Multi-trait genome-wide association study of opioid addiction: OPRM1 and beyond
Nathan Gaddis, Ravi Mathur, Jesse A. Marks, Linran Zhou, Bryan C. Quach, Alex Waldrop, Orna Levran, Arpana Agrawal, Matthew Randesi, Miriam Adelson, Paul W. Jeffries, Nicholas G. Martin, Louisa Degenhardt, Grant W. Montgomery, Leah Wetherill, Dongbing Lai, Kathleen K. Bucholz, Tatiana Foroud, Bernice Porjesz, Valgerður Rúnarsdóttir, Þórarinn Tyrfingsson, Guðmundur Einarsson, Daníel F. Guðbjartsson, Bradley T. Webb, Richard C. Crist, Henry R. Kranzler, Richard Sherva, Hang Zhou, Gary Kenneth Hulse, Dieter B. Wildenauer, Erin Kelty, John Attia, Elizabeth Holliday, Mark McEvoy, Rodney J. Scott, Sibylle G. Schwab, Brion S. Maher, Richard Gruza, Mary Jeanne Kreek, Elliot C. Nelson, Thorgeir E. Thorgeirsson, Kāri Stefánsson, Wade H. Berrettini, Joel Gelernter, Howard J. Edenberg, Laura J. Bierut, Dana B. Hancock, Eric O. Johnson
Abstract
Abstract Opioid addiction (OA) is moderately heritable, yet only rs1799971, the A118G variant in OPRM1 , has been identified as a genome-wide significant association with OA and independently replicated. We applied genomic structural equation modeling to conduct a GWAS of the new Genetics of Opioid Addiction Consortium (GENOA) data together with published studies (Psychiatric Genomics Consortium, Million Veteran Program, and Partners Health), comprising 23,367 cases and effective sample size of 88,114 individuals of European ancestry. Genetic correlations among the various OA phenotypes were uniformly high (r g > 0.9). We observed the strongest evidence to date for OPRM1 : lead SNP rs9478500 ( p = 2.56 × 10 –9 ). Gene-based analyses identified novel genome-wide significant associations with PPP6C and FURIN . Variants within these loci appear to be pleiotropic for addiction and related traits.