SAN FRANCISCO — A handful of years later, transcatheter aortic Valve replacement (TAVR) still compared well against surgery in the PARTNER 3 and Evolut Low Risk trials.

PARTNER 3 appeared to show a convergence of the TAVR and surgical aortic valve replacement (SAVR) arms of the study by 5 years, when the composite endpoint of death, stroke, or rehospitalization numerically favored TAVR but ultimately came down equal between groups (22.8% vs 27.2%, HR 0.79, 95% CI 0.61-1.02), reported Martin Leon, MD, of NewYork-Presbyterian/Columbia University Irving Medical Center in New York City.

In contrast, diverging event curves put TAVR patients right on the cusp of statistical significance for a reduction in 4-year combined all-cause mortality and disabling stroke in Evolut Low Risk (10.7% vs 14.1% with SAVR, HR 0.74, 95% CI 0.54-1.00, P=0.05), according to Michael Reardon, MD, of Houston Methodist Debakey Heart & Vascular Center.

Both reports — extension studies of the PARTNER 3 and Evolut trial that had been pivotal in FDA approving TAVR’s extension across all surgical risk categories — were presented as late-breaking studies here at the Transcatheter Cardiovascular Therapeutics (TCT) meeting hosted by the Cardiovascular Research Foundation.

During a TCT press conference, Bernard Prendergast, MD, of St. Thomas’ Hospital and Cleveland Clinic London, called the new data “very exciting” as they “importantly reaffirm guideline recommendations from both sides of the Atlantic.” In a “bold step” at the time 3 years ago, the U.S. guidelines had lowered the age for arbitrating TAVR vs SAVR down to 65 years, Prendergast recalled.

Leon and Reardon both emphasized that the long-term performance of TAVR still needs to be assessed for people with severe, symptomatic aortic stenosis and low surgical risk.

“Valve durability, one of the key things in younger patients, is not a class effect,” Reardon stressed. “We really need to get out to 10 years.”

“I’m encouraged by these data but certainly would need to look at the 10-year endpoint to draw any firm conclusions,” agreed Kendra Grubb, MD, of Emory University in Atlanta. “We’ve seen surgical valves where the problem starts at 7, 8 years with newer technology.”

PARTNER 3 at 5 Years

By 5 years, clinical outcomes from TAVR with the balloon-expandable Sapien 3 valve seemed to converge with those of surgery, the PARTNER 3 group reported.

The study’s first primary endpoint could be broken down into individual endpoints all supporting a convergence between TAVR and SAVR:

• Death: 10.0% and 8.2% (HR 1.23, 95% CI 0.79-1.90)
• Stroke: 5.8% and 6.4% (HR 0.87, 95% CI 0.51-1.48)
• Rehospitalization: 13.7% and 17.4% (HR 0.75, 95% CI 0.54-1.05)

PARTNER 3’s 5-year report was simultaneously published in the New England Journal of Medicine.

These data are less favorable for TAVR than the shining 1-year report that had found a significant lowering of deaths, strokes, and rehospitalizations with this approach (8.5% vs 15.1%, HR 0.54, 95% CI 0.37-0.79). Two years into the PARTNER 3 trial, the event curves were already starting to narrow (11.5% vs 17.4%, HR 0.63, 95% CI 0.45-0.88); at that point, rehospitalization appeared to be the only endpoint component still significantly in TAVR’s favor.

The changes in mortality 5 years after TAVR were attributed to disproportionately more noncardiovascular deaths in this arm, Leon explained during a TCT press conference.

As for hemodynamic performance of Sapien 3 at 5 years, Leon reported mean valve gradients nearly superimposing at 12.8 mm Hg in the TAVR group and 11.7 mm Hg in the SAVR group. Bioprosthetic valve failure rates reached 3.3% and 3.8%, respectively.

Valve thrombosis occurred in substantial excess after TAVR (2.5% vs 0.2%, HR 10.52, 95% CI 1.37-80.93).

“The reasons for the greater incidence of valve thrombosis among TAVR patients remain speculative, but this event did not appear to affect valve durability at 5 years. It is possible that the differences between the groups in the use of anticoagulation therapies during the first years after the procedure may have contributed to the higher incidence of thrombosis in the TAVR group, but this is also unknown,” study authors wrote.

Leon added that the trial’s second primary endpoint — counting death, disabling stroke, nondisabling stroke, and the number of rehospitalization days at 5 years — yielded a nonsignificant win ratio of 1.17 for TAVR (95% CI 0.90-1.51).

Atrial fibrillation and bleeding were less frequent in the TAVR group than in the surgery group, whereas surgery patients had the advantage for lower paravalvular aortic regurgitation, valve thrombosis, and pacemaker implantation rates. Functional and quality of life outcomes appeared similar among survivors at 5 years.

Evolut Low Risk at 4 Years

Whereas Sapien TAVR seemed to converge with surgical outcomes in PARTNER 3, TAVR with the self-expanding Evolut platform pulled away from surgery in a separate low-risk trial.

Its borderline advantage in the primary endpoint at 4 years was arguably more driven by death (9.0% with TAVR vs 12.1% with SAVR, log-rank P=0.07) and less so by disabling stroke (2.9% vs 3.8%, log-rank P=0.32).

Reardon also reported that significantly better hemodynamics were maintained with Evolut TAVR vs SAVR: at 4 years, mean gradients were 9.8 vs 12.1 mm Hg, respectively (PP=0.002).

Rates of severe patient-prosthesis mismatch reached 1.1% vs 3.5%, also favoring TAVR (P=0.008).

“In low-risk patients, the Evolut platform [is a good] first choice due to valve performance and associated excellent clinical outcomes,” Reardon concluded, citing the lower clinical event rates, superior hemodynamics, and lower rates of structural valve deterioration in previous research.

There were no between-group differences in severe paravalvular regurgitation, valve endocarditis, or valve thrombosis in Evolut Low Risk. These findings from 4 years were reported in a brief letter published in the Journal of the American College of Cardiology.

Notably, primary endpoint events were on track for a separation of curves between TAVR and SAVR. “The absolute difference between treatment arms for the primary endpoint continued to increase over time: 1.8% at 1 year, 2.0% at 2 years, 2.9% at 3 years, and 3.4% at 4 years,” the authors reported.

Three years into the trial, the TAVR arm was just shy of reaching statistical significance for reduced less all-cause mortality and disabling stroke compared with SAVR recipients (7.4% vs 10.4%, HR 0.70, 95% CI 0.49-1.00, P=0.051).

Reardon cautioned against comparing the Sapien 3 and Evolut devices directly based on the newest data, citing the differences in baseline risk between the study cohorts.

“Both of these trials are wins for TAVR,” agreed Pinak Shah, MD, of Brigham and Women’s Hospital, Boston, another panelist at the press conference. “Which device I’m going to choose for younger low-risk patients is entirely dependent on patients’ anatomy.”

Reardon also noted that the trial excluded many low-risk individuals owing to their bicuspid anatomy or heavily calcified left ventricular outflow tracts, among other criteria limiting its applicability to real-world practice.

“That’s excluding a whole lot of people I see anyway,” he said. “I have 55-year-olds coming in asking for TAVR. I have no idea what’s going to happen to them.”