Yunxiang Wang, Sheng Xu, Junfeng Liu, Pan Qi

Head and Neck Breast Department, Xinxiang Central Hospital, The Fourth Clinical College of Xinxiang Medical University, Xinxiang, Henan, 453000, People’s Republic of China

Correspondence: Pan Qi, Head and Neck Breast Department, Xinxiang Central Hospital, The Fourth Clinical College of Xinxiang Medical University, No. 56 Jinsui Avenue, Weibin District, Xinxiang, Henan, 453000, People’s Republic of China, Email 413960747@qq.com

Background: Peroxisomes are increasingly linked to cancer development, yet the prognostic role of peroxisome-related genes (PRGs) in breast cancer remains unclear.
Objective: This study aimed to construct a prognostic model based on PRG expression in breast cancer to clarify their prognostic value and clinical implications.
Methods: Transcriptomic data from TCGA and GEO were used for training and validation cohorts. TME characteristics were analyzed with ESTIMATE, MCP-counter, and CIBERSORT algorithms. qPCR validated mRNA expression levels of risk genes, and data analysis was conducted in R.
Results: Univariate and multivariate Cox regression identified a 7-gene PRG risk signature (ACBD5, ACSL5, DAO, NOS2, PEX3, PEX10, and SLC27A2) predicting breast cancer prognosis in training (n=1069), internal validation (n=327), and external validation (merged from four GEO datasets, n=640) datasets. While basal and Her2 subtypes had higher risk scores than luminal subtypes, a significant prognostic impact of the PRG risk signature was seen only in luminal subtypes. The high-risk subgroup exhibited a higher frequency of focal synonymous copy number alterations (SCNAs), arm-level amplifications and deletions, and single nucleotide variations. These increased genomic aberrations were associated with greater immune suppression and reduced CD8+ T cell infiltration. Bulk RNA sequencing and single-cell analyses revealed distinct expression patterns of peroxisome-related genes (PRGs) in the breast cancer TME: PEX3 was primarily expressed in malignant and stromal cells, while ACSL5 showed high expression in T cells. Additionally, the PRG risk signature demonstrated efficacy comparable to that of well-known biomarkers for predicting immunotherapy responses. Drug sensitivity analysis revealed that the PRG high-risk subgroup was sensitive to inhibitors of BCL-2 family proteins (BCL-2, BCL-XL, and MCL1) and other kinases (PLK1, PLK1, BTK, CHDK1, and EGFR).
Conclusion: The PRG risk signature serves as a promising biomarker for evaluating peroxisomal activity, prognosis, and responsiveness to immunotherapy in breast cancer.

Keywords: invasive breast cancer, peroxisome, cancer metabolism, transcriptomic analysis, biomarker, immunotherapy