Towards a network control theory of electroconvulsive therapy response

Area of Science:

• Computational neuroscience
• Neuroimaging
• Psychiatry

Background:

• Electroconvulsive therapy (ECT) is a highly effective treatment for severe, treatment-resistant depression.
• Predicting individual patient response to ECT remains a significant clinical challenge due to interindividual variability.
• Existing theories lack a mechanistic framework to explain differential ECT outcomes.

Purpose of the Study:

• To develop and validate a quantitative, mechanistic framework for predicting ECT response using Network Control Theory (NCT).
• To establish a formal association between brain network controllability and ECT treatment outcomes.
• To explore the mediating role of the Postictal Suppression Index (PSI) in the relationship between controllability and ECT response.

Main Methods:

• Derived whole-brain modal and average controllability metrics from pre-ECT white-matter structural connectome data.
• Formally associated these controllability metrics with the Postictal Suppression Index (PSI), an ECT seizure quality measure.
• Tested the hypothesis that controllability metrics predict ECT response, mediated by PSI, in a cohort of 50 patients with depression undergoing ECT.

Main Results:

• Whole-brain controllability metrics, derived from pre-ECT structural connectome data, significantly predicted ECT treatment response.
• The hypothesized mediation effect via PSI was empirically confirmed.
• The NCT-based metrics demonstrated predictive performance comparable to or exceeding that of advanced machine learning models using the same connectome data.

Conclusions:

• A control-theoretic framework based on individual brain network architecture can quantitatively predict ECT response.
• This approach provides testable predictions for personalized ECT interventions.
• The findings suggest a potential foundation for a comprehensive, quantitative theory of personalized ECT rooted in control theory.