Identification of Hub Genes in Idiopathic Pulmonary Fibrosis and NSCLC Progression:Evidence From Bioinformatics Analysis

Abstract

Background: Lung cancer is the most common comorbidity of idiopathic pulmonary fibrosis. Thus there is an urgent need for the research of IPF and carcinogenesis Objective: The objective of this study was to explore hub genes which are common in pulmonary fibrosis and lung cancer progression through bioinformatic analysis. Methods: All the analysis was performed in R software. Differentially expressed genes (DEGs) were explored by comparing gene expression profiles between IPF tissues and healthy lung tissues from GSE24206, GSE53845, GSE101286 and GSE110147 datasets. Venn Diagram analysis was used to identify the overlapping genes, while GO and KEGG pathway enrichment analysis were used to explore the biological functions of the DEGs using clusterprofiler package. Hub genes were identified by analyzing protein-protein interaction networks using Cytoscape software. Nomogram was constructed using the rms package. Tumor immune dysfunction and exclusion (TIDE) and Genomics of Drug Sensitivity in Cancer (GDSC) analysis was used to quantify the immunotherapy and chemotherapy sensitivity of non-small cell lung cancer (NSCLC) patients. Results: COL1A1, COL3A1, MMP1, POSTN1 and TIMP3 were identified as the top five hub genes. The five hub genes were used to construct a diagnostic nomogram that was validated in another IPF dataset. Since the hub genes were also associated with lung cancer progression, we found that the nomogram also had diagnostic value in NSCLC patients. These five genes achieved a statistically difference of overall survival in NSCLC patients (p < 0.05). The expression of the five hub genes was mostly enriched in fibroblasts. Fibroblasts and the hub genes also showed significant ability to predict the susceptibility of NSCLC patients to chemotherapy and immunotherapy. Conclusion: We identified five hub genes as potential biomarkers of IPF and NSCLC progression. This finding may give insight into the underlying molecular mechanisms of IPF and lung cancer progression and provides potential targets for developing new therapeutic agents for IPF patients.

Keywords: bioinformatics analysis; hub genes; idiopathic pulmonary fibrosis; lung cancer progression; pathways.

FIGURE 1

Identification of DEGs in IPF patients. Notes: (A–D): DEGs analysis was performed between IPF and normal lung tissue with the threshold of p < 0.05 and |logFC| > 1 based on the GSE24206, GSE53845, GSE101286 and GSE110147; (E): A Venn graph analysis intersected 31 common DEGs in GSE24206, GSE53845, GSE101286 and GSE110147 database.

FIGURE 2

Five hub genes were identified in IPF samples. Notes: (A): PPI network of the DEGs; (B): Top ten central elements calculated by connectivity degree in the PPI network; (C): Top five central elements named hub genes calculated by connectivity degree in the PPI network; (D): Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment pathways were explored based on the 31 common DEGs.

FIGURE 3

Expression level of the five hub genes in IPF and NSCLC patients. Notes: (A): Heatmap comparing hub gene expression levels between IPF tissue and normal lung tissue in GSE24206, GSE53845, GSE101286 and GSE110147 datasets, ns = p > 0.05, * = p < 0.05, ** = p < 0.01, *** = p < 0.001.; (B): The expression difference of hub genes between cancer tissue and normal lung tissue in TCGA-LUAD patients, ns = p > 0.05, * = p < 0.05, ** = p < 0.01, *** = p < 0.001.; (C): The expression difference of hub genes between cancer tissue and normal lung tissue in TCGA-LUSC patients, ns = p > 0.05, * = p < 0.05, ** = p < 0.01, *** = p < 0.001.

FIGURE 4

Prognosis value of hub genes. Notes: (A): Kaplan-Meier curves showed the prognosis value of COL1A1 in NSCLC patients; (B): Kaplan-Meier curves showed the prognosis value of COL3A1 in NSCLC patients; (C): Kaplan-Meier curves showed the prognosis value of MMP1 in NSCLC patients; (D): Kaplan-Meier curves showed the prognosis value of POSTN in NSCLC patients; (E): Kaplan-Meier curves showed the prognosis value of TIMP3 in NSCLC patients.

FIGURE 5

Analysis of the association between the five hub genes and clinicopathological features of NSCLC patients. Notes: (A): The expression level of five hub genes in NSCLC patients with different T stage (T1-2 vs T3-4; T1-2: 872 individuals, T3-4: 162 individuals), ns = p > 0.05, * = p < 0.05, ** = p < 0.01, *** = p < 0.001; (B): The expression level of five hub genes in NSCLC patients with different N stage (N0 vs N1-3; N0: 668 individuals, N1-3: 347 individuals), ns = p > 0.05, * = p < 0.05, ** = p < 0.01, *** = p < 0.001; (C): The expression level of five hub genes in NSCLC patients with different M stage (M0 vs M1; M0: 773 individuals, M1: 32 individuals), ns = p > 0.05, * = p < 0.05, ** = p < 0.01, *** = p < 0.001; (D): The expression level of five hub genes in NSCLC patients with different clinical stage (Stage I-II vs Stage III-IV; Stage I-II: 824 individuals, Stage III-IV: 201 individuals), ns = p > 0.05, * = p < 0.05, ** = p < 0.01, *** = p < 0.001; (E): The expression level of five hub genes in NSCLC patients with different gender (Female vs Male; Female: 417 individuals, Male: 620 individuals), ns = p > 0.05, * = p < 0.05, ** = p < 0.01, *** = p < 0.001; (F): The expression level of five hub genes in NSCLC patients with different age (≤ 65 vs >65; ≤ 65: 446 individuals, >65: 563 individuals), ns = p > 0.05, * = p < 0.05, ** = p < 0.01, *** = p < 0.001; (G): The expression level of five hub genes in NSCLC patients with smoking years (< 40 vs ≥40; < 40: 332 individuals, ≥40: 472 individuals), ns = p > 0.05, * = p < 0.05, ** = p < 0.01, *** = p < 0.001; (H): The expression level of five hub genes in smokers and non-smokers (No vs Yes; No: 93 individuals, Yes: 918 individuals), ns = p > 0.05, * = p < 0.05, ** = p < 0.01, *** = p < 0.001.

FIGURE 6

Analysis of the expression landscape of the five hub genes using Human cell landscape analysis. Notes: (A): Single-cell analysis evaluating the expression of hub genes in Kidney1 based on the Human cell landscape data; (B): Single-cell analysis evaluating the expression of hub genes in Kidney2 based on the Human cell landscape data; (C): Single-cell analysis evaluating the expression of hub genes in Kidney3 based on the Human cell landscape data.

FIGURE 7

A diagnostic nomogram model was created based on the hub genes and verified in IPF and NSCLC patients. Notes: (A–B): Correlation between cancer-associated fibroblasts and patients prognosis (OS and DSS); (C): A diagnostic nomogram model was constructed based on the hub genes; (D): The AUC curve of the model in predicting the diagnostic value of IPF; (E): The AUC curve of the model in predicting the diagnostic value of NSCLC.

FIGURE 8

Association between the five hub genes and response to immunotherapy and chemotherapy. Notes: (A): The difference of cancer-associated fibroblasts infiltration level between immunotherapy responder and non-responder patients based on the TCGA-LUAD cohort, ns = p > 0.05, * = p < 0.05, ** = p < 0.01, *** = p < 0.001; (B): The difference of hub gene expression between immunotherapy responder and non-responder patients based on the TCGA-LUAD cohort, ns = p > 0.05, * = p < 0.05, ** = p < 0.01, *** = p < 0.