A glycolysis-based three-gene signature predicts survival in patients with lung squamous cell carcinoma

Abstract

Background: Lung cancer is one of the most lethal and most prevalent malignant tumors worldwide, and lung squamous cell carcinoma (LUSC) is one of the major histological subtypes. Although numerous biomarkers have been found to be associated with prognosis in LUSC, the prediction effect of a single gene biomarker is insufficient, especially for glycolysis-related genes. Therefore, we aimed to develop a novel glycolysis-related gene signature to predict survival in patients with LUSC.

Methods: The mRNA expression files and LUSC clinical information were obtained from The Cancer Genome Atlas (TCGA) dataset.

Results: Based on Gene Set Enrichment Analysis (GSEA), we found 5 glycolysis-related gene sets that were significantly enriched in LUSC tissues. Univariate and multivariate Cox proportional regression models were performed to choose prognostic-related gene signatures. Based on a Cox proportional regression model, a risk score for a three-gene signature (HKDC1, ALDH7A1, and MDH1) was established to divide patients into high-risk and low-risk subgroups. Multivariate Cox regression analysis indicated that the risk score for this three-gene signature can be used as an independent prognostic indicator in LUSC. Additionally, based on the cBioPortal database, the rate of genomic alterations in the HKDC1, ALDH7A1, and MDH1 genes were 1.9, 1.1, and 5% in LUSC patients, respectively.

Conclusion: A glycolysis-based three-gene signature could serve as a novel biomarker in predicting the prognosis of patients with LUSC and it also provides additional gene targets that can be used to cure LUSC patients.

Keywords: Gene signature; Glycolysis; Lung cancer; Prognosis.

Fig. 1

The flow chart of study

Fig. 2

GSEA results for enrichment plots of five gene sets which were significantly differentiated between in LUSC and normal tissues. A, BIOCARTA_GLYCOLYSIS_ PATHWAY; B, GO_GLYCOLYTIC_PROCESS; C, HALLMARK_GLYCOLYSIS; D, KEGG_GLYCOLYSIS_GLUCONEOGENESIS; E, REACTOME_ GLYCOLYSIS. Abbreviations: GSEA, gene set enrichment analysis; LUSC, lung squamous cell carcinoma

Fig. 3

Differential expression of three genes in the normal tissues (n = 49) and tumor tissues (n = 501). (* p < 0.05, **p < 0.01, ***p < 0.001). A, ALDH7A1; B, HKDC1; C, MDH1

Fig. 4

A risk of three-gene signature predicted the overall survival in patients with LUSC. A, Distribution of risk score per patient; B, Survival status of each patients; C, A heatmap of three genes expression profile; D, Kaplan-Meier survival curve analysis for LUSC patients divided into the high-risk and low-risk groups. Abbreviation: LUSC, lung squamous cell carcinoma

Fig. 5

The time-independent ROC curve of the risk score for prediction the 1, 3, 5-year overall survival

Fig. 6

Validation for prognostic value of the risk score in different subgroups. A, Subgroup for age ≤ 65; B Subgroup for age > 65; C, Subgroup for Female; D Subgroup for Male; E, Subgroup for AJCC stage I-II; F, Subgroup for AJCC stage III-IV; G, Subgroup for T1 + T2; H, Subgroup for T3 + T4; I, Subgroup for M0; J, Subgroup for M1; K, Subgroup for N0; L, Subgroup for N1 + N2 + N3. Abbreviation: AJCC stage, American Joint Committee on Cancer stage