The overall objective of this proposal is to investigate circular extrachromosomal DNA (ecDNA) as a potential
driver of intratumoral heterogeneity and treatment resistance in medulloblastoma, the most common pediatric
malignant brain tumor. Intratumoral heterogeneity is one of the leading determinants of therapeutic resistance
and treatment failure and one of the main reasons for poor overall survival in cancer patients. However, the
functional relevance of ecDNA as a driver of tumor heterogeneity and treatment resistance in medulloblastoma
has hardly been studied. To analyze the clinical impact of ecDNA in the different molecular subgroups of
medulloblastoma, we assembled a multi-institutional cohort of Whole Genome Sequencing data from 468
medulloblastoma patient samples. Using novel computational methods for the detection and reconstruction of
ecDNA, we found ecDNA in 82 patients (18%) and observe that the presence of ecDNA is associated with
significantly poorer outcomes. In addition, we find that individual medulloblastoma tumors often harbor multiple
variants of ecDNA, each containing different amplified oncogenes along with co-amplified non-coding
regulatory DNA (‘enhancers’). Based on our preliminary results, we propose the central hypothesis that ecDNA
drives intratumoral heterogeneity and treatment resistance in high-risk medulloblastoma patients. The central
hypothesis will be tested through the following three specific aims: To investigate the molecular evolution of
ecDNA as a potential driver of treatment resistance (Aim 1); To evaluate combinatorial therapies targeted
against mechanisms of ecDNA formation and clustering to reduce treatment resistance (Aim 2); To probe
medulloblastoma tumor-dependencies by functional inhibition of coding and non-coding regulatory DNA co-
amplified on ecDNA (Aim 3). The research proposed in this application has technical, conceptual, and
biological innovations, including the analysis of ecDNA on the single-cell level using novel imaging and
multiome single-nucleus sequencing methods in medulloblastoma tumors and in patient-derived xenograft
(PDX) models. The proposed research is significant, because the current standard treatment for children with
medulloblastoma causes developmental disorders, neurological damage, and secondary metastases. Novel
therapeutic approaches are urgently needed. Our approach will test the impact of standard-of-care treatments
on the molecular evolution of ecDNA and functionally test novel combination treatments targeted against
ecDNA genesis and clustering. These preclinical studies have the potential to uncover novel mechanisms by
which ecDNA contributes to the pathogenesis of medulloblastoma and to identify new scientific leads for the
development of improved treatments. We expect that our studies will expose the contribution of ecDNA
variants to the emergence of therapy resistance, reveal their selection advantages, validate recently described
properties of ecDNA and their therapeutic susceptibilities, and identify novel tumor-dependency genes
amplified on ecDNA in some of the most aggressive medulloblastoma tumors.
