What are some of the most important breaking trials to come out of TCT this year?

Dr. Stone: I think there were several pharmacology trials that were very important, as well as several others that provided some very important information.

TRITON TIMI-38
TRITON TIMI-38 looked at long-term clopidogrel use versus the recently approved prasugrel in patients with acute coronary syndrome. These clinical results had been previously known; that is, that the more potent thienopyridine agent, prasugrel, compared to clopidogrel reduced rates of myocardial infarction and stent thrombosis and improved overall ischemic outcomes but had a higher rate of bleeding. The question TRITON TIMI-38 intended to answer was if this was cost effective. David Cohen, MD, presented the results of the formal TRITON TIMI-38 economic substudy, which showed that it was indeed cost effective. The approximate 20% increase in the price of prasugrel compared to clopidogrel was worth the cost because of the reduction in myocardial infarction, stent thrombosis, and also repeat revascularization.

COGENT
A major issue has been raised in that thienopyridines, especially clopidogrel but also prasugrel, are metabolized by the cytochrome P450 system in the liver. Certain proton pump inhibitors, such as omeprazole, are known to reduce the pharmacodynamic potency of thienopyridines. Some studies have suggested that this leads to worse outcomes in patients who are taking clopidogrel; they essentially become resistant if they are also taking proton-pump inhibitors. This is a problem because patients who are on aspirin and clopidogrel therapy have an increased risk of gastrointestinal bleeding, and their physicians prescribe proton pump inhibitors as a prophylactic measure. COGENT is a large randomized trial of 3,700 patients receiving a combination agent of clopidogrel and omeprazole or clopidogrel alone. COGENT showed that there was no difference whatsoever in the overall survival curves and in freedom from cardiovascular death, myocardial infarction, repeat revascularization, or stroke, with follow-up out to 1 year. In contrast, there was a fairly marked reduction in a composite measure of adverse gastrointestinal events, including bleeding from the gastrointestinal tract, bleeding of unknown origin, ulcers, obstruction, and perforation in patients who were receiving proton pump inhibitors. This is very interesting because recently, based on nonrandomized trials, the FDA actually changed the labeling for clopidogrel, suggesting that it not be taken commonly with proton pump inhibitors. However, COGENT possibly suggests that clopidogrel should be taken with proton pump inhibitors. Keep in mind that this combination of omeprazole and clopidogrel works in such a way that it is basically timed release; there is not a release of both agents into the bloodstream at the same time. This is not the same as taking two pills (one clopidogrel and one omeprazole) at the same time, and it may be that in trying to translate these results into the real world, it will be necessary to stagger dosing by 6 to 12 hours.

SPIRIT IV AND COMPARE
SPIRIT IV is, to date, the largest completed head-to-head comparison of any two drug-eluting stents (Xience V [Abbot Vascular, Santa Clara, CA] vs Taxus Express2 [Boston Scientific Corporation, Natick, MA]). SPIRIT IV (sponsored by Abbott Vascular) is an important study because of its size; it was appropriately powered for its clinical endpoints (3,690 patients). Also, there was absolutely no angiographic follow-up in the study, which is important because most previous studies comparing drug-eluting stents have incorporated routine angiographic follow-up that can potentially bias the study toward the stents that have lower late loss. Additionally, because the study was so large, it was able to look at some very important subgroups, particularly patients with diabetes. The subgroups of patients with diabetes is one group in which paclitaxel-eluting stents have always done very well, however, no previous studies have been large enough to look at the diabetic group. SPIRIT IV found that Xience V compared to Taxus Express2 was superior in terms of the primary composite safety and efficacy endpoint of target lesion failure, which is a composite of cardiac death, target vessel myocardial infarction, or ischemia-driven target lesion revascularization. There was a 38% reduction in target lesion failure at the end of 1 year with Xience V compared to Taxus Express2.

SPIRIT IV was also significant in terms of the second major endpoint of target lesion revascularization, which was reduced by approximately 45%. The Taxus Express2 stent actually performed very well—the target lesion revascularization rate was only about 4.5% at the end of 1 year, which is quite good, but the Xience V stent had a target lesion revascularization rate of 2.3%.

The third primary endpoint of the SPIRIT IV trial was a composite of cardiac death or target vessel myocardial infarction. In this endpoint, the Xience V and Taxus Express2 stents were not inferior to each other, although there was a trend toward somewhat better results with Xience V compared to Taxus Express2 (2.2% vs 3.2%).

Another important finding of the SPIRIT IV trial was that stent thrombosis was significantly reduced with Xience V compared to Taxus Express2 (Academic Research Consortium definition of definite/probable stent thrombosis was reduced from 1.1% to 0.3%, an approximate 75% reduction).

