Bioresorbable Scaffolds: Lingering Challenges for a Device that Goes Away

Drug-eluting bioresorbable vascular scaffolds proved to be noninferior to best-in-class drug-eluting stents at one year for target lesion failure in a randomized clinical trial. The data were sufficient to earn the support of a U.S. FDA advisory panel, but will “as good” be good enough to win over cardiologists and payers?  

Design tradeoffs

The multicenter, randomized controlled clinical trial ABSORB III was designed to evaluate the safety and effectiveness of a device that potentially offers the benefits of a stent without its permanence. Sponsored by Abbott Vascular, the trial enrolled 2,008 patients at 193 sites with stable or unstable angina and relatively simple coronary lesions to be treated with either Abbott’s everolimus-eluting polymer scaffold (Absorb) or its everolimus-eluting cobalt-chromium stent (Xience). One-year results, unveiled in late 2015, met the bar set by the FDA for noninferiority for the primary endpoint.

[[{"fid":"22927","view_mode":"media_original","type":"media","attributes":{"height":512,"width":600,"style":"width: 180px; height: 154px; margin: 5px; float: left;","alt":" - d-kereiakes","class":"media-element file-media-original"}}]]

The one-year rate of target lesion failure, defined as the composite of cardiac death, target vessel myocardial infarction (MI) or ischemia-driven target lesion revascularization, totaled 7.8 percent for Absorb and 6.1 percent for Xience (N Engl J Med 2015;373[20]:1905-1915). The 1.7 percentage point difference was within the prespecified noninferiority margin. Xience also edged out Absorb on death from any cause, any MI, any revascularization and definite or probable device thrombosis. The two devices more or less tied for patient-reported angina.

“The fact that it is numerically different by a nonsignificant degree is not surprising,” says Dean J. Kereiakes, MD, medical director at the Christ Hospital Heart and Vascular Center in Cincinnati and a co-principal investigator in ABSORB III. At least at the outset, first-generation bioresorbable scaffolds may be handicapped compared with the latest-generation drug-eluting stents. To provide radial strength equivalent to a metal stent, the polymer-based scaffold requires relatively thicker struts—150 microns vs. less than 100 microns for Xience—plus another 7 microns for the coating that elutes everolimus. “I thought a wider, thicker strut would be associated with greater incidence of periprocedural MI. It wasn’t.”

The two devices virtually tied for MI during a procedure. But the one-year rate of device thrombosis was twice as high with Absorb, at 1.5 percent vs. 0.7 percent with Xience.

“Bioresorbable scaffolds, with devices now being first generation, are somewhat thicker strut devices compared with current contemporary drug-eluting stents and for that reason they are prone to a slightly higher rate of stent thrombosis unless they are implanted really well,” says Gregg Stone, MD, of Columbia University Medical Center in New York City and ABSORB III study chair. “I do believe you can overcome most of any increased risk, but it does require a little bit more attention to operator technique.”

[[{"fid":"22926","view_mode":"media_original","type":"media","attributes":{"height":242,"width":577,"alt":" - absorb-procedure","class":"media-element file-media-original"}}]]

Lessons & learning curves

The one-year results gave researchers insights about the device’s learning curve and other physician factors. The average operator in the trial deployed two scaffolds in the trial, Stone says, and the device’s deployment technique and profile differ from metallic stents. “You can be a moderately high-volume operator and be surprised when you put the device in by its profile,” Kereiakes says. “It is not a big deal for those of us who have tried all types of devices, basically test-pilot devices, but it is very different from putting the Xience stent in.”

Kereiakes, Stone and co-principal investigator Stephen G. Ellis, MD, of the Cleveland Clinic, reviewed every case of scaffold thrombosis and identified some common themes, many related to deviations from the study protocol. For instance, 20 percent of the lesions in patients enrolled in ABSORB III were below the 2.5 mm lower limit for the reference vessel diameter, according to Stone.

