PROTECT: Lower Stent Thrombosis Risk with Endeavor Emerges at 4 Years

Download this article's Factoid (PDF & PPT for Gold Subscribers)


At 4 years, patients with sirolimus-eluting stents (SES) have a higher risk of definite or probable stent thrombosis and death compared with those who received zotarolimus-eluting stents (ZES), according to results of the PROTECT trial published online August 8, 2014, ahead of print in the European Heart Journal. The differences were not apparent earlier in the trial, emerging only between the third and fourth years. 

The data were originally presented at the Transcatheter Cardiovascular Therapeutics scientific symposium in October 2013.

Edoardo Camenzind, MD, of University of Geneva (Geneva, Switzerland), and colleagues say their findings illustrate “the importance of appropriately powered, long-term large studies to assess safety and efficacy of two typologically different drug-eluting stents, particularly as hard outcomes (death, MI, or stent thrombosis) are relatively rare.”

For the current analysis, the investigators looked beyond the main 3-year results of PROTECT (Patient Related OuTcomes with Endeavor versus Cypher stenting Trial) to examine outcomes at 4 years. The all-comers trial randomized 8,709 CAD patients between May 2007 and December 2008 at 196 international sites to Endeavor ZES (Medtronic; Santa Rosa, CA; n = 4,357) or Cypher SES (Cordis/Johnson & Johnson, Warren, NJ; n = 4,352). Baseline characteristics were similar between the arms, with about 26% of patients presenting with MI and 19% with unstable angina.
Use of dual antiplatelet therapy was similar between groups at all follow-up intervals (6 months and years 1-4), with a mean prevalence of 27.5% at 4 years.


More Stent Thrombosis with Cypher

Academic Research Consortium-defined definite or probable stent thrombosis (primary endpoint) was higher with Cypher at 4 years, driven by more very late (>1 year) events. The combined rate of large MI and death also was higher with Cypher, as was death and any nonfatal MI (table 1).

Table 1. Outcomes at 4 Years


(n = 4,357)

(n = 4,352)

P Value

Definite/Probable Stent Thrombosis




Very Late Stent Thrombosis



< .001

Death/Large MI




Death/Nonfatal MI





Of note, the Cypher group had a greater increase in definite or probable stent thrombosis from year 3 to year 4 (from 1.8% to 2.6%) than the Endeavor group (from 1.4% to 1.6%; log rank P = .003).

While clinically driven TLR was lower with Cypher compared with Endeavor (4.5% vs 5.9%; P = .002), clinically driven TVR was not (P = .37). Combined safety and efficacy endpoints, including TLF, TVF, and MACE, did not differ between groups. Additionally, rates of definite or probable stent thrombosis with Endeavor declined progressively over time and were consistent across 12 prespecified subgroups.


The 4-year findings, the study authors note, stand in contrast to the first year of follow-up, when Endeavor showed a higher incidence of TLR and TVR and a numerically higher rate of definite or probable stent thrombosis.

“Had the evaluation of outcomes been done after 1 year, the conclusions would have been that [Cypher] was more efficacious and safer [than Endeavor],” they write. “Likewise, differences in definite or probable stent thrombosis and the main secondary endpoints were not apparent at 3 years, when the primary trial outcomes was assessed, but emerged between year 3 and year 4 for all the various stent thrombosis definitions and consistently for the various composites of death and MI.” 

Response to the 2006 DES ‘Firestorm’?

In 2006, Dr. Camenzind was at the center of a ‘firestorm’ after presenting data at the European Society of Cardiology Congress suggesting a higher ongoing rate of stent thrombosis, death, and MI with DES compared with BMS. At the same meeting, a meta-analysis of randomized trials found an increased risk for overall mortality through 4 years in patients treated with DES vs BMS. Amid this controversy, PROTECT was designed to explore the incidence of stent thrombosis using 2 commercially available DES with different efficacy levels and healing characteristics, the study authors write.

According to Dr. Camenzind and colleagues, “long-term follow-up of clinical safety is critical for patient safety given that these devices are permanently implanted.” Importantly, they say, extended follow-up provides the ability to demonstrate “that differences in the pharmaco-polymeric properties of 2 drug-eluting stents translate into differences in clinical adverse events that prior smaller randomized trials did not discern.”

Current Relevance Questionable

But in a telephone interview with TCTMD, Sorin J. Brener, MD, of Weill Cornell Medical Center (New York, NY), was more pragmatic, noting that since Cypher is no longer on the market and Endeavor is much less commonly used now than Resolute, the findings are of interest more from a historical perspective.

“Some of this information just isn’t as clinically relevant anymore. For patients who have Cypher stents, we know that there is a small, fixed, [and] apparently unending risk of stent thrombosis, which appears to be unrelated to dual antiplatelet therapy. It’s an excess of about .2% to about .5%, but it’s there and we can’t change that,” he said.

“What’s interesting is that even though stent thrombosis continues to occur in a very small proportion of patients with Cypher stents, the impact is less and less,” Dr. Brener continued, noting that at 6 months mortality is 20%-30%, but by year 5 or 7 it is much lower for reasons that remain unclear.

Wijns W, Steg PG, Mauri L, et al. Endeavour zotarolimus-eluting stent reduces stent thrombosis and improves clinical outcomes compared with Cypher sirolimus-eluting stent: 4 year results of the PROTECT randomized trial. Eur Heart J. 2014;Epub ahead of print.  


  • The PROTECT trial was funded by Medtronic. 
  • Drs. Camenzind and Brener report no relevant conflicts of interest. 

Related Stories: