IGFBPs were associated with stemness, inflammation, extracellular matrix remodeling and poor prognosis of low-grade glioma

Zhihui Liu, Hang Ji, Wenchao Fu, Shuai Ma, Hongtao Zhao, Fang Wang, Jiawei Dong, Xiuwei Yan, Jiheng Zhang, Nan Wang, Jiasheng Wu, Shaoshan HuFrontiers in Endocrinology 2022 · DOI

TL;DR

This study identifies a novel "IGFBPScore" that effectively predicts prognosis in low-grade glioma (LGG). The score is linked to key tumor characteristics like stemness, inflammation, and extracellular matrix remodeling, offering new therapeutic targets for this challenging cancer.

What Problem Does This Paper Address?

Low-grade gliomas (LGGs) are a type of brain tumor that, despite being "low-grade," can be highly aggressive and have a poor prognosis. Current treatments like surgery, radiation, and chemotherapy have limitations, and understanding the underlying biological mechanisms driving LGG progression is crucial for developing better therapies. The extracellular matrix (ECM) and its associated proteins, like the Insulin-like Growth Factor Binding Proteins (IGFBPs), are known to play roles in tumor development, but their specific involvement in LGG and their potential as prognostic markers have not been fully explored. This paper aims to fill that gap by investigating the IGFBP family's role in LGG and developing a predictive tool based on their expression.

Methods

The researchers employed a multi-faceted approach:

  • Differential Expression Analysis: They compared IGFBP gene expression in normal brain tissue versus LGG samples using RNA-sequencing data from The Cancer Genome Atlas (TCGA) and the Chinese Glioma Genome Atlas (CGGA).
  • Prognostic Analysis: Kaplan-Meier survival analysis and Cox regression were used to assess the relationship between IGFBP expression levels and patient survival.
  • IGFBPScore Development: A LASSO (Least Absolute Shrinkage and Selection Operator) regression model was used to identify the most significant IGFBP genes and construct a risk score (IGFBPScore) for predicting overall survival (OS) in LGG. This score was then validated in independent datasets.
  • Correlation Analysis: The IGFBPScore was correlated with various clinical traits, molecular subtypes, genomic alterations (mutations, copy number variations), epigenetic profiles (methylation subtypes), stemness indices, and the tumor microenvironment (immune cell infiltration, extracellular matrix components).
  • Single-Sample Gene-Set Enrichment Analysis (ssGSEA): This technique was used to quantify the enrichment of specific biological pathways, including immune functions, metabolic pathways, and DNA damage repair processes, within individual tumor samples.

Datasets

The study utilized several publicly available datasets:

  • TCGA (The Cancer Genome Atlas): Provided RNA-seq and clinical data for LGG patients.
  • CGGA (Chinese Glioma Genome Atlas): Provided RNA-seq and clinical data for two cohorts of LGG patients (mRNAseq_325 and mRNAseq_693).
  • GTEx (Genotype-Tissue Expression): Provided RNA-seq data for normal cerebral cortex tissues.
  • PanCanAtlas Publications & MEXPRESS database: Used for obtaining genomic and molecular subtype information.

No new benchmark datasets were introduced.

Key Insights and Ideas

  • IGFBP family as prognostic markers: The study strongly suggests that the expression levels of most IGFBP family members are significantly altered in LGG and are directly correlated with patient prognosis, with higher expression generally indicating a poorer outcome.
  • IGFBPScore as a powerful predictor: The developed IGFBPScore, derived from a parsimonious set of IGFBP genes, demonstrated significant predictive power for overall survival in LGG, outperforming individual IGFBP genes.
  • High IGFBPScore = Aggressive phenotype: Tumors with a high IGFBPScore exhibited a distinct molecular profile resembling glioblastoma (GBM), characterized by specific gene mutations (e.g., EGFR, PTEN, NF1), hypomethylation, and a more aggressive stem cell profile.
  • Microenvironment differences: High IGFBPScore tumors were associated with a more inflamed tumor microenvironment, increased extracellular matrix, and altered metabolic pathways, all contributing to a more malignant phenotype.
  • Stemness and EMT/MET: The high-IGFBPScore subgroup showed elevated stemness indices and enhanced epithelial-mesenchymal transition (EMT) and mesenchymal-epithelial transition (MET) processes, suggesting a more plastic and migratory cell population.

Key Results

  • Most IGFBP family members were found to be aberrantly expressed in LGG compared to normal brain tissue.
  • IGFBP1, IGFBP3, IGFBP4, IGFBP5, and IGFBP7 were significantly overexpressed in LGG.
  • The IGFBPScore was constructed using IGFBP2 and IGFBP5, with LASSO coefficients of 0.0165918577 and 0.0004779568, respectively.
  • The IGFBPScore was significantly associated with overall survival in both TCGA and CGGA cohorts, with higher scores correlating with shorter survival times.
  • High IGFBPScore tumors showed a distinct mutation pattern (IDH1wt, EGFRmut, PTENmut, NF1mut) compared to low IGFBPScore tumors (IDH1mut, TP53mut, CICmut, ATRXmut).
  • High IGFBPScore was linked to increased immune cell infiltration, higher tumor mutational burden, and enhanced antigen presentation.
  • Metabolic pathways, except for energy metabolism and the tricarboxylic acid cycle, were significantly elevated in high-IGFBPScore tumors.

Limitations and Future Work

The authors acknowledge several limitations:

  • Correlation vs. Causation: The study primarily demonstrates correlations, and further experimental validation is needed to establish causal relationships.
  • Intra-tumoral Heterogeneity: The analysis was based on bulk tumor tissue, which may not fully capture the complex intra-tumoral heterogeneity of LGG.
  • Pathway Analysis Limitations: The functional classification score (FCS) approach, while useful, can lead to gene overlap issues when analyzing individual pathways.

Future work should focus on:

  • Experimental Validation: Conducting in vitro and in vivo experiments to confirm the functional roles of IGFBPs and the IGFBPScore in LGG.
  • Therapeutic Targeting: Investigating the IGFBP family as potential therapeutic targets for LGG, particularly in combination with existing treatments.
  • Mechanism Elucidation: Further exploring the precise molecular mechanisms by which IGFBPs influence stemness, inflammation, ECM remodeling, and other tumor-promoting processes in LGG.

Summary updated

Abstract

Our study provides a summary analysis and a theoretical basis for the biological role and clinical value of the IGFBP family in LGG, providing an important therapeutic target for LGG.
BibTeX
@inproceedings{liu2022igfbps,
  author = {Zhihui Liu and Hang Ji and Wenchao Fu and Shuai Ma and Hongtao Zhao and Fang Wang and Jiawei Dong and Xiuwei Yan and Jiheng Zhang and Nan Wang and Jiasheng Wu and Shaoshan Hu},
  title = {{IGFBPs were associated with stemness, inflammation, extracellular matrix remodeling and poor prognosis of low-grade glioma}},
  booktitle = {Frontiers in Endocrinology},
  year = {2022},
  doi = {10.3389/fendo.2022.943300},
}
Figure