The CURE Trial

Abstracts & Commentary

Three major pharmacologic advances have occurred over the past several years for the therapy of individuals with the acute coronary syndromes (ACS) of unstable angina or non-ST-segment elevation myocardial infarction (MI). These agents include the IIb/IIIa receptor antagonists; low molecular weight heparin (LMWH); and the thienopyridine derivatives, ticlopidine, and clopidogrel. The latter are potent platelet inhibitors blocking adenosine modulated aggregation and subsequent thrombosis. The Clopidogrel in Unstable angina to prevent Recurrent Events (CURE) study sought to determine whether clopidogrel is beneficial in ACS. Some years ago, the clopidogrel vs. aspirin in patients at risk of ischemic events (CAPRIE) trial demonstrated modest efficacy of clopidogrel over aspirin in a mixed group of subjects that had peripheral vascular disease, stroke, or quiscent coronary artery disease. For some years, ticlopidine, and subsequently clopidogrel, has been aggressively used by interventional cardiologists in the United States and elsewhere during and after percutaneous coronary interventions (PCI), particularly with stent placement. The optimal duration of therapy with clopidogrel after PCI is unknown; it is currently in the range of 2-4 weeks.

CURE investigated long-term outcomes in a large cohort of individuals who were hospitalized with ACS, aiming for centers that performed PCI relatively infrequently. Patients were randomized to clopidogrel (loading dose of 300 mg/d, then 75 mg/d), or matching placebo for 3-12 months. The mean trial duration was 9 months. Low-dose aspirin was also mandated. Eligible subjects were hospitalized within 24 hours of symptoms with acute chest pain, but no ST elevation. Initially, patients older than 60 without ECG changes, but a history of CAD, were enrolled. However, because of low event rates after the first 3000 patients, the requirements changed to enroll only patients who had ischemic ECG changes or elevation in biochemical markers/cardiac enzymes. This was an international study, with patient recruitment between late 1998 and late 2000, in almost 500 centers in 28 countries. The major primary outcome was a composite of cardiovascular death, nonfatal MI, or stroke. The second primary outcome was the primary composite or refractory ischemia. The secondary outcomes included severe ischemia, revascularization, or heart failure. The safety outcomes were related to bleeding complications. A total of 12,549 individuals were randomized and had complete follow-up.

Results

The major primary outcome of death, nonfatal MI, or stroke was decreased in the clopidogrel cohort (9.3%) vs. placebo (11.4%); P = 0.001, risk reduction (RR) = 20%. The second primary outcome, which included refractory ischemia, was comparably decreased by clopidogrel, 16.5% vs. 18.8%; RR = 14%; P = 0.001. Rates of subsequent MI were decreased by the platelet inhibitor, RR = 32%; P = 0.007. Rehospitalization for unstable angina was not different between the groups; however, there was slightly less severe ischemia, 2.8% vs. 3.8%; r = 26%; P = 0.003, and recurrent angina, RR = 9%; P = 0.01. In the clopidogrel group, major events were reduced by beginning within hours of randomization, reaching a 21% RR at 30 days. Between 1 month and the study end, an average of 9 months, there was an additional 18% RR. Subgroup analyses demonstrated remarkable consistency of results across a number of subgroups, including patients receiving beta blockers, heparin, and/or ACE inhibitors. Individuals with prior revascularization fared better, with a primary event RR of 44%. The 12-month Kaplan-Meier curves remained parallel after the third month.

Side Effects

Major bleeding, as expected, was more common with clopidogrel than placebo, 3.7% vs. 2.7%; P = 0.001. However, there was no excess rate of fatal hemorrhage or hemorrhagic stroke. Moderate bleeding was almost twice as frequent with the platelet inhibitor.

The research group concluded that clopidogrel reduced the risk of MI and recurrent ischemia in patients with non-ST elevation ACS. They emphasize the absence of a routine policy for an early invasive approach in the CURE institutions; however, 44% of the entire cohort underwent angiography, and 38% had a revascularization procedure. They also observed that the benefits of clopidogrel were found in a wide variety of patients, including those at low, medium, and high risk for cardiovascular events, and as well as those on proven therapies for CV. The increased bleeding risk was felt to be acceptable. (The Clopidogrel in Unstable Angina to Prevent Recurrent Events Trial Investigators. N Engl J Med. 2001;345:494-502.)

