Tumor Expression Profile Analysis Developed and Validated a Prognostic Model Based on Immune-Related Genes in Bladder Cancer

Bingqi Dong; Jiaming Liang; Ding Li; Wenping Song; Shiming Zhao; Yongkang Ma; Jinbo Song; Mingkai Zhu; Tiejun Yang

doi:10.3389/fgene.2021.696912

Tumor Expression Profile Analysis Developed and Validated a Prognostic Model Based on Immune-Related Genes in Bladder Cancer

Front Genet. 2021 Aug 27:12:696912. doi: 10.3389/fgene.2021.696912. eCollection 2021.

Authors

Bingqi Dong¹, Jiaming Liang^{2

3}, Ding Li⁴, Wenping Song⁴, Shiming Zhao¹, Yongkang Ma¹, Jinbo Song¹, Mingkai Zhu¹, Tiejun Yang¹

Affiliations

¹ Department of Urology, Affiliated Cancer Hospital of Zhengzhou University, Henan Cancer Hospital, Zhengzhou, China.
² The Second Affiliated Hospital of Guangzhou Medical University, Guangzhou, China.
³ State Key Laboratory of Respiratory Disease, The First Affiliated Hospital of Guangzhou Medical University, National Clinical Research Center for Respiratory Disease, Guangzhou, China.
⁴ Department of Pharmacy, Affiliated Cancer Hospital of Zhengzhou University, Henan Cancer Hospital, Zhengzhou, China.

Abstract

Background: Bladder cancer (BLCA) ranks 10th in incidence among malignant tumors and 6th in incidence among malignant tumors in males. With the application of immune therapy, the overall survival (OS) rate of BLCA patients has greatly improved, but the 5-year survival rate of BLCA patients is still low. Furthermore, not every BLCA patient benefits from immunotherapy, and there are a limited number of biomarkers for predicting the immunotherapy response. Therefore, novel biomarkers for predicting the immunotherapy response and prognosis of BLCA are urgently needed. Methods: The RNA sequencing (RNA-seq) data, clinical data and gene annotation files for The Cancer Genome Atlas (TCGA) BLCA cohort were extracted from the University of California, Santa Cruz (UCSC) Xena Browser. The BLCA datasets GSE31684 and GSE32894 from the Gene Expression Omnibus (GEO) database were extracted for external validation. Immune-related genes were extracted from InnateDB. Significant differentially expressed genes (DEGs) were identified using the R package "limma," and Gene Ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis for the DEGs were performed using R package "clusterProfiler." Least absolute shrinkage and selection operator (LASSO) regression analysis were used to construct the signature model. The infiltration level of each immune cell type was estimated using the single-sample gene set enrichment analysis (ssGSEA) algorithm. The performance of the model was evaluated with receiver operating characteristic (ROC) curves and calibration curves. Results: In total, 1,040 immune-related DEGs were identified, and eight signature genes were selected to construct a model using LASSO regression analysis. The risk score of BLCA patients based on the signature model was negatively correlated with OS and the immunotherapy response. The ROC curve for OS revealed that the model had good accuracy. The calibration curve showed good agreement between the predictions and actual observations. Conclusions: Herein, we constructed an immune-related eight-gene signature that could be a potential biomarker to predict the immunotherapy response and prognosis of BLCA patients.

Keywords: Gene Expression Omnibus; The Cancer Genome Atlas; bladder cancer; immune-related signature; immunotherapy.