Bladder cancer is the fifth most common cancer in the US and the fourth most common in men. Radical surgery remains the standard of care for patients with high grade, muscle-invasive disease but despite multi-modality treatment, approximately half of such patients develop metastatic disease, which is with rare exception fatal. Here, a custom next generation sequencing assay will be employed to define the spectrum of co-mutational events in muscle-invasive bladder cancers with a focus on defining the prevalence and prognostic relevance of mutations in the PI3 kinase/AKT/mTOR pathway. To avoid selection bias, this analysis will be performed using a large, prospectively collected, sequential cohort of patients with muscle-invasive disease undergoing radical cystectomy. A field cancerization effect is observed in patients with bladder cancer whereby multiple primary tumors develop within the urinary tract. To explore a genetic basis for this phenomenon, the analysis will be extended by comparing the genomic profile of normal appearing bladder epithelium to primary tumors to matched metastatic lymph nodes and distant metastatic sites. One goal of these studies will be to determine the temporal sequence of mutational events in bladder cancer with a focus on the timing of PI3 kinase alterations in disease progression. Functional studies will focus on genes that are commonly co-mutated with PI3 kinase pathway alterations to identify aberrations that enhance or abrogate tumor invasion and/or PI3 kinase and mTORC1-dependence. Finally, preliminary genomic data indicate that PI3 kinase pathway alterations are common and occur in a mutually exclusive pattern in patients with bladder cancer, suggesting overlapping functional effects. To directly compare the functional consequences of PTEN and TSC1 loss in bladder cancer in depth, we will compare the phenotype of genetically engineered mouse (GEM) models with conditional inactivation of the Pten and Tsc1 genes in the bladder epithelium. Mice with bladder specific and inducible expression of shRNAs will also be generated to determine whether continued suppression of Pten and/or p53 is required for tumor maintenance in mice with established tumors. The long-term objective will be to develop GEM mice that model the pattern of co-mutations identified in human bladder cancer with the goal of using these mice to understand the contribution of specific genomic alterations to bladder cancer progression and as models to study novel therapeutic strategies.