Toward Clinical Implementation of Polygenic Scores for Substance Use Disorders: A Multi-Ancestry Study

Area of Science:

• Genetics
• Psychiatry
• Pharmacogenomics

Background:

• Substance use disorders (SUDs) represent a significant public health challenge.
• Developing accurate predictive tools for SUDs is crucial for early intervention and personalized treatment.
• Genetic factors play a substantial role in an individual's susceptibility to SUDs.

Purpose of the Study:

• To develop and validate clinically relevant polygenic scores (PGS) for alcohol (AUD), cannabis (CanUD), opioid (OUD), tobacco (TUD), and polysubstance use disorders (polySUD).
• To assess the performance of these PGS across African (AA), European (EA), and Latinx (LA) ancestry populations.
• To evaluate the clinical utility of PGS for risk prediction in diverse populations.

Main Methods:

• Utilized genome-wide association study (GWAS) summary statistics and multiple PGS methods.
• Developed and evaluated SUD PGS in the Indiana Biobank (IB) and validated top performers in the All of Us Research Program (AOU).
• Defined cases and controls using ICD-9/10 codes, with clinical relevance set at an odds ratio (OR) >=2 for the highest PGS group.

Main Results:

• In EA and LA populations, all PGS achieved clinically relevant performance in both IB and AOU datasets (ORs: 2.00-9.10).
• In AA populations, PGS met the clinical relevance threshold in IB (ORs: 2.02-2.71) but not in AOU (ORs: 1.28-1.56).
• Opioid use disorder (OUD) PGS demonstrated the strongest associations, followed by cannabis use disorder (CanUD) and polysubstance use disorder (polySUD) PGS.

Conclusions:

• PGS show clinically meaningful risk prediction for SUDs in EA and LA populations, supporting potential clinical implementation for population-level screening.
• The reduced performance of PGS in AA populations highlights the critical need for increased genetic research within this demographic.
• Further research is warranted to improve PGS accuracy and utility across all ancestral groups for equitable SUD risk assessment.