Development and validation of a molecular prognostic index of bladder cancer based on immunogenomic landscape analysis

Abstract

Background

Bladder cancer (BCa) is one of the important tumors that have been proven to be treatable with immunotherapy. This study aims to identify and validate a molecular prognostic index of BCa based on immunogenomic landscape analysis.

Methods

The cancer genome atlas (TCGA) database and immunology database and analysis portal (ImmPort) database were used to identified differentially expressed immune-related genes (IRGs). Prognostic IRGs were screened and protein–protein interaction (PPI) network was constructed. Multivariate Cox analysis was performed to develop a molecular prognostic index of BCa. Internal and external validation were then performed in TCGA cohort and GEO cohort, respectively. Besides, we also explore the relationship between this index and clinical characteristics, immune cell infiltration and tumor microenvironment.

Results

A total of 61 prognostic IRGs were identified and a molecular prognostic index was developed. The top four hub genes included MMP9, IGF1, CXCL12 and PGF. The difference in overall survival between high-risk group and low-risk group was statistically significant. The area under curve of the receiver operating characteristic (ROC) curve was 0.757, suggesting the potential for this index. Besides, Internal validation using TCGA cohort and external validation using GEO cohort indicated that this index was of great performance in predicting outcome. T cells CD8, T cells CD4 memory activated, T cells follicular helper, macrophages M0, macrophages M2 and neutrophils were significantly associated with prognosis of BCa patients. Female, high grade, stage III&IV, N1-3 and T3-4 were associated significantly with higher risk score compared with male, low grade, stage I&II, N0 and T1-2, respectively. High risk score had a positive association with higher stromal score and ESTIMATE score while high risk score had a negative association with tumor purity.

Conclusions

This study identified several prognostic immune-related genes of clinical value. Besides, we developed and validated a molecular index based on immunogenomic landscape analysis, which performed well in predicting prognosis of BCa. Further researches are needed to verify the effectiveness of this index and these vital genes.