Landscape of Immune Cell Infiltration and Inflammatory Response-related Gene Signature in Breast Cancer: an Experimental Validation Study

Abstract

Abstract Purpose The underlying role of inflammatory response-related genes (IRGs) in the tumor microenvironment (TME) of breast cancer (BC) remains unknown. Here, we comprehensively investigated the correlations of IRGs with prognosis and immune cell infiltration in BC patients. Methods IRGs expression profiling were acquired from The Cancer Genome Atlas (TCGA, N = 993) and Gene Expression Omnibus (GEO, N = 3,256) database. Independent prognostic IRGs were identified via Lasso-Cox regression analyses and used to established a predictive model. Thus, time-dependent receiver operating curve (ROC), calibration curve, decision curve, subgroup analysis, drug sensitivity and immune microenvironment analysis were executed to assess the performance of nomogram. Results Three IRGs (CCR7, NFKBIA, and KCNMB2) were identified as prognostic independent predictors of BC and were used to build an inflammation-score. Then, the cluster1 and cluster 2 subtypes were determined by consensus clustering for these three IRGs. Cluster 2 had better overall survival, and higher CD8 T cell infiltration but lower M2 macrophage levels than cluster 1. Subsequently, an IRGs-based nomogram was constructed using the inflammation-score, clinical TNM stage, age, and tumor subtype. Patients in the high-risk group exhibited aggressive clinicopathological features and suffered poor overall survival. Patient in the high-risk groups exhibited lower expression levels of immune checkpoints such as PD-1/PD-L1, lower immune scores, higher infiltration of M2 macrophages, and lower chemotherapeutic sensitivity. Besides, the expression of the three IRGs were validated using IHC experiment. Conclusion The IRGs-based nomogram might open up new horizons in the understanding of TME and provide therapeutic strategies for precision therapy.