Integrating bulk, single-cell, and spatial transcriptomics to identify a novel pyroptosis-related gene signature for predicting prognosis and tumor immune landscape in triple-negative breast cancer.

INTRODUCTION

Triple-negative breast cancer (TNBC) features significant heterogeneity and a complex tumor immune microenvironment (TIME). Pyroptosis strongly influences this environment, yet the roles of pyroptosis-related genes (PRGs) remain unclear. Since single transcriptomic methods obscure the full clinical value of PRGs, multi-omics identification of PRGs in TNBC was used to predict the prognosis and immune landscape of TNBC.

METHODS

We integrated TNBC transcriptomic data from the TCGA and GEO databases. We constructed a PRG prognostic signature using the LASSO algorithm. This signature was validated in an independent GEO cohort. We used single-cell RNA sequencing (scRNA-seq) to analyze the expression heterogeneity of signature genes across different cell subpopulations. We also evaluated their association with the TIME. Spatial transcriptomics (ST) was used to map the spatial distribution of these genes. Finally, we performed immunohistochemistry (IHC) on 48 clinical TNBC samples. This step validated the protein expression of six core genes (PINK1, GZMB, PFKFB3, RSPO3, TREM1, and VEGFA) .

RESULTS

The PRG signature demonstrated robust prognostic predictive performance. It effectively distinguished TNBC patients with different prognoses and immune phenotypes. ScRNA-seq analysis revealed a predominant enrichment of signature genes in T cells. Pseudotime trajectory analysis delineated a continuous T-cell state transition landscape characterized by progressive GZMB upregulation. Cell communication analysis indicated extensive interactions between T cells and macrophages. This interaction occurred via the MIF-CD74-CXCR4 axis. ST confirmed significant expression of signature genes in immune cell enriched regions. IHC results showed that high GZMB and RSPO3 expressions correlated with lower recurrence risk and favorable survival outcomes. Conversely, elevated PINK1, PFKFB3, TREM1, and VEGFA predicted higher recurrence and poorer survival.

CONCLUSION

We developed a reliable PRG prognostic signature for TNBC. The signature genes demonstrate significant cellular and spatial heterogeneity within the TIME. They drive interactions between T cells and macrophages through the MIF pathway to remodel the TIME. This signature robustly predicts clinical outcomes for patients. It also offers tremendous translational value by providing promising targets for personalized treatment.