“That is really where the stent thromboses clustered, which is not surprising given that when you have a very thick strut device and you put it in a very small vessel, you magnify any of these space-occupying concerns with the technology,” Stone says. “If you look at the larger vessels, the ones for which the device was intended, then the stent thrombosis rates and the overall outcomes were very similar between Absorb and Xience.”

The study recommended that operators use high-pressure postdilation with a noncompliant balloon, with the goal of 10 percent or less residual stenosis. Instead, in the cases of scaffold thrombosis, they often found undersizing of the scaffold and suboptimal postdilation. “Believe or not, they used a balloon smaller than the scaffold; they used a compliant balloon,” Kereiakes observes. “Or if they used a noncompliant balloon, they didn’t even go into 12 atmospheres, which is nominal pressure. …Most if not all of the cases with scaffold thrombosis had over 20 to 30 percent residual stenosis within the scaffold segment.”

ABSORB III will follow patients through five years to determine if the scaffolds can offer the near-term benefits of drug-eluting stents and avoid the long-term disadvantages of a permanent implant. Scaffolds’ opportunity to outshine metallic stents is expected to emerge between one and three years of implant as the scaffold dissolves and eventually disappears. The lack of a metal cage eliminates the risk of stent fracture and may reduce the risk of thrombosis and restenosis. It also may allow the vessel to regain its natural shape and vascular function, among other advantages.

“What is so important is we have the capacity for adaptive remodeling,” Kereiakes says. “When you have a caged or restrained vessel with a metal scaffold, there is only one way the artery can go. It has to go inward.” 

But the scaffolds may not completely snuff the risk of very late events. Although rare, cases of very late scaffold thrombosis have been reported in Europe, where Absorb received CE mark in 2011 (J Am Coll Cardiol 2015;66:1901-1914). Contributing factors may include underexpansion during implantation, passing devices such as catheters through the scaffolds or occurrences during the scaffolds’ dismantling process. “It is theoretically possible that if you don’t have tissue constraining the scaffold, which usually does occur to some degree after three to six months, possibly you could have larger pieces of the scaffold that abut in the lumen,” Stone explains.

Taking cost into account

If bioresorbable scaffolds receive FDA approval, they still will face the challenge of being accepted in a healthcare system focused on cost containment. Establishing noninferiority at one year may not be enough to convince Medicare to pick up the tab for a device that likely will cost more than stents. Based on historical trends with stents, though, that might not hinder its uptake.

Elise Berliner, PhD, director of the technology assessment program at the Agency for Healthcare Research and Quality’s Center for Outcomes and Evidence, studied the early use of bare-metal stents under a Medicare policy that in 1994 folded stenting into the diagnosis-related group (DRG) code for angioplasty. Nonetheless, between 1995 and 1996 stenting procedures in Medicare beneficiaries doubled.

“If you performed the procedure with a balloon and didn’t use a stent at all, or if you used a stent, you’d get paid the same,” she says. “The hospitals decided to use the stents anyway.”

Stenting received a separate DRG in fiscal year 1998 with reimbursemeant of $2,120 per procedure. Drug-eluting stents, on the other hand, were considered a breakthrough technology and the Centers for Medicare & Medicaid Services (CMS) immediately assigned them a separate DRG in 2004 with a payment of $1,500 above a bare-metal stent. Today if a technology isn’t covered, companies can apply for pass-through payment status to counteract higher costs until CMS recalibrates its DRGs.  

Acceptance and reimbursement ultimately may hinge on data from ABSORB IV. The five-year trial will enroll an additional 3,000 patients and will be powered to demonstrate superiority. It also will include a cost analysis to determine if the benefits offset the price of the device.

“If the device going away materially influences device-related events, we should see that,” says Kereiakes, who also is an ABSORB IV co-principal investigator. “The curves should diverge. ABSORB III didn’t have a snowball’s chance in you-know-where of showing superiority for that endpoint because the device is still present and [enrollment] wasn’t large enough.”