GAITHERSBURG, MD.– A Food and Drug Administration advisory panel narrowly voted to support approval of the Watchman left atrial appendage closure device for patients with nonvalvular atrial fibrillation, a sharp contrast to the panel’s nearly unanimous vote to recommend approval just 10 months earlier.
At the meeting on Oct. 8, the FDA’s Circulatory System Devices panel voted 6-5 with 1 abstention that the benefits of the device outweighed the risks for the proposed indication, to prevent thromboembolism from the left atrial appendage (LAA), with the following statement: The device “may be considered for patients with nonvalvular atrial fibrillation (AF) who, based on CHADS2 or CHA2DS2-VASc scores, would be recommended for warfarin therapy to reduce the risk of stroke and systemic embolism.”
Voting separately on safety and effectiveness, the panel voted 12-0 that there was “reasonable assurance” that the device was safe for use in this population, but voted 6-6 that there was not reasonable assurance that it was effective; the panel chair then voted no to break the tie. (Device panel chairs vote only when the vote is tied). The shift in the panel’s support for approving the device was because of an increase in new ischemic strokes reported in patients treated with the Watchman device in the PREVAIL study, between January and June 2014, after the study was reviewed at the previous panel meeting in December 2013. It was to review these data that the FDA convened the latest meeting.
Panelists voting on both sides of the risk-benefit question said the indication was too broad and should be revised to describe the device as a second-line alternative to warfarin, making clear it is not appropriate for all warfarin-eligible patients. Several of those voting no said they would support approval of a second-line indication.
“As excited as we were last time about the benefits of the device – to the point that the indication was expanded to include all Coumadin-eligible patients now, with these new concerning data from PREVAIL about not preventing ischemic strokes, the device seems to be more appropriate as a second-line therapy for patients who cannot take Coumadin,” one of the panelists, Dr. Ralph Brindis, said in an interview after the meeting. Dr. Brindis, of the University of California, San Francisco, abstained on the risk-benefit question.
The Watchman is a percutaneously delivered permanent cardiac implant placed in the LAA to prevent the embolization of thrombi formed in the LAA, reducing the risk of ischemic stroke and systemic embolism. It was approved in Europe in 2005 and is now approved in more than 70 countries, according to Boston Scientific, the manufacturer.
Dr. Ralph Brindis
At the meeting in December, the panel voted 13-1 to recommend approval of the device, based on data from the PREVAIL and PROTECT-AF studies, which compared the device to chronic warfarin, and information from the Continued Access to PREVAIL (CAP2) registry. PREVAIL compared implantation of the device – with 45 days of warfarin plus 81 mg of aspirin for 45 days, followed by 325 mg of aspirin and 75 mg of clopidogrel through 6 months, followed by 325 mg of aspirin a day indefinitely – to chronic warfarin. PREVAIL, which was designed to address limitations of PROTECT-AF, enrolled higher-risk patients.
At that time, patients in PREVAIL had been followed up for an average of 12 months, and only 28% had reached the 18-month follow-up period. The safety endpoint (all-cause death, ischemic stroke, systemic embolism, or device- or procedure-related events requiring surgery or endovascular intervention within 7 days of the procedure) and one of the two primary effectiveness, noninferiority endpoints (18-month rate of ischemic stroke and systemic embolism at least 7 days after randomization) had been met. But the other effectiveness endpoint (a composite of all stroke, systemic embolism, and CV or unexplained death) was not met.
Between January and June 2014, when all the patients had been followed for at least 18 months, there were 8 additional cases of ischemic strokes in the Watchman group, all at least 1 year after the device was implanted, for a total of 13, and no new ischemic strokes in the warfarin-treated group, yielding event rates per 100 patient-years of 2.30 in the Watchman group and 0.34 in the control group. These additional cases “raise questions about the long-term effectiveness of the Watchman device for protecting patients from ischemic stroke,” according to the FDA.
With the additional data, as of June, the safety endpoint was still met, but the data on the device did not meet the noninferiority criteria for either of the two primary effectiveness endpoints, and the FDA convened the meeting for the panel to address whether the totality of the data would change the panel’s conclusion in December that the device had a favorable risk-benefit profile.