Article In Brief
Vascular neurologists are questioning the US Food and Drug Administration decision to expand the indication of a stent for carotid endarterectomy. They contend that there are not enough data to support the decision and that it is premature.
The US Food and Drug Administration's (FDA) recent decision to expand the indication for carotid revascularization using the TransCarotid Artery Revascularization (TCAR) procedure is puzzling some stroke neurologists.
The procedure, which places an Enroute stent after entering via the common carotid artery, was originally approved in 2015 for patients with either symptomatic or asymptomatic carotid stenosis who are at high risk from surgical carotid endarterectomy (CEA), long the gold standard for treatment of carotid stenosis that poses a significant risk of stroke. The FDA's May 2 decision expands the indication to include patients at standard risk from CEA as well.
The decision was evidently based on a supplemental premarket approval application from the manufacturer Silk Road Medical, which included surveillance data extracted from the Vascular Quality Initiative (VQI). According to a company release, those data, which include real-world outcomes for 20,264 patients considered to be at standard anatomic and physiological risk from surgery, found TCAR to be statistically non-inferior in stroke and death outcomes to CEA, while showing a ninefold reduction in cranial nerve injury: 2.7 percent vs 0.3 percent, p=<0.001.
“Pairing the right patient with the right treatment results in significantly improved physician and patient experiences and outcomes. I speak for the vascular community in welcoming this label expansion for TCAR and recognizing this as a vital advancement in the treatment paradigm for patients at risk of stroke,” said Dr. Marc L. Schermerhorn, chief of the division of vascular and endovascular surgery at Beth Israel Deaconess Medical Center, in a statement accompanying the FDA decision.
“The decision regarding which patients to treat with TCAR is no longer restricted to patients at high surgical risk, providing a greater opportunity for the care-team to pursue the less invasive approach for a broader set of their patients. At Beth Israel Deaconess, going forward, we expect that the right treatment for the majority of our atherosclerotic carotid disease patients will be TCAR.”
Is the Expanded Indication Premature?
But not everyone in the vascular community agrees, with some arguing that the expansion is premature and that there is not enough high-quality, independent data to support it. While the original TCAR approval in high-risk patients was based on the results of a randomized controlled trial, the 2015 ROADSTER study, the expanded indication is based on registry data.
“The VQI is used for a lot of vascular data collection, and it is certainly higher-quality data than just hospital self-reporting, but it's not a randomized controlled trial,” said Ashutosh P. Jadhav, MD, PhD, a vascular and interventional neurologist at Barrow Neurological Institute in Phoenix.
“When our committee looked at the totality of the data, we found that the evidence that TCAR is useful is uncertain. That is definitely a concern.”—DR. SEEMANT CHATURVEDI
“We can't really know with appropriate rigor whether TCAR is non-inferior in this setting,” Dr. Jadhav said. “Non-inferiority is a pretty strong claim and is usually based on a non-inferiority randomized controlled clinical trial design. Registry data like VQI is attendant with the typical problems of non-randomized data—patient selection is a major confounder as there may be bias to treat patients with a particular modality based on their individual anatomy, age, morbidity, and acuity.”
Seemant Chaturvedi, MD, FAAN, Stewart J. Greenebaum Endowed Professor of Stroke Neurology at the University of Maryland School of Medicine in Baltimore, was a co-author of the American Heart Association/American Stroke Association's 2021 guideline for the prevention of secondary stroke in patients with stroke and transient ischemic attack (TIA), which included an assessment of the evidence for TCAR. “When our committee looked at the totality of the data, we found that the evidence that TCAR is useful is uncertain,” Dr. Chaturvedi said. “That is definitely a concern.”
“They do have a large amount of registry data which is certainly supportive, and that is one of the criteria that the FDA can use for the PMA [premarket] process with a device,” says Larry B. Goldstein, MD, FAAN, the Ruth L Works Professor and Chair of the department of neurology at the University of Kentucky, co-director of the Kentucky Neuroscience Institute, and interim director of the UK-Norton Healthcare Stroke Care Network.
“That said, I'm always hesitant because when you don't have contemporaneous control data, there is always a risk of reaching inappropriate conclusions,” Dr. Goldstein said.
“We went through a similar process with stenting for symptomatic cranial stenosis. It appeared to make sense considering the natural history, until a prospective, randomized controlled trial showed that not only was it not effective, but it also put patients at a significantly increased risk of stroke or death in the first 30 days compared with the best medical therapy.”
That study, the SAMPRISS trial, is the only one to date that has compared the risks and benefits of stenting with best medical management for intracranial arterial stenosis. No trial has compared stenting with best medical management for carotid stenosis.
