MUNICH, Germany—After 3 years, no difference exists between stent thrombosis rates for a first-generation sirolimus-eluting stent and a second-generation zotarolimus-eluting stent. Findings form PROTECT, the largest randomized head-to-head trial of drug-eluting stents (DES) were reported August 27, 2012, at the European Society of Cardiology Congress and simultaneously published online ahead of print in the Lancet.

According to presenter William Wijns, MD, PhD, of Cardiovascular Center Aalst (Aalst, Belgium), the Patient Related OuTcomes with Endeavor versus Cypher stenting Trial (PROTECT) is also the first powered to compare differences in stent thrombosis risk between stents.

For the all-comers trial, investigators randomized 8,800 CAD patients at 196 sites around the world to the Endeavor zotarolimus-eluting stent (Medtronic Santa Rosa, CA; n = 4,400) or the Cypher sirolimus-eluting stent (Cordis/Johnson & Johnson, Warren, NJ; n = 4,400). Baseline characteristics were similar between the arms, with about 26% of patients presenting with MI, 28% having diabetes, and 58% exhibiting complex disease.

DAPT Use Similar in Both Arms

Use of dual antiplatelet therapy (DAPT) was 96% at discharge and 88% at 1 year, after which use fell sharply to 37% at 2 years and 30% at 3 years. However, the proportion of patients on DAPT did not differ significantly between the stent arms at any time point.

At 3 years, the incidence of Academic Research Consortium-defined definite or probable stent thrombosis, the primary endpoint, was equivalent between the Endeavor and Cypher arms, as were rates of death and large MI and cardiac death and large MI. On the other hand, rates of TLR and MACCE (composite of death, MI, TLR, emergent CABG, and stroke) were lower with Cypher (table 1).

Table 1. Clinical Outcomes at 3 Years

An exploratory analysis showed that though there was no difference in stent thrombosis between the DES during the first year after the index procedure (P = 0.06), between the first and third years the Cypher group experienced more very late events than the Endeavor group (1.1% vs. 0.3%; P < 0.001).

This suggests that a difference in stent thrombosis between the DES is emerging over time, Dr. Wijns said, and in fact 5-year follow-up is under way to determine if the rates continue to diverge and translate into variations in clinical safety.

Interestingly, the investigators found a low incidence of definite stent thrombosis in the Cypher arm at 1 year when DAPT use was high, but a more than threefold increase in the annual rate when DAPT use decreased to less than 40%. By contrast, in the Endeavor arm, the incidence of definite very late stent thrombosis decreased twofold after 1 year despite decreasing DAPT use. According to Dr. Wijns, this trend is consistent with the hypothesis that DAPT use may not have the same long-term clinical relevance for all stent types.

Why PROTECT Is Still Relevant 

Asked in a press conference about the relevance of the study to clinical practice today in light of Cypher’s exit from the market and declining use of Endeavor, Dr. Wijns replied that “millions of patients have been implanted with these devices, so from a patient perspective, data like these on long-term safety are essential.

“From a physician’s perspective,” he continued, “the hypothesis subtending the trial was a comparison between 2 devices with very different antiproliferative profiles: Cypher nearly annihilates neointimal proliferation but with excess risk of late stent thrombosis vs. bare-metal stents, while the vascular healing response of Endeavor is more like that of bare-metal stents with no excess risk of stent thrombosis. So from a physician’s viewpoint, this study shows that the trade-off between efficacy and safety is different when comparing first-generation DES with bare-metal stents vs. 2 different DES.”

Dr. Wijns concluded that “large pragmatic trials such as PROTECT are essential to determine differential outcomes between DES and to capture the impact of improved procedural and clinical practice on stent thrombosis rates over time.”

Commenting on the presentation, Stephan Windecker, MD, of Bern University Hospital (Bern, Switzerland), lauded PROTECT for several strengths, including its size and global representation, evaluation of safety rather than efficacy, and long-term follow-up.

Considering Why the Trial Was Negative

However, because the trial was negative, it is important to determine if it was conclusive, Dr. Windecker said. And in that regard the 4% loss to follow-up, if skewed toward one device rather than the other, could affect the findings, especially when viewed against the background that PROTECT was powered under the assumption that the 3-year stent thrombosis rate would be 2.5% for Cypher and 1.5% for Endeavor.

Dr. Windecker offered several possible reasons why the trial might have failed to find the expected superiority of Endeavor in regard to stent thrombosis, including comparatively lower enrollment of STEMI patients compared with other trials, practice changes that might have weeded out higher risk patients, and improvements in technique. Moreover, the endpoint of definite or probable stent thrombosis is more sensitive but less specific than definite stent thrombosis, resulting in dilution of a device-specific treatment effect, he commented.

Although PROTECT failed to show a significant difference between stent types for the primary endpoint, findings for secondary outcomes, namely definite and very late definite stent thrombosis, confirm previous concerns of delayed arterial healing with the early-generation Cypher, Dr. Windecker said.

Less Import for New-Generation Devices

Dr. Windecker noted that PROTECT was initiated against the backdrop of the so-called “ESC firestorm” of 2006, which was touched off when a pair of meta-analyses presented at the annual congress showed higher rates of death and MI for first-generation DES compared with BMS in long-term follow-up, attributing the events to stent thrombosis.

But stent thrombosis is no longer an issue with new-generation DES, Dr. Windecker observed, and previous safety concerns have been resolved without compromising efficacy. The cardiology community recognized the safety issue before regulators did, he said, and device manufacturers solved the problem by developing new technology. “Self-regulation of the health care profession to protect patients is intact,” he concluded.

2. Camenzind E, Wijns W, Mauri L, et al. Stent thrombosis and major clinical events at 3 years after zotarolimus-eluting or sirolimus-eluting coronary stent implantation: A randomized, multicenter, open-label, controlled trial. Lancet. 2012;Epub ahead of print.

Related Stories:

• RESOLUTE: Zotarolimus-Eluting Stent Fares Well at 2 Years
• Endeavor Imparts Durable Efficacy, Even Among High-Risk Patients
• ZES Match or Exceed Performance of Earlier DES