Extrachromosomal DNA as a driver of heterogeneity and plasticity in glioblastoma

PROJECT SUMMARY Glioblastoma multiforme (GBM) is the most common malignant primary brain tumor and, despite incremental advances over the past few decades, has a five-year survival rate less than 10%. The failure of targeted drug therapies and this stagnation in therapeutic development is due, in part, to its characteristic heterogeneity and plasticity––“multiforme”––highlighting a critical need to understand the biological processes underlying these GBM hallmarks. Extrachromosomal DNA (ecDNA), large circular DNA elements that harbor oncogenes and enhancers, are a class of amplifications with oncogenic properties due to their unique structure. Without centromeres, ecDNA segregate randomly into daughter cells, which promotes genetic heterogeneity and exceptionally high copy number. ecDNA is present in over 50% of GBM tumors, the most of any cancer type, and the most prevalent genes found on ecDNA are well characterized cellular state drivers such as EGFR and PDGFRA. Despite the well-documented individual relationships between glioblastoma, extrachromosomal DNA, and intratumoral heterogeneity, the interplay between the three is not well understood. The overall goal of this proposal is to characterize the role of ecDNA in GBM transcriptional heterogeneity––diverse distributions of GBM cellular states––and plasticity––frequent cell state interconversions––during tumor initiation and upon drug treatment. Given its nonequal inheritance and intra-tumoral copy number variation, the central hypothesis of this proposal is that ecDNA enable cellular state heterogeneity and plasticity. By comparing engineered models that differ only in how EGFRvIII and PDGFRAΔ8–9 are inherited, Aim 1 will determine how ecDNA shapes the distribution of GBM cellular states during tumor initiation. Aim 2 will characterize the heritability and plasticity of cellular states during and after targeted drug therapy with an EGFR inhibitor in barcoded isogenic patient derived xenograft models that differ in how they amplify EGFRvIII: on ecDNA vs. chromosomes. In addition to determining whether cellular state transitions occur more frequently (i.e., high plasticity) in ecDNA+ models, this aim will also directly establish whether ecDNA enables drug resistance to targeted therapy in vivo. Heterogeneity and plasticity are thought to enlarge the total fitness landscape promoting adaptation during evolution and drug treatment, and as such, clarifying the mechanisms underlying these GBM hallmarks, as expected with this proposal, is essential. The expertise in glioma modeling present within the Furnari lab will come together with the expertise in genomics of the Ren lab to form an ideal environment to execute this research plan at UC San Diego. Through graduate coursework, a diverse mentorship team, and hands-on research, this proposal will provide the training to become a successful independent physician-scientist and domain expert in cancer genomics.