Dr. Jadhav noted that two major trials have established the non-inferiority of stenting in general (via the traditional transfemoral approach) compared with CEA: the 2016 Asymptomatic Carotid Trial (ACT) I, which randomized patients with asymptomatic severe carotid stenosis who were not at high surgical risk to carotid stenting with embolic protection versus CEA, and the 2010 Carotid Revascularization Endarterectomy vs. Stenting Trial (CREST), which included both symptomatic and asymptomatic patients. “I think most practitioners in this space believe the non-inferiority question for stenting vs. CEA has been largely settled,” he said.
The more critical question, he suggested, is whether revascularization via either CEA or stenting is superior to medical management in asymptomatic patients. The majority of the estimated 100,000 carotid interventions performed per year in the US —either surgical or stenting—are in asymptomatic cases.
“They do have a large amount of registry data which is certainly supportive, and that is one of the criteria that the FDA can use for the PMA [premarket] process with a device. That said, I'm always hesitant because when you don't have contemporaneous control data, there is always a risk of reaching inappropriate conclusions.”—DR. LARRY B. GOLDSTEIN
“If you look at the National Inpatient Sample data, for every 100 carotid interventions being performed, 92 of them are in asymptomatic patients,” Dr. Jadhav noted. “That's the highest rate of any country in the world. There are countries in Europe where these interventions are not performed in asymptomatic individuals at all.”
The benefit of CEA in asymptomatic patients was established in the mid-1990s with the Asymptomatic Carotid Atherosclerosis Study (ACAS), which found a significantly lower five-year risk of ipsilateral stroke and any perioperative stroke or death for the surgical group (5.1 percent) vs medical therapy (11 percent). The reduction in five-year ipsilateral stroke risk in the surgical group was 53 percent of the estimated five-year risk in the medical group (p = 0.004). But that trial was conducted in a different era for medical management, Dr. Jadhav pointed out. “At the time, we had no Plavix, no statins, and even blood pressure control medical therapy was not as well attended to.”
The Carotid Revascularization and Medical Management for Asymptomatic Carotid Stenosis Trial (CREST), currently underway, is comparing intensive medical management to stenting or CEA in combination with medical management, in two parallel multicenter randomized, observer-blinded trials enrolling a total of 2,480 asymptomatic patients with carotid stenosis of 70 percent or greater.
“Is best medical therapy today as good as stenting or CEA? That's a major question to answer given the way we practice now,” Dr. Jadhav said. Many people think it probably is based on registry data.”
A comparative effectiveness study published in JAMA Neurology in 2020, for example, found that the absolute reduction in the risk of fatal and nonfatal strokes associated with early CEA was less than half the risk difference in trials from 20 years ago, and was no longer statistically significant when the competing risk of nonstroke deaths was accounted for in the analysis.
Dr. Goldstein agrees. “Medical therapy has evolved significantly in the decades since the trials of endarterectomy compared with medical therapy were done, and we don't know that either procedure is necessarily of benefit in the current environment, “he said.
The intensive medical management regimen in CREST-2 consists of aspirin 325 mg/day and intensive risk factor management, primarily targeting systolic blood pressure less than 130 mm Hg (initially systolic blood pressure less than 140 mm Hg) and -density lipoprotein cholesterol less than 70 mg/dL. Secondary risk factor targets focus on tobacco smoking, non-HDL (high-density lipoprotein), HbA1c (hemoglobin A1c), physical activity, and weight.
CREST-2 has not yet finished enrollment, and Dr. Chaturvedi expressed some concern that the new expanded indication for TCAR could hamper that enrollment. “CREST-2 needs about 400 more patients, and it's possible that some patients who are eligible for the trial might instead be funneled to TCAR,” he said. “We don't know for a fact this will happen, but it is a potential concern.”
He encouraged colleagues to recommend that patients with asymptomatic carotid stenosis enroll in CREST-2 if possible. “If they are not eligible, there needs to be an individual discussion as to the potential benefits and risks of revascularization. For symptomatic patients who meet the criteria, and especially those with recent symptoms and more than 70 percent stenosis, I would favor endarterectomy over TCAR until we have more solid, independent data.”
“When CREST-2 is completed, depending upon what it shows, one might question whether any procedure should be done for patients with asymptomatic disease, despite the FDA approval,” Dr. Goldstein said. “But once something is approved and being used, the genie is out of the bottle, and it's hard to put it back. It would have been helpful if these 20,000 patients in the registry had been randomized. To my mind, this is a lost opportunity.”
Dr. Jadhav had no disclosures.