Towards MRI-based detection and categorization of invasive high grade glioma subtypes - PROJECT SUMMARY The goal of this work is to detect and differentiate with MRI the invasive NEU and GPM high-grade glioma subtypes based on their divergent microstructure, glycolytic metabolism, and glutaminergic synapse profiles. High grade glioma (HGG) brain tumors rank among the most lethal of all human cancers, with a median survival of ~17 months. Current therapies for HGG, like surgery and chemoradiation, rely on contrast-enhanced (CE) MRI to demarcate treatment targets based on tumor-associated blood brain barrier (BBB) disruption. CE-MRI is a powerful tool to detect these so-called “enhancing” lesions. However, infiltrating tumor cells extend well beyond the boundaries of BBB disruption and these “non-enhancing” tumoral regions remain invisible on CE-MRI. This failure of standard-of-care CE-MRI to detect infiltrating tumor leads to undertreatment of HGG and greater incidence of recurrence. Recently, four HGG subtypes have been identified based on developmental and metabolic features, including the proliferative/progenitor (PPR), neuronal (NEU), mitochondrial (MTC), and glycolytic/plurimetabolic (GPM) phenotypic subtypes and confers clinical implications. Expanding on this and using 313 image localized biopsy tissue samples across 68 HGG patients, recent work has shown that the majority of non-enhancing (invasive) tumor cells are of NEU or GPM subtype, and that they cluster spatially. The NEU subtype uniquely forms unmyelinated neurite-like microtubes and glutamatergic synapses with neighboring HGG and host cells, while the GPM subtype has a uniquely high glycolytic activity involving interaction with astrocytes, other glial cells, and infiltrating immune cells. The predominance of one subtype over the other influences biological behavior and therapeutic susceptibilities. For this reason, non-invasive methods to spatially resolve clusters of these invasive HGG subtypes would be a major advance in the diagnosis and treatment of HGG. To spatially resolve clusters of invasive NEU and GPM HGG subtypes, we will leverage well-established advancements in quantitative diffusion MRI (dMRI)-based unmyelinated neurite microstructure mapping and deuterium (2H) MRI-based mapping of glycolysis and neurotransmitter (glutamate) synthesis. We hypothesize that: 1. The GPM subtype will be made MRI-detectable based on low neurite-like microtube density, high glycolytic metabolism, and low glutamate synthesis, and 2. The NEU subtype will be made MRI-detectable based on high neurite-like microtube density, low glycolytic metabolism, and high glutamate synthesis. The Specific Aims of this work are: Aim 1. Establish and validate quantitative dMRI microstructural signatures of NEU and GPM HGG subtypes based on their known differential expression of neurite-like microtubes; Aim 2. Establish and validate quantitative metabolic 2H MRI signatures of NEU and GPM HGG subtypes based on their known divergent glycolytic and glutaminergic synapse phenotypes.