Majority of FDA Advisors Give Green Light to Amplatzer PFO Closure Device for Stroke

(UPDATED) Patent foramen ovale (PFO) closure using the Amplatzer device for the prevention of recurrent stroke should be approved as an option for patients with cryptogenic stroke, the majority of a team of experts assembled by the US Food and Drug Administration (FDA) agreed Tuesday. The mixed vote comes after two decades of controversy, failed trials, and abandoned devices for a strategy that all but the most enthusiastic operators had stepped away from, given the apparent futility of studies that have sought to show a benefit for the therapy.

After a day spent listening to presentations from manufacturer St. Jude Medical and agency representatives, as well as input from patients and professional organizations, the Circulatory System Devices Panel voted 15 “yes” to 1 “no” on a question of device safety, 9 to 7 on the question of effectiveness, and 11 to 5 on the question of whether the potential benefits of the device outweigh the risks. There were no abstentions on any of the questions.

“I was encouraged by the fact that the term used in [the FDA’s] questions was ‘reasonable assurances’ rather than statistical certainty, if such a thing can be attained,” Jeffrey Brinker, MD (Johns Hopkins Hospital, Baltimore, MD), commented. “There is a real clinical need—even if the group turns out to be much smaller than what was studied—for some device to be helpful for the pathophysiology that most certainly exists in some portion of these patients with cryptogenic stroke.”

Support for the Amplatzer device comes from the RESPECT trial, first revealed at TCT 2012, which failed to show a benefit in the intention-to-treat population, although “as-treated” and “per-protocol” analyses were positive.

At TCT 2015, investigators returned with mean follow-up in the range of 5 years for both the medically managed and device-treated groups, and zeroed in specifically on the outcome of cryptogenic stroke. With the extended follow-up, investigators saw a 54% relative risk reduction in stroke recurrence among device-treated patients.

Not surprisingly interventional cardiologists on Tuesday’s panel were almost unanimously in favor of recommending approval, while neurologists Karen Furie, MD, PhD (Rhode Island Hospital, Providence); Seemant Chaturvedi, MD (University of Miami Miller School of Medicine, Miami, FL); and Patrick Noonan, MD (Scott and White Memorial Hospital, Temple, TX), all voted “no” on efficacy, although Furie voted yes on the benefit/risk question. “Although it may seem illogical,” she said, “I voted yes . . . because I also see a clinical need for this device and it’s my hope that a population that would benefit from its use can be identified.”

By contrast, Chaturvedi voted no on all three questions, saying his concerns over the increased rate of DVT and pulmonary embolism in the device-treated patients were not assuaged by the data he saw, nor was he remotely impressed by the data on efficacy and benefits versus risks. “I would just reiterate that we are in the world of evidence-based medicine and we have two trials with this device, both of which failed the primary endpoint,” he stressed.

Both statisticians on the panel also voted no on the efficacy and benefit/risk questions, citing similar reasons.

During their daylong deliberations, the FDA panel delved deep into the high dropout rate in the study and the lack of benefit in the intention-to-treat population as compared with the statistically significant benefit seen in the per-protocol, as-treated, and device-in-place analyses. Also troubling to the panel was how cryptogenic stroke was determined at the outset, what role anticoagulation and adherence to other medications may have played, the possible importance of DVT and atrial arrhythmias—both of which were higher among device-treated patients—and the completeness of PFO closure. As the FDA review showed, residual shunts were not uncommon, occurring in 29% of RESPECT subjects.

Ultimately, the bulk of the panel members concluded that they wanted the device to be available for patients, particularly given the number of products that have long used off-label for this indication.

“Patients and clinicians need to have this device available as opposed to using other devices, that probably are not as good, in an off-label fashion,” said Ralph Brindis, MD (Oakland Kaiser Medical Center, Oakland, CA). “If I was on the writing committee for this particular guideline, I would give it a IIb indication, rather than a III. And my hopeful plea is that aggressive patient choice, especially looking [to rule out patients with] occult or atrial fibrillation, is done in making sure we chose the right patients for this device.”

Brindis was referring, in part, to current American Heart Association/American Stroke Association (AHA/ASA) guidelines that assign PFO closure a class III, or “do not do,” recommendation. During the open public comment period, renowned PFO researcher Shunichi Homma, MD (Columbia University Medical Center, New York, NY), pointed to the AHA/ASA position and urged FDA reviewers to say no to Amplatzer, calling it “premature.” He also noted that new recommendations from the American Academy of Neurology are pending and suggested that the FDA review those before making its final decision.

This is the fourth time PFO closure for stroke has been considered in some fashion by the FDA’s Circulatory System Devices Panel over a span of nearly two decades. The entire strategy has proved to be one of the most polarizing topics in cardiovascular medicine, often pitting cardiologists in favor of closure against neurologists dead set against it. The field has seen at least one device manufacturer declare bankruptcy and other companies abandon device development; clinical trials that struggled to enroll sufficient patients only to deliver negative and ambiguous results; and untold numbers of patients who sought PFO closure off-label or in countries where the devices are approved for use.


CORRECTION: An earlier version of this story incorrectly credited Homma with implying that the forthcoming American Academy of Neurology position does not support PFO closure. In fact, Homma did not refer to specific content in the upcoming guidelines but rather urged the FDA to review the guidance before making its final decision.

Related Stories:

We Recommend