Cardiovascular Device Update Associate

A blockbuster technology faces significant pushback as studies begin to gather longer-term data concerning real-world use.

When the FDA in April 2003 approved the first drug-eluting stent (DES) for sale in the United States — the Cypher made by Johnson & Johnson (New Brunswick, New Jersey) subsidiary Cordis (Miami Lakes, Florida) — this new technology appeared to be nothing short of miraculous when compared with its bare-metal stent (BMS) forebears. In study after study DES devices were shown to reduce restenosis by more than 70% compared to BMS devices, and to reduce the need for repeat procedures to less than 5%, compared to the redo procedures of 20% to 30% for BMS.

Prior to commercialization DES devices were described as a “disruptive technology,” poised to revolutionize interventional cardiovascular procedures. Though occasional cautions were expressed concerning the need for long-term data of at least 10 years, the FDA based its approval decisions on months of data, that data primarily focusing on the trial endpoint of reduced restenosis. There was a clear absence of long-term studies concerning adverse events and mortality.

With commercialization, market uptake of the Cypher, and then the Taxus DES from Boston Scientific (Natick, Massachusetts) — which garnered FDA approval in 2004 — was perhaps the most rapid in medical device history. The technology was quickly adopted by interventionalists, rapidly moving to 60% of total stent usage and now representing up to 100% of stent placements in some hospitals. Up to 6 million patients worldwide have received DES devices, creating a $5-billion-a-year business for market leaders J&J and Boston Scientific, and they, along with other companies, have continued to roll out new DES products in Europe and other countries.

Now, with use of DES in real-world clinical practice, questions about these devices are being raised. Was the FDA too hasty in approving the technology? Does DES technology provide clinical benefit beyond the avoidance of redundant procedures to open clogged arteries? Have the important questions been answered concerning the appropriate uses of these devices in the appropriate patients? And the appropriate drug doses? And what are the proper dosages of follow-on drugs? What is the relationship between DES and mortality?

In sum, are DES devices as safe and effective as they were originally bannered to be?

Recent studies put safety in question

In May of this year at the EuroPCR in Paris, Jean Marco, MD, of the interventional cardiology unit at Clinique Pasteur (Toulose, France) and William Winjs, MD, co-director of the Cardiovascular Center at OLV Ziekenhuis (Aalst, Belgium) laid out an important argument: that simply reducing the occurrence of restenosis is not the ultimate or the most important endpoint for DES technology. Rather, they focused on the need for evidence concerning mortality. Marco noted that cardiologists are already on the lookout “to resolve certain of the problems that are inherent in these devices,” primarily thrombosis and the possibility of late or “precocious” thromosis.

Then last month, at the annual meeting of the European Society of Cardiology (ESC)/World Congress of Cardiology (WCC; Sophia, Antipolis, France), the clinical bombshell was dropped, seeming to give substance to the many early warnings. At the meeting researchers issued reports indicating that DES devices may increase the risk of potentially fatal blood clots.

A meta-analysis of results from past clinical trials of first-generation DES devices presented at the meeting indicated that DES-implanted patients had a greater risk of heart attack or death than patients BMS-implanted. The increased relative risk was greatest for J&J’s Cypher stent, at a statistically significant 38%, said Edoardo Camenzind, MD, of University Hospital (Geneva, Switzerland).

For Boston Scientific’s Taxus stent the increased risk was 16%, a figure not meeting statistical significance. The actual risk of death or heart attack was still low in both groups, at 6.3% for patients given Cypher and 2.6% for Taxus. But because DES devices are so widely used, thousands of patients could be affected.

Salim Yussuf, MD, of McMaster University (Hamilton, Ontario), said the findings were disconcerting. “Now that we are having this concern, I would urge limited use,” he said. “As clinicians we seem to have lost our clinical judgment, let alone our ability to view data and evidence,” Yusuf added. “We therefore need a thoughtful and selective approach to PCI, complementing full medical therapy ... The whole field of angioplasty has been led astray by a preoccupation with restenosis, for which study after study has shown has no prognostic value.

“We’re chasing problems that are iatrogenic that naturally would not exist in people. We’ve had a perverse financial incentive on the practice of cardiology. It is time to stop and reevaluate.”

