Integrated Transcriptomic and Single-Cell Analyses Identify a T Cell-Mediated Tumor Killing Gene Signature for Prognostic Stratification and Immune Characterization in Breast Cancer

T cell-mediated tumor killing (TTK) is a central mechanism for antitumor immunity. However, the prognostic value of genes regulating TTK sensitivity (GSTTKs) in breast cancer (BC), along with their impact on the immune microenvironment and T-cell dynamics, remains to be systematically elucidated.

Using gene expression data from public databases, we identified prognosis-related key GSTTKs. A risk model was developed through LASSO regression and validated in an independent external dataset. We then evaluated the independent prognostic value of the risk score and clinical features, and constructed a nomogram for survival prediction. Immune characteristics and immunotherapy response were assessed using CIBERSORT and TIDE algorithms. Furthermore, scRNA-seq data were integrated to elucidate T-cell communication networks and differentiation trajectories. Immunohistochemistry was performed to validate the protein expression patterns of the key GSTTKs.

We constructed a stable prognostic risk model comprising 5 GSTTKs (KCTD5, COL17A1, LRTM2, DOK7, and CXCR3). Patients were classified into high- and low-risk groups, with significantly different survival outcomes. Risk score, age, stage, and M were independent prognostic factors, and integrating them into a nomogram effectively predicted 1-, 3-, and 5-year survival. The two risk groups differed significantly in immune infiltration and immune checkpoint expression, with high-risk group showing elevated TIDE scores. scRNA-seq analysis further revealed a higher T-cell proportion in the low-risk group, interacting with macrophages via the CCL5-CCR1 pathway. Additionally, CXCR3 was dynamically expressed along T-cell differentiation trajectories.

This study identified five GSTTK-related signatures in BC, revealing their association with immune heterogeneity and T-cell-specific regulation. These preliminary findings may inform prognostic evaluation, but warrant further validation before clinical translation, given the retrospective design, reliance on public datasets, lack of functional validation, and limited immunohistochemical samples.