The Hole Story: Interatrial Shunt Companies Are—Mostly—Pushing Ahead

Even since April when the RELIEVE-HF trial came out showing no benefit to an investigational interatrial shunt for heart failure (HF) patients, experts have been scratching their heads as to how best to proceed—and how cautiously—in this truly novel space.

As reported by TCTMD, RELIEVE-HF showed that implanting the Ventura shunt (V-Wave) failed to improve clinical outcomes or reduce symptoms in a wide range of HF patients and, alarmingly, raised the possibility that the shunt, designed to decrease pressure in the left atrium (LA), might actually be harmful in patients with preserved ejection fraction (HFpEF).

REDUCE-LAP-HF II, an earlier randomized trial testing a different interatrial shunt (Corvia Medical) in patients with HFpEF, also missed the mark. Here, placement of the Corvia shunt did not reduce HF events or improve health status in the overall trial. A group of HF patients did benefit, however: those without evidence of pulmonary vascular disease (PVD) identified on invasive exercise testing and without a cardiac rhythm management device. Post hoc analyses of the trial also showed favorable changes in cardiac structure and function in the responder group when compared with nonresponders. 

These mixed results, and particularly the signal of harm in RELIEVE-HF, have given pause to the several companies enrolling HFpEF patients into pivotal randomized trials, including Corvia Medical, Alleviant, and Edwards Lifesciences, among others.

James Udelson, MD (Tufts Medical Center, Boston, MA), who is involved in testing a no-implant shunt (Alleviant) in the ALLAY-HF trial, said RELIEVE-HF threw many for a loop. “The HFpEF subgroup looked pretty bad,” he told TCTMD. “It was pretty stunning sitting there.”

Within 24 hours or so, he said, “the dust settled.” Udelson, like others, came to the conclusion that the HFpEF patients studied in RELIEVE-HF were a “pretty sick bunch” who simply were unlikely to benefit from the shunt as opposed to being harmed by the procedure.

In the HFpEF patients in RELIEVE-HF, placement of a shunt was associated with a significant 68% higher risk of all cardiovascular events and more than a threefold higher risk of all-cause mortality compared with those treated with a sham procedure. In those with HFrEF, shunt implantation was associated with a 45% lower risk of all cardiovascular events compared with the sham arm.

Given those findings, Gregg Stone, MD (Icahn School of Medicine at Mount Sinai, New York, NY), who presented the study at the American College of Cardiology Scientific Sessions in April, said at the time that if he were designing future trials in this space, he’d only randomize HFrEF patients.

William Abraham, MD (Ohio State University Wexner Medical Center, Columbus, OH), the chief medical officer at V-Wave, told TCTMD they are still evaluating the RELIEVE-HF findings, and that they couldn’t comment about their results or future plans, including the HF patient population they envision best suited for interatrial shunting.

The RELIEVE-HF researchers are planning to publish their full results as well as a crucial “responder analysis” later this year, Stone told TCTMD.  

Responding, or Not

Nearly a dozen companies have been developing interatrial shunt devices or working on procedures to unload the LA. Since HF is characterized by increased LA pressures and pulmonary venous congestion, the working theory is that creating a small hole between the LA and right atrium (RA) might reduce LA pressures at rest and during exercise.

Corvia Medical, V-Wave, Occlutech, and Edwards Lifesciences, as well as others in early feasibility testing, have developed permanent shunts. Others, such as Alleviant, NoYA, and InterShunt, are going with the no-implant approach. Edwards Lifesciences is working on a leave-behind shunt in multiple sizes that preserves the septum and instead creates a channel between the LA and coronary sinus. Shunts vary from 4 mm to 12 mm, with some companies working across a range of sizes while others are going with a single shunt size. For example, the Corvia shunt is 8 mm, but the V-Wave and Alleviant shunts are smaller (5.1 mm and 6.0 mm, respectively).   

Based on the responder group in its randomized trial, Corvia Medical launched RESPONDER-HF, a 260-patient confirmatory, randomized trial of patients with HFpEF and those with mildly reduced EF (HFmrEF) that will exclude participants with hemodynamic evidence of PVD and cardiac rhythm devices. 

Jan Komtebedde, DVM, the company’s senior vice president and chief medical officer, said that considering the types of patients enrolled, there were few findings in RELIEVE-HF that took them by surprise, including the suggestion of harm in those with HFpEF. In general, though, they can’t make too much of RELIEVE-HF just yet because it’s not been published, and subgroup and responder analyses have not been presented. Based on what they’ve seen so far, Komtebedde said, Corvia has no plans to alter their inclusion/exclusion criteria. Enrollment began in April 2023.

“We have not seen any evidence that the protocol should be changed because the patients that we excluded following our pivotal trial are the patients that were included in the RELIEVE-HF trial,” said Komtebedde. “That’s why we expected the RELIEVE-HF [patients with HFpEF] to not have good outcomes: because there's overlap with our nonresponders.”

