Comprehensive integration of multi-omics data identifies novel immunological biomarkers and repurposable drug targets for human aging and frailty

Aging and frailty are complex biological processes characterized by a gradual decline in physiological functions and increased vulnerability to stressors. Although immunological dysregulation is recognized as a key feature of both, studies identifying shared biomarkers and therapies are limited. We conducted a detailed multi-omics analysis combining data from Gerontome (2,347 aging-related genes), a significant frailty GWAS, and transcriptome data from aging immune cells (2,843 DEGs). We employed advanced bioinformatics tools, including WGCNA, machine learning, immune cell deconvolution, and drug repurposing . We identified 28 core genes common across all datasets, including TP53, APOE, IL6, FOXO3, and SIRT1. The turquoise module, comprising 1,872 genes, showed the strongest correlation with frailty (r=0.78, p=2.1e-16) among the 23 modules identified via WGCNA. Using a 12-gene panel, our machine learning model achieved an AUC of 0.91 for frailty prediction. Immune cell analysis revealed that frail patients had significantly higher levels of senescent T cells (p=1e-06) and M1 macrophages (p=7e-04) compared to healthy controls. Drug repurposing identified 47 FDA-approved drugs targeting 18 key genes, with metformin, rapamycin, and tocilizumab showing particular promise . This research provides a comprehensive catalog of biomarkers for frailty and aging, highlighting immune pathways and opportunities for drug repurposing. Our findings emphasize the role of immunosenescence and inflammation in frailty and offer a foundation for targeted therapies.