COAPT at 5 Years: MitraClip Still Ahead, but Deaths and Crossovers Cast Shadows

NEW ORLEANS, LA—Benefits of transcatheter edge-to-edge repair (TEER) over optimal medical therapy in patients with severe secondary mitral regurgitation (MR) continue to be seen over long-term follow-up in the COAPT trial. But clip therapy is no panacea: at 5 years in, nearly three-quarters of patients treated with MitraClip (Abbott) had died or been hospitalized for heart failure.

Those sobering event rates, notably, were lower in the device group than in the control group, where more than nine out of 10 patients had been hospitalized for heart failure or died.

“Despite the favorable risk-benefit profile of the MitraClip in this setting, adverse events continued to accrue,” said Gregg Stone, MD (Icahn School of Medicine at Mount Sinai, New York, NY), who presented the 5-year data here at the American College of Cardiology/World Congress of Cardiology (ACC/WCC) 2023 meeting. Results were simultaneously published in the New England Journal of Medicine.

“These findings,” he said, “emphasize the need for further therapies to address the underlying left ventricular dysfunction in this high-risk population.”

Despite the favorable risk-benefit profile of the MitraClip in this setting, adverse events continued to accrue. Gregg Stone

But Nathan Mewton, MD, PhD (Hôpital Cardiovasculaire Louis Pradel, Lyon, France), senior investigator for the only other randomized trial of the MitraClip in this population—MITRA-FR—was unconvinced. Speaking with TCTMD, he pointed out that over the last 5 years there have been multiple, much larger randomized trials testing medical therapies in heart failure that have shown meaningful LV improvements. In COAPT, meanwhile, so many patients have died, withdrawn from the study, or crossed over from control therapy to a clip repair, that it’s too “confusing” to draw any solid conclusions from this follow-up.

“I'm concerned that this trial being published in the New England Journal Medicine is encouraging people to be monosynaptic about a complex question. Secondary MR is not that simple,” Mewton said. “We cannot forget that secondary MR is first a disease of the myocardium, and that's more where we should act, not on the mitral valve.”

COAPT Then and Now

Back in 2018, when the long-running, randomized, 614-patient COAPT trial was first presented, TEER on top of maximally tolerated guideline-directed medical therapy (GDMT) in patients with moderate or severe mitral regurgitation was superior to GDMT alone in reducing annualized rates of hospitalizations for heart failure: 35.8% versus 67.9% (P < 0.001) at 2 years. Mortality, the bigger surprise, powered as a prespecified secondary endpoint, was also significantly lower among MitraClip treated patients: 29.1% versus 46.1% (HR 0.62; 95% CI 0.46-0.82).

Now, nearly 5 years later, there have been 447 heart failure hospitalizations among 208 patients in the control group and 314 among 151 patients in the TEER group, yielding an annualized rate of hospitalizations of 57.2% and 33.1%, respectively (HR 0.53; 95% CI 0.41-0.68).

More than half of device-treated patients were dead (57.3%), compared with more than two-thirds in the control group (67.2%; HR 0.72; 95% CI 0.58-0.89). But that gap is narrowing, in part due to the almost 45% of surviving control group patients (n = 67) who crossed over, electing to undergo TEER after 2 years, an option specified in the protocol.

I'm concerned that this trial being published in the New England Journal Medicine is encouraging people to be monosynaptic about a complex question. Nathan Mewton

Death or hospitalizations combined reached 91.5% in the control group, as compared with 73.6% in the MitraClip group (HR 0.53; 95% CI 0.44-0.64).

Strikingly, the event curves for heart failure hospitalizations, mortality, and the two combined over 3 years of follow-up in patients who crossed over at 2 years (permitted in the study protocol) were markedly similar to those of patients treated with TEER at the study outset.

Also seen at 5 years: symptomatic heart failure class improved, and MR reduction was typically durable over extended follow-up. Among the 57 device-treated patients alive at 5 years, the vast majority had MR grade 1+ and roughly 20% had no residual MR.

