Association of Irreversible Electroporation Treatment with Overall Survival in Stage III Pancreatic Ductal Adenocarcinoma: DIRECT Registry Results

Abstract

Abstract Background Irreversible electroporation (IRE) is used in selected patients with Stage III pancreatic ductal adenocarcinoma (PDAC), but its association with survival relative to standard-of-care (SOC) therapy remains uncertain. Methods This multicenter observational study included 137 patients from the DIRECT registry: 99 treated within an IRE-based strategy and 38 receiving SOC without IRE. Overall survival (OS) from diagnosis was evaluated using propensity-score overlap weighting as the principal adjusted analysis. Conventional inverse probability of treatment weighting (IPTW), propensity-score matching, 3- and 6-month landmark analyses, and a time-dependent Cox model treating IRE as a time-varying exposure were sensitivity analyses. Restricted mean survival time (RMST) at 24 months was estimated. Results Median time from diagnosis to enrollment was 8 months (IQR, 6-10) with IRE and 4 months (IQR, 3-6) with SOC (P < 0.0001). Unadjusted median OS was 28 months (95% CI, 23-33) and 15 months (95% CI, 12-21), respectively. Overlap weighting was associated with lower mortality with IRE (HR, 0.28; 95% bootstrap CI, 0.12-0.48). Twenty-four-month RMST was 21.22 versus 16.24 months, an absolute difference of 4.99 months (95% bootstrap CI, 1.15-8.12). The time-dependent analysis attenuated but retained the association (HR, 0.59; 95% bootstrap CI, 0.37-0.94). Other sensitivity analyses were directionally consistent. Conclusions An IRE-based treatment strategy was associated with longer OS than SOC without IRE. Attenuation in the time-dependent analysis indicates that treatment timing and selection contributed to the observed difference. Given the nonrandomized design and potential residual confounding, these findings demonstrate association rather than an independent causal effect of IRE.