Effects of Alzheimer’s disease plasma marker levels on multilayer centrality in healthy individuals

Abstract

AbstractFinding early and non-invasive biomarkers that help identify individuals in the earliest stages of the Alzheimer’s disease continuum is paramount. Electrophysiology and plasma biomarkers are great candidates in this pursuit. Furthermore, the combination of functional connectivity metrics with graph-theory analyses allows for a deeper understanding of network alterations. Despite this, this is the first MEG study to assess multilayer centrality considering inter-band connectivity in an unimpaired population at high risk of Alzheimer’s disease. Our objective is twofold. First, to address the relationship between a compound centrality score designed to overcome previous inconsistencies stemming from the use of various individual metrics, and plasma pathology markers of Alzheimer’s disease in unimpaired individuals with elevated levels of the latter. Lastly, to evaluate whether hubs’ centrality is more affected by the pathology.33 individuals with available MEG recordings and elevated plasma pathology markers were included. A compound centrality score for each brain source of every subject was calculated combining widely used centrality metrics, considering intra- and inter-band connections.Spearman correlations were carried out to address the association between each node’s centrality score and biomarkers levels. Next, to test whether greater associations were found in hubs, a correlation between the obtained rho and the grand-average of the centrality score was carried out.Increasing concentrations of p-tau231 were associated with greater centrality within the network of posterior areas, which increased their connectedness in the theta range with the remaining areas, regardless of the latter’s frequency range. The opposite relationship was found for left areas, that decreased their connectedness in the gamma frequency range. Hubs’ centrality was significantly more affected by p-tau231 levels.Our results expand previous literature demonstrating early network reorganizations associated with elevated plasma p-tau231 in cognitively unimpaired individuals. Multilayer centrality increases in the theta band in posterior regions are congruent with previous results and theoretical models, that predict a longitudinal evolution towards a loss of centrality. On the other hand, the changes in multilayer centrality found in the gamma band could be associated with inhibitory neuron dysfunction, classical in AD pathology. Lastly, hubs were more likely to increase their centrality in association to p-tau231, thus corroborating hubs’ vulnerability.