PROJECT SUMMARY / ABSTRACT The goal of this project is to dissect the immune evasion mechanisms and enable immunotherapy for children with high-risk Group 3 medulloblastoma (G3MB) via systems immunology approaches. Brain tumors are the leading cause of cancer-related deaths in children. Medulloblastoma is the most prevalent malignant pediatric brain tumor and is characterized by four major molecular subgroups, among which G3MB is the most aggressive form and features MYC overexpression. The immunosuppressive tumor microenvironment (TME) is poorly understood in G3MB, and no immunotherapy is available for children with this high-risk disease. Systems immunology approaches—especially single-cell and spatial multi-omics profiling and in vivo CRISPR-based functional screening—have proven powerful in dissecting tumor–TME interactions and identifying novel immunotherapy targets in various cancer types, but very few studies have integrated these approaches. In our preliminary studies, we applied two unique immunocompetent genetically-engineered mouse models (GEMMs) of MYC-driven G3MB and performed scRNA-seq, scATAC-seq and spatial transcriptomics profiling. We enriched immune cells from the TME by sorting CD45 positive cells for single-cell studies. Our preliminary analysis of single-cell and spatial omics data revealed striking interactions of neural stem cell-like tumor cells with macrophages and other immune cells that potentially create a suppressive TME and drive immune evasion in mouse G3MB. We also performed in vivo CRISPR screening in tumor cells using the GEMMs to identify modulators of tumor development, which demonstrated the feasibility of in vivo functional genomics screening in our preclinical models. In this project, first, we propose to utilize cutting-edge single-cell and spatial omics technologies to characterize the two G3MB GEMMs at different stages of tumor progression. We will use our network-based tools to integrate these multi-omics data to dissect the dynamic tumor–immune interactions and underlying “hidden” drivers that drive the immune exclusion and suppression during G3MB progression. We will also validate discoveries of G3MB from mouse studies in patient samples. We will develop a cloud-based portal to visualize and explore our single-cell and spatial data and tumor–TME interactomes of G3MB. Second, we will establish the mechanistic basis of tumor–T cell interactions and strategies to enable adoptive T cell therapy for G3MB by discovering functional drivers and putative targets in both tumor cells and T cells. To this end, we will apply both candidate approach and in vivo CRISPR screening in immunocompetent GEMMs to identify tumor- intrinsic modulators that will remodel the suppressive TME and sensitize G3MB tumors to adoptive T cell and CAR-T cell therapies. We will also test if targeting inhibitory factors for T cell function will enable and optimize effective adoptive T cell therapies against such tumors. Our studies promise to provide new insights into mechanisms of tumor–TME interactions in G3MB and manifest legitimate immunotherapeutic opportunities.

PROJECT NARRATIVE Medulloblastoma is the most common malignant pediatric brain tumor and Group 3 medulloblastoma is the most aggressive form of this disease, which is mostly refractory to immunotherapies. This project will dissect the crosstalks between tumor and tumor microenviorment and the mechansims of immune evasion in Group 3 medulloblsatoma. We will also investigate cellular therapy strategies that will circumvent tumor-mediated immunosuppressive mechanisms and reinvigorate anti-tumor immunity in this devastating disease.