Carotid stent makers and interventionalists plan to use recently reported CREST trial data to argue for expanded Medicare coverage that allows more patients to receive the devices, but potential obstacles remain.

The long-awaited 2,502-patient, randomized CREST trial data, reported Feb. 26 at the American Stroke Association annual meeting in San Antonio, found carotid stenting and carotid endarterectomy surgery to be statistically equivalent on a composite endpoint.

The results are considered landmark by interventionalists because the study is the largest to compare the interventions and also because it included a broad array of patients at risk for stroke who were candidates for open surgery. The trial was sponsored by the National Institutes of Health and stent maker Abbott.

"I think this will close the door on the [question] of whether stenting is appropriate," said Christopher Cates, director of vascular interventions at Emory University, Atlanta, and VP of the Society for Cardiovascular Angiography and Interventions, in an interview. "We now have two good therapies."

But for parity between stenting and surgery to transfer to routine clinical practice, CMS needs to significantly expand its coverage policy, which currently limits stenting to a small minority of patients deemed to be at especially high risk for surgery.

For that expansion to occur, the data will first need to be published. Investigators say the study is currently under review at a peer-reviewed publication. But beyond that, the Medicare agency, which has displayed significant caution about expanding the practice of carotid stenting in the past, will need to be convinced.

Also, almost in parallel to last month's CREST presentation, interim data from a large European trial comparing stenting to surgery was published online in The Lancet. That data, from the 1,713-patient International Carotid Stent Study, found a higher event rate in the stent group compared to surgery.

Nonetheless, CREST researchers and carotid stenting advocates say the NIH-funded trial was so well designed that it will be enough to expand percutaneous intervention to many new carotid artery disease patients.

"This is going to be the trial we lean back on as medical practitioners," said Cates, who is on the multi-specialty society committee that is writing clinical practice guidelines for carotid artery disease, expected out later this year.

CREST Conclusion: Two Reasonable Carotid Therapies

There are about seven carotid stents approved by FDA for high-risk surgical patients with stenosis levels as low as 50% if accompanied by clinical symptoms and at 80% or more if no symptoms are evident. But the current market for the products is further limited by Medicare, which only covers symptomatic patients with 70% or greater blockage.

The CREST trial enrolled a much broader patient population at 117 U.S. and Canadian centers. Most of the subjects were at average risk for surgery. About half of the patients had symptoms and half did not. Most had more than 70% blockage, but about 15% did not.

The stenting group received Abbott's Acculink carotid stent system, originally developed by Guidant.

A statistically similar proportion of patients in the stenting and surgery groups (7.2% vs. 6.8%) exhibited the primary endpoint of death, heart attack or stroke 30 days after the procedure, as well as stroke associated with the clogged artery following an average of 2.5 years of follow-up.

The data is significant because prior to this, surgery had been judged to be the better approach in most cases.

The outcomes were similar by patient gender and by whether they were symptomatic or asymptomatic, and by the type of specialist performing the intervention.

The researchers observed that patients younger than 70 did better with stents while patients older than 70 had better results with surgery.

Notably, they found that the stroke rate was higher in the stent group, but that this was balanced by a higher myocardial infarction rate in the surgery arm. Also, they point out that the event rates in both groups are the lowest that have been observed in any major trial.

"The CREST trial results show that we now have two safe and effective methods to treat carotid artery disease directly, the tried and true CEA [surgery] and the new kid on the block, CAS," said Thomas Brott, CREST's principal investigator and Mayo Clinic neurologist.

FDA, CMS Submissions Forthcoming

But from a market perspective, carotid artery stents are not completely new. The first one, Acculink, was approved in the U.S. in 2004. With more entries from Abbott, Boston Scientific, ev3, Johnson & Johnson and Medtronic, the domestic market has reached about $80 million-$100 million and has settled there due to the coverage restrictions, analysts estimate.

If reimbursement were expanded to the population represented in CREST, the market potential would shoot north of $500 million, according to Bob Hopkins, an industry analyst with Bank of America.

Abbott plans to submit the CREST data to FDA to support expanded labeling of its Acculink stent. The firm also will use the data to request that CMS expand its national coverage policy to include all or most patients at average risk for surgery.

Many agree that CREST will be an impressive data set for CMS, but not everyone thinks that across-the-board coverage for stenting is a slam dunk.

CMS has shown significant caution with carotid stenting in the past. Three times since it set its current policy in 2005 the agency has considered and declined to incrementally expand coverage to more high-risk patients included in FDA labeling (¹ Also see "CMS Clarifies Medicare Policy For Embolic Protection With Carotid Stenting" - Medtech Insight, 14 December, 2009.).