A companion article, published simultaneously in the Lancet, assessed the role of clopidogrel and aspirin only in those subjects who underwent a percutaneous coronary intervention (PCI-CURE). A total of 2658 patients with ACS who received PCI were randomized to clopidogrel or placebo. PCI was performed at the discretion of the investigator after randomization; open label clopidogrel or ticlodopine was given for 2-4 weeks after coronary artery stenting, after which randomly assigned medication was resumed for 3-12 months. The primary outcome of this analysis study was a composite of cardiovascular death, MI, or urgent target lesion revascularization (TVR) at 30 days after the PCI. Major and minor bleeding was assessed. The 1313 clopidogrel and 1345 placebo subjects received a total of 1730 PCIs during the hospitalization and 9200 after discharge. The median number of days prior to PCI was 6 days, with an average duration of follow-up of 8 months. Baseline characteristics of the 2 groups were comparable. Stent use occurred in more than 80% of both cohorts. Unfractionated heparin was used in almost 90% of patients; the remainder received LMWH. Target vessels for PCI were well balanced for 2 groups with respect to proximal or distal location, and distribution in the coronary tree. A small number of sapheneous vein grafts were addressed. Forty-two percent presented with ST-segment depression. About a quarter of the patients received an open-label thienopyridine before PCI and 80% afterward for a median of 30 days.

There was less MI or severe ischemia in individuals receiving clopidogrel. In addition, the primary outcome of death, MI, or urgent TVR at 30 days was 6.4% placebo and 4.5% clopidogrel (30% RR, P = 0.03). The events from PCI to the end of follow-up were reduced, with a composite end point of 21.7% placebo and 18.3% clopidogrel, RR = 17%. Events prior to PCI were also decreased, 15% placebo vs. 12% clopidogrel, RR = 24%; P = 0.008. When all events were included, before and after PCI, there was a major difference in cardiovascular death or myocardial infarction, 12.6% placebo and 8.8% clopidogrel, RR = 31%; P = 0.002. There were little differences in bleeding rates between placebo and clopidogrel. Metha and colleagues concluded that clopidogrel reduced the risk of cardiovascular death or MI by approximately one-third, with benefits noted prior to PCI, within the first month after PCI, and long term. They speculate that the early benefits were due to mainly the reduction of thrombus mediated events, such as MI and urgent TVR. They suggest that the benefit at 30 days was to a great extent due to pretreatment with open-label clopidogrel prior to the PCI, and that the actual benefit was likely to be an underestimate of the true treatment effects, in that 25% of placebo patients received open-label thienopyridine prior to PCI. They emphasize that PCI-CURE supports the use of clopidogrel in centers with and without ready access to cardiac catheterization and revascularization; benefit was found whether PCI was done urgently or later during hospitalization. They believe that the PCI-CURE data support the long-term efficacy of dual antiplatelet therapy, with less events occurring after 30 days than in RCT registries of PCI patients. There was less IIb/IIIa antagonist use in the clopidogrel patients, consistent with a clinical benefit from oral antiplatelet therapy alone. They suggest that a strategy of clopidogrel and aspirin pre-PCI as well as long-term is appropriate regardless of whether glycoprotein IIb/IIIa inhibitors are used. They recommend initiating and maintaining clopidogrel therapy in all patients with non-ST segment elevation ACS in whom an invasive strategy with PCI is undertaken. (Mehta SR, et al. Lancet. 2001;358:527-533.)

Comment by Jonathan Abrams, MD

The CURE trial, and its subsidiary observational study, PCI-CURE, represent a significant advance in the treatment of ACS, and suggest that clopidogrel should be initiated upon diagnosis of an ACS and continued for at least 1 month, possibly longer. Furthermore, this drug, or ticlodipine, should not be withdrawn if an angioplasty, with or without stent placement, is likely to occur. These data support a large number of observations and randomized clinical trials indicating that the platelet is indeed the "Darth Vader" of the coronary vascular tree in ACS without ST elevation. Over the past few years, there have been a remarkable number of clinical research studies focusing on various platelet inhibitors in individuals undergoing PCI with or without an ACS, as well as in ACS without PCI. The 2 major recent advances in antiplatelet therapy are the thienopyridines, ticlodopine and clopidogrel, and the IIb/IIIa inhibitors, (3 available in the United States). In CURE, clopidogrel on top of aspirin was effective in reducing major cardiovascular events in ACS patients who were not clearly heading to the catheterization laboratory within a short period of time. Thus, a relatively conservative approach to ACS was elected. Of note, in the PCI-CURE cohort, the median duration from randomization to PCI was 6 days. This is rather different from the approach used in most institutions in the United States, where earlier intervention is the rule. Clopidogrel proved to be beneficial in the period between randomization and PCI, as well as following PCI. Patients with and without coronary intervention benefited, and no specific subgroups were identified who did not do well.

