Early deviations from normative brain morphology and cortical microstructure in schizophrenia spectrum disorders

In schizophrenia spectrum disorders (SSD), structural alterations of the gray matter (GM) and white matter (WM) have been widely described. However, the complex interplay between early disease-related changes and ongoing brain maturation challenges our ability to identify early biomarkers. In this study, we investigated structural abnormalities and their association with symptoms in a drug-naïve or minimally medicated sample comprising 113 patients with SSD and 112 neurotypical controls. Specifically, we derived centile scores using normative modelling from cortical thickness (CT), and subcortical volumes derived from structural MRI images, and diffusion tensor imaging-derived (DTI) WM tract fractional anisotropy (FA). In addition, we derived raw cortical mean diffusivity (cMD) metrics from DTI. Compared to controls, SSD participants showed reduced CT centiles, ventricular enlargement, and subcortical centile reductions in the hippocampus, thalamus, amygdala, and nucleus accumbens. SSD was also associated with widespread increases in cMD. We also explored associations among these structural markers, identifying significant relationships between CT centiles and raw cMD, as well as between subcortical centiles and FA tract-based centiles. No interaction with SSD diagnosis was found. Furthermore, positive symptoms correlated negatively with CT and FA tract-based centiles, and showed widespread positive associations with raw cMD, whereas negative symptoms showed no associations. These findings underscore multimodal brain abnormalities that originate early in the course of SSD and their distinct associations with symptoms. Our results support the potential of both normative modelling and diffusion imaging markers to identify individualized early brain changes in SSD Among these, cMD emerges as a potential marker of SSD-related microstructural alterations.