The high mortality rate of pancreatic ductal adenocarcinoma (PDA) is mainly due to a lack of highly sensitive and specific tools to detect the disease at an early stage and, as such, tumors are typically diagnosed after metastasis. There is an urgent need to further improve our understanding of molecular events leading to PDA development and to identify strategies for diagnosing and treating PDA in its preinvasive state before metastasis occurs. Intraductal papillary mucinous neoplasias (IPMNs) are macroscopically identifiable, non- invasive, preneoplastic cystic lesions that are precursors to PDA. IPMNs as a group can be categorized into different histological subtypes that are associated with different survival outcomes. While knowledge of the IPMN subtype in patients could be of prognostic value, preoperative diagnostic subtyping of IPMNs is difficult. Identifying biomarkers for IPMN subtypes, such as unique gene mutations that can be detected in aspirates from pancreatic cysts or serum, would help distinguish cysts with low and high malignant potential. To further our understanding of IPMN biology and identify markers predictive of malignant potential, we have developed the first mouse model of IPMN that fully recapitulates the clinicopathological features of human IPMN formation and progression. Based on observations in human patients showing that dysregulation of the PI3K/PTEN signaling pathway is associated with IPMN-PDA, we employed a genetic strategy to inducibly ablate the duct- enriched tumor suppressor Pten specifically in pancreatic ductal cells in mice. These mice spontaneously form IPMN lesions of multiple histological subtypes. Interestingly, in a subset of mice that progress to invasive PDA, we detected spontaneous activating mutations of Kras in the associated IPMN lesions. These activating Kras mutations were specifically found in pancreatobiliary IPMNs (PB-IPMN), suggesting that combined loss of Pten and Kras activation drives initiation and/or progression of this IPMN subtype. To test this hypothesis, we will genetically determine if activation of Kras and loss of Pten synergize in mouse pancreatic ductal cells to drive the formation and/or progression of PB-IPMNs (Aim 1). Further, we propose to employ this IPMN mouse model to identify novel mutations involved in IPMN initiation and progression by performing exome sequencing on tissue samples from our mouse model (Aim 2). Finally, we will employ a human tissue microarray of pancreatic lesions to examine whether the same mutations are present in human IPMNs and associate with specific IPMN subtypes (Aim 2). Results from our proposed studies will not only identify candidate diagnostic biomarkers for pre-operatively assessing malignant potential of IPMNs, but may also reveal novel signaling pathways to target for drug discovery.