Lung adenocarcinoma (LUAD) is one of the world's commonest malignancies with a high fatality rate. Chemokines not only regulate immune response but also participate in tumor development and metastasis and yet the mechanism of chemokines in LUAD remains unclear. In this study, transcriptional expression profiles, mutation data, and copy number variation data were downloaded from The Cancer Genome Atlas (TCGA). Risk gene protein expression was assessed by the Clinical Proteomic Tumor Analysis Consortium (CPTAC) and the Human Protein Atlas (HPA). Gene Expression Omnibus (GEO) data was used to validate the prognostic model. We summarized the genetic mutation variation landscape of chemokines. The risk prognosis model was developed based on differentially expressed chemokines, and patients in the high-risk score (RS) group had lower survival rates. Gene Set Enrichment Analysis (GSEA) revealed that high-RS patients were associated with metabolic transformation pathways, while low-RS patients were associated with immune-related pathways. Compared with the high-RS group, the low-RS group had higher immune/stromal/estimate scores calculated by the ESTIMATE package. The proportion of immune cells obtained using the CIBERSORT package was significantly different between the two groups. Most of the immune checkpoints were highly expressed in low-RS samples. Finally, we discovered that the lncRNA MIR17HG/AC009299.3/miR-21-5p/CCL20 regulatory network might be crucial in the pathogenesis of LUAD. In conclusion, we developed a risk signature and chemokine-related competing endogenous RNA (ceRNA) network.© 2022 The Authors.