| Title: | A protein-based prediction model for fragility fracture risk in individuals with diabetes |
| Journal: | Journal of Molecular Cell Biology |
| Published: | 5 Jan 2026 |
| Pubmed: | https://pubmed.ncbi.nlm.nih.gov/41489455/ |
| DOI: | https://doi.org/10.1093/jmcb/mjaf058 |
Publication 16879
WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
Abstract
Individuals with diabetes are at high risk of fragility fractures. We aimed to develop and validate a protein-based model to predict fragility fractures in individuals with diabetes and to explore whether a protein risk score (ProRS) would improve the risk prediction. A total of 3535 individuals with diabetes from the UK Biobank Pharma Proteomics Project were included in the study. During a median follow-up period of 13.3 years, 5.2% (185) of the individuals with diabetes experienced a fragility fracture. Of 2902 unique proteins, 139 exhibited significant associations with fragility fracture risk. A protein-based model that included 10 proteins was then developed using the machine learning model. Compared with the low ProRS tertile, medium and high tertiles were strongly associated with increased fragility fracture risk. The ProRS achieved a C-index of 0.739 and a 10-year area under the curve (AUC) of 0.733 for the fragility fracture prediction. Adding ProRS to the traditional prediction model (the fracture risk assessment tool [FRAX]) improved the prediction performance with a C-index increase of 0.080 (0.673 [FRAX] versus 0.754 [FRAX + ProRS]) and a 10-year AUC increase of 0.064 (0.693 versus 0.757), thereby promoting early monitoring and prevention in individuals with diabetes.</p>
7 Authors
- Suna Wang
- Li Shen
- Weituo Zhang
- Jingyi Guo
- Wei Chen
- Xiangtian Yu
- Cheng Hu
1 Application
| Application ID | Title |
|---|---|
| 213095 | Associations between body compositions with the development of cardiometabolic diseases and cognitive disorders |
Enabling scientific discoveries that improve human health