麻豆传媒

This article is a collaboration between 麻豆传媒 and:

SAN DIEGO -- The last of the new trials to report on the Absorb bioresorbable vascular scaffold (BVS) stent added certainty to the decision to halt sales of the first-generation device.

The ABSORB IV trial, which expanded on prior trials by including patients with biomarker-positive acute coronary syndromes, again showed the numerically higher risk with the Absorb GT1 stent that had been seen in the pivotal ABSORB III trial, and led to an FDA cautionary statement, after which Abbott moved to end sales in September 2017.

The primary outcome of target lesion failure -- encompassing cardiac death, target vessel MI, and ischemia-driven target lesion revascularization -- at 30 was more common but met non-inferiority against the conventional metallic Xience stent (5.0% vs 3.7%, P=0.0244 for noninferiority), with optimized implantation technique.

In the COMPARE-ABSORB trial presented at the same session with a high-risk patient or complex-lesion population, the target lesion failure rate was a relative 24% higher at 5.1% at 1 year versus 4.2% with Xience, although it did meet noninferiority criteria.

The major driver in the difference was target-vessel MI (4.0% vs 2.1%), largely in the first week, reported Pieter Smits, MD, PhD, of Maasstad Hospital in Rotterdam, the Netherlands.

ABSORB IV was designed with careful blinding and adjudication of angina to address a signal seen in an early study, noted Gregg Stone, MD, of New York-Presbyterian Hospital/Columbia University Medical Center in New York City, who presented the findings at the Transcatheter Cardiovascular Therapeutics (TCT) conference.

But the 1,296-patient randomized trial showed angina at 1 year similarly common with Absorb as Xience (20.3% vs 20.5%, P=0.0008 for non-inferiority).

"This is real angina," Stone said, highlighting that two-thirds of these cases recurred "within about 4 to 6 weeks, way before the restenosis period, suggesting either incomplete revascularization or microvascular disease or spasm or something else."

Cardiac death rates are low; angina is why most stable patients are stented, Stone noted. "We're going to need to figure this better if we're going to keep improving quality of life for PCI patients."

Overall, though, despite efforts to optimize lesion and patient selection and implantation technique, "the themes are consistent; the excess thrombosis is consistent with every single study been done," commented David Cohen, MD, of Saint Luke's Mid America Heart Institute in Kansas City, Missouri. "We've learned lessons but they still don't overcome whatever issues there are with the technology."

A second-generation Absorb device is still under development, and the panel at a TCT press conference for the late-breaking clinical trials indicated that hope persists.

This particular stent has no commercial future, but "there are a lot of lessons that will be useful for future stent development," commented Deepak Bhatt, MD, of Brigham and Women's Hospital in Boston. "Ultimately, I do think there will be a bioabsorbable stent in clinical use, and all of these results will be useful in those development programs."

For example, Reva and Amaranth have devices in development with, respectively, better expansion characteristics and a thinner strut on par with conventional metallic stents, Stone noted. "The scaffolds are getting better and better and hopefully that will eliminate most of these differences."

But "it's much more complicated than we imagined," Cohen acknowledged. While there are plans to continue follow-up of first-generation Absorb recipients out to 7 to 10 years, "the benefit is all promise. It's hard to sell promise. It's hard as a clinician to do something for a promise that, so far, we have absolutely no evidence is realized."

Bhatt said the lesson should be that it was a mistake to approve the stent in the first place without 5-year data. "I think as a field we've got to learn that, yeah, things might look exciting, but don't approve them until there's good safety data and preferably efficacy data."