Sex- and age-related heterogeneity of phase-based dynamic functional connectivity in autism spectrum disorder

Autism spectrum disorder (ASD) is characterized by atypical brain functional organization, which may vary depending on sex and age. This study aimed to explore how sex and age influence dynamic functional connectivity (dFC) in ASD. This study analyzed resting-state functional magnetic resonance imaging data from 349 individuals with ASD (284 males/65 females) and 459 typically developing controls (340 males/119 females). A phase-based dFC analysis was first performed to extract time-varying functional brain organization. The Leading Eigenvectors Dynamics Analysis was then used to reduce the dimensionality of dFC, and K-means clustering was conducted to obtain dFC states. The permutation-based three-way analysis of variance was finally used to investigate sex- and age-related changes in ASD. This study identified three dFC states and sex-specific developmental trajectories of dFC were observed in different cognitive-related states in ASD. Moreover, sex- and age-related dFC alterations in individuals with ASD were primarily distributed in the default mode network across all three states. In addition, the state-specific dFC alterations can predict different clinical symptoms in males and females with ASD. These findings demonstrate sex- and age-related heterogeneity of functional connectivity dynamics in ASD, offering phase-based insights into the neural mechanisms underlying the disorder. • Three distinct dFC states characterize dynamic brain organization in ASD. • Sex- and age-related dFC alterations in ASD primarily involve the DMN. • Sex-specific dFC alterations predict divergent clinical symptom profiles.