Shidong Hu, Ling Wu, Yi-Xiang Wang, Fengsheng Dai
Background Osteosarcoma (OS) is characterized by significant intratumoral heterogeneity and a remodeled immune microenvironment. However, the mechanisms through which metabolic reprogramming drives tumor progression are not fully understood. Lactylation, an epigenetic modification derived from lactate, has recently emerged as a key regulator of tumor metabolism and immune modulation. Methods We employed an integrative multi-omics approach, combining single-cell RNA sequencing, spatial transcriptomics, and machine learning-based prognostic modeling to decipher lactylation-associated networks in OS. Key drivers were identified through co-expression network analysis and SHapley Additive exPlanations (SHAP) interpretability frameworks, followed by experimental functional validation and virtual screening for potential therapeutics. Results We identified a distinct tumor subpopulation exhibiting high lactylation activity, genomic instability, and stem-like properties, which is associated with unfavorable patient prognosis. Network and SHAP analyses pinpointed hepatoma-derived growth factor (HDGF) as a central regulator within this axis. Functional studies demonstrated that HDGF knockdown potently inhibited OS cell proliferation, migration, invasion, and tumor growth in vivo . Spatial transcriptomics confirmed the colocalization of elevated lactylation activity and HDGF expression within tumor regions, linking metabolic reprogramming to the local microenvironment. Furthermore, virtual screening identified several FDA-approved compounds with high predicted binding affinity for HDGF. Conclusion Our study unveils a novel lactylation–HDGF regulatory association that promotes OS progression and modulates the tumor microenvironment. These findings highlight HDGF as a promising prognostic biomarker and therapeutic target, offering new avenues for precision therapy in osteosarcoma.