Abstract
BackgroundBreast cancer is a malignant tumor with high morbidity and mortality, which seriously endangers the health of women around the world. Biomarker-based exploration will be effective for better diagnosis, prediction and targeted therapy.ObjectiveTo construct biomarker models related to glycolysis and gluconeogenesis in breast cancer.MethodsThe gene expression of 932 breast cancer patients in the Cancer Genome Atlas (TCGA) database was analyzed by Gene Set Variation Analysis (GSVA) using glycolysis and gluconeogenesis-related pathways. Differential expression genes were searched for by the T-test. Univariate Cox proportional hazards model (COX) regression, Least Absolute Shrinkage and Selection Operator (LASSO) regression, and Multivariate COX regression were used to find clinically significant genes for prognostic survival. After that, the constructed gene signature was externally validated through the Gene Expression Omnibus (GEO). Finally, a nomogram was constructed to predict the survival of patients. In addition, analyzing the role of biomarkers in pan-cancer.ResultsA risk scoring model associated with glycolysis and gluconeogenesis was developed and validated. A nomogram was created to predict 2-, 3-, and 5- survival.ConclusionsThe predictive model accurately predicted the prognosis of breast cancer patients.