Independent Versus Joint Effects of Polygenic or Family-Based Schizophrenia Risk in Diverse Ancestry Youth in the ABCD Study

Abstract

Background: Behavioral and cognitive signs often precede schizophrenia (SCZ) onset, and measures of SCZ risk, such as family history of psychosis (FH) and SCZ polygenic risk scores (SCZ-PRS), have been linked to cognition and mental health symptoms in childhood. However, studies using SCZ-PRS have largely focused on European ancestry youth. In addition, the extent to which FH reflects common variant based polygenic risk for SCZ, or broader risk factors for SCZ is unclear. We investigated whether SCZ-related behavioral and cognitive signs and symptoms are associated with SCZ-PRS or FH across diverse ancestry children in the Adolescent Brain Cognitive Development (ABCD) study.Method: Using baseline ABCD data, we analyzed associations of FH and SCZ-PRS, with cognitive, behavioral and emotional measures derived from the NIH-Toolbox, Child Behavior Checklist (CBCL), and Kiddie Schedule for Affective Disorders and Schizophrenia (KSADS) for 9,636 children (mean age=9.92 yrs, 47.4% female), specifically 5,636 European, 2,093 African, and 1,477 Admixed American ancestry individuals. SCZ-PRS were refined using PRS-CSx and normalized within ancestry groups, before being assessed for associations with the different phenotypes. Associations were adjusted for age, sex, study site, and genetic relatedness. Results: FH showed positive associations with SCZ-PRS across all groups, with a significant association in European ancestry youth (b=0.39, FDR-p=0.02). FH was linked to higher CBCL scores across all ancestries (b range=2.90-4.81, FDR-p<0.001), and higher PQB Distress scores in European ancestry youth (b=3.74, FDR-p=0.01). SCZ-PRS was associated with lower NIH-TB total cognition across ancestries (b=-0.43, FDR-p=0.023), particularly in Europeans (b=-0.78, FDR-p=0.001), but not with CBCL or PQB Distress. FH was associated with higher odds of depressive disorders, conduct disorder, anxiety, PTSD, and ADHD (OR=2.17-5.09, FDR-p<0.001), while SCZ-PRS was only linked to a greater likelihood of depressive disorders in Admixed American youth (OR=1.37, FDR-p=0.02). Income-to-needs adjustment slightly altered some associations, but overall patterns remained unchanged. Conclusion: This study underscores the independent effects of FH and SCZ-PRS on childhood SCZ-related emotional, cognitive and behavioral signs across ancestries. FH was consistently linked to greater psychopathology but not cognitive functioning, while SCZ-PRS showed strong associations with cognition and specific emotional and behavioral symptoms in European ancestry youth. Findings highlight the complementary role of FH and PRS in early SCZ risk assessment and call for improved PRS accuracy across diverse populations.