Colorectal cancers (CRC) are the second leading cause of cancer and cancer death in adult Americans. A number of research efforts are ongoing trying to elucidate the cellular and molecular mechanisms that cause the onset of CRC, and its progression through increasingly more severe pathological stages (0-IV). We are using a proteomics approach to further our understanding of the mechanistic cause of CRC, and to potentially identify biomarkers that indicate the stage, severity, and prognosis of the disease. A pathologic database at the Case Cancer Center contains matched (normal/tumor) tissue samples collected during surgery from an extensive variety of patients diagnosed with varying stages of CRC. As an initial attempt to use discovery proteomics to examine CRC, we obtained twelve matched tissue samples (normal/stage IV) and performed differential gel electrophoresis (2D-DIGE) to evaluate significant changes in the expression of proteins common to both the normal and diseased tissues. 58 spots are differentially expressed (± 50%) between normal and cancer tissue with statistical significance (p ≤ 0.05). These proteins have been identified by mass spectrometry, and the involved metabolic and cell signaling pathways will be reported. Such pathways may present new targets for adjuvant chemotherapy. Using the protein abundance values on the gels, a supervised learning technique is under development to train a support vector machine (SVM), which in turn could be used to classify an electrophoretic gel pattern as “cancer” or “no-cancer.”

98th AACR Annual Meeting-- Apr 14-18, 2007; Los Angeles, CA