The ferroptosis signature predicts the prognosis and immune microenvironment of nasopharyngeal carcinoma

Ji Zhou, Tianyao Guo, Lin Zhou, Meihua Bao, Lin Wang, Wenhu Zhou, Shan Tan, Guangyi Li, Binsheng He, Zhen GuoScientific 2023 · DOI

Abstract

Nasopharyngeal carcinoma (NPC) is a cancer with a high metastatic rate and poor prognosis. Growing studies suggest that ferroptosis take part in the development of tumours. At the same time, the connection between ferroptosis-related genes (FRGs) and the prognosis of NPC remains unclear. In this study, we explored the dysregulated FRGs between normal control and tumour samples of NPC. Firstly, 14 of 36 differentially expressed FRGs were identified in NPC tissues compared to normal tissues, among which ABCC1, GLS2, CS and HMGCR were associated with poor prognosis for patients. The four ferroptosis genes were used for consensus cluster analysis and two risk-related FRGs (ABCC1 and GLS2) were used in a risk model. The ROC curve revealed the good predictive performance of this risk signature. Multivariate analysis revealed that risk score and intratumoral TILs were independent risk factors linked to prognosis. Additionally, our results suggested that the risk signature was attached to the immune microenvironment. Moreover, the NPC patients with high risk were sensitive to chemotherapeutic drugs including axitinib, docetaxel, embelin, epothilone.B, parthenolide, thapsigargin, tipifarnib, vinorelbine. Finally, the expression of ABCC1 and GLS2 was validated in NPC tissues using immunohistochemistry. Together, these results revealed ferroptosis may be a potential biomarker in NPC and representing a promising future direction in prognosis and therapeutic strategy for the treatment of NPC.

Introduction

Nasopharyngeal carcinoma (NPC) is a challenging malignancy characterized by high metastatic potential and a tendency for recurrence following standard treatments. Ferroptosis, an iron-dependent form of regulated cell death driven by lipid peroxidation, has emerged as a critical factor in tumor progression and therapeutic response. However, its specific role and prognostic value in NPC have remained largely unexplored. This study aimed to identify ferroptosis-related genes (FRGs) that could serve as reliable biomarkers to predict patient prognosis and guide personalized therapeutic strategies.

Methods

The researchers employed a multi-stage workflow integrating bioinformatics and experimental validation. They analyzed three Gene Expression Omnibus (GEO) datasets: GSE53819 and GSE12452 for discovery, and GSE102349 for prognostic analysis. Differentially expressed FRGs were identified, and LASSO and Cox regression analyses were used to construct a two-gene risk model (ABCC1 and GLS2). Immune cell infiltration was estimated using the CIBERSORT algorithm, and drug sensitivity was predicted via the pRRophetic R package. Finally, the expression of key genes was validated in a cohort of 11 NPC and 3 normal samples using immunohistochemistry (IHC).

Results

Identification of Differentially Expressed FRGs

The study identified 36 differentially expressed FRGs between NPC and normal tissues. Among these, 14 genes showed significant prognostic potential. The differential expression levels of these genes in the GSE12452 dataset highlight the widespread dysregulation of ferroptosis pathways in NPC.

Differential expression of ferroptosis-related genes in the GSE12452 dataset, comparing normal (blue) and tumor (red) tissues.

Heatmap analysis across multiple datasets further confirmed consistent expression patterns for these FRGs, providing a robust basis for model construction.

Heatmaps illustrating the expression patterns of ferroptosis-related genes across two independent datasets.

Prognostic Signature and Clinical Independence

The final risk model, based on ABCC1 and GLS2, demonstrated high predictive accuracy with an Area Under the Curve (AUC) of 0.837 for 3-year overall survival. Patients categorized as high-risk based on this signature had significantly shorter progression-free survival (PFS). Multivariate analysis confirmed that the risk score is an independent prognostic factor, providing value beyond traditional clinical staging.

Immune Microenvironment and Therapeutic Sensitivity

The risk signature correlated significantly with the tumor immune microenvironment. High-risk scores were positively associated with follicular helper T cells and resting dendritic cells, while negatively correlated with memory and naive B cells. Furthermore, the model predicted that high-risk patients are more sensitive to specific chemotherapeutic agents, including docetaxel, axitinib, and vinorelbine, offering a potential path for personalized treatment.

Discussion

This study establishes a novel two-gene ferroptosis signature as a robust prognostic tool for NPC. By linking ferroptosis to both survival outcomes and the immune landscape, the findings suggest that ABCC1 and GLS2 are not only biomarkers but potential therapeutic targets. While limited by a small validation cohort and the use of proxy data for some analyses, this research provides a strong foundation for future prospective clinical trials. Understanding the precise molecular mechanisms by which these genes modulate ferroptosis in NPC remains a priority for future work.

Summary updated

Abstract

Nasopharyngeal carcinoma (NPC) is a cancer with a high metastatic rate and poor prognosis. Growing studies suggest that ferroptosis take part in the development of tumours. At the same time, the connection between ferroptosis-related genes (FRGs) and the prognosis of NPC remains unclear. In this study, we explored the dysregulated FRGs between normal control and tumour samples of NPC. Firstly, 14 of 36 differentially expressed FRGs were identified in NPC tissues compared to normal tissues, among which ABCC1, GLS2, CS and HMGCR were associated with poor prognosis for patients. The four ferroptosis genes were used for consensus cluster analysis and two risk-related FRGs (ABCC1 and GLS2) were used in a risk model. The ROC curve revealed the good predictive performance of this risk signature. Multivariate analysis revealed that risk score and intratumoral TILs were independent risk factors linked to prognosis. Additionally, our results suggested that the risk signature was attached to the immune microenvironment. Moreover, the NPC patients with high risk were sensitive to chemotherapeutic drugs including axitinib, docetaxel, embelin, epothilone.B, parthenolide, thapsigargin, tipifarnib, vinorelbine. Finally, the expression of ABCC1 and GLS2 was validated in NPC tissues using immunohistochemistry. Together, these results revealed ferroptosis may be a potential biomarker in NPC and representing a promising future direction in prognosis and therapeutic strategy for the treatment of NPC.
BibTeX
@inproceedings{zhou2023ferroptosis,
  author = {Ji Zhou and Tianyao Guo and Lin Zhou and Meihua Bao and Lin Wang and Wenhu Zhou and Shan Tan and Guangyi Li and Binsheng He and Zhen Guo},
  title = {{The ferroptosis signature predicts the prognosis and immune microenvironment of nasopharyngeal carcinoma}},
  booktitle = {Scientific},
  year = {2023},
  doi = {10.1038/s41598-023-28897-2},
}
Figure