SCAI's Cates, who has been closely involved with the prior coverage requests, is optimistic CMS will see things differently this time, but, he acknowledges that "it is hard to know, because I would have said there would have been some modification of the very limited coverage policy the last three times we tried to open this up."

Also, some surgeons say the higher stroke rate in the CREST stent group should not be ignored.

"From the standpoint of the practitioner and the patient, the complication you most worry about in a carotid intervention is stroke," said Richard Cambria, chief of vascular and endovascular surgery at Massachusetts General Hospital, and VP of the Society of Vascular Surgeons.

He also points out that the safety track record for surgery goes back longer and that similar surgical results to CREST have been reported in smaller trials, while this is the best result ever for stenting.

"I do think it is reasonable for some expansion of reimbursement, but I do not think it is reasonable for a blanket, unqualified, unrestricted reimbursement," he said in an interview.

At minimum, it might make sense to reinstitute a previous mandate requiring patients to see both an interventionalist and a surgeon prior to getting a procedure, he suggested.

"The interventionalist who has one treatment to offer the patient is going to offer that treatment, but vascular surgeons are in the very favorable position with respect to the patient because we can offer either procedure," Cambria said.

European Data Supports Surgery Over Stenting

The other challenge before CMS will be to analyze CREST in the context of other evidence. Large studies in average-surgical-risk patients reported prior to CREST have primarily come out of Europe and support surgery over stenting.

The French multi-center EVA-3S study was halted early in response to safety concerns in the stent group. The 1,183-patient SPACE trial data, reported in 2006, failed to show non-inferiority for stenting compared to surgery (² Also see "SPACE For Carotid Stents? Data Does Not Support Expanded Use In Near Term" - Medtech Insight, 16 October, 2006.).

And, most recently, the interim 120-day analysis from the ICSS study, posted online by The Lancet Feb. 26, found a statistically significant higher rate of the death, stroke or heart attack endpoint in the stent group compared to surgery (8.5% vs. 5.2%). That study, designed to collect three-year follow-up data, includes only symptomatic patients.

Further, the ICSS researchers included a meta-analysis of the EVA-3S, SPACE and ICSS trials in their study, which clearly favored surgery.

And in an ICSS subset analysis published online the same day in Lancet Neurology, magnetic resonance imaging showed that about three times more patients had new carotid artery lesions following treatment in the stent group than in the surgery arm.

Longer-term data is needed, but, "In the meantime, carotid endarterectomy should remain the treatment of choice for patients suitable for surgery," the ICSS team writes. Carotid stenting advocates suggest, however, that the CREST data is superior to the European trials, and should change that conclusion.

For one, they stress CREST's extensive lead-in phase during which interventionalists, surgeons and others were credentialed on the techniques before being allowed to participate in the study.

"This is a little different than recent European trials, where you have to be an experienced surgeon in many ways, but you [can be] a much less experienced interventionalist to use carotid stenting, which we know ... is a very skill-dependent therapy," Cates said.

Also, CREST supporters note that the different outcomes between the two treatment groups is principally due to an excess of "non-disabling" strokes in the stented patients, while the rate of disabling stroke or death is similar in the two groups.

Finally, they point to the fact that embolic protection devices, designed to prevent potential clot-causing debris from entering the blood stream during an intervention, were not required in EVA-3S, SPACE and ICSS trials, but were mandated in CREST.

Embolic protection systems are now standard in the U.S., and are required for Medicare coverage, but the Lancet Neurology MRI subset analysis questions their utility. That study found higher rates of post-procedure lesions in patients treated at facilities that required protection devices than in patients treated elsewhere.

CMS will need to weigh all of these factors. And Klaus Gröschel, a neurologist at the University of Göttingen, Germany, who wrote an editorial in Lancet Neurology touting the ICSS results prior to the CREST presentation, suggested more consideration is needed before declaring the debate settled.

"In my humble opinion, one should wait for the final publication of the CREST study to appropriately incorporate the results of this high quality data into the current context of literature," he said in an interview.

But Cates is confident that CREST will have the strongest impact on CMS because of the trial design and because it will heavily influence the pending carotid artery disease clinical guidelines.

He notes that the committee delayed completing the guidelines expressly to wait on CREST. "That," he declared, "is the importance of the CREST trial."

- David Filmore ( ³ d.filmore@elsevier.com )

Carotid Stenting: CREST Makes A Splash, But Market Will Depend On CMS

More from Archive

More from Medtech Insight