COMPARE and SPIRIT IV: "Durable, Progressive Relative Benefit" of Xience V Over Taxus

September 27, 2010

September 27, 2010 (Washington, DC) — Two-year results from the COMPARE and SPIRIT IV trials confirm the superiority of the everolimus-eluting Xience V stent (Abbott) over two iterations of the paclitaxel-eluting Taxus stent (Boston Scientific). In COMPARE, an investigator-initiated, all-comers trial, researchers showed that the everolimus-eluting stent remained superior to the Taxus Liberté stent, with significant differences in the composite end point of death, nonfatal MI, and target vessel revascularization (TVR) observed at two years.

Similarly, the SPIRIT IV investigators, led by Dr Gregg Stone (Columbia University, New York), also showed sustained benefit of the Xience V stent over the older Taxus Express stent. At two years, treatment with the everolimus-eluting stent resulted in a relative 30% reduction in target lesion failure, a composite end point reflecting cardiac death, target vessel MI (TVMI), or ischemia-driven target lesion revascularization (TLR).

Dr Dean Kereiakes

Presenting the results of the SPIRIT IV study during a morning press conference at TCT 2010, investigator Dr Dean Kereiakes (The Christ Hospital, Cincinnati, OH), said both trials show a "durable, progressive relative benefit" of the Xience V stent over time.

"The other revelation is that the very late stent-thrombosis rates with the Xience V are low, either with or without extended dual antiplatelet therapy," Kereiakes told heartwire . "Our trial had 72% of patients taking dual antiplatelet therapy at two years, while their trial [COMPARE] had just about 10% of the Xience V cohort on dual therapy at two years. Either way, the very late stent-thrombosis rates were very low with the Xience V stent. Not so in the Taxus arm."

Dr Stephan Windecker (Bern University Hospital, Switzerland), who was not affiliated with the two trials, agreed, telling the media the message from the COMPARE everolimus-vs-paclitaxel comparison was the difference in stent-thrombosis rates between the two stents at one and two years.

"In terms of efficacy, I don't think we're surprised to see the superiority of this newer-generation stent," said Windecker. "The new message relates to the one-to-two year interval, a very late period, and the significant reduction in very late thrombosis. This is an important advance, because it was the principal limitation of both Cypher and Taxus."

COMPARE and SPIRIT IV at Two Years

Presented last year at TCT 2009 in San Francisco, CA and reported by heartwire at that time, SPIRIT IV investigators randomized 3690 patients in a 2:1 fashion to the Xience V or the Taxus Express 2 stent. Regarding the primary end point at 12 months, there was a 2.6% absolute difference in the rate of target lesion failure between the two stents, statistically favoring the Xience V. A statistically significant difference in stent thrombosis between the two arms was also observed at one year: 1.06% for Taxus vs 0.29% for Xience V (p=0.003).

With the two-year data presented today, there was an absolute 3.0% difference in target lesion failure, again statistically favoring the Xience V stent. There were also significant reductions in ischemia-driven TLR and TVMI with the everolimus-eluting stent, but no significant differences in mortality. The differences in definite or probable stent thrombosis were also in favor of the Xience V stent, with a two-year stent thrombosis rate of 1.23% among the Taxus-treated patients and 0.42% for the Xience-treated patients.

SPIRIT IV: Primary and Secondary End Points at Two Years

End point Xience V (n=2458), % Taxus (n=1229), % p
Target lesion failure 6.9 9.9 0.003
Ischemia-driven TLR 4.5 6.9 0.004
Cardiac death 0.9 1.3 0.34
Target vessel MI 2.3 3.5 0.04
Definite or probable stent thrombosis 0.42 1.23 0.008

Dual antiplatelet therapy was maintained in a majority of patients, 72% among patients treated with Xience and Taxus, until two years, note the researchers. They also note that diabetics did not fare as well as those without diabetes. At two years, as with the one-year data, there was no significant difference in the primary end point of target lesion failure among diabetic patients treated with the two stents.

Dr Ian Meredith

Speaking during the late-breaking clinical-trials session, Dr Ian Meredith (Monash Medical Center, Melbourne, Australia) said the sustained benefit of Xience over its Taxus counterpart is an important finding. He stressed, however, that SPIRIT IV was not an all-comers trial and that while investigators treated a wide range of vessels and lesion lengths, many patients were excluded.

