Project Summary/Abstract Solid tumors arise as the consequence of accumulation of oncogene and/or tumor suppressor mutations. How these mutations arise and accumulate in one cell over a lifetime remains a mystery. It is likely that an improved understanding of the early events that form a pre-tumorigenic cell could have implications for analysis of mature tumor cells and insights into improved therapy development. We have developed fully penetrant genetically engineered mouse models of glioblastoma multiforme (GBM) by mutation of three tumor suppressors commonly found mutated in human GBM (P53, PTEN, & NF1). Using our combined background in developmental biology and neuroscience, we have traced the origin of these tumors to the adult stem/progenitor cell population. We have developed tools to uncover functional GBM subtypes that are predicated on the tumor cell of origin rather than on specific driver mutations (Alcantara, Cancer Cell, 2015). These studies will be extended to identify cell of origin and relationship to genotype and phenotype. Moreover, using gene expression signatures from the novel mouse GBM subtypes, we have identified human GBM counterpart signatures that suggest similar biological origins and a novel strategy for human GBM molecular stratification. Our data provide evidence for additional human GBM subtypes that may also relate to novel cells of origin. The mouse models demonstrate an endogenous GBM tumor cell hierarchy placing a cancer stem cell at the apex. Our ongoing studies suggest that each of the new stratified GBM subtypes are governed by a cancer stem cell pattern of growth. We will expand and confirm these observations. Using a phenotypic high throughput small chemical compound screen we have identified small molecules that have nanomolar toxicity on primary low passage GBM derived cells but not on primary normally dividing cells such as mouse embryo fibroblasts or neonatal astrocytes. In addition, lead compounds including a benzimidazolium compound and its derivatives demonstrate toxicity on primary human GBM derived tumor spheres. These compounds hold promise for in vivo studies and identifying novel key GBM dependency pathways for therapeutic development. We are developing a comprehensive GBM patient derived xenograft program that will be employed to further validate our mouse model finding and to identify, isolate and neutralize human GBM cancer stem cells.