Sentinel Embolic-Protection Device at TAVR Fails to Show Brain Benefit

November 03, 2016

WASHINGTON, DC — Are surrogate end points enough to recommend a treatment that seems safe but hasn't "worked" by clinical-trial standards, but many agree ought to work?

A pair of catheter-delivered safety nets positioned before transcatheter aortic-valve replacement (TAVR) in arteries connecting the heart to the brain consistently trapped potentially embolic debris released by the procedure in a small randomized trial[1]. The study nonetheless failed to show prospectively that the strategy limits cerebral injury or preserves cognitive function.

But an analysis of the study testing the device (Sentinel TCEP, Claret Medical), which Dr Susheel Kodali (Columbia University Medical Center, New York, NY) included with his presentation of the SENTINEL study here at TCT-16, was not prespecified but associated it with a reduced volume of cerebral lesions by MRI, compared with TAVR without transcatheter cerebral embolic protection (TCEP).

Post Hoc Significance

Sentinel TCEP [Source: Claret Medical]

That finding came after post hoc adjustment of the primary results for possible confounders that had emerged: the type of TAVR valve implanted and baseline cerebral lesion volume in brain territories shielded by the TCEP device.

In particular, reduction in new cerebral lesion volume with TCEP vs no TCEP, the primary efficacy end point, reached 42% but was not significant (P=0.33). But, Kodali said in his presentation, "When you look at the adjusted model, the difference was 49% and the confidence intervals were smaller, and it did result in a significant difference."

He also pointed to a reduction in risk of stroke at 30 days from 9.1% without to 5.6% with TCEP, "which was not statistically significant, but it's important to note that the study was not powered for clinical end points."

The SENTINEL study was published in the Journal of the American College of Cardiology, with Dr Samir R Kapadia (Cleveland Clinic, OH) as lead author, at about the same time as Kodali's presentation.

Claret Medical announced on September 20, 2016 that it has filed a market application for the device by the FDA.

The study's evidence that the Sentinel might protect the brain from injury during TAVR, although not prospective or from a big trial with clinical end points, led some experts to voice support for offering embolic-protection devices to some patients slated for TAVR.

Some Hopeful, Others Enthusiastic

In general, but not always, such support for having TCEP available for patients was contingent on a good showing for the device in bigger, more definitive trials. It also stemmed from the relative safety it showed in the trial, the perception that such a device ought to protect the patients from distal embolization of debris emitted by the valve procedure, and the catastrophic nature of stroke should one occur.

As a panelist after Kodali's presentation, Dr Eric Peterson (Duke University Medical Center, Durham, NC) said that "the study is inconclusive because of the small sample size and the fact that the P value doesn't reach significance, but it is remarkably promising. If you look at reduction in the clinical end point from 9.1% to 5.6%, again, that's very hopeful."

Still, "What I would like to see is a larger follow-up study that would confirm these results. In the interim its nice to see that the safety profile actually looks pretty good," Peterson said.

"So it is going to be a challenge for clinicians to know whether to implement this now and wait a little bit for the real evidence to come behind it or wait for the full larger study."

Dr Robert O Bonow (Northwestern University Feinberg School of Medicine, Chicago, IL), also on the panel, said, "I would also agree that this is very promising thus far."

Pressed further by moderator, Dr Gregg W Stone (Cardiovascular Research Foundation, New York, NY), Bonow said he is "leaning toward" favoring use of the TCEP device for TAVR neuroprotection, but "I still think as an academic person as well as one who sees patients, I'd like to see more data."

At a press briefing on the study, discussant Dr Jeffrey J Popma (Beth Israel Deaconess Medical Center, Boston, MA) said, "These are very small numbers to be making any kind of conclusions about whether this is actually going to be useful."

But at the same venue, Dr David R Holmes Jr (Mayo Clinic College of Medicine, Rochester, MN) seemed enthusiastic about the study's findings and urged that consideration go beyond P values to include the patient's point of view.

"Patients are going to say, why wouldn't you use a device that captures 99% of stuff that goes to the brain, that at least in this small underpowered study decreases the rate of neurologic events by about 40%?" he proposed. "Well, why wouldn't you use that in everybody?"

Holmes continued, "I think cerebral protection is the real deal. I think its going to make a huge impact" and that patients are going to ask for it.

Thrombus, Calcium, Bits of Valve, and Artery

The Sentinel trial randomized 363 patients undergoing TAVR for severe, symptomatic aortic stenosis to have same-procedure placement of the TCEP device (n=123), Sentinel placement with before-and-after cerebral MRI (n=121), or such imaging without TCEP (n=119).

Dr Susheel Kodali

As Kodali described, 99% of the retrieved filters contained debris liberated by the valve procedure, most of which consisted of thrombus and bits of arterial wall, but almost half was made up of calcification particulates and valve tissue. The published report says that >80% of the debris had a maximum diameter of 150 to 500 µm, with <5% wider than 1000 µm.

The rate of major adverse cardiac and cerebrovascular events (MACCE) at 30 days, the primary safety end point, was 7.3% in the two TCEP arms combined and 9.9% in the non-TCEP control arm (P=0.41).

