Dataset Information

Association between a literature-based genetic risk score and bone mineral density of African American women in Women Health Initiative Study.

ABSTRACT: Genetic risk of low BMD in African American women remains unclear. Based on SNPs discovered from a predominantly Caucasian sample, genetic profile was summarized and was found to be significantly associated with BMD variation in African American women.

Introduction

Osteoporosis is largely under-recognized and undertreated in African-American women, the post-fracture morbidity and mortality rates in this racial group is rather high. Since BMD was proved to be highly heritable, based on a comprehensive genome-wide meta-analysis that reported 63 BMD-related single nucleotide polymorphisms (SNPs), we aim to unravel the overall genetic risk for decreased BMD and osteoporosis in African-American women.

Methods

Genotype data of 842 African American women in a Women's Health Initiative cohort were analyzed. Comprehensive genotype imputation was conducted at the Sanger Imputation Server. Multi-locus genetic risk scores (GRSs) based on 62 BMD-related single-nucleotide polymorphisms (SNPs) were calculated. The association between GRS and BMD was assessed by regression analysis. Longitudinal data was further analyzed using a generalized estimating equation, which helps achieve more efficient and unbiased regression parameters by accounting for the within-subject correlation of responses on dependent variables.

Results

After adjusting for age, body weight, hormone use, and previous fracture, for every unit increase of GRS.FN and GRS.LS, BMD at hip and lumbar spine decreased 0.124 g/cm² and 0.086 g/cm², respectively. Collectively, the model accounted for 34.95% of the femoral neck BMD variation and 25.79% of lumbar spine BMD variation. Notably, GRS.FN and GRS.LS accounted for 2.03% and 2.39% of the total explained variance, respectively. The proportion of BMD variation can be explained by GRSs increasing as participants aged.

Conclusions

Genetic risk score was significantly associated with lower BMD in the current study, suggesting that SNPs discovered from prior meta-analysis based on primarily Caucasian population can also explain a considerable proportion of BMD variation in African Americans.

SUBMITTER: Xiao X

PROVIDER: S-EPMC7176548 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:UnlabelledThere is a paucity of normative bone mineral density (BMD) data in healthy African women. Baseline total hip and lumbar spine BMD was measured in premenopausal women. BMD distribution was comparable to that of a reference population and was impacted by several factors including contraception and duration of lactation.IntroductionNormative data on bone mineral density (BMD) and the cumulative impact of lactation, contraceptive use, and other factors on BMD in healthy African women have not been well studied.ObjectivesThe objective of this study was to determine the factors associated with BMD in healthy premenopausal women in Uganda and Zimbabwe.MethodsBaseline total hip (TH) and lumbar spine (LS) BMD was measured by dual x-ray absorptiometry in 518 healthy, premenopausal black women enrolling in VOICE, an HIV-1 chemoprevention trial, at sites in Uganda and Zimbabwe. Contraceptive and lactation histories, physical activity assessment, calcium intake, and serum vitamin D levels were assessed. Independent factors associated with BMD were identified using an analysis of covariance model.ResultsThe study enrolled 331 women from Zimbabwe and 187 women from Uganda. Median age was 29 years (IQR 25, 32) and median body mass index (BMI) was 24.8 kg/m(2) (IQR 22.2, 28.6). In univariate analyses, lower TH BMD values were associated with residence in Uganda (p < 0.001), lower BMI (p < 0.001), and any use of and duration of depot-medroxyprogresterone acetate. Use of oral contraceptives, progestin-only implants, and higher physical activity levels were protective against reduced BMD. Similarly, lower LS BMD values were associated with these same factors but also higher parity and history of breastfeeding. In a multivariable analysis, lower TH and LS BMD values were associated with enrollment in Uganda, lower BMI, and lower physical activity level; contraceptive use was associated with lower spine BMD, and breastfeeding contributed to lower total hip BMD.ConclusionsAmong healthy premenopausal women, TH and LS BMD was higher in Zimbabwe than Uganda. Additional factors independently associated with BMD included BMI, physical activity level, contraceptive use, and lactation.

