Mutation patterns in colorectal cancer and their relationship with prognosis

Colorectal cancer (CRC) is a prevalent malignancy and a leading cause of cancer-related mortality. Extensive research into the aetiology of CRC has revealed that somatic mutations in certain genes play a crucial role in CRC development.AIM: In this study, we utilized data from public databases to investigate prevalent mutation patterns in CRC and developed a prognostic predictive model for CRC patients based on mutant genetic characteristics and other relevant clinical features.

We initially gathered mutation information from CRC patients by analysing data from 15 datasets to identify genes with a mutation frequency of ≥10 %. Next, log-rank analyses were used to determine the relationship between prognosis and the mutational status of the most commonly mutated genes; the SIGnaling database was utilized to generate a protein‒protein interaction network. We consolidated and classified the gene mutation patterns of CRC patients in the database based on frequently mutated genes related to prognosis. A predictive nomogram was constructed, including age, sex, TNM stage, and mutation partner, based on available clinical, mutational, and prognostic information for CRC patients at our institution. Finally, the reliability of the model was verified using time-dependent ROC curve analysis.

The top 7 genes somatically mutated ≥10 % in 4477 samples from 4255 patients were TP53 (67 %), APC (66 %), KRAS (43 %), PIK3CA (18 %), FBXW7 (14 %), SMAD4 (14 %), and BRAF (10 %). Log-rank analysis demonstrated that the mutation status of 5 genes, namely, TP53, APC, PIK3CA, SMAD4, and BRAF, correlated significantly with prognosis. Protein‒protein interaction analysis confirmed functional interactions between these 5 genes, implicating them in tumorigenesis. We exhaustively enumerated the mutation patterns involving these five genes in 4255 patients, resulting in identification of 32 mutational patterns. After consolidation and classification, these patterns were divided into 3 grades based on patient prognosis. Next, a predictive nomogram based on the clinical, mutational, and prognostic information of 107 CRC patients treated at University Medical Center Rostock was constructed. The area under the curve (AUC) values for the model for predicting 1-, 3-, and 5-year overall survival were 0.779, 0.721, and 0.815, respectively.

Common mutational patterns based on frequently mutated genes are associated with prognosis in CRC patients. Our study provides a valuable and concise prognostic predictor for determining outcomes in patients with CRC.