Pancreatic cancer is a devastating disease with one of the lowest survival rates of all cancers, partly due to the difficulty of detection in its early stages. Recent developments in liquid chromatography – mass spectrometry (LC-MS) technology have made it possible to profile proteins associated with disease in a comprehensive manner. This work sought to apply current LC-MS methodology to the problem of pancreatic cancer.

A comprehensive list of proteins secreted by the tumor microenvironment was compiled in the first aim. Culture of human pancreatic cell lines, both cancerous and stromal, was used to simulate the tumor microenvironment in vitro and produce a proteome rich in potential biomarkers. The proteins identified in this analysis demonstrated the complex signals pancreatic tumor cells use to communicate with their environment and might encompass therapeutic targets and biomarkers.

To measure the levels of these proteins in human serum, a relative quantitative LC-MS method was developed in the second aim. Traditional bioanalytical techniques were incorporated to enhance the utility of the method. Multiple reaction monitoring (MRM)/MS provided sensitivity and specificity to the method and stable isotope dilution (SID) provided correction for losses during sample preparation and analysis that enabled precise quantitation. The multiplexed method efficiently determined the concentration of over 70 proteins in a single analysis.

In the third aim, the relative quantitative method was used to compare a panel of putative protein biomarkers in human pancreatic cancer and control serum. This analysis identified 15 differential proteins with the ability to discriminate pancreatic cancer subjects from their age-matched controls with sensitivity and specificity superior to any other reported pancreatic cancer biomarkers. Future validation of this panel could change the way we diagnose pancreatic cancer.