Muscle-invasive bladder cancer is an aggressive disease that leads to death in approximately half of patients despite optimal multimodal therapy. The mainstay of treatment in bladder cancer is surgical resection with or without multi-agent chemotherapy, based wholly on staging. This project aimed at utilizing whole transcriptome profiling technology to develop and validate specific molecular signatures of bladder cancer progression that can be applied to clinical decision-making. A final signature and model has been selected and evaluated on the testing set, alone and in combination with clinical variables. The selected genomic classifier for prediction of progression after radical cystectomy comprises 89 markers corresponding to 80 genes. This classifier was able to stratify patients into those who would or would not develop recurrence. It will be further developed into a robust clinical tool by validating its performance in independent patient cohorts.