Trouble in Trialville: Plans for Post-DES Clopidogrel Trial Get Bogged Down Over Trial Design, Leadership

Shelley Wood

April 11, 2008

April 11, 2008 (Chicago, IL) - Almost 16 months after an FDA hearing into drug-eluting stent (DES) safety emphasized the need for a randomized clinical trial to determine, once and for all, the optimal duration of clopidogrel (Plavix, Sanofi-Aventis) plus aspirin after DES implantation, researchers, government agencies, and industry sponsors cannot agree on how it should move ahead.

The clinical trial proposed by Dr Mitchell Krucoff and colleagues at the Duke Clinical Research Institute and dubbed Clopidogrel: Optimal Duration of Antiplatelet Therapy (CODA) is currently in limbo while the various parties try to agree on key aspects of trial design, duration, and who should head up the study.

In an interview with heartwire , Krucoff explained that the CODA trial, designed with a patient-oriented end point of death, MI, and stroke vs bleeding, was developed within the larger private-public partnership between the FDA and Duke University, dating back to a 2006 'memorandum of understanding' that created a collaborative Cardiovascular Safety and Research Consortium, originally to evaluate drug effects on cardiac repolarization. According to Krucoff, apart from answering the public-health question of optimal duration of clopidogrel therapy post-DES, CODA, as designed, will also help fill a key gap in the "regulatory landscape": namely, how to test drugs necessary for the safe use of drug-device combinations.

Companies not convinced by CODA

Despite the urgent nature of the question--with roughly 10 million DES-treated patients worldwide and no clear answers as to how long they should take clopidogrel and aspirin--CODA has stalled. Krucoff and colleagues, on the FDA's advice, have filed an investigational device exemption (IDE) for the trial, and they've held "think-tank meetings" with the four major DES manufacturers as well as smaller DES companies; thienopyridine developers Sanofi-Aventis/Bristol-Myers Squibb and Eli Lilly (responsible for clopidogrel and the next-generation thienopyridine, prasugrel, respectively); the National Institutes of Health, and the FDA. Ideally, all of these parties would be funding the CODA trial.

But the device companies have balked at the notion of coughing up support for a trial with a patient-oriented end point that won't specifically establish the safety of a DES or determine the ability of dual antiplatelet therapy to reduce the risk of late stent thrombosis and how long such therapy is warranted to mitigate this risk. The CHARISMA study, for example, suggested that long-term dual antiplatelet therapy in patients with vascular disease reduces death, MI, and stroke, but that benefit may not be specific to DES. Acting via their umbrella trade organization, Advamed, the companies have met with the FDA to propose changes to the CODA trial design.

"We have not yet seen the Advamed proposal," Krucoff told heartwire . "My understanding is that the Advamed proposal [seeks to] change the primary end point from what the current CODA IDE submission proposes. We tried to make CODA a big, simple, clinical trial because there is nobody tumbling head over heels to fund this. Our goal was to make this as logistically feasible and inexpensive as possible and have it still provide some prospective insights, on an expedited basis, as to what we should be doing with the 10 million people with DES in their hearts. Our understanding is that the Advamed proposed trial would be a more complex, slightly larger trial, with a more device-specific rather than patient-oriented end point."

Advamed and its member companies are pushing for a primary end point of late stent thrombosis, as defined by the Academic Research Consortium, which would require a larger trial and more funding. Given the uncertainty over just how long the stent-thrombosis risk endures, trial follow-up would likely also need to be extended. Sources also tell heartwire that the DES manufacturers are also calling for a bare-metal-stent arm, since the risk of late stent thrombosis following bare-metal-stent implantation is also unknown.

"Such a trial would, indeed, be feasible if the device industry is willing to put up additional funds," Krucoff said. "Our hope is that these will not go forward as separate efforts but as a collaborative effort with both the short-term issue of what to do with the millions of patients with DES and also the longer-term issue of helping us try to rearrange the regulatory landscape so that it is more accommodating and defined for this drug-device interaction."

Heading up the trial

A separate issue is whether Duke would still head up the trial if the CODA trial design is rejiggered. Krucoff believes that, given the FDA-Duke collaboration already in place, Duke should have a key role to play. Others have told heartwire that the sponsors are asking the FDA to reopen the decision as to which research group would lead the trial and permit other academic research centers to bid.

