Increased Early Risk of Stent Thrombosis With the Absorb Bioresorbable Scaffold

For some, dissolving coronary stents represent one of the most promising  technologies in interventional cardiology, but in recent months, multiple reports examining the performance of the leading bioresorbable scaffold suggest that while the device performs as well as a permanent metallic stent clinically, it carries an increased risk of stent thrombosis.

The mounting evidence is enough to convince some researchers in this field that the rate of stent thrombosis with the new technology is real, prompting discussions as to the importance of this signal, what steps could be taken to mitigate the problem, and what the implications are for ongoing research.

At least 3 recently published studies attempt to shed light on safety outcomes with the Absorb everolimus-eluting bioresorbable vascular scaffold (Abbott Vascular), specifically the issue of stent thrombosis. The first was a meta-analysis of 6 clinical trials published online in the Lancet in November. The second was an expanded meta-analysis, which included additional registry data, published online in JACC: Cardiovascular Interventions this week. The third was real-world outcomes from 2 high-volume centers in Munich, Germany, where the device—still investigational in the United States—is already approved for use.

Aloke Finn, MD, from the CVPath Institute (Gaithersburg, MD), who wrote an editorial accompanying the Lancet publication, said the findings do raise concerns, particularly given the devastating consequences of stent thrombosis. In light of the meta-analysis, as well as the other newly published data, there “definitely appears to be a signal with the Abbott Absorb stent,” Finn told TCTMD.

The Lancet Meta-Analysis

In the Lancet paper, Salvatore Cassese, MD, of Deutsches Herzzentrum (Munich, Germany), conducted a meta-analysis of 6 clinical trials, including ABSORB China, ABSORB Japan, ABSORB II and III, EVERBIO II and TROFI II, and found that the rate of definite or probable stent thrombosis after a median of 12 months was significantly increased among patients who received the Absorb bioresorbable scaffold compared with those who received an everolimus-eluting metallic stent (OR 1.99; 95% CI 1.00-3.98). Stents included Xience V, Prime, and Expedition (Abbott Vascular).

Specifically, among 2309 patients with the bioresorbable stent, the rate of definite or probable stent thrombosis was 1.3% compared with 0.5% in the 1382 patients who received a Xience stent. There was also a trend toward more MI—5.2% with the Absorb device vs 3.5% with the everolimus-eluting stent—but the difference was not statistically significant. Rates of target lesion revascularization, target lesion failure, and all-cause mortality were all similar at one year.

The JACC: Cardiovascular Interventions report, by Michael Lipinski, MD, and Ron Waksman, MD, of the MedStar Cardiovascular Research Network (Washington, DC), included an expanded meta-analysis of trials testing the Absorb stent as well as data from other registries, such as GHOST-EU. In all, this included data on 8,351 patients treated with the Absorb scaffold and 2,159 patients receiving a metallic DES. Overall, the rate of definite/probable stent thrombosis, at 1.2%, was twice as high for the Absorb-treated patients compared with DES-treated patients (OR 2.06; 95% CI 1.07-3.98). Moreover, this analysis showed a statistically significant doubling of MI risk with the bioresorbable stent (OR 2.06; 95% CI 1.31-3.22), but no differences in revascularizations.

And lastly, the analysis of patients treated with the Absorb stent at 2 high-volume centers in Munich, Germany, published in Catheterization and Cardiovascular Interventions, showed that definite stent thrombosis at 12 months occurred in 2.6% of patients who received the bioresorbable vascular scaffold—that’s roughly twice what was seen in both of the meta-analyses.

Expert Reaction

According to investigators who conducted the German analysis, dubbed “ISAR-ABSORB,” the rate of definite stent thrombosis is “not insignificant” and needs further study. More patients with longer follow-up are required “before we can be sure of the role of these devices in clinical practice,” said senior investigator Robert Byrne, MD, and colleagues. 

“This is a new technology,” added Finn. “It’s a new area for interventional cardiology that has been explored to some extent in the past, but it’s by far the most advanced attempt making it into the clinical arena. So, it’s going to encounter unexpected issues, just the way first-generation DES, or even stents in general, encountered events that were unexpected. We’re still learning how this is all going to work, what are the vascular responses, what’s important in terms of the design of bioresorbable stents, and how do we make them better.”

Finn believes that despite concerns about the risk of stent thrombosis, he does not think the risk should preclude the technology from entering clinical practice. But he stressed that researchers and physicians need to understand more about what is causing these events.

Byrne made similar comments, but noted, at present, the bioresorbable stent simply is not beating the current gold standard of care.

