Finding Biomarkers of Metastatic Potential in Colorectal Cancer

Tese de mestrado, Bioinformática e Biologia Computacional, 2023, Universidade de Lisboa, Faculdade de CiênciasColorectal cancer (CRC) is among the most common cancers and globally one of the deadliest. Treatments for this disease have been improving, increasing the survival rates for CRC patients. On the contrary, when the cancer progresses and metastasis occurs, the overall survival rates are quite low, reflecting the role of metastasis as the leading cause of death. CRC is a very heterogeneous disease that can be classified in various types depending on how it develops, and in which are involved many different pathways. Despite its complexity, CRC molecular mechanisms are each year being better understood, including events like cell proliferation, immune surveillance blockage, cell adhesion disturbances. These mechanisms may be related to cancer progression from primary to metastatic. Biomarkers are molecules found in tissue, blood, or stool samples and have been used to identify diseases like cancer. These markers have proved to be very beneficial in the aid of CRC treatment and other cancers. Despite that, the identification of reliable biomarkers for metastatic CRC remains poor, mainly biomarkers that predict metastasis development. Here we uncovered promising biomarkers in early stages that reflect the tumour potential to evolve from primary to metastasis. Bioinformatic analyses were conducted with transcriptomic data to investigate gene expression differentiation, gene sets enrichment, and to create a predictive model. Genomic data was also analysed to find correlations between mutations and metastasis occurrence, although this last step was inconclusive. The most differential expressed genes found have been also identified to be related to metastasis in other studies and the same happened with some of the enriched pathways. The predictive model had an insufficient efficacy but revealed to be promising. Here we showed the impact that gene expression analysis can have in the important field of biomarker research and the need of future studies with this type of data