"It's important to recognize that there were certain patients that weren't treated in this study, bearing in mind that we can't necessarily generalize the stent-thrombosis rates to these other patients, namely those patients with ostial lesions, chronic total occlusions, calcified vessels, acute MIs, saphenous vein grafts, and patients with thrombus, all of whom have a significant rate of early and late stent thrombosis," said Meredith.

Regarding the diabetes data, Meredith said that the Xience V stent appears to lose some efficacy among diabetic patients, and the reason is not entirely clear. By his interpretation, he said the results suggest that both stents fared equally well in the diabetic cohort.

Taking on All-Comers in COMPARE

In contrast with SPIRIT IV, the COMPARE investigators did take on all patients and included 1800 complex patients, such as those with acute MI and STEMI, as well as those with relatively high proportions of calcified lesions, bifurcations, multivessel disease, and diabetes. As reported previously by heartwire , the Xience V stent performed better than the Taxus stent, with significant reductions in the primary composite end point of death, nonfatal MI, and TVR. Similarly, there were reductions in secondary end points, as well as reductions in the risk of definite or probable stent thrombosis.

Presenting two-year data today, lead investigator Dr Peter Smits (Maasstad Ziekenhuis, Rotterdam, the Netherlands) reported that the difference in the primary end point was maintained at two years, with larger absolute differences between the Xience V and Taxus stents than at 12 months.

COMPARE: Primary and Secondary End Points at Two Years

End point Xience V (%) Taxus (%) p
MACE 9.0 13.7 0.0016
Cardiac death, nonfatal MI, and TLR 7.4 11.4 0.0038
Nonfatal MI 3.9 7.6 0.0009
Definite/probable stent thrombosis 0.9 3.9 <0.0001
Early stent thrombosis (<30 d) 0.2 1.7 0.002
Late stent thrombosis, (30 d–1 y) 0.3 0.9 0.13
Very late thrombosis (1–2 y) 0.3 1.5 0.013

MACE=major adverse cardiac events (all death, nonfatal MI, and TVR)

Unlike SPIRIT IV, where a majority of patients continued to take dual antiplatelet therapy at two years, just 13% of COMPARE patients were taking clopidogrel and aspirin after two years. Despite this, there were significantly lower rates of definite/probable stent thrombosis among those treated with the Xience stent.

Speaking during the late-breaking clinical-trials session, Dr Jeffrey Popma (Beth Israel Deaconess Medical Center, Boston, MA) said the stent-thrombosis data suggest that "there is something clearly there" with the Taxus Liberté stent. To heartwire , Kereiakes said the relative lack of dual antiplatelet therapy in the COMPARE trial highlights the differences between the two stents and exposes the "relative hazard" of Taxus, especially when patients stop taking clopidogrel.

Like the SPIRIT IV study, there was no benefit of Xience in diabetic patients.

A spokesperson for Boston Scientific noted that the Taxus Express stent, tested in SPIRIT IV, is no longer commercially available in the US. He also said the event rates observed in COMPARE, "specifically for early stent thrombosis, TLR, and MI," are not in line with results observed in other trials.

Stone disclosed receiving research grants from Atrium, Volcano, and TherOx; consulting fees from Osprey Medical, InfraRedx, Reva Medical, Boston Scientific, Abbott Vascular, the Medicines Company, AstraZeneca, Eli Lilly, Bristol-Myers Squibb, Edwards, Evalve, Ortho McNeil, and Otsuka; holding equity in Devax; BioStar I and II funds; MedFocus I, II, and Accelerator funds; FlowCardia; MediGuide; Guided Delivery Systems; Arstasis; Micardia; and Access Closure. Kereiakes reports research support from Abbott Vascular, Amylin Pharmaceuticals, Cordis, Boston Scientific, Medtronic, and Daiichi Sankyo and consulting fees from Abbott, Boston Scientific, Eli Lilly, Medpace, and Reva Medical. Windecker reports speakers' bureau fees from Abbott, Cordis, Medtronic, Boston Scientific, and Biosensors. Meredith reports consulting for Medtronic, Boston Scientific, and Abbott Vascular. Popma disclosed research support from Cordis, Medtronic, Boston Scientific, and Abbott and consulting fees from Cordis and Boston Scientific. Smits disclosed receiving unrestricted research grant support from Abbott, Boston Scientific, and Terumo Medical.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.