Mapping Alzheimer's disease stages toward it's progression: A comprehensive cross-sectional and longitudinal study...

Functional brain connectivity of resting-state networks varies as Alzheimer's disease (AD) progresses. However, our understanding of the dynamic longitudinal changes that occur in the brain over the course of AD is sometimes contradictory and lacking.

In this study, we analyzed whole-brain networks connectivity using longitudinal resting-state fMRI data from 132 participants from ADNI dataset. The cohort was divided into four groups: 20 AD, 35 CN, 46 Early MCI, and 31 Late MCI Cross-sectional analyses were conducted at baseline and follow-up (approximately two years apart), with longitudinal changes assessed within and between groups.

Cross-sectional analyses revealed that all groups differed significantly from AD in global network properties at both time points, with EMCI also showing disrupted topological metrics compared to CN. Longitudinal analyses highlighted notable changes in small-worldness (σ), global clustering coefficient (Cp), and normalized characteristic path length (λ) across groups. Both EMCI and LMCI groups showed significant alterations in global efficiency (Eglob), Cp, and σ over time. Pairwise comparisons also revealed significant interaction effects, particularly between CN-EMCI and CN-AD groups. All groups showed notable changes in σ, λ, and Cp, according to within-group longitudinal changes. Furthermore, distinct changes in Eglob over time were observed in the LMCI and EMCI groups. Almost all subnetwork attributes demonstrated significant changes between patients at various phases in both time intervals.

Our findings emphasize significant connectivity alterations across all groups at both baseline and follow-up, with longitudinal analyses underscoring the progression of these changes. Graph theory metrics provide valuable insights into the transition from normal cognition to AD, potentially serving as biomarkers for disease progression.