Boston Scientific Corp.believes the results from the PREVAIL trial are enough to finally secure FDA approval for the Watchman left-atrial appendage closure device to prevent stroke in patients with atrial fibrillation, but skepticism about the device remains among some cardiologists.

The company released preliminary results from PREVAIL, which compared Watchman to long-term warfarin therapy in patients with nonvalvular atrial fibrillation, March 9 in conjunction with the 2013 American College of Cardiology Scientific Sessions in San Francisco.

The data was supposed to be reported by David Holmes of the Mayo Clinic, as part of the late-breaking clinical trial presentations that day, but Boston Scientific announced the trial outcomes in advance of ACC’s embargo, causing the meeting organizers to pull the trial from the program. Thus, the data did not get the same type of formal public vetting it would have as an official ACC study.

According to the data released by the company and slides posted on www.tctmd.com, the trial met the pre-specified criteria of all-cause death, ischemic stroke, systemic embolism, or device or procedure-related events necessitating open surgery or major endovascular intervention.

However, the trial did not meet the pre-specified second co-primary endpoint – a composite of all stroke, cardiovascular death, and systemic embolism at 18 months. Both the Watchman and the control groups had similar 18-month adverse event rates of around 0.064, but the confidence interval was too wide to show a statistically significant result.

Investigators and Boston Scientific staff worked around-the clock to complete the data analysis for all three endpoints in time for the meeting, before they inadvertently broke the embargo, according to the company.

Not enough patients (30 in the control group and 58 in the Watchman group) have reached the 18-month follow-up to show a statistically significant difference in this endpoint, Joe Fitzgerald, Boston Scientific’s senior VP and president for cardiology, rhythm, and vascular, told “The Gray Sheet.” But he said the data is trending in favor of Watchman and that the trial will likely meet this endpoint by the end of the planned five-year follow-up.

PREVAIL also met its pre-specified non-inferior third co-primary endpoint, a composite of late ischemic stroke and systemic embolism between eight days post randomization and 18 months. The rates were 0.0253 and 0.0201 in the Watchman and control groups, respectively.

All of the patients in PREVAIL had a Calculated CHADS2 score, which estimates the stroke risk in patients with atrial fibrillation, of 2 or greater -- indicating their annual risk of stroke was at least 4 percent – or a CHADS2 score of one (annual risk, 2.8 percent) plus other important risk factors.

All of the data was labeled as preliminary. Boston Scientific originally planned to report outcomes only from one of the three endpoints. But, according to the company, investigators and Boston Scientific staff worked around-the clock to complete the data analysis for all three endpoints in time for the meeting, before they inadvertently broke the embargo.

PREVAIL PI: New safety data key to FDA submission

The company will include the final PREVAIL results in a new FDA submission for Watchman. The firm will also submit six-year data from PROTECT AF pilot study, results from a 150-patient ASA Plavix (ASAP) contraindication registry in Europe, and continued access protocol (CAP) registry results, Fitzgerald said. (See Also see "Research In Brief" - Medtech Insight, 14 May, 2012..) Boston Scientific plans to update the status of the FDA submission during their next quarterly earnings call in April.

Boston Scientific has submitted Watchman for FDA approval before. In 2007, an FDA advisory panel voted 7-5 in favor of approving Watchman based on results of the 800-patient PROTECT AF trial, which showed that that left-atrial appendage closure with Watchman reduced cardiovascular death and systemic embolism by 38% relative to long-term anticoagulation medication.(See Also see "Panel Endorses Atritech Watchman Stroke-Prevention Device In Close Vote" - Medtech Insight, 27 April, 2009..) Despite the positive efficacy results, FDA asked Boston Scientific to collect more clinical data to overcome questions about the device’s safety, especially the relatively high rate of pericardial effusions and procedure related strokes observed in the early experience with the device. (See Also see "Left Atrial Appendage Closure Studies: FDA Offers Tips" - Medtech Insight, 20 February, 2012..)

“PROTECT AF showed that, from an efficacy perspective, the Watchman did as well as [warfarin] and, in fact, if you look at patients who received the Watchman successfully, they did better. So from the efficacy side there was never a problem,” PREVAIL co-primary investigator Vivek Reddy, an electrophysiologist at Mount Sinai Hospital in Manhattan, told “The Gray Sheet”.

“The real issue was on the safety side, there were a number of events and when we presented the data to FDA, we showed that over the course of the [PROTECT AF] study, the safety events decreased over time,” Reddy said. “So the appropriate question FDA asked was ‘what will happen when you introduce it to the market? Will we have a rash of big problems?’ So the major reason for PREVAIL was to really understand if all of the learnings from PROTECT [could] be translated to new operators.”

Watchman left-atrial appendage closure device

Photo: Boston Scientific

Boston Scientific and the investigators designed PREVAIL to confirm the PROTECT AF results and show that relatively inexperienced users could safely implant Watchman if adequately trained. PREVAIL randomized 407 randomized patients at 41 U.S. centers, including 18 centers without prior experience with Watchman, to either treatment with Watchman or warfarin therapy only. The trial protocol required that at least one quarter of the patients be enrolled at these new centers.

