TAXUS-IV: The Pivotal Results
A look at the study design, technology, and 9-month results that secured Boston Scientific a November date for FDA panel review.
To view the table related to this article, please refer to the print version of our October issue, page 36.
At last month’s Transcatheter Cardiovascular Therapeutics meeting, investigators from the TAXUS-IV pivotal trial of Boston Scientific Corporation’s (Natick, MA) TAXUS Express2 stent presented the much-anticipated results of the device’s pivotal clinical trial, TAXUS-IV. One day after Principal Investigator Gregg W. Stone, MD, presented the 9-month data, Boston Scientific announced that the FDA had scheduled a November 20, 2003, panel hearing to review the Pre-Market Approval application for the paclitaxel-eluting TAXUS device. The favorable results of the trial, combined with the late-Fall review date, are a strong indicator that the US drug-eluting stent market may soon have its second entrant.
TAXUS-IV is a prospective, randomized, double blind trial of 1,326 patients undergoing elective stenting at 73 US sites. Single de novo lesions coverable by one stent were treated with 1 µg/mm2, slow-rate-release paclitaxel-eluting Express2 (TAXUS) stents in the study arm; patients in the control arm received bare Express2 stents. Clinical follow-up was scheduled for 1, 4, and 9 months, and annually for 5 years. Angiographic follow-up was scheduled for 9 months (n=732), as was IVUS follow-up (n=268).
Key Exclusion Criteria
Patients were excluded from the study if their history included PCI in the target vessel within 9 months of evaluation; myocardial infarction (MI) within 72 hours or CK-MB >2x ULN on the day of the procedure; planned use of atherectomy or a cutting balloon; and previous or planned brachytherapy in the target vessel. Target vessel and lesion characteristics that warranted exclusion included having ostial lesions within 3 mm of the lesion; a bifurcation branch >2 mm in diameter with >50% DS; initial TIMI flow of 0 or 1; probable or definite thrombosis; and excessive proximal vessel or lesion tortuosity or calcification.
TAXUS-IV’s primary endpoint was 9-month ischemia-driven target vessel revascularization.
A total of 1,326 patients were initially randomized for the trial; five patients were deregistered in the TAXUS arm and seven in the control arm, resulting in a study population of 1,314 patients (TAXUS, n=662; control, n=652). This population included 226 patients receiving 32-mm stents. Angiographic follow-up was preselecteded for 732 patients; 559 of those patients returned for angiographic follow-up at 9 months (76%).
TAXUS and control groups were well-matched for baseline patient characteristics. Patients averaged 62.8 years of age in the TAXUS group, which was not markedly different from their counterparts in the control group (62.1%). The majority of patients in both groups were male (TAXUS, 71.8%; control, 72.4%). Twenty-three percent of the TAXUS-treated patients had medically treated diabetes mellitus, with 7.7% requiring insulin, versus 25% (8.3% insulin-requiring) in the control group. History of medically treated hypertension was observed in 70.5% of TAXUS patients and 69% of controls; 65% of TAXUS patients had medically treated hyperlipidemia versus 65.6% of controls. Current smokers comprised 23.4% of the TAXUS group and 20.1% of the control group; 30.5% of TAXUS patients and 29.1% of controls had experienced previous MI.
Target lesions ranged from 2.5 mm to 3.75 mm in diameter and 10 mm to 28 mm in length. Devices used in both cohorts were available in 2.5-mm, 3-mm, and 3.5-mm diameters, and 16-mm, 24-mm, and 32-mm lengths. The majority (91%) of patients in both groups required the implantation of only one study stent. The most commonly implanted stent length was 16 mm, which was used in more than three times as many cases as the 24-mm and 32-mm lengths. The 3.0-mm diameter stent was implanted most frequently, followed by the 3.5-mm diameter (2.5-mm length stents are not available in lengths of 32 mm).
The target lesion revascularization (TLR) rate in the TAXUS group was 3% versus 11.3% in the control group. The TAXUS group also experienced a significantly lower rate of target vessel revascularization (TVR) than did the control group (4.7% vs 12%, respectively) (Table 1).
The binary in-segment (includes 5 mm proximal and distal to stent) restenosis rate in the TAXUS group was 7.9%, compared with 26.6% in the control group; in-stent binary restenosis rates were 5.5% and 24.4% respectively. In-segment percent diameter stenosis was significantly better in the TAXUS groups (26.3% vs 39.8%), and in-segment minimum lumen diameter was 2.03 mm in the TAXUS group compared with1.68 mm in the control group. In-segment late lumen loss results also favored the TAXUS group (0.23 mm vs 0.61 mm), as did in-stent late loss (0.39 mm vs 0.92 mm) (Table 1).
The overall rates of major adverse cardiac events (MACE) were 8.5% in the TAXUS group and 15% in the control group. Stent thrombosis rates were low overall, and were not significantly different in the TAXUS group versus the control (0.6% vs 0.8%, respectively). Target vessel failure was 47% lower in the TAXUS group (7.6%) than in the control group (14.4%).
Efficacy in Diabetic Patients
Diabetic patients represented approximately 24% of the overall population. Diabetic patients in the TAXUS group had a TLR rate of 5.2%, whereas the rate for diabetic controls was 16.0%. In-stent late loss for diabetic patients treated with the TAXUS stent was 0.37 mm, compared with 0.96 mm for those in the control group.
TAXUS-IV Versus SIRIUS?
Earlier this year, Cordis Corporation (a Johnson & Johnson company, Miami, FL) received FDA approval for the CYPHER Sirolimus-eluting Coronary Stent, opening the doors of the DES market in the US. Naturally, there will be a desire to compare the results of TAXUS-IV to those of SIRIUS, the pivotal trial that convinced the FDA of CYPHER’s safety and efficacy, in the hopes of determining which DES platform is most effective, and consequently, which will be the platform of choice in the US. Although the trials had many similarities, it is important to note the significant areas in which they differ.
Stephen G. Ellis, MD, Co-Principal Investigator for TAXUS-IV, gave a presentation highlighting some of its finer details, wherein he discussed why comparing the two landmark studies is not exactly an apples-to-apples discussion. For instance, randomization was performed predilation in TAXUS-IV and postdilation in SIRIUS. Also, TAXUS-IV’s guidelines permitted non-target-vessel stenting, whereas it was prohibited in SIRIUS. More obvious differences such as using different platforms resulted in variability of sizes and numbers of stents being implanted.
Further discrepancies in study design included different medication durations and definitions for MACE. Clopidogrel was administered to TAXUS-IV patients for 6 months and to SIRIUS patients for 3 months. MACE in SIRIUS was a composite of death, MI, and TLR, whereas death, MI, and TVR (which includes TLR) comprised MACE in TAXUS-IV. The average patient age did not vary considerably between the two trials, nor did lesion length. Mean vessel diameters were also similar (approximately 2.8 mm). Dr. Ellis further explained that the DES restenosis rates in the trials were comparable, as were revascularization rates.