Abstract
BACKGROUND: Breast cancer (BC) is a significant malignancy characterized by a high global incidence and a propensity for recurrence. Absorption, distribution, metabolism, and excretion (ADME) genes comprise a collection of genes that participate in the drug ADME. Understanding the role and prognostic value of ARGs (ADME related genes) in BC advancement is critical for personalized therapy. Therefore, an ARPS (ADME related prognostic signature) was created in this study to examine the clinical implications of ARGs in patients with BC. METHODS: A multi-omics investigation of ADME-related genes in BC was conducted using bulk RNA sequencing, single-cell RNA sequencing, and spatial transcriptome data. According to the expression profiles of ADME-related differentially expressed genes (DEGs), the ARPS was calculated, and all patients were stratified based on their risk scores. A prediction model was then created using Cox regression and stepAIC analyses. This model divided all patients into HR (High risk) and LR (Low risk) groups following the median risk score. Bioinformatics analyses were conducted to estimate the risk signature's predictive capacity. RESULTS: This study identified five ARGs (SLC7A5, HSD11B1, ADHFE1, GSTM2, and TAP1) correlated with BC prognosis. The risk signature in the TCGA-BRCA, METABRIC, and GSE58812 cohorts revealed robust predictive accuracy for 1-, 3-, and 5-year survival. Compared to the gene signature alone, the nomogram integrating the ARPS and clinical parameters demonstrated improved prognostic performance. Immune infiltration analysis revealed a high level of immune checkpoint related gene expression and immune score in patients with ARPS LR, suggesting potential implications for immunotherapy responses. CONCLUSION: The findings highlight the prognostic significance of ARPS in BC and its potential utility in guiding personalized treatment strategies. Combining ARPS with clinical parameters enhances prognostic accuracy and may help patients with BC make clinical decisions.