The one area in which there were no overall major differences in the SPIRIT IV study was in patients with diabetes. It is very interesting that almost every subgroup benefitted, except patients with diabetes. I think there are different restenotic pathways in diabetic patients compared to nondiabetic patients, and it is likely that the two types of drugs are somewhat differentially effective in these two patient groups.

The COMPARE trial was interesting in that it was also a large trial (approximately 1,800 patients), and it was a real-world trial. For the most part, there were no restrictions on the types of lesions that were allowed into the study. The COMPARE trial used the Taxus LibertŽ (Boston Scientific Corporation) instead of the Taxus Express and was sponsored not by industry but by an academic group. The results of the COMPARE trial were almost identical to the findings of the SPIRIT IV trial in that there was a reduction in overall safety and efficacy, as measured by major adverse cardiovascular events. There was also a reduction in myocardial infarction, target lesion revascularization, and a 75% reduction in stent thrombosis. Importantly, the results of the COMPARE trial apply to the Taxus LibertŽ platform, whereas the results of all the earlier SPIRIT trials applied to the older Taxus Express platform. COMPARE did not report the results of the diabetic versus nondiabetic comparison, but the results of both trials were extraordinarily consistent.

PROSPECT
PROSPECT was the study presented at TCT that I think can be considered the most novel from the perspective of pure scientific interest. PROSPECT was a natural history study of atherosclerosis. In fact, it is the first prospective natural history study of atherosclerosis in which multimodality intracoronary imaging was used to determine whether it is possible to prospectively identify vulnerable plaque. There has been much discussion about vulnerable plaque, but until now, it has primarily been a pathologic entity seen during autopsy of patients who have died from a ruptured plaque. We know that this happens to more than 1 million patients per year, but we currently have no way to prophylactically identify these patients. PROSPECT used advanced imaging to see if it was possible to identify (1) how many patients were at risk, (2) what kind of events were occurring, and (3) if it was possible to predict where in the coronary tree these events were likely to occur.

PROSPECT studied 700 patients who presented with an acute coronary syndrome (basically a myocardial infarction), because they are a homogenous group who already are likely to have one ruptured plaque. We know from autopsy studies that this type of patient is more likely to have another one somewhere, even if you cannot see it. All patients' critical lesions were successfully treated. In addition to angiography performed in all three coronary arteries, intravascular imaging was performed with grayscale intravascular ultrasound and virtual histology. Virtual histology uses radiofrequency analysis of the ultrasound information to provide a different picture of the composition of the plaque, which has been rather highly correlated with histology. All patients were treated with aspirin, clopidogrel, and high-dose statins (optimal medical therapy). PROSPECT identified approximately 12% of patients who had an unexpected adverse cardiovascular event over 3 years of follow-up that was due to a relatively mild portion of the coronary tree. In other words, it was an unexpected vulnerable plaque that put the patient at risk. The PROSPECT investigators then went back and examined the individual problematic lesions to determine what they looked like 3 years earlier, as well as the other areas that did not develop a problem. By using proprietary methodology to link the angiographic, ultrasound, and virtual histology data, it was possible to characterize every area in the coronary tree. PROSPECT found that when reviewing the angiograms of these patients, the areas that were causing these problems looked very mild. However, when reviewing the ultrasound data from the same areas, they often looked like severe plaques. When looking at the virtual histology data, they often looked like a particular type of plaque—a thin-cap fibroatheroma. A thin-cap fibroatheroma is the type of lesion that is usually associated with thrombotic occlusion and sudden death in autopsy series. The combination of using ultrasound and virtual histology allowed for more accurate identification (than any other parameters) of where the problem areas were going to occur.

An interesting finding was that only about 12% of patients during 3 years developed a problem (approximately 4% per year), which is relatively low. Most of the episodes were not death or heart attack and were mostly angina (often progressive or unstable angina). This goes to show that as we initially treat patients and stent their severe lesions and follow them closely with good medical therapy, we do a pretty good job in keeping their events relatively under control. Of course, this does not help the 1 million people per year who are having heart attacks and who we are not following closely and are not on optimal medical therapy.

HORIZONS-AMI
HORIZONS-AMI is the largest study of drug-eluting stents versus bare-metal stents in any setting, and this study focused on patients who suffered a heart attack who were undergoing primary angioplasty. There was also a second study (a two-by-two randomization) in which 3,602 patients with ST-segment elevation myocardial infarction were also randomized to either unfractionated heparin and glycoprotein IIb/IIIa inhibitors versus bivalirudin alone. The 1-year data from the HORIZONS-AMI study had previously been reported, but TCT was the first time the 2-year data were presented.

