<p>Osteosarcopenia—the concurrent presence of osteoporosis and sarcopenia—affects approximately 18.5% of older adults globally, yet its shared genetic basis remains poorly understood. We applied genomic structural equation modeling (Genomic SEM) to integrate genome-wide association study (GWAS) summary statistics across five phenotypes spanning the pathophysiological spectrum of osteosarcopenia: appendicular lean mass (ALM), bone mineral density (BMD), handgrip strength (HGS), walking pace, and fracture. Fine-mapping, transcriptome-wide association study (TWAS), pathway enrichment, cell-type enrichment, and spatial transcriptomic mapping were performed to functionally annotate the identified loci. A single-factor model (CFI = 0.976) captured the shared genetic liability, with HGS and ALM loading most strongly. A two-factor sensitivity analysis confirmed partial separability of muscle and bone dimensions, though the single common factor was retained for integrated downstream annotation. We identified 58,696 genome-wide significant single-nucleotide polymorphisms (SNPs) condensed into 1078 independent lead variants, including 29 novel loci. Fine-mapping prioritized 317 high-confidence causal variants, encompassing key genes including BMP6, ACAN, IHH, LRP5, and SOX5. TWAS and MAGMA converged on IGF1R, FOXO3, and IRS1 as dual susceptibility genes. Pathway analysis revealed significant enrichment in endochondral ossification and growth hormone/insulin-like growth factor-1 signaling. Cell-type enrichment localized genetic risk to mesenchymal stem cells and skeletal muscle satellite cells, while spatial mapping identified cartilage primordium as the most enriched developmental context. This study systematically elucidates the shared genetic architecture of osteosarcopenia, highlighting developmental, endocrine, and stem cell-related pathways as core mediators. These findings provide a theoretical foundation for precision geroscience and the development of dual-target therapeutic strategies.</p>