Share on Facebook
Share on Twitter
Share on LinkedIn
Share by Email
Print
The performance of the World Health Organization’s reference standard for osteoporosis diagnosis and prediction of fracture risk using bone density T-scores varies by race in postmenopausal women, according to study results published in the Journal of Clinical Medicine.
Current osteoporosis definitions and risk assessment tools were developed using data from cohorts composed primarily of white women. Investigators aimed to evaluate how T-scores perform in osteoporosis classification and fracture prediction in women of different races and polygenic risk profiles.
Data for 2417 postmenopausal women (aged 50-79 years) were obtained from the Women’s Health Initiative (WHI) nationwide longitudinal study. Bone mineral density (BMD) was measured using dual-energy x-ray absorptiometry and T-scores were calculated using reference databases. Osteoporosis was defined as a BMD ≥2.5 standard deviations below the mean value for young, healthy women (T-score <-2.5). Polygenic risk scores were generated based on 63 femoral neck BMD-associated single nucleotide polymorphisms.
Over the course of follow-up (19 years), 289 women (11.96%) sustained ≥1 fracture. T-scores were significantly lower in women who experienced fracture (mean, -1.36) compared with those who did not (mean, -0.72; P <.0001), whereas polygenic score did not significantly differ between those who did and did not experience fracture (mean score, 2.27 for both).
The fracture incidence predicted based on T-scores significantly underestimated fracture risk across all polygenic risk score groups. In women with a low polygenic risk score, the predicted incidence of major osteoporotic fracture was 3.83% based on T-scores compared with an observed incidence of 8.8% (predicted to observed ratio [POR], 0.43; 95% CI, 0.28-0.64). Similar results were observed in the medium (POR, 0.71; 95% CI, 0.56-0.90) and high polygenic risk score groups (POR, 0.72; 95% CI, 0.52-0.98).
The T-score-based prediction of major osteoporotic fracture risk significantly underestimated the fracture incidence in black women (POR, 0.52; 95% CI, 0.30-0.83) and white women (POR, 0.63; 95% CI, 0.50-0.78). Moreover, the incidence of any fracture was significantly underestimated in black (POR, 0.19; 95% CI, 0.11-0.31), white (POR, 0.47; 95% CI, 0.37-0.59), and Hispanic women (POR, 0.48; 95% CI, 0.33-0.67).
Using the definition for osteoporosis of a T-score <-2.5 as a major osteoporotic fracture risk predictor, the investigators quantified the false positive and negative rates across polygenic risk score groups and races. The false positive rate was low across all groups (<10% for all). The false negative rate was highest in black women (96.15%) and the low polygenic score group (92.86%) and was lowest in white women (79.44%) and the medium polygenic score group (80.95%).
After adjusting for T-score, polygenic risk score was not associated with major osteoporotic fracture incidence. Race, however, remained a significant predictor of subsequent fracture. Compared with white women, black women (hazard ratio [HR], 0.41; 95% CI, 0.33-0.52), Native American women (HR, 0.59; 95% CI, 0.35-0.99), and Hispanic women (HR, 0.45; 95% CI, 0.35-0.58) had a lower hazard for major osteoporotic fracture. When adjusted for polygenic risk score, the effect of race on fracture risk was attenuated but remained significant.
The researchers noted that the study only comprised women aged 50 to 79 years and that the results may not be generalizable to men or women outside of the included age group. In addition, many genetic factors related to fracture risk remain unclassified and could not be included in the analysis.
“Our findings demonstrated that T-score performed differently in different races and [polygenic risk score] groups, and thus the effect of race and genetic determinants in osteoporotic fracture prediction should be taken into account beyond the T-score classification,” the study authors concluded. “Fully integrating genetic profiling and racial factors into the existing fracture assessment model is very likely to improve the accuracy of osteoporosis diagnosis.”
Reference
Wu Q, Xiao X, Xu Y. Evaluating the performance of the WHO international reference standard for osteoporosis diagnosis in postmenopausal women of varied polygenic score and race. J Clin Med. 2020;9(2):499.