At the very least, said Yussuf, large registries should be mandated to track adverse DES events. But Yussuf also made a plea to the major cardiovascular organizations to revisit not only DES use but also the role of PCI in the treatment of stable, non-drug-refractory angina. He added that PCI and DES use play a key role in the treatment of unstable angina and acute coronary syndrome — it is as a treatment for stable angina to treat non-life-threatening restenosis that Yussuf singles out as a “myth” and a “man-made disease.”

This latest research follows an earlier Swiss study that found the rate of heart attacks and deaths was more than three times higher in patients with DES devices who stopped taking blood-thinning drugs than those who had bare-metal stents.

Drug coating on stents apparently works by producing arterial scarring after a stent is implanted, but the drug coating may delay healing around the stent, creating a risk of clots forming, which can trigger a heart attack, the researchers suggested. The real worry is what happens after a year or more, with evidence growing that the risk of late stent thrombosis continues long after DES placement.

Arguing for DES safety at the conference, Antonio Colombo, MD, of San Raffaele Hospital (Milan), said that data from the four-year TAXUS II trial indicated “low target lesion revascularization rates [and] low cardiac death rates.” The reported cardiac death rates for both dose levels was 1.6%, not a meaningful difference from the 1.5% in the bare-metal stent control group. Additionally, TAXUS II showed that both dose levels cut restenosis sharply. The controls, who received bare-metal stents, experienced a restenosis rate of 15.7% compared to the 7.2% level in low-dose stent and 3.7% in the medium-dose stent.

Colombo said that DES technology “has truly changed the way physicians treat coronary artery disease.” (For more on the data as presented at the Barcelona conference, see next article.)

NEJM weighs in

The New England Journal of Medicine (NEJM) has published articles covering two other studies as well as an editorial on those studies that jointly do little to clarify these matters. A study of the sirolimus-eluting Cypher did not find any significant difference between the controls and the study groups for death or reinfarction, nor did thrombosis become more problematic with the test than with the control. However, the study showed “a significantly reduced ... rate of target-vessel revascularization at one year.”

As for the paclitaxel-eluting Taxus made by Boston Scientific, Laarman, et al. report that “there was a trend toward a lower rate of serious adverse events ... than in the uncoated-stent group, but that the P score, which was 0.09, was not robust enough to be deemed significant despite a 4% drop in adverse events.

The study also demonstrated no difference in stent thrombosis after one year.

Boston Scientific wants broader look

On the point of serious adverse cardiac events, Charles Rudnick, a media relations representative with Boston Scientific, told Cardiovascular Device Update that the inferential power of the NEJM study could have been improved with a larger enrollment, which was less than 700.

“The body of evidence available demonstrates that Taxus provides a significant benefit as compared to bare-metal stents and is as safe as bare-metal stents. Both these studies have to be placed in the perspective of a broader body of clinical data available.”

In his NEJM editorial, Frans Van de Werf, MD, wrote that DES technology does indeed “significantly reduce the risk of both restenosis and target-vessel revascularization after elective PCI (percutaneous coronary intervention) as compared with uncoated stents.” However, he cautioned against comparisons of the two studies on several accounts.

Still, while design and inclusion criteria were different, “the results seem to be in line with those of studies” comparing the two, namely “lower rates of restenosis and repeated intervention, with the sirolimus-eluting stent [failing to demonstrate] significant differences in myocardial infarction or death.”

Cordis acknowledges ‘challenge’

Mariela Melendez, the director of communications at Cordis Cardiology, told CDU that “[I]n four controlled randomized trials of Cypher vs. bare metal stents involving about 1,800 patients followed over four years, we have seen five cases of very late stent thrombosis with the Cypher Sirolimus-eluting stent, compared to no such cases observed with bare metal stents.”

However, she said that while these cases did not constitute a “statistically significant difference, it is an important clinical challenge that we continue to investigate.”

Cordis’ data showed long-term mortality rates of 6.5% for patients with a Cypher stent against 5.1% for bare metal stents, not a statistically significant difference.

Kaiser to examine the issue

In a response to the new — and often contradictory — studies of DES technology, healthcare giant Kaiser Permanente (Oakland, California) last month reported that it will conduct a retrospective study of the safety and efficacy of DES devices in its patient population, which totals 8.5 million. Kaiser obviously has what might be called a “real-world” interest, rather than simply a research interest, in wanting to determine the appropriate use of DES devices and bare metal stents.

