Exploring the Pattern Associated With Longitudinal Changes of β-Amyloid Deposition During Cognitively Normal Healthy Aging

Abstract

The aim of this study was to determine a pattern associated with longitudinal changes of β-amyloid (Aβ) deposition during cognitively normal(CN) healthy aging. We used 18F-florbetapir (AV-45) PET images of the brains of 207 cognitively normal subjects (CN1), obtained through the Alzheimer's Disease Neuroimaging Initiative (ADNI), to identify the healthy aging pattern and 76 cognitively normal healthy subjects (CN2), obtained through the Xuanwu Hospital of Capital Medical University, Beijing, China, to verify it. A voxel-based correlation analysis of standardized uptake value ratio (SUVR) map image and age was conducted using the DPABI (Data Processing & Analysis of Brain Imaging) software to identify the pattern. The sum of squares due to errors (SSE), R-square (R2) and the root-mean-square error (RMSE) were calculated to assess the quality of curve fitting. Among them, R2 was proposed as the coherence coefficient, which was as an index to assess the correlation between SUVR value of the pattern and subjects' age. The pattern characterized by age-associated longitudinal changes of Aβ deposition was mainly distributed in the right middle and inferior temporal gyrus, the right temporal pole: middle temporal gyrus, the right inferior occipital gyrus, the right inferior frontal gyrus (triangular portion), and the right precentral gyrus. There were a significant positive correlation between the SUVR value of the pattern and age for each CN group (CN1: R2 = 0.120, p < 0.001 for quadratic model; CN2: R2 = 0.152, p = 0.002 for quadratic model). These findings suggest a pattern of changes in Aβ deposition that can be used to distinguish physiological changes from pathophysiological changes, constituting a new method for elucidating the neuropathological mechanism of Alzheimer's disease.