Validation of biomarkers for risk prediction and early diagnosis of Pancreatic Adenocarcinoma

摘要

Abstract Pancreatic adenocarcinoma (PDAC), with an overall 5-year survival of 11%, is the 3rd most common cause of cancer deaths in the United States, despite accounting for only 2% of all malignancies. Only a minority of patients (~11%) are diagnosed with “localized” disease (I or IIA), which has a 5-year survival rate of about 40% in setting of a node-negative, margin negative pancreatic resection. An obvious strategy for improving the dismal survival would be to detect PDAC when localized and thus at a more curable stage. Since screening the general population for PDAC is not feasible, current efforts have focused on identifying a subset of the people at an increased risk for PDAC development. Currently, only up to 25% of individuals who develop PDAC are candidates for pancreatic cancer surveillance. About 10% are individuals with a strong family history or a combination of family history and germline mutations associated with the risk of PDAC development. The other ~15% are individuals with cystic neoplasms of the pancreas, including IPMNs and MCNs. The inability to predict the malignant transformation of mucinous cysts and thus identify the cysts that should be surgically removed requires appropriate surveillance. Despite developing multiple consensus guidelines on managing cystic lesions, it is still challenging to determine which mucinous cysts will undergo malignant transformation. During the previous funding cycle, we identified and evaluated novel serum biomarker panels comprising mucins (MUC4, MUC5AC), mucin-associated glycoepitopes (STRA), TGM2, THSP2, TIMP2, and autoantibodies that were validated in a blinded case-control cohort aimed at detecting resectable PDAC. Preliminary mutational profiling of pancreatic cyst fluid (phase I and II) identified a unique panel (PancreaSeq) that helped define the malignant risk of pancreatic cysts. The goal of the current Clinical Validation Center (CVC) is to validate further these panel(s) in a prospective-specimen-collection, retrospective-blinded-evaluation (PRoBE) compliant manner in cohorts of high-risk individuals who are current or potential candidates for early detection or diagnosis of PDAC. Aim 1 will validate cyst fluid and blood-based biomarkers in patients with cystic lesions to identify advanced precursor lesions and differentiate low-grade dysplasia and no lethal potential. In this aim, we will validate candidate biomarkers, including PancreaSeq mutational panel (Phase 3), inflammatory markers (Phase 2), and mucins (MUC4, MUC5AC, STRA) (Phase 3) in the cyst fluid (high specificity) and serum samples (high sensitivity) to identify interval malignancy during surveillance imaging. Aim 2 will evaluate the performance of optimized biomarker panel(s) for early detection of malignant disease in patients with a hereditary predisposition undergoing surveillance for PDAC development and in patients with new-onset diabetes and chronic pancreatitis, defined as within two years of initial diagnosis, who are at a significantly increased risk compared to the general population for having an undiagnosed PDAC. Impact: The proposed studies will validate the clinical utility of promising biomarker panels for early detection of PDAC and risk stratification of pancreatic cystic lesions, which can be used in phase IV clinical utility studies.