Objective To establish a prognostic model for lung adenocarcinoma (LUAD) based on genes associated with the disulfidptosis (DS) pathway, and to elucidate its potential biological mechanisms.

Methods LUAD-related gene sequencing and clinical information were sourced from public databases.The correlation between results of gene set variation analysis (GSVA) and mRNA expression in The Cancer Genome Atlas (TCGA) dataset was used to screen genes that were significantly active in the disulfur death (DS) pathway.The Least Absolute Shrinkage and Selection Operator (LASSO) analysis and Random Forest (RF) algorithm were employed to screen out DS pathway prognosis-related genes (DPRGs) and multivariate Cox regression analysis was used to construct risk score (RS) model, which was validated using external GEO datasets.The samples were divided into high and low-risk groups based on the median score of RS.A protein-protein interaction (PPI) network corresponding to 7 DPRGs was established, with LDHA identified as the protein with the most interactions, thereby further investigating its function and expression patterns.

Results In this study, 7 DPRGs were screened, including SLC2A1, LDHA, SNAI2 and ACO2, FGF12, ANP32B and ST13.The prognostic model constructed based on these genes exhibited high validation efficiency.Kaplan-Meier survival analysis revealed significant differences in overall survival of patients between high-risk group and low-risk group in four datasets.Differential expression gene enrichment analysis between the high-risk and low-risk groups showed that these genes were enriched in pathways such as the p53 signaling pathway and cell cycle. Results of real-time quantitative polymerase chain reaction (qRT-PCR) and immunohistochemistry indicated that LDHA expression levels were elevated in LUAD tissue compared to normal tissues.

Conclusion The LUAD model established based on DPRGs can effectively predict patients'prognosis, potentially offering insights into the treatment and prognosis of LUAD patients.