Project description:ObjectiveThe objectives of this study were to describe the trajectories of bone mineral density (BMD) and trabecular bone score (TBS) changes throughout pre-menopause (reproductive phase and menopausal transition) and post-menopause (early and late menopause) in women with HIV (WWH) undergoing different antiretroviral therapies (ARTs) and explore the risk factors associated with those changes.MethodsThis was an observational longitudinal retrospective study in WWH with a minimum of two DEXA evaluations comprising BMD and TBS measurements, both in the pre-menopausal and post-menopausal periods. Menopause was determined according to the STRAW+10 criteria, comprising four periods: the reproductive period, menopausal transition, and early- and late-menopausal periods. Mixed-effects models were fitted to estimate the trajectories of the two outcomes (BMD and TBS) over time. Annualized lumbar BMD and TBS absolute and percentage changes were calculated in each STRAW+10 time window. A backward elimination procedure was applied to obtain the final model, including the predictors that affected the trajectories of BMD or TBS over time.ResultsA total of 202 WWH, all Caucasian, were included. In detail, 1954 BMD and 195 TBS data were analyzed. The median number of DEXA evaluations per woman was 10 (IQR: 7, 12). The median observation periods per patient were 12.0 years (IQR = 8.9-14.4) for BMD and 6.0 years (IQR: 4.3, 7.9) for TBS. The prevalence of osteopenia (63% vs. 76%; p < 0.001) and osteoporosis (16% vs. 36%; p < 0.001) increased significantly between the pre-menopausal and post-menopausal periods. Both BMD (1.03 (±0.14) vs. 0.92 (±0.12) g/cm2; p < 0.001) and TBS (1.41 (IQR: 1.35, 1.45) vs. 1.32 (IQR: 1.28, 1.39); p < 0.001) decreased significantly between the two periods. The trend in BMD decreased across the four STRAW+10 periods, with a slight attenuation only in the late-menopausal period when compared with the other intervals. The TBS slope did not significantly change throughout menopause. The delta mean values of TBS in WWH were lower between the menopausal transition and reproductive period compared with the difference between menopause and menopausal transition.ConclusionsBoth BMD and TBS significantly decreased over time. The slope of the change in BMD and TBS significantly decreased in the menopausal transition, suggesting that this period should be considered by clinicians as a key time during which to assess bone health and modifiable risk factors in WWH.

Project description:Genetic risk of low bone mineral density in women remains unclear. This study found that a large percentage of Caucasian women have a high genetic risk of osteoporosis, and genetic risk scores are significantly associated with BMD variation in a bone healthy sample of Caucasian women. INTRODUCTION:We aimed to examine the distribution of risk alleles in an independent sample and to determine if such genetic components are associated with bone mineral density (BMD) variation in the sample. METHODS:Existing genotype data of 1205 women in the cross-sectional Genomic Wide Scans for Female Osteoporosis Gene Study (GWSFO) were analyzed. Multi-loci genetic risk scores (GRSs) based on 62 BMD-associated single nucleotide polymorphisms (SNPs) were calculated. Regression analysis was employed to assess the association between GRSs and BMD. To examine the effect of SNPs clustered within key pathways associated with the development of osteoporosis, subtype weighted GRS specific to WNT signaling (6 SNPs), RANK-RANKL-OPG (3 SNPs), and mesenchymal stem differentiation (3 SNPs) were generated for analysis. RESULTS:The unweighted GRS ranged from 48 to 80. One third of the women carried 66% risk alleles. After adjusting for age, height, and body weight, each unit increase of weighted GRS was associated with a decrease in BMD of 0.097 at femur (p?<?0.0001) and 0.110 (p?<?0.0001) at lumbar spine. The weighted GRS accounted for only 3.17-4.52% of BMD variance. The WNT signaling pathway GRS (6 SNPs) and the RANK-RANKL-OPG signaling pathway GRS (3 SNPs) both were significantly associated with decreased BMD at femur neck (p?=?0.0004 and p?=?0.0063, respectively) and lumbar spine (p?<?0.0001 and p?=?0.0001, respectively), while the mesenchymal stem cell differentiation pathway (3 SNPs) GRSs were associated only with the lumbar spine BMD (p?=?0.045). CONCLUSIONS:A substantially large percentage of healthy Caucasian women have a high genetic risk of osteoporosis. Weighted GRS was significantly associated with decreased BMD. The contribution of subtype GRS to the BMD variation differs by specific biological pathway and skeletal regions.