“When you look at all the data comparing bioabsorbable stents with everolimus-eluting stents, which has been the control stent in most of these studies, bioabsorbable stents aren’t better in any of the parameters,” he told TCTMD. “You look at the angiographic parameters, late lumen loss, target lesion failure, or stent thrombosis, it always seems to be favoring the conventional metallic stent with no signal of a benefit for the bioabsorbable stent. This probably dampens people’s enthusiasm to some extent or gives them pause for thought.” 

Based on the registry data, the rates of stent thrombosis observed with the Absorb stent are higher than what interventional cardiologists have come to expect from the drug-eluting, metallic stents typically used in clinical practice, said Byrne. His group recently completed an analysis of more than 150 randomized trials with DES showing the rate of stent thrombosis to be approximately 0.5% at 12 months. In a paper published in the European Heart Journal in 2015, Byrne pointed to data from clinical registries and randomized trials with broad inclusion criteria showing the rate of stent thrombosis is less than 1.0% in the first 12 months with DES and approximately 0.2% to 0.4% per year beyond then.

Byrne, who was also a co-author on the Lancet study, said meta-analyses are best interpreted as a “summary of treatment effects.” In that sense, the treatment effect of the Absorb stent when compared with the everolimus-eluting metallic stent suggests a doubled risk of stent thrombosis, a finding that is consistent in the 2 reports. Waksman made a similar comment, noting that there are always calls for randomized controlled trials to get to the bottom of a clinical question, but the stent thrombosis data with Absorb are pretty consistent across the registries and studies.

Early Days With Bioresorbable Scaffolds

The promise behind the new bioresorbable technology has been to shift coronary interventions away from a permanent stent to one where the vascular device would deliver the antirestenotic drug and then disappear. To combat some of the late vascular consequences associated with the permanent metallic cage, such as impaired coronary vasomotion or the development of neoatherosclerosis, a bioresorbable stent could provide transient arterial support until drug elution was complete and then degrade over time (Absorb requires approximately 3 years for complete degradation). This would allow the vessel to regain vasoreactivity and minimize the risk of late stent thrombosis, say the experts.

In Europe, 2 bioresorbable stents have received CE Mark approval—the Absorb stent and the novolimus-eluting DESolve stent (Elixir Medical, Sunnydale, CA)—but Absorb is further ahead in clinical testing and is the most widely used technology. Neither stent is approved by the US Food and Drug Administration.

To TCTMD, Finn said he likens the early reports of stent thrombosis with the bioresorbable stents to the early days with DES. He noted that the rates of restenosis and thrombosis with second- and third-generation stents with a permanent polymer have declined significantly when compared with the earliest devices.

Gregg W. Stone, MD, Columbia University Medical Center (New York, NY), said the stent thrombosis signal was observed in the registry data, although it has declined over time, and was also seen again in the randomized trials, including ABSORB III, for which Stone was the study chairman. “On an absolute level, it’s a relatively small increase, but it does seem to be there,” Stone said in an interview. Still, the hazard needs to be put in perspective, particularly given the learning curve with this technology, he argued. “Physicians were using this device for the first time in the setting of these studies. This scaffold, this first-generation scaffold, requires different techniques than standard stents, which I don’t think was appreciated at the time as much as it is now.”

The Absorb stent, with its balloon-expandable bioresorbable poly-L-lactide scaffold, has relatively thick struts (150 µm) to increase radial strength. Finn said that lesion preparation in patients receiving the technology is “demanding.” There are limits on balloon expansion and postdilatation because, being nonmetallic, the stent can fracture more easily than conventional devices.

“We can’t extrapolate what we know from metallic stents and use that for implanting bioresorbable stents,” he said. “It’s a very different animal. There certainly seems to be some issue surrounding the implantation and the early rates of stent thrombosis. Whether that’s due to stent fracture, malapposition, incomplete expansion—we can’t answer that right now.”

Timing of Stent Thrombosis

To TCTMD, Waksman said that stent thrombosis, or “scaffold thrombosis” in the case of the bioresorbable vascular device, is happening relatively early, meaning it is likely a mechanical issue rather than one of delayed healing.

In the Lancet meta-analysis, patients with bioresorbable vascular scaffolds had a significant time-dependent risk of early stent thrombosis compared with patients who received the metallic stent. Overall, the risk of definite/probable stent thrombosis was tripled in the first 30 days after implantation (OR 3.11; 95% CI 1.24-7.82). In the meta-analysis by Lipinski and Waksman, the risk of acute and subacute stent thrombosis was 0.27% and 0.57%, respectively.

