Prognostic impact of chemoradiation-related lymphopenia in patients with locally advanced pancreatic cancer.

Abstract

439 Background: Chemoradiation (CRT) induced lymphopenia is common and associated with poorer survival in multiple solid malignancies. The objective of this study was to evaluate the prognostic impact of lymphopenia in patients with nonmetastatic, unresectable pancreatic ductal adenocarcinoma (PDAC) treated by neoadjuvant FOLFIRINOX (5-fluorouracil [5FU]/leucovorin/irinotecan/oxaliplatin) followed by CRT. We hypothesized that severe lymphopenia would correlate with worse survival. Methods: The inclusion criteria for this single-institution retrospective study were: 1) biopsy-proven diagnosis of unresectable PDAC, 2) absence of distant metastasis, 3) receipt of neoadjuvant FOLFIRINOX followed by CRT, and 4) absolute lymphocyte count (ALC) available prior to and two months after initiating CRT. In general, CRT consisted of 5FU or capecitabine and RT with 58.8 Gy over 28 fractions. Lymphopenia was graded according to CTCAE v5.0. The primary variable of interest was lymphopenia at two months, dichotomized by ALC of < 0.5/μl (Grade 3 lymphopenia). The primary endpoint was overall survival (OS). Cox modeling and Kaplan-Meier methods were used to perform survival analyses. Results: A total of 138 patients were identified. Median follow-up for the entire cohort was 16 months. Median age was 65. Fifty-six percent were female, 86% were Caucasian, and 97% had ECOG ≤1. Median tumor size was 3.8 cm. Tumor location was pancreatic head in 63%, body in 22%, tail in 8%, and neck in 7%. Median baseline ALC for the entire cohort was 1.5 k/ul. Two months after initiating CRT, 106 (77%) had severe (Grade 3 or worse) lymphopenia. While there were no significant differences in baseline patient or disease characteristics, patients with severe lymphopenia received higher doses of RT with longer duration of treatment compared to those without severe lymphopenia. On multivariable Cox model, severe lymphopenia at two months was significantly associated with increased hazards of death (HR = 4.00 [95% CI 2.03-7.89], p < 0.001). Greater number of neoadjuvant FOLFIRINOX cycles received prior to CRT was associated with lower hazards of death (HR = 0.84 [95% CI 0.77-0.92], p < 0.001). The 12-month OS was 73% vs. 90% in the cohort with vs. without severe lymphopenia, respectively (log-rank p < 0.001). Conclusions: Treatment-related lymphopenia is common and severe lymphopenia may be a prognostic marker of poorer survival in locally advanced pancreatic cancer. Closer observation in high-risk patients and minimization of RT dose and duration are potential approaches to mitigating CRT-related lymphopenia. Our findings also suggest an important role of the host immunity in pancreatic cancer outcomes, supporting the ongoing efforts of immunotherapy trials in pancreatic cancer.