Prognostic Association of Handgrip-Defined Probable or Possible Sarcopenia Status and Polygenic Risk with 10-Year Fracture Incidence among Black, Hispanic, and White Women: A Women’s Health Initiative Study
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Purpose
The Fracture Risk Assessment Tool (FRAX) excludes objective skeletal muscle health and genetic variables. We evaluated the prognostic associations of handgrip-defined probable/possible sarcopenia and genome-wide polygenic scores (GPS) with 10-year fracture risk, and their incremental predictive value beyond FRAX across racial/ethnic groups and GPS strata.
Methods
We analyzed 2,051 postmenopausal women from the Women’s Health Initiative. Race-specific analyses focused on Black, Hispanic, and White participants (n=2,009), excluding American Indian/Alaska Native and Asian/Pacific Islander individuals due to sparse fracture events. Sarcopenia status was operationalized by low handgrip strength alone via EWGSOP2 (<16.0 kg) and AWGS 2025 (<18.0–20.0 kg) criteria. Fine-Gray models estimated subdistribution hazard ratios (sHR), treating death as a competing risk. Predictive performance at 10 years was assessed using time-dependent AUC, Brier scores, and decision curve analysis (DCA).
Results
Handgrip-defined probable or possible sarcopenia prevalence was 4.4% (EWGSOP2) and 6.4% (AWGS 2025). Black women demonstrated lower risk for major osteoporotic fractures (MOF) (adjusted sHR=0.19, 95% CI: 0.08–0.48) and hip fractures (adjusted sHR=0.07, 95% CI: 0.01–0.52) compared to White women. Neither sarcopenia status nor high GPS showed statistically significant independent associations with fractures after FRAX adjustment. Adding sarcopenia status to baseline FRAX (AUC: 0.71 for MOF; 0.69 for hip) yielded near-identical AUCs, Brier scores, and within-sample net benefit.
Conclusion
Handgrip-defined probable/possible sarcopenia and current GPS do not provide independent or incremental predictive value beyond the clinical FRAX framework within this genomic sub-sample of older women.
Mini Abstract
The Fracture Risk Assessment Tool (FRAX) currently excludes skeletal muscle and genomic metrics. In 2,051 postmenopausal women, incorporating handgrip-defined sarcopenia and polygenic scores provided no incremental predictive benefit beyond standard FRAX. Within this study population, separate integration of these markers may not enhance routine fracture risk stratification.
