Combining Brain Structural Changes To Improve The Prediction Of Alzheimer’s Disease And Investigation Of Metabolite Changes In Early Alzheimer’s Disease Using4.0T~1H Magnetic Resonance Spectroscopy

Part I Combining structural brain changes to improve the prediction of Alzheimer’s diseaseBackground and Objective:Several structural brain changes including global atrophy and white matter lesions have been associated with Alzheimer’s disease (AD), in addition to the hallmark medial temporal lobe atrophy (MTA). The potential role of these changes in AD diagnosis is not well understood. This study investigated the effect of combining various structural changes in the prediction of AD. Methods: Baseline and24-month follow-up high resolution MRI data of people with mild AD (n=39), mild cognitive impairment (MCI; n=82), and of healthy controls (HC; n=58) were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Bilateral MTA was evaluated using a medial temporal lobe atrophy scale (MTAS). Various common changes of the aging brain, including atrophy and lesions in the supratentorial and infratentorial regions, were summarized using a brain atrophy and lesion index (BALI). The performance of the MTAS, BALI, and a score combining both, for the discrimination of MCI and AD, and for the prediction of two-year MCI to AD conversion, were assessed. Results:The MTAS and BALI scores were highly correlated (r2>0.56, p<0.001), each differed significantly between AD, MCI, and HC Having an unfavorable MTAS score was associated with an increased risk of MCI-AD conversion (odds ratio=3.71, p=0.039), adjusted for age, sex, and education; while having an unfavorable BALI score also marginally contributed to such risks (odds ratio=4.08, p=0.080). Combining MTAS and BALI resulted in a greater odds ratio (8.99, p=0.007) and an improved predictive accuracy (75.9%, p=0.002). Conclusion:MTA is predominating in AD; even so, other structural changes can cause a significant additive effect. The data support potential future roles of combining multiple structural changes to benefit AD diagnose, progression monitoring, and/or treatment evaluation. Part II Investigation of Metabolite changes in Early Alzheimer’s disease using4.0T1H Magnetic Resonance SpectroscopyBackground and Objective:Alzheimer’s disease (AD) is a neurodegenerative disorder that involves deficits of several important brain metabolites. Proton magnetic resonance spectroscopy (’H MRS) is a unique neuroimaging method, which can be used non-invasively to obtain information about the neurochemical changes in AD. Although measurement of tissue metabolism has been used to help with more accurate early diagnose of AD and with the evaluation of response to treatment, the metabolite differences involved in multiple brain regions during the process of AD are not well understood. Recently, a neurocompensatory response has been observed in brain region important for executive functioning. Such compensatory in early AD, regarding biochemical metabolic response, has not yet to be studied. The objective of this study was to apply4.0Tesla1H MRS to uncover the neurochemical metabolite changes, and to correlate these changes in two different brain regions, either compromised or preserved in early AD. The project was also aimed to study how MRS-based metabolite changes can be used to distinguish early AD patients from healthy control subjects and to detect the effect of the treatment with cholinesterase inhibitors. Methods:Sixteen subjects with early AD (mean age=76.8±8.0yrs), and20age-matched healthy controls (HC; mean age=73.6±6.3yrs) subjects participated in the study. Subjects had one MRS scan at baseline and one after6months. For the patients with early AD, the follow-up scans were performed after the administration of standard cholinesterase inhibition treatment. MRS signals were sequentially acquired within two single voxels of the same size, respectively placed in the posterior cingulate gyri (PCG) and the left dorsolateral prefrontal cortex (DLPFC). Data post-processing and analysis were performed using fitMAN, AFNI and FSL, and SPSS software packages. Results:Compared to HC subjects, decreased NAA (N-acetylaspartate)/Cr (creatine/phosphocreatine) and NAA/Cho (choline-containing compounds), and increased Cr, Gly (glycine), Cho/NAA, Cr/NAA and Gly/NAA were detected in the PCG voxel in the patients with early AD. Meanwhile, decreased NAA, NAA/Cr, NAA/Cho, NAA/MI (myo-inositol) and Glu (glutamtate)/Cr, and increased MI/NAA, Cr/NAA and Glc (glucose)/NAA were detected in the DLPFC voxel. The Cho level was lower in PCG than in DLPFC, while Glu/Cr was higher in PCG than in DLPFC. No interaction between time (i.e., baseline vs. follow-up) and brain region (i.e., PGC vs. DLPFC) was detected in the metabolites under investigation, except for the level of tau. In addition, after cholinesterase inhibitor treatment, early AD patients demonstrated a reduced level of GPC (glycerophosphorylcholine) and an increased level of PC (phosphorylcholine), PC/Cr and PC/NAA in PCG; they also showed a reduced level of Tau (taurine), Tau/Cr, Tau/Cho, and Tau/NAA in DLPFC. Combining the levels of multiple metabolites resulted in an increased accuracy in the discrimination of people with early AD vs. HC (88.0%for the PCG and88.4%for the DLPFC, compare with using any individual metabolite:64.8%～86.7%). The diagnostic accuracy was further increased to94.4%(95%confidence intervals:78.6%～98.1%) when the levels of metabolites in both PCG and DLPFC were used collectively. Conclusion:High-field ’H MRS has been successfully used to assess metabolite changes and to evaluate responses to medical treatment in early AD. Data suggest that early AD is associated with characteristic patterns of neuronal metabolite abnormalities, which are in line with the underlying pathological and biochemical changes of the disease. The ability of performing basic cognitive functions is related with the plasticity in energy metabolism in the brain, which was indicated by detectable metabolite changes. The differences in the level of the metabolites in different regions and their interactions reflect the heterogeneity in disease progression among individual patients with early AD. Combining multiple metabolites in different brain regions can help improve the accuracy of early diagnosis of AD.