Identification and Validation of Immune-Related Gene Signature for Predicting Lymph Node Metastasis and Prognosis in Lung Adenocarcinoma

摘要

Lung cancer is a serious malignancy, and lung adenocarcinoma (LUAD) is the most common pathological subtype. Immune-related factors play an important role in lymph node metastasis. In this study, we obtained gene expression profile data of LUAD and normal tissues from TCGA database and analyzed their IRGs, and observed that 459 IRGs were differentially expressed. Further analysis of the correlation between differentially expressed IRGs and lymph node metastasis identified 18 lymph node metastasis-associated IRGs. in addition, we analyzed the mutations, function and pathway enrichment of these IRGs, and regulatory networks established through TF genes. We then identified eight IRGs (IKBKB, LTBR, MIF, PPARD, PPIA, PSME3, S100A6, SEMA4B) as the best predictors by LASSO Logistic analysis , then we used the eight IRGs to construct a model to predict lymph node metastasis in LUAD patients (AUC 0.75;95% CI: 0.7064-0.7978), and survival analysis showed that the risk score independently affected patients' survival. We validated the predictive effect of risk scores on lymph node metastasis and survival using the GEO database as a validation cohort and showed good agreement. In addition, the risk score was highly correlated with immune cell infiltration (mast cells activated, macrophages M2, macrophages M0 and B cells na?ve), immune and stromal scores, and immune checkpoint genes (LTBR, CD40LG, EDA2R, and TNFRSF19). We identified key IRGs associated with lymph node metastasis in LUAD and constructed a reliable risk score, which may provide valuable biomarkers for LUAD patients and further reveal the mechanism of its occurrence.