Identification and validation of a glycolysis-related gene signature for predicting metastasis and survival rate in patients with thyroid cancer

Background: Thyroid cancer (THCA) is one of the most commonly malignant endocrine tumors worldwide. This study aimed to explore new gene signatures to better predict the metastasis and survival rate of patients with THCA. Methods: mRNA transcriptome date and clinical characteristics of THCA were obtained from the Cancer Genome Atlas (TCGA) database to identify the expression and prognostic implications of glycolysisrelated genes. Gene Set Enrichment Analysis (GSEA) of differentiated expressed genes was performed, and relationship between genes and glycolysis was observed in the Cox proportional regression model. Using the cBioPortal, mutations were subsequently identified in model genes. Results: A 3-gene (HSPA5, KIF20A and SDC2) signature based on glycolysis related genes was identified and used to predict metastasis and survival rate in patients with THCA. Further expression analysis revealed that KIF20A was a poor prognostic gene, while HSPA5 and SDC2 were good prognostic genes. Evaluating the prognosis of patients with THCA could be more effective with this model. Conclusions: The study reported a three-gene signature of THCA, including HSPA5, KIF20A and SDC2, which were found to be closely correlated with the glycolysis of THCA, and it showed a high efficacy to the prediction of metastasis and survival rate of THCA.