Super-Responders in COAPT: Improving MR and QoL Is Key

Fewer than one in five patients enrolled in the COAPT trial of the MitraClip (Abbott) were “super-responders” to their assigned therapies, with more seen in the percutaneous intervention group than in the medical control group. But understanding at the outset which patients will benefit most from the clip remains murky.

Writing in an early-online publication of the Journal of the American College of Cardiology, Paul A. Grayburn, MD (Baylor University Medical Center, Dallas, TX), and fellow COAPT investigators state that baseline predictors of being a super-responder in the trial included lower serum creatinine, low quality-of-life score, and treatment with the MitraClip itself.

This last finding, said Grayburn, drives home the point that the key to clinical improvement in functional mitral regurgitation (MR) is reducing that MR.

“When we looked at MR grade and right ventricular systolic pressure at 30 days, they were a lot better in the super-responders compared to the nonresponders, which implies that the biggest predictor of being a super-responder is getting a successful MitraClip procedure that reduces your MR grade and reduces your RV systolic pressure,” he told TCTMD.

COAPT Responders

As previously reported by TCTMD in 2018, COAPT was a randomized clinical trial of more than 600 patients with heart failure and severe functional MR showing that transcatheter mitral valve repair using the percutaneous clip on top of guideline-directed medical therapy (GDMT) significantly reduced not only the primary endpoint of heart failure (HF) rehospitalizations by 47%, but also mortality at 2 years, by 38%, as compared to people treated with GDMT only. It was at odds, however, with the French MITRA-FR trial of 304 patients, which found no benefit with the MitraClip in reducing death or HF hospitalizations.

In the wake of those two trials, there have been multiple efforts to explain the different outcomes as well as a push to figure out which patients would truly benefit from clip therapy on top of optimal medical care. This latest study, which was post hoc, attempted to get at answers to this question by stratifying patients as super-responders, responders, or nonresponders according to prospective definitions.

Super-responders were defined as patients alive at 12 months with no hospitalizations and an improvement in Kansas City Cardiomyopathy Questionnaire overall summary (KCCQ-OS) of 20 points or more. Responders were defined with the same clinical criteria and a KCCQ-OS improvement of 5-20 points, while nonresponders were those who had died, been hospitalized for heart failure, or saw their scores improve by less than 5.

Overall, after excluding 41 patients with missing KCCQ-OS scores, 27.2% of the trial participants in the MitraClip group were categorized as super-responders, 19% as responders, and 53.8% as nonresponders. By contrast, in the medical therapy arm, 10.2% were super-responders, 16.3% were responders, and 73.5% were nonresponders.

In multivariate analyses, the only three baseline predictors of super-response in the device arm were lower serum creatinine, low KCCQ-OS score, and MitraClip treatment. At 30 days, decreased MR severity as well as reduced RV systolic pressure were associated with better outcomes over the longer term.

That KCCQ-OS scores helped to predict a good response to therapy was not particularly surprising, Grayburn said—and has been hinted at in earlier analyses. “That's expected because if you really feel bad, you could get better, so the people with the worst KCCQ scores have the greatest improvement,” he said. “And in a way that highlights the fact that if you're asymptomatic, it's hard to get better with the MitraClip.” COAPT, he noted, did not enroll asymptomatic patients.

That assignment to the MitraClip device was such a strong predictor of being a super-responder reinforces what’s known about the mechanism, Grayburn continued. He pointed to an analysis by COAPT investigator Saibal Kar, MD—reported by TCTMD at EuroPCR last year—showing that regardless of treatment assignment in the trial, having a lower residual MR at 30 days was strongly associated with reduced heart failure hospitalizations, all-cause mortality, and improved quality of life as compared to higher degrees of residual MR.

The fact that one in ten patients in the medical therapy group were super-responders also carries an important message, Grayburn said. “We didn't really expect that, but I think it speaks to the fact that functional MR is dynamic and it can continue to respond to medical therapy over time. So even if we optimize medical therapy today and wait a month and then enrolled you in a trial, you can still get better 6 months or 1 year down the road.”

And that’s another key message from COAPT, he continued, especially for interventional cardiologists keen on using the clip: to be eligible for this trial, patients had to be truly symptomatic despite having maxed out on meds. “What we're calling ‘guideline-directed medical therapy’ requires uptitration of ACE inhibitors and ARBs or ARNI and beta-blockers to the highest doses that were successful in clinical trials of heart failure,” said Grayburn.

A recent study from the CHAMP-HF registry showed that only 1% of the eligible heart failure population in the United States is getting all of their medications at the recommended doses, he noted, “which is very, very sad because those drugs reduce your mortality 30%, and your heart failure, and they're not being used properly. So what we really need is for physicians to recognize that functional MR is a disease of the left ventricle and [that] properly and aggressively treating the left ventricular dysfunction will improve a lot of patients in terms of mortality and symptoms.”

Still Work to Be Done

Commenting on the analysis for TCTMD, Alec Vahanian, MD (Hôpital Bichat, Paris, France), who was on the steering committee for MITRA-FR, said it is “very interesting and adds to the current knowledge.”

What’s clear, he said in an email, is that predicting outcomes requires an understanding of both the patient’s global condition as well as the potential success of the intervention. Less obvious is how to identify “responders” more broadly, “which is less catchy than ‘super-responders,’” he noted. “This is not clear today because of the discordant results of the two randomized controlled trials, so the discussion about what could be the definition of ‘response to [GMDT]’ is also open. But I congratulate the COAPT investigators for looking closely at KCCQ, which is rather unique and necessary from a patient's perspective.”

A quality-of-life score such as the KCCQ was not, unfortunately, part of MITRA-FR, Vahanian added, which will complicate efforts to combine the cohorts—as proposed by Grayburn and colleagues—to help better identify patient-level characteristics that could help predict a response to therapy.

Rebecca Hahn, MD (NewYork-Presbyterian/Columbia University Irving Medical Center, New York, NY), who wrote an editorial accompanying the new paper, notes that given the very positive results for the MitraClip device in the COAPT trial, the results of this analysis “seem a forgone conclusion.” But the analysis does raise additional issues, she points out, including whether or not being a super-responder translates into meaningful outcomes down the road, and whether changes in right or left ventricle size in top responders in both groups translates into long-term benefit.

Hahn also points out that serum creatine (as well as KCCQ-OS) predicted super-response not only in patients randomized to MitraClip but also in those assigned to GDMT and as such might be a marker for patients in whom GDMT was not maximized due to concerns over renal function.

Finally, Hahn notes, “the high nonresponder rate in both treatment groups (53.8% vs 73.5% for transcatheter mitral valve repair plus GDMT and GDMT, respectively) is surprising and deserves our attention.”

A “deeper look” is warranted to determine how the pathoanatomy of MR, optimization of GDMT, and other factors play a role in responsiveness, she writes. Doing so “may help clinicians make informed decisions regarding appropriate treatment options, understanding that both transcatheter devices and GDMT continue to improve.”

Vahanian also zeroed in on the high proportion of patients who saw no benefit from the device. “The relatively large percentage of nonresponders in the MitraClip arm after only 1 year suggests that there is a lot of work to be done to improve the efficacy of the strategy: earlier and better detection of MR/HF, better medical therapy before and after the procedure, and better interventions” may be needed, he predicted, including newer devices or devices in combination.