TY - JOUR
T1 - Genome-wide association study of phenotypes measuring progression from first cocaine or opioid use to dependence reveals novel risk genes
AU - Sherva, Richard
AU - Zhu, Congcong
AU - Wetherill, Leah
AU - Edenberg, Howard J.
AU - Johnson, Emma
AU - Degenhardt, Louisa
AU - Agrawal, Arpana
AU - Martin, Nicholas G.
AU - Nelson, Elliot
AU - Kranzler, Henry R.
AU - Gelernter, Joel
AU - Farrer, Lindsay A.
N1 - Publisher Copyright:
© The Author(s) 2021.
PY - 2021
Y1 - 2021
N2 - Aim: Substance use disorders (SUD) result in substantial morbidity and mortality worldwide. Opioids, and to a lesser extent cocaine, contribute to a large percentage of this health burden. Despite their high heritability, few genetic risk loci have been identified for either opioid or cocaine dependence (OD or CD, respectively). A genome-wide association study of OD and CD related phenotypes reflecting the time between first self-reported use of these substances and a first DSM-IV dependence diagnosis was conducted. Methods: Cox proportional hazards regression in a discovery sample of 6,188 African-Americans (AAs) and 6,835 European-Americans (EAs) participants in a genetic study of multiple substance dependence phenotypes were used to test for association between genetic variants and these outcomes. The top findings were tested for replication in two independent cohorts. Results: In the discovery sample, three independent regions containing variants associated with time to dependence at P < 5 × 10-8 were identified, one (rs61835088 = 1.03 × 10-8) for cocaine in the combined EA-AA meta-analysis in the gene FAM78B on chromosome 1, and two for opioids in the AA portion of the sample in intergenic regions of chromosomes 4 (rs4860439, P = 1.37 × 10-8) and 9 (rs7032521, P = 3.30 × 10-8). After meta-analysis with data from the replication cohorts, the signal at rs61835088 improved (HR = 0.87, P = 3.71 × 10-9 and an intergenic SNP on chromosome 21 (rs2825295, HR = 1.14, P = 2.57 × 10-8) that missed the significance threshold in the AA discovery sample became genome-wide significant (GWS) for CD. Conclusions: Although the two GWS variants are not in genes with obvious links to SUD biology and have modest effect sizes, they are statistically robust and show evidence for association in independent samples. These results may point to novel pathways contributing to disease progression and highlight the utility of related phenotypes to better understand the genetics of SUDs.
AB - Aim: Substance use disorders (SUD) result in substantial morbidity and mortality worldwide. Opioids, and to a lesser extent cocaine, contribute to a large percentage of this health burden. Despite their high heritability, few genetic risk loci have been identified for either opioid or cocaine dependence (OD or CD, respectively). A genome-wide association study of OD and CD related phenotypes reflecting the time between first self-reported use of these substances and a first DSM-IV dependence diagnosis was conducted. Methods: Cox proportional hazards regression in a discovery sample of 6,188 African-Americans (AAs) and 6,835 European-Americans (EAs) participants in a genetic study of multiple substance dependence phenotypes were used to test for association between genetic variants and these outcomes. The top findings were tested for replication in two independent cohorts. Results: In the discovery sample, three independent regions containing variants associated with time to dependence at P < 5 × 10-8 were identified, one (rs61835088 = 1.03 × 10-8) for cocaine in the combined EA-AA meta-analysis in the gene FAM78B on chromosome 1, and two for opioids in the AA portion of the sample in intergenic regions of chromosomes 4 (rs4860439, P = 1.37 × 10-8) and 9 (rs7032521, P = 3.30 × 10-8). After meta-analysis with data from the replication cohorts, the signal at rs61835088 improved (HR = 0.87, P = 3.71 × 10-9 and an intergenic SNP on chromosome 21 (rs2825295, HR = 1.14, P = 2.57 × 10-8) that missed the significance threshold in the AA discovery sample became genome-wide significant (GWS) for CD. Conclusions: Although the two GWS variants are not in genes with obvious links to SUD biology and have modest effect sizes, they are statistically robust and show evidence for association in independent samples. These results may point to novel pathways contributing to disease progression and highlight the utility of related phenotypes to better understand the genetics of SUDs.
KW - FAM78B
KW - GWAS
KW - Substance use disorders
KW - cocaine use disorders
KW - genetics
KW - opioid use disorders
UR - http://www.scopus.com/inward/record.url?scp=85113726389&partnerID=8YFLogxK
U2 - 10.37349/emed.2021.00032
DO - 10.37349/emed.2021.00032
M3 - Article
C2 - 34124712
AN - SCOPUS:85113726389
SN - 2692-3106
VL - 2
SP - 60
EP - 73
JO - Exploration of Medicine
JF - Exploration of Medicine
IS - 1
ER -