PT  - JOURNAL ARTICLE
AU  - Peter B. Barr
AU  - Albert Ksinan
AU  - Jinni Su
AU  - Emma C. Johnson
AU  - Jacquelyn L. Meyers
AU  - Leah Wetherill
AU  - Antti Latvala
AU  - Fazil Aliev
AU  - Grace Chan
AU  - Samuel Kuperman
AU  - John Nurnberger
AU  - Chella Kamarajan
AU  - Andrey Anokhin
AU  - Arpana Agrawal
AU  - Richard J. Rose
AU  - Howard J. Edenberg
AU  - Marc Schuckit
AU  - Jaakko Kaprio
AU  - Danielle M. Dick
TI  - Polygenic Prediction of Substance Use Disorders in Clinical and Population Samples
AID  - 10.1101/748038
DP  - 2019 Jan 01
TA  - bioRxiv
PG  - 748038
4099  - http://biorxiv.org/content/early/2019/08/30/748038.short
4100  - http://biorxiv.org/content/early/2019/08/30/748038.full
AB  - Genome-wide, polygenic risk scores (PRS) have emerged as a useful way to characterize genetic liability using genotypic data. There is growing evidence that PRS may prove useful to identify those at increased risk for developing certain diseases. The current utility of PRS in relation to alcohol use disorders (AUD) remains an open question. Using data from both a population-based sample [the FinnTwin12 (FT12) study] and a high risk sample [the Collaborative Study on the Genetics of Alcoholism (COGA)], we examined the association between PRSs derived from genome-wide association studies (GWASs) of 1) alcohol dependence/alcohol problems, 2) alcohol consumption, and 3) risky behaviors with AUD and other substance use disorder (SUD) symptoms. Individuals in the top 20%, 10%, and 5% of PRSs had increasingly greater odds of having an AUD compared to the lower end of the continuum in both COGA (80th % OR = 1.95; 90th % OR = 2.03; 95th % OR = 2.13) and FT12 (80th % OR = 1.77; 90th % OR = 2.27; 95th % OR = 2.39). Those in the top 5% reported greater levels of licit (alcohol and nicotine) and illicit (cannabis) SUD symptoms. PRSs can predict elevated risk for SUD in independent samples. However, clinical utility of these scores in their current form is modest. As these scores become more predictive of SUD, they may become useful to practitioners. Improvement in predictive ability will likely be dependent on increasing the size of well-phenotyped discovery samples.