Using a Polygenic Score to Predict the Risk of Developing Primary Osteoporosis

Area of Science:

• Genetics
• Metabolic Bone Diseases
• Osteoporosis Research

Background:

• Osteoporosis is a complex metabolic bone disease.
• Genetic factors significantly influence osteoporosis pathogenesis.
• Previous research identified DNA loci associated with bone mineral density (BMD).

Purpose of the Study:

• To develop predictive models for osteoporosis-related outcomes using polygenic scores (PGS).
• To assess the efficacy of PGS in predicting fracture risk and low BMD in women from the Volga-Ural region.
• To evaluate a combined model for predicting comorbid fracture risk and low BMD in postmenopausal women.

Main Methods:

• Utilized the polygenic score (PGS) approach.
• Integrated DNA loci data from the GEFOS/GENOMOS consortium GWAS meta-analysis.
• Developed and validated predictive models for fracture risk, low BMD, and comorbid conditions.

Main Results:

• A model predicting fracture risk achieved 74% efficacy (AUC = 0.740).
• A model for low BMD formation showed 79% efficacy (AUC = 0.790).
• A combined model for comorbid fracture risk and low BMD demonstrated 85% accuracy (AUC = 0.850) in postmenopausal women.

Conclusions:

• Polygenic scores are effective tools for predicting osteoporosis-related outcomes.
• The developed models offer significant potential for early risk identification and intervention in at-risk populations.
• The study highlights the utility of PGS in personalized medicine for metabolic bone diseases.