Stone reported that medical therapy usage was mostly similar between the two arms over follow-up. Just three patients started taking an SGLT2 inhibitor (all in year 5), and very few patients were taking sacubitril/valsartan (Entresto; Novartis) over the trial follow-up, although notably, use of this class of therapies was greater in the device group than in controls.

What’s needed now, believes Mewton, is a redo of COAPT in which all patients were optimized first on four-pillar heart failure therapy and cardiac resynchronization therapy as needed.

Stone, asked if a do-over is in order, is convinced the trial results would be unchanged, even with modern-day GDMT. “I'm personally very confident you would see the same results,” he said. “What the effect of those agents would be [is] to more effectively medically treat some patients to reduce their severe MR, so it would decrease the pool of the patients that would require transcatheter edge-to-edge repair with the MitraClip, but the ones in whom it was not effective would still need this.”

Eric Velazquez, MD (Yale School of Medicine, New Haven, CT), said he doesn’t think a new trial is warranted, although registry analyses tracking TEER outcomes, as well as concomitant medical therapy and how well it is uptitrated prior to TEER referral, will be essential.

While it’s “disappointing” that SGLT2 inhibitor and sacubitril/valsartan use was not more ubiquitous in the latter years of COAPT, said Velazquez,  the heart failure community has gotten the message that four-pillar therapy is the place to start.

“The strategy that I employ, and I would recommend that we take from the COAPT data, is that we have to initiate GDMT in the population with moderate/severe or severe MR and heart failure as quickly as possible. . . . I think in those patients who continue to have risk of future events and/or persistent, severe MR—and I think that's going to be a sizable population—then the TEER program is their best pathway to reducing events.”

I think in those patients who continue to have risk of future events and/or persistent, severe MR—and I think that's going to be a sizable population—then the TEER program is their best pathway to reducing events. Eric Velazquez

Wayne Batchelor, MD (Inova Heart and Vascular Institute, Falls Church, VA), speaking with the press, had a similar take, noting that the high mortality in the medical arm “speaks volumes” and supports the early identification of patients who would benefit most from TEER. “We don’t know exactly how this would pan out” in the setting of routine ARNI and SGLT2 inhibitor use, Batchelor acknowledged. “Medical therapy has come a long way, but the dramatic difference between the treated arm and the medical therapy arm in this trial makes me feel that this [procedure] is here to stay—we just need to figure out how to deploy it properly.”

Indeed, Tariq Ahmad, MD (Yale University School of Medicine), also commenting on the study for TCTMD, stressed that how and when TEER is used are key. “At first blush, these are incredibly impressive findings,” he agreed. “I think it shows that in the very select population that was studied in COAPT, the impact of a MitraClip can be very impressive on both ‘hard’ outcomes and symptoms.”

That said, Ahmad continued, “I think the bigger question is: how does this population compare to the patients who get most of the MitraClips in the US? I would guess that they are very different and the benefit they are reporting here is nowhere close to what we have seen in the real world; in fact, there is a very real possibility that it might be causing harm in many patients. . . . I hope this results in a serious effort to study who is currently getting MitraClip.”

What to Take From COAPT?

Stone offered two “main messages” from these follow-up data. “First, treatment of secondary mitral regurgitation in a high-risk heart failure population with the MitraClip led to a persistent durable reduction in both deaths and heart failure hospitalization at 5 years, with improved quality of life, and it was a very safe therapy,” he told TCTMD. “But number two, the effects sizes that we saw were mitigated, or were lessened, after 2 to 3 years, in large part because of crossover of patients in the control group that were allowed to be treated with the MitraClip after 2 years.”

Those crossovers—nearly half of the survivors—are also revealing, Stone insisted. “The major take-home message from that is that we should be identifying these patients with heart failure and severe secondary MR who can benefit from transcatheter edge-to-edge repair and treat them as early as possible, to try to prevent both mortality and heart failure hospitalization and improve their quality of life.

