Individualized, multimodal prediction and preemptive intervention for neuropsychiatric comorbidities in epilepsy: A randomized controlled trial

OBJECTIVE: To evaluate the feasibility and preliminary clinical signals of a risk-stratified care pathway integrating multimodal markers to support the management of epilepsy with co-occurring depressive or anxiety symptoms. METHODS: In this single-center, assessor-blinded, pilot randomized study (January 2023-January 2025), adults with epilepsy and clinically relevant depressive or anxiety symptoms (PHQ-9 ≥ 10 or GAD-7 ≥ 8) were allocated 1:1 to either a prespecified risk-stratified stepped-care pathway or guideline-based usual care. The pathway used a multimodal risk score integrating electrophysiological, neuroimaging, pharmacogenomic, and clinical features to inform tiered monitoring and supportive interventions. Outcomes were assessed at 3 and 6 months. RESULTS: Implementation of the risk-stratified pathway was feasible within routine clinical workflows. Compared with usual care, participants in the intervention group showed greater improvement in depressive symptoms and quality of life at 6 months, alongside favorable trends in seizure outcomes and treatment tolerability. Multimodal biomarkers demonstrated directionally consistent changes over follow-up. All efficacy and biomarker findings are considered exploratory given the pilot design. CONCLUSION: This pilot study demonstrates the feasibility of embedding a multimodal, risk-stratified pathway into epilepsy care and provides hypothesis-generating signals supporting further evaluation in adequately powered multicenter trials. The findings do not establish causal efficacy or warrant immediate changes to clinical practice.