ReCor’s Renal Denervation System Gets Thumbs Up From FDA Advisory Panel

Advisors on the US Food and Drug Administration’s Circulatory System Devices Panel voted today in support of the premarket approval (PMA) application for the Paradise Ultrasound Renal Denervation System (ReCor Medical). With 12 members voting, all agreed that there was reasonable assurance of safety, while votes for reasonable assurance of efficacy were 8-to-3 in favor, with 1 abstention, and votes for benefits outweighing risk coming in at 10-to-2 in favor of the technology.

The panel spent the day considering the PMA; hearing from the sponsor and various experts and patient and industry representatives. Paradise is supported by data from three sham-controlled trials: RADIANCE-HTN SOLO, RADIANCE-HTN TRIO, and RADIANCE-II.

In 2020, Paradise met criteria for breakthrough device designation, but before making a decision on the device, the FDA sought the input of its advisory committee to be assured that it meets safety and efficacy standards based on the totality of data.

The catheter-based system delivers ultrasound energy to disrupt renal sympathetic nerve activity through ablation. The goal is to reduce systemic arterial blood pressure in patients with treatment resistant hypertension.

Despite hours of debate regarding questionable efficacy endpoints and durability issues that could not be answered because of the short follow-up duration of the supporting studies, there was a fair amount of consensus on the panel that safety was not an issue and that the device fills an unmet clinical need, voting member Robert W. Yeh, MD (Beth Israel Deaconess Medical Center, Boston, MA), told TCTMD after the session.

“When the studies aren't designed to answer certain questions, then we're sort of left looking for the signals in imperfect data,” said Yeh, who voted “yes” on all three questions. “We saw that many were concerned that too broad-based an indication will drive inappropriate use. Then there’s another concern, which I share, and that is whether the indications for use actually reflect what they studied. In some ways, it wasn't exactly the population that they had described in their research studies.”

ReCor is asking that the device be indicated for use in patients who may be inadequately responsive to, or who are intolerant of antihypertensive medications.

Supporting Data Questioned

RADIANCE II (n = 224) and RADIANCE-HTN SOLO (n = 146) enrolled US and European patients with mild-to-moderate hypertension who were on no or up to two antihypertensives. During the study, these patients were off antihypertension medications until the 2-month primary endpoint was assessed. RADIANCE-HTN TRIO (n = 136) enrolled patients with resistant hypertension despite taking three or more antihypertensives that were replaced with a single triple combination pill during the study period until the 2-month primary endpoint was assessed. Crossover to renal denervation was allowed at 6 months in the SOLO and TRIO studies and at 12 months in RADIANCE II.

The SOLO and RADIANCE II trials met their primary effectiveness endpoints for baseline adjusted difference of daytime ambulatory systolic BP at 2 months versus sham control, with an adjusted mean difference of 6.3 mm Hg in both trials.

Two statistical methods were used to examine the primary effectiveness endpoint in TRIO. In one, the mean difference reduction was 4.5 mm Hg favoring renal denervation (P = 0.0809). In the other more nuanced analysis of ANCOVA on the ranks, the median differences in reduction (median of all pairwise differences of BP reduction) was again 4.5 mm Hg favoring renal denervation (P = 0.0223). The reason for this second analysis was because of non-normal distribution of the data due to the number of outliers, and was part of the study’s prespecified statistical plan, ReCor told the panel.

But at least one advisory committee consultant and statistician, Benjamin Saville, PhD (Berry Consultants, Austin, TX), questioned the ANCOVA analysis, calling it “outdated.” Saville was one of three members who ultimately voted no on the efficacy question.

However, Julia Lewis, MD (Vanderbilt University School of Medicine, Nashville, TN), noted that when the data were examined by the proportions of patients in the intervention versus sham groups who achieved reductions of ≥ 5 mm Hg, ≥ 10 mm Hg, or ≥ 15 mm Hg in daytime systolic BP at 2 months, rather than in terms of mean differences between groups, the data were more strongly supportive of renal denervation in all three trials.

“One of the challenges of these data, and for continuous data in general that we often see, is the comparison of means [may] hide a more substantive benefit than we might see if we look at it a different way,” Yeh noted. “That’s why the raw data are important, and we heard that from the panel today asking for the waterfall plots and other distributions [rather than] just looking at the mean difference.”

When the studies aren't designed to answer certain questions, then we're sort of left looking for the signals in imperfect data. Robert Yeh

The panel also questioned what they saw as a similar level of BP control between the denervation and sham groups when meds were reintroduced to the sham group; the difference was < 1 mm Hg at 6 and 12 months. There also was virtually no difference in the mean number of antihypertensive agents that patients in the denervation and sham groups in TRIO were taking at 6 and 12 months.

