Gaithersburg, MD - In a vote of nine to one, the FDA Circulatory System Devices Panel has recommended approval of Abbott Vascular's Xience V everolimus-eluting stent, surprising skeptics who had predicted there were not enough long-term data to support a yes vote for the device. After 10 hours of presentations, discussion, and occasional disagreement, the panel had only two suggested conditions for approval, pertaining to dual antiplatelet therapy labeling and a postmarketing study.

Dr John C Somberg (Rush University Medical Center, Lake Bluff, IL) the sole no vote, had initially proposed a third condition: that any recommendation for approval be pending completion of two-year follow-up in the remaining patients enrolled in the pivotal SPIRIT III trial. None of the other panelists, however, backed him up.

"I thought the safety data from 12 to 24 months was inadequate, and it was a bad precedent to establish," Somberg said to explain his no vote. "The pivotal study not having been fully evaluated was of concern to me, given the recent tumult over late stent thrombosis, which may or may not be an issue. To have inadequate data leaves this issue really unaddressed for many years."

Two-year follow-up data from the SPIRIT trial program is available for only 422 patients who have received the Xience V stent and for only 26 patients have been followed for three years.

Many of the remaining nine panel members said that they could justify their yes votes because SPIRIT II and III had met their primary end points for efficacy and because the safety record for the stent, at least out to one year, was acceptable. Panel members, for the most part, seemed to agree with Dr Mitchell Krucoff (Duke University, Durham, NC), who presented data earlier in the day on behalf of the sponsor, when he suggested that, considering all of the preclinical and clinical data so far, the Xience V stent was "unlikely to throw us a time bomb."

"We have very little in the way of statistical certainty; that doesn't mean it isn't reassuring," Krucoff said.

Indeed, panel member Dr John Hirshfeld (University of Pennsylvania, Philadelphia) said he thought Xience V "looks like it will prove to be a very nice adjunct to our armamentarium" but added that he will be paying "very close attention to ongoing studies and follow-up data."

Those studies include the randomized SPIRIT IV study, which started in August 2007 and has already enrolled roughly 2000 of the proposed 3690 subjects, with a primary end point of ischemia-driven MACE at 270 days. The sponsor, Abbott, has also proposed a 5000-patient registry study, dubbed XIENCE V USA, which will have stent thrombosis as its primary end point.

In discussing the conditions for approval, the panel debated the merits of insisting on a concurrent control group for this postmarketing registry; a vote on this condition failed to pass by a narrow margin. In the end, the panel agreed, almost unanimously, that one condition for approval be the requirement for a postmarketing study, "the details of which are to be determined at a later date by the FDA."

The panel also agreed unanimously on the condition that the device label contain the same recommendations for dual antiplatelet therapy duration as the other drug-eluting stents already on the US market, as set out by the AHA/ACC.

Despite the lack of long-term clinical data, panel members appeared to be impressed by preclinical work presented by the sponsor and by the repeated assertions that the Xience V's design represents a step forward for drug-eluting stents (DES). The XIENCE V platform is Abbott's Multilink Vision, a low-profile, thin-strut, cobalt-chromium stent.

During today's hearing, Dr Gregg Stone (Columbia University, New York, NY), also presenting on behalf of the sponsor, pointed out that in the Cypher-vs-Taxus trial (REALITY), late loss was significantly lower for the Cypher, but rates of binary restenosis and target lesion revascularization (TLR) were similar between the two FDA-approved stents. But Stone speculated that the late-loss differences between the Xience V and the Taxus in SPIRIT III (0.14 mm vs 0.28 mm at nine months) might actually be clinically meaningful. "Here, possibly as a result of thinner stent struts, better endothelialization, greater polymer integrity, and perhaps less stent-strut fracture, we're not only seeing less late loss, but in two consecutive trials we've actually seen reduced binary restenosis and now reduction in TLR. I would never directly compare the Cypher and the Xience V—those trials have not been done—but I think you can speculate they might be similar."

Stone also characterized the Xience V as a "true next-generation product . . . a true medical advance."

The panel seemed to agree: "I believe reasonable assurance of safety was demonstrated," Dr Richard L Page (University of Washington, Seattle) commented after delivering his yes vote. "I think this represents a step forward for interventional cardiologists and our patients."

Today's panel decision was something of a déjà vu for those who keep a close watch on regulatory proceedings: in October, a panel made up of many of the same members recommended approval of Medtronic's Endeavor stent with similar conditions. Analysts predict both the Endeavor and the Xience V will join the Cypher and Taxus in the US marketplace sometime in 2008. In fact, when Abbott's Xience V stent is approved, an identical stent, named Promus, made by Abbott but distributed on a private-label basis by Boston Scientific, will also be marketable in the US. The unique arrangement stems from Boston Scientific's 2006 acquisition of Guidant, the original developer of the Xience V device. As part of the deal, Abbott purchased Guidant's stent business to assuage antitrust concerns, but Boston Scientific negotiated the right to sell its own version of the stent, although the company will, for the time being, pay a royalty of 40% to Abbott for every Promus stent it sells.