The Impact Of ABSORB II: Down The Slippery Slope

When it was approved earlier this year in July the Absorb (Abbott) bioresorbable stent provoked a fair amount of discussion about the “slippery slope of noninferiority.” Although the stent met the FDA criteria for approval, the available trial results went in the wrong direction and there was a particularly concerning warning sign of a higher rate of stent thrombosis.

Absorb represents a remarkable technical achievement, propping open a clogged artery and then gradually, at least theoretically, disappearing and leaving a healthy artery wall intact.

Now, with a presentation at the Transcatheter Cardiovascular Therapeutics meeting and a simultaneous publication in the Lancet, 3-year results from the ABSORB II trial are helping to confirm the worst fears raised at the time of approval.

ABSORB II, which is the longest randomized comparison to date, failed to show any benefit for the device. The trial did not meet its coprimary endpoints—superior vasomotor reactivity and noninferior late luminal loss— which were designed to highlight the main theoretical advantage of the bioresorbable stent. Even more troubling, although the trial was not powered for clinical endpoints, there was a significantly higher rate of target vessel MI.

The ABSORB II findings put a spotlight on the FDA’s increasing reliance on post-approval studies. In recent years, many cardiac devices— including the Watchman, Impella, CardioMems, Mitraclip, and Amplatzer—have been approved with little or no data demonstrating efficacy. Instead the FDA has required the companies to perform extensive post-approval studies.

I asked a number of cardiologists how the ABSORB II findings would likely affect the field and whether the FDA should do anything different in the future in response to the trial.

Sripal Bangalore, MD (NYU-Langone Medical Center), said that, if ABSORB II had been the pivotal trial, the FDA would not have approved the device. “However, the trial was not the pivotal trial and was not powered for hard endpoints.” His main takeaway is that, “as acknowledged by the authors, the trial emphasizes that operators must be very meticulous in implanting these devices. A scaffold thickness of 150 microns will be non-forgiving if it is not deployed meticulously.”

Sanjay Kaul, MD (Cedars-Sinai Medical Center) pointed out that the FDA had previously reviewed 2-year data from ABSORB II that showed that differences in target lesion revascularization, stent thrombosis, or target lesion failure worsened from 1 to 2 years. “The only thing new at 3 year follow-up is vasomotor reactivity and IVUS data.” Kaul agreed with Bangalore that the FDA would be unlikely to change its decision and will almost certainly wait until Abbott completes a series of post-approval studies. All bets are off, however, if the studies are not reassuring. “It remains to be seen what the FDA will do if the post-approval studies are uniformly negative.”

As to the larger implications of the story, Kaul said that the findings call into question the FDA’s increased reliance on post-approval studies in the absence of compelling efficacy data. “The results of ABSORB II support the argument that the FDA should have asked for 2 or 3 year data prior to approval.”

David Kandzari, MD (Piedmont Heart Institute) also thinks that the FDA is unlikely to do anything right away, “other than await longitudinal follow-up of ongoing ABSORB III and IV trials.” Kandzari said that he thought that “these very late events in ABSORB II are unsettling, not just because of the promise that the resorbable scaffolding holds but also because the reason for these events are uncertain.” But, he added, “for now, I don’t think there’s enough insight to state how this would influence the approval process.”

Robert Harrington, MD (Stanford University) questioned the use of the noninferiority design for the initial approval of a first-generation device. “I think of noninferiority as needing to have other ‘benefits.’ This might include, safety, ease of administration or delivery, or price. Otherwise, why would you give up something?” Harrington said he was “surprised” but not “shocked” by the approval of Absorb. The new data “raise questions about this first-generation technology.” We need more studies and better next generation devices, he said. For now, though, he sees little reason for using Absorb. “With these data, I don’t see a role until the next generation device.”

Sanket Dhruva, MD (Yale University) said that to get more information about the troubling clinical findings in ABSORB II, the FDA “should push for an interim analysis or expedite the next planned analysis” of the ABSORB III study, which is an ongoing larger study powered for clinical endpoints. He also said that “the FDA should also consider placing a black box warning that the device may have a higher rate of myocardial infarction compared to drug-eluting stents, at least until longer-term data are available from ABSORB III.”

Dhruva also suggested that “for any patient receiving a coronary angiogram with the possibility of PCI where the interventional cardiologist is considering placing a BVS [bioresorbable vascular scaffold], the FDA should mandate a discussion about the pros and cons of BVS versus a drug-eluting stent or bare-metal stent and ensure patients are informed of the clinical results from ABSORB II. This could be formalized in a checklist that each patient must sign prior to any angiogram where PCI may follow, ensuring that a fully-informed patient consents to receiving the BVS.”

Dhruva said he hoped that a long-term effect of the case will be that the FDA will “demand longer follow-up of high-risk permanently implanted devices to ensure benefit versus standard of care on clinically meaningful endpoints before approval.”

Dhruva also agreed with Kaul that since the FDA knew about the 2-year ABSORB II results “it would have been prudent for the FDA to wait an additional 4 months for longer-term data” given that “all the clinically-meaningful endpoints suggested that Absorb BVS was no more effective – and very possibly less safe — than standard drug-eluting stents.”


  1. jason weinstock says

    the real question is given the data, would you or these Drs want this implanted. and the idea that putting these in inferior stents in unsuspecting patients to further demonstrate their inferiority is deeply disturbing.

  2. Hello everyone
    I received the ABSORB Stint in 2016 and now it has failed.
    I didn’t know I was getting it till after the surgery. I am a disabled vet. It is in a main artery, and now going through more test to see how much damage is done. Any and all help on this subject would be helpful. I fell like I took one step forward and three steps backward. After the stint I developed AFIB. I feel I was let down by the system and the doctors that I put my trust in.

Speak Your Mind