Forerunners of conduct disorder and psychosis in the developing brain - insights across time from the ABCD study

Le résumé fourni par la source

Conduct disorder (CD) typically emerges in late childhood and early adolescence, a period that also marks the onset of prodromal psychosis symptoms. Their co-occurrence may exacerbate aggression and increase risk of adverse outcomes. While white matter (WM) abnormalities have been linked to both, little is known about their shared and distinct developmental WM signatures and trajectories over time. We examined cross-sectional and longitudinal associations between WM microstructure, CD traits, psychotic-like experiences (PLEs), and their interaction in youth from the Adolescent Brain Cognitive Development Study (ABCD) (up to 20,206 scans). WM microstructure was measured with diffusion tensor imaging (DTI) across major WM tracts at ages 9–10 (baseline), and at 2- and 4y follow-ups. Linear mixed-effect models were used to test cross-sectional and longitudinal associations controlling for age, sex, ancestry, relatedness, scanner/site, and parental education. CD traits and PLEs were positively correlated at all timepoints (r s = 0.12–0.14, p < 0.0001). Cross-sectional analyses revealed several associations with WM microstructure, particularly for CD traits at baseline and 2y, and for PLEs at 2y ( p FDR < 0.05), though effects were absent at 4y. No significant longitudinal associations between WM microstructure or CD × PLE interactions were detected after FDR-correction. To our knowledge, this is the first study to examine potentially shared and distinct WM correlates of CD and PLEs during late childhood and adolescence. The findings highlight subtle, time-limited WM alterations and underscore the need for larger, high-risk samples with extended follow-up to clarify neurodevelopmental trajectories underlying early symptoms of antisocial behavior and psychosis. • First study of shared and distinct WM correlates of PLEs and CD in youth. • Cross-sectional WM associations found with PLEs and CD traits. • No longitudinal WM effects or CD × PLE interactions were detected. • Larger, high-risk samples are needed to clarify WM-based trajectories.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.