No Very Late Scaffold Thrombosis Risk With Absorb BVS Studies in China, Japan

PARIS, France—Long-term follow-up data of two studies with highly selected patient populations provide some reassurance that there is no very late risk of scaffold thrombosis or device-oriented myocardial infarction with the Absorb bioresorbable vascular scaffold (BVS) when compared with the metallic everolimus-eluting stent (Xience, both Abbott Vascular).

The two studies—ABSORB-China and ABSORB-Japan—were presented yesterday at EuroPCR 2017 in Paris, France, and for the first time showed very low rates of scaffold thrombosis between 2 and 3 years, with just one case of definite/scaffold thrombosis in the Absorb arm (and none in the Xience-treated patients).

The late findings, along with a number of other Absorb BVS studies and registries, should help flesh out which components of the so-called PSP protocol are key for reducing the scaffold thrombosis risk that has plagued the device in 1-, 2-, and 3-year follow-up, experts said here.

Absorb in Japan and China

In ABSORB-Japan, the overall 3-year rate of definite/probable scaffold thrombosis was 3.6% in the Absorb-treated patients and 1.6% among patients who received the Xience stent.

Ken Kozuma, MD (Teikyo University Hospital, Tokyo, Japan), who presented the results, said the sole case of late scaffold thrombosis between 2 and 3 years occurred at day 810 in a patient with a high degree of coronary restenosis. The rate of target lesion failure—a composite of cardiac death, target-vessel MI, and ischemia-driven target lesion revascularization—was 1.6% in both treatment arms between 2 and 3 years.

In ABSORB-China, the 3-year rate of cardiac death, target-vessel MI, and ischemia-driven target lesion revascularization was 5.5% in the Absorb arm and 4.7% in the Xience-treated patients, a difference that was not statistically significant. The rate of target-vessel MI was 2.5% and 0.9% in the Absorb- and Xience-treated patients, respectively, while the rate of ischemia-driven target lesion revascularization was 4.2% and 2.6% in the Absorb and Xience arms, respectively. These differences were not statistically significant.

Importantly, ABSORB-China also showed a low risk of scaffold thrombosis in the extended follow-up. Among the 238 patients treated with Absorb BVS, the low rate of probable/definite scaffold thrombosis was striking at 0.9% at 3 years, with no events observed between 2 and 3 years. Comparatively, there were no cases of stent thrombosis among the Xience-treated patients.

All Aboard the Absorb Roller-Coaster Ride

Davide Capodanno, MD (University of Catania, Italy), who chaired the morning press conference where the results were presented, noted that the Absorb BVS story has been a bit of a “roller coaster.” These new positive findings from China and Japan, when interpreted in light of the negative news that has come to light this past year, are part of the development process as new technologies are put through a “stress test” to determine long-term viability, said Capodanno.

He also noted that ABSORB-China has always been outlier amongst clinical trials testing Absorb BVS.

For example, the 3-year results from ABSORB-China contrast with extended follow-up data from ABSORB II, a study recently presented at TCT 2016. In that analysis, there was a twofold increased risk of device-oriented clinical events, specifically an increased risk of target-vessel MI, as well as an increased risk of late scaffold thrombosis with Absorb when compared with Xience.

Runlin Gao, MD (Fu Wai Hospital, Beijing, China), who presented the results of ABSORB-China during a session devoted to Absorb BVS, noted their study differs from other clinical trials with the bioresorbable technology in that less than 10% of implantations were in vessels smaller than 2.25 mm.

“The ABSORB-China trial selected relatively larger vessels,” said Gao. “The mean reference vessel diameter was 2.81 mm and in other trials, such as ABSORB-Japan, ABSORB II, or ABSORB III, the reference vessel diameter was around 2.6 mm or 2.7 mm. Another thing is that only 9.2% of patients who received BVS had a vessel diameter less than 2.25 mm.” In other trials, approximately 15% of patients treated with Absorb received a device in vessels smaller than 2.25 mm, said Gao.

