Brain features of nearly drug‐naïve female monozygotic twins with first‐episode schizophrenia and the classification accuracy of brain feature patterns: A pilot study

Abstract

Background Data on differences in brain features between monozygotic (MZ) twins with and without schizophrenia are scarce. Methods We compared brain features of female MZ twins with and without first‐episode schizophrenia and healthy controls (n = 20 each). Voxel‐based morphometry and tract‐based spatial statistics were used to analyze differences in brain structure. Whole‐brain effective connectivity (EC) and functional connectivity (FC) networks were constructed using resting‐state functional magnetic resonance imaging (rs‐fMRI) data. Results Female twins with schizophrenia exhibited abnormal gray matter volume (GMV) in the basal ganglia and prefrontal and parietal cortices, impairments in the arcuate fasciculus, and significant disruptions (primarily decreases) in nine EC networks. They exhibited rs‐EC alterations involving the limbic areas and subcortex. Combined rs‐EC and rs‐FC data distinguished twins with first‐episode schizophrenia with high accuracy. Combined consideration of structural and functional features enabled the distinction of female MZ twins with schizophrenia from those without schizophrenia and healthy controls with 100% accuracy. Conclusions Female MZ twins with schizophrenia exhibited increased GMV, white matter impairment, and disruptions in EC and FC networks. The combination of rs‐EC + rs‐FC data could distinguish female twins with schizophrenia from twins without schizophrenia and healthy controls with 97.4% accuracy, and the addition of structural brain features yielded a 100% accuracy rate. These findings may provide pivotal insight for further study of the mechanisms underlying schizophrenia.

Brain‐based sex differences in schizophrenia: A systematic review of fMRI studies

“…After removing duplicates ( n = 138), the remaining 653 articles were screened for abstracts and titles, which resulted in the exclusion of 626 studies. After full‐text assessment, 17 additional papers were excluded due to: the population under study was not schizophrenia ( n = 6) (Anderson et al, 2013 ; Brébion et al, 2021 ; Coman et al, 2010 ; Garcia et al, 2020 ; Sen & Parhi, 2019 ; Sokhi et al, 2005 ); evaluating only one sex ( n = 8) (Abel et al, 2003 ; Das et al, 2012 ; Kircher et al, 2008 ; Kubicki et al, 2003 ; Kumari et al, 2009 ; Sommer et al, 2003 ; Zhang et al, 2021 ; Zhou et al, 2016 ); not being an original paper ( n = 1) (Mendrek, 2013 ); not reporting fMRI findings ( n = 1) (Bang et al, 2019 ); no full‐texts available ( n = 1) (Mendrek, 2007 ). By screening reference lists of the included articles, we also identified two additional articles (Iraji et al, 2022 ; Lakis & Mendrek, 2013 ) not detected in the literature search using our a priori keywords.…”