Neurodegeneration targets brain networks
Neurodegenerative disease atrophy patterns mirror healthy neuronal network activity and gray matter volume variation.
This paper was described as the fundamental framework anchoring a lot of neurodegenerative disease modeling within the Memory and Aging Center at UCSF. Essentially showing how 1) synchronous baseline activity can define function-linked neural networks and 2) these networks are uniquely targeted by different neurodegenerative diseases.
Alzheimer’s disease (AD), behavioral variant frontotemporal dementia (bvFTD), semantic dementia (SD), progressive nonfluent aphasia (PNFA), and corticobasal syndrome (CBS) all showed associations with unique clinical syndromes and atrophy across specific multiple regions. Using most atrophied regions as ROI seeds and recording brain network patterns in healthy controls, the resulting networks matched disease atrophy patterns. Essentially, people who were sick had tissue atrophy in the same connected brain regions that healthy individuals showed network activation in. Not only that, the atrophy areas and healthy networks also showed correlated gray matter measurements across sample populations, meaning not only did neuronal activity show similar patterns, but physical structure also showed similarity. Lastly, these connected networks were unique and specific for each studied disease.
These experiments, while rigorous and thorough, do not prove that disease travels in a “growing” trajectory from respective ROI seeds. It could be equally likely that these regions experience independent, simultaneous degeneration. The authors suggest a number of possible propagation theories, and also that none of them are definitively supported by these experiments. The small sample sizes across diseases also warrant closer looks at these patterns with larger and more diverse populations. In fact, a later article (Zhou et al, Neuron 2012) tests this concept of simultaneous degeneration, called shared vulnerability. Finally, correlating patients’ longitudinal network patterns with disease trajectory is an important area to consider.
Most of my AD-related experiments these days look at region- or lobar-level effects. If degeneration truly follows network association patterns, then perhaps these patterns should also be a default experimental regimen. For example, AD-related free water fraction analyses could look at default mode network-aggregated measurements in addition to just anatomically connected white-matter bundle tracts.