With the thickness of the Absorb BVS struts—157 µm—putting the device in too small a vessel would “result in some problems,” he added.

Gao said that nearly every patient received predilatation and 63% of the Absorb-treated patients underwent postdilatation following device implantation. In addition, 40% of patients were taking dual antiplatelet therapy at 2 years, and 20% were still on both therapies at 3 years.

Size Does Matter

In March 2017, the US Food and Drug Administration reiterated their concern about physicians implanting the Absorb technology in inappropriately sized vessels. The agency issued a “Dear Doctor” letter reminding physicians to closely follow package instructions regarding target vessel selection—avoiding BVS use in vessels smaller than 2.5 mm and ideally using intravascular ultrasound and optical coherence tomography to optimally size the vessel. 

Robbert de Winter, MD (Academic Medical Center, Amsterdam, the Netherlands), one of the authors of the recently published Amsterdam Investigator-Initiated Absorb Strategy All-Comers Trial (AIDA), which showed an increased risk of scaffold thrombosis and target-vessel MI with Absorb BVS, also emphasized the importance of appropriate sizing. Of the three variables often cited for obtaining good outcomes with the technology—the so-called PSP protocol (preparation, sizing, and postdilatation)—appropriate sizing may be the most imperative.

“Patient selection is a very important aspect,” he told TCTMD. “We’ve done a preliminary PSP analysis as well, and I tend to agree that sizing may be more important than postdilatation.”

Capodanno also believes sizing is critical. “We’re seeing more and more postdilatation in these trials, but it seems the real factor that affects the outcome is vessel size,” he said. “If you put such thick struts in a small vessel, you’ll occlude the vessel. So, it’s very reassuring to see that when good selection was done, like in ABSORB-China, they had no events between 2 and 3 years with both devices. That would be expected with Xience but not with Absorb—so this is very good data.”

Ron Waksman, MD (MedStar Heart and Vascular Institute, Washington, DC), also said the data are reassuring to a certain extent, but cautioned against making too much of relatively short-term clinical outcomes.

It’s kind of reassuring, but the story is not complete. Ron Waksman

“I think the real interest will be seeing what happens beyond 3 years after the scaffold is gone,” Waksman told TCTMD. “Obviously the less scaffold you have, the less scaffold thrombosis. I would expect to see fewer events between 2 and 3 years than between 1 and 2 years. So, it’s kind of reassuring, but the story is not complete. What’s going to happen between 3 and 10 years?”

Highly Selected Patient Population

Speaking to the results of ABSORB-China and ABSORB-Japan, de Winter said that if Absorb BVS is used in a wider range of patients, such as those with complex coronary artery disease including calcified or tortuous lesions, then physicians are likely to see higher rates of scaffold thrombosis and adverse clinical outcomes than observed here.

He told TCTMD that after they performed a data sweep of the AIDA trial, they had to return to patients to tell them about the risk of MI and scaffold thrombosis.

“Some of the patients were very mad,” said de Winter. “You know, we told them, there’s a device we’re testing and we think there’s great promise in having a dissolvable one, and of course that’s something that’s very appealing to patients. Now we’ve had to come back to tell them that although it’s a small risk, there is a risk involved.”

In this situation, de Winter said they give the treating cardiologists the option of putting patients back on dual antiplatelet therapy, allowing them to balance the risks and the potential benefit of treatment. “Of course, there’s no evidence that putting them back on dual antiplatelet therapy protects them from very late stent thrombosis, but we think it’s wise to at least give them that option,” he suggested.

In addition to the 3-year data from China and Japan, the EuroPCR session highlighted 12-month data from a number of European registries, including the Italian BVS-RAI and IT-DISAPPEARS registries, the Swedish SCAAR registry, the German-Austria Absorb Registry (GABI-R), the France Absorb registry, and the UK registry.

Following the presentations, Waksman, who chaired the session, polled the room to ask those in attendance if they agreed or disagreed with the recent European regulatory decision to restrict Absorb BVS to clinical trials and registries. Of those who voted, approximately 60% felt the decision is the correct one.