Assessment of Colorectal Cancer Risk Factors through the Application of Network-Based Approaches in a Racially Diverse Cohort of Colon Organoid Stem Cells

SIMPLE SUMMARY: Risk factors for colorectal cancer (CRC) include age, body mass index, race, smoking history, and sex. However, molecular mechanisms underlying these associations remain unclear. To better define transcriptional mechanisms impacting risk, we hypothesized that stem-cell-enriched colon organoids derived from the healthy epithelium of a diverse, average-risk patient population would provide insight into genes associated with these risk factors. We used network-based approaches to better define transcriptional mechanisms involved in CRC risk and incorporated external, publicly available datasets to prioritize likely biological drivers. ABSTRACT: Numerous demographic factors have been associated with colorectal cancer (CRC) risk. To better define biological mechanisms underlying these associations, we performed RNA sequencing of stem-cell-enriched organoids derived from the healthy colons of seven European Americans and eight African Americans. A weighted gene co-expression network analysis was performed following RNA sequencing. Module–trait relationships were determined through the association testing of each module and five CRC risk factors (age, body mass index, sex, smoking history, and race). Only modules that displayed a significantly positive correlation for gene significance and module membership were considered for further investigation. In total, 16 modules were associated with known CRC risk factors (p < 0.05). To contextualize the role of risk modules in CRC, publicly available RNA-sequencing data from TCGA-COAD were downloaded and re-analyzed. Differentially expressed genes identified between tumors and matched normal-adjacent tissue were overlaid across each module. Loci derived from CRC genome-wide association studies were additionally overlaid across modules to identify robust putative targets of risk. Among them, MYBL2 and RXRA represented strong plausible drivers through which cigarette smoking and BMI potentially modulated CRC risk, respectively. In summary, our findings highlight the potential of the colon organoid system in identifying novel CRC risk mechanisms in an ancestrally diverse and cellularly relevant population.