Questions and Answers (May 15, 2002)

May 15, 2002 Scott McMahon: Our first question is as follows: Do drug-eluting stents eliminate the need for IIb/IIIa platelet inhibitors and their associated costs which are reimbursed in DRG 516? William O’Neill: Dean, could you comment on that? You have done an enormous amount of work on glycoprotein receptor blockers. Could you explain to the audience how you plan to incorporate the additional cost of the stents into the incremental costs of glycoprotein receptor blockers? Dean Kereiakes: I think these products are not mutually exclusive — and correct me if I’m wrong. In the SIRIUS trial, even in the drug-eluting stent arm, the peri-procedural myocardial infarction rate defined as a three-fold bump in CKMB, is as least 6.5%. The use of IIb/IIIa inhibitors will be the primary method of reducing peri-procedural myocardial infarction events. Periprocedural infarction correlates to mortality at one and three years. William O’Neill: So there is really nothing about the drug-eluting stents that would decrease peri-procedural risk. In fact, whatever practitioners are currently doing for peri-procedural management will probably need to be maintained with drug-eluting stent. Dean Kereiakes: That is correct. If you think that specific agents such as abciximab have anti-inflammatory properties related in part to MAC 1 and ALPHA-V, BETA 3 Receptor Blockade, then the anti-inflammatory effect of agent with respect to blocking the inflammatory response to athero-embolization in the microvasculature, is still a very viable effect of this drug and is distinct from the target site where the stent is placed. William O’Neill: Right. I have a question for Greg Braden. Greg, you presented some superb results from your clinical experience. This is one of the few cases where the clinical experience ends up more favorable than the trial results. Oftentimes, the trials show better results because they involve lower-risk patients, but in your clinical experience, the TVR rate is something like 5%. In our early experience at Beaumont Hospital involving over 150 patients, there is about a 3% TVR rate, and Dean has told us that his center’s is roughly 2.5–3.0%, so these TVR rates are all well under the 10% figure. I am curious as to why you think the current clinical results appear to be so much better than what the randomized trial showed. Greg Braden: Our more favorable results may be due to the fact that we were doing more complete treatments compared to what was done in the clinical trials. Longer source trains are now available and I think we are better able to completely treat lesions, especially longer ones, than was the case with the pullbacks. Certainly, we can treat longer segments — I think we are probably getting more effective vascular brachytherapy treatment to the injured zone. Furthermore, there is now an understanding about the importance of treating the entire zone and there is a more systematic approach to ensuring that those areas are well treated. Clinical results have also improved due to the awareness of the problem of under-expanded stents. We also now know that the patients who return with restenosis either need to be evaluated by IVUS or routinely evaluated again with high pressures. Our practices could actually image most of these patients. I personally believe that debulking is a good idea and has decreased the need for placing additional stents, which can cause edge dissections. Thus, we are better at treating the entire injured zone. William O’Neill: Steve, you also had some thoughts about why we seem to be performing much better now than we did in the original randomized trial. Stephen Ramee: In the randomized trial, we were squeezing patients in who had failed all other treatment options and may have stretched the limits of inclusion and exclusion criteria. We were including some patients with lesions that were a little longer than what the trial specified and were not allowed to do pullback so we couldn’t treat the entire circumflex. For instance, if the patient had a 28 mm-long lesion, we measured 25 mm and treated it. As a result, we may have missed the edge. Thus, some of the edges may in fact have been a geographic miss rather than an edge effect. It is very easy to spend an extra 3 minutes to treat the whole vessel if the lesion is longer than the stated length of the radiation catheter. William O’Neill: Those are great suggestions. The technique has dramatically changed from its original system using a 30 mm source train which only effectively treated up to 20 mm lesions. There were no pullbacks and we were rather limited in what we could do. We were also treating a very high-risk cohort with a multiple recurrences and those patients are less numerous now. Also, the availability now of the 30 mm and 40 mm source train in pullback has significantly changed our results. There were eight failures of the first 155 patients we treated. Five of those eight failures were due to geographic miss in our early experience. We could only identify three patients who truly had restenosis inside an adequately treated zone. The results for in-stent restenosis are superb, in my opinion. Steve, I wonder if you could comment on that. During