Enrollment in the first trial designed to directly compare the risks and benefits of intracranial stenting over medical management was halted early on April 5 after patients in the stenting group were found to have more than double the 30-day risk of death or stroke than those in the medical therapy group.
Stenting patients did worse than those previously reported in registries, while patients on medical management did better than those in previous studies. Yet neurologists familiar with the findings called the results not entirely surprising, and said they should lead to a sharp drop in the popularity of intracranial stenting, at least until full details of the study can be analyzed and published.
NINDS Deputy Director Walter J. Koroshetz, MD, cautioned, however, against making too much of the decision to suspend recruitment until follow-up of all patients can be completed and the full results analyzed and published.
“I don't know that this decision [to suspend enrollment] closes the book on anything,” said Dr. Koroshetz. “The study was designed with the goal of seeing a 35 percent reduction in the rate of stroke in the stenting group. That was the bar they set for themselves. The probability of hitting that bar became so low that the concern about risk dominated their decision to close it.”
The Stenting and Aggressive Medical Management for Preventing Recurrent stroke in Intracranial Stenosis (SAMMPRIS) study had recruited 451 (59 percent) of the planned 764 patients at 50 participating sites in the United States when recruitment was stopped by the NINDS. The decision was based on the study's statisticians' calculation that 14 percent of patients treated with angioplasty combined with stenting experienced a stroke or died within the first 30 days after enrollment compared with 5.8 percent of patients treated with medical therapy alone — a highly significant difference.
Beyond 30 days, the rates of stroke in the area of the stenotic artery were similar in the two groups, but because fewer than half of the enrolled patients had been followed for one year, follow-up will continue and the final results published when available, NINDS noted in a press release announcing the suspension of SAMMPRIS.
“It will be interesting to see if those curves ever do cross over,” said Wade S. Smith, MD, PhD, director of the neurovascular service at the University of California-San Francisco.
Until those final results are published, however, the findings “are a strong word of caution that this [intravascular stenting] might not be the best thing to do,” said Dr. Smith, who was not involved in the trial, although UCSF was a participating site.
But Dr. Koroshetz said that until the study is completed and published, he will remain agnostic about why higher than expected event rates were seen in the intervention group, and lower than expected rates in the medical group.
“We'll have to sort that out. This is the end of the enrollment phase, but just the beginning of analysis.”
The chair of the Data and Safety Monitoring Board agreed that the final verdict on the study must await its publication, but emphasized that the low rates of stroke and death for patients on medical therapy were important.
“Medical therapy appears to be improving the outcomes of patients,” said James C. Grotta, MD, professor and chair of neurology at the University of Texas Medical School in Houston and director of the stroke program at Memorial Hermann Hospital.
“These data will provide a benchmark for expected outcomes with aggressive medical therapy,” he added. “Any future study will have to beat some pretty tough competition.”
Medical management in both arms of SAMMPRIS consists of 325 mg of aspirin per day for the entire follow-up, 75 mg of clopidogrel for 90 days after enrollment, intensive management of vascular risk factors, with a target of systolic blood pressure less than 140 mm Hg (less than 130 mm Hg if diabetic) and LDL less than 70 mg/dl), and provision of a lifestyle modification program to all study patients.
The angioplasty and stenting system used in SAMMPRIS was the Gateway-Wingspan system.
Compared to the estimated 30-day rate of stroke or death of between 5.2 percent and 9.6 percent seen in stenting registries, Dr. Grotta called the 14 percent rate in SAMMPRIS “slightly surprising.” But, he said, “This is why randomized trials are necessary. When you consider that this is intracranial disease, and these are small vessels, we need studies like this to demonstrate the true risk of these procedures.”
An interventional neurologist who was not involved in the study said that while he is disappointed by the high complication rate seen in SAMMPRIS, the findings are important enough to change practice patterns.
“This is really big news within the field of stroke, particularly because there has been a recent surge in neurologists wanting to train to do these procedures,” said Robert Taylor, MD, assistant professor of neurology, neurosurgery and radiology at the University of Minnesota. “I do this procedure, and I'm very disappointed to see these high complication rates. But it means that intracranial stenting is not yet ready to be a widespread procedure. The frequency of this procedure is going to go way down, and it will likely reduce the number of neurointerventionalists we need to train.”
While some centers might have lower complication rates than those seen in SAMMPRIS, he said, “There really ought to be strict monitoring of centers to show that they meet a minimal procedure-related complication rates, prior to their participation in this kind of trial in the future.”
Marc I. Chimowitz, MBChB, professor of neurology at the Medical University of South Carolina and the neurology principal investigator of the SAMMPRIS executive committee, declined to comment until final results of the study are published.
Another multicenter, randomized clinical trial is continuing despite the results seen in SAMMPRIS. The Vitesse Intracranial Stent Study for Ischemic Therapy (VISSIT), designed to include up to 250 patients at 30 sites in the US, Europe and China, is comparing use of the Pharos Vitesse stent system to aspirin and clopidogrel.
“VISSIT is going to present very useful data, to find out if the medical arm achieves the same benefit, and whether there's the same excess morbidity in the stenting arm,” said Dr. Smith. “One might even be able to pool the data of the two trials, once they're both published, to get a better sense overall.”