Prediction of prognosis and immunotherapy response in lung adenocarcinoma based on CD79A, DKK1 and VEGFC

BackgroundTumor immune microenvironment (TIME) is crucial for tumor initiation, progression, and metastasis; however, its relationship with lung adenocarcinoma (LUAD) unknown. Traditional predictive models screen biomarkers that are too general infrequently associated genes.MethodsRNA sequencing data of LUAD patients immune-related gene sets were retrieved from public databases. Using the common genes shared by The Cancer Genome Atlas (TCGA) Immunology Database Analysis Portal (ImmPort), differential expression analysis, survival Lasso regression univariate multivariate Cox analyses performed to generate a novel risk score model. cohort in International Consortium (ICGC), GSE68465 Gene Expression Omnibus (GEO) an immunohistochemical assay used validate key constructed score. LUAD-related prognosis, clinical indicators, infiltrate characteristics, response immunotherapy, chemotherapeutic agents different groups evaluated CIBERSORT, ImmuCellAI, pRRophetic other tools.ResultsThe model was using CD79a molecule (CD79A), Dickkopf WNT signaling pathway inhibitor 1 (DKK1), vascular endothelial growth factor C (VEGFC). High identified as negative predictor overall (OS) subgroup stage, TNM classification, therapy outcome, ESTIMATE scores (P < 0.05). Low positively plasma cells, memory B CD8 T regulatory cells γδT In low-risk group, programmed cell death receptor (PD1), cytotoxic T-lymphocyte protein 4 (CTLA4), lymphocyte activating 3 (LAG3) indoleamine 2,3-dioxygenase (IDO) more robustly expressed treatment responses checkpoint blockade (ICB) chemotherapy likewise superior group had significantly higher levels CD79A, DKK1, VEGFC than adjacent normal 0.01).ConclusionsCD79A, DKK1 important related LUAD, could reliably predict composition TIME immunotherapy patients. excellent performance shows strong broad application potential.