83. REPLICABLE ASSOCIATIONS OF POLYGENIC SCORES WITH SEVERE MENTAL ILLNESS ONSET IN DIVERSE POPULATIONS

Background Polygenic scores (PGS) for psychiatric traits are associated with the onset of severe mental illness (SMI) among youth of European ancestry; however, the replicability of these associations in ethnically diverse populations remains unclear. This study examined whether associations between PGS and SMI onset identified in a primarily European sample replicate in a diverse cohort of youth. Methods Participants were drawn from international cohorts enriched for familial risk of SMI across Australia, Brazil, Canada, the Netherlands, Spain, the United Kingdom, and the United States. Youth were followed from a mean age of 11.8 years (SD = 4.3) to 21.5 years (SD = 4.9). The discovery sample (n = 1,862; primarily European ancestry) and replication sample (n = 2,508; ethnically diverse) included 459 (24.7%) and 647 (25.8%) SMI onsets, respectively. Polygenic scores for schizophrenia, major depressive disorder (MDD), bipolar disorder (BD), anxiety, neuroticism, attention-deficit/hyperactivity disorder (ADHD), addiction risk factor, suicidal ideation, sleep duration, and physical activity were calculated using PRSice-2, LDpred2-auto, and SBayesRC and averaged across methods. Familial high-risk (FHR) status was established via semi-structured interviews with relatives. SMI onset was defined as a new diagnosis of MDD, BD, or schizophrenia spectrum disorder. Cox regression models examined associations between PGS and time to SMI onset. Results In the discovery sample, PGS for MDD, BD, schizophrenia, anxiety, neuroticism, ADHD, addiction risk factor, suicidal ideation, short sleep, and lower physical activity were associated with SMI onset. In the replication sample, PGS for MDD, anxiety, neuroticism, lower physical activity, and suicidal ideation were independently associated with SMI onset. In the combined sample (n = 4,370), higher PGS for suicidal ideation (HR = 1.37, 95% CI 1.23-1.52), MDD (HR = 1.34, 95% CI 1.20-1.48), anxiety (HR = 1.29, 95% CI 1.18-1.40), ADHD (HR = 1.18, 95% CI 1.10-1.28), neuroticism (HR = 1.18, 95% CI 1.10-1.26), short sleep (HR = 1.14, 95% CI 1.06-1.23), addiction risk factor (HR = 1.13, 95% CI 1.05-1.22), and BD (HR = 1.09, 95% CI 1.02-1.17) were associated with increased risk of SMI onset, while higher physical activity PGS was associated with reduced risk (HR = 0.85, 95% CI 0.80-0.92). Discussion Five PGS demonstrated replicable associations with SMI onset across European and ethnically diverse youth. In the combined sample, associations of eight PGS made a unique contribution to early risk identification beyond family history and early psychopathology, and higher scores were generally linked to increased risk and earlier illness onset. These findings suggest that PGS may help identify youth at elevated risk for SMI and support their potential role in improving early risk stratification and prevention strategies.