Washington Editor

WASHINGTON — Abbott Vascular (Temecula, California), held center stage for much of the morning Sept. 14 sessions at TCT 2014 thanks in part to publication of early data from the Absorb II trial matching the company's BVS bioresorbable stent against the company's flagship Xience drug-eluting stent. To no one's surprise, the one-year data left a number of questions unanswered, but principal investigator Patrick Serruys hinted that the device's performance seems to address one big question about this class of product, stating that the BVS "curvature is better than we expected."

Abbott snared a CE mark for its BVS scaffold in 2011, and resorted to calling the device a scaffold rather than a stent shortly thereafter as it ramped up production capacity. Because of the inherently lower grade of radial strength in the polylactide used to manufacture the struts compared to metal, these bioresorbable stents are thicker than their metallic counterparts, thus raising questions about ease of implant and whether these scaffolds would be sufficiently flexible to find widespread adoption. There are questions about the effects of shear associated with the device's higher strut profile as well.

Serruys, who practices cardiology at Erasmus University (Rotterdam), remarked that the BVS "should have the mechanical properties sufficient to prevent acute recoil" despite concerns about the radial strength of polylactic acid devices. He added that dissemination following the CE mark "has until now occurred without a randomized comparison with a metallic counterpart."

Absorb II randomized 501 subjects in a 2:1 ratio to the study article and the everolimus-eluting Xience, also made by Abbott, and the estimated primary completion date for the study is July 2016, with the study expected to be wrapped up in full 24 months later. Patients must have no more than two de novo lesions of differing epicardial locations and at least 50% stenosis in the target lesion[s].

Serruys said the steering committee opted to publish one-year endpoints in order to clarify some of the issues surrounding the device, explaining further, "the majority of the patients have stable angina" at enrollment, with only about 21% exhibiting unstable angina. More than 80% presented with single-artery disease, he remarked.

Of the 335 on the BVS and the 166 on the Xience, 98.2% and 98.8% were available for follow-up at a year, respectively. The study was designed to establish superiority at three years for vasomotion as assessed by mean lumen diameter before and after nitrate therapy, while non-inferiority for minimum lumen diameter (also at three years) serves as the trial's second co-primary endpoint.

Nearly 12% of the BVS patients had undergone prior intervention in the target vessel compared to slightly less than 9% of controls, although the difference did not hit statistical significance via a 95% confidence interval. The rate of prior infarcts was similar, 28% for the study article vs. 28.9% for the comparator.

Operators pre-dilated all the lesions on the BVS (364 lesions total) whereas 1.1% of the 182 lesions treated with Xience were not pre-dilated. The nominal diameter of the last balloon used was 3.08 millimeters for BVS and 3.16 for Xience, but the Xience arm saw more pressure used (15.03 atmospheres) compared to the 14.23 atmospheres used on average in the BVS arm, an outcome that generated a p value of 0.01.

Serruys' numbers indicated that post-procedure reference vessel diameter was 2.64 millimeters for the BVS and 2.80 mm for the Xience, a difference that met statistical significance. Minimum lumen diameter was also significantly larger on the Xience, (2.22 mm versus 2.50 for Xience).

Acute clinical device success was 99.2% for BVS and 100% for the Xience. Acute clinical procedural success favored Xience again (98.8% vs. 96.1% for BVS). Another factor that has sparked some noise is the three incidents of thrombosis on the study article, one each at day one, between days two and 30, and between 31 and 365 days, although there is some uncertainty about the last of those three cases.

Device makers square off over trial

Abbott did not present the results of Absorb II without an adversary waiting to pounce. Boston Scientific (Natick, Massachusetts), which as a bioresorbable stent program of its own in place, also had a lot to say, and it is apparent that the war of words has only just started.

David Rizik, the principal investigator for the Absorb II trial at one of the study's U.S. sites, told Medical Device Daily that it is difficult to infer whether there is any connection between multiple scaffold placement and the incidence of thrombosis in Absorb II. "The more overlapping stents you put in, certainly the more problems you're going to have with restenosis and stent thrombosis," he observed, a predicament of some historical documentation.

