Failure to Undergo Resection Following Neoadjuvant Therapy for Resectable Pancreatic Cancer: A Secondary Analysis of SWOG S1505

Abstract

Background: Neoadjuvant therapy (NT) is increasingly used for patients with pancreatic ductal adenocarcinoma (PDAC), and yet reasons for not undergoing subsequent pancreatectomy are poorly understood. Given the importance of completing multimodality therapy, we investigated factors associated with failure to undergo surgical resection following NT for PDAC. Methods: SWOG S1505 was a multicenter phase II randomized trial of preoperative mFOLFIRINOX or gemcitabine/nab-paclitaxel prior to planned pancreatectomy for patients with potentially resectable PDAC. Associations between clinical, demographic, and hospital-level characteristics and receipt of surgical resection were estimated via multiple logistic regression. Differences in overall survival from 18 weeks postrandomization (scheduled time of surgery) according to resection status were assessed via Cox regression models. Results: Among 102 eligible patients, 73 (71.6%) underwent successful pancreatectomy, whereas 29 (28.4%) did not, primarily because of progression (n=11; 10.8%) or toxicity during NT (n=9; 8.8%). Weight loss during NT (odds ratio [OR], 0.34; 95% CI, 0.11–0.93) and the hospital’s city size (small: OR, 0.24 [95% CI, 0.07–0.80] and large: OR, 0.28 [95% CI, 0.10–0.79] compared with midsize) were significantly associated with a lower probability of surgical resection in adjusted models, whereas age, sex, race, body mass index, performance status, insurance type, geographic region, treatment arm, tumor location, chemotherapy delays/modifications, and hospital characteristics were not. Surgical resection following NT was associated with improved overall survival (median, 23.8 vs 10.8 months; P<.01) even after adjusting for grade 3–5 adverse events during NT, performance status, and body mass index (hazard ratio, 0.55; 95% CI, 0.32–0.95). Conclusions: Failure to undergo resection following NT was relatively common among patients with potentially resectable PDAC and associated with worse survival. Although few predictive factors were identified in this secondary analysis of the SWOG S1505 randomized trial, further research must focus on risk factors for severe toxicities during NT that preclude surgical resection so that patient-centered interventions can be delivered or alternate treatment sequencing can be recommended.