Associations of osteoprotegerin (OPG) TNFRSF11B gene polymorphisms with risk of fractures in older adult populations: meta-analysis of genetic and genome-wide association studies

The meta-analysis of osteoprotegerin (OPG) (TNFRSF11B) polymorphisms from genetic association studies and genome-wide association studies was performed in order to test the hypothesis of association between OPG polymorphisms and fracture. The findings showed a significant 13% to 37% protective effect of OPG on fractures in postmenopausal women (PSM) (rs2073618), overall, ≥ 60y and Western subjects (rs3134069 and rs3134070). Fractures in older people usually result from compromised bone integrity. The multifactorial aetiology of fractures includes both genetic and environmental factors. Inconsistency of reported associations of osteoprotegerin (OPG) (TNFRSF11B) polymorphisms with fracture in the older adult population warranted a meta-analysis to determine more precise estimates. We searched for all available literature on OPG (TNFRSF11B) and fracture. Four polymorphisms were examined, one exonic (rs2073618) and three intronic (rs3134069, rs3134070 and rs3102735). The first two intron polymorphisms were combined (OPGI: osteoprotegerin intron) on account of complete linkage disequilibrium. Risks were estimated with odds ratios (ORs) and 95% confidence intervals (CIs) using the allele-genotype model that included variant (var), wild-type (wt) and heterozygote (het). Multiple comparisons were Bonferroni-corrected. We used meta-regression to examine sources of heterogeneity. Zero heterogeneity (homogeneity: I2 = 0%) and high significance (Pa < 0.00001) were the criteria for strength of evidence. Significant outcomes were subjected to sensitivity analysis and publication bias assessment. From 13 articles (11 genetic association and two genome-wide), this meta-analysis generated five significant pooled ORs, all indicating reduced risks (ORs 0.44–0.87). Of the five, four highly significant comparisons (Pa ≤ 0.00001–0.002) survived the Bonferroni correction, one in rs2073618 het model of the postmenopausal women (OR 0.87, 95% CI 0.81–0.92, I2 = 0%) and the other three in OPGI wt model of the overall analysis, ≥ 60 y and Western subjects (ORs 0.63–0.71, 95% CI 0.47–0.86, I2 = 97–99%). These findings were consistent, had high significance and high statistical power and were robust and without evidence of publication bias. Four covariates (year of publication, study quality, fracture type/site and sample size) were the sources of heterogeneity in the OPGI overall outcomes (Pa = 0.0001–0.03). Evidence showed that the OPG (TNFRSF11B) polymorphisms reduced the risk for fracture in older adults, particularly protective among postmenopausal women, ≥ 60 y and Western subjects.