J&J's Cypher Beats Medtronic Endeavor Stent In SORT-OUT III: Does It Matter?

Medtronic directly disputed the validity of the Cypher results, but at least one interventional cardiologist publicly urged a halt in use of Endeavor based on the data. Several other physicians at the ACC meeting, though, said the study should not have much of an impact on clinical practice in the context of all available data.

Johnson & Johnson's Cypher sirolimus-eluting stent decidedly outperformed Medtronic's Endeavor zotarolimus-eluting stent in 18-month, head-to-head data reported last week at the American College of Cardiology annual meeting.

The significance of the outcome, however, is in the eye of the beholder.

Medtronic directly disputed the validity of the Cypher results. But at least one interventional cardiologist publicly urged a halt in use of Endeavor based on the data.

Several other physicians at the meeting, though, said the study should not have much of an impact on clinical practice in the context of all available data.

And, industry analysts who focus on the stent business did not pay the data much mind. The significance of both Cypher and Endeavor in the market has fallen precipitously with the increasing dominance of Abbott's Xience /Boston Scientific's Promus and the expectation for release of more next-generation products.

Doubled Event Rate For Endeavor

Researchers from five hospitals in Denmark reported 18-month results from the 2,332-patient "all-comer" SORT-OUT III trial March 15 at ACC, and in The Lancet.

Use of Endeavor resulted in about double the rate of the primary composite endpoint of cardiac death, heart attack or target vessel revascularization compared to Cypher (9.7% versus 4.5%). The results were similar to nine-month outcomes reported at the 2008 Transcatheter Cardiovascular Therapeutics conference, which led to some reduction in Endeavor market share (Also see "SORT-OUT III Trial Shows More Stent Thrombosis With Endeavor Than Cypher" - Medtech Insight, 20 Oct, 2008.).

The main differences between the nine- and 18-month results is that Cypher no longer has a statistical advantage in stent thrombosis (two cases in Cypher versus none from Endeavor in the final eight months), and that J&J's stent now does have a statistical advantage in all-cause mortality.

"If you have to compare the two stents, the Cypher stent is a better stent," concluded Michael Maeng, an interventional cardiologist at Aarhus University Hospital in Denmark, who presented the data at ACC. Maeng says Aarhus no longer uses Endeavor.

Despite the nine-month data, this was not an expected result, the SORT-OUT III researchers said. Other data suggest Endeavor exhibits a more stable long-term performance than Cypher. And Endeavor has matched or bettered Cypher's safety performance in previous randomized studies.

Maeng pointed out, however, that the different results likely reflect the different design of SORT-OUT III, which, as an "all-comer" trial, enrolled a broader range of patients with more complex lesions compared to more restricted studies. Also, he stressed that the trial is powered to assess clinical, as opposed to only angiographic, endpoints.

"If you want to assess the clinically relevant differences between the various drug-eluting stents, you have to compare the stents in routine clinical care patients," he said.

The design of SORT-OUT III is unusual. The researchers looked at outcomes in patients who received stents at the five high-volume angioplasty centers in Denmark, with relatively light exclusion criteria. As with all patients in the Danish systems, the subjects were registered in national registries, which the researchers used to track outcomes.

Unlike most stent studies, SORT-OUT III did not include any additional monitoring or angiograms for the sake of data collection. Only information logged into the registries as part of standard patient care was included.

Cardiologists Disagree On Interpretation

"This is a study that we have been waiting for," said Harvey White, an interventional cardiologist and director of coronary care and cardiovascular research at Green Lane Cardiovascular Service, Auckland, Australia. He responded during the ACC "late-breaking" trial session for SORT-OUT III.

"The trial had fewer exclusions, and we should be able to apply this to our patients," he said. "Should the [Endeavor] stent be used? My answer is no."

But others were not as won over.

"I think for the clinical audience it is important to recognize that these results are different from other trials that have been presented of similar size," said Laura Mauri, an interventionalist at Brigham and Women's Hospital, Boston, and chief scientific officer of the Harvard Clinical Research Institute, during the same session.

"We have to take into account the totality of evidence before thinking about changing practice," she said.

Mauri also pointed out that although the coronary artery lesions in the Danish study are more complex than FDA pivotal studies that included Cypher and Endeavor, the complexity is not as high as you would expect in a truly real-world population.

The Lancet paper specifies that 5,676 patients were eligible for the study, but that 3,344 patients were ultimately not enrolled, "mainly because the all-comer principle was not used by all investigators."

In an interview, David Kandzari, director of interventional cardiology research at Scripps Clinic, La Jolla, Calif., pointed out that the lack of individual patient follow-up by the researchers led to some gaps, including no measure of events such as heart attacks that occurred during hospitalization for the procedure. Endeavor has demonstrated some advantage in a number of procedural events in other studies compared to Cypher and Taxus, he said.

Kandzari was chief medical officer for Cypher-manufacturer Johnson & Johnson/Cordis from 2006 to 2008. He is currently the primary investigator for Medtronic's Endeavor III and IV studies.

Kandzari reported five-year outcomes at ACC from the 436-patient ENDEAVOR III trial, which compared Endeavor and Cypher. In results reported in 2006, Endeavor did not meet its non-inferiority endpoint to Cypher for late lumen loss at eight months. But, over a longer period, Cypher has displayed a faster rate of growth of events such as late stent thrombosis, cardiac death and myocardial infarction, and target lesion revascularization.

The five-year analysis, reported March 14, concluded that emerging advantages for Endeavor on cardiac death/MI and major adverse cardiovascular events are "achieving statistical significance."

Another study reported at the meeting, ISAR-TEST-2, a 1,007-patient randomized comparison of Cypher, Endeavor and a next-generation dual-drug, polymer-free stent, found overall higher target lesion revascularization rates from Endeavor at two years, but more growth in TLR between year one and two with Cypher.

Investigator Robert Byrne, Technische Universitat, Munich, referred to this as "evidence of a slight decrement in anti-restenotic efficacy with Cypher."

Medtronic: Trial "Biased" Against Endeavor

Medtronic, concerned with relinquishing some of its 20% share of the worldwide coronary stent market, immediately came out challenging the validity of SORT-OUT III, referring to its "unconventional design and analytical methods" as "biased" against Endeavor in favor of Cypher.

Among the firm's criticisms are the exclusion of 3,344 eligible patients and the variation in adherence to anti-platelet drugs. Medtronic says the study reported "unexpectedly low event rates" for Cypher compared to other trials, while Endeavor rates were generally consistent with prior studies.

Kandzari agrees with some of Medtronic's critiques, but was skeptical of its charge of "biased" results. "Whatever limitation there is in the study, it should exist equally for Cypher and Endeavor," he said.

Ultimately, both SORT-OUT III and ENDEAVOR III reflect "very favorable results in both stents," Kandzari said, and, especially with the nine-month SORT-OUT data already considered, he does not expect practice to change now.

The study is a reminder, he said, that so much stent data is available that no study on its own should substantially change behaviors. "It is probably more appropriate for us to look to themes of consistency for a drug-eluting stent across different trial designs and different populations studied."

Another broader message from SORT-OUT III, he explained, is that data for currently available drug-eluting stents is very good and that it will be difficult for new technologies, such as bioresorbable stents, to statistically prove they are better.

"In interventional cardiology, the psychology has been ... there is a new product always around the corner and that new is better. I think trials like SORT-OUT III remind us that this is not always the case."

-David Filmore ( [email protected] )

[Editor's note: This story appears courtesy of the editorial staff of 'The Gray Sheet,' your source for coverage of devices and diagnostics. For a free trial, call customer service at 800-332-2181.