Neoadjuvant treatment in pancreatic cancer: Evidence-based medicine? A systematic review and meta-analysis.

Neoadjuvant treatment in non-metastatic pancreatic cancer (PaC) has the theoretical advantages of downstaging the tumor, sterilizing any present systemic undetectable disease, selecting patients for surgery and administering therapy to each patient. The aim of this systematic review is to analyze the state of the art on neoadjuvant protocols for non-metastatic PaC. A literature search over the last 10 years was conducted, and papers had to be focused on resectable, borderline resectable (BLR) or locally advanced (LA) histo- or cytologically proven PaC; to be prospective studies or prospectively collected databases; to report percentage of protocol achievement and survival data at least in an intention-to-treat (ITT) analysis. Twelve studies were eligible for systematic review. Studies included a total of 624 patients: 248 resectable, 268 BLR, 71 LA and 37 non-specified. All studies were included for meta-analysis. ITT overall survival (OS) was 16.7 months (95% CI 15.16-18.26 months); for resected patients OS was 22.78 months (95% CI 20.42-25.16), and for eventually non-resected patients it was 9.89 months (95% CI 8.84-10.96). Neoadjuvant approaches for resectable, BLR and LA PaC are spreading. Outcomes tend to be better outside an RCT context, but strong evidences are lacking. Actually such treatments should be performed only in a randomized clinical trial setting.