Antibody Microarray Interrogation of Tissue and Plasma for the Improved Early Detection of Pancreas Cancer

Abstract

Pancreas cancer, most frequently pancreatic ductal adenocarcinoma (PDA), is the tenth most common cancer but is the fourth leading cause of cancer deaths in the United States, with a fiveyear survival rate of just 6%. This is due to its asymptomatic nature and resistance to current treatment regimens. The majority of patients present with unresectable disease upon diagnosis of PDA, and, even with chemotherapy, have a median survival of only 6-11.1 months. Those who are diagnosed with resectable disease and undergo surgery, the only current potential curative option, have improved five-year survival rates but will also still eventually succumb to recurrent metastatic disease. Thus, efforts are being directed towards both developing better early detection strategies and improving treatments of advanced disease. The focus of this doctoral dissertation was to find biomarkers, or measurable biological characteristics, that can be used for early detection of pancreas cancer. I utilized our laboratory's experience in the design and fabrication of antibody microarrays to create a pancreas cancer-specific platform consisting of >4000 unique features to identify protein biomarkers. We have also used a variation on our platform to identify and preliminarily validate markers in pre-diagnostic triple negative breast cancer plasma samples. In efforts to find proteins that could serve as true early detection biomarker of PDA, I took a dual approach, interrogating the primary tumor tissue and plasma proteome of a highly faithful mouse model of pancreas cancer as well as mining human plasma samples collected from a cohort of women before they were diagnosed with and eventually succumbed from PDA. Using this approach, I identified and validated a panel of 7 tissue markers in the mouse model, which were also able to distinguish early stages of pancreas cancer from chronic pancreatitis. This included STK4, a novel biomarker of PDA that is up regulated at preinvasive disease stages and that I show is also up regulated in human PDA. Furthermore, the dynamic expression pattern of STK4 at early disease stages suggests it may functionally contribute to pancreas cancer progression. Plasma biomarkers discovered in the mouse model were cross-referenced with biomarkers identified in pre-diagnostic human plasma samples to further validate and better define their utility as markers for human PDA. These parallel array experiments yielded a panel of 3 proteins, ERBB2, TNC and ESR1, all of which have been previously identified as markers of PDA in samples collected at diagnosis, and that thus warrant further validation in patients at high risk of developing pancreas cancer. Collectively, these antibody microarray experiments identified promising plasma and tissue markers that call for further follow up as potential diagnostics and could help us gain insight into the molecular processes underlying disease progression.