Midterm TAVR Data Turn Muddy: Will 10 Years Even Suffice for a Verdict?

NEW YORK CITY — It was increasingly clear that clinical trials alone won’t tell the full story of how transcatheter aortic valve replacement (TAVR) performs in the real world, according to midterm studies presented here.

In a 7-year report of PARTNER 3 — in which low-risk patients with symptomatic severe aortic stenosis were randomized to TAVR with the Sapien 3 valve or surgery — the two groups showed comparable and sustained valve durability, with low and similar rates of structural valve deterioration and bioprosthetic valve failure (BVF).

In particular, there were favorable results in the proportion of patients alive and free of all-cause BVF (73.4% with TAVR vs 74.8% with surgery, P=0.69) and the rate of valve reintervention (6.0% vs 5.5%, P=0.77) in the trial, reported Philippe Pibarot, DVM, PhD, of the Institut Universitaire de Cardiologie et de Pneumologie de Québec.

This analysis of PARTNER 3 was presented at the New York Valves annual conference hosted by the Cardiovascular Research Foundation, and was simultaneously published in JAMA Cardiology.

However, it was a different story when it came to real-world TAVR outcomes.

Among low-risk patients undergoing TAVR with the balloon-expandable Sapien 3 and its more contemporary iterations (i.e., Sapien 3 Ultra and Sapien 3 Ultra Resilia) in a registry analysis, valve reinterventions occurred in 2.2% of patients, and a whopping 45.1% died by 7 years. Even in the youngest TAVR patients (ages 65-74 years), the mortality rate was still 19.1%, while reinterventions occurred in 2.5%, reported Sreekanth Vemulapalli, MD, of Duke University Medical Center in Durham, North Carolina.

A high mortality rate was also evident following self-expandable TAVR in real-world practice. Registry data for the Evolut TAVR platform showed the proportion of deaths exceeding 50% by 7 years within the Veterans Affairs healthcare system, said Neel Butala, MD, of the Rocky Mountain Regional VA Medical Center in Aurora, Colorado.

Session co-moderator Martin Leon, MD, of NewYork-Presbyterian/Columbia University Irving Medical Center in New York City, noted that these analyses were complementary, and suggested it would be a situation of comparing apples and oranges in trying to resolve the huge discrepancy between the TAVR trial and registry data.

“We’re desperate for data. You’ve seen the rarefied environment of a prospective, carefully conducted randomized trial, but that’s not realistic, that’s not what we deal with every day,” Leon said. “I think if we’re going to try to reference this real world to randomized trials, we’ve got to be doing much, much better at really assigning populations that are reasonably equivalent to actually take the numbers seriously.”

For one, “low risk” in clinical practice was not the same as low risk in the PARTNER 3 trial population — the latter having excluded anyone with frailty, Leon noted.

Discussant Nimesh Desai, MD, PhD, of the Hospital of the University of Pennsylvania in Philadelphia, pointed out that “the mortality rate just seems to be too high versus our surgical cohort.”

“Put that in context with the recent COMMENCE data for surgical valves, the 10-year data with the Resilia valve in a surgical platform … and a 93% survival rate in a similar age population,” he said. “So I think a lot of this has to do with patient selection of surgical versus transcatheter therapies for patients with aortic stenosis.”

Fellow panelist Harold Dauerman, MD, of the University Of Vermont Medical Center in South Burlington, nevertheless stressed the “importance of registries and long-term follow-up, because it really does shape our conversations with patients. I don’t want you to control out the frail patients. When you look at the registries, those are the patients we’re seeing, and that’s the real-world mortality that we have to address.”

In any case, the present data are still, in their totality, wholly inadequate for a verdict on TAVR’s durability.

“We can’t just rely on 7-year data. We know that we’re going to need [longer-term data], especially because we know our patients are living longer,” said Suzanne Baron, MD, MSc, of Massachusetts General Hospital in Boston.

Fellow discussant Patrick O’Gara, MD, of Brigham and Women’s Hospital in Boston, added that “we know from our surgical experiences that we should be patient. We need to wait another 10 years.”