Classification of first-episode psychosis using cortical thickness: A large multicenter MRI study.

Machine learning classifications of first-episode psychosis (FEP) using neuroimaging have predominantly analyzed brain volumes. Some studies examined cortical thickness, but most of them have used parcellation approaches with data from single sites, which limits claims of generalizability. To address these limitations, we conducted a large-scale, multi-site analysis of cortical thickness comparing parcellations and vertex-wise approaches. By leveraging the multi-site nature of the study, we further investigated how different demographical and site-dependent variables affected predictions. Finally, we assessed relationships between predictions and clinical variables. 428 subjects (147 females, mean age 27.14) with FEP and 448 (230 females, mean age 27.06) healthy controls were enrolled in 8 centers by the ClassiFEP group. All subjects underwent a structural MRI and were clinically assessed. Cortical thickness parcellation (68 areas) and full cortical maps (20,484 vertices) were extracted. Linear Support Vector Machine was used for classification within a repeated nested cross-validation framework. Vertex-wise thickness maps outperformed parcellation-based methods with a balanced accuracy of 66.2% and an Area Under the Curve of 72%. By stratifying our sample for MRI scanner, we increased generalizability across sites. Temporal brain areas resulted as the most influential in the classification. The predictive decision scores significantly correlated with age at onset, duration of treatment, and positive symptoms. In conclusion, although far from the threshold of clinical relevance, temporal cortical thickness proved to classify between FEP subjects and healthy individuals. The assessment of site-dependent variables permitted an increase in the across-site generalizability, thus attempting to address an important machine learning limitation.

Disturbed glutathione antioxidative defense is associated with structural brain changes in neuroleptic-naïve first-episode psychosis patients.

BACKGROUND: Oxidative stress and impaired antioxidant defense are reported in schizophrenia and are thought to be associated with disturbed neurodevelopment, brain structural alterations, glutamatergic imbalance, negative symptomatology, and cognitive impairment. To test some of these assumptions we investigated the glutathione (GSH) antioxidant defense system (AODS) and brain structural abnormalities in drug-naïve individuals with first acute episode of psychosis (FEP). METHOD: The study involved 27 drug-naïve FEP patients and 31 healthy controls (HC). GSH AODS markers and TBARS (thiobarbituric acid reactive substances) were measured in blood plasma and erythrocytes. High-resolution T1-weighted 3T MRI were acquired from all subjects. To investigate brain structural abnormalities and effects of illness on interactions between GSH metabolites or enzyme activities and local grey matter density, voxel-based morphometry (VBM) with the computational anatomy toolbox (CAT12) was used. Symptomatology was assessed using the Positive and Negative Syndrome Scale (PANSS) and the Symptom Checklist 1990 revised (SCL-90-R). RESULTS: (i) In FEP patients, glutathione reductase activity (GSR) was lower than in the HC group. GSR activity in plasma was inversely correlated with SCL-90-R scores of depression and PANSS scores of the negative symptom subscale. (ii) A reduction of GM was observed in left inferior frontal, bilateral temporal, as well as parietal cortices of FEP patients. (iii) Interaction analyses revealed an influence of illness on GSR/GM associations in the left orbitofrontal cortex (BA 47). CONCLUSION: Our findings support the notion of altered GSH antioxidative defense in untreated acute psychosis as a potential pathomechanism for localized brain structural abnormalities. This pathology relates to a key brain region of social cognition, affective motivation control and decision making, and is clinically accompanied by depressive and negative symptoms.