Clinical trial lessons from the V-Wave and Cardiac Dimensions heart failure pivotal studies

After leading V-Wave‘s pivotal trial ahead of the device developer’s acquisition by Johnson & Johnson MedTech, Dr. William Abraham is working on another heart failure trial as the new chief medical officer at Cardiac Dimensions.

V-Wave’s U.S. pivotal trial, RELIEVE-HF (short for Reducing Lung CongestIon Symptoms Using the V-Wave Shunt in Advanced Heart Failure), demonstrated the Ventura interatrial shunt’s safety but fell short of its efficacy endpoint across the entire patient population.

Related: Why V-Wave’s heart failure tech is worth $1.7B to J&J

The study included heart failure with reduced ejection fraction (HFrEF) patients as well as patients with heart failure with preserved ejection fraction (HFpEF). The results showed benefits for that first group but not the second, and the trial’s design meant it wasn’t the end of the road for V-Wave’s technology.

“It was a very well designed and really well executed study,” he said in a Medical Design & Outsourcing interview.

This is the second article from that interview, following the first, which focused on Abraham’s perspective on the future of heart failure devices, the tech that will enable them, and opportunities to meet major needs for cardiac patients.

In this second part, we discuss lessons from V-Wave’s pivotal trial and what’s ahead for Cardiac Dimensions as it enrolls functional mitral regurgitation (FMR) patients in the Empower study of its implantable, nitinol-enabled Carillon Mitral Contour System, which already has a CE mark in Europe.

Related: How Cardiac Dimensions developed the Carillon Mitral Contour System — and makes the nitinol implant in-house

The following has been lightly edited for clarity and space.

MDO: What did you learn from the V-Wave pivotal trial that could help other device developers?

Abraham: “The key to every successful trial is first, you need to have a therapy that works. We think interatrial shunting works, it benefits patients — at least some patients — with heart failure. And then you have to execute a trial well. That includes selecting the right patients. One of the challenges is we don’t always know exactly who those right patients are going in. One of the things we wisely chose to do in the RELIEVE-HF trial was to stratify randomization on the basis of left ventricular ejection fraction (LVEF). There was debate within the heart failure community: Is this a therapy likely to help HFpEF patients, or HFrEF patients, or all heart failure patients? And at the time the trial was designed no one really knew. Some companies were studying HFpEF, some HFrEF, some both. V-Wave chose to study both groups of heart failure. Structurally and functionally these are two very different forms of heart failure. HFpEF is associated with small hearts that are thick and stiff. HFrEF is associated with enlarged hearts that are sort of baggy and more compliant and weakened. Will they respond the same or differently? What we did in the design of that trial was we stratified randomization in the trial by LVEF. That creates two strata of patients within the study — the HFrEF group and the HFpEF group — and allows one to not only look at the overall results, but to look at the results in the strata.”

How is that different from subgroups?

Abraham: “We specify lots of subgroups. Oftentimes subgroup specification in clinical trials is based on what we did in the last trial or what we think is appropriate to do. We stratify by sex, men and women, we stratify by age, older and younger, and so on and so forth. We don’t expect differences among those subgroups. But when we stratify randomization, we do that because we think it’s possible that the two strata may not be poolable. We may not see the same results, and we want to have the ability to independently analyze the two strata in a way that is perhaps statistically a bit more valid than just the typical subgroup analysis.”

What’s the trade off from that approach? Time and cost to get the enrollment numbers?

Abraham: “To some extent. You don’t have to individually power to either strata. A lot depends on how it all turns out at the end. There’s no crystal ball. We test the interaction between the strata using something called an interaction test. You’ve heard of the interaction p-value. Based on statistical principle, if the interaction p-value is insignificant, it means that the two strata are poolable and you analyze the results in the entire group. If the interaction p-value is positive, it indicates the two strata are not poolable. They behave very differently in the study. Then, statistical principles say you go on and you analyze the separate strata. You can’t stratify randomization on too many factors. We try to do it on the basis of one or at most two clinical factors. But it at least gives you an opportunity to come up with a hypothesis that we don’t think this therapy is going to depend on whether or not you’re a man or a woman, but we do think this therapy might depend on whether or not you have HFrEF or HFpEF. So let’s stratify on that basis, and based on these statistical principles, do the right test at the end of the day to see who really responds to this therapy.”

What’s your biggest challenge in the Carillon system’s pivotal trial and what’s the plan to overcome it?

Abraham: “The main challenge — but this is a challenge for every study — is enrollment. It’s getting the word out, it getting the referrals in. Oftentimes with trials like this it may be perceived as being a structural heart or interventional trial because it requires a talented implanting physician to put the device in place, so we don’t always get the message to the clinicians who actually have the patients and need to refer them into the trial: the heart failure specialist, the general cardiologist. Many of these patients are not living in the interventional cardiologist clinic, but they’re living in the heart failure clinic, the general cardiology clinic, and so on and so forth. So one of the challenges of enrollment in these trials that involve the convergence of heart failure and interventional cardiology or heart failure and electrophysiology is really engaging all of the stakeholders in the trial in a way that will lead to better or faster enrollment.”

What kind of tactics can device developers use to get the right people in their trial?

Abraham: “One of the most effective tactics we’ve used over the years is promoting the heart team approach and the co-equals at sites between implanting physician and managing physician to designate co-PIs (principal investigators). Usually institutions have to designate a single PI of record for IRB and contracting purposes, but there’s no reason why a study sponsor can’t designate a lead implanter and a lead heart failure specialist as co-PIs, or an implanter and heart failure specialist at the site. You get engagement from both specialties that are vitally important in executing the trial. It’s developing that partnership with the managing physicians and the implanting physicians to make sure there’s a smooth flow of patients from one to the other.”

Nitinol innovation: Download our free special report featuring nitinol expertise and tips from medical device OEMs