Abstract: Objective To establish a bioinformatics model for evaluating the prognosis of bladder cancer by immune genes,and screen the biomarkers influencing the occurrence and prognosis of bladder cancer. Methods The transcriptome data and clinicopathological data of bladder cancer and adjacent normal tissues were downloaded from the Cancer Genome Atlas database,immune gene data were downloaded in the ImmPort database,and the differentially expressed immunogenes were screened.Univariate and multivariate Cox regression models were used to analyze the prognosis of bladder cancer by differentially expressed immune genes.The correlation between risk score,clinical and pathological features of bladder cancer and prognosis of bladder cancer were further analyzed. Results Totally 260 differentially expressed bladder tumor-related immune genes were screened out based on "Wilcox.Test" in R language.Univariate Cox analysis showed that 25 differentially expressed immune genes were related to the prognosis of bladder tumor (P<0.01).Then the genes were inputted in the multivariate Cox regression analysis,and a prognosis model including 11 immune genes (TAP1,MMP9,ADIPOQ,PDGFRA,AHNAK,KCNH2,OLR1,RAC3,NAMPT,PPY,SH3BP2) differentially expressed in bladder cancer was achieved.The samples were divided into high risk group and low risk group by the median value of the model risk score.Kaplan-meier survival curve results showed that the 5-year overall survival rate (OS) of the high risk group was 28.6%,and that of the low risk group was 59.7%.The ROC curve analysis confirmed the accuracy of the prediction model (AUC=0.732).The clinicopathological features were included in Cox regression analysis,and the results showed that the high risk score of the immune gene prognostic assessment model was an independent predictor of the prognosis of patients with bladder cancer. Conclusion The bladder cancer prognosis model based on immune genes is reliable,which is expected to help clinical physicians to predict the prognosis of bladder cancer.