Abstract 2506: Differentiating pancreatic cystic neoplasms by global protease specificity

Abstract

Abstract Decision-making for patients with pancreatic cystic lesions represents the greatest area of clinical uncertainty in the management of pancreatic disease today. Short of pancreas removal, there is no definitive technology that differentiates benign, pre- malignant, and malignant pancreatic cystic neoplasms. As a result, some patients die from undiagnosed cancers, while others undergo unnecessary pancreatic resections with considerable morbidity. The cyst fluid from these neoplasms is a rich resource for the discovery of new biomarkers because it is in direct contact with the involved cells, and it is sampled routinely as a part of clinical care. Amylase activity and CEA concentration are typically measured, but they lack prognostic value. Protein levels of proteases and their inhibitors differ between benign and malignant cystic neoplasms, however, effects on enzymatic activity and specificity profiles are unknown. We sought to identify and characterize proteolytic signatures specific to different stages of pancreatic carcinogenesis. We used a new protease substrate profiling technology recently developed in our lab, Multiplex Substrate-Profiling by Mass Spectrometry (MSP-MS), that can detect and classify the specificity of endo- and exopeptidases in a complex sample. This assay consists of a physiochemically diverse library of 228 tetradecapeptides designed by incorporating all combinations of neighbor and near-neighbor amino acid pairs to provide maximum specificity information. We have previously used this technology to determine the “proteolytic signatures” of mouse pancreatic cancer cell secretions and human neutrophil extracellular traps. MSP-MS has the unique ability to monitor all peptide bonds cleaved from this library of peptides, thereby simultaneously detecting aminopeptidase, carboxypeptidase and endopeptidase activity. This assay was applied to pancreatic cyst fluid derived from 4 low risk and 4 high risk lesions as determined by current clinical guidelines. The cleavage data from the respective categories was combined and compared. The proteases in the samples generated 773 cleavage sites in the low risk category and only 285 in the high risk category. Interestingly, only 220 of the sites were shared between groups. The signature from the low risk group had a trypsin-like specificity with arginine and lysine preferred at the P1 position, whereas the high risk group had minimal trypsin-like specificity and predominantly showed amino- and carboxypeptidase activity. These data suggest that protease profiling of pancreatic cyst fluid may facilitate risk stratification among pancreatic cystic lesions and thereby guide clinical decision-making. MSP-MS offers a unique technology to both identify active proteases in clinical samples of interest and measure total and specific proteolytic activity despite the presence of endogenous inhibitors. Citation Format: Dana A. Dominguez, Anthony J. O'Donoghue, Kimberly S. Kirkwood, Charles S. Craik. Differentiating pancreatic cystic neoplasms by global protease specificity. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 2506. doi:10.1158/1538-7445.AM2014-2506