled in TRILOGY ACS, about two-thirds were patients with a non-STEMI and one-third were patients with unstable angina. The unstable angina group had to have a significant degree of ST-depression on the ECG, which by itself puts the unstable angina patients in at least in the moderate risk category. The high event rates observed also speak to the fact that the patient population studied was not low risk.
Prasugrel was not superior to clopidogrel in the first 12 months, a different result than TRITON-TIMI 38 (which looked at PCI patients). Why do you think this was the result?
We believe these findings can be explained by a mixture of things. We have not put our finger on the exact reasons why the trial showed those findings, but I will go through our list of what we think may be involved. The first is that when a patient has a stent implanted, especially in the ACS setting, we know that their platelet reactivity increases as a result of stenting. It is already increased because the patient has an ACS, but then a stent is placed, rupturing the plaque further and activating platelets further. These patients are uniquely susceptible to more potent platelet inhibitors such as prasugrel, especially early on, because many of the early events are related to the stent itself, with stent thrombosis or a procedure-related MI. The second is that in the TRITON trial, the drug was started before PCI began. In our study, most patients were randomized about 4-5 days after presentation and the majority of patients were treated with clopidogrel for a few days before they were randomized. In that setting, it may be that intensified platelet inhibition is not going to have an early effect, because these patients do not have augmentation of their platelet reactivity by stenting. They may have “quieted down” somewhat by getting a few days of clopidogrel before randomization. The third is that there may also be more of a natural disease progression, meaning that we need to wait until the patient has other events occurring later in the timeline of the disease process. These later events may be more susceptible to a more potent drug like prasugrel, and thus we see the potential benefit, as TRILOGY ACS showed, only after 12 months.
Those are our theories; we cannot prove them, but we do feel there was more potent platelet inhibition in prasugrel patients, because of increases in minor and moderate bleeding.
All patients received clopidogrel initially.
Yes. The challenge was that we knew it would take a few days, for the most part, to determine that someone would be medically managed. It is usually not a decision made in the first 24 hours. We also did not want to have a high number of patients cross over, meaning patients initially received medical management, but later in the hospitalization, needed revascularization. This meant that we had to wait for four to five days for that decision to be made. Based on practice guidelines and the evidence around using clopidogrel, clearly patients could not wait for days and receive no treatment before randomization. That is why the trial was designed the way it was, and perhaps the results would have been different if we had been able to start prasugrel in the first 24 hours. We just don’t know that.
How did age impact trial results?
Based on the experience of the TRITON trial, in TRILOGY ACS, those patients at least 75 years of age or older and randomized into the trial were treated with either 5mg of prasugrel or 75 mg of clopidogrel. We have not yet presented the specific data on this group of patients, but we did show the overall data from the trial in the NEJM paper. The data on the elderly patients will be forthcoming soon, but it should be noted that we did not power the trial in the elderly population for efficacy purposes. We decided to explore the lower prasugrel dose in at least 2,000 elderly patients to provide adequate exposure. This was a very important group to look at, because we do not really know what to do in the elderly population, and we were concerned about the findings in the TRITON trial, where the 10 mg prasugrel dose led to an increase in fatal, life-threatening bleeding in elderly patients. Thus, in TRILOGY ACS, we made important dose adjustments to prasugrel to address prior safety concerns and we evaluated a unique patient population, including one of the largest evaluation of elderly ACS patients in a dedicated clinical trial.
Could you clarify the bleeding scales used in the trial and how patients fared?
