Ependymomas are tumors of the brain and spinal cord. Treatment approaches and mortality rates for this disease have changed little over the last twenty years, highlighting the great need for new therapies. Histologic similarities among ependymomas have led investigators to treat these tumors as a single entity; but we have shown that ependymomas from different regions of the central nervous system include discrete clinical and molecular subtypes, suggesting they are different diseases. Thus, contemporary efforts to cure all patients with ependymoma must be concerned with understanding the biological basis of these disease subtypes, and where necessary, developing subtype-specific therapies. During the last funding cycle we developed a cross- species genomics approach that characterized genomic subtypes of human ependymoma and matched these with neural stem cells in the mouse to generate accurate models of the disease. We will build on this work to complete three Specific Aims designed to test the central hypothesis that 'Ependymoma subtypes are driven by distinct cell signals that can be targeted for therapeutic gain.' Aim 1 will employ new 'virus-pool' screens to test the in vivo oncogenic role of the top 90 candidate oncogenes and 40 tumor suppressor genes that we previously identified in ependymoma. We will thereby generate a 'clinic' of mice that recapitulate the full spectrum on human ependymoma subtypes. Aim 2 will interrogate the models developed in Aim 1 using high- throughput drug screens, kinome-wide binding and cell biology assays to pinpoint the key cell signals that maintain each ependymoma subtype that might therefore serve as therapeutic targets. Integration of these data with genetic analyses of human and mouse tumors will further validate candidate drug targets. Future clinical trials of therapies that target signals identified in Aim 2 will demand the selection of patients with the appropriate disease subtype. Therefore, Aim 3 will develop a robust Affymetrix Quantigene assay to reliably and rapidly diagnose ependymoma subtypes using formalin fixed paraffin embedded tissue. PUBLIC HEALTH RELEVANCE: The failure to progress the treatment of patients with ependymoma has resulted from a lack of preclinical models and a drug development process that does not account for the discrete subtypes that comprise the disease. Our proposal will develop highly accurate mouse models of the full spectrum of ependymoma and use these to identify robust drug targets and diagnostic tools for subtype-specific clinical trials.