EVOLVED: Early AVR Fails to Help Asymptomatic AS Patients With Fibrosis

WASHINGTON, DC—An early aortic valve replacement (AVR) strategy with either TAVI or surgery fails to lower the risk of all-cause mortality or unplanned aortic stenosis-related hospitalizations in asymptomatic patients with evidence of myocardial fibrosis, the EVOLVED study shows.

The 244-patient study, which was presented today during the first late-breaking clinical trial session of TCT 2024, throws cold water on the possibility of using cardiac magnetic resonance (CMR) to find and treat asymptomatic patients with severe aortic stenosis and early evidence of structural remodeling.

Despite not reducing the primary endpoint, investigators see some positives in their data, particularly the reduction in unplanned hospitalizations for aortic stenosis-related reasons.

“We do a lot of public engagement, and we chat to patients in these groups about what's important to them,” senior investigator Marc Dweck, MBChB, PhD (University of Edinburgh, Scotland), told TCTMD. “What’s important is quality of life, not quantity of life. They are much more interested in staying healthy, keeping well, and staying out of hospital.”

Although there are differences in trial design and patient population, Dweck sees similarities with EARLY TAVR, another study testing a strategy of early intervention, using TAVI only, in asymptomatic patients. In that trial, presented right before EVOLVED at TCT 2024, there was a significant reduction in the primary endpoint of death, stroke, or unplanned cardiovascular hospitalizations when patients were treated with TAVI instead of a conservative strategy of watchful waiting. The robust reduction in the primary endpoint in EARLY TAVR was driven by reduced hospitalizations. 

“I think that's the story here,” said Dweck. “Intervening early isn't necessarily going to make you live longer because there's other things that come into play, but it can make you live better. It keeps you out of hospital. Our patients were less symptomatic in a year—the people who had an early intervention, their symptomatic status was the same a year down the line as it was at baseline whereas the people who had delayed intervention were more symptomatic.”

Intervening early isn't necessarily going to make you live longer because there's other things that come into play, but it can make you live better. Marc Dweck

Interventional cardiologist Chetan Huded, MD (Saint Luke’s Mid America Heart Institute/University of Missouri—Kansas City), who wasn’t involved in the study, said he was “a little surprised” that EVOLVED was negative in light of two small trials—RECOVERY and AVATAR—that showed a benefit of intervening early with surgical aortic valve replacement (SAVR) in patients with asymptomatic, severe aortic stenosis.

“[EVOLVED] is a small trial and clearly testing a hypothesis that I think a lot of people believe to be true, which is that earlier treatment of patients with aortic stenosis prior to the development of symptoms is probably beneficial,” Huded told TCTMD. In contrast with RECOVERY and AVATAR, though, patients in EVOLVED were older and had less severe aortic stenosis, which may explain the different results.

As a result, it’s difficult to draw hard conclusions about the early treatment strategy used in EVOLVED. “It’s hard to show a really important difference in clinical events like mortality in such a small study,” said Huded.

Benjamin Hibbert, MD, PhD (Mayo Clinic, Rochester, MN), another interventional cardiologist not involved in EVOLVED, said that based on its results, stratifying asymptomatic patients for early intervention with AVR based on the presence of cardiac fibrosis is not the way to go.

While there are some plausible explanations for why the study was neutral, “I’m not sure I would want to put a lot more effort into chasing this strategy, because it does look like a bit of a bust,” he said. “It’s really hard to take someone who’s doing well and make them better.”  

He noted that while myocardial fibrosis is associated with adverse outcomes, “it’s a great lesson that we learn over and over again: just because a surrogate tracks with poorer outcomes, it doesn’t mean that intervening in the disease alters the natural history of the disease.”

The EVOLVED results were published today in JAMA, with Krithika Loganath, MD (University of Edinburgh), as lead author.

What the US Guidelines Say

In the US valvular heart disease guidelines, aortic valve replacement (AVR) with TAVI or surgery is a class IA recommendation for patients with symptoms of exertional dyspnea, heart failure, angina, syncope, or presyncope (by history or on exercise testing). In asymptomatic patients, earlier AVR is indicated for those with severe aortic stenosis and LVEF less than 50% or those undergoing cardiac surgery for other indications (both class I, level of evidence B).

The guidelines also state that AVR is a reasonable decision—a class 2A recommendation—in selected asymptomatic patients with severe aortic stenosis, such as those with elevated serum B-type natriuretic peptide (BNP) levels or decreased exercise tolerance, as well as in those with rapid disease progression or very severe aortic stenosis.

“The big question in cardiology over the last 50 or 60 years is when we should replace the valve,” said Dweck. “Our current approach is to wait for symptoms, but it’s difficult to assess symptoms in these elderly patients. They've often got comorbidities and how do you know their breathlessness is due to their valve disease and not due to lung disease? At the other end of the spectrum, they're often poorly mobile and unable to exercise so it can be difficult to know if they are symptomatic or not.”

Myocardial fibrosis on CMR is an irreversible marker of LV decompensation in patients with aortic stenosis. With EVOLVED, investigators wanted to determine if preemptive AVR would be beneficial in an enriched population at high risk of events given the early signs of myocardial scarring. Patients were eligible for the trial based on high-sensitivity troponin I concentrations or 12-lead ECG criteria and underwent cardiac MR to confirm midwall myocardial fibrosis.

It’s really hard to take someone who’s doing well and make them better. Benjamin Hibbert

The study included 224 patients (mean age 73 years; 28% female) with asymptomatic severe aortic stenosis and midwall scarring randomized to an early AVR intervention (surgery or TAVI determined by the local heart team) or guideline-directed conservative management.

