Rapid evolution of field means transcatheter valves ‘obsolete’ by long

In the field of medicine, long-term data have often been considered the gold standard of outcomes data.

And while this holds true in cardiology as well, for the burgeoning practice of transcatheter aortic valve replacement, there have been signs that the release of contemporarily relevant long-term data may prove elusive for the foreseeable future.

"The science is such a rapidly evolving field that there is no steady state where you can say, ‘Aha, I am going to follow this transcatheter valve ... because it hasn't changed,’" Dharam J. Kumbhani, MD, SM , said in an interview. "Our field is evolving so quickly that ... by the time we get meaningful longer-term data, that valve has become obsolete. A true Sisyphean challenge for all of us."

Kumbhani, who is associate professor of medicine at UT Southwestern Medical Center and cath lab director at William Clements University Hospital, spoke with Healio about this conundrum of long-term data, as well as the most clinically relevant long-term trials on TAVR and the challenges of treating patients who need a second TAVR procedure due to the tissue valves breaking down over time.

To date, one of the trials that Kumbhani said has provided the most clinically meaningful long-term results for low-risk patients with aortic stenosis is the Evolut Low Risk trial, the 3-year outcomes of which were presented at the American College of Cardiology Scientific Session in March.

In the trial, patients at low surgical risk were randomly assigned to TAVR with a self-expanding, supra-annular valve (n = 730; CoreValve, Evolut R or Evolut PRO; Medtronic) or surgery (n = 684).

At 3 years, the primary endpoint of all-cause mortality or disabling stroke occurred in 7.4% of patients receiving TAVR compared with 10.4% of patients receiving surgery (HR = 0.7; 95% CI, 0.49-1; P = .051). Furthermore, among the surgical group, the rates of mild paravalvular regurgitation (2.5% vs. 20.3%) and pacemaker placement (9.1% vs. 23.2%; P <. 001) were lower.

With this trial, Kumbhani said, "There is a little bit of yin and yang with each incremental year. Some years, surgery comes out on top for one endpoint, and for other years, TAVR comes out on top. The big picture I have been telling my patients is that TAVR and (surgical aortic valve replacement [SAVR]) are very similar. These 3-year data certainly support that."

Another trial Kumbhani said will help answer the long-term question for the majority of patients currently undergoing TAVR is PARTNER 3. The long-term, 5-year outcomes from this trial are expected to report out later this year; the 2-year data were released in 2021.

In all, 1,000 patients were randomly assigned to TAVR with a balloon-expandable valve (Sapien 3 [S3], Edwards Lifesciences) or surgery. At 2 years, TAVR demonstrated a lower rate of the primary endpoint of all-cause death, all stroke or CV rehospitalization with TAVR (11.5% vs. 17.4%; HR = 0.63; 95% CI, 0.45-0.88; P = .007).

Although current long-term results have shown promise for TAVR, they have yet to elucidate on some important aspects of the technology, including how long the tissue valves will last once implanted. Kumbhani said a 10- to 15-year shelf life may prove the correct ballpark, but cautioned that we do not yet have data to confirm this figure. A chief reason for this, he said, is the initial population of high-risk patients who received the valve.

"If you look at the patients who first received TAVR in the 2010 PARTNER trial, they had such a high competing risk of dying from other factors," Kumbhani said. "A successful TAVR was basically a patient who didn't die and walked out of the hospital."

In the years that followed, as the FDA approved TAVR in lower-risk populations, younger, less frail individuals began receiving the procedure; in essence, patients who could outlive the valves were receiving them, ultimately changing the definition of success.

"A successful TAVR is now a much more complicated issue because of the younger, lower-risk patients," he said. "It is not enough to just put the valve in and say, ‘OK, you are done.’"

As a result, Kumbhani said, "Five-year data, like the ones we are anticipating from PARTNER 3, are not long term if you take it in the perspective of a 70-year-old patient who is getting the valve today. [The trials] need to go at least 10 years because a 70-year-old patient, assuming they have no other issues, should live longer than 5 years."

While 10-year data may prove necessary to answer the durability question, the evolution of the technology seems likely to outpace such data in the years ahead.

"If you look at the PARTNER 3 trial, it is great that it features the S3 valve, but in practice we have already started routinely using the S3 Ultra Resilia, which is a different valve in many ways," Kumbhani said. "If this trial goes out to 10 years, we may be using the Sapien 4 valve by the time the results are published. So whether those 10-year results are going to be truly relevant to where we are at that time, I don't know."

Despite this, Kumbhani said he expects that at least some of the long-term durability questions will be answered this decade. In the meantime, he said, "There has been no clear signal that there is going to be a long-term durability issue."

Among patients who outlast their transcatheter valves, those who are healthy enough will need to undergo a second TAVR procedure. For Kumbhani, he has already begun to see an uptick in patients needing TAVR-in-TAVR in his practice.

"In the last year, we have had to treat three patients who had a failed TAVR valve, and while the procedures can be straightforward, some of them are very challenging," he said. "In fact, some of them are prohibitive, primarily due to an exceedingly high risk for coronary occlusion with TAVR-in-TAVR. In such cases, there is no way to put a second valve in without either leaflet modification or some other complex procedure (or SAVR). ... We are going to see this big epidemic this decade of patients needing TAVR-in-TAVR."

Consequently, Kumbhani has changed the way he approaches implanting the first TAVR valve. With this initial procedure, he asks himself a question: If this patient needs a TAVR-in-TAVR in the future, will the anatomy based on the first implant support the second procedure? "If so," he said," I then consider whether I should alter how deep or shallow I place the valve, so that if I did have to come back in 10 or 15 years, I haven't shot myself in the foot. Alternatively, if TAVR-in-TAVR looks challenging or prohibitive in the future and the patient is otherwise a good surgical candidate, SAVR may need to be preferentially considered. So those are questions we, as clinicians, need to start asking ourselves."

Despite the challenges that come with working in a new and fast-moving field, the technological advances have mostly been a tremendous boon for both clinicians and patients.

"Initially, all the conversation in TAVR was about paravalvular leak (PVL)," Kumbhani said, "but then PVL rates got really good, so it changed to durability. Now people are starting to talk about other aspects, like commissural alignment and coronary re-access."

These improvements and the overall change in discussion are the direct result of the progress made in the field of TAVR during the last 12 years.

"Now it is not just enough to show your valve is going to last 10 years and is not going to have PVL," Kumbhani said. "A lot more is being asked, and rightfully so, because TAVR has become the mainstream of how patients with aortic stenosis are being treated."

Dharam J. Kumbhani, MD, SM , can be reached at [email protected].