Potential Limits for Drug-Coated Stents

William O’Neill: We will begin by spending about 40 minutes discussing some exciting new advances in the treatment of restenosis. Also, we will discuss drug-eluting stents and their issues in the cath lab. We will then devote a good deal of time and attention to vascular brachytherapy. My focus will be to discuss what is clearly a huge breakthrough in the field of restenosis: drug-eluting stents. I will talk about some clinical data that have recently become available, followed by various drug-eluting stent issues. Finally, I will offer some of my personal views about where the field of drug-eluting stents is headed. There are a number of ongoing clinical trials regarding drug-eluting stents and there is absolutely no question that they represent a huge breakthrough in the treatment of coronary restenosis. We know that, apart from vascular brachytherapy, there really has not been any definitive method to decrease fibrinogen hyperplasia after coronary intervention. I think there is enormous interest and excitement among interventional cardiologists, analysts and patients about these promising new treatments. However, the jury is definitely still out about the efficacy of these breakthrough therapies. We know that a variety of drugs have been tested in various programs, the most advanced of which is Sirolimus. We are all very familiar with the RAVEL study, and have all probably been involved with the pivotal SIRIUS trial in the U.S. The early data from the SIRIUS trial are being presented in two days at the Paris Course on Revascularization, and I think the results will be very interesting. Paclitaxol has also been studied. I am very fortunate and honored to be a co-principal investigator for the important DELIVER trial, which is being conducted in the United States. Not all the trial results have been good, however. We need to understand that just because a stent is coated with a drug, it does not mean it will magically cure restenosis. Tinomycin D was very promising in some early animal data, but proved to be ineffective in early clinical trials, so it was abandoned. Bactimostat in the BRILLIANT study and dexmethosone in the STRIDE trial were also proven to be ineffective. With that in mind, I think we need to be very cautious about assuming that any drug put on a stent will automatically decrease restenosis. We must keep our enthusiasm in check, both in terms of early clinical efficacy as well as the long-term impact of these treatments. When you examine the angiographic data from the studies that have been conducted thus far: the RAVEL study with rapamycin; the TAXUS study; and the ASPECT high-dose and low-dose study; it is clear that the MLDs obtained in these vessels varied. They are anywhere from 2.4 to 2.9 prior to the coronary intervention. The most interesting and important data available are from the RAVEL study, which showed dramatically low late-loss and one of the most impressive early clinical results in terms of decreased restenosis rates. The other studies, such as TAXUS and high-dose ASPECT, have also shown significant reductions in restenosis rates. The dosing amounts that will be put forward in U.S. clinical trials are shown in Table 1. As a result of this, the binary restenosis is 0% in RAVEL — the one that everyone seems to remember — 0% in TAXUS, and 4% in the high-dose ASPECT. These data truly represent dramatic advances in what had previously been reported for angiographic restenosis. I think this is the reason why these numbers have caught the attention of all the scientific and media worlds. There is still a substantial number of major adverse cardiac events, and we need to recognize that if you place a drug-eluting stent across a major sidebranch and occlude it, there will still be CPK elevations or reprised distal embolization. Furthermore, we must understand that because of the small patient cohorts studied in these trials, the confidence intervals for acute complication rates, particularly subacute thrombosis rates, have not been adequately addressed. The RAVEL study has been the most impressive and widely promulgated. The European study conducted by Cordis in which the Cordis Velocity stent was coated with rapamycin was very interesting. This randomized study involved 120 patients in the treatment group and 118 in the control group. There were very low risks for major adverse clinical events. There was one TVR in the drug group and one in the control group. Repeat target lesion revascularization rates were 0% in the treatment group and 22% in the control group — a rate which I think is rather significant considering that these patients had single, short lesions and rather large coronaries. Irrespective of the control group, the 0% restenosis rate in the treatment group has definitely grabbed our attention. The question one can ask is if these 0% and 22% numbers are real. It is important to note that all patients in the RAVEL trial underwent angiographic follow-up. The BENESTENT II trial previously showed that re-catheterizing everyone resulted in a much higher rate of target lesion revascularization compared to simply following the patients clinically. The large VENUS trial, which studied the efficacy of the Velocity stent, followed patients clinically and resulted in only a 4% target lesion revascularization rate. I think the 22% in the control group of RAVEL is probably artificially elevated, but it still does not account for the very low rate of