Identification and validation of a seven-gene metastasis-associated prognostic model in breast cancer

Metastasis is the primary cause of mortality in patients with breast cancer. This study aimed to develop a prognostic signature based on metastasis- and cancer-associated differentially expressed genes (M-CA-DEGs) and to identify potential novel therapeutic genes.

Data were acquired from the TCGA-BRCA, AURORA US Network, SCAN-B, and GEO databases. M-CA-DEGs were identified, and a prognostic risk model was constructed via univariate Cox and LASSO regression analyses. The prognostic value was verified using calibration curves and decision curve analysis (DCA). Independent prognostic factors were subsequently validated. Functional enrichment was assessed through GSEA and ssGSEA. Immune infiltration and mutation profiles were compared between risk groups. A TF-mRNA regulatory network was constructed. Single-cell analysis was performed to characterize gene expression patterns. The mRNA levels of prognostic genes were examined in MCF-10A mammary epithelial cells and breast cancer cell lines.

We developed and validated a novel seven-gene, metastasis-associated prognostic signature (IGJ, CXCL14, PTGER3, RTN1, EGOT, TLR10, PANX2) for breast cancer. The risk score emerged as a powerful independent prognostic factor. The low-risk group exhibited superior survival, an immunologically "hot" phenotype with enriched activated CD8⁺ T cells, and higher immune activity, whereas the high-risk group showed T-cell exclusion and enrichment in kinase signaling and metabolism. Somatic mutation landscapes differed significantly between groups. Crucially, we identified two previously under-characterized genes (RTN1 and TLR10) as potential novel drivers of tumor progression. Single-cell transcriptomics unveiled their cell type-specific expression patterns, and in vitro assays confirmed differential expression in cancer cell lines.

This study establishes a robust, biologically grounded, metastasis-related seven-gene prognostic model for breast cancer. Beyond prediction, our work identifies two novel therapeutic targets and reveals distinct immune and metabolic phenotypes across risk groups, thereby providing novel mechanistic insights into tumor heterogeneity and actionable targets for future therapies.