Retrospective cohort analysis of neoadjuvant treatment and survival in resectable and borderline resectable pancreatic ductal adenocarcinoma in a high volume referral centre

Abstract

BACKGROUND: Pancreatic ductal adenocarcinoma (PDAC) is a deadly disease. Neoadjuvant therapy (NA) with chemotherapy (NAC) and radiotherapy (RT) prior to surgery provides promise. In the absence of prospective data, well annotated clinical data from high-volume units may provide pilot data for randomised trials. METHODS: Medical records from a tertiary hospital in Sydney, Australia, were analysed to identify all patients with resectable or borderline resectable PDAC. Data regarding treatment, toxicity and survival were collected. RESULTS: Between January 1 2010 and April 1 2016, 220 sequential patients were treated: 87 with NA and 133 with upfront operation (UO). Forty-three NA patients (52%) and 5 UO patients (4%) were borderline resectable at diagnosis. Twenty-four borderline patients received NA RT, 22 sequential to NAC. The median overall survival (OS) in the NA group was 25.9 months (mo); 95% CI (21.1-43.0 mo) compared to 26.9 mo (19.7, 32.7) in the UO; HR 0.89; log-ranked p-value = 0.58. Sixty-nine NA patients (79%) were resected, mOS was 29.2 mo (22.27, not reached (NR)). Twenty-two NA (31%) versus 22 UO (17%) were node negative at operation (N0). In those managed with NAC/RT the mOS was 29.0 mo (17.3, NR). There were no post-operative deaths with NA within 90-days and three in the UO arm. DISCUSSION: This is a hypothesis generating retrospective review of a selected real-world population in a high-throughput unit. Treatment with NA was well tolerated. The long observed survival in this group may be explained by lymph node sterilisation by NA, and the achievement of R0 resection in a greater proportion of patients.