Identification of the key genes and characterizations of Tumor Immune Microenvironment in Lung Adenocarcinoma (LUAD) and Lung Squamous Cell Carcinoma (LUSC)

Lemeng Zhang, Jianhua Chen, Tianli Cheng, Hua Yang, Haitao Li, Changqie PanJournal of Cancer 2020 · DOI

TL;DR

This study delves into the tumor immune microenvironment (TME) of two major lung cancer subtypes, LUAD and LUSC, across different stages. By analyzing gene expression and clinical data, researchers identified distinct immune cell profiles and key genes that correlate with tumor progression and patient survival, offering potential therapeutic targets for non-small cell lung cancer (NSCLC).

What Problem Does This Paper Address?

Lung cancer, particularly non-small cell lung cancer (NSCLC), remains a leading cause of cancer-related deaths. While significant progress has been made with immunotherapies, understanding the intricate interplay between the tumor and its surrounding immune cells (the tumor immune microenvironment or TME) is crucial for developing more effective treatments. The TME can vary significantly based on cancer type and stage, influencing treatment response and patient prognosis. This paper aims to systematically characterize these differences between lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) across various TNM stages, identifying key genes and immune cell populations that are critical for understanding disease progression and predicting outcomes.

Methods

The researchers employed a multi-faceted bioinformatics approach. They began by downloading gene expression and clinical data for LUAD and LUSC from the Genomic Data Commons (GDC) database.

  • Immune Microenvironment Analysis: They calculated immune scores and immune cytolytic activity scores using established R packages (ESTIMATE, GZMA, PRF1). Tumor Mutation Burden (TMB) was assessed using Maftools. Tumor-infiltrating immune cells (TIICs) were quantified using the CIBERSORT algorithm, which estimates the abundance of 22 different immune cell types.
  • Differential Gene Expression (DEG) Analysis: They identified genes that were differentially expressed between tumor and normal tissues, and more importantly, between early (Stage I/II) and advanced (Stage III/IV) stages within LUAD and LUSC.
  • Correlation and Survival Analysis: Spearman correlation was used to link the abundance of TIICs with differentially expressed genes to identify immune-related DEGs. These immune-related DEGs were then analyzed for their correlation with patient survival (Overall Survival - OS) using Kaplan-Meier curves and log-rank tests.
  • Functional Enrichment and Network Analysis: Gene Ontology (GO) and KEGG pathway enrichment analyses were performed on the DEGs to understand their biological functions. Protein-protein interaction (PPI) networks were constructed using STRING and visualized with Cytoscape to identify key gene modules.

Datasets

The primary dataset used in this study was obtained from the Genomic Data Commons (GDC) database. This included:

  • RNA-seq gene expression data (FPKM format) for LUAD and LUSC.
  • Corresponding clinical phenotype data (including TNM staging, survival status, smoking history, etc.).

The data was downloaded in August 2019 and re-annotated using the Gencode database.

Key Insights and Ideas

  • Distinct Immune Signatures: LUAD and LUSC, despite being subtypes of NSCLC, exhibit notable differences in their TME composition and dynamics across disease stages. This highlights the need for subtype-specific therapeutic strategies.
  • Early Stage Immune Advantage: In LUAD, earlier stages show a more robust immune response (higher immune scores, more memory B cells and M0 macrophages) compared to advanced stages. Conversely, LUSC shows an advantage in early stages with higher cytolytic activity and neoantigen burden.
  • Key Immune Cell Players: M0 and M2 macrophages are consistently abundant in both LUAD and LUSC. However, specific immune cell types like memory B cells (LUAD) and plasma cells (LUSC) are also prominent.
  • TMB Trends Differ: The tumor mutation burden (TMB) generally decreases with advancing stage in LUAD but increases in LUSC. This divergence is significant for understanding treatment response, particularly to immunotherapies.
  • Survival-Associated Genes: A substantial number of immune-related genes were identified as differentially expressed and correlated with survival, particularly in LUAD. This provides a rich source for potential prognostic biomarkers and therapeutic targets.

