Construction of prognosis model of bladder cancer based on transcriptome

• Published in: Zhejiang da xue xue bao. Journal of Zhejiang University. Medical sciences. Yi xue ban Vol. 51; no. 1; pp. 79 - 86
• Main Authors: Chen, Qiu, Cai, Liangliang, Liang, Jingyan
• Format: Journal Article
• Language: English
• Published: 25.02.2022
• DOI: 10.3724/zdxbyxb-2021-0368
• ISSN: 1008-9292

OBJECTIVETo screen for prognosis related genes in bladder cancer, and to establish prognosis model of bladder cancer. METHODSThe clinical information and bladder tissue RNA sequencing data of 406 bladder cancer patients, and the bladder tissue RNA sequencing data of 28 healthy individuals were downloaded from The Cancer Genome Atlas (TCGA) database, Genotype-Tissue Expression (GTEx) database through the UCSC Xena platform. The weighted gene co-expression network analysis (WGCNA), univariate Cox regression, LASSO regression analysis and multivariate Cox regression analysis were used to screen the prognosis-related genes of bladder cancer and the prognostic model was established. The prognostic model was evaluated with receiver operator characteristic curve (ROC curve). RESULTSA total of 2308 differentially expressed genes related to bladder cancer were obtained from the analysis. Six gene modules were obtained by WGCNA, and 829 genes with significant effect on bladder cancer prognosis were screened out. Univariate Cox regression and LASSO regression analysis showed that 24 genes were related to the prognosis of bladder cancer patients. Multivariate Cox regression analysis revealed 9 genes as independent predictors in training set, namely ADCY9, MAFG_DT, EMP1, CAST, PCOLCE2, LTBP1, CSPG4, NXPH4, SLC1A6, which were used to establish the prognosis model of bladder cancer patients. The 3-year survival rates of the high-risk group and the low-risk group in the training set were 31.814% and 59.821%, respectively. The 3-year survival rates of the high-risk group and the low-risk group in the test set were 32.745% and 68.932%, respectively. The areas under the ROC curve of the model for predicting the prognosis of bladder cancer patients in both the training set and the test set were above 0.7. CONCLUSIONThe established model in this study has good predictive ability for the survival of bladder cancer patients.