Multi-omics approaches for diagnosis, prognosis and response to treatment of colorectal cancer.

Abstract

The high heterogeneity of colorectal cancer (CRC) in the disease biology, therapy response, and prognosis has become evident in the recent few years. For this reason, an extensive disease stratification is required. The advent of Big Data analysis in medical research has revolutionized the traditional hypothesis-driven approach. Big Data analysis provides an invaluable opportunity to improve individual and public health. In fact, the availability of large databases to capture and store the genomic landscape of patients with CRC, provides information on the genes that are frequently deregulated in CRC. Moreover, the possibility of using gene-expression profiling and highly sensitive next-generation sequencing analyses to differentiate the subtypes of CRC into prognostic groups can also lead to a better understanding of adequate CRC treatment, improving prognosis and patients’ quality of life. The aim of this study, conducted in collaboration with the Italian Institute for Genomic Medicine (Turin, Italy) and the Department of Computer Science, University of Turin (Turin, Italy), was to elucidate relationships between multiple relevant markers identified by a multi-omics approach (coding and non-coding transcriptome, metagenomics, and mutational status) investigated in different types of biospecimens from the same subjects (stool, plasma, primary tissue) for the diagnosis and prognosis of patients with CRC, identifying specific signatures. Samples collected from patients with diagnosis of CRC, recruited in our outpatient clinic and undergoing surgery during the study period were analyzed. Furthermore, the demographic characteristics as well as the clinical-pathological information obtained at both pre- (imaging, i.e. TC-scan, MRI, and colonoscopy) and post-operative (histopathological examination) diagnosis were related to the clinical outcomes of patients. Patients were regularly followed-up in accordance to current guidelines. Lastly, these data were compared with those obtained in our previous cross-sectional studies including CRC patients, healthy controls and subjects with different types of polyps. The Omics techniques used were small RNA-sequencing (sRNA seq), RNA-sequencing (RNA-seq), shotgun metagenomics sequencing and genomic profile based on Next Generation Sequencing (NGS) assay (for the mutational and MSI status). In this PhD project we have obtained several results. Firstly, we demonstrated that fecal miRNome analysis identified a predictive signature accurately discriminating CRC and precancerous lesions for a non-invasive diagnosis aimed at improving the effectiveness of current screening programs. Secondly, we pointed out the crucial role of the altered expression of 8q24-related microRNAs (miRNAs) for the initiation and/or progression of cancer as well as the correlation with the consensus molecular subtypes (CMS) classification system. Lastly, miRNA profiles in stool may reflect common traits and lifestyle habits and should be considered in relation to disease and association studies based on faecal miRNA expression.