Longitudinal changes in the functional connectivity of individuals at risk of Alzheimer's disease

Abstract

First-degree relatives of Alzheimer's disease patients constitute a key population in the search for early markers. Our group identified functional connectivity differences between cognitively unimpaired individuals with and without a family history. In this unprecedented follow-up study, we examine whether family history is associated with a longitudinal increase in the functional connectivity of those regions. Moreover, this is the first work to correlate electrophysiological measures with plasma p-tau231 levels, a known pathology marker, to interpret the nature of the change. We evaluated 69 cognitively unimpaired individuals with a family history of Alzheimer's disease and 28 without, at two different time points, approximately 3 years apart, including resting state magnetoencephalography recordings and plasma p-tau231 determinations. Functional connectivity changes in both precunei and left anterior cingulate cortex in the high-alpha band were studied using non-parametric cluster-based permutation tests. Connectivity values were correlated with p-tau231 levels. Three clusters emerged in individuals with family history, exhibiting a longitudinal increase of connectivity. Notably, the clusters for both precunei bore a striking resemblance to those found in previous cross-sectional studies. The connectivity values at follow-up and the change in connectivity in the left precuneus cluster showed significant positive correlations with p-tau231. This study consolidates the use of electrophysiology, in combination with plasma biomarkers, to monitor healthy individuals at risk of Alzheimer's disease and emphasizes the value of combining noninvasive markers to understand the underlying mechanisms and track disease progression. This could facilitate the design of more effective intervention strategies and accurate progression assessment tools.

Keywords: Alzheimer’s disease risk; Functional connectivity; Longitudinal; P-tau231, MEG.

Fig. 1

Schematic representation of the assessment protocol. Baseline assessment is identified as A1; the longitudinal follow-up, approximately 3 years afterwards, as A2

Fig. 2

Visual representation of the analysis pipeline. (a) The 1210 × 1210 PLV matrix is calculated, containing the connectivity values between all cortical sources and, based on the AAL atlas, (b) only PLV values of the sources that correspond to the seeds are selected. (c) The connectivity average of the sources of the seed is calculated, obtaining two seed-with-all connectivity vectors, one for the baseline assessment and one for the follow-up. (d) A cluster-based permutation test based on a t-test is performed

Fig. 3

Significant clusters of FH + longitudinal PLV changes. Clusters with significant differences in functional connectivity at A1 and A2. The seeds are (a) right precuneus, (b) left precuneus, (c) left anterior cingulate cortex. Black areas represent the seed area under study, and the purple ones represent the clusters with significant differences in connectivity with the seed area. Colored in purple are the clusters that comprise the areas where a longitudinal increase in PLV with the seed area is seen

Fig. 4

Functional connectivity in the resulting clusters. PLV values for the FH + (purple) and FH- (yellow) groups at A1 and A2 in the significant cluster of increased longitudinal connectivity with the (a) right precuneus, (b) left precuneus, and (c) left anterior cingulate cortex

Fig. 5

Relationship between PLV values at A2 and p-tau231 levels. Illustration of the relationship between the levels of p-tau231 and the FC values for the FH + (purple) and FH- (yellow) groups in the significant clusters of increased longitudinal connectivity with the (a) right precuneus, (b) left precuneus, and (c) left anterior cingulate cortex