(OA09) Neoadjuvant Versus Adjuvant Radiation Therapy for Resectable Pancreatic Cancer: A Propensity Score Matched Analysis

Abstract

Both neoadjuvant and adjuvant therapy are widely used for resectable pancreatic adenocarcinoma. Prospective studies with mature data to compare these regimens are limited. The aim of this study is to compare the clinical outcome of patients with resectable pancreatic adenocarcinoma who underwent either neoadjuvant or adjuvant radiation therapy in a large population-based database over 10 years. The Surveillance, Epidemiology, and End Results (SEER) database was queried for cases of resected Stage I-III pancreatic cancer treated with radiation therapy between 2004 and 2014. Neoadjuvant and adjuvant radiation therapy cohorts were matched in a 1:1 ratio using propensity score. Our endpoints were overall survival (OS) and cancer-specific survival (CSS). R software (version 3.4.1) was used for statistical analysis. Of 1881 total cases of resected pancreatic cancer, including 160 cases treated with neoadjuvant radiation therapy, 146 matched pairs were constructed. All baseline characteristics were well balanced. There was no statistically significant difference in OS (P=0.56) and CSS (P=0.86) between neoadjuvant and adjuvant radiation therapy arms. Median OS was 16 months (IQR 10-35.8) for the neoadjuvant treatment arm and 18 months (IQR 11-27) for the adjuvant treatment arm. Multivariate analysis showed that age above 60 (HR 1.69, P<0.001), male gender (HR 1.53, P=0.0035), tumor grade (HR 1.41, P=0.0016), and nodal stage (HR 1.43, P=0.013) were associated with poor OS. The same variables were also associated with poor CSS. Neoadjuvant and adjuvant radiation therapy regimens showed no difference in clinical outcomes. Poorly differentiated tumor grade and positive nodal status were independent adverse prognostic factors for survival.