While the overall results of the 2 CURE reports support the use of clopidigrel as standard therapy in ACS, a number of problems are raised by the data and trial design. First, the initial study was designed to enroll ACS patients without ST elevation, as well as those older than 60, with a history of CAD, but no ECG changes. However, after enrolling 3000 subjects, the criteria were changed to include only the higher-risk patients who had either ECG changes or elevated cardiac enzymes or markers. There is little or no discussion of these important criteria in either of the 2 CURE papers. This is of interest, in that a great deal of attention has been paid to risk stratification in ACS over the past several years, with elevated troponin as well as ST depression emerging as the most important factors separating high from low risk. In fact, some of the IIa/IIIb and LMWH trials have suggested that the benefit with these agents accrues or occurs mostly in individuals with these high-risk features. No specific data are provided about troponin levels or cardiac enzymes in CURE or PCI-CURE. Thus, it is difficult to know the frequency of the various criteria for which the many patients were enrolled. In both papers, the number of patients with ST depression was approximately 42%. No discussion is provided as to whether individuals with or without enzyme elevation and/or ST depression had an altered risk for major events, with or without clopidogrel. It would appear that electrocardiographic changes also included individuals without ST- segment depression but with T wave abnormalities. Other studies suggest that the latter group is at much lower risk. Thus, in that about 60% of CURE patients did not have ST depression, and apparently 74% did not have elevated enzymes, one is uncertain about the risk status of CURE cohort. However, the patients were put into 3 risk categories (tertiles of primary event frequency). I calculate that relative benefit appears lower in the higher-risk cohort (18.5% placebo vs 16.3% clopidogrel, RR = 14%), as opposed to the low- and intermediate-risk patients, where the relative RR was more robust (23.8% and 30.8%, respectively).

What is the physician to do when presented with an individual with protracted chest pain who requires hospitalization? In spite of the various moving targets in ACS, clearly, the platelet is a key factor and vigorous antiplatelet therapy is essential. All patients should be treated with aspirin. Current data still support that individuals who clearly are to undergo a coronary angiogram and possible intervention should be placed on a IIb/IIIa inhibitor as soon as possible. Whether clopidogrel can be used instead of a IIb/IIIa drug, and still achieve the same reduction in clinical end points, has not been tested. The CURE data represent a more conservative strategy to ACS, with much less angiography and PCI compared to the United States, but the trial does not resolve the comparative efficacy of IIb/IIIa receptor blockers vs. thienopyridines. It does not appear that excessive bleeding was a big problem in the CURE individuals who received triple therapy. Finally, the optimal length of therapy with clopidogrel is unclear. Clearly, CURE confirms that for the first 30 days, with or without PCI, clopidogrel is an important addition. Nevertheless, the event curves for placebo and clopidogrel remain relatively parallel after 30 days, and in fact, in the PCI-CURE cohort, no additional benefit was found of reduction in the primary end point between day 30 and the end of the study.

We await more detailed analyses of the CURE data with respect to a number of these issues. At the present time, another weapon has become available in our armamentarium for ACS. Just as clinicians are still sorting out which IIb/IIIa blocker is best and for which patient, or whether LMWH is truly superior to unfractionated heparin (or rather just easier to use), we now have the conundrum of whether to use clopidogrel in all ACS patients and for how long. Risk stratification remains important, such as using the TIMI risk score or other conventional markers of high risk (not particularly emphasized in the CURE papers). One should consider these 2 reports as a useful addition to the ACS guidelines. For instance, it may be more appropriate to use clopidogrel in low and medium risk individuals, reserving IIb/IIIa blockers for the high-risk patients, or perhaps using triple therapy in the latter group.