Metabolic neuroimaging of ApoE and APP mutational status in mouse models of Alzheimer’s disease

Motivation: As metabolic impairment is key in AD, metabolic imaging could potentially improve diagnosis and monitoring of AD. Goal(s): Our goal is to determine which metabolic imaging approach, or combination of approaches, provide the optimal set of biomarkers for AD. Approach: We combined three metabolic imaging methods, 1H MRS, HP 13C MRSI and 18F-FDG PET, with machine learning to characterize the neurometabolic profiles linked to AD-related risk factors, namely ApoE mutation, APP mutation, and sex in AD mouse models. Results: Combining metabolic neuroimaging and machine learning can help discriminate between AD-related mutational status (APP and ApoE) and provide information of AD-related sexual dimorphism. Impact: Knowing which metabolic imaging approach(es) is/are optimal to monitor progression in each subset of AD patients, based on sex and mutational status, would improve patient-centric clinical care and potentially create new avenues for assessment of new metabolism-targeting therapies.