Using the TCGA Database to Predict and Analyze Tumor Microenvironment Genes Related to Poor Prognosis of Colon Cancer.

BACKGROUND Colon cancer (COAD) is a highly malignant gastrointestinal cancer. The existence of the TCGA database allows us to more easily perform gene expression profiling and data mining on colon cancer patients worldwide, and to more easily discover the correlation between genes and survival prognosis of colon cancer. Related reports show that the degree of infiltration of tumor immune cells and stromal cells in tumor microenvironment cells has a significant impact on the prognosis of cancer patients. MATERIAL AND METHODS The immune and stromal components in colon cancer can be quantitatively analyzed using relevant scores obtained by use of the ESTIMATE calculation method. To better explain the effect of relevant genes of cells associated with immunity and stroma on the survival prognosis of colon cancer, we divided the data from 191 downloaded case into high and low groups according to their scores of immunity and stroma, and identified differentially expressed genes. RESULTS The results showed that immune and stromal scores were significantly associated with survival prognosis. After performing biological function enrichment analysis and protein interaction network on the target genes, the results showed that these genes are mainly involved in inflammatory response, immune response, and chemotaxis. We then performed relevant survival prognosis analysis of these genes. CONCLUSIONS We found a number of genes that possess the properties of tumor immune microenvironment and can predict poor prognosis of colon cancer.

Prediction and identification of immune genes related to the prognosis of patients with colon adenocarcinoma and its mechanisms.

BACKGROUND: Colon adenocarcinoma (COAD) is a gastrointestinal tumor with a high degree of malignancy. Its deterioration process is closely related to the tumor microenvironment, and transcription factors (TF) play a regulatory role in this process. Currently, there is a lack of exploration between the genes related to the COAD tumor microenvironment and the survival prognosis of patients. Models composed of multiple genes usually predict the survival prognosis of patients more accurately than single genes. We can analyze the multigene models that can predict the prognosis of COAD from the current database. METHODS: The limma package of the R programming language is used for gene differential expression analysis. Kaplan-Meier curve is used to analyze the relationship between the patient risk score model and survival data. The hazard model is used to analyze the relationship between the risk score and the clinical data of COAD patients. The information of immune genes and immune cells is obtained from IMMPORT database and TIMER database. Receiver operating characteristic (ROC) curve is used to judge the stability of the model. RESULTS: We found 7 immune genes, which can built a risk score model to predict the survival prognosis of COAD. According to univariate and multivariate analysis, the risk score can be used as an independent predictor. The content of some immune microenvironment cells will also increase as the risk score increases. CONCLUSIONS: We found 7 immune genes, such as SLC10A2 (solute carrier family 10 member 2), CXCL3 (C-X-C motif chemokine ligand 3), IGHV5-51 (immunoglobulin heavy variable 5-51), INHBA (inhibin subunit beta A), STC1 (stanniocalcin 1), UCN (urocortin), and OXTR (oxytocin receptor), can constitute a model for predicting the prognosis of COAD. They may provide potential therapeutic targets for clinical treatment of COAD.