AUTHOR=Liu Shenting , Xu He , Feng Ying , Kahlert Ulf D. , Du Renfei , Torres-de la Roche Luz Angela , Xu Kai , Shi Wenjie , Meng Fanshuai
TITLE=Oxidative stress genes define two subtypes of triple-negative breast cancer with prognostic and therapeutic implications
JOURNAL=Frontiers in Genetics
VOLUME=14
YEAR=2023
URL=https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2023.1230911
DOI=10.3389/fgene.2023.1230911
ISSN=1664-8021
ABSTRACT=Introduction: Oxidative stress (OS)-related genes have been confirmed to be closely related to the prognosis of triple-negative breast cancer (TNBC) patients; despite this fact, there is still a lack of TNBC subtype strategies based on this gene guidance. Here, we aimed to explore OS-related subtypes and their prognostic value in TNBC.
Methods: Data from The Cancer Genome Atlas (TCGA)-TNBC and Sequence Read Archive (SRA) (SRR8518252) databases were collected, removing batch effects using a combat method before analysis. Consensus clustering analysis identified two OS subtypes (clusters A and B), with cluster A showing a better prognosis. Immune infiltration characteristics were analyzed using ESTIMATE and single-sample gene set enrichment analysis (ssGSEA) algorithms, revealing higher ImmuneScore and ESTIMATEscore in cluster A. Tumor-suppressive immune cells, human leukocyte antigen (HLA) genes, and three immune inhibitors were more prevalent in cluster A.
Results: An eight-gene signature, derived from differentially expressed genes, was developed and validated as an independent risk factor for TNBC. A nomogram combining the risk score and clinical variables accurately predicted patient outcomes. Finally, we also validated the classification effect of subtypes using hub markers of each subtype in the test dataset.
Conclusion: Our study reveals distinct molecular clusters based on OS-related genes to better clarify the reactive oxygen species (ROS)-mediated progression and the crosstalk between the ROS and tumor microenvironment (TME) in this heterogenetic disease, and construct a risk prognostic model which could provide more support for clinical treatment decisions.