Screening and verification of potential gene targets in the metastasis of breast cancer by bioinformatics analysis and immunohistochemistry

Abstract

Abstract BACKGROUND Breast cancer (BC), one of the most common cancers among females worldwide, has a high mortality rate, especially for patients with BC metastasis. However, BC metastasis pathogenesis has not yet been completely elucidated. METHODS We integrated multiple databases to clarify and verify potential key candidate biomarkers and signal transduction pathways in BC metastasis. Differentially expressed genes (DEGs) of breast cancer metastasis and non-metastasis sites were screened using microarray data from three publicly available datasets (GSE14776, GSE103357 and GSE32489). GEO2R, DAVID 6.8, STRING, Cytoscape, GEPIA 2.0 and R 4.0.5 were utilized in this study. The correlations between hub genes and clinical value were validated through the GEPIA online tool, UALCAN and immunohistochemistry (IHC) stain. RESULTS A total of 295 DEGs were identified, which were significantly enriched in terms related to focal adhesion, and cell division. KEGG pathway analysis showed that significant pathways included the MAPK signaling pathway, the Rap1 signaling pathway, cell adhesion molecules. Eight hub genes (TYMS, SKA1, ADCY7, POLR3H, CDCA8, PRC1(ASE1), KIF14, and MX1 and two biomarkers (PRC1 and POLR3H) with significantly prognostic values were screened by multi-omics data analysis and verified by IHC stain. CONCLUSIONS In this study, we identified a robust set of potential candidate biomarkers in BC metastasis, which would provide potential value for its early diagnosis and prognosis, and would promote molecular targeting therapy for BC metastasis.