Key Results

  • Immune Scores: LUAD showed significantly higher immune scores in early stages compared to advanced stages. LUSC did not show a significant difference in immune scores between stages.
  • Immune Cytolytic Activity: LUSC showed significantly higher immune cytolytic activity in early stages compared to advanced stages, while LUAD did not show a significant difference.
  • Tumor Mutation Burden (TMB): The median TMB decreased from stage I to IV in LUAD, while it increased from stage I to IV in LUSC.
  • Immune-Related DEGs: 83 immune-related DEGs were identified in LUAD, and 9 in LUSC.
  • Survival Correlation: Of the 83 immune-related DEGs in LUAD, 23 genes were significantly correlated with survival. In LUSC, only 2 out of 9 immune-related DEGs were significantly correlated with survival.
  • Specific Genes: Genes like PLK1 and RRM2 were identified as immune-related DEGs significantly associated with survival in LUAD. SHANK1 was identified as a crucial immune-related DEG for survival in LUSC.

Limitations and Future Work

The authors acknowledge several limitations:

  • Statistical Rigor: The use of multiple hypothesis tests in statistical analyses could be a concern.
  • Clinical Data Heterogeneity: The TCGA clinical data for staging includes different AJCC staging editions, which could introduce variability.
  • Immunotherapy Application: While immune characteristics were analyzed, their direct clinical application in immunotherapy requires further investigation with larger sample sizes.
  • SCLC Exclusion: The study focused on LUAD and LUSC, excluding Small Cell Lung Cancer (SCLC), which constitutes about 15% of lung cancers. Further research into SCLC's TME is needed.

Future work should focus on:

  • In Vitro and In Vivo Validation: Experimental validation of the identified key genes (PLK1, RRM2, SHANK1) and their roles in proliferation, metastasis, and invasion is crucial.
  • Prognostic Biomarker Development: Determining if these genes can serve as independent prognostic biomarkers and their association with immunotherapy efficacy requires further clinical studies.
  • Broader TME Analysis: Investigating the clinical application of TME characteristics in immunotherapy with larger cohorts.
  • SCLC Research: Expanding the study to include SCLC to provide a more comprehensive understanding of lung cancer TME.

Summary updated

Abstract

This study aimed to investigate the key genes and immune microenvironment involved in different TNM stages of lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC). The gene expression and clinical characteristics data were downloaded from the genomic data commons (GDC) database. After initial data processing, the characteristics of the immune microenvironment were analyzed. The differentially expressed genes (DEGs) in tumor vs. normal, and in early vs. advanced stages were screened, followed by Spearman correlation test for tumor infiltrating immune cells (TIICs) to identify immune-related genes. Finally, functional enrichment, protein-protein interaction, and survival analyses were performed. In LUAD, early stage was with higher immune scores, greater number of memory B cells and M0 macrophages compared to advanced stage. M0 and M2 macrophages, and resting memory CD4+ T cells accounted for a large proportion of TIICs in LUAD. The abundance of M0 macrophage infiltration was significantly correlated with the TNM stage and survival. In LUSC, early stage was with higher cytolytic activity and neoantigen burden compared to advanced stage. M0 and M2 macrophages, and plasma cells accounted for a large proportion of TIICs in LUSC. The abundance of resting and activated mast cells was significantly correlated with TNM stage, while resting dendritic cells, eosinophils, activated memory CD4 T cells, and mast cells were significantly correlated with prognosis. Tumor mutation burden analysis revealed that the median of variants per sample decreased from stage I to IV in LUAD, while it increased in LUSC. Further, 83 and 9 immune-related DEGs were identified in LUAD and LUSC, respectively, of which 23 genes in LUAD and 2 genes in LUSC correlated with survival. In conclusion, we identified the key genes, and characterized the tumor immune microenvironment in LUAD and LUSC which may provide therapeutic targets for the treatment of NSCLC.
BibTeX
@article{zhang2020identification,
  author = {Lemeng Zhang and Jianhua Chen and Tianli Cheng and Hua Yang and Haitao Li and Changqie Pan},
  title = {{Identification of the key genes and characterizations of Tumor Immune Microenvironment in Lung Adenocarcinoma (LUAD) and Lung Squamous Cell Carcinoma (LUSC)}},
  journal = {Journal of Cancer},
  year = {2020},
  doi = {10.7150/jca.42531},
}
Figure