Abstract
BACKGROUND: With the aging population, osteoporosis and sarcopenia have emerged as two prevalent age-related degenerative diseases that pose significant public health challenges. Although clinical studies increasingly report the co-occurrence of these conditions, the underlying molecular mechanisms linking them remain poorly understood. METHODS: We adopted a systems biology approach to identify key biomarkers and explore their molecular roles in the interplay between osteoporosis and sarcopenia. Transcriptomic datasets were systematically analyzed to identify candidate genes. The expression patterns of core biomarkers were validated using independent datasets and in vitro cellular models of both diseases. Furthermore, a machine learning-based diagnostic framework was constructed using the identified biomarkers, and model interpretability was enhanced using Shapley Additive Explanations (SHAP). RESULTS: We identified DDIT4, FOXO1, and STAT3 as three central biomarkers that play pivotal roles in the pathogenesis of both osteoporosis and sarcopenia. Their expression patterns were consistently validated across multiple independent transcriptomic datasets, and their differential expression was further confirmed using quantitative reverse transcription polymerase chain reaction (RT-PCR) in disease-relevant cellular models. A diagnostic model constructed based on biomarker genes achieved high classification accuracy across diverse validation cohorts. Moreover, SHAP analysis quantified the individual contribution of each biomarker to the model's predictive performance. CONCLUSION: This study uncovers key molecular links between osteoporosis and sarcopenia, highlighting DDIT4, FOXO1, and STAT3 as shared biomarkers. The findings provide novel insights into their common pathophysiology and lay the groundwork for developing more accurate diagnostic tools and targeted therapeutic strategies.