An FDA advisory committee voted in favor of approving St. Jude Medical's Amplatzer device, despite an agency review suggesting inconclusive evidence for its safety and effectiveness.
The panel voted 15-1 in favor of the safety of the Amplatzer and 11-5 that its benefits outweigh the risks. Votes also leaned towards its effectiveness, although fairly split at 9-7.
Members spent much of the day evaluating the RESPECT trial, which failed to find superiority of the patent foramen ovale (PFO) occluder over medical management for the prevention of recurrent cryptogenic strokes (risk reduction 0.534, 95% CI 0.234-1.220). Among other problems, RESPECT had been plagued by a high rate of patient withdrawal, a low number of analyzable events, and the lack of confirmation that the PFO was actually closed during all interventions.
"Do we throw away 9 years of effort because the study was underpowered?" posed discussant , of the Cleveland Clinic. "Or is it more rational to try to get a better understanding and reality of the treatment effect and magnitude?"
"I don't think we'll see another randomized trial like this in the near term. There is a directional change towards reduction in unexplained stroke," , of Atlanta's Piedmont Heart Institute, said, citing the approximate halving of such events in the long run.
"Doing a randomized trial on this is extremely difficult, and doing another one now wouldn't change the results," agreed , of Rush University Medical Center in Chicago, who had advocated Amplatzer on behalf of the Society for Cardiovascular Angiography and Interventions.
Kavinsky ultimately told 51˶ that he was "pleasantly surprised" by the voting results. "But this was right."
Naftali Z. Frankel, MBA, the consumer representative on the committee, made the point that a non-statistically significant reduction can still signal a benefit -- no matter how small -- to patients.
"There is no small number when it comes to the catastrophic result of a stroke," he said. "From a statistical point of view, in a sterile sense, something may not seem significant; but from the patient's vantage point it is significant."
"Clinicians have to be crystal clear that by no means is this an alternative to medical treatment," he concluded. "The overall trend is that there is a benefit, and the question is to what extent?"
And for whom?
"I am convinced that there is a benefit in selected patients," said committee member , of California's Oakland Kaiser Medical Center. With so little information from RESPECT, he suggested, "the trouble is selecting the proper patients."
Even so, there are slight signals that younger patients and those with large PFOs or atrial septal aneurysms stand to derive the most benefit from PFO occlusion, the panel overall agreed.
Then there was the suggestion that PFOs may not really be behind unexplained strokes.
"The mechanism for risk is still largely hypothetical, and 'parodoxical embolism' is being used synonymously with having a PFO. The causal association is often drawn and may not be accurate," said , of Rhode Island Hospital in Providence.
Still, Furie later voted in agreement with the statement that the Amplatzer's benefits can outweigh its risks. "I see a clinical need to this device," she explained.
As was the case with many committee members who voted yes, she tempered that by saying effectiveness hangs on the ability to find the right patients for the device.
Unexplained strokes mark "a strong clinical need that needs to be met," Kavinsky concluded. He added that FDA approval of the PFO occluder should include mechanisms to curb overuse and abuse, such as mandatory participation in a registry.
When asked how much time is enough in the real world to determine the right patients for PFO closure, he responded: "With event rates so low, I almost don't have an answer, but I would think 5 years at least."
There is no standard of care for patients who do suffer cryptogenic strokes. Medical management strategy -- including antiplatelets and/or anticoagulation -- is varied and controversial, according to Kandzari, despite society recommendations.
Disclosures
Kavinsky declared no relevant conflicts of interest.