We assessed bleeding with two different scales: the GUSTO bleeding scale and the TIMI bleeding scale. The GUSTO bleeding scale has severe, life-threatening bleeding and moderate bleeding, and in that scale, severe, life-threatening bleeding was <1%, with no difference by treatment. When we added moderate bleeding to severe, life-threatening bleeding, then there was a trend to an increase with prasugrel. GUSTO severe, life-threatening bleeding is hemodynamically significant bleeding that is either fatal, intercranial or causes a drop in blood pressure, and moderate bleeding is bleeding that requires a transfusion. TIMI major bleeding is related to the location as well as the amount of hemoglobin drop, and minor bleeding is a lower hemoglobin drop, and this bleeding scale also accounts for transfusion, etc. With the TIMI scale, we showed no difference in TIMI major bleeding by treatment, but when we added TIMI minor bleeding together with major bleeding, then we saw an increase in the prasugrel group. We believe this finding demonstrates more potent platelet inhibition with prasugrel, but not so much that it caused an increase in severe bleeding as was seen in the TRITON study. This finding also speaks to the fact that we made adjustments in the dose, lowering the prasugrel dose for younger patients who were at low body weight and for all patients over the age of 75.
Can you describe the post hoc analysis and why it was performed?
The divergence of the event curves after 12 months was an unexpected finding, but was surprising and intriguing. We saw it in the primary endpoint, which was cardiovascular death, MI or stroke, and we saw it with each of the component endpoints: cardiovascular death, MI by itself, and stroke by itself. We didn’t anticipate this finding, so in a post hoc analysis that was not pre-specified, we looked at the landmark timeframe of 12 months, and the risk of events by treatment before and after 12 months. In each case, with the primary endpoint and each of the three component endpoints, we found no difference by treatment before 12 months, yet after 12 months, there did appear to be evidence of a treatment effect. Sometimes it was a trend, never a significant p-value, but certainly consistent across all the endpoints. We think it is an authentic finding. Many have asked us, why didn’t you have the trial go on for longer than two and a half years so this could be explored further? Well, there was no way we knew this was going to happen. The DSMB was meeting, but they never saw data at the very end of the trial, when they would have picked this finding up, and we never charged them with actually looking at efficacy and giving us recommendations on the duration; we charged the DSMB only with monitoring trial safety. While this was a very interesting finding, it was not what the trial set out to do at the beginning, so it is not definitive, but it is suggestive of a late treatment effect. Further studies are needed.
Did patients receiving clopidogrel undergo any platelet function testing?
We did have a large platelet function sub study in the trial for patients on both types of treatment, about one-third of the patients who are enrolled. We are going to present those data at the American Heart Association meeting in early November. I think they are going to be very intriguing data, but I can’t speak to them now, obviously. These people had measurements of platelet function with the VerifyNow device, starting at baseline and throughout the study, for the duration of follow-up — in some cases, up to two and a half years.
What about the issue of cost, since clopidogrel is now generic in the U.S.?
We have a cost effectiveness study as part of TRILOGY ACS, and one of our colleagues here, Dr. Dan Mark at the Duke Clinical Research Institute, has led that study. Results are still forthcoming.
A small percentage of patients <75 years of age underwent revascularization.
We had <8% of patients undergo revascularization, which is what we had hoped, because we wanted them to stay medically managed. What it has shown is if the physician makes a decision not to perform revascularization of an ACS patient similar to the population we studied, then it is very unlikely they will undergo revascularization during the follow-up period; meaning, there are reasons the patient didn’t get it in the first place, and those reasons probably don’t change that much.
You make the point that medically managed patients have not received the same research attention as revascularized patients. Will that change?
I hope so. It has been very traditional that we study ACS patients in the cath lab who are getting revascularized, because that has been the model. It has also been something that is much easier to do. It was difficult to enroll in this trial, because sites were not aligned in a fashion to identify these patients, and follow them during the hospitalization before and after the cath to find out who was being medically managed, so it was more work. I do hope we will study these patients in the future, either in dedicated trials or by making sure in future ACS trials to have a substantial component of patients who are medically managed without revascularization. This is the group of patients that deserves the most work to figure out what the right treatment strategies are.
Any final thoughts?
