[Objective] To investigate the clinical utility of ATP-binding cassette transporter family in immunotherapy for bladder cancer based on IMvigor210 and UNC-108 (GSE176307) datasets. [Methods] Gene expression data of 348 patients with bladder urothelial carcinoma were downloaded from the IMvigor210 database.Firstly, consensus clustering was performed to the gene expression levels of the ATP-binding cassette transporter family, resulting in two clusters: Cluster 1 and Cluster 2.Survival analysis was conducted between the two clusters.Next, univariate Cox regression was employed to identify ATP-binding cassette transporter family genes significantly affecting prognosis.A predictive model was constructed using the random survival forest algorithm to predict treatment response and survival outcomes in patients with bladder cancer receiving immunotherapy.The model's accuracy was validated with UNC-108 dataset.Then, ESTIMATE and ssGSEA were applied to analyze differences in the tumor microenvironment.Furthermore, oncoPredict algorithm was used to predict the sensitivity of patients to cisplatin. [Results] A total of 9 ATP-binding cassette transporter family genes were selected to construct the ABC score predictive model.In the IMvigor210 datasets, the Kaplan-Meier analysis revealed that patients with low ABC score had significantly longer overall survival compared to those with high ABC score (P<0.001). The ABC score model demonstrated good performance with area under the ROC curve (AUC) of 0.80, 0.87, and 0.88 for predicting 0.5-year, 1-year, and 1.5-year survival, respectively.The predictive ability of the ABC score for immunotherapy response, with an AUC of 0.78, outperformed that of TMB (AUC: 0.72) and PD-L1 (AUC: 0.58). The ABC score also exhibited favorable predictive performance in the UNC-108 validation cohort.ABC score was also significantly correlated with tumor microenvironment immune scores, various immune cells, and the expression of immune checkpoint genes. [Conclusion] ATP-binding cassette transporter family is closely associated with the tumor microenvironment and the efficacy of immunotherapy for bladder cancer, making it a potential novel biomarker for immunotherapy.