Deconvolution of the Meningioma Microenvironment
Date of Award
Spring 2024
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Department
Genetics
First Advisor
Gunel, Murat
Abstract
Meningiomas, the most common primary intracranial tumor, provide a unique opportunity to study the intricate interplay between genetic mutations, cellular interactions, and the tumor microenvironment. This dissertation offers a comprehensive examination of meningioma biology, dissecting their complex microenvironment to reveal the environmental influences that shape their morphological characteristics and epidemiological observations. By analyzing data from 309 genetically and clinically characterized tumors, 56,020 spatial transcriptomic spots, and 220,043 single cells, this research uncovers signaling dynamics that foster spatial compartmentalization of genetic and phenotypic variance within these tumors. Unsupervised clustering of spatial transcriptomics uncovers shared marker genes in meningioma whorls and lobules enriched in glycolysis and hypoxia-associated pathways. We identify differing cell type abundances within distinct malignant cell populations and immune cell profiles across genomic subgroups and discover that meningiomas exhibit metabolic and immune spatial compartmentalization. Similarly, we identify that chromosome copy number variations localize to regions marked by structural and metabolic variation, including hypoxia. We developed spatial Affinity-graph Recovery of Counts (spARC), a tool that applies principles of manifold geometry and graph signal processing to spatial transcriptomics data. spARC uses biological constraints to denoise and refine high-dimensional data into a lower-dimensional, denoised form that allows the modeling of spatial compartment signaling dynamics. Employing spARC, landscape ecology principles, and novel signal processing techniques, we identified APOE-LRP1, THBS1-LRP1, COL1A2-CD36, and PDGFD-PDGFRB as the predominant signaling pathways in meningiomas. This dissertation revisits the epidemiological profile of meningiomas, noting a higher female-to-male incidence ratio in reproductive years, suggesting a role for gonadal steroid hormones. This observation dates back to Harvey Cushing's "Meningiomas: Their Classification, Regional Behaviour, Life History, and Surgical End Results" published in 1938. We conducted a multilevel meta-analysis of 114 studies that spanned nearly seven decades and included 6,092 tumors belonging to 5,810 patients to resolve long-standing questions about hormone receptor status, revealing patterns linked to patient demographics, tumor location, and histological characteristics, thus informing hormonal therapy strategies in clinical trials where receptor stratification has historically been sub-optimal. For the first time, we first describe the identification and characterization of all three estrogen receptors (ERα, ERβ, GPER) in meningiomas, underscoring the complex hormonal regulation within the tumor microenvironment. With NF2/22q loss accounting for the majority of sporadic meningioma mutations and a genetic vulnerability to ferroptosis, our multi-omic perturbation experiments demonstrate that estrogen and its modulators can inhibit ferroptosis in NF2/22q loss meningiomas, revealing a complex interplay between estrogen, metabolism, and cell fate. This dissertation advocates for a systems-based therapeutic approach, acknowledging the difficulty of targeting individual mutated cells amidst the heterogeneous and metabolically constrained tumor microenvironment. Meningiomas, characterized by a low mutation burden and rich intercellular interactions, emerge as an exemplary model for probing genetic, developmental, and evolutionary interrelations, thus providing a microcosmic perspective on tumor diversity and evolution. This work describes the present understanding of meningioma biology and establishes a new trajectory for future research and therapeutic interventions considering the breadth of factors in the tumor ecosystem.
Recommended Citation
Miyagishima, Danielle F., "Deconvolution of the Meningioma Microenvironment" (2024). Yale Graduate School of Arts and Sciences Dissertations. 1448.
https://elischolar.library.yale.edu/gsas_dissertations/1448