First, TRILOGY ACS was a global study, with 52 countries involved. It is applicable to care around the world. Second, the trial was the result of a strong academic collaboration with many people around the world. It is a model for how trials can be conducted effectively in partnership with funding organizations like the pharamaceutical industry. Third, it speaks to the fact that future studies in this area probably need to consider more than just one dose of a given therapy. There needs to be potential design adaption for different groups of patients who may require a higher or lower dosing strategy, like the elderly or low body weight patients. All those things are worthwhile concepts that have come out of TRILOGY ACS. We have learned a great deal. It is a novel study and we hope it has an influence on the conduct of future studies, so we can try to further improve the care of patients, but also design unique studies that don’t simply replicate previous trials.
Dr. Matthew Roe can be contacted at matthew.roe@dm.duke.edu.
References
- Roe MT, Armstrong PW, Fox KA, White HD, et al.; the TRILOGY ACS Investigators. Prasugrel versus clopidogrel for acute coronary syndromes without revascularization. N Engl J Med 2012 Oct 4;367(14):1297-1309.
- Gershlick A, Roe MT. TRILOGY-ACS with Dr. Matthew Roe. Theheart.org. Available online at https://www.theheart.org/podcast/video/trials-and-pis/matthew-roetrilogyacs.do. Accessed October 10, 2012.
- Califf R, Ohman M, Roe MT. TRILOGY ACS in review: A discussion with Rob Califf, Magnus Ohman, and Matthew Roe. Duke Clinical Research Institute Videos. Available online at https://www.dcri.org/research/videos/trilogy-acs-in-review-a-discussion-with-rob-califf-magnus-ohman-and-matthew-roe. Accessed October 10, 2012.
TRILOGY ACS Trial News
TRILOGY ACS Angiographic Cohort Presented at TCT 2012: Prasugrel May Reduce Cardiovascular Events Among Patients Managed Medically After an Angiogram for Acute Coronary Syndrome
(Miami, Fl. – October 24, 2012) – Results of the TRILOGY ACS trial Angiographic Cohort were presented by Stephen D. Wiviott, MD, on October 24th at the Transcatheter Cardiovascular Therapeutics (TCT) scientific symposium.
Angiography was not required for enrollment in TRILOGY ACS, but if performed, evidence of coronary disease had to be demonstrated (at least one lesion with >30% stenosis or prior percutaneous coronary intervention/coronary artery bypass graft surgery). Approximately 43% of the patients had angiography performed and coronary anatomy determined before being triaged to medical therapy. The results of this angiographic cohort were presented for the first time at TCT.
Prasugrel reduced cardiovascular events among patients presenting with an acute coronary syndrome who were managed medically after an angiogram was performed to determine coronary anatomy. Overall, in the TRILOGY ACS trial, prasugrel did not reduce cardiovascular events among patients managed medically for ACS. However, when treated with prasugrel compared to clopidogrel, patients who were triaged to medical therapy in the trial following angiography tended to have:
- Lower rates of the combined endpoint of cardiovascular disease/heart attack/stroke;
- Lower rates of heart attack, stroke alone, and recurrent ischemic events;
- Higher rates of bleeding.
“Though hypothesis-generating, these results are consistent with previous trials and suggest that when angiography is performed and coronary disease is confirmed, the benefits and risks of intensive antiplatelet therapy exist whether medical therapy or percutaneous coronary intervention is elected,” said Dr. Wiviott. Dr. Wiviott is a senior investigator with the TIMI Study Group, associate physician, Cardiovascular Division at Brigham and Women’s Hospital, and associate professor of Medicine at Harvard Medical School.
The trial was funded by Eli Lilly & Company and Daiichi Sankyo. Dr. Wiviott reported grant/research support from Eli Lilly & Company, AstraZeneca, Merck, and Eisai; and consulting fees/honoraria from Eli Lilly & Company, Daiichi Sankyo, AstraZeneca, BMS, Sanofi-Aventis, and Eisai.
