ECTS2016 Poster Presentations Genetics and Epigenetics (25 abstracts)
1Division of Bone Disease, University Hospital and Faculty of Medicine, Geneva, Switzerland; 2Department of Internal Medicine and Medical Disciplines, Sapienza, University, Rome, Italy; 3Department of Geriatrics, Department of Internal Medicine, Rehabilitation and Geriatrics, University Hospital, Geneva, Switzerland.
Background: We previously reported that serum periostin levels are determined by additive genetic effects. Whether serum levels and/or SNPs in the periostin gene (Postn) contribute to bone microstructure however remains unknown.
Aim: To investigate the association between periostin levels, Postn SNPs, bone-mass and bone microarchitecture in a cohort of postmenopausal women.
Methods: A total of 648 postmenopausal women from the Geneva Retirees Cohort were analyzed for six periostin SNPS (rs9547952, rs9603226, rs7322993, rs9576308, rs7338244, rs9547970). Periostin serum levels were determined by ELISA. Areal bone mineral density (aBMD) was measured by dual-energy X-ray absorptiometry at radial, lumbar and femoral sites (Discovery A, Hologic® inc, Waltham, MA, USA). Distal radius, tibia volumetric BMD and bone microstructure were measured by high-resolution peripheral quantitative computed tomography (XtremCT, Scanco Co, Bruttisellen, CH). Regression analyses were carried out to determine the association between periostin serum level and SNPs with bone traits.
Results: Mean age of the cohort was 64±1.4 years. Periostin levels were positively associated with 1/3 radius aBMD (beta=0.10, P=0.009) and negatively with radius cortical porosity (beta=−0.09, P=0.02). Periostin SNPs rs7322993, rs9576308, rs7338244, rs9547970 were associated with aBMD at lumbar, femoral and ultradistal radius sites. Those SNPs were also associated with radius and tibia microarchitecture (trabecular number and cortical porosity). Once adjusted for age, height, weight and years since menopause only rs7322993, rs9576308, rs9547970 remained significantly associated with both lumbar aBMD (P=0.02−0.04) and radius trabecular number (P=0.008−0.04). There was no association between periostin serum level and periostin SNPs.
Conclusion: Periostin levels and SNPs are significantly associated with aBMD and with radius bone microstructure. If confirmed in independent cohort, these results would contribute to understand the genetic determinants of bone fragility.