Identification of a Glucose Metabolism-related Signature for prediction of Clinical Prognosis in Clear Cell Renal Cell Carcinoma

Sheng Wang; Ling Zhang; Zhihong Yu; Kequn Chai; Jiabin Chen

doi:10.7150/jca.45296

Identification of a Glucose Metabolism-related Signature for prediction of Clinical Prognosis in Clear Cell Renal Cell Carcinoma

J Cancer. 2020 Jun 21;11(17):4996-5006. doi: 10.7150/jca.45296. eCollection 2020.

Authors

Sheng Wang^{1

2}, Ling Zhang^{1

2}, Zhihong Yu², Kequn Chai², Jiabin Chen²

Affiliations

¹ The Second Clinical Medical College, Zhejiang Chinese Medicine University, Hangzhou, Zhejiang.
² Department of Oncology, Tongde Hospital of Zhejiang, Hangzhou, Zhejiang 310053, P.R. China.

Abstract

Background: Clear cell renal cell carcinoma (ccRCC) is one of the most prevalent and invasive histological subtypes among all renal cell carcinomas (RCC). Cancer cell metabolism, particularly glucose metabolism, has been reported as a hallmark of cancer. However, the characteristics of glucose metabolism-related gene sets in ccRCC have not been systematically profiled. Methods: In this study, we downloaded a gene expression profile and glucose metabolism-related gene set from TCGA (The Cancer Genome Altas) and MSigDB, respectively, to analyze the characteristics of glucose metabolism-related gene sets in ccRCC. We used a multivariable Cox regression analysis to develop a risk signature, which divided patients into low- and high- risk groups. In addition, a nomogram that combined the risk signature and clinical characteristics was created for predicting the 3- and 5-year overall survival (OS) of ccRCC. The accuracy of the nomogram prediction was evaluated using the area under the receiver operating characteristic curve (AUC) and a calibration plot. Results: A total of 231 glucose metabolism-related genes were found, and 68 differentially expressed genes (DEGs) were identified. After screening by univariate regression analysis, LASSO regression analysis and multivariable Cox regression analysis, six glucose metabolism-related DEGs (FBP1, GYG2, KAT2A, LGALS1, PFKP, and RGN) were selected to develop a risk signature. There were significant differences in the clinical features (Fuhrman nuclear grade and TNM stage) between the high- and low-risk groups. The multivariable Cox regression indicated that the risk score was independent of the prognostic factors (training set: HR=3.393, 95% CI [2.025, 5.685], p<0.001; validation set: HR=1.933, 95% CI [1.130, 3.308], p=0.016). The AUCs of the nomograms for the 3-year OS in the training and validation sets were 0.808 and 0.819, respectively, and 0.777 and 0.796, respectively, for the 5- year OS. Conclusion: We demonstrated a novel glucose metabolism-related risk signature for predicting the prognosis of ccRCC. However, additional in vitro and in vivo research is required to validate our findings.

Keywords: clear cell renal cell carcinoma; glucose metabolism; signature.