Objective To identify the immune cell-related biomarkers in lung adenocarcinoma using weighted gene co-expression network. Methods In this study, based on TCGA database, gene co-expression network was constructed in TCGA-LUAD by WGCNA package, and different gene modules were formed by clustering. At the same time, ESTIMATE analysis was performed on lung adenocarcinoma tumor samples in the TCGA-LUAD dataset. Gene enrichment pathways in the most significant related modules were evaluated by GO and KEGG assays. Candidate hub genes in the selected key modules were used to construct protein-protein interaction (PPI) network for intersection to obtain hub genes. The prognostic properties of these hub genes and patient immune cell infiltration were verified by Kaplan-Meier curve and TIMER algorithm. Multivariate Cox regression analysis was performed on the acquired hub gene and a prognostic risk model was constructed. Results In the co-expression network, we observed that the brown module was closely related to ImmuneScore, StromalScore and ESTIMATE Score. Five immune-related hub genes CD53, PLEK, SPI1, IL10RA and C3AR1 were obtained. The enrichment analysis of brown module found that module genes were mainly enriched in GO items such as innate immune response regulation and KEGG pathways such as NF-kappa B signaling pathway. In addition, the results of this study also found that the expression levels of 5 hub genes were significantly positively correlated with the infiltration abundance of immune cells. IPS and TIDE validated the immune relevance of the model. At the same time, we found that the RiskScore we established has great potential in predicting immunotherapy. Conclusion In summary, the five key genes related to immune cells obtained may provide new and effective potential targets for the immunotherapy of lung adenocarcinoma, which is also beneficial to provide personalized diagnosis and treatment strategies for patients with lung adenocarcinoma in the later stage.