乌鸦传媒

More data revealed more failure for the Watchman left atrial appendage (LAA) closure device, but, again, interventional cardiologists attending the meeting were scouring the rubble for positive signals.

The latest run at Watchman comes from a meta-analysis of the pivotal PROTECT-AF trial and the confirmatory PREVAIL trial presented at TCT by , of the Mayo Clinic.

The PREVAIL patients were slightly older (74.0 versus 71.7 years), had more diabetes (33.8% versus 24.4%), and more had a prior stroke or transient ischemic attack (27.5% versus 17.7%).

The analysis showed no efficacy benefit for Watchman overall, with a hazard ratio of 0.79 and P=0.22. The same was true for:

• Stroke or systemic embolism (HR 1.02, P=0.94)
• Ischemic stroke or systemic embolism (HR 1.95, P=0.05)
• Major bleeding (HR 1.00, P=0.98)
• All-cause death (HR 0.73, P=0.07)

But Watchman did beat warfarin on hemorrhagic stroke (HR 0.22, P=0.004), major bleeding beyond 7 days (HR 0.51, P=0.002), and cardiovascular or unexplained death (HR 0.48, P=0.006).

When "PREVAIL-like" patients in PROTECT-AF were compared with PREVAIL patients, the efficacy was similar on both efficacy endpoints.

That was the reasoning favored by , director of invasive structural interventions at Columbia University Medical Center and a PREVAIL investigator.

"Why did PROTECT-AF show a clear benefit of [LAA] closure versus warfarin and PREVAIL did not?" he said in an interview. "The ischemic stroke rate [with warfarin] in PREVAIL was substantially lower ... than the rates in a series of other trials using warfarin."

"Now, we have this larger experience, and the difference in PREVAIL versus PROTECT-AF is that in PROTECT-AF the warfarin patients had the number of strokes they should have had."

Sommer said he is convinced that Watchman is effective and he would "absolutely" use it if it were approved by the FDA.

"PROTECT-AF has 5-year data. Five years later it is not only noninferior, it is superior to warfarin," he said. "PREVAIL is interesting only because the Coumadin patients did so much better than anticipated, and that is probably the big reason why there is a difference between the two studies.

"PREVAIL is newer and doesn't have as much long-term follow-up. I am quite convinced that, going forward, the curves of the event rates -- survival curves, etc. -- will diverge over time ... Provided that the data are what PROTECT-AF show -- that ischemic stroke rates are no different -- it has to be a win in the long run."

Sommer said PREVAIL was not undertaken to prove efficacy, just safety.

"The whole reason they did PREVAIL was the early complication rate, the early operator experience [in PROTECT-AF], which were all procedural complications. The real reason for PREVAIL was to prove safety got better [with experience]," he said.

"As they went through PROTECT-AF, each third had lower [complication] rates, so they showed there was a learning curve ... In PREVAIL, 75% of the sites were old sites, and 25% were new sites, and still the safety was improved.

Will the FDA see it the same way?

The FDA has scheduled that meeting of the Circulatory Systems Devices Advisory Committee for Oct. 8, 2015. Watchman has had a long, hard road, and it doesn't look like it is getting any easier.

In 2009, an FDA advisory committee voted 7-5 to recommend approval. But in 2010, the FDA rejected Watchman, issuing a complete response letter saying another trial was needed because of a high rate of complications in the pivotal PROTECT-AF trial.

The confirmatory trial, PREVAIL, missed one of the three primary endpoints, failing to show noninferiority to warfarin at 18 months on prevention of the composite of ischemic stroke, systemic embolism, and cardiovascular or unexplained death.

However, a second advisory committee, on Dec. 11, 2013, was convinced by the totality of the evidence and voted 13-1 that the device is safe, effective, and has a positive risk-to-benefit for use in preventing stroke in patients with atrial fibrillation.

So, why the third advisory committee meeting?

In the original PREVAIL data, reviewed by the advisory committee in 2013, one efficacy endpoint had been missed.

In the updated analysis at 5 years, both the primary efficacy endpoints were missed, , of the Mount Sinai Hospital in New York City, reported.

The worse outcomes were blamed on the eight ischemic strokes with Watchman versus none with warfarin in the new data.

The question remains: will the FDA be convinced? It has repeatedly said, it is always a matter of "reasonable assurance of safety and efficacy."

From of the American Heart Association:

Comment