Identification of key tumor stroma-associated transcriptional signatures correlated with survival prognosis and tumor progression in breast cancer
Md Nazim Uddin,Xiaosheng Wang
DOI: https://doi.org/10.1007/s12282-022-01332-6
Breast Cancer
Abstract:Background: The aberrant expression of stromal gene signatures in breast cancer has been widely studied. However, the association of stromal gene signatures with tumor immunity, progression, and clinical outcomes remains lacking. Methods: Based on eight breast tumor stroma (BTS) transcriptomics datasets, we identified differentially expressed genes (DEGs) between BTS and normal breast stroma. Based on the DEGs, we identified dysregulated pathways and prognostic hub genes, hub oncogenes, hub protein kinases, and other key marker genes associated with breast cancer. Moreover, we compared the enrichment levels of stromal and immune signatures between breast cancer patients with bad and good clinical outcomes. We also investigated the association between tumor stroma-related genes and breast cancer progression. Results: The DEGs included 782 upregulated and 276 downregulated genes in BTS versus normal breast stroma. The pathways significantly associated with the DEGs included cytokine-cytokine receptor interaction, chemokine signaling, T cell receptor signaling, cell adhesion molecules, focal adhesion, and extracellular matrix-receptor interaction. Protein-protein interaction network analysis identified the stromal hub genes with prognostic value in breast cancer, including two oncogenes (COL1A1 and IL21R), two protein kinases encoding genes (PRKACA and CSK), and a growth factor encoding gene (PLAU). Moreover, we observed that the patients with bad clinical outcomes were less enriched in stromal and antitumor immune signatures (CD8 + T cells and tumor-infiltrating lymphocytes) but more enriched in tumor cells and immunosuppressive signatures (MDSCs and CD4 + regulatory T cells) compared with the patients with good clinical outcomes. The ratios of CD8 + /CD4 + regulatory T cells were lower in the patients with bad clinical outcomes. Furthermore, we identified the tumor stroma-related genes, including MCM4, SPECC1, IMPA2, and AGO2, which were gradually upregulated through grade I, II, and III breast cancers. In contrast, COL14A1, ESR1, SLIT2, IGF1, CH25H, PRR5L, ABCA6, CEP126, IGDCC4, LHFP, MFAP3, PCSK5, RAB37, RBMS3, SETBP1, and TSPAN11 were gradually downregulated through grade I, II, and III breast cancers. It suggests that the expression of these stromal genes has an association with the progression of breast cancers. These progression-associated genes also displayed an expression association with recurrence-free survival in breast cancer patients. Conclusions: This study identified tumor stroma-associated biomarkers correlated with deregulated pathways, tumor immunity, tumor progression, and clinical outcomes in breast cancer. Our findings provide new insights into the pathogenesis of breast cancer.