An 8-gene diabetes-related signature predicts survival and immunotherapy response in breast cancer.

OBJECTIVETo develop a signature of Diabetes-Related Genes (DRGs) using data from the Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases to predict the prognosis of Breast Cancer (BRCA) patients and identify potential therapeutic targets.

METHODSA list of DRGs was sourced from the GeneCard database. Analyses of differential expression and consensus clustering were conducted to identify DRGs, and Cox regression and Least Absolute Shrinkage and Selection Operator (Lasso) regression were used to construct prognostic risk signatures based on DRGs. High- and low-risk groups were classified using median risk scores, and nomogram plots were created to visualize prognostic signatures. The relationship between immune infiltration, chemotherapeutic agents, and risk scores was evaluated, along with the expression of DRGs in individual immune cells.

RESULTSA total of 1231 RNA-seq samples (cancer: 1118, paraneoplastic: 113), 1046 clinical data, and 965 DRGs were obtained from the databases. Differential expression and consensus clustering analyses identified 224 differentially expressed DRGs. Cox regression and Lasso regression analyses led to the establishment of a prognostic signature based on 8 DRGs. Kaplan-Meier curves demonstrated that patients in the low-risk group had a significantly better prognosis compared to those in the high-risk group (p < 0.001). The calibrated curves of the nomogram at 1-, 2-, 3-, and 5-years all aligned with the diagonal, indicating that our nomogram has high predictive performance. Immune infiltration analysis showed significant correlations between our constructed signature and the relative abundance of immune cells (p < 0.05). Single-cell data analysis revealed that DRGs were predominantly expressed in CD4+ conventional T-cells and regulatory T-cells.

CONCLUSIONThis study has successfully developed a prognostic signature for DRGs, which can be utilized as an efficient tool for risk stratification and prognostic prediction in BRCA patients. Additionally, this research offers a novel basis for investigating potential immunotherapy targets for BRCA patients.