Resting-state functional MRI signal fluctuations are correlated with brain amyloid-β deposition

Mounting evidence suggests that amyloid-{beta} (A{beta}) and vascular etiologies are intertwined in the pathogenesis of Alzheimers disease. Spontaneous fluctuations of the brain blood-oxygen-level-dependent (BOLD) signal, as measured by resting-state functional MRI (rs-fMRI), have been shown to be associated with neuronal activities as well as cerebrovascular hemodynamics. Nevertheless, it is unclear if rs-fMRI BOLD fluctuations are associated with brain A{beta} deposition in individuals with an elevated risk of Alzheimers disease. We recruited 33 patients with amnestic mild cognitive impairment who underwent rs-fMRI and positron emission tomography (PET). The A{beta} standardized uptake value ratio (SUVR) was calculated with cortical white matter as the reference region to improve sensitivity for cortical A{beta} quantification. We calculated the amplitudes of low-frequency fluctuations (ALFF) of local BOLD signals in the frequency band of 0.01-0.08 Hz. Applying physiological/vascular signal regression in stepwise increasing levels on the rs-fMRI data, we examined whether local correlations between ALFF and brain A{beta} deposition were driven by vascular hemodynamics, spontaneous neuronal activities, or both. We found that ALFF and A{beta} SUVR were negatively correlated in brain regions involving the default-mode and visual networks, with peak correlation at the precuneus, and angular, lingual, and fusiform gyri. Regions with higher ALFF had less A{beta} accumulation. The correlated cluster sizes in MNI space were reduced from 3018 mm3 with no physiological/vascular regression to 1072 mm3 with strong physiological/vascular regression, with mean cluster r values at approximately -0.47. Results demonstrate that both vascular hemodynamics and neuronal activities, as reflected by BOLD fluctuations, are negatively associated with brain A{beta} deposition. These findings further imply that local brain blood fluctuations due to either vascular hemodynamics or neuronal activities can affect A{beta} homeostasis.