Changqi Fu, Zhifang Jia, Yingli Fu, Yangyu Zhang, Yanhua Wu, Xueyuan Cao, Donghui Cao, Jing Jiang
High Resolution Image Download MS PowerPoint Slide Gastric cancer (GC) is a leading cause of cancer-related mortality globally. Histone lactylation, an emerging post-translational modification, holds promise as a therapeutic target and prognostic biomarker, though its expression patterns and clinical relevance in GC are underexplored. Pan-lactylation was quantified by IHC in tumor/adjacent normal tissues and validated via Western blot in GC/normal cell lines. Systematic profiling of lactylome and proteome of tumor-normal pairs ( n = 3) identified 127 lactylation-regulated genes. Four prognostic models (stepCOX, LASSO, RSF, GBM) were built using these lactylation-regulated genes from GPL6947 data sets and validated independently in GPL570 data sets. IHC results confirmed elevated pan-lactylation in GC tissues and high pan-lactylation level correlated with larger tumor diameter ( P = 0.041) and vascular invasion ( P = 0.032). In addition, pan-lactylation level was an independent prognostic predictor (HR = 2.06, 95%CI: 1.51–2.80). Four prognostic models were constructed with different advantages, while the RSF model achieved superior training performance (5-year C-index = 0.837), while COX regression demonstrated optimal validation robustness (5-year C-index = 0.624). Pan-lactylation is a novel prognostic biomarker for GC. The lactylation-driven prognostic signature facilitates precise risk stratification in GC patients, providing a framework for personalized therapeutic intervention.