By the afternoon, panel member Randall Starling, MD, MPH (Cleveland Clinic, OH), said he remained troubled by this.

Ajay J. Kirtane, MD (NewYork-Presbyterian/Columbia University Irving Medical Center, New York, NY), the primary investigator for RADIANCE II and SOLO, noted that because the sham group was aggressively treated with “the exact same goal for blood pressure under blinded circumstances,” this lack of difference is not necessarily a reflection of a lack of efficacy of denervation compared with medical therapy.

“We entirely expected that to be the case; in fact, we told our investigators that the goal should be to lower blood pressure in both groups,” he added.

In pooled mixed-model data that Kirtane showed from all three trials, there also was evidence of persistence of effect of renal denervation after accounting for a change in the number of medications, but Bram Zuckerman, MD, director of the Office of Cardiovascular Devices in the FDA’s Center for Devices and Radiological Health, interjected to say that those data had not been independently verified by FDA statisticians prior to being presented to the panel.

Another issue that the panel asked ReCor to address was the percentage of patients with an eGFR < 60. There were 14 such patients in the treatment group and 13 in the sham group in the pooled analysis, with no indication that renal denervation had less of an effect on daytime ambulatory systolic blood pressure in those with lower versus higher eGFR values (interaction P = 0.1073).

Safety, Durability, and Unmet Need

Advisory committee consultant John Hirshfeld Jr, MD (University of Pennsylvania, Philadelphia), said while there were no concerns raised in the trials with regard to safety, long-term follow-up is appropriate to ensure that restenosis complications do not occur.

Similarly, Eric Bates, MD (University of Michigan, Ann Arbor), said it might be “a little too rosy” to suggest that there are no safety concerns related to the procedure, noting that as it rolls out to a wider number of operators, there could be a small increase in the rate of access-site complications.

Throughout the day, members of the panel expressed interest in seeing longer term data on the impact of renal denervation on reducing medication burden. While Starling said he was satisfied that the medication burden was lessened with the procedure, Keith Allen, MD (St. Luke’s Hospital of Kansas City MO), said he was grappling with whether an invasive procedure is a reasonable alternative to medication.

“I personally don’t think denervation will be durable,” he said. “I can’t think of a single area where denervation has been used, whether it’s lumbar sympathectomy or even cardiac transplantation, where the organ is completely denervated that re-enervation doesn’t typically or always recur. I don’t think this is better than medicines at 6 months.”

Jay Giri, MD, MPH (University of Pennsylvania, Philadelphia), who spoke during the open public comment, contended that many patients struggle with hypertension despite the availability of good oral medications.

“It can be difficult to get on these multimedicine regimens and to keep them straight and take them as directed,” he noted, adding that cost and side effects of individual or combination therapies are other important considerations for these patients.

“For all these reasons, we’re at a place where we need a better solution, at least as an adjunct to what we’re doing to help patients deal with hypertension, which is causing all these terrible outcomes down the line,” he added.

In a patient preference survey of 258 people conducted by ReCor, 42% of respondents said they would prefer renal denervation to an additional pill to control their hypertension.

We’re at a place where we need a better solution, at least as an adjunct to what we’re doing to help patients deal with hypertension. Jay Giri

While some of the panelists felt respondents may not have had enough information on what it may be like to actually undergo the procedure to make that decision, others like Abdulla Damluji, MD, PhD, MPH (Inova Health Fairfax, VA), said if it makes a difference for patients who suspect, or who their physicians suspect, wouldn’t be fully compliant with multiple medications, it might be worth considering.

ReCor has plans for ongoing follow-up of all patients in all three of its trials, and the company has initiated a Continued Access Study/Continued Access Protocol (CAP), which is designed to include patients similar to those enrolled in RADIANCE II in one arm and patients similar to those enrolled in TRIO in another. They also intend to initiate a US global registry similar to one currently ongoing in Europe and the UK.

The advisory committee as a whole made clear that if the device is approved they hope postmarket studies will provide needed information to close many of the gaps addressing questions they still have, particularly with regard to the efficacy and duration of benefit, as well as the sponsor’s willingness to include underrepresented minority populations. Several members indicated that their “yes” votes were cast in hopes of seeing resolution to those issues.

Yeh pointed to another important point of clarity that he and others would be looking for in postapproval data, which is the type of patient who actually benefits most from undergoing renal denervation. Within the limited data, he said, there are indications of wide variation in the magnitude of benefit.

The panel reconvenes tomorrow to do this all over again for Medtronic's Symplicity Spyral Renal Denervation System.