Basically, when looking at the bivalirudin versus the heparin and glycoprotein IIb/IIIa inhibitor arm, there continues to be impressive results in patients treated with bivalirudin. There was a marked reduction in cardiac mortality and a significant reduction in overall mortality at 2 years. These findings could be very much related to the relatively early prevention of bleeding. Additionally, as these patients were followed over time, there was a significant reduction in reinfarction rates in the bivalirudin-treated patients. The curves are gradually spreading, favoring the bivalirudin group. It has been described that there can be long-lasting effects of bleeding, usually on mortality, but now we must consider if it extends to reinfarction as well.

When looking at the drug-eluting stent versus the bare-metal stent arm of the HORIZONS-AMI trial (Taxus Express2 paclitaxel-eluting stent vs an otherwise equivalent bare-metal Express stent), the rates of target lesion revascularization were nicely reduced, not only at 1 year, but the difference had spread nicely between 1 and 2 years. There was a significant 42% reduction in ischemic target lesion revascularization with the Taxus Express2 paclitaxel-eluting stent compared to the Taxus bare-metal stent.

There were no observed safety concerns. Importantly, the rates of myocardial infarction, reinfarction, and stent thrombosis were very similar between the two arms. If anything, there tended to be less mortality in the Taxus group between 1 and 2 years (P = .04; 0.8% of Taxus Express2 patients died between 1 and 2 years vs 1.8% of Express bare-metal stent patients). I think we have to consider this to be hypothesis-generating, but regardless, the safety profile out to 2 years looks very good, and the anti-ischemic benefits were maintained or were slightly spread.

FAME
The FAME trial examined the best way to revascularize patients with multivessel disease. FAME used an angiographically guided strategy versus a fractional flow reserve (FFR) strategy. FFR utilizes a pressure wire to measure any hemodynamic significance across a lesion. The FAME investigators randomized approximately 1,000 patients with either two- or three-vessel disease to either FFR-guided or angiographically guided drug-eluting stents. For the first time, they presented their 2-year data at TCT, and they saw fewer major adverse cardiovascular events in the FFR-guided strategy, mostly due to fewer myocardial infarctions resulting from prevention by not treating lesions up front and not having restenosis occur. There were also some trends toward benefit in terms of improved quality of life. They also showed that the lesions they initially deferred did quite well, for the most part, remaining rather stable without need for intervention 2 years later. These results suggest that severe lesions need to be treated, but in more intermediate lesions, our patients might benefit from using physiologic lesion assessment with FFR to help decide which ones do, in fact, need to be treated.

One of the big topics at TCT this year was transcatheter aortic valve implantation (TAVI). What is your take on the current status of TAVI, and what is your gut feeling on when we might see its application in clinical practice?

Dr. Stone: I certainly agree that TAVI was one of the hottest topics at TCT this year. The valve rooms were just packed throughout the entire week. The whole issue of valves can be divided into transcatheter approaches for aortic valve implantation (particularly aortic stenosis) and, emerging even more rapidly, for degenerated mitral valves (or patients with so-called functional mitral regurgitation). There are two valves that have become widely used in Europe for TAVI—the Edwards Sapien valve (Edwards Lifesciences, Irvine, CA) and the CoreValve (Medtronic, Inc., Minneapolis, MN). TAVI has basically become a routine procedure in Europe, primarily used in patients who are at high risk for surgery or who are inoperable. However, these are still large devices, and there can be complications that result from using them (in particular, vascular complications), and we do not yet know the long-term durability of these valves. The pivotal trial for the Edwards Sapien valve, the PARTNER trial, has completed enrollment, and hopefully we will see the results of this eagerly anticipated trial at TCT next year. If the results of the PARTNER trial are beneficial, it would presumably lead to approval of the device in the United States, likely in 2011, for use in either high-risk or surgically inoperable patients with aortic stenosis.

There is also a lot of excitement in the mitral valve space, particularly involving the MitraClip device (Evalve, Inc., Menlo Park, CA), which has also completed its pivotal trial—EVEREST II. The results of EVEREST II will be presented in 2010, and assuming there are positive results, they could result in approval for use in the United States in patients with mitral regurgitation.

TCT moved to a new venue in San Francisco this year. How did the meeting go from your perspective?

Dr. Stone: We went into this year's TCT with some cautious apprehension. After 20 years in Washington, DC, San Francisco was our first new venue. TCT is a very complex meeting. We've been told that it is the most complex multimedia production outside of the Olympics, due to the high-definition transmissions from more than 20 sites around the world. We were somewhat nervous about moving to a new site, but we thought it would be stimulating to shake things up a little. We were very pleased to see that attendance was up approximately 20%, and in this economy and with most other meetings shrinking by 20% to 30%, we think the meeting was a tremendous success.

Interestingly, the increase in attendance was almost all physicians; approximately 2,000 to 3,000 more physicians came to TCT this year from all over the world, not just a boost from the Asia-Pacific area. Attendance by United States physicians was up 15%, and the rest of the world was up 75%. We are very pleased with how the meeting turned out.