ABSTRACT Pancreatic ductal adenocarcinoma (PDAC) is one of the most lethal cancers, primarily due to most cases being diagnosed at an advanced, incurable stage. While 5-year survival of metastatic PDAC is <5%, outcomes dramatically improve for localized PDAC. Poor prognosis is due to a lack of biomarkers for diagnosing PDAC at an early, asymptomatic stage when cure is possible. Effective diagnosis of early stage PDAC depends on identification of accurate, non-invasive biomarkers in combination with a strategy for screening increased risk populations. Our primary objective is to identify non-invasive protein biomarkers in serum, urine, and exosomes that accurately distinguish between patients with and without early stage resectable PDAC that is amenable to curative surgery. Novel diagnostics would also improve discrimination between PDAC and benign pancreatic pathologies. The goal of the proposed research is to develop clinically translatable noninvasive biomarkers- based tests for screening (in high risk groups) and differential diagnosis of PDAC. Our central hypothesis is that combinations of urinary, serum, and exosome derived biomarkers could be synergistic offering a superior classification power. We have used urine and serum samples from retrospective and prospective cohort studies to identify a range of strong candidate combinatorial multimarker algorithms for early detection and diagnosis of PDAC. In Aim 1, we will optimize the performance of a PDAC differential diagnosis algorithm and will validate the optimized algorithm in samples collected prior to clinical diagnosis in the Pancreatic Adenocarcinoma Gene Environment Risk (PAGER) study. In Aim 2, we will optimize the performance of an early detection algorithm for resectable PDAC in pre-diagnostic samples from three prospectively collected cohorts and validate the optimized EDA in blinded parallel serum/urine samples from the Southern Community Cohort Study (SCCS). If successful, our project will yield novel, validated algorithms for risk assessment and early detection and for differential diagnosis of PDAC. These algorithms when combined will result in a new pioneering screening paradigm for PDAC allowing for timely live-saving interventions. Our strong preliminary data, powerful and synergistic investigative team, and the availability of parallel urine and serum samples from unique prospective cohorts contribute to the high probability of successful accomplishing the proposed studies.

NARRATIVE We propose to develop a novel screening strategy to detect patients at high-risk for developing pancreatic cancer at least 6 months prior to the development of symptoms. The approach is based on following innovative premises: (i) using urine biomarkers to better enrich the screening population, and thereby, identify the subset of patients in whom more expensive and invasive surveillance strategies are warranted, which should significantly reduce pancreatic cancer mortality and patient care costs; (ii) using urine exosomes rather than total urine for some tests to further concentrate disease specific biomarkers, and (iii) using samples collected before diagnosis in prospective studies.