Association analyses of 100,720 individuals reveal new loci associated with body fat percentage providing new insights in related cardiometabolic traits. Y. Lu1, F. Day2, S. Gustafsson3, T. Kilpeläinen4, R. Loos1,2 on behalf of the Genetics of Body Fat Consortium 1) The Charles Bronfman Institute for Personalized Medicine, Icahn School of Medicine at Mount Sinai, New York, NY 10029, USA; 2) MRC Epidemiology Unit, University of Cambridge, Institute of Metabolic Science, Addenbrookes Hospital, Hills Road, Cambridge, CB2 0QQ, UK; 3) 3, Department of Medical Sciences, Molecular Epidemiology and Science for Life Laboratory, Uppsala University, Uppsala 75185, Sweden; 4) Novo Nordisk Foundation Center for Basic Metabolic Research, University of Copenhagen, Copenhagen 2100, Copenhagen, Denmark.

   Large-scale meta-analyses of genome-wide association studies (GWAS) for readily-available adiposity measures, such as BMI, waist-to-hip ratio adjusted for BMI (WHRadjBMI) and obesity risk have identified at least 75 loci that contribute to body weight and fat distribution in adults and children of diverse ancestry. While these commonly studied adiposity traits are easily assessed in large populations and thus allow statistically well-powered meta-analyses, they represent heterogeneous phenotypes and do not distinguish between lean and fat mass. To increase our understanding of the genetic basis of adiposity and its links to cardiometabolic disease risk, we conducted a genome-wide association meta-analysis of body fat percentage (BF%), which more accurately assesses adiposity. In our primary meta-analysis, we combined the results of genetic associations with BF% for up to 100,720 individuals from 43 GWAS (n up to 76,138) and 13 MetaboChip studies (n up to 24,582), which were predominantly of European ancestry (n up to 89,300). In secondary analyses, we stratified by sex and/or ancestry. For loci that reached genome-wide significance (P<5x10-8), we examined their association with cardiometabolic traits. SNPs in 12 loci reached genome-wide significance. Two (near IRS1, SPRY2) of the 12 loci had been identified in previous GWAS for BF%, and six (in or near FTO, MC4R, TMEM18, TOMM40, TUFM/SH2B1, and SEC16B) have previously been reported for BMI. Four of the 12 loci, near GRB14/COBLL1, IGF2BP1, PLA2G6, and CRTC1, were novel. SNPs in the 12 established loci increase body fat percentage by 0.24 to 0.51 SD/allele, explaining 0.58% of the BF% variance (between 0.03- 0.13% per locus). Some loci had association signatures with other cardiometabolic traits that were discordant with observed phenotypic correlations. E.g. the BF% increasing allele of the COBLL1/GBR14 locus was associated with reduced WHRadjBMI, an improved lipid profile, and decreased risk of type 2 diabetes (T2D). The BF%-increasing allele of the PLA2G6 locus was associated with reduced triglyceride levels and that of the TOMM40/APOE locus with reduced risk of cardiovascular disease, but increased risk of T2D. Our meta-analysis for BF% identifies novel loci, and reveals patterns of association that suggest a role of peripheral mechanisms involved in adipocyte and lipid metabolism and insulin sensitivity, complementing the central nervous pathways that are highlighted in GWAS for BMI and obesity risk.

You may contact the first author (during and after the meeting) at