Introduction: Evidence supporting adjuvant therapy for resected pancreatic cancer is limited primarily to head tumors. We analyzed data from the National Cancer Database (NCDB) to evaluate the relationship of tumor site with benefit from adjunctive (adjuvant, neoadjuvant, perioperative) therapy (Rx).
Methods: All NCDB patients with clinical stage I and II pancreatic cancer, diagnosed from 2003 to 2013, who underwent surgical resection and had data on site of primary were included. Overall survival (OS) analyses with hazard ratios (HR), 95% confidence intervals (CI), and two-sided p-values are presented.
Results: A total of 27,930 patients met inclusion criteria; median age 66 years, 51% males, 86% white. Primary site was coded as head (74.4%), body (9.3%), or tail (16.3%). Pathologic stage was predominantly stage II (77%); 81% had negative margins. Perioperative Rx was used in 4%, neoadjuvant in 8%, adjuvant in 48%. Median OS for the cohort was 24 months; for head, body and tail tumors, it was 21.6, 34.5, and 42.5 months, respectively. In univariable analyses, adjunctive Rx was associated with improved OS in head tumors (HR, any Rx vs. no Rx: 0.87; 95% CI 0.84-0.91; p < 0.0001) but not in body (1.82; 1.59-2.08; <0.0001) and tail (2.28; 2.05-2.53; <0.0001) tumors; multivariable models including statistically significant predictors (Charlson-Deyo comorbidity score, tumor grade and stage, positive resection margin) confirmed these results.
Conclusions: Our study suggests that the benefit of adjunctive Rx is restricted to pancreatic head tumors; body and tail tumors have a much better prognosis. These results warrant further evaluation in prospective studies.
Copyright © 2018 Elsevier Ltd. All rights reserved.