Nine glycolysis-related gene signature predicting the survival of patients with endometrial adenocarcinoma

Cancer Cell International

SiYue Li¹  JinHui Liu¹  Jing Yang¹  WenJun Cheng¹  SiPei Nie¹  HuangYang Meng¹  Rui Sun¹  Gao Feng² 

[1] Department of Gynecology, The First Affiliated Hospital of Nanjing Medical University;Department of Orthopedic Surgery, The First Affiliated Hospital of Nanjing Medical University;

关键词: Endometrial cancer;    Glycolysis;    Prognostic model;    TCGA;    GSEA;   

DOI  :  10.1186/s12935-020-01264-1

来源: DOAJ

【 摘 要 】

Abstract Background Endometrial cancer is the fourth most common cancer in women. The death rate for endometrial cancer has increased. Glycolysis of cellular respiration is a complex reaction and is the first step in most carbohydrate catabolism, which was proved to participate in tumors. Methods We analyzed the sample data of over 500 patients from TCGA database. The bioinformatic analysis included GSEA, cox and lasso regression analysis to select prognostic genes, as well as construction of a prognostic model and a nomogram for OS evaluation. The immunohistochemistry staining, survival analysis and expression level validation were also performed. Maftools package was for mutation analysis. GSEA identified Glycolysis was the most related pathway to EC. qRT-PCR verified the expression level of hub gene in clinical samples. Results According to the prognostic model using the train set, 9 glycolysis-related genes including B3GALT6, PAM, LCT, GMPPB, GLCE, DCN, CAPN5, GYS2 and FBP2 were identified as prognosis-related genes. Based on nine gene signature, the EC patients could be classified into high and low risk subgroups, and patients with high risk score showed shorter survival time. Time-dependent ROC analysis and Cox regression suggested that the risk score predicted EC prognosis accurately and independently. Analysis of test and train sets yielded consistent results A nomogram which incorporated the 9-mRNA signature and clinical features was also built for prognostic prediction. Immunohistochemistry staining and TCGA validation showed that expression levels of these genes do differ between EC and normal tissue samples. GSEA revealed that the samples of the low-risk group were mainly concentrated on Bile Acid Metabolism. Patients in the low-risk group displayed obvious mutation signatures compared with those in the high-risk group. The expression levels of B3GALT6, DCN, FBP2 and GYS2 are lower in tumor samples and higher in normal tissue samples. The expression of CAPN5 and LCT in clinical sample tissues is just the opposite. Conclusion This study found that the Glycolysis pathway is associated with EC and screened for hub genes on the Glycolysis pathway, which may serve as new target for the treatment of EC.

【 授权许可】

Unknown