A null findings study: graph theoretical analysis of the fetal functional connectome shows no relationships with future autistic traits

Autism spectrum disorder (ASD) is a neurodevelopmental condition, with <i>ex vivo</i> studies suggesting its neurobiological origin as early as the first and second trimester of pregnancy. Functional MRI studies using graph-theoretical approaches have isolated features in the global connectome architecture that distinguish toddlers with ASD from their typically developing peers. Additionally, functional connectivity patterns in the infant brain have shown to be predictive of later ASD diagnosis. An important yet unexplored question in the literature is whether graph-theoretical differences are evident prior to infancy, in the brain of <i>fetuses</i> who will later exhibit ASD traits in early childhood. In this study, we address this question using a sample of 88 children with both quality-assured <i>fetal</i> brain resting-state functional MRI data and standardized parent assessment of ASD traits including social-emotional and social communication skills and repetitive and restricted behaviors at age 3. Multiple regression analyses revealed no significant associations between fetal global graph features (e.g., network segregation, integration, and small-world architecture) and ASD traits at age 3 (<i>p</i>'s > 0.1). Therefore, our findings do not provide support for prenatal emergence of global topographical differences of brain functional organization in fetuses who later develop ASD traits. However, this does not rule out the possibility of other neural signatures in the fetal functional connectome that may predict autistic traits and future ASD diagnosis.