Mortality Assessment of Paclitaxel-Coated Balloons: Patient-Level Meta-Analysis of the ILLUMENATE Clinical Program at 3 Years

A recent summary-level meta-analysis comprising randomized, controlled trials (RCTs) of femoropopliteal paclitaxel-coated balloon and stent intervention identified excess late mortality in the paclitaxel-treated patients. METHODS: We evaluated the safety of the Stellarex drug-coated balloon (DCB) for femoropopliteal artery disease with an independently performed meta-analysis of patient-level data from all patients in the Stellarex femoropopliteal clinical program. To compare mortality after DCB or uncoated percutaneous transluminal angioplasty (PTA), we aggregated data from 2 RCTs comprising 419 patients treated with DCB and 170 patients treated with PTA. In an additional analysis, data were aggregated from 6 poolable Stellarex DCB studies (2 RCTs, 3 single-arm studies, and 1 registry). All serious adverse events including deaths were adjudicated by a blinded, third-party, independent Clinical Events Committee. Kaplan–Meier estimates in the RCTs were compared with restricted mean survival time. Predictors of death were assessed with hazard ratios (HRs) and Cox proportional hazards modeling. RESULTS: Baseline characteristics were similar in the patients treated with DCB and PTA in the pooled RCT analysis, with the exception that the DCB cohort was younger (67.4±9.7 versus 69.4±9.4 years, P=0.02), smoked more frequently (86.6% versus 78.8%, P=0.02), and were less often treated for recurrent lesions (8.8% versus 14.7%, P=0.04). In the RCTs, patients treated with DCB had all-cause mortality rates that were not different from those of patients treated with PTA (Kaplan–Meier estimates 1.8±0.7% versus 1.3±0.9%, 6.5±1.2% versus 5.9±1.9%, and 9.3±1.5% versus 9.9±2.4% at 1, 2, and 3 years, respectively, P=0.86). All-cause mortality rates were similar in a 1906-patient pooled nonrandomized DCB data set (Kaplan–Meier estimates of 2.1%, 4.9%, and 7.0% at 1, 2, and 3 years, respectively). Clinical Events Committee–adjudicated causes of death were balanced between the DCB and PTA cohorts. Multivariable Cox modeling identified age (HR, 1.06; 95% CI, 1.04–1.08; P<0.001), diabetes mellitus (HR, 1.42; 95% CI, 1.01–2.00; P=0.04), congestive heart failure (HR, 1.88; 95% CI, 1.12–3.16; P=0.02), and renal insufficiency (HR, 2.00; 95% CI, 1.33–3.01; P<0.001) as predictors of mortality. Paclitaxel exposure was unrelated to mortality (HR, 1.04; 95% CI, 0.98–1.10; P=0.23). CONCLUSIONS: The mortality rates for patients treated with the DCB and uncoated PTA were indistinguishable over 3-year follow-up. Additional patient-level, adequately powered meta-analyses with larger RCT data sets will be needed to confirm the generalizability of these findings. CLINICAL TRIAL REGISTRATION: URL: http://www.clinicaltrials.gov. Unique identifiers: NCT02110524, NCT01858363, NCT01858428, NCT03421561, NCT01912937, NCT01927068, and NCT02769273.