Integrative multi-omics analyses unravel the immunological implication and prognostic significance of CXCL12 in breast cancer

Background: CXCL12 is a key chemokine regulating migration of multiple cell types and has roles in physiological and pathological processes. This study aimed to comprehensively delineate the role of CXCL12 in breast cancer and to discover CXCL12-related biomarkers through integrative multi-omics analyses to build a prognostic model. Methods: Immunohistochemistry (IHC) on a tissue microarray evaluated associations between CXCL12 protein levels and outcomes. Combined single-nucleus RNA-seq (snRNA-seq) and spatial transcriptomics defined the spatial distribution of CXCL12-expressing cells. Weighted gene co-expression network analysis (WGCNA) identified CXCL12-related genes; univariate Cox and LASSO regression selected prognostic genes to construct a CXCL12-related prognostic signature, which was validated in external cohorts. Functional enrichment, mutation landscape, immune checkpoints and immune infiltration analyses compared high/low-risk groups. Paired single-cell RNA-seq (scRNA-seq) and bulk RNA-seq further explored tumor immune microenvironment associations. Drug-gene correlations and immunotherapy response were analyzed to identify potential therapies. Results: CXCL12 expression was reduced in breast tumors versus normal tissue yet higher CXCL12 was associated with prolonged survival. CXCL12 localized predominantly to cancer-associated fibroblasts by snRNA-seq/spatial analyses. WGCNA identified 402 CXCL12-related genes; 11 genes (VAT1L, TMEM92, SDC1, RORB, PCSK9, NRN1, NACAD, JPH3, GJA1, BMP8B, ADAMTS2) formed a prognostic signature that robustly stratified patients into high/low-risk groups across cohorts, with high-risk showing worse outcomes. Combining risk score with tumor mutational burden (TMB) improved prognostic performance. High-risk tumors had higher M2-like macrophage infiltration and distinct pathway enrichment. Several candidate drugs with differential sensitivity were identified.

Zhi-Jie Gao, Zhou Fang, Jing-Ping Yuan, Sheng-Rong Sun, Bei Li