Peggy Peck

March 07, 2005

March 7, 2005 (Orlando) -- Opening day of the American College of Cardiology 2005 Annual Scientific Sessions was dominated by news about drug-eluting stents, and at the end of the day, "it was clearly a winning day for Cypher," said Valentin Fuster, MD, PhD, a spokesperson for the American Heart Association.

Dr. Fuster said that three separate head-to-head trials of Cypher, a sirolimus-eluting stent made by Cordis, a Johnson & Johnson Company, and Taxus, the paclitaxel-eluting stent made by Boston Scientific, "all tended to favor Cypher." Dr. Fuster is director of both the Zena and Michael A. Wiener Cardiovascular Institute and the Marie-Josée and Henry R. Kravis Center for Cardiovascular Health at Mount Sinai Medical Center in New York City.

Moreover -- and this was a finding that Dr. Fuster characterized as "most significant" -- one of the three studies presented the first report of safety concerns with the paclitaxel stent. "The thrombus findings are, I would say, the most unexpected finding," Dr. Fuster said in a telephone interview with Medscape. Dr. Fuster was not involved in the study.

Dr. Fuster was referring to data from the REALITY study, which was a head-to-head, multicenter, international, randomized study of the sirolimus- and paclitaxel-eluting stents. Marie-Claude Morice, MD, from the Institut Cardiovasculaire Paris Sud, France, who presented the data for the REALITY investigators, said the eight-month stent thrombosis rate (acute and subacute) in the actually treated patients was 0.4% in the Cypher group and 1.8% in the Taxus group. "This was significant at P = .0196, and this is the first study to report this safety concern for Taxus," said Dr. Morice.

While Cordis immediately issued a press release highlighting the Taxus thrombosis data, Eberhard Grube, MD, from the Heart Center in Siegberg, Germany, a coinvestigator of the study, questioned the finding during his discussion of the paper. (Gregg Stone, MD, from Columbia University in New York City, was originally slated to be the discussant for the REALITY trial, but Dr. Grube stepped in at the last minute.)

Dr. Grube noted that when considering the intent-to-treat analysis, the difference in stent thrombosis was not statistically significant (0.6 vs 1.6; P = .0723). It was only when the data were analyzed by actually treated patients is it significant. "Is there a real difference here, or is this a non-end point?" he asked.

Despite the unexpected finding on safety, the REALITY study, which was sponsored by Cordis, failed to meet its primary end point of in-lesion binary restenosis at eight months.

That, however, was not the case for SIRTAX, which was the last of the comparison studies reported.

This study, by a team of researchers from the Swiss Cardiovascular Center in Bern and Inselspital University Hospital Bern and Zurich, delivered the strongest evidence to date that Cypher is superior to Taxus.

"Cypher is superior. And this study took all comers, it did not select out the easiest patients," said Bernhard Meier, MD, professor and chairman of cardiology at Inselspital University Hospital and one of the study authors. He told Medscape that the results from SIRTAX -- which was funded by the university and received no commercial support -- "are likely to be more important because this is what we see in real practice."

The study randomized 1,012 patients to either Cypher or Taxus. Unlike other stent studies that limited the number of blocked arteries or the length of the blockage, the Swiss study had no limit on the number of lesions, the number of vessels, or the length of the lesions. In addition, about one third of patients had acute coronary syndromes including non-ST elevation myocardial infarction (MI) and ST-elevation MI.

There were 503 patients were randomized to the Cypher stent and 509 to the Taxus stent. The researchers reported on nine months of follow-up in 500 Cypher patients and 505 Taxus patients. The primary end point was cardiac death, MI, or ischemia-driven target lesion revascularization at nine months.

Among the findings from SIRTAX:

  • Target lesion revascularization rate was 4.8% in the Cypher group vs 8.3% in the Taxus group ( P = .025)

  • Target vessel revascularization rate was 6% for Cypher vs 9.2% for Taxus ( P = .009).

  • Target vessel failure: cardiac death, MI, or target vessel revascularization rate was 7% for Cypher vs 11.6% for Taxus ( P = .012).

  • Late lumen loss was 0.13 ± 0.37 in the Cypher group vs 0.25 ± 0.49 in the Taxus group ( P < .001).

When the results were analyzed by sex and presence of diabetes or acute coronary syndrome, all findings favored Cypher, said Stephan Windecker, MD, who presented the results during a late-breaking clinical trials session. He added that unlike the REALITY study, there was no evidence of increased acute and subacute thrombus associated with the Taxus stent.

Dr. Morice, who presented the REALITY data, served as discussant for the SIRTAX study. She said the findings affirm that while "one stent [Taxus] is excellent, the second one [Cypher] is a little better."

Dr. Meier said that based on the study results, "We are using the Cypher stent, not Taxus."

REALITY, which was the largest of the comparison studies, was presented earlier in the day at the first late-breaking clinical trials session. The trial enrolled 1,353 patients with two or fewer de novo native coronary lesions. The lesions were at least 2.25 mm in diameter, but no more than 3 mm in diameter. There were 684 patients with 970 lesions randomized to the Cypher stent and 669 patients with 941 lesions randomized to the Taxus stent. The patients had both angiographic and clinical follow-up at eight months after the procedure.

The primary end point was in-lesion binary stenosis rate by quantitative angiography at eight months. The investigators assumed the eight-month in-lesion restenosis rate for Cypher would be 8.0% and 14.0% for Taxus.

The success rate was "excellent for both stents, 99.4% for each," said Dr. Morice.

While the trial did not meet its primary end point, of a 43% reduction in angiograpic binary restenosis, in-stent late loss was significantly less for Cypher than for Taxus (0.09 vs 0.31; P < .001).

There were no significant differences in clinical events at eight months, although there was "a suggestion of a trend in Q-wave MI with one in the Cypher arm and six in the Taxus arm ( P = .067)."

The ISAR DIABETES study, a smaller study that randomized 250 patients with diabetes to Cypher or Taxus, was presented by Adnan Kastrati, MD, from Deutsches Herszzentrum in Germany.

Diabetic patients are considered high risk for restenosis and other major cardiac events after undergoing percutaneous procedures, said Dr. Kastrati. In this study the primary end point was difference in late lumen loss, and here Cypher again fared better than Taxus.

Asked if the findings will alter his clinical practice, Dr. Kastrati told Medscape, "These data push us to select Cypher in the diabetic patient." He cautioned, however, that the trial included a small patient population and that larger prospective studies are needed to conclusively distinguish the benefits of one stent from another in these high-risk patients.

John Hirshfeld, MD, from the University of Pennsylvania in Philadelphia, said that "statistically speaking, you can give the diabetic patients greater protection with the Cypher stent. But these data need to be put in context." Dr. Hirshfeld, the moderator of the ACC press conference at which the findings were discussed, was not involved in the study.

The ISAR-DIABETES study was funded by the institution with no commercial support.

ACC 2005 Annual Scientific Session: ACCIS O5 Spotlight 22, ACCIS 05 Spotlight 25. Presented March 6, 2005.

Reviewed by Gary D. Vogin, MD