Characterizing Sex Differences in Functional Connectivity Changes Across the Alzheimer’s Disease Clinical Continuum

Abstract

Treatments that are currently available for Alzheimer’s disease (AD) are largely ineffective, likely because they are delivered following diagnosis, when significant neurodegeneration has already occurred and cannot be reversed (Waite, 2015). Another key element that may contribute to therapeutic failure is the “one-treatment-fits-all” approach, which inherently considers AD as a homogenous state, ignoring the significant interindividual variability that is observed in risk profiles (Reitz, 2016). Characterizing the influence of factors that contribute to the observed heterogeneity in AD, such as biological sex, on pathological brain changes may reveal more individualized biomarkers to aid early detection efforts and more effective treatment targets. Sex differences in AD prevalence and manifestation are well-documented (Hebert et al., 2013), yet the neural correlates underlying these clinical and epidemiological observations are relatively unknown. The purpose of the current study was to characterize sex differences in functional connectivity alterations across the AD clinical continuum, in cognitively normal (CN), MCI, and AD groups. Data used in preparation of this study were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database (http://adni.loni.usc.edu). A comprehensive panel of functional connectivity (FC) markers was utilized to examine network topology (graph theory) and resilience to simulated insult, network connectivity, and temporal fluctuations in FC across the scan period. Static FC analyses revealed that regardless of diagnostic group, males showed higher intra- and inter-network connectivity than females. The dynamic FC analysis showed that the full scan session was characterized by five re-occurring patterns FC and that regardless of diagnosis, males spent more time in connectivity states characterized by high global connectivity across the network, whereas females spent more time in connectivity states with high intra- and low inter-network connectivity. FC variability analysis showed that connectivity patterns within the right fusiform and left middle frontal gyrus were more stable in MCI females than MCI males, CN females, and AD females. Graph theory analyses revealed in CN and MCI groups, but not in AD, female networks were more efficiently organized to achieve global integration and were more resilient to targeted hub attacks than males. This study demonstrates that alterations to network topology and brain dynamics across the AD clinical continuum are sex specific and further emphasizes the need to include sex in AD research to uncover neural correlates underlying sex differences apparent in risk profiles, manifestation, and progression of AD.