Structural Covariance Network Topology in Individuals with Clinical High Risk for Psychosis

615 

Abstract Submission 

Siwei Liu¹, Paul Thompson², Dennis Hernaus³, Maria Jalbrzikowski⁴, Jimmy Lee⁵, Juan Helen Zhou¹, ENIGMA Clinical High Risk for Psychosis Working Group⁶

¹National University of Singapore, Singapore, Singapore, ²Imaging Genetics Center, Keck School of Medicine of University of Southern California, Los Angeles, CA, ³Maastricht University, Maastricht, Limburg, ⁴Boston Children's Hospital, Boston, MA, ⁵Department of Psychosis, Institute of Mental Health, Singapore, Singapore, ⁶USC's Mark and Mary Stevens Neuroimaging and Informatics Institute, Southern California, CA

Siwei Liu  
National University of Singapore
Singapore, Singapore

Paul Thompson, PhD  
Imaging Genetics Center, Keck School of Medicine of University of Southern California
Los Angeles, CA

Dennis Hernaus  
Maastricht University
Maastricht, Limburg

Maria Jalbrzikowski  
Boston Children's Hospital
Boston, MA

Jimmy Lee  
Department of Psychosis, Institute of Mental Health
Singapore, Singapore

Juan Helen Zhou  
National University of Singapore
Singapore, Singapore

ENIGMA Clinical High Risk for Psychosis Working Group  
USC's Mark and Mary Stevens Neuroimaging and Informatics Institute
Southern California, CA

There are widespread, subtle brain structural abnormalities in individuals at Clinical High Risk (CHR) for developing psychosis [Yung et al., 1996]. While the network approach was proposed to explain deficit propagation in schizophrenia [Chopra et al., 2021; Chopra et al., 2023; Shafiei et al., 2020], studies describing the grey matter structural network properties in CHR [Das et al., 2018] are scarce and suffer from small sample sizes, leaving gaps in understanding brain networks underlying disease progression and symptom development.

Here, we sought to test whether global and network-level structural covariance topographical properties differed between a) CHR and healthy controls, b) CHR who transitioned to psychosis (CHR-T) and those who did not (CHR-NT), and c) CHR subtypes. We further hypothesized that less efficient structural covariance topology would be related to more severe symptoms. Cross-sectional structural scans of 2864 individuals (1842 CHR and 1417 controls) from 31 Enhancing NeuroImaging Genetics through Meta-Analysis (ENIGMA) consortium [Baldwin et al., 2022; Haas et al., 2023; Jalbrzikowski et al., 2021] sites were included.

At the global level, CHR individuals exhibited lower structural covariance (false discovery rate corrected p value q<0.001) and less optimal structural network configuration than controls, including lower global efficiency (q=0.013), local efficiency (q=0.001), and clustering coefficient (q=0.025). At the network level, the system segregation indexes (i.e., network distinctiveness) of distinct frontotemporal surface area networks were higher in CHR-T than CHR-NT and controls (all q<0.013). The system segregation index of frontal cortical thickness network was lower in CHR-T than CHR-NT (q=0.012) and controls (q=0.065). Furthermore, frontal networks also showed unique associations with positive symptoms in CHR-NT (surface area q=0.008) and negative symptoms in CHR-T (thickness q=0.063).

Our findings provide new insights in network-level alterations in CHR, and how such alterations may contribute to symptoms and, ultimately, conversion. Taken together, network topology offers a valuable perspective on pathology across different stages of illness in early psychosis.

Psychiatric (eg. Depression, Anxiety, Schizophrenia) ¹

Connectivity (eg. functional, effective, structural) ²

DISORDERS

MRI

Psychiatric

Psychiatric Disorders

Schizophrenia

STRUCTURAL MRI

Structures