Kaiser doctors reportedly have implanted an average of 1.5 DES (which cost on average about $2,500 vs. $900 for a bare-metal stent) in each of the 3,500 angioplasties performed in 2003. Calvin Weisberger, chief cardiologist for Kaiser, estimated that the rate of implantation currently is about three times the 2003 rate.

“There are relatively few people, to start with, who are even a candidate” for a DES, Weisberger said, noting that seemingly “everyone got one.” And he said that with introduction of DES devices, there may have been a “me-too” syndrome among patients who were saying “if he gets it, I want it.” This then put significant patient pressure on physicians as well as creating some similar peer pressure.

DES benefit vs. BMS brethren over-hyped?

Another area that needs to be investigated is the purported benefit of the substantially costlier DES vs. the traditional workhorse BMS.

A study published in the August 2006 issue of Clinical Cardiology showed that the benefits of switching from BMS to DES devices to open/reopen blocked coronary arteries may be smaller than previous trials have indicated. The findings were based on data provided by Goodroe Healthcare Solutions (Norcross, Georgia), which is focused on helping hospitals improve clinical quality and economic performance.

The Study looked at detailed clinical data from 17,000 people BMS-implanted from December 1998 through March 2003. The information was collected from 17 hospitals across the country that use Goodroe’s CathSource Enterprise software, which feeds information into the Goodroe Data Warehouse, that the company says is one of the nation’s largest warehouses of data on outcomes, costs and productivity measures for cardiac catheterization labs, as well as for cardiac and orthopedic surgery procedures.

The researchers — interventional cardiology expert Spencer King, III, MD, from the Fuqua Heart Center of Piedmont Hospital (Atlanta); Cynthia Yock from the Center for Primary Care and Outcomes Research at the School of Medicine at Stanford University (Palo Alto, California); and Mike Isbill of Goodroe — found that less than 8% of the patients needed additional blockage treatment after the first follow-up month. This is less than half the rate originally reported in most DES trials.

Joane Goodroe, founder of Goodroe Healthcare Solutions, told CDU that the researchers were stunned at these findings. “We felt like our statistics would be more in line with what the drug-eluting stent studies had shown, and we were surprised that there was such a difference.”

Goodroe said that physicians’ rapid acceptance of DES in 2002 was based on evidence from manufacturer-sponsored clinical trials that showed they reduced the need for revascularization by as much as 81% compared with bare-metal stents. However, the Stanford study, along with meta-analysis, published in The Lancet, shows there is no significant difference in serious events, such as mortality and post-stent heart attack, between patients receiving DES and those receiving bare-metal stents.

Goodroe did not suggest that the companies that manufacture DES devices had gamed the research data to demonstrate a dramatic difference between the DES and its BMS counterpart. Rather, she noted an important and common characteristic of much trial research: that it may look at “a very narrow piece in time with controlled variables. And in real-life practice, that’s not how things work.” Thus, she emphasized a need for “deeper” real-life, long-term data to ultimately resolve the issue.

The study utilizing Goodroe data “provides a real-life example of the importance of collecting and evaluating clinical data,” she said. “Now, we have information that questions the value of using expensive drug-eluting stents, rather than bare-metal stents, during cardiac catheterization procedures.”

Fooling mother nature?

Noting that the increase in thrombosis associated with DES has been a concern from the beginning, Dr. King of Piedmont Hospital — a former president of the American College of Cardiology (Bethesda, Maryland) and a pioneer in the development of the angioplasty procedure — said that he had noted this possibility from the reports released following approval and roll-out of DES to general use.

“There was a concern, not in the individual randomized trials, since none of them showed a difference, but in the experience outside the clinical trials people saw this,” he told CDU. He compared the use of DES with the application of radiation to prevent restenosis — most notably deployed by now-defunct Novoste (Norcross, Georgia) via its Beta-Cath system — a precursor to DES, in the similar necessity to administer an extended regimen of antiplatelet therapy to counter the potential for latent clotting.