restenosis in the drug-eluting stent arm of this trial. Regardless, the news is very good: four of the lesion subsets have been studied, and we have made a very substantial impact on restenosis. We know that drug-eluting stents represent a breakthrough technology, but there are limitations. It is very important to understand that all of the early studies have primarily involved patients with single-lesion, single-vessel disease. Multi-vessel disease patients have not been systemically studied at this point. Also, left main small vessel disease has not been definitively studied. Thus, we don’t yet know whether the benefit of drug-eluting stents will surpass that offered by coronary bypass surgery. A number of trials are in the planning stages. For example, a European trial in the works will compare stenting with drug-eluting stents and coronary bypass surgery. The U.S. substudy of this European trial will add diabetics. This study promises to offer important data, but you cannot logically assume that drug-eluting stents will replace bypass surgery until these scientific randomized trials are completed. With the likelihood of decreased restenosis rates, we can predict that interventional procedures will increase by 20% as more patients become available to treat percutaneously. However, we know that 0% restenosis will not continue as we begin to treat more complex lesions, bifurcated lesions, long diffused disease, ostial lesions, and so forth. There is no doubt that there will be some instances of restenosis in these cases. In fact, I am told that the SIRIUS trial restenosis rates which will be presented at the upcoming Paris Course on Revascularization will be somewhere between 2–8%. All in all, there has certainly been a marked decrease in restenosis rates in a very specific sub-group of patients. We must all be cognizant of the fact that the FDA has some concerns about drug-eluting stents and will want to see the one-month and six-month pre-clinical data. More importantly, we must understand the lesion subset that we will be treating with these stents. The drugs are on the stent and there is a dose-response group in these drugs. What happens when you place two or three stents in a patient? Also, what happens when we treat very long lesions? These questions have yet to be adequately studied — and overlap is certainly a major concern as well. These types of issues are what significantly delayed the TAXUS IV study. Initially, a very broad complex lesion subset was recommended for TAXUS IV, yet there were significant concerns about dosing. We should all be concerned as well until the data from our investigational trials are finalized, and we must be careful to limit our treatment with these drug-coated stents to the types of patients studied in these trials. Also, there is a concern about angiogenicity. These are drugs, and as such, there is a small but distinct possibility that patients may develop allergic reactions to them. Issues such as these will require further long-term evaluation. Another concern was raised by Van Giesson in a paper published in Circulation regarding the need to find the right polymers. He stated that we must find the right drug, the right dose, the right polymers, and the right method of releasing the kinetics. Van Giesson demonstrated that some polymers result in very little hyperplasia, some cause substantial hyperplasia and some have an inflammatory effect. These complex issues need to be carefully studied in the various patient subsets before we can be fully confident that drug-eluting stents should become the standard of care. Furthermore, we must talk about the uncertainties involved in the various lesion subsets: small vessels, long lesions, chronic total occlusions, bifurcations. For instance, there were early concerns about restenosis complications in the European trial involving the use of rapamycin. Again, we must understand how these very complex drug-eluting stents are going to work in diffusely diseased restenotic vessels. It is expected that diabetic patients will substantially benefit from these drug-eluting stents. Finally, we need to look carefully at acute coronary syndromes, because it is unknown at present whether these drug-coated stents are more predisposed to subacute thrombosis. In acute coronary syndromes, we know that patients who have a large thrombus burden have some biological differences, which may actually cause the drug to diffuse differently. We need to be very concerned about acute coronary syndromes and acute myocardial infarction. I will close my comments by addressing the economic implications of drug-eluting stents. If 900,000 interventions are performed annually in the U.S., and if 1.4 stents are placed per lesion, assuming the price will be about $3,000, then we can estimate a $2.4 billion increase in U.S. healthcare costs. Where will that money come from? Budgets will not increase very much, so resources will have to be taken from other areas. With this in mind, we must be as certain as we can that this is the appropriate way to spend this amount of money. Also, I think we must all be cautious in the early days not to be overly exuberant about drug-eluting stents. I would now like to turn the program over to Dr. Dean Kereiakes. Dean has been one of the key interventional cardiologists in numerous percutaneous coronary intervention trials, and he has some very important data to share with us this evening concerning the economic implications of these new therapies.