Medulloblastoma is the most common malignant brain tumor in children that is treated with surgery, chemotherapy, and cranio-spinal radiation. Such treatment is associated with grim long-term side-effects including secondary malignancies. Further, these treatments are often ineffective in high-risk, infant, and recurrent medulloblastomas. Therefore, there is a critical need to develop novel therapies to treat medulloblastomas. The main objective of the dissertation is to identify vulnerabilities within the genetics and microenvironment of medulloblastomas that could be targeted therapeutically. Two major strategies were used to identify such therapeutic targets. First, next generation sequencing data from cases of medulloblastomas were used to computationally identify mutational signatures informing about underlying mutagenesis in medulloblastoma. Secondly, immune histochemical approaches were used to characterize the immune microenvironment of murine and human medulloblastomas. Broadly, the dissertation highlights three major findings. First, though medulloblastomas are hypomutated, I identified a small subgroup of hypermutated medulloblastomas. These findings indicated the need to assess mutational burden in medulloblastomas, as hypermutation is a marker for response to immune check point inhibitions. Secondly, the dissertation defines the role of homologous recombination defect in majority of medulloblastoma cases. This finding provides preliminary evidence indicating testing for homologous recombination defects in larger cohorts of medulloblastomas.

This may help direct therapy, as tumors defective in homologous recombination often respond to carboplatin therapy and PARP inhibition. Finally, immunohistochemical data concluded that a subgroup of medulloblastomas are immune-inflamed and more extensive-studies are required to test the efficacy of immunotherapy in such tumors.