Use of Diffusion Magnetic Resonance Imaging for Detection of Early Neurodegeneration in Asymptomatic Individuals with Early Alzheimer's Disease

Abstract

Neurodegeneration associated with Alzheimer’s disease (AD) can start years before cognitive symptoms, but is difficult to identify in presymptomatic individuals. In order to delay the onset and/or slow down the progress of AD-associated cognitive decline, early detection within presymptomatic individuals can lead to improved treatment outcomes. Diffusion magnetic resonance imaging (dMRI) has the capability to probe the brain microstructure and the potential to extract imaging biomarkers related to early AD-associated neurodegeneration. Within this dissertation, we aim to develop novel dMRI biomarkers to detect early microstructural neurodegeneration in: the hippocampus by using structural connectomes derived from fiber tractography to assess changes in connectivity (Aim 1) and in the gray matter by using a cortical column-based analysis to assess changes in radial diffusivity across different cortical depths and regions (Aim 2). The hippocampal connectivity analysis was first applied to dMRI data from the Alzheimer’s Disease Neuroimaging Initiative repository. While the spatial resolution was standard (2 mm isotropic), the large amount of participant data allowed us to apply both statistical and machine-learning models to differentiate cognitively normal participants from participants diagnosed with AD. We also applied these models towards predicting the classification of participants who remained cognitively normal vs. participants who were later diagnosed with AD.Additionally, we used high-resolution (1 mm isotropic) dMRI data from the Duke-UNC Alzheimer’s Disease Research Center to extract imaging biomarkers of early neurodegeneration for both aims. With the high-resolution data, we compared four groups of participants across a spectrum of cognitive decline (n=10 in each group): cognitively normal amyloid-negative (normal controls), cognitively normal amyloid-positive (stage-1 AD), mild cognitive impairment (MCI), and AD dementia. Many of these participants also had follow-up clinical exams to assess cognitive decline, which allowed us to evaluate the ability of our biomarkers to detect early neurodegeneration in participants who were later diagnosed with MCI or AD dementia.When applying the hippocampal connectivity analysis to the high-resolution dMRI data, a group-wise analysis reported a progressively lower connectivity across groups 1–4 for regions affected by neurodegeneration in AD. Based on this differentiation, an individual participant analysis then reported a lower connectivity within multiple asymptomatic stage-1 AD participants who were later diagnosed with MCI. When applying the cortical column analysis to the high-resolution dMRI data, our results showed progressive increases in radial diffusivity across groups 1–4 in cortical regions associated with AD. More importantly, our analysis in the individual asymptomatic participants from the stage-1 AD group was able to differentiate different degrees of diffusivity increases to predict risk of future cognitive decline, as confirmed by follow-up clinical exams.Within these endeavors, we have demonstrated that biomarkers extracted from dMRI data have the potential to identify early neurodegeneration in the hippocampus and gray matter of asymptomatic individuals with early AD, which will be highly valuable to guide early and more effective treatments.