In total, 94% of patients randomized to early intervention underwent AVR, with 86% receiving the new valve within 12 months of randomization. The median time to intervention was 5.0 months, with 75% of patients treated with SAVR. In the guideline-directed conservative treatment arm, the median time to intervention was 20.2 months. Here, 77% received a new valve, with 28% treated within the year. In this group, SAVR was performed in 55% of patients. 

Early intervention failed to lower the risk of all-cause mortality or unplanned hospitalization for aortic stenosis compared with conservative care (HR 0.79; 95% CI 0.44-1.43). There was no significant difference in the risk of all-cause mortality, but there was a significant 63% lower risk of unplanned hospitalizations with the early intervention (HR 0.37; 95% CI 0.16-0.88). In this trial, the aortic stenosis-related hospitalization was defined as any unplanned admission before or after aortic valve replacement with syncope, heart failure, chest pain, ventricular arrythmia, or second- or third-degree heart block attributed to aortic valve disease (adjudicated independently by two blinded investigators).

An improvement in NYHA symptom class from baseline to 12 months occurred more frequently with the early intervention. Overall, 80% and 18% of patients in the early-intervention group had NYHA functional class I and II symptoms at 1 year, respectively, compared with 62% and 29% in the conservative treatment arm (OR for higher NYHA classification 0.37; 95% CI 0.20-0.70).

Underpowered, but That Likely Didn’t Matter

EVOLVED investigators originally planned to include 356 patients, but recruitment slowed due to the COVID-19 pandemic, leaving the trial underpowered for the primary endpoint. Despite this, Dweck doesn’t think that having more patients in the trial would change the results.

“The trial may be underpowered, but EARLY TAVR showed basically the same thing,” said Dweck. “In these elderly people with severe aortic stenosis, average age in the mid-70s, it's really hard to shift mortality. If you look at the causes of death in our population, only a third were due in any way to their valve disease and we used a very liberal definition of aortic stenosis-related death.”

With two-thirds of patients dying from other causes, fixing the aortic valve is not going to impact mortality, he added. “What you can affect is heart failure and hospitalizations.”

Hibbert agreed that adequate statistical power wouldn’t “necessarily” move the needle. “There was an up-front risk with the early-intervention strategy—all-cause mortality was higher in the first year—before the event curves crossed and then ran parallel. Moreover, there was a large difference in the promptness of AVR in EVOLVED compared with EARLY TAVR—5 months to AVR in the former versus 14 days to TAVI in the latter—and he wondered if the extensive delay in EVOLVED might explain the neutral results.

As for the endpoint of aortic stenosis-related hospitalizations, Hibbert called it a bit of an “odd duck,” noting that it’s not something he’s ever seen used in prior clinical trials. Beyond 1 year, the Kaplan-Meier curve for this endpoint is flat, meaning there were no hospitalizations related to aortic stenosis from year 1 to 5, which he found difficult to comprehend. Apart from that outlier, Hibbert only had praise for the investigators’ interpretation of the data, especially their conclusion that risk-stratification using myocardial fibrosis in asymptomatic patients isn’t effective.

If we're going to follow somebody with asymptomatic, severe aortic stenosis with surveillance, which is kind of the conventional way, we make sure that that person is safe and we don’t see any risk. Chetan Huded

To TCTMD, Huded pointed out that there are entirely asymptomatic patients with high functional capacity despite the presence of severe aortic valve disease. The majority, however, are slowly beginning their decline, and symptoms of aortic disease can be subtle.

“Our job is to try to tease that out,” said Huded, noting they rely on careful histories and use of patient-reported outcomes to quantify symptom burden. They will also utilize treadmill testing, if possible. “If we're going to follow somebody with asymptomatic, severe aortic stenosis with surveillance, which is kind of the conventional way, we make sure that that person is safe and we don’t see any risk.”  

In 2023, Huded led a study published in JACC: Cardiovascular Interventions based on more than 230,000 patients treated with TAVI between 2015 and 2021 in the TVT Registry. Only 20% of those treated were minimally symptomatic, but they had more favorable outcomes than those with moderate or more severe symptoms.

“There's an overwhelming amount of evidence moving in this direction that we're waiting a little bit too long if we wait for symptoms in severe aortic stenosis,” he said. “The tricky part now is finding out which patients and how early [do we intervene]? What type of valve replacement? Those are all the questions that are made to be answered.”

Cardiothoracic surgeon Suyog Mokashi, MD (Temple University, Philadelphia, PA), said close surveillance is the goal in asymptomatic patients with severe aortic stenosis.

“I’d bring them back in a very short interval,” he told TCTMD. “We would do a real deep dive to make sure they don't have symptoms, and if they truly don't have symptoms, then we just bring them back in a shorter period of time. Instead of saying, ‘Hey, we'll see you in 3 months,’ we might see them in a month. We tend to follow them quite closely if their aortic stenosis is severe. We don't want to miss the window when they start becoming symptomatic.”

As for the future of CMR, Dweck said there are challenges with the technology, including around access, but that it is quite routinely used for assessment of patients with heart failure and cardiomyopathy. Based on EVOLVED, CMR-assessed fibrosis will not impact clinical decisions around AVR, he said, though preponderance of data—EVOLVED, EARLY TAVR, RECOVERY, and AVATAR—suggest there is a benefit to early intervention in some asymptomatic patients.

“I’ve been thinking about how I'm going to approach this issue with my patients coming up,” he said. “It’s going to be an individualized discussion, because I think we should be offering early intervention to our patients.  And I think some will [say], ‘Leave me alone, doctor. I feel absolutely fine. I don't want to have an operation or a procedure.’ Other people are more proactive about their health and they feel well and they want to keep well. In those patients, we should be offering them early intervention.”