 |
| Abbott's Absorb bioresorbable stent--Courtesy of Abbott |
Abbott's ($ABT) Absorb bioresorbable stent took a step closer to FDA approval, winning the overwhelming backing of an FDA advisory panel of experts, who voted 9-0 (with one abstention) to recommend approval of the device based on an analysis of its risks and rewards.
The backing of what is on track to become the first FDA-approved bioresorbable stent came despite an agreement that the benefits over the comparator device are unproven. There were also concerns about the use of the Absorb in patients with small vessels, namely those less than 2.25 millimeters in diameter.
The panel also voted 9-1 that the device is safe and 10-0 that it is effective.
Abbott met its noninferiority endpoint against the company's Xience metallic drug-eluting stent based on one-year data from the 2008-person Absorb III trial. The benefits over permanent drug-eluting stents are not experienced until the device is resorbed by the body, a milestone that occurs after about two years.
Abbott said 5-year superiority data will be presented in 2020, from its 5-year Absorb IV trial. It presented 5-year data from its small, first-in-man study of 101 patients to preliminarily demonstrate that the promised clinical benefits like restored vasomotion and late lumen enlargement do occur in long-term patients. A definite benefit is the ability of clinicians to conduct reinterventions; permanent metallic stents can become physical barriers to certain procedures such as the insertion of bypass grafts.
Dr. Warren Laskey of the University of New Mexico School of Medicine said that he voted against the device's safety because several components of the composite endpoint went in the wrong direction, saying the Absorb's underperformance against the Xience on several safety metrics (though not by enough to prove statistical inferiority) didn't sit well with him.
For example, the 1,313 patients who received the Absorb had a 1.54% rate of definite or probable stent thrombosis, compared to 0.74% for 677 who randomly received the Xience. The difference was not statistically significant, but the Absorb also unperformed Xience on cardiac death rate and target vessel myocardial infarction, at least numerically.
Another area of concern coming into the meeting was the 94.3% device success rate, referring to the successful delivery and deployment of the Absorb at the intended target lesion. Abbott said the relatively low rate was a result of a "low threshold" for crossing over to the Xience in cases where the operator experienced any difficulties with implantation.
Use of the Absorb in small vessels
The improper deployment of the Absorb in patients with small vessel sizes dominated the discussion. Almost 20% (or 375) Absorb patients in the pivotal trial had a reference vessel diameter of less than 2.25 millimeters, despite the fact that the Abbott has proposed indicating the device for those vessels that are at least 2.5 millimeter in diameter.
The discrepancy reflects the challenges of accurately estimating the metric.
Those Absorb patients with vessels smaller than 2.25 millimeters had a one-year rate of stent thrombosis of 4.6%, compared to 1.5% for the comparator device, Abbott's permanent drug-eluting Xience. The difference was not statistically significant, but Dr. Gregg Stone, of the Columbia University Medical Center, said it is mechanistically and biologically plausible that the one-year event rate is higher among Absorb patients with small vessels, in part because the device's struts are thicker than those of the Xience.
Abbott has proposed labeling that recommends quantitative coronary angiography or intravascular imaging in patients with visually assessed vessel diameters estimated to be less than 2.75 millimeters. That's because visual assessment routinely results in an overestimate of vessel size, Abbott and the FDA say.
The panelists came to a consensus that the labeling should be changed to recommend imaging in vessels assessed to be smaller than 3 millimeters.
All patients in the trial received angiograms on a post hoc basis at the angiographic core lab. Based on that data, Abbott said that about half of patients with vessels visually estimated to be smaller than 2.75 millimeters actually had vessels smaller than 2.25 millimeters (based on the imaging data), while among those whose vessels were estimated to be smaller than 3 millimeters, 26% fell into the very small vessel group.
Abbott said that preliminary data from the Absorb IV trial indicates that due to operator education and stricter adherence to the treatment protocol, only 5.2% of the Absorb patients in that trial have vessel sizes smaller than 2.25 millimeters. The results of the trial remain blinded.
It appears that the panelists were convinced by Abbott's argument that most of the underperformance against the Xience can be attributed to improper usage of the device in small vessels, and that the problem can be ameliorated through training and use of imaging.
That should be a relief to the company. The Absorb had previously been met with a fair share of skepticism from the medical community, as evidenced by a critical editorial in the New England Journal of Medicine, which said that the clinical relevance of the finding of statistical noninferiority to the Xience is open to question.
The overwhelming panel vote, and lack of skeptical undertones from the FDA during the meeting or in its meeting documents, makes approval quite likely.
Absorb achieved a CE mark in 2010 and is used widely abroad, with India and Brazil accounting for slightly more than half its 2014 revenue of $132 million, according to Global Data, which projects U.S. revenues of $96 million by 2017, assuming approval. If approved, it will compete most directly with Boston Scientific's ($BSX) Synergy partially bioresorbable stent.
- read the March 15 meeting's literature
Special Report: Leaders emerge in the race for the first U.S. bioresorbable stent - Abbott's Absorb BVS leads the pack