Year: 2022
Type: dissertation
Abstract: Glioblastoma (GBM) is the most common brain tumour amongst adults with a dismal median survival of 15 months, despite multimodal therapy, including maximal resection and chemotherapy. Classically, GBM has been defined by hallmark histomorphological features, including cellular (CT) and infiltrating tumour (IT) areas, regions of microvascular proliferation (MVP), and hypoxia where the tumour cells palisade around necrosis (PAN). These diverse histomorphologic niches, create a heterogeneous tumour microenvironment (TME) that contributes to tumour evolution and treatment resistance. To better understand the molecular tenants of these histomorphological features, I created an anatomical proteomic atlas of human glioblastoma by excising these regions, using laser capture microdissection techniques, and performing liquid-chromatography based proteomics. This unique resource highlighted that GBM intra-tumour heterogeneity can be defined across a› three-dimensional axis including MYC-, KRAS- and hypoxia programs. Furthermore, utilizing these reference proteomic profiles, I leveraged machine learning to deconvolute and estimate niche proportions in independent bulk based proteogenomic efforts. This approach revealed a strong association of the proneural subtype with a diffusely infiltrating phenotype providing new insights for the genetic drivers and poor response of this subtype. Finally, I compiled these proteomic profiles to develop a publicly available bioinformatics portal for knowledge dissemination in hopes of accelerating our understanding of glioblastoma biology. Together these works highlight the importance of accounting for regional proteomic profiles in GBM to nominate phenotypic programs and unify biological theories. Generalization of the approaches and data outlined in this thesis can help explore other intra-tumoural aspects of tumour heterogeneity including divergent spatially distributed tumour subpopulations and structures.
Source: TSpace (University of Toronto)
Author: Lam, Ka Hang Brian
Institution: University of Toronto
Language: English
Subfield: Oncology
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W7132991659
