Alzheimer's disease (AD) accounts for the majority of dementia cases worldwide. The number of persons with AD in the US is expected to reach 13.5 million by 2050, and the medical costs associated with AD are expected to exceed $20 trillion over the next 40 years. Available drugs offer only moderate symptom alleviation. Most importantly, no therapeutic strategies have demonstrated clinically significant disease- modifying benefits, with more than 100 failed clinical trials in the last 20 years. Emerging evidence suggests striking similarities between energetic adaptations observed in central nervous system (CNS) cells in sporadic AD and those shown by cancer cells. These include a shift from oxidative phosphorylation (OxPhos) to upregulation of aerobic glycolysis (AG) as an adaptive mechanism against neurotoxic conditions (e.g., the accumulation of amyloid beta [A?] oligomers in the case of AD). This energetic shift could explain the paradoxical observation that many elderly individuals remain cognitively normal (NL) despite the presence of high levels of A? deposition. While AG upregulation may serve as a stopgap to rescue select neurons and preserve cognition, its long-term upregulation reduces neuron viability due to improper processing of lactate, increased levels of which cause oxidative stress and neuronal loss. Subsequently, greater energy demands are placed on the diminished neuron population, which initiates a new shift to increased OxPhos that is also consistent with reduced efficiency in ATP synthesis by mitochondrial dysregulation (a hallmark of AD) and neuronal death. This model predicts: i) increased AG in A?+ NL subjects compared to A?? NL controls, ii) decreased OxPhos in A?+ NL subjects compared to A?? NL controls, and iii) increased OxPhos in neuronal cells that have survived apoptosis in amnestic mild cognitive impairment (aMCI) or AD patients compared to A?+ NL controls. Combined multinuclear MR/PET is a uniquely suitable tool to directly test such neuroenergetic models. While our study focuses on the compensatory energy pathways and A? in AD, the tools we develop can provide vital insight on a range of amyloid cascade and/or neuroenergetic hypotheses. We will use our MR/PET system to measure amyloid deposition with PiB (Pittsburgh compound)-PET, OxPhos with phosphorus (31P) Magnetic Resonance Spectroscopic Imaging (MRSI), and lactate with proton (1H)-MRSI. This work has been propelled by our recent instrumentation grant that has enabled multinuclear MRSI/PET on our clinical scanner. In addition, we engineered and tested a highly sensitive dual/tuned (31P/1H) head coil array with low PET attenuation, and developed a software pipeline that uses anatomical MR images to create volumes of interest that selectively include cortical regions with high AG in the presence of adequate oxygen in healthy subjects. Together, these tools will enable simultaneous co-localized 31P/1H-MRSI and PET imaging for pioneering AD neuroenergetic evaluation in a group of 15 elderly participants (age range: 70?85 y/o) with aMCI and 30 age- and sex-matched NL controls.
We have recently engineering a highly sensitive dual/tuned (31P/1H) head coil array with low PET attenuation. This tool will enable simultaneous co-localized 31P/1H-MRSI and PET imaging for studying neuroenergetic adaptations in a group of 15 elderly participants (age range: 70?85 y/o) with amnestic mild cognitive impairment and 30 cognitively normal controls. Multimodal (MR/PET) and multinuclear (31P/1H) neuroimaging will allow us to gain access to a uniquely comprehensive and highly consistent view of neuroenergetic adaptations in both the clinical and preclinical stages of Alzheimer's disease.
