Date of Award

Spring 5-15-2023

Author's School

Graduate School of Arts and Sciences

Author's Department

Biology & Biomedical Sciences (Immunology)

Degree Name

Doctor of Philosophy (PhD)

Degree Type



Fueled by a deeper understanding of the fundamental mechanisms of anti-tumor immune responses, cancer immunotherapy has experienced a resurgence over the last decade with remarkable success in the treatment of certain malignancies. However, despite a multitude of treatment approaches ranging from immune checkpoint blockade to adoptive cellular therapy to neoantigen vaccines, certain cancer types have shown limited response to immunotherapy. Glioblastoma multiforme (GBM), the most common primary malignancy of the central nervous system, represents a particularly challenging tumor where new treatment approaches are desperately needed. Despite multimodality treatment consisting of surgical resection, chemotherapy, and radiation therapy, patients with GBM face a median survival of less than two years. This standard of care treatment has not changed in over fifteen years, with numerous immunotherapy trials providing no survival benefit. However, fundamental aspects of the cancer immunity cycle in GBM and other malignant brain tumors remain unclear. Specifically, the prevalence of T cell responses and the nature of tumor antigens recognized within human brain tumors has not been described. Furthermore, the role of T cells targeting tumor-specific neoantigens within preclinical systems has not been well-defined. This work is focused upon characterizing the genomic and immunological state of malignant brain tumors, developing approaches to probe for reactivity among tumor-infiltrating T cells, and establishing a murine system to target tumor neoantigens through T cell therapy. In contrast to the lack of success in GBM, immunotherapy has demonstrated clear benefit in the treatment of secondary brain metastases arising from distant primaries such as lung carcinoma or melanoma. To characterize the immunogenomic landscape of primary and secondary brain malignancies, we performed multi-region whole-exome, RNA, and TCR sequencing on a cohort of primary gliomas and secondary brain metastases. Our analyses identified significant differences between primary and secondary malignancies with gliomas displaying substantially more spatial heterogeneity at the genomic and neoantigen level. Additionally, this increased spatial diversity was recapitulated in the distribution of T cell clones where some gliomas harbored highly expanded but spatially restricted clonotypes. Finally, we comprehensively profiled a hypermutated recurrent tumor, a phenotype observed in 20% of recurrent gliomas, finding evidence of significant intratumoral T cell clonal expansion and gene expression changes indicative of T cell activation. Following the appreciation that T-cell mediated recognition of tumor-specific neoantigens is a crucial component of both cancer immunoediting and the effective immune response following checkpoint blockade, there has been extensive interest in characterizing the antigen reactivity of tumor infiltrating lymphocytes in a variety of tumor types. However, this analysis has not been extended to malignant brain tumors and is complicated by the relatively sparse T cell infiltrate within gliomas. To address this, we generated and validated mouse and human T cell reporter systems with the goal of defining the antigen reactivity of tumor infiltrating lymphocytes. The specificity of expanded intratumoral T cell clones identified through single-cell RNA-sequencing could be screened by introduction of their T cell receptor (TCR) into the reporter cell lines prior to subsequent co-culture with MHC-matched cells loaded with candidate antigens. We are actively applying these systems to explore the antigen reactivity within our murine preclinical systems and cohorts of human patients. Prior work in our lab identified the Imp3D81N mutation (mImp3) as a tumor neoantigen recognized by infiltrating CD8 T cells in the murine glioma model GL261. To develop a system for targeting this endogenous neoantigen, we cloned a TCR specific for the mImp3 epitope and generated a transgenic mouse expressing this TCR, hereafter referred to as the Mutant Imp3-Specific TransgenIC (MISTIC) mouse. MISTIC T cells proliferate and secrete cytokines in response to the mutant but not wild-type peptide and recognize mImp3-bearing tumor targets. When in vitro activated MISTIC T cells are transferred into tumor-bearing mice in a model of cellular therapy, they localize to the tumor microenvironment, and a majority of mice are cured of their tumor. To our knowledge, the MISTIC mouse represents the first TCR transgenic mouse specific for a brain tumor neoantigen, paving the way for future studies on neoantigen-specific T cell responses in endogenous and therapeutic settings. Collectively, these studies provide insight into the immunogenomic landscape of malignant brain tumors, develop systems to probe for neoantigen reactivity within mouse and human brain tumors, and display the potential of cellular therapy targeting tumor-specific neoantigens.


English (en)

Chair and Committee

Gavin P. Dunn

Committee Members

Todd Fehniger, Malachi Griffith, Albert Kim, Naresha Saligrama,

Available for download on Monday, February 17, 2025