ASSOCIATIONS OF POLYGENIC LIABILITY FOR EXTERNALIZING AND SUBSTANCE USE WITH PRENATAL AND ADOLESCENT SUBSTANCE USE

Prenatal substance exposure (i.e., maternal substance use during pregnancy) and adolescent substance use represent distinct but developmentally linked periods of vulnerability, with implications for maternal health, fetal development, and later offspring substance use trajectories. Genetic liability for externalizing (EXT) behaviors and substance use disorders (SUD) may increase risk across these stages by elevating the likelihood of maternal substance use during pregnancy and, independently, by predisposing offspring to earlier initiation of substance use during adolescence. However, few studies have examined, within a unified analytic framework, how shared genetic liability for EXT and SUD contributes to risk across developmental stages. Here, we leverage genetic and phenotypic data from two independent longitudinal cohorts, the Healthy Brain and Child Development (HBCD) Study (Release 2.0; N = 1,170 mothers) and the Adolescent Brain Cognitive Development (ABCD) Study (Release 6.1; N = 10,977 adolescents, ages 9-17), to evaluate how genetic liability is associated with prenatal substance use in mothers and substance use initiation in adolescents, providing complementary evidence on how these risk processes operate across generations. In HBCD, maternal polygenic scores (PGS) for EXT and tobacco use disorder (TUD), but not cannabis use disorder (CUD), were associated with increased likelihood of prenatal nicotine use (EXT: aOR = 1.61, 95% CI: 1.08-2.40; TUD: aOR = 1.62, 95% CI: 1.10-2.37; CUD: aOR = 1.01, 95% CI: 0.71-1.46). For prenatal cannabis use, only EXT PGS was associated with increased risk (e.g., aOR = 1.22, 95% CI: 1.05-1.41). In ABCD, higher adolescent PGS for EXT, TUD, and CUD, as well as prenatal substance exposure, were associated with earlier initiation of nicotine and cannabis use in time-to-event models. For nicotine initiation, EXT PGS (HR = 1.45, 95% CI: 1.38-1.54), TUD PGS (HR = 1.24, 95% CI: 1.18-1.31), and CUD PGS (HR = 1.21, 95% CI: 1.14-1.27) were associated with earlier onset. Female sex (HR ≈ 1.49-1.50, 95% CI: 1.34-1.67) and prenatal substance exposure (HR ≈ 1.55-1.62, 95% CI: 1.34-1.86) were also associated with earlier nicotine initiation. For cannabis initiation, EXT PGS (HR = 1.53, 95% CI: 1.42-1.65), TUD PGS (HR = 1.24, 95% CI: 1.17-1.35), and CUD PGS (HR = 1.30, 95% CI: 1.21-1.40) were associated with earlier onset. Prenatal substance exposure showed a stronger association with earlier cannabis initiation (HR ≈ 1.96-1.99, 95% CI: 1.54-2.44), whereas sex was not significantly associated. Together, these findings indicate that both broad liability for EXT and substance-specific genetic liability operate across developmental stages, linking maternal risk during pregnancy to adolescent nicotine and cannabis use trajectories. This work highlights the value of integrating genetic data across longitudinal cohorts to characterize developmental patterns of risk and inform early prevention strategies.