Work

Neuroimaging of Clinical High-Risk Youth From the Kenya Psychosis-Risk Outcomes Study (KePROS): Regional Structural Analyses and Vertexwise Cortical Deviation Mapping (VCDM)

Year: 2026

Type: article

Abstract: BACKGROUND AND HYPOTHESIS: There are no neuroimaging studies of individuals at clinical high risk (CHR) for psychosis in Africa. Using the Kenya Psychosis-Risk Outcomes Study (KePROS), we examined cortical and subcortical morphometry in CHR individuals using conventional regional measures and Vertexwise Cortical Deviation Mapping (VCDM), a surface-based method quantifying extreme cortical thickness deviations. STUDY DESIGN: Seventy-four CHR participants and 42 healthy controls (HCs), aged 18-26, underwent structural magnetic resonance imaging on a 3 T scanner in Nairobi. Regional cortical thickness and subcortical volumes were derived using FreeSurfer. Vertexwise Cortical Deviation Mapping quantified the proportion of cortical vertices exhibiting abnormally thin or thick cortex across standardized deviation thresholds (1-3 SD). Clinical symptoms were assessed over 2-17 months using the Brief Psychiatric Rating Scale, with principal component analysis used to derive latent symptom dimensions. Associations with symptoms, cognition, functioning, and symptom trajectories were examined. STUDY RESULTS: Clinical high risk and HC groups did not differ in regional cortical thickness or subcortical volumes. In contrast, VCDM revealed robust group differences: CHR participants showed greater proportions of extremely thin and thick cortical vertices, particularly at ≥ 2.5 SD thresholds (d = 0.53-0.80). Greater proportions of extremely thin cortical vertices were associated with better social functioning, whereas regional morphometric measures showed no functional associations. Cortical deviation metrics were not associated with symptom severity or trajectories. Smaller nucleus accumbens volume predicted worsening negative-somatic symptoms longitudinally. CONCLUSIONS: Vertexwise deviation metrics are more sensitive than conventional regional measures for detecting cortical abnormalities in CHR individuals, capturing biologically meaningful heterogeneity relevant for stratification.

Source: Schizophrenia Bulletin

DOI: 10.1093/schbul/sbag211