GAITHERSBURG, MD – An expert panel narrowly voted in favor of recommending Food and Drug Administration approval of a novel, percutaneously implanted device that repairs the mitral valve for use in selected patients with significant symptomatic mitral regurgitation who are poor surgical candidates.
At a meeting on March 20, the FDA’s Circulatory Systems Devices panel, it voted 5-3 that the benefits of the MitraClip Delivery System outweighed the risks for patients who met the criteria specified in the indication under review: "percutaneous reduction of significant symptomatic mitral regurgitation [moderate to severe (3+)MR] in patients who have been determined by a cardiac surgeon to be too high risk for open mitral valve surgery and in whom existing co-morbidities would not preclude the expected benefit from correction of the mitral regurgitation."
The panel was more comfortable with the safety data, unanimously voting that there was "reasonable assurance" that the device was safe for patients who met the criteria in the indication.
Largely because of limitations of the studies submitted by manufacturer Abbott Vascular – which included combining nonpivotal data from two high-risk patient registries – the panel was less confident with the efficacy data, voting 5-4 that there was not reasonable assurance that the device was effective. (The extra vote was the panel chair, who votes in case of a tie, and he voted no.)
The panelists generally agreed with the FDA’s conclusions that there were numerous problems with the data – the basis of the FDA’s conclusion that the device should not be approved at this time – but they said that they had a sense there were patients who could benefit from treatment.
Dr. Craig Selzman, a cardiothoracic surgeon at the University of Utah, Salt Lake City, voted in favor of approval "with trepidation" because he believed the benefit-risk ratio was likely favorable, based on the safety profile and the apparent benefit. He stressed, however, that the device should be limited to "truly inoperable patients" and warned about "indication creep," when a device is rapidly used for treating patients outside the approved indication once it is marketed. He also recommended that the patient should be on optimal medical treatment and that an experienced mitral valve surgeon should identify which patients are candidates for treatment.
Also voting in favor of approval, Dr. George Vetrovec, professor of medicine and director of the adult cardiac catheterization laboratory at the Medical College of Virginia, Richmond, like several others, cited the improvements in New York Heart Association class in those treated with the device, "which got so much better it couldn’t just be happenstance. ... It seemed to me that had to be real and overall risks seemed to be reasonable for a very selected, limited population."
Those voting against approval said that treatment with the device appeared to be potentially beneficial but that it was difficult to identify which patients would benefit from treatment. Therefore, a randomized controlled trial was needed before approval.
The MitraClip system includes a delivery catheter and the MitraClip device, a mechanical clip designed to reduce regurgitation by clipping together the leaflets of the mitral valve.
In its application, Abbott combined two studies of nonpivotal data – a single-arm study of 78 high-risk patients that was an adjunct to EVEREST II (a randomized controlled study of patients with mitral regurgitation who were surgical candidates that did not find a benefit of MitraClip over surgery) and another registry of 273 high-risk patients (REALISM). The 351 patients had functional or degenerative MR, their mean age was 76 years, mean ejection fraction was 48%, 85% were NYHA functional class III/IV, and their mean predicted surgical mortality risk was 18%.
Implantation was successful in about 95% of patients, and the major effectiveness end point, change in left ventricular size from baseline to 1 year, significantly improved, with about a 10% reduction in LV volumes and LV dimensions. Other changes included improvements in NYHA class (including a two-class improvement in 36% of patients at 1 year) and significant improvement in a quality-of-life questionnaire. There were also reductions in heart failure hospitalization rates in the year after implantation, compared with the year before; the mean hospital stay was about 3 days. The primary safety end point, 30-day mortality, was 4.8%, which was lower than the mean predicted mortality risk of 18%, according to the company. Three of the 17 deaths within the first 30 days were device related.
The company also compared mortality in a Duke University cardiovascular database of high-surgical-risk patients who were treated medically (31% at 1 year), with mortality in a matched cohort of 211 MitraClip patients in the registry studies (24%), which the company concluded indicated that there was no increased mortality of the procedure compared with the natural history of the disease.