FDA Advisors Cool on Watchman Approval Amid Ischemic-Stroke Data

Shelley Wood

October 08, 2014

GAITHERSBURG, MD ( updated ) — The FDA will once again have its work cut out for itself after a very mixed vote from its Circulatory System Device Panel Wednesday on the Watchman (Boston Scientific) left atrial appendage occluder.

The panel voted unanimously that the device is safe for the proposed indication but were split 6:6 whether there was reasonable assurance that the device is effective for the patients covered by the proposed indication—namely, those who, based on CHADS2 or CHA2DS2-VASc scores, would be recommended for warfarin therapy to reduce the risk of stroke and systemic embolism. Panel chair Dr Richard Page (University of Wisconsin School of Medicine & Public Health, Madison) voted to break the tie with a "no" vote, swinging the decision against voting in favor of device efficacy.

The third and final vote, on whether the benefits of the Watchman outweighed the risk, was again a split vote, six in favor to five against, with one abstention.

Several panel members acknowledged their "no" vote on efficacy and on the safety/benefit trade-off might change if the Watchman were approved as an option for second-line therapy, and not for patients who would be recommended for warfarin therapy.

Third Time, No Charm

It was a third trip to Gaithersburg, MD for the Watchman left atrial appendage closure device, and it's clear that FDA advisors are perturbed by the increase in ischemic strokes beyond seven days seen among Watchman-treated patients.

The sponsor, Boston Scientific, is seeking an approval for the prevention of thromboembolism in patients with nonvalvular atrial fibrillation.

Back in 2009, the FDA's Circulatory System Devices Panel, as reported by heartwire , voted 7 to 5 in favor of approval, prompting the FDA to ask for an additional study. With partial PREVAIL results in hand, a December 2013 advisory voted 13 to 1 in favor of approval. Wednesday's third meeting was called in order to give the FDA advisory panel more complete follow-up from PREVAIL. That data include the eight new ischemic strokes seen among the Watchman-treated patients in PREVAIL group (and zero in the comparator group). And while the original PREVAIL results indicated that one of two primary end points were met, longer follow-up, which included the ischemic-stroke data, show that neither of the primary efficacy end points met the prespecified noninferiority end point.

Indication Discomfort

The key issue for panel members was the patient group covered in the proposed indication, which reflected the patients studied, with many panelists saying they were uncomfortable with the labeling but that they believed the device might have a role as second-line therapy or for patients who couldn't take warfarin.

Dr Kristen Patton (University of Washington, Seattle) who voted yes on safety, yes on efficacy, but no on the question of benefit/risk trade-off, said she couldn't vote yes on the final question because the indication, as written, is too broad. Pressed to explain, she said she believes that patients should be able to "avail themselves of this therapy," but "it has to be made clear [to physicians that] it's not responsible to offer this device as first-line therapy, equivalent to warfarin."

Others used the exact same concerns to explain why they voted one way or the other on the different questions. That was the strategy taken by Dr David Kandzari (Piedmont Heart Institute, Atlanta, GA), for example, voted yes on safety, no on efficacy, but yes on the benefit/risk question and had similar arguments to others who did the opposite for questions 2 and 3.

Kandzari said he was concerned with the "trajectory" of higher ischemic strokes for a technology whose primary purpose "intuitively" is to reduce this very problem.

"But on the risk/benefit balance, I'm swayed by the composite  . . . of reduced hemorrhagic stroke compared with warfarin and combined antithrombotic therapy, and like others, I struggle with the indication [as currently written]."

Dr Ralph Brindis (University of California, San Francisco), who cast the "abstain" vote on question 3, thereby tilting the results marginally in favor of the "yes" side, said he was "wrestling between voting yes or abstaining" on this benefit/risk question but that his reasons echoed Kandzari's.

"I abstained because I'm looking forward to a label revision and hopefully can change that to a yes."

Panel chairs at these meetings are only called upon to vote if a tiebreak is needed. Page, in his closing remarks, said he "always hopes I won't need to vote" and thanked Brindis, prompting appreciative chuckles from the panel, for abstaining on question 3, saving him from having to cast another tie-breaker.

But had he been called upon to decide the balance on the risk/benefit question, Page said, had he needed to, he would have voted "yes."

"I think this device is important technology," Page said, and "we've got to have it available, but it's got to be available for the right people. . . . I think this device has a home, and we just need the FDA and sponsor to work together on a description and indication that's appropriate and gives this important technology a place in our armamentarium to care for patients with atrial fibrillation at risk of . . . embolic stroke."


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.