Longitudinal Associations Between Depressive Symptoms and Brain Structure Across Late Childhood and Adolescence: A...

Major depressive disorder is the leading cause of nonfatal disability in youth and disproportionately affects adolescent females. Structural magnetic resonance imaging (MRI) studies of adolescent depression have yielded inconsistent findings, potentially reflecting symptom heterogeneity and rapid developmental changes in brain morphology.

We examined associations between specific depressive symptoms and structural brain MRI measures in 9761 youth (53.1% male, ages 8.32-18.0 years) from the Adolescent Brain Cognitive Development (ABCD) Study across 4 waves. A panel graphical vector autoregression model separated within-person (contemporaneous and temporal) effects from between-person effects. Brain measures included cortical thickness in 4 regions, hippocampal volume, and global surface area. Depressive symptoms included parent-reported depressed mood, anhedonia, lethargy, worthlessness, and sleep problems. Sex differences in network structure were also examined.

Strong within-domain associations were observed among brain measures and symptoms. Cross-domain associations emerged at both within-person and between-person levels. Within persons, higher depressed mood predicted subsequent decreases in global surface area, whereas higher worthlessness predicted subsequent increases; depressed mood was also linked to medial orbitofrontal cortex thickness over time and to lower global surface area concurrently. At the between-person level, individuals with higher average anhedonia showed lower average global surface area. Sex differences were confined to the within-person networks.

Brain-symptom associations were subtle, symptom specific, and present across network levels with differential edge stability. Global surface area emerged as the primary structural correlate. Longitudinal within-person approaches capture the coupling of within-person deviations from expected brain and symptom trajectories, complementing between-person approaches in understanding brain-depression co-fluctuation during adolescence.