Striatal Connectivity and Clinical Outcome in Psychosis

In this proposed study, the study will examine treatment-related effects on functional brain circuitry in first episode schizophrenia. Converging lines of evidence suggest a key role for striatal disconnectivity in the pathophysiology of psychosis. The proposed study will utilize resting state functional magnetic resonance imaging (rs-fMRI), as well as fMRI tasks derived from the Research Domain Criteria (RDoC) framework, to: 1) develop and validate a prognostic biomarker to predict antipsychotic treatment response; and 2) to model the underlying neural circuitry changes associated with state changes in psychotic symptomatology. As a prognostic biomarker, a neuroimaging assay of striatal connectivity can potentially provide a clinically useful tool to advance the goal of precision medicine. As a longitudinal index of symptom change, our model can serve as an objective index against which to measure potential efficacy of newly developed antipsychotic treatments. A large, well-characterized cohort of patients presenting with a first episode active psychosis (regardless of DSM diagnosis) will be recruited, along with matched controls. The study will utilize two well-validated fMRI tasks capturing two portions of the positive valence system: probabilistic category learning and reward responsiveness; these tasks are designed to interrogate dorsal and ventral corticostriatal circuits, respectively. The design will be longitudinal, with two scanning sessions performed for each patient: at baseline, and after 12 weeks of treatment. Treatment will be standardized across all patients to reduce potential confounds, and healthy controls will also be scanned at baseline and 12 weeks in order to control for effects of time and practice. Level of psychotic symptomatology (hallucinations, delusions, and thought disorder) will be measured at regular intervals using a comprehensive battery of rating scales. As secondary measures, electroencephalography (EEG) will be performed coinciding with neuroimaging on a subset of patients who provide consent. We will utilize Kaplan-Meier estimators and hierarchical linear modeling to examine the association of baseline striatal connectivity, and changes in connectivity over time, with clinical response of psychotic symptoms to antipsychotic treatment. Deliverables will include both baseline and longitudinal biomarkers that can subsequently be tested in broader, more heterogeneous populations of patients with psychosis.