Exploring the key genes associated with breast cancer radiotherapy sensitivity based on the stromal-immune score and analysis of the WGCNA and ceRNA network

Background Breast cancer is a highly malignant disease worldwide, but there are currently no sufficient molecular biomarkers to predict patient prognosis and guide radiotherapy. The tumor microenvironment (TME) is an important factor affecting tumor biological function, and changes in its composition are equally relevant to tumor progression and prognosis during radiotherapy. Methods Here, we performed bioinformatic analyses using data obtained from The Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus (GEO) databases to screen for molecular biomarkers related to the TME that may influence radiotherapy sensitivity. By combining immune scores and stromal scores and performing weighted coexpression network analysis (WGCNA), we identified key modules and hub genes to construct competing endogenous RNA (ceRNA) networks. Then, key pathways and genes were identified using Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses. Furthermore, we explored the effect of THBS2 on the malignant phenotype of breast cancer through cellular experiments. Results Genes in the PI3K-AKT pathway in the blue module were significantly enriched. Among the hub genes in the blue module, COL1A1, COL1A2, COL6A3, THBS2 and PDGFRB were negatively associated with RT sensitivity. Cellular experiments confirmed that knockdown of THBS2 inhibited the proliferation and invasion of breast cancer cells. Conclusion These findings may provide new insights into the mechanisms of radiotherapy sensitivity in breast cancer patients, offering hope for the discovery of new therapeutic targets.