218 Background: The development of sensitive and specific biomarkers for the early diagnosis of human pancreatic ductal adenocarcinoma (PDAC) represents a major unmet clinical need. GEMMs that faithfully recapitulate the clinical, histological and molecular features of PDAC and capture its genomic complexity (including that triggered by a truncating germline Brca2 mutation) coupled with state of the art proteomic technologies can overcome some of the challenges associated with the analysis of a heterogeneous human disease. We have conducted a large scale screen for differentially expressed glycoproteins in pancreatic tissue and serum from mice with PDAC, pancreatic intra-epithelial neoplasia (PanIN) and matched healthy controls, using the well validated KPCB model of PDAC. Methods: Tryptic digests of protein extracts from snap frozen PDAC, pre-malignant and normal pancreata were subjected to solid-phase extraction of N-glycopeptides and the samples were analysed on a LTQ Orbitrap XL mass spectrometer. LC-MS feature maps were compared employing SuperHirn software and used for the label-free quantification of corresponding proteins. Primary mPDAC cell lines and their supernatants were also analysed. Selected candidate proteins were then quantified in mouse serum by SRM-based targeted proteomics. Results: 146 glycoproteins were identified as overexpressed by >2 fold (p<0.05, Student’s t test) in PDAC compared to normal pancreas and 60 were prioritised for targeted identification in mouse serum. 12 proteins were significantly upregulated (p<0.01, Mann-Whitney U test ) in the serum of tumour-bearing mice with PDAC compared to controls. Conclusions: Using a state of the art GEMM of PDAC, we have successfully applied a two-stage strategy for the tissue-based identification followed by quantification in murine serum of a panel of glycoproteins that are consistently over-expressed during mPDAC development. Validation in serum collections from patients with PDAC, chronic pancreatitis and healthy controls is underway and updated results will be presented at the conference.
Read full abstract