"There's all kinds of levels of politics here, everything from ego, to money, to whatever," Krucoff admitted. "As in any hot-topic area, there may be some competition as to who runs the trial. That for me, frankly, is a secondary issue. The important thing is for us to get as immediate an answer as possible for patients and also help evolve the landscape for evaluating these issues long term."

Asked if the CODA trial would still go ahead if the FDA agrees to a new, Advamed-supported trial, headed by a different group, Krucoff hedged.

"There are a lot of hypotheticals," he said. "If our trial were totally redundant, and they had all the money and we had none, would we go ahead? No. But what we're most enthusiastic about right now is that these efforts are progressing. They are trying to achieve the same definition and information, and our hope is that this will come forward as a single, collaborative effort. . . . The details, frankly, of who does what with regard to the trial itself I'm sure are going to be subject to all of the usual issues that make people decide to do a trial at this center or that center, or with this PI or that PI, with this research organization or that research organization. So nothing is written in stone."

MATRIX registry results

Results from the MATRIX registry study, presented at last week's American College of Cardiology 57th Annual Scientific Session/i2 Summit-SCAI Annual Meeting, reinforced the need for a randomized controlled trial to settle the DES-safety/clopidogrel-duration issue. According to Dr George Dangas (Columbia University, New York, NY), who presented the landmark analysis, patients treated with drug-eluting stents who are no longer on clopidogrel after one year are more likely to die than patients still taking the drug. But that signal of increased risk is muddied somewhat by a lack of information on patients who stopped and restarted the drug over the study period, investigators for the study noted.

Among the 1510 patients treated with a sirolimus-eluting stent in MATRIX, a low but measurable rate of clopidogrel discontinuation over time was associated with higher all-cause mortality. At two years, patients off clopidogrel had a statistically significant, threefold higher increase in all-cause mortality and a twofold, statistically significant increase in death/MI.

But the devil is in the details: patients still on clopidogrel at one year were also more likely to have a target lesion revascularization or target vessel revascularization--a chicken-and-egg scenario that suggests some patients were taking clopidogrel at one year or beyond because they had undergone repeat procedures. Indeed, more than one-third of patients who were off clopidogrel at 30 days were back on it at six months and at one year; almost half of the patients who were off clopidogrel at six months were back on it at one year; while 44% of patients who were off at six months were back on it at two years. Even more striking, one out of 10 patients who were off clopidogrel at one year were back on it at two years. The numbers likely speak to patient-level finances, disease progression, and comorbidities, as well as uncertainty among physicians as to whether one year of dual antiplatelet therapy is sufficient or whether a lifetime on both aspirin and a thienopyridine may be the safest bet for avoiding stent thrombosis post-DES. To confuse matters further, stent thrombosis after one year occurred in 0.3% of patients still taking clopidogrel at 12 months, but in no patients not taking it at this cutoff.

Discussing the results after Dangas's presentation, Dr Bill Knopf (Piedmont Hospital, Atlanta, GA) urged the audience to take the findings with a grain of salt, since follow-up to date on the 1510 patients is incomplete: just 88% at one year and 70% at two years. He called the lack of difference in late subacute stent-thrombosis rates based on 12-month clopidogrel coupled with the higher rate of death and MI "a bit of a disconnect" and said that the zigzagging of patients on and off clopidogrel and the effect of this pattern "needs to be sorted out."

"There is a higher death/MI rate with discontinuation of Plavix at one year, leaving open the question of what is the optimal dosing of Plavix and what is the mechanism for this death/MI-rate increase." Knopf concluded. "At the end of the day, with a registry study, it opens a lot of questions that don't all get answered, and that leads us to larger randomized trials to determine what might be the optimal strategy."

Also commenting on the MATRIX results, Dr Gregg Stone (Columbia University), a coinvestigator for the MATRIX registry, emphasized the difficulties of using a landmark analysis to address whether the benefits of long-term clopidogrel outweigh bleeding risks in a group of patients who go on and off the drug. In MATRIX, for example, there was a trend toward being off clopidogrel being associated with both cardiac and noncardiac deaths, suggesting that those patients may just be sicker, Stone noted.

"Registries such as this, like the Duke/Eisenstein registry, suggest that long-term Plavix may be beneficial and may be associated with reduced mortality; however, given the potential for unmeasured confounders to cloud the result, a large-scale randomized clinical trial is clearly warranted if we're going to know the answer to this very important question."

The complete contents of Heartwire , a professional news service of WebMD, can be found at, a Web site for cardiovascular healthcare professionals.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.