As to the evidence concerning the association between DES and thrombosis, he said that the jury must be considered as still deliberating. But, he said that the possibility for that association has an immediately intuitive basis. “We are doing something [using DES] to block the normal protective endothelial healing of the artery. It’s not surprising that if you fool mother nature that way, you’ve got to fool her again to keep from forming of a clot.”

The questions to be asked, and answered, he said, are the ones that are increasingly being asked for both devices and drugs that involve the basic benefit/risk ratio: What are the benefits of DES, and do those benefits significantly outweigh the possible contraindications?

King estimates that by switching completely from BMS to DES, the interventionalist saves an average of about five redo procedures. But he said this must be balanced against the increased risk of heart attack and death for some patients.

Importantly, he noted that “everyone is not average. There are situations where there’s clear-cut, dramatic benefit in reducing restenosis, and there are other situations where restenosis is rather trivial and not worth taking the risk of all. That’s why we’ve argued for the selective use of drug-eluting stents.”

Some patient populations that King believes would benefit substantially from DES use include diabetics, patients with smaller vessels and longer lesions, those having problems in the vicinity of the anterior descending artery and patients who present with total occlusions.

King said that at Piedmont Hospital, his group currently uses DES in about 82% of its patients, which he characterized as “pretty conservative,” when compared with a national utilization average for DES of more than 90%. Somewhat tellingly, however, King cited one study recommending that the ratio of patients who truly need a DES to those who would best be BMS-implanted was 60% to 40%. “That number is one that I think is probably reasonably accurate,” he said.

Another large question still to be answered is whether physician use of DES devices will decrease as a result of the latest studies.

“Most cardiologists don’t really know yet what to do with all of this information,” King said. He said the marketing and the hype surrounding the DES has convinced many of them that “the bare metal stent is the old Model T Ford and the DES is the Lamborghini.” He added its probably good to “sober up” and actually think about which one to put in the patient “rather than giving you the latest and the greatest.”

He predicted a bit of pushback in the near term on DES use, but he also expressed hope that doctors could work with the medical device industry “to fix this defect.”

Enter the FDA

Is there a defect — or defects? If so, should these possible risks have been seen early on?

Such questions, of course, turn the spotlight onto the FDA, which is fond of bannering itself as the “gold-standard” in world medical regulation, but increasingly criticized for approval of products — primarily on the drug side — with significantly greater risks than benefits.

Thus, last month — in an obvious response to the new studies questioning the broad use of DES — it issued a lengthy statement saying that it was reiterating its position that DES devices are effective and safe. But, in a rather revealing comment, it added that reiteration of policy with a rather large caveat: that it is studying the new data concerning the devices and that it will convene a meeting of its cardiovascular advisory panel “later this year” for additional review, including the possibility of labeling change for DES.

One wonders if the agency itself is convinced of DES safety — or is it planning to “reposition” itself on the issue?

In a summary statement that appeared to be its most definitive position, the agency said that there are still “important questions” concerning the devices ... for example, what causes drug-eluting thrombosis, how often it occurs, under what circumstances it occurs, or what the risk of occurrence is in a given patient?

The agency statement said, “The specific studies that have prompted recent media inquiries are the BASKET-LATE study (presented at the March 2006 American College of Cardiology Scientific Sessions in Atlanta, Georgia) and more recently, the Camenzind meta-analysis presented at the September 2006 European Society of Cardiology Annual Meeting/
World Congress of Cardiology Meeting in Barcelona, Spain.” It goes on to say that in these studies, targeting patients following 18 months to three years after stent implantation, there was found a “small but significant increase in the rate of death and myocardial infarction.”

Promised — more formal evaluation

These research results “have raised important questions,” the statement said, while adding that “the data we currently have do not allow us to fully characterize the mechanism, risks, and incidence of DES thrombosis.” It said it intends to study the data presented at the Atlanta and Barcelona meetings with “a more formal evaluation.”

In terms of regulatory concerns, it said the agency is “keenly interested in the long-term follow-up of patients enrolled in the original pivotal DES randomized trials, as well as those in the more complex patient and lesion subsets (for example, patients with diabetes; acute myocardial infarction or multiple vessel disease; or lesions involving arterial bifurcations, the left main coronary artery, and long arterial segments) who are currently being treated in “real world” randomized and registry studies.