Irreversible electroporation versus chemotherapy with or without radiotherapy for unresectable locally advanced pancreatic adenocarcinoma: a systematic review and meta-analysis

Abstract

Presenter: Kavin Sugumar MD | Case Western Reserve University School of Medicine Background: Multi-agent chemotherapy (CT) +/- radiotherapy (RT) is the standard treatment for patients with non-metastatic, locoregionally unresectable pancreatic adenocarcinoma. Irreversible electroporation (IRE) has been described as a locoregional treatment modality for LAPC patients, but postoperative morbidity and mortality rates are high for an invasive, non-curative intervention. Few studies compare IRE with chemotherapy +/- RT. We performed a comprehensive meta-analysis of the available literature comparing overall survival, progression-free survival, and treatment-related complications between these groups. Methods: A systematic literature search was performed in Medline and Embase in January 2020. Due to the paucity of studies (n = 3) that compare IRE to chemotherapy +/- RT, two separate literature searches were performed. Studies evaluating the outcomes of IRE combined with or without CT were included and grouped into the IRE cohort. Studies in which CT +/- RT only were utilized were grouped as the chemotherapy (CT) cohort. The primary outcomes evaluated were overall survival (OS; at 6/12/24 months) and progression free survival (PFS; at 6/12 months) defined from the time of diagnosis. Additional OS/PFS analyses in the IRE group were performed starting from the time of procedure. Forest plot analyses were used to calculate the weighted average survival estimates (OS and PFS) at the specified time points. Secondary outcomes included treatment related morbidity and mortality. Results: Of 585 published articles, 27 and 21 studies met inclusion criteria for the IRE and CT groups, respectively. All studies were observational cohort studies except for 4 phase I-II trials (8%). Combined, these studies included clinical data on 1420 (IRE) and 1348 (CT) patients. The pooled 6-, 12-, and 24-month OS estimates for the IRE group were 99%, 84%, and 28%. The pooled 6-, 12-, and 24-month OS estimates for the CT group were 99%, 80%, and 12% (Figure). There was overlap in the 95% confidence intervals of OS among the treatment groups. The pooled 6- and 12-month PFS estimates for IRE were 98% and 63%, and for CT were 61% and 29% (Figure). From the IRE procedure, 6-, 12-, and 24-month OS was 89%, 55%, and 12%, and 6-/12-month PFS was 58% and 27%. From the IRE procedure, 6- and 12-month PFS was 58% and 27%. The median major complication (Clavien-Dindo > = 3) and 90-day mortality rates reported in the IRE group were 12% (range: 0-53%) and 2% (range: 0-17%), repectively. The median grade 3-4 adverse effect score in the CT group was 24% (range: 14-52%), and there were no associated treatment-related deaths. Conclusion: There is a striking paucity of studies comparing IRE with the standard of care treatment approaches for LAPC. From time of diagnosis, it appears that IRE has similar OS compared to multiagent CT +/- RT and may have better PFS at 6- and 12-months. However, the majority of patients progress and nearly half die within 1 year of the IRE procedure. Reported non-curative IRE-related morbidity and mortality rates approach those associated curative-intent pancreatectomy. Until high quality, prospective studies with standardized indications are conducted, IRE should be used with caution and remains experimental in the treatment of pancreatic adenocarcinoma.