Advisory Panel Balks at Neovasc Reducer for Refractory Angina

More than 10 hours of testimony and debate failed to convince a US Food and Drug Administration (FDA) expert advisory panel that existing evidence reasonably supports a premarket approval (PMA) application for the Neovasc Reducer device.

The Reducer is intended for people suffering from refractory angina pectoris despite guideline-directed medical therapy, who are unsuitable for revascularization by coronary bypass grafting or percutaneous coronary intervention.

The FDA’s Circulatory System Devices Panel advisory committee generally agreed in a 14-to-4 vote that the evidence provides reasonable assurance the Reducer is safe, but took a dim view on assurances of its effectiveness in a vote of 1 to 17.

On the question of a benefit/risk ratio, the vote was 3 yes, 13 no, and 2 abstentions. The panel did not vote on the approval of the PMA itself but heard several impassioned pleas from patients calling for its approval in the United States.

“People talked about the urgent need and the patients need hope, but we shouldn’t give them false hope,” Richard Page, MD, University of Vermont Medical Center, South Burlington, said. “We need to provide them something we truly believe is going to be effective and that was not proven today.”

Erik Magnus Ohman, MD, Duke University School of Medicine, Durham, North Carolina, who voted no on all three counts, said, “I voted no for efficacy because I couldn’t link ischemia, which is objective, with a device that we put in permanently when nearly half the patients had very little treatment benefit.”

The primary data set in support of the PMA was from the phase 2 COSIRA study, in which 18 of 52 patients treated with the Reducer and 8 of 52 patients treated with a sham procedure had improved by at least two Canadian Cardiovascular Society (CCS) classes, the primary efficacy outcome, at 6 months (34.6% vs 15.4%, P = .024). 

In 28.8% of the Reducer group and 57.7% of the control group, no change in CCS was seen.

“This is an angina study and for 50 years the standard for angina for success or magnitude of success in minimizing or preventing angina has been a quantitative exercise test, but that’s not what we’re talking about here,” Jeffrey Borer, MD, SUNY Downstate Health Sciences University in Brooklyn, New York, said. “We’re talking about a subjective endpoint and I think that’s a real problem and just magnifies all the other problems that have been discussed.”

Several panelists questioned whether the trial truly enrolled patients with CCS class 3 or 4 angina with limited options, given that 27% of Reducer patients and 25% of controls were on none or one antianginal medication. Information was also not provided about compliance or whether patients were on therapeutic or maximally tolerated doses.

Others pointed out the potential for a placebo effect and that patients were largely satisfied with treatment despite the marked discrepancy in results. Also, the trial lacked a formal blinding assessment for investigators.

Wayne Batchelor, MD, Inova Heart & Vascular Institute, Fairfax, Virginia, took issue with the statistical robustness of the results with such small patient numbers and wide confidence intervals.

“If one patient was misclassified in each arm, the statistical results would have been completely removed and, in fact, if one patient was misclassified in the Reducer arm, with the Yates correction, also statistical significance would be lost,” he said. “I think this is important when dealing with a trial that is naturally prone to regression to the mean, placebo effects, and also blinding concerns.”

Other sticking points were the lack of a prespecified primary safety endpoint; difficulty in interpreting secondary endpoints due to high rates of missing data, which was roughly twice as high among controls; and that the trial was conducted in Europe and Canada and halted prematurely.

John Somberg, MD, Rush Medical College, Lake Bluff, Illinois, said there was nothing to suggest COSIRA wasn’t blinded. “So, I’m less concerned about blinding but I’m most concerned about the missing data. It wasn’t in the Canadian questionnaire [results], so that’s good, but everything else seems to be Swiss cheese.”

Joaquin Cigarroa, MD, Oregon Health & Science University, Portland, said he voted yes for safety, but with regard to efficacy, “the primarily male, Caucasian European, and non-US approach made it very difficult to extrapolate to our US population and then the challenges of a small sample size, further cemented my response for voting no.”

Data were limited beyond 6 months, which made it difficult to assess a lot of the long-term safety issues such as the impact the device might have on diastolic heart failure, which many of these patients get, as well as the need for antithrombotic therapy, Cigarroa and others noted.

Speaking on behalf of the sponsor, Gregg Stone, MD, Icahn School of Medicine at Mount Sinai in New York City, said the COSIRA data was still blinded when the trial was stopped early due to slow enrollment and good retention, and that postoperative management was also blinded.

He noted that not all patients can tolerate multiple antianginal medications and that these patients truly had few options, having already undergone multiple procedures. The alternatives of surgical transmyocardial revascularization or heart transplant carry substantial safety concerns, whereas serious adverse events favored the device arm in COSIRA.

The Reducer device received CE Mark in 2011 and was granted breakthrough device designation by the FDA in October 2018.

The FDA approved plans for the COSIRA-II trial, which was to include 380 patients and a primary endpoint of objective assessment of ischemia with exercise tolerance testing. The agency, however, suggested additional data may be required to support PMA approval, and the trial was never started.

To support the PMA, the sponsor provided results from 44 coronary angiograms in Reducer patients and additional data from the first-in-man study and the ongoing REDUCER-1 observational study. FDA officials, however, concluded that the lack of controls in REDUCER-1 made it difficult to draw efficacy conclusions.

The Reducer is a balloon-expandable, hourglass-shaped, stainless steel mesh device that is designed to obstruct flow and increase coronary-sinus pressure, thereby increasing the perfusion of oxygenated blood to ischemic areas of the heart muscle.

Animal studies were also presented but again the data were limited and the studies were unable to confirm coronary-sinus narrowing or an elevated coronary-sinus pressure, FDA officials said.

“While we don’t necessarily need to understand fully the mechanism of a device’s action, aren’t we obligated to consider scientific plausibility as we assess the data overall and come to a conclusion as to a reasonable assurance of effectiveness?” asked Richard Page, MD, University of Vermont Medical Center, South Burlington.

Bram Zuckerman, MD, director of the FDA’s Cardiovascular Devices Office of Product Evaluation and Quality, said the totality of the evidence should be considered to determine the risk/benefit ratio but that a clear understanding of the mechanism of action is not needed for approval.

The panel was also asked to comment on a postapproval, double-blind, sham-controlled study proposed by the company to “reduce any remaining uncertainty the FDA may have.” In a show of hands, all but one member felt a US-based preapproval study was needed.

“I find it unusual we are asked to approve a product and then do the study we would all like to see done,” Keith Allen, MD, NorthBay Healthcare, Fairfield, California, observed dryly.

Circulatory System Devices Panel of the Medical Devices Advisory Committee. October 27, 2020.

Follow Patrice Wendling on Twitter: @pwendl. For more from theheart.org | Medscape Cardiology, join us on Twitter and Facebook.