Multimodal Clustering Analysis of meta-analytically derived brain regions in Schizophrenia Spectrum Disorders

BackgroundSchizophrenia spectrum disorders (SSD) present substantial clinical and biological heterogeneity, impeding advances in diagnostic precision and personalised treatment. Despite consistent evidence of brain alterations, the identification of subgroups reflecting the disorders complexity remains challenging, due to interindividual variability and the inherent limitations of studying neuroimaging modalities in isolation. We developed a novel meta-analytically anchored clustering approach integrating structural (T1-weighted, diffusion tensor imaging) and functional (resting-state fMRI) brain data to derive multimodal subgroups. MethodsWe analysed data from 146 SSD patients and 129 healthy controls, initially replicating gray matter volume alterations reported in meta-analyses. We then examined these regions across neuroimaging modalities and performed multimodal clustering on 104 patients with complete data across modalities. The clusters were probed for validity and robustness and characterised by clinical features, polygenic risk scores (PRS) and gene expression pathways. ResultsWe successfully replicated gray matter volume alterations in 30/33 regions, with approximately half showing significant cross-modal abnormalities. Thalamic dysfunction emerged as particularly prominent. Clustering identified five distinct SSD subgroups with divergent brain-symptom-genetics profiles. Most notably, one subgroup exhibited pronounced white matter decline with aberrant neuroinflammation and myelin gene expression, while another subgroup showed increased gray matter volumes and elevated PRS for intracranial volume, even exceeding levels in our healthy controls. ConclusionThese findings highlight the relevance of analyzing meta-analytically validated regions across multiple imaging modalities. Our clustering approach successfully identified neurobiologically distinct SSD subgroups, offering a promising framework for addressing the challenge of heterogeneity in severe mental illness and advancing precision psychiatry.