Background <p>This study aimed to investigate the relationship between disulfidptosis-related genes (DRGs) and lung squamous cell carcinoma (LUSC).</p> Methods <p>Using the TCGA-LUSC dataset, WGCNA identified LUSC-related genes, and hub genes were obtained after intersecting these with DRGs. Consensus clustering analysis divided LUSC patients into different subtypes and differentially expressed genes were then identified. LASSO-Cox regression was applied to establish a prognostic model, followed by analyses of immune cell infiltration and drug sensitivity. Finally, in vitro experiments confirmed the results of bioinformatics analysis.</p> Results <p>After intersecting 9,157 LUSC-related genes with 24 DRGs, 12 hub genes were identified. Two subtypes (Cluster 1 and Cluster 2) for LUSC were recognized, and 219 differentially expressed genes were screened. A prognostic model was constructed based on 7 genes (<i>CD300LG</i>, <i>PCDHGA12</i>, <i>HPD</i>, <i>KANK4</i>, <i>PPBP</i>, <i>ZCCHC5</i>, and <i>CSMD3</i>), with good predictive ability. Neutrophils, plasma cells, T cells CD4 memory resting, and T cells follicular helper displayed significant differences between different risk groups and subtypes. <i>ZCCHC5</i> and <i>PCDHGA12</i> were negatively correlated with all 20 screened drugs. In vitro experiments confirmed the occurrence of disulfidptosis and the levels of prognostic genes in LUSC.</p> Conclusion <p>The study identified two LUSC subtypes and established a 7-gene prognostic model based on DRGs, which can be used as a new prognostic tool for LUSC.</p> Graphical Abstract <p></p>