The primary efficacy end point, median total new lesion volume by before-and-after MRI, was 102.8 mm3 for TCEP vs 178.0 mm3 in controls (P=0.33). That applied to the "protected" brain territories served by the vessels containing the TCEP filters, the brachiocephalic artery and left common carotid artery.

Nor was postprocedure lesion volume across all brain territories statistically different, 294 mm3 for TCEP vs 309.8 mm3 for controls (P=0.81), according to the published report.

In the post hoc multivariate analysis, white-matter lesion volume at baseline and valve type were significant predictors of new-lesion volume after TAVR. The trial used four different valves, including the Sapien 3 (52.4% of cases) and Sapien XT (17.8%) (Edwards Lifesciences); and the Evolut R (25.9%) and CoreValve (3.9%) (Medtronic). They were similarly distributed across the randomization groups, write the authors.

The finding doesn't say one valve type is better than another, not with these numbers, Kodali said. And the analysis couldn't adjust for physician decisions to use one TAVR valve over another.

"We saw that there was [a protective] effect at all valve types. It's just the magnitude of which was different," he said. "But the reality is that these valves interact differently" with the TCEP part of the procedure.

Neurocognitive Protection?

Neurocognitive scores weren't significantly different at baseline or at 30 or 90 days after TAVR. But changes in neurocognitive scores to 30 days were significantly related to mean new lesion volume in the "protected" territories and in all territories (P= 0.011 and P=0.003, respectively).

"We have a clinical correlation between neurocognitive function and embolic debris, [and] we have other neurologic literature suggesting that a decrease in silent infarctions does improve clinical outcomes," Kodali pointed out at the press briefing.

Commenting for heartwire from Medscape on other aspects of the procedure, Kodali said TCEP placement and retrieval by operators without a lot of experience adds about 9 minutes to the overall TAVR procedure. When experienced operators do it, he said, TCEP adds only 3 to 5 minutes.

Any small additional risk from adding TCEP to TAVR should be well worth the likely protection against strokes, according to Kodali. Without TCEP, "one out of 11 patients had a stroke—that's not trivial."

More Trial Scrutiny?

Stone tried to crystallize the panelists' collective take on the TCEP device. "I think we are all absolutely intrigued and hopeful about the results of this study. I think it's going to come down to, do we have enough data to identify which patients in the current era might benefit from cerebral protection, as opposed to a one-size-fits-all strategy?"

At the press briefing, Kodali emphasized that, given the seriousness of stroke as a possible TAVR complication and the proof that the device catches a lot of debris, "for a lot of people, it's the safety of the device" that matters most.

For perspective, he said, "There's no randomized trial showing a benefit of embolic protection in carotid stenting. There are reasons for that; I don't think it's justified to do a randomized trial with that now."

Proposed Holmes, "You can look at a study like this statistically, and say it did not meet its primary end point, so everything else is hypothesis-generating; or you could look at it from the patients' standpoint."

A conversation with the patient might include that the device is safe, and "we know that it captures things that go to your brain. It cannot be true that debris going to your brain is a good thing," Holmes said. "So patients are going to say, 'Well why would you then not use a device like this?' "

And, he said, "nobody's going to do a study that's going to wait 20 years to see if there's dementia. That's not going to happen."

A patient-level pooled analysis of studies of cerebral embolic protection at TAVR "suggests that the Sentinel double-filter significantly reduces total new lesion volume in protected regions by approximately 100 mm3 of damaged brain," according to an accompanying editorial[2].

Given the likely difficulties of a randomized trial of the device, "the small imaging sample size is probably the main reason the SENTINEL study is negative, but this does not equate to saying that there is no role for [TCEP]," write Dr Azeem Latib (San Raffaele Scientific Institute, Milan, Italy) and Dr Matteo Pagnesi (EMO-GVM Centro Cuore Columbus, Milan, Italy).

The SENTINEL Trial was funded by Claret Medical. Kodali discloses consulting for Edwards Lifesciences. Kapadia had no relevant financial relationships. Disclosures for the coauthors are listed in the paper. Popma disclosed grant support, consulting fees, and equity in various companies. Stone reports receiving consulting fees from Velomedix, Toray, Matrizyme, Miracor, TherOx, Reva, V-Wave, Vascular Dynamics, Ablative Solutions, Neovasc, and Medical Development Technologies; serving as a consultant on prasugrel patient litigation paid for by Lupin Pharmaceuticals; and holding equity, stock options, or both in the MedFocus family of funds, Guided Delivery Systems, Micardia, Vascular Nanotransfer Technologies, Cagent, Qool Therapeutics, Caliber Therapeutics, Aria, and the Biostar family of funds; and that Columbia University receives royalties from Abbott Vascular for the sale of the MitraClip. Peterson reports receiving grant support from Eli Lilly, Janssen, and the American Heart Association and consulting fees from Boehringer Ingelheim, Bristol-Myers Squibb, Janssen, Pfizer, and Genentech.

Follow Steve Stiles on Twitter: @SteveStiles2 . For more from, follow us on Twitter and Facebook .


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.