ECTS2013 Poster Presentations Genetics (17 abstracts)
1Pharmacology and Clinical Neurosciences, Umeå University, Umeå, Sweden; 2Public Health and Clinical Medicine, Umeå University, Umeå, Sweden; 3Fred Hutchinson Cancer Research Center, Seattle, Washington, USA; 4Center for Bone and Arthritis Research, Institute of Medicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden; 5Surgical and Perioperative Sciences, Umeå University, Umeå, Sweden; 6Department of Radiation Sciences, Oncology, Umeå University, Umeå, Sweden; 7Radiation Epidemiology Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute, NIH, Bethesda, Maryland, USA; 8Division of Cancer Epidemiology and Genetics, National Cancer Institute, Rockville, Maryland, USA; 9Translational Genomics Reserach Institute, Phoenix, Arizona, USA; 10Department of Pharmacogenomics, College of Medicine, Ohio State University, Columbus, Ohio, USA; 11Genomics Core Facility, University of Gothenburg, Gothenburg, Sweden; 12Department of Internal Medicine, Erasmus Medical Center, Rotterdam, The Netherlands; 13Division of Endocrinology, Diabetes and Metabolism, Ohio State University, Columbus, Ohio, USA.
Hip and forearm fractures are the two clinically most important non-vertebral fractures. Twin studies have demonstrated a high heritability of these fractures and the heritable component of fracture risk is largely independent of BMD.
To identify common genetic variants associated with hip and forearm fractures, we performed a genome-wide association study (GWAS ~ 2.5 million SNPs) meta-analysis of two large fracture data sets within the well-characterized UFO cohort (UFO-hip; 1014 hip fractures and 862 controls, and UFO-forearm; 1060 forearm fractures and 1055 controls). All fractures were confirmed through radiographic reports. Replication was performed in the Womens Health Initiative (WHI) cohort (1845 hip fractures verified by medical records and 2120 controls).
We identified one SNP within the SOAT1/AXDND1 locus (1q25.2) that was associated with fracture risk at genome wide significance (OR per allele=1.33; P=3.1×10−8) in the UFO discovery meta-analysis. This SNP was associated with fracture risk both in the WHI replication cohort (OR 1.16, P=2.1×10−3) and in the combined analyses comprising 7956 subjects (3919 cases and 4037 controls; OR=1.24, P=5.6×10−10). However, it was not associated with BMD or biochemical bone markers, suggesting that its association with fractures is BMD-independent. A genetic score (GS), including information from 63 SNPs earlier shown to be reproducibly associated with BMD, was significantly associated with both hip (P=7.9×10−4) and forearm (P=8.6×10−5) fractures. Models including both the SNP in the SOAT1/AXDND1 locus and the GS demonstrated that the impact of the SNP in the SOAT1/AXDND1 locus on fracture risk was independent of the BMD-associated GS.
In summary, both a BMD-associated GS and a non-BMD associated genetic variant in the SOAT1/AXDND1 locus are associated with hip and forearm fractures.