CoreValve Trial Shows Transcatheter Aortic Valve Replacement Better than Surgery in High-Risk Patients

At the joint ACC/JACC Late-breaking Clinical Trials presentation at the American College of Cardiology 2014 Scientific Sessions, held in Washington DC on March 29, 2014, David Adams, MD, professor and chair of cardiothoracic surgery at Mount Sinai Hospital, presented findings on behalf of the US CoreValve investigators. Speaking to an overflow audience in the main exhibit hall, Adams presented trial outcomes demonstrating superiority of a self-expanding transcatheter aortic valve replacement (TAVR) over surgical aortic valve replacement (SAVR) in patients with severe aortic valve stenosis who were deemed high-risk for surgery. The multi-center CoreValve study follows a study now in press that examined use of the self-expanding TAVR inpatients deemed to be at extreme high risk for surgery.

Noting that previous studies have shown that TAVR with a balloon expandable valve improved survival compared to medical therapy and had similar survival compared to surgery in patients who were inoperable or deemed at high risk for surgery, Adams outlined the aims of the CoreValve study, which targeted high-risk surgical candidates and employed a self-expanding transcatheter prosthesis.

Participant inclusion criteria included risk of death at 30 days after surgery of >/= 15% and <50% risk of death or significant complications within 30 days after surgery. Participants were excluded if they had had a recent active GI bleed, stroke, or myocardial infarction, as well as any recent stenting, renal failure, LVEF of less than 20%, or comorbidities which would result in a life expectancy of less than one year.

With the primary endpoint being all-cause mortality at one year, the study set a 7.5% noninferiority margin of TAVR with the CoreValve prosthesis over SAVR, with a subsequent test for superiority at the one-sided 0.05 level to follow if non-inferiority testing was also met at that level. Secondary endpoints were set hierarchically, and included change from baseline to the one-year point of the mean flow gradient, effective orifice area, NYHA heart failure classification, and frailty and quality of life as measured by the KCCQ and SF-12, all of which save the SF-12 were evaluated for non-inferiority. Additionally, the secondary endpoint of major adverse cardiovascular and cerebrovascular events (MACCE) was evaluated for superiority.

The TAVR arm enrolled 394 patients, with 390 eventually being treated. Of those, 323 received the prosthesis via an iliofemoral route. Four hundred and one patients were enrolled in the SAVR arm. Of those, 44 exited the study before the procedure, yielding an as-treated group of 357 patients for the SAVR arm. Mean age for both arms was just over 83 years, with 52% of the SAVR group and 53% of the TAVR group being male.

More than 85% of patients in both arms were NYHA class III or IV. Other comorbidities and assessments of frailty and disability did not differ significantly, with the exception of incidence of diabetes mellitus (34.9% TAVR, 45.4% SAVR, p<0.01).

Study compliance was very high, with 98.5% of TAVR-arm patients following through to one year, and 93.6% of the SAVR patients achieving one-year follow-up. Study authors reported excellent results for both treatment arms.

The primary endpoint of one-year all-cause mortality was 14.2% for TAVR, compared with 19.1% in the SAVR group (p=0.04 for superiority). For this endpoint, noninferiority (p<0.001) and superiority thresholds were met in both the as-treated and intention-to-treat cohorts. Low numbers in both treatment arms suffered stroke (8.8% TAVR vs. 12.6% SAVR) and major stroke (5.8% TAVR vs. 7.0% SAVR).

The secondary endpoint of one-year MACCE was 20.4% for the TAVR arm vs. 27.3% for the SAVR arm (p=0.03). All other secondary endpoints showed statistically significant noninferiority of TAVR over SAVR. Limitations noted by the study authors include a larger number of patients withdrawing from the surgical arm (though Adams noted that treated and withdrawn patients did not have importance differences); also, since the 30-day mortality rate was lower than anticipated, the study population may actually not have been as high-risk as the study design intended.

In discussion after the presentation, an expert panel lauded the authors for taking care to create a study design that captured measures of frailty and quality of life, as well as for carefully tracking stroke and other cerebrovascular events. Several panelists questioned Adams closely about differences seen between the CoreValve study and the PARTNERS study, which studied another transcatheter aortic valve prosthesis. These two devices are compared in the CHOICE trial, to be presented Sunday, March 30.