Abstract
OBJECTIVE: Pediatric anxiety disorders are common and predict adult psychopathology, yet current treatments, such as cognitive-behavioral therapy (CBT), produce lasting remission in less than 50% of affected youths. To support the search for improved, mechanistically grounded interventions, this study evaluated neural efficiency, defined as similarity in functional connectivity between a threat task and rest, as a potential biomarker. The study evaluated neural efficiency in relation to anxiety diagnosis and treatment response. METHODS: The authors compared 103 youths with an anxiety disorder diagnosis (mean age, 12.5 years [SD=2.91], 62% female) to 103 youths with no psychiatric diagnosis (mean age, 13.4 years [SD=2.58], 53% female). Participants completed functional MRI while resting and during a dot-probe task with threatening faces. Neural efficiency was calculated as partial correlations between intrinsic and task-related functional connectivity patterns across the whole brain. Four-month test-retest reliability as well as relationships with anxiety and response to exposure-based CBT were examined. RESULTS: Neural efficiency demonstrated satisfactory test-retest reliability (intraclass correlation coefficient=0.65) in healthy youths over a period of 11 to 18 weeks. Neural efficiency was significantly negatively related to anxiety as both a diagnostic category (t=2.62, d=0.29) and a symptom dimension (r=-0.18). Although it did not change after CBT, lower neural efficiency at baseline was significantly associated with poorer treatment response in a subset of 80 anxious youths who underwent CBT (β=-11.88, χ(2)=9.20). CONCLUSIONS: Neural efficiency, measured as network reconfiguration between rest and task, holds promise as a biomarker in pediatric anxiety. Its association with CBT response suggests that it might aid in patient stratification and offer a target for interventions aimed at enhancing CBT efficacy.