Research Paper|Volume 13, Issue 2|pp 1859—1871

Development of prognostic signature based on immune-related genes in muscle-invasive bladder cancer: bioinformatics analysis of TCGA database

Kun Jin¹, Shi Qiu^1,², Di Jin¹, Xianghong Zhou¹, Xiaonan Zheng¹, Jiakun Li¹, Xinyang Liao¹, Lu Yang¹, Qiang Wei¹

¹Department of Urology, Institute of Urology and National Clinical Research Center for Geriatrics, West China Hospital, Sichuan University, Chengdu, Sichuan, China
²Center of Biomedical Big Data, West China Hospital, Sichuan University, Chengdu, Sichuan, China

* Equal contribution

Received: April 29, 2020Accepted: July 14, 2020Published: January 19, 2021

Copyright: © 2021 Jin et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Abstract

Background: Muscle-invasive bladder cancer (MIBC) with high tumor stages accounts for most bladder cancer patient mortality. Platinum-based chemotherapy provides insufficient survival benefits; however, immunotherapy is a promising option for MIBC.

Results: There were 31 differentially expressed IRGs that significantly correlated with the clinical outcomes of MIBC patients. A prognostic signature based on 12 IRGs (MMP9, RBP7, ADIPOQ, AHNAK, OAS1, RAC3, SLIT2, EDNRA, IL34, PDGFD, PPY, IL17RD) performed moderately in prognostic predictions with area under the curve (AUC) equal to 0.76. The high-risk patient group presented worse survival outcomes (hazard ratio 1.197, 95% confidence interval 1.103–1.299, p < 0.001). Furthermore, immune cell infiltration analysis showed increased tumor infiltration of macrophages in the high-risk group.

Conclusion: This novel prognostic signature can effectively divide MIBC patients into different risk groups, allowing for intensive treatment of high-risk individuals who have worse predicted survival outcomes.

Methods: Bioinformatics analyses were conducted using the Cancer Genome Atlas (TCGA) database. Differentially expressed genes and survival-associated immune-related genes (IRGs) were analyzed through a computational algorithm and Cox regression. The potential mechanisms of IRG expression were explored with transcription factors, and a prognosis classification based on IRG expression was developed to stratify patients into distinct risk groups.