A phenome-wide association study of externalizing polygenic risk in the Adolescent Brain Cognitive Development (ABCD) study
Sarah E. Paul, Nathaniel Thomas, Maia Choi, Nastasia S. Courchesne-Krak, I-Tzu Hung, Travis T. Mallard, Nicole R. Karcher, Aaron J. Gorelik, Genevieve F. Dash, John Meredith, Sarah Brislin, Abraham A. Palmer, Sandra Sanchez-Roige, Danielle Dick, Alexander S. Hatoum, Jill A. RabinowitzAbstract
Background
Externalizing disorders encompass a range of disinhibited behaviors that typically manifest in youth and are linked with adverse psychological and social outcomes. Polygenic liability to externalizing behaviors has been associated with psychiatric and disease phenotypes in adults. Identifying correlates of genetic risk in youth—prior to the emergence of more impairing manifestations of externalizing behaviors (e.g. substance use disorders)—could improve our etiological understanding and inform targeted prevention efforts.
Methods
We conducted a phenome-wide association study (PheWAS) to test whether externalizing (EXT) polygenic scores (PGS) were associated with up to 1,697 psychosocial, behavioral, environmental, and health-related phenotypes among youth of European-like genetic ancestry. PheWAS data came from the baseline ( n = 4,931, mean age = 9.93 ± 0.63 years, 46.9% female) and two-year follow-up ( n = 4,507, 12.02 ± 0.66 years, 47.0% female) data collection waves of the Adolescent Brain Cognitive Development Study SM (ABCD Study®). Secondary analyses examined within-family effects and moderation by biological sex.
Results
EXT PGS was associated with 268 and 352 phenotypes at baseline and follow-up, respectively, following Bonferroni correction at each wave. The strongest associations were observed with externalizing behaviors, screen time/media use, caffeine consumption, sleep disturbances, and poor school performance. Effects were attenuated by 59%–64% within family, on average, with five predominantly externalizing phenotypes remaining significantly associated with within-family PGS. There was limited evidence of moderation by biological sex.
Conclusions
These results broadly characterize the early phenotypic correlates of an externalizing polygenic score and can be used to generate hypotheses about phenotypes that might represent mechanisms or future prevention targets.