A Novel Ferroptosis-Related Biomarker Signature to Predict Overall Survival of Esophageal Squamous Cell Carcinoma

Abstract

Esophageal squamous cell carcinoma (ESCC) accounts for the main esophageal cancer (ESCA) type, which is also associated with the greatest malignant grade and low survival rates worldwide. Ferroptosis is recently discovered as a kind of programmed cell death, which is indicated in various reports to be involved in the regulation of tumor biological behaviors. This work focused on the comprehensive evaluation of the association between ferroptosis-related gene (FRG) expression profiles and prognosis in ESCC patients based on The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO). ALOX12, ALOX12B, ANGPTL7, DRD4, MAPK9, SLC38A1, and ZNF419 were selected to develop a novel ferroptosis-related gene signature for GEO and TCGA cohorts. The prognostic risk model exactly classified patients who had diverse survival outcomes. In addition, this study identified the ferroptosis-related signature as a factor to independently predict the risk of ESCC. Thereafter, we also constructed the prognosis nomogram by incorporating clinical factors and risk score, and the calibration plots illustrated good prognostic performance. Moreover, the association of the risk score with immune checkpoints was observed. Collectively, the proposed ferroptosis-related gene signature in our study is effective and has a potential clinical application to predict the prognosis of ESCC.

Keywords: GEO; TCGA; esophageal squamous cell carcinoma; ferroptosis; gene signature; prognosis.

FIGURE 1

Construction of the seven-ferroptosis-gene signature. (A) Cross-validation for tuning parameter screening upon LASSO regression analysis. (B) LASSO coefficient profiles for those intersected genes. (C) Forest plot of hazard ratios exhibiting the prognostic worth of seven FRGs.

FIGURE 2

Risk score of the prognostic signature that comprises seven FRGs for OS in four cohorts. (A) Risk score distribution and survival status of high- and low-risk patients. (B) Kaplan–Meier analysis on high- and low-risk patients. (C) Time-dependent ROC curve analyses on the GEO training set, GEO validation set, entire GEO set, and TCGA validation set.

FIGURE 3

Subgroup analyses based on different clinical features of ESCC cases: (A) age ≤65, (B) age >65, (C) male, (D) female, (E) stage I-II, (F) stage III, (G) T1-T2, (H) T3-T4, (I) N0, and (J) N1–N3.

FIGURE 4

Prognostic signature in combination with clinical parameters for predicting prognostic outcomes for ESCC cases. (A) Univariate analysis and (B) multivariate analysis containing the risk score and clinical factors. (C) Nomogram for predicting one-, three-, and five-year OS. (D–F) Calibration curves of nomogram on consistency between predicted and observed one-, three-, and five-year survival.

FIGURE 5

GSEA in high- and low-risk patients: (A) adhesion molecules, (B) chemokine signaling pathway, (C) KRAS signaling, and (D) IL-2/STAT5 signaling.

FIGURE 6

Immune checkpoint analysis. (A–B) Heatmap of immune checkpoints between high- and low-risk patients in GSE53625 and TCGA datasets. Differential expression of immune checkpoints in high- vs. low-risk patients: (C) BTLA, (D) CTLA4, (E) CD40, (F) CD40LG, (G) TIM-3, and (H) HHLA2.

FIGURE 7

Differentiation analysis of the signature genes among ESCCs and normal tissues based on GSE53625 dataset.

FIGURE 8

Protein expression of signature genes in HPA database. (A) protein expression of ALOX12. (B) protein expression of ALOX12B. (C) protein expression of MAPK9. (D) protein expression of SLC38A1. (E) protein expression of ZNF419.

FIGURE 9

Effects of inhibiting the expression of SLC38A1 on ESCC cell proliferation and migration. (A) SLC38A1 was upregulated within KYSE-150 and Eca109 cells relative to HEECs by the qRT-PCR assay and western blot analysis. (B) The expression of SLC38A1 was downregulated in KYSE-150 and Eca109 cells by siRNAs. (C–E) KYSE-150 and Eca109 cell proliferation after anti-SLC38A1 siRNA transfection was measured using CCK-8 assays (C, D) and colony formation assays (E). (F) Transwell assays for ESCC cell migration (scale bars: 100 μm).