TCT: New Definition and Data Make Drug-Eluting Coronary Stents Seem Safer

WASHINGTON, Oct. 24 -- The maker of the Cypher (sirolimus-eluting) coronary stent said today that a new independent analysis has given it an even playing field compared with bare-metal devices.

The analysis of data from four randomized trials demonstrated that there was no excess mortality, myocardial infarction, or thrombosis associated with the Cypher drug-eluting device compared with bare metal stents, said Cordis, the Johnson & Johnson subsidiary that markets Cypher.

That analysis used a new definition of stent thrombosis that moved bare metal devices from a late-thrombosis rate of 0.6% to a rate of 3.3%, according to findings reported by Donald Cutlip, M.D., of the Harvard Clinical Research Institute.

Dr. Cutlip, who also co-chaired the Academic Research Consortium (ARC) group that drafted the new definition during a series of meetings convened after recent studies raised questions about drug-eluting stent safety, presented the Cypher data at the Transcatheter Cardiovascular Therapeutics meeting here.

When possible thrombosis -- defined by ARC as an unexplained death that occurs more than 30 days after procedure -- was factored in, the four-year rate was 3.5% for Cypher and 3.4% for bare metal, Dr. Cutlip said.

In other words, the new definition made bare metal stents more thrombosis-prone, but left the thrombosis risk essentially unchanged for Cypher.

After four years, 1.6% of Cypher patients and 1.7% of the bare-metal stent patients had a definite or probable thrombosis as defined using the new ARC definition.

A spokesperson for Cordis said the FDA asked drug-eluting stent makers to use this new definition when they present data at an FDA meeting on stent safety in early December.

The new analysis, along with additional data reported at an industry analyst's meeting today by Brian G. Firth M.D., Ph.D., worldwide vice president of medical affairs and health economics at Cordis, contradict results from two independent meta-analysis reported in Barcelona at the European Society of Cardiology/World Cardiology Congress -- the so-called ESC firestorm.

But the FDA seemed skeptical when the new analysis was presented today during a TCT hotline session titled, "The ESC Firestorm," The FDA's Bran Zuckerman said that while these new data can provide some reassurance to concerned physicians and patients, the agency "continues to dynamically follow [drug-eluting stent safety] and we do there think there is a signal there."

He added that the "exact nature [of that signal] and the best way to frame the risk for an individual patient remains unclear." Those issues, he said, will be addressed at the safety panel meeting in December.

In Barcelona researchers analyzed published data and concluded that there was a 2.4% increased of death or MI among patients who received Cypher stents versus patients with bare metal stents (P=0.03). A second analysis reported an increased risk of non-cardiac mortality at three years versus bare metal stents. Both meta-analyses found no significant increase for Taxus, a paclitaxel-eluting stent made by Boston Scientific, but the company later released data indicating that late-stent thrombosis was also a risk with Taxus.

Dr. Cutlip's pooled analysis was based on four-year data from RAVEL, SIRIUS, E-SIRIUS, and C-SIRIUS, which are the four major Cypher randomized controlled trials of Cypher that were sponsored by Cordis. The trials included 878 patients who received Cypher stents and 870, who were treated with bare metal stents.

The way it worked was that under the old or traditional definition -- which was the definition used by Cordis when it designed its clinical trials of Cypher -- late stent thrombosis was defined as stent thrombosis that occurred more than 30 days post-procedure in a target vessel that had not undergone revascularization after the procedure.

If stent thrombosis caused an MI, which resulted in a revascularization with percutaneous coronary intervention or bypass surgery, that thrombosis was not listed as a late thrombosis because it occurred in a vessel that was "not free from revascularization."

The ARC said that was a flaw in the definition that prevented an accurate assessment of stent thrombosis risk. Moreover, the flaw benefited bare metal stents that have much higher target vessel revascularization rate than drug-eluting stents.

The new definition also raises questions about conventional wisdom concerning bare metal versus drug eluting stents, which has held that thrombosis is a risk for the first 30 days after a bare metal stent is placed, but not a factor thereafter because the stent has completely re-endothelialized and thus clots are unlikely to form.

Dr. Cutlip said that while thrombosis risk with Cypher increases with time, there was also late risk with bare metal stents.

That "wisdom" drives current antiplatelet recommendations, which call for 30 days of dual antiplatelet therapy -- aspirin plus Plavix (clopidogrel) -- for bare metal stents, versus a minimum of three months for Taxus and six months for Cypher. Moreover, in the absence of excess bleeding risk, American College of Cardiology/American Heart Association guidelines recommend that all drug eluting stent patients can benefit from 12 months of dual antiplatelet therapy.

Within the first 30 days there the probable and definite thrombosis rate was 0.5% for Cypher and 0.3% for bare metal stents, at one year the rate was 0.6% for Cypher and 1.3% for bare metal stents, he said.