your talk, you highlighted your concerns about using drug-eluting stents for treatment of in-stent restenosis. Could you expound on that issue? Stephen Ramee: We are learning as we go along because it is still early in the treatment process. Let’s take for example a stent that has an obligate margin of real intimal hyperplasia inside of it and you place a second stent — in this case a drug-eluting stent. As the drug-eluting stent works over time, it destroys the scar tissue that is embedded in it. Theoretically, the stent could be left hanging in space and not apposed to the wall. The exception would be if you had a self-expanding stent that continued to expand as the scar tissue regressed. Thus, I think there are some real concerns about using drug-eluting stents for in-stent restenosis. They showed tremendous promise for the prevention of restenosis, but I don’t think they make much sense for the treatment of in-stent restenosis. William O’Neill: I think you’re right, Steve. It is a new insight for me because I had not really understood why some of the results in the early experiences in Rotterdam, for example, were so poor. It is possible that a sort of metal-on-metal rattling is occurring inside the coronary artery which will surely be a potential substrate for subacute thrombosis. Scott McMahon: The following question was sent in by Gary Fishbein: Dean’s financial analysis is quite disconcerting, especially if you anticipate that every stent will be a drug-eluting one. Can we define a strategy of restenosis risk stratification that would, for instance, use bare-metal stents in the low-risk patients — i.e., those with local disease and large vessels — and reserve drug-eluting stents for the higher-risk patients — i.e., diabetics, patients with smaller vessels — and then use vascular brachytherapy for those who do restenose? William O’Neill: Dean, why don’t you tackle that question? What do you think about that type of strategy? Dean Kereiakes: I think the easiest approach would be to try to recreate the clinical trials and limit the application of this technology to patients who have lesions similar to those treated in the clinical trials, specifically, single-lesion, single-vessel disease. That strategy would markedly limit proliferation. The problem is how to police this. Another strategy would be to construct an algorithm that looks at the major predictors such as minimal lumen diameter, stent length, diabetes, smoking and other factors that contribute to restenosis. You could even just go with the top three factors: MLD, stent length and diabetes. That’s what we found in our multivariate analysis of the entire Multi-link family of stents. William O’Neill: Right. You did a beautiful job of constructing a 2 x 2 table that looks at length, MLD, short lesions, large vessels, and very low risk of restenosis. Dean Kereiakes:Yes, a very low risk of restenosis. Now we are faced with an entirely new problem: these patients read the Wall Street Journal. The first patient I placed a SIRIUS Cypher stent in was a 75-year-old retired attorney in Cincinnati who asked to receive this new stent. I asked him if he would like to be the first one. I’m making a joke of it, but in reality, the American public is very well informed thanks to extensive media coverage on new medical technologies. What are you going to do about the patients who want the new stent? Will you say no? I think physicians want to have a “perceived competitive advantage” as well. If you put the product on the shelf and there are randomized trial data suggesting its superiority, the stent will be used. It will be difficult to limit use of this stent. Greg Braden: The financial aspect of this new technology may be the limiter. As you mentioned in your discussion, at $3,000 a pop, it will be hard to place these stents in patients — or anyone with multi-vessel disease. Furthermore, if more than one stent is required, the hospital will lose money. Stephen Ramee: Ron Waksman and I were talking to the JP Morgan analysts this year at Washington Hospital Center and he blocked that out for the first time. I give him credit. It really opened my eyes to the fact that this technology, which you intuitively think is going to drive multi-vessel PCI, may actually drive referrals to surgery because of the cost. First of all, this technology has not been tested in multi-vessel disease; let’s make that very clear. Secondly, you saw the diagram I constructed. The practitioners in our centers will have to reduce the number of stents they place to stay within our current profit margin for PCI procedures, whatever it may be — even in DRG 516 becomes a reality on day-one. William O’Neill: Dean, we need to bring this session to a close. It is incredibly interesting and obviously very crucial to discuss the major scientific, medical and financial implications of this new technology. The great news is that both vascular brachytherapy and drug-eluting stents have finally cured the restenosis nemesis we have been battling in coronary intervention for nearly 30 years. Now the challenge for us is how to responsibly integrate this technology into our local practice environment. I want to thank all of you for participating this evening.