Prostate cancer is characterized by large genomic rearrangements and deletions. We show that the genes CHD1 and MAP3K7 are co-deleted in ERG translocation negative prostate cancer. To demonstrate a functional cooperativity we used a novel mouse prostate stem cell developmental model and showed that collaborative loss of CHD1 and MAP3K7 promotes an aggressive prostate cancer phenotype with altered lineage differentiation, abnormal secretory products, massive nuclear atypia, loss of E-cadherin and enrichment in neuronal and neuroendocrine markers. Profound alterations in AR expression were also observed. Using multiple human cell line models we also demonstrate that loss of CHD1 and MAP3K7 promotes castrate-resistant prostate cancer. This project will evaluate downstream targets of AR altered in tumors with loss of MAP3K7 and CHD1 using functional genomics in vitro and in animal models and assess the clinical impact of these targets on patient outcome. This work could have impact on the management of the most aggressive prostate cancer. Co-deletion of CHD1 and MAP3K7 occurs in 10-15 % of primary tumors. Relapse occurs in approximately 50% of patients with co-deletion. If these deletions occur in primary tumors and predict poor survival, men could be stratified based on MAP3K7 and CHD1 status. A functional understanding of this variant of prostate cancer could lead to novel therapeutic targeting strategies in the future.