"In Absorb II, one of the thromboses was on a vessel that had a scaffold in a main branch and one in a side branch," Rizik stated, but he said the data from outside the U.S. offers little information on the subject. "It's hard to know what to make of the European data – at least my read of it – and that's because some of the reporting is incomplete. I would not go so far as to draw any conclusions" about overlapping device placement and thrombosis, he said. In Absorb II, "it would appear that one of the cases had multiple locations" of BVS placement, Rizik explained, but he said the data are not robust enough to infer a connection.

Rizik said the Xience is "a very good stent in terms of safety, but it's still unusual to have a zero rate [of thrombosis] in the control arm of any study."

"The fact that we have a 0.9% one-year rate in the Absorb arm . . . really does go a long way to put to rest the safety concerns" about the BVS, Rizik argued. He said the lower level of pressure used to inflate the BVS suggests the patients on the study article were "in my opinion systematically under-treated. And to have a less than 1% rate of thrombosis when these were systematically under-treated really does dispel those concerns," he said.

Target vessel infarcts and all infarcts seemed higher on the BVS compared to Xience, but there was a higher total rate of revascularization on Xience (7.3% compared to 3.6% on BVS). "I think this is a complex issue," Rizik observed regarding infarcts. "If you look at the per-protocol procedural myocardial infarction (MI) rate, there is no statistically significant difference" between the two arms.

Rizik said the Society for Cardiovascular Angiography and Interventions (SCAI; Washington) "has a very specific definition of procedural MI, and many of us do subscribe to that definition" rather than the definition used by the World Health Organization to categorize all target vessel infarcts and all infarcts regardless of location for the per-protocol trial population. Rizik said that the SCAI approach is a reading for creatinine kinase or creatinine kinase MB at 10 times the upper limits of normal. Both arms of the study "are exactly equal in terms of the peri-procedural MI rate," he argued, asserting that the SCAI definition is "the most clinically relevant definition."

Craig Thompson, chief medical officer at Boston Scientific (Natick Massachusetts), took a more jaundiced view of the results of the trial. "I think what we've seen today is that Absorb II reinforces that this technology is not ready for routine clinical use," Thompson told MDD, adding that he sees "a big concern with the safety signal" in reference to "numerically higher" thromboses and infarcts compared with everolimus." These numbers "tend to reinforce a variety of real-world registry data we've seen over the past several months," he said.

Thompson said there is "not a real clinical upside that I see at the moment," in part because the BVS is "a bit more cumbersome to use. We're still interested in this space," he added, but mused, "we need to be pragmatic about how to move forward and solve unmet needs in a way that's safe." He argued that the registry data for OUS usage "would suggest that in more widespread application, the thrombosis rates are higher than we would expect with the current generation of drug-eluting stents," citing everolimus-eluting devices as the standard.

When asked what he thought had triggered thrombosis in the device, Thompson said, "it's hard to know. The data were from a patient population that uniformly had IVUS," and hence procedurally the patients were handled optimally. He noted further that in OUS use, "I think there has been variability in practice, particularly with additional imaging," along with vessel preparation and post-implant dilation. Because of these variables, "it's difficult to say there is any one opportunity to improve that particular outcome."

Given his skepticism, Medical Device Daily asked Thompson why there is any interest in such devices, and he responded, "essentially aesthetics."

"It doesn't fill a clear, unmet clinical need," Thompson continued, stating, "the underlying theory is that it could reduce angina, but the data today suggest that the angina score out to a year is identical" to DES. "I think there may be some acceptance by patients about not having a permanent implant," Thompson continued, but he said of DES, "the bar for replacing this workhorse technology is quite high because their outcomes in this day and age are outstanding."

"One of the things we do not know is very long-term opportunities in terms of outcomes" for this kind of technology, Thompson mused, but he said BSX is interested "if there can be next-generation technologies that can perform to the level we expect" from current offerings "and can fully resorb." Assuming the company "can provide [the technology] in a cost-conscious way, there are opportunities to meet the needs of the market," but he said there is a need for "more investigations into which subset of patients" are best suited for bioresorbables.

However, when asked whether he thought bioresorbables would present value via a substantial clinical improvement at a higher price vs. an "aesthetic" improvement at the same price as DES, Thompson said, "it's not appropriate for me to speculate on market dynamics." //