Gastric cancer is one of the most common malignant tumors of the digestive tract. However, there are no adequate prognostic markers available for this disease. The present study used bioinformatics to identify prognostic markers for gastric cancer that would guide the clinical diagnosis and treatment of this disease.
Gene expression data and clinical information of gastric cancer patients along with the gene expression data of 30 healthy samples were downloaded from the TCGA database. The initial screening was performed using the WGCNA method combined with the analysis of differentially expressed genes, which was followed by univariate analysis, multivariate COX regression analysis, and Lasso regression analysis for screening the candidate genes and constructing a prognostic model for gastric cancer. Subsequently, immune cell typing was performed using CIBERSORT to analyze the expression of immune cells in each sample. Finally, we performed laboratory validation of the results of our analyses using immunohistochemical analysis.
After five screenings, it was revealed that only three genes fulfilled all the screening requirements. The survival curves generated by the prognostic model revealed that the survival rate of the patients in the high-risk group was significantly lower compared to the patients in the low-risk group (P-value < 0.001). The immune cell component analysis revealed that the three genes were differentially associated with the corresponding immune cells (P-value < 0.05). The results of immunohistochemistry also support our analysis.
CGB5, MKNK2, and PAPPA2 may be used as novel prognostic biomarkers for gastric cancer.