UCSF

Alzheimer’s disease (AD) is the most prevalent form of dementia associated with impaired cognitive ability and abnormal neural synchrony. Electrophysiological studies done by magnetoencephalography (MEG) reveal abnormal neural synchrony, reflected in brain signals with elevated power in the delta-theta band (1-7 Hz) and reduced power in the alpha-beta-gamma band (8-51 Hz), accompanied by altered neural driving activity that is scale-free in nature due to the lack of a dominant temporal scale. However, despite being a fundamental component of brain activity, the role of scale-free neural driving dynamics and their influence on altering brain network organization in AD remains poorly understood. To address this, a biophysical modeling framework called the Spectral Graph Model (SGM) is used in this study to deconvolve the scale-driving signals, analyze how they are altered in AD compared with healthy controls, and examine their effects on excitatory-inhibitory (E/I) neuronal dynamics. MEG recordings from a cohort of 77 AD and 77 healthy control subjects are used in this study. AD is accompanied by altered scale-free driving signals across frequency bands compared to healthy controls, with a significant increase in delta-theta power, especially in the temporo-parietal cortices and frontal cortices; a biphasic pattern in the alpha band with increased power in the medial temporo-parietal and frontal regions and reduced power in the lateral posterior-parietal cortices; the beta band shows reduced power in the lateral posterior-parietal cortices, while the gamma band shows an overall reduction in power across brain regions. These spatial patterns of altered scale-free driving signals, especially the delta-theta, alpha, and beta bands, recapitulate the AD-vulnerable cortices revealed by hypometabolism, amyloid-beta, and tau accumulation in this AD patients’ cohort. Altered scale-free driving signals significantly impaired the temporal profiles of E/I dynamics, with AD showing a significant increase in local and long-range excitatory time constants, reflecting local and global slowing of excitatory neural dynamics. Together, these findings demonstrate that scale-free neural driving signals are profoundly disrupted in AD, contributing to alterations in both local E/I dynamics and long-range inter-regional network dynamics, paving the way for a potential new biomarker of AD.