Mewton, however, pushed back on this, pointing out that there is “nothing” in this long-term COAPT data at 5 years to support intervening earlier in disease—something he said should be “urgently” tested. Instead, he observed there have been somewhere in the order of 30 papers generated from COAPT, an Abbott-sponsored trial, while MITRA-FR, the only other randomized controlled trial, was investigator initiated and found opposite results. There has been no new randomized trial data to settle the question left open by the disparate results of these two trials.

Indeed, he argued, there’s not enough in this 5-year data to persuade him that an edge-to-edge approach provides a meaningful long-term solution for patients whose primary disease is left ventricular dysfunction, not mitral disease.

Given the high mortality rates, the high rate of crossovers among survivors, as well as the fact that 15% of the original control group and 10% of the TEER group withdrew from the study or were lost to follow-up, Mewton is scratching his head.

“You don’t really know what you’re looking at,” he said. “You’re lost because of all of this loss of data. . . . The only scientifically reasonable conclusion that you can take is that the benefits shown at 2 years of follow up remain, but then that data is also confusing.”

I think the bigger question is: how does this population compare to the patients who get most of the MitraClips in the US? I would guess that they are very different. Tariq Ahmad

In another distracting finding that speaks to the ongoing discussions of how best to identify who will benefit from TEER long-term, half of the survivors in the control arm were no longer eligible for the MitraClip at 5 years. Questioned on this, Stone pointed to some selection bias: patients with the most severe MR at baseline were the ones most likely to die over 5 years, while others actually saw their mitral regurgitation improve beyond the moderate/severe cutoff established in the original trial, possibly due to sacubitril/valsartan use.

“There is also the fact that secondary mitral regurgitation is a dynamic process—sometimes it's more severe than other times—and so it may just be that at that exact 2-year period the mitral regurgitation was less than 3+, and it may have been greater than 3+ later,” Stone said.

Discussing the results after Stone’s late-breaking presentation in New Orleans, Suzanne Baron, MD (Lahey Hospital & Medical Center, Burlington, MA), asked about the proportion of deaths adjudicated to be cardiac in nature. Stone, in response, said approximately 75% of the deaths were cardiovascular , and about two-thirds of those were adjudicated as being heart failure.

“So that really does seem to suggest that this mortality benefit is related to what we're doing here within the heart,” Baron responded.

Pointing to the very similar early outcomes in the crossover patients, Baron speculated that “waiting to treat patients with MR may be harmful. [But] if we're going to consider treating earlier, how do we identify the right patient?. . . How do you recommend to hospital programs and implanters out there to ensure that we are picking the right patients for this treatment?”

Stone, in response, referenced the hypothesis he’s raised in the past, that the ideal patient would be one with severe MR, but with left ventricular dysfunction not too severe, which he maintains is what differentiates MITRA-FR patients from those in COAPT. Investigators hope to publish 5-year echocardiographic data after the analyses are done.

“This is a complex analysis for a lot of reasons,” said Stone. “For one, there's a lot of selection bias. Two . . . relieving mitral regurgitation actually increases afterload, and so there's kind of countervailing effects. While you'll end up with increased cardiac output, you may not see the effects on left ventricular ejection fraction and some other simple-to-measure parameters that you would expect.”

Velazquez offered TCTMD the reminder that epidemiological studies, including some that he himself has been involved with, really documented the poor disease course of these heart failure patients with severe MR in years past. For the right patients, TEER may be lifesaving, so long as other, proven therapies are tried first—he has seen a tendency on the part of some enthusiastic centers to treat TEER as a go-to option before opportunities for medical therapy and CRT have been explored in full.

“The community has to hold itself to a very high standard. We have to get these patients on the right therapies. We have them, they're very evidence-based for heart failure and LV dysfunction, but if that MR persists and is symptomatic, I think it's a no-brainer to provide patients with this option,” he said.