RESUMO
Background: In recent years, significant molecules have been found in gastric cancer research. However, their precise roles in the disease's development and progression remain unclear. Given gastric cancer's heterogeneity, prognosis prediction is challenging. This study aims to assess patient prognosis and immune therapy efficacy using multiple key molecules. Method: The WGCNA algorithm was employed to identify modules of genes closely related to immunity. A prognostic model was established using the Lasso-Cox method to predict patients' prognosis. Single-sample gene set enrichment analysis (ssGSEA) was conducted to quantify the relative abundance of 16 immune cell types and 13 immune functions. The relationship between risk score and TMB, MSI, immune checkpoints, and DNA repair genes was examined to predict the effectiveness of immune therapy. GO and KEGG analyses were performed to explore potential pathways and mechanisms associated with the genes of interest. Single-cell RNA sequencing was utilized to investigate the expression patterns of key genes in different cell types. Results: Through the WGCNA algorithm and Lasso-Cox algorithm selected KL, SERPINE1, and STK40 as key genes for constructing the prognostic model. The SSGSEA algorithm was employed to evaluate the infiltration of immune cells and immune functions in different patients, and their association with the risk score was investigated. The high-risk group exhibited lower TMB and MSI compared to the low-risk group. MMR and immune checkpoint analysis revealed a significant correlation between the risk score and multiple molecules. Finally, we also believe that STK40 is the most critical senescence-related gene affecting the progression of gastric cancer. In vitro experiments showed that ROS accumulation and cell proliferation ability of gastric cancer cells were impaired when STK40 was knocked down. Conclusion: In summary, we've constructed a prognostic model utilizing key genes for gastric cancer prognosis, while also showcasing its efficacy in predicting patient response to immunotherapy.