Thus, it may be that the increased risk of thrombosis might be the result of inadequately matching the scaffold to the vessel or being placed in small vessels, Waksman said. For example, in ABSORB III, the overall rate of stent thrombosis was 1.5% for patients treated with the Absorb stent and 0.7% in patients treated with the everolimus-eluting metallic stent, a nonsignificant difference. However, if vessels with a reference diameter less than 2.25 mm were excluded, the rates of device thrombosis were 0.9% in the Absorb arm and 0.6% in the Xience arm.

“There are some issues related to understanding the technology,” said Waksman. “Imaging during the procedure can be helpful, as can sizing to match the device to the vessel. Smaller vessels should probably be avoided right now.”

Deliverability and Predilatation

Byrne said the bioresorbable stent is thought to be tougher to deliver, especially with the thicker stent struts, but he personally has been surprised by its ease of use. For their group, deliverability is not a sticking point. Like Stone, he said high-pressure lesion preparation is key, as is the liberal use of adjunctive technologies such as scoring balloons, rotablation, and IVUS. “Our experience is that if you don’t get the predilatation right, it doesn’t matter what you do with postdilatation,” said Byrne.

Stone, for his part, suspects attention to such procedural detail was not followed routinely in the early days with the technology, or not fully appreciated, and this likely contributes to the higher rate of stent thrombosis. “Overall, we’ve seen this small signal, but I think it’s explainable, and I think it can be eliminated with good technique and the selection of appropriate vessels,” he said.

As for what happens next, Finn said while there are presently concerns about the early risks of stent thrombosis, long-term outcomes are lacking. He pointed out that complete degradation of the polymer takes as long as 32 months in animal models and inflammation is observed for 1 year. The thicker stent struts means the vessel heals less quickly leaving some unanswered questions about the optimal use of dual antiplatelet therapy.

Rates of stent thrombosis observed early might be offset by a significant reduction in the long-term risks associated with a permanent metallic stent, although that remains unknown at present. The other possibility is that future iterations of the device, coupled with improved physician education on how to use the newer device, will reduce the early hazard of stent thrombosis.

“This is the promise of the technology,” said Byrne. “With metallic stents, we still have a small issue of late-occurring safety events.”

Overall, Byrne suspects that in the next 3 to 5 years, physicians might treat 10% to 15% of coronary lesions with the bioresorbable stent, although he notes others think the percentage of candidates for the stent could be larger. The penetration in Germany is presently around 5%, but as operators become more accustomed to the “wrinkles” of the bioresorbable stent technology and the technology improves, use will increase, he predicted.

Stone noted that a second-generation bioresorbable scaffold from Abbott is about to start testing in humans later this year. This stent has considerably thinner stent struts than the current device. For now, the bioresorbable stent could be a good choice for patients with large vessels or noncalcified lesions. Young patients, those with another 20, 30, or 40 years of living left would also benefit from not having a permanent stent left in place, he said. He also thinks that ACS patients could benefit from the stent given that these lesions are softer and a metallic DES does not heal very well in this setting.

Waksman said that if the stent is approved in the United States—it is expected to go before an FDA advisory panel this year—it would need to be carefully used in selected patients and selected lesions.

“There is an issue,” he said. “If you’re not careful, you pay the price. There is a tax. Stent thrombosis is a serious event.” Despite these concerns, he believes the bioresorbable stent in an important advance, one that should continue to move forward in testing and development. While Absorb is not a “regular” stent and cannot be used as such, he said, the risks can be mitigated.

---

Sources: 
1. Cassese S, Byrne RA, Ndrepepa G, et al. Everolimus-eluting bioresorbable vascular scaffolds versus everolimus-eluting metallic stents: a meta-analysis of randomized controlled trials. Lancet. 2015;Epub ahead of print.
2. Lipinski MJ, Escarcega RO, Baker NC, et al. Scaffold thrombosis after percutaneous coronary intervention with ABSORB bioresorbable vascular scaffold. J Am Coll Cardiol Intv. 2016;9:12-24.
3. Hoppmann P, Kufner D, Cassese S, et al. Angiographic and clinical outcomes of patients treated with everolimus-eluting bioresorbable stents in routine clinical practice: results of the ISAR-ABSORB registry. Catheter Cardiovasc Intv. 2015;Epub ahead of print.
4. Finn AV, Virmani R. The clinical challenge of disappearing stents [editorial]. Lancet. 2015;Epub ahead of print.
5. Byrne RA, Joner M, Kastrati A. Stent thrombosis and restenosis: what have we learned and where are we going? Eur Heart J. 2015; 36: 3320-31.

 

Related Stories:

• ABSORB III: Novel Bioresorbable Scaffold as Good as Standard-of-Care DES
• ABSORB Trials Pit BRS Against Second-Generation DES
• ABSORB II: Similar Clinical Outcomes, Less Angina with Bioresorbable Scaffold vs EES