The overall procedural success rate in PREVAIL was 95.0 percent, which compares very favorably to the 90.9 percent rate in PROTECT AF. Both new users and experienced users showed statistically significant success rates (93.2 percent versus 96.3 percent). The rates of vascular complications within one week of the procedure were also similar, 2.9 percent versus 5.4 percent, for new and experienced operators, respectively. Pericardial effusion requiring intervention was only experienced among patients treated by new users, and there was no device embolization, peri-procedural strokes, or cardiac perforation, according to the firm.

“We’re very pleased [with PREVAIL] because the largest question over the past three years has been FDA’s concern on procedural safety with new operators,” Fitzgerald said. He said the company has learned from early experience how to train new users more effectively.

The overall seven-day serious procedure/device related complication rate in PREVAIL was 4.4 percent compared to 8.7 percent in PROTECT AF. The pericardial effusion rate in PREVAIL was 1.9 percent vs. 4.0 percent in PROTECT AF.

New users will have to complete the same learning process that the PREVAIL investigators went through, including didactic sessions, case demonstrations, and proctored procedures, Reddy explained.

In particular he stressed the importance of careful sheath management to avoid introducing air into the patients’ circulation which can cause strokes. “That’s something that all the new operators are very careful about.” The PROTECT investigators have also learned how to avoid pericardial perfusions or cardiac tamponade (build-up of fluid between the myocardium and pericardium) with the device and Boston Scientific incorporated those lessons into the firm’s Watchman training program.

Reddy acknowledged that some observers may be disappointed by the PREVAIL results, but only because they were “expecting a PROTECT-like study,” he said. “But PREVAIL was never designed to do that. PREVAIL was designed to assess if we could be safer and it actually showed that. So if you understand what PREVAIL was designed to be, you’ll actually be very happy with the outcome.”

“If you look at the three major studies that were done in the U.S., PROTECT, PREVAIL, and the CAP registry, it’s over a 1,000 patients undergoing this first-in-class procedure. Not a single procedure-related mortality,” Reddy said. “That’s pretty good. You don’t see that often in a first-in-class procedure.”

[Sidebar#01130318023]

Skepticism remains

Despite Boston Scientific’s optimism about the left-atrial appendage closure market opportunity, some cardiologists attending the ACC meeting remain skeptical about the potential benefits of the device.

In his blog on website theheart.org, John Mandrola, an electrophysiologist at the Louisville Cardiology Group in Kentucky, gave eight reasons to remain doubtful about Watchman.

He points out, for example, that PREVAIL did not include patients who are ineligible for anticoagulation drugs because of their high risk of bleeding complications. But this is the atrial fibrillation patient population most in need of an alternative to those drugs, Mandrola noted. Also, the control patients in the trials of Watchman took warfarin, but newer anticoagulants, such as dabigatran, rivaroxaban, and apixaban have all shown advantages to warfarin in higher-risk patients with atrial fibrillation, so it remains to be seen how Watchman would compare to therapy with those drugs, he said.

“The device is a permanent indwelling device lodged in the human heart. That's significant. Once endothelialized, it isn't coming out easily. As tools go, placing a device in the left atrial appendage is a big hammer,” Mandrola said.

And Gordon Tomaselli, the director of cardiology at Johns Hopkins School of Medicine in Baltimore and former president of the American Heart Association, said he would not generally recommend Watchman for patients with new onset atrial fibrillation. “I would refer patients under certain circumstances, particularly those who cannot take oral anticoagulants due to problems with certain bleeding complications and those that have a high CHAD2 score,” said Tomaselli, as quoted by the publication Cardiovascular Business. “This patient population equates to approximately 5-10 percent of those I see in practice.

“If I were able to look in a crystal ball, I would speculate that [FDA] will approve Watchman in a restricted set of circumstances.”

New indications, new devices

But, according to Boston Scientific’s Fitzgerald, “Watchman has been trialed in the toughest patient population -- those are the warfarin/anticoagulant tolerant patient population. So the indicated population that we will ask for approval on is the largest: a-fib and warfarin-eligible.”

“[But] the other question that you get is ‘what about the intolerant or the contraindicated?’” he said.

Watchman earned CE mark labeling for atrial fibrillation patients who also have a contraindication to warfarin and newer oral anticoagulants based on results of the ASAP registry. (See Also see "New Products In Brief" - Medtech Insight, 3 September, 2012..) ASAP demonstrated a 77 percent reduction for ischemic stroke in patients contraindicated to warfarin. Fitzgerald indicated the company may eventually seek a similar indication for Watchman in the U.S.

Boston Scientific is also working on new versions of Watchman which it hope will perform more consistently than the current version by conforming better to the shape of the patient’s left atrial appendage, Fitzgerald said. Also, the new version will be retractable so it can be recaptured and redeployed if the first attempt is yields a suboptimal placement, he said.

Reddy said the new device should further lower the risks of tamponade and device related thrombus. “There’s a small but real rate of thrombus with the device. Fortunately, it doesn’t translate to clinical events, but we still don’t want it to be there.”

Boston Scientific Says PREVAIL Results Affirm Safety of Watchman LAA Anti-Stroke Device

More from Clinical Trials

More from R&D