This article is more than 5 years old. Information may no longer be current.
ABSORB III 2-year results suggest optimal techniques may be key
WASHINGTON — At 2 years, the cumulative rate of target lesion failure was higher with a bioresorbable vascular scaffold compared with an everolimus-eluting stent, but the difference was smaller in patients with appropriately sized vessels in the ABSORB III trial.
Longer-term data will help determine whether better patient selection and technique improves overall outcomes with the bioresorbable vascular scaffold (Absorb BVS, Abbott Vascular) and in comparison with an everolimus-eluting stent (EES; Xience, Abbott Vascular), Stephen G. Ellis, MD, Director of Interventional Cardiology in the Robert and Suzanne Tomsich Department of Cardiovascular Medicine in the Sydell and Arnold Miller Family Heart and Vascular Institute, Cleveland Clinic, said here.
The 2-year rate of TLF was 11% with the BVS vs. 7.9% with the EES (HR = 1.42; 95% CI, 1.04-1.94; P = .03). In patients meeting inclusion criteria for the study (with reference vessel diameter >2.25 mm by quantitative coronary angiography), 2-year TLF — a composite of cardiac death, target vessel MI and ischemia-driven target lesion revascularization — was 9.4% vs. 7%, respectively (HR = 1.35; 95% CI, 0.93-1.96; P = .11). The overall rate of definite/probable stent thrombosis was 1.9% with the BVS vs. 0.8% with the EES at 2 years, which was slightly reduced among patients with appropriate sized vessels (1.3% vs. 0.6%). The scaffold thrombosis rate between 1 and 2 years was 0.3% (number needed to harm = 317).
On Saturday, in advance of the late-breaking clinical trial presentation, the FDA issued a letter to health care providers stating that the agency is investigating the increased 2-year rate of MACE observed with the Absorb BVS vs. the Xience EES in ABSORB III.
“The FDA is working with Abbott Vascular to conduct additional analyses to better understand the cause(s) of the higher cardiac event rate and device thrombosis rates in patients treated with the BVS compared to the Xience stent,” according to the letter.
Until further data are available, the FDA recommends that health care providers “follow the instructions for target heart vessel selection (eg, avoiding BVS use in small heart vessels) and optimal device implantation that are included in the BVS physician labeling” and “advise BVS patients to follow the recommendations for DAPT prescribed by their health care providers.”
ABSORB III was a prospective, multicenter, single-blind trial that randomly assigned 2,008 patients (mean age, 63 years; 70% men) in a 2:1 fashion to the BVS (n = 1,322) or EES (n = 686). At 2 years, follow-up data were available for 1,296 and 671 patients, respectively. Baseline patient and lesion characteristics were similar between the groups. One-third had diabetes and one-quarter had unstable angina. Mean lesion length was 12.6 mm in the BVS group and 13.1 mm in the EES group. Mean reference vessel diameter was 2.6 mm in both groups.
The study protocol was revised and the landmark TLF endpoint was revised from 1 to 5 years to between 3 and 7 years, or up to 10 years, because “superiority of Absorb BVS is not likely to emerge before the bioresorption process is complete (approximately 3 years), consistent with emerging reports of very late events between 1 to 3 years with BVS in small studies,” Ellis said during a presentation.
One-year results from ABSORB III demonstrated noninferiority of the BVS to EES for the primary endpoint of TLF. The new results show that noninferiority was retained between years 1 and 2, but risk for TLF was higher at the end of 2 years.
Ellis and colleagues also analyzed use of “PSP” — pre-dilatation, appropriate vessel sizing and high-pressure post-dilatation — in the cohorts and found that results improved in the BVS cohort with PSP to 8.7% for TLF and 1.1% for stent thrombosis.
A preliminary analysis of pooled rates with BVS and EES in the ABSORB IV study showed stent thrombosis rates of 0.4% at 30 days (n = 2,397) and 0.5% at 1 year (n = 1,415). The anticipated enrollment of ABSORB IV is 3,000 patients, according to a company press release.
Ellis noted several limitations of the study, one being that the BVS is a first-generation device and was used for the first time by most operators within the trial and the optimal implantation technique was still evolving during the initiation and enrollment of ABSORB III. Additionally, “results should be viewed in the context that Xience was the control device, which has been associated with low rates of stent thrombosis and TLF,” he said.
During the discussion portion of the late-breaking clinical trial presentation, Antonio Colombo, MD, FACC, FESC, FSCAI, said, “the bright side is that there are new scaffolds on the horizon [with] thinner struts [and] much more friendly delivery. The field should stay on with all the limitations that we see. But we can overcome the limitations and I see this data in a positive fashion.”
Reference:
Ellis SG, et al. Late-Breaking Clinical Trials: Interventional.
Disclosure: The study was funded by Abbott Vascular. Ellis reports receiving consultant fees and research grants from Abbott Vascular, Boston Scientific and Medtronic.