As for the marked benefits in HFrEF patients, prior open-label studies using the Ventura shunt or the Occlutech device have not shown such benefits, said Komtebedde. Of note, in Corvia Medical’s REDUCE-LAP-HF II, approximately 10% had HFmrEF, and this group clearly benefited less than those with HFpEF, he added.  

We’re trying to pick out the strike zone, to find the patients most likely to respond. James Udelson

The RELIEVE-HF investigators have published their results from the open-label, roll-in cohort, which gives a sense of the types of patients randomized in the trial. As Udelson pointed out, the vast majority had NYHA class III symptoms, markedly elevated NT-proBNP levels, and evidence of more right ventricular dysfunction than in previous shunt studies. Invasive exercise hemodynamic testing was not performed in the trial.

“I think there [is] an explanation, at least in some ways, for the negative results in the HFpEF group,” Udelson told TCTMD. “The HFpEF group looks very much like the nonresponders in REDUCE-LAP II. Sicker patients, higher pulmonary vascular resistance. I think we were all thinking that [RELIEVE-HF] almost validates what we’re doing with these other trials. We’re trying to pick out the strike zone, to find the patients most likely to respond.”

The ALLAY-HF trial, said Udelson, is modelled after the REDUCE-LAP II responder group. Researchers plan to enroll 400 to 700 patients with HFpEF (left ventricular EF ≥ 40%) who have elevated pulmonary capillary wedge pressure during exercise ≥ 25 mm Hg and exercise pulmonary vascular resistance < 1.8 WU.

“Essentially, ALLAY-HF and RESPONDER-HF are studying the same type of patients,” Udelson told TCTMD.

Christopher Meduri, MD (Karolinska Institute, Stockholm, Sweden), who is involved in ALLAY-HF, said that while the first two randomized, controlled trials were neutral overall, they highlighted the importance of identifying the right patient who stands to benefit from an atrial shunt.

“Corvia’s responder group in REDUCE-LAP II taught us the importance of exercise phenotyping in HFpEF,” he told TCTMD—an idea the investigators have been insisting on for years.

Last year, Occlutech paused enrollment in the FROST-HF, which was their sham-controlled, randomized trial, in light of the many unknowns. The company initially had plans to include patients with HFpEF and did not have exclusion criteria based on pulmonary vascular resistance during exercise. According to an Occlutech spokesperson, they are waiting to learn more about RELIEVE-HF before starting up again. 

“An in-depth understanding of [the RELIEVE-HF] data, alongside comparisons with data from other stakeholders, will inform us on how to outline the most impactful clinical development plan, including consideration of a targeted randomized controlled trial,” the Occlutech spokesperson told TCTMD.

Edwards Lifesciences would not comment on whether they are making any changes to their randomized trials in light of RELIEVE-HF or REDUCE-LAP II. Earlier this year, researchers presented favourable 1-year outcomes from their early feasibility study and have since launched ALT-FLOW II, a 100-patient, sham-controlled trial that includes HFpEF and HFmrEF patients. Key exclusion criteria include patients with pulmonary vascular resistance ≥ 5.0 WU at rest and greater than mild RV dysfunction.

Other companies, including NoYA, InterShunt, NXT, Adona Medical, and D-Shant, are still in early feasibility testing or first-in-human studies.

Staying the Course

To TCTMD, Komtebedde said the development of interatrial shunting for HF is complex, with multiple variables—patient selection, shunt size, type of shunt, among others—that factor into a successful therapy for appropriate patients.

“We have excellent scientific advisors that are involved in protocol development and patient selection criteria,” he said. “Based on our open label study, we thought we had the right criteria going into the pivotal trial. Lo and behold, we were wrong with respect to pacemakers and high pulmonary vascular resistance. There was no evidence these would be problematic or associated with nonresponders. And there’s still things to be learned.”

There’s still things to be learned. Jan Komtebedde

For example, it’s unclear why patients with pacemakers fared worse or if they’ve selected the optimal cutoff point with peak exercise pulmonary vascular resistance (< 1.75 WU). At the moment, Komtebedde said the company has no plans to study HFrEF patients in a separate randomized trial, instead choosing to focus on the responder phenotype they identified in REDUCE-LAP-HF II and are studying in RESPONDER-HF.

“That's really the top priority for our company,” he said. “We hope that other companies will try to understand if patients with lower EF can benefit from a shunt.”

Indeed, all of the ongoing studies will help to hone patient selection.

Based on what’s known so far, Udelson said he and his coinvestigators are not too concerned about patients with HFpEF being harmed by shunt placement in the ALLAY-HF trial.

“This whole thing comes down to, whatever kind of shunt you use, you’re unloading the left atrium, which is good, and you’re loading the right side and pulmonary vasculature,” he said. “It’s all about finding who benefits from that unloading and isn’t harmed by the right-sided loading. Essentially, who can handle the shunt.”