Individual‐specific functional connectome biomarkers predict schizophrenia positive symptoms during adolescent brain maturation. Issue 5 (2nd December 2020)

• Record Type: Journal Article
• Authors: Fan, Yun‐Shuang; Li, Liang; Peng, Yue; Li, Haoru; Guo, Jing; Li, Meiling; Yang, Siqi; Yao, Meng; Zhao, Jingping; Liu, Hesheng; Liao, Wei; Guo, Xiaonan; Han, Shaoqiang; Cui, Qian; Duan, Xujun; Xu, Yong; Zhang, Yan; Chen, Huafu
• Is Part Of: Human brain mapping. Volume 42:Issue 5(2021)
• Page Start: 1475
• Page End: 1484
• Publication Date: 2020-12-02
• Subjects: adolescent -- biomarker -- functional connectome -- individual level -- schizophrenia; Brain mapping -- Periodicals; 611.81
• DOI: 10.1002/hbm.25307
• Languages: English
• ISSNs: 1065-9471

Abstract

Abstract: Even with an overarching functional dysconnectivity model of adolescent‐onset schizophrenia (AOS), there have been no functional connectome (FC) biomarkers identified for predicting patients' specific symptom domains. Adolescence is a period of dramatic brain maturation, with substantial interindividual variability in brain anatomy. However, existing group‐level hypotheses of AOS lack precision in terms of neuroanatomical boundaries. This study aimed to identify individual‐specific FC biomarkers associated with schizophrenic symptom manifestation during adolescent brain maturation. We used a reliable individual‐level cortical parcellation approach to map functional brain regions in each subject, that were then used to identify FC biomarkers for predicting dimension‐specific psychotic symptoms in 30 antipsychotic‐naïve first‐episode AOS patients (recruited sample of 39). Age‐related changes in biomarker expression were compared between these patients and 31 healthy controls. Moreover, 29 antipsychotic‐naïve first‐episode AOS patients (analyzed sample of 25) were recruited from another center to test the generalizability of the prediction model. Individual‐specific FC biomarkers could significantly and better predict AOS positive‐dimension symptoms with a relatively stronger generalizability than at the group level. Specifically, positive symptom domains were estimated based on connections between the frontoparietal control network (FPN) and salience network and within FPN. Consistent with the neurodevelopmental hypothesis of schizophrenia, the FPN–SN connection exhibited aberrant age‐associated alteration in AOS. The individual‐level findings reveal reproducible FPN‐based FC biomarkers associated with AOS positive symptom domains, and highlight the importance of accounting for individual variation in the study of adolescent‐onset disorders. Abstract : Reliable frontoparietal control network‐based functional connectome biomarkers underlying positive symptom domains could be identified at the individual level, but not at the group level. Replication results indicated relatively better generalization ability of the prediction model by using individual‐based strategy compared with group‐level approach.