An investigation to identify tumor microenvironment-related genes of prognostic value in lung squamous cell carcinoma based on The Cancer Genome Atlas

Huan Liu; Boxuan Liu; Lei Zhang; Mingzhen Li; Cheng Chen; Shaohua He; Tingting Luo; Xiaohui He; Liming Tan

doi:10.21037/tcr-21-401

An investigation to identify tumor microenvironment-related genes of prognostic value in lung squamous cell carcinoma based on The Cancer Genome Atlas

Transl Cancer Res. 2021 Apr;10(4):1885-1899. doi: 10.21037/tcr-21-401.

Authors

Huan Liu¹, Boxuan Liu¹, Lei Zhang^{2

3

4}, Mingzhen Li¹, Cheng Chen¹, Shaohua He¹, Tingting Luo¹, Xiaohui He¹, Liming Tan⁵

Affiliations

¹ Department of Precision Medicine Center, The Second People's Hospital of Huaihua, Huaihua, China.
² Key Laboratory of Ministry of Education for Medicinal Plant Resource and Natural Pharmaceutical Chemistry, Xi'an, China.
³ National Engineering Laboratory for Resource Development of Endangered Chinese Crude Drugs in Northwest China, Xi'an, China.
⁴ College of Life Sciences, Shaanxi Normal University, Xi'an, China.
⁵ Clinical Pharmacy Center, The Second People's Hospital of Huaihua, Huaihua, China.

Abstract

Background: Lung squamous cell carcinoma (LUSC) is a prevalent and lethal malignancy with a poor clinical prognosis. Major constituents of the tumor microenvironment (TME) include infiltrating immune cells and stromal cells, which play a pivotal role in the progression and growth of the disease. To improve the understanding of the prognostic influence of immune and stromal cell-related genes for patients with the disease, we performed a comprehensive bioinformatics analysis to identify TME-relevant biomarkers, and investigated the potential role of these candidate signatures in LUSC.

Methods: Estimation of STromal and Immune cells in MAlignant Tumor tissues using Expression data (ESTIMATE) assessed the samples of LUSC obtained from The Cancer Genome Atlas (TCGA). The samples were grouped according to their immune/stromal scores (high or low). Multivariate cox regression and receiver operating characteristic curves (ROC) were implemented to construct the risk assessment model for prognosis prediction. The co-upregulated differentially expressed genes (DEGs) in the immune and stromal groups were used for further analyses. Overall survival (OS) curves were used to determine the prognostic value of the DEGs, and the TME-related DEGs were verified with Gene Expression Omnibus (GEO) datasets. The functional assessments were performed include Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and protein-protein interaction (PPI) analyses.

Results: The immune/stromal scores calculated by ESTIMATE showed significant associations with OS (log-rank P<0.05). In addition, the prognostic risk score model based on immune and stromal scores also showed significant correlations with OS. A total of 94 TME-related genes were obviously related to poor OS. Among them, BHMT2, FES, HSPB7, NOVA2, LPAP2, and SEMA3B (BFHNLS) were confirmed using GSE4573 and GSE17710 datasets. The functional assessments exhibited those TME-related genes mostly participate in immune response, cytokine-cytokine receptor interaction, and metabolic pathways, which elucidated the probable correlation of TME with tumorigenesis in LUSC.

Conclusions: In this study, 6 potential biomarkers named BFHNLS were identified as TME-related genes with prognostic value based on immune and stromal scores of LUSC patients of TCGA, and were verified using GEO datasets, which might serve as therapeutic targets.

Keywords: ESTIMATE; Lung squamous cell carcinoma (LUSC); biomarker; tumor microenvironment (TME).