Integrated Analysis of Gene Expression Data for Colon Cancer Biomarker Discovery

The identification of molecular markers with prognostic value in colorectal cancer is a challenging task that is needed to define therapeutic guidelines. Despite recent advances in the screening, diagnosis, and treatment of colorectal cancer, an estimated 608,000 people die every year from this form of cancer, which is 8% of all cancer deaths. We performed two staged integrated bioinformatics analytics on gene expression data sets of three latest developed studies of colon cancer. We identified two groups of integrated signatures from the comparison of normal versus tumor and tumor versus mets patients samples. Functional analysis of the diagnostics 267-genes shows over-representation of signaling-related molecules and also significantly involved in cancers related regulatory pathways. The metastatic 124-gene signature shows functionally involved in immune-response, lipid metabolism and PPAR signaling pathways. Kaplan-Meier estimates of 124-genes using independent data sets shows that higher grade/stage patients have significantly better overall-survival (p=0.001, HR=2.61 (CI 1.43-4.79)) and disease-specific survival rate (p=0.00, HR=2.41 (CI 1.28-4.53)) compare to low grade patients. Further biological validation of genes identified in this study may provide vital biomarker targets for colon cancers.