Abstract
Gene mutations play an important role in tumor progression. This study aimed to identify genes that were mutated in colorectal cancer (CRC) and to explore their biological effects and prognostic value in CRC patients. We performed somatic mutation analysis using data sets from The Cancer Genome Atlas and International Cancer Genome Consortium, and identified that FREM2 had the highest mutation frequency in patients with colon adenocarcinoma (COAD). COAD patients were divided into FREM2-mutated type (n = 36) and FREM2-wild type (n = 278), and a Kaplan-Meier survival curve was generated to perform prognostic analysis. A FREM2-mutation prognosis model was constructed using random forest method, and the performance of the model was evaluated using receiver operating characteristic curve. Next, the random forest method and Cox regression analysis were used to construct a prognostic model based on the gene expression data of 36 FREM2-mutant COAD patients. The model showed a high prediction accuracy (83.9%), and 13 prognostic model characteristic genes related to overall survival were identified. Then, the results of tumor mutation burden (TMB) and microsatellite instability (MSI) analyses revealed significant differences in TMB and MSI among the risk scores of different prognostic models. Differentially expressed genes were identified and analyzed for functional enrichment and immune infiltration. Finally, 30 samples of CRC patients were collected for immunohistochemical staining to analyze the FREM2 expression levels, which showed that FREM2 was highly expressed in tumor tissues. In conclusion, CRC patients had a high level of FREM2 mutations associated with a worse prognosis, which indicated that FREM2 mutations may be potential prognostic markers in CRC.