Four-year patient-level pooled mortality analysis of the ILLUMENATE US Pivotal and EU randomized controlled trials

Abstract

ObjectiveTo perform a meta-analysis of two concordant randomized controlled trials (RCTs) examining the long-term, 4-year safety profile of the Stellarex drug-coated balloon (DCB) vs percutaneous transluminal angioplasty (PTA) for the treatment of peripheral artery disease. MethodsAn independent, third-party, meta-analysis of homogeneous, patient-level data from the ILLUMENATE Pivotal and ILLUMENATE EU RCTs was performed to assess mortality (time to death) in patients treated for symptomatic femoropopliteal disease. The Kaplan-Meier (KM) methodology was used to estimate hazard rates [HRs] of all-cause mortality, and Cox proportional hazard modeling was used to assess predictors of mortality. All serious adverse events, including deaths, were adjudicated by an independent, blinded clinical events committee. ResultsIn total, 589 (419 DCB; 170 PTA) patients were included in the pooled analysis of the ILLUMENATE Pivotal and ILLUMENATE EU RCTs. The median follow-up was 1735 days (interquartile range, 1434-1829 days), equivalent to 4.75 years. Vital status compliance was >95% in each RCT. The total number of deaths through 4 years was 81 of 589 (13.8%): 58 of 419 (13.8%) in the DCB arm and 23 of 170 (13.5%) in the PTA arm. The 1-year KM estimate of all-cause mortality was 1.9% ± 0.7% (estimate ±standard error) in those treated with DCB vs 1.2% ± 0.9% in those treated with PTA. At 2, 3, and 4 years, the respective KM estimates were 6.6% ± 1.2% vs 4.9% ± 1.7%, 9.3% ± 1.4% vs 9.9% ± 2.4%, and 14.0% ± 1.7% vs 14.4% ± 2.8% (P = .864). There were no significant differences in clinical events committee-adjudicated deaths between the two cohorts. In multivariate analysis, predictors of 4-year mortality were age (HR, 1.048; 95% confidence interval [CI], 1.026-1.071; P < .0001), renal insufficiency (HR, 2.440; 95% CI, 1.566-3.800; P < .0001), and lesion length (HR, 1.004; 95% CI, 1.000-1.008; P = .041). Neither paclitaxel exposure (DCB vs PTA; HR, 1.086; 95% CI, 0.709-1.664; P = .705) nor dose (mg; HR, 1.043; 95% CI, 0.971-1.119; P = .248) was the predictor of all-cause mortality at 4 years. ConclusionsThis systematic meta-analysis of two concordant ILLUMENATE RCTs shows no difference in all-cause mortality through 4 years between Stellarex DCB and PTA, confirming the acceptable, long-term safety profile of the Stellarex DCB.