Development and Validation of a Hypoxia-Associated Prognostic Signature Related to Osteosarcoma Metastasis and Immune Infiltration

Yucheng Fu, Qiyuan Bao, Zhuochao Liu, Guoyu He, Junxiang Wen, Qi Liu, Yiqi Xu, Zhijian Jin, Weibin ZhangFrontiers in Cell and Developmental Biology 2021 · DOI

Abstract

The current study established and validated a novel hypoxia-associated gene signature in osteosarcoma. It could act as a prognostic biomarker and serve as therapeutic guidance in clinical applications.

Introduction

Osteosarcoma remains the most prevalent primary bone malignancy among children and adolescents. While advancements in surgical techniques and multi-agent chemotherapy have improved the 5-year survival rate for localized cases to over 70%, the prognosis for patients with metastatic disease is significantly poorer, with survival rates dropping below 20%. A critical driver of this malignancy is hypoxia—a state of low oxygen common in solid tumors that promotes metastasis, chemotherapy resistance, and immune evasion. This study aimed to develop a validated hypoxia-related gene signature to predict metastasis and survival outcomes in osteosarcoma patients.

Methodology and Dataset Selection

Researchers utilized a multi-stage bioinformatics pipeline to identify key prognostic markers. The study drew data from two major public repositories: the TARGET database (97 patients) for the training cohort and the Gene Expression Omnibus (GEO) dataset GSE21257 (53 patients) for external validation.

The analytical workflow involved:

  1. Differential Expression Analysis: Identifying genes associated with metastasis.
  2. Hypoxia Filtering: Cross-referencing these genes with hypoxia-related sets from the Molecular Signature Database (MSigDB).
  3. Statistical Refinement: Using Univariate and Multivariate Cox regression to isolate independent prognostic factors.
  4. Immune Profiling: Employing single-sample gene set enrichment analysis (ssGSEA) to correlate risk scores with the tumor microenvironment.
Figure 4: The bioinformatics pipeline used to identify hypoxia-related differentially expressed mRNAs (DEmRNAs) and analyze their relationship with immune infiltration.

The Two-Gene Prognostic Signature

From an initial pool of hundreds of candidates, the study distilled a streamlined two-gene signature consisting of DCN (Decorin) and P4HA1 (Prolyl 4-hydroxylase subunit alpha 1). These genes appear to play opposing roles in tumor progression:

  • P4HA1: Identified as an oncogenic driver. Higher expression correlates with increased risk, likely due to its role in remodeling the extracellular matrix to facilitate invasion.
  • DCN: Identified as a protective factor. Its downregulation is a hallmark of high-risk, hypoxic tumors.

The resulting mathematical model for the risk score is: $Risk score = (4.01 \times P4HA1) + (-4.46 \times DCN)$

Hypoxia as an Immune Shield

A key finding of the research is the link between high hypoxia risk scores and an immunosuppressive tumor microenvironment. Patients classified as high-risk exhibited significantly lower infiltration of essential immune cells, including dendritic cells (DCs), macrophages, and neutrophils. This suggests that hypoxic conditions in osteosarcoma do not merely drive growth but actively shield the tumor from the host's immune response.

Figure 5: Comparison of immune infiltration features between high-risk (red) and low-risk (blue) patient groups, showing significant downregulation of several immune cell types in high-risk cases.

Clinical Results and Validation

The model demonstrated high predictive accuracy across both cohorts:

  • Survival Outcomes: High-risk patients had significantly lower survival rates (p < 0.05) compared to low-risk patients.
  • Predictive Power: The Area Under the Curve (AUC) for 1-year survival was 0.81.
  • Clinical Integration: A nomogram combining the gene signature with clinical factors (age and gender) achieved a C-index of 0.85, providing a reliable tool for clinical decision-making.

Conclusion and Future Directions

This study provides a robust, two-gene hypoxia signature that effectively predicts metastasis and survival in osteosarcoma. However, the authors note that the relatively small sample sizes and the retrospective nature of the data require further validation through larger, prospective clinical trials. Future research will also need to explore the specific molecular mechanisms by which DCN and P4HA1 interact within the hypoxic bone microenvironment.

Summary updated

Abstract

The current study established and validated a novel hypoxia-associated gene signature in osteosarcoma. It could act as a prognostic biomarker and serve as therapeutic guidance in clinical applications.
BibTeX
@inproceedings{fu2021development,
  author = {Yucheng Fu and Qiyuan Bao and Zhuochao Liu and Guoyu He and Junxiang Wen and Qi Liu and Yiqi Xu and Zhijian Jin and Weibin Zhang},
  title = {{Development and Validation of a Hypoxia-Associated Prognostic Signature Related to Osteosarcoma Metastasis and Immune Infiltration}},
  booktitle = {Frontiers in Cell and Developmental Biology},
  year = {2021},
  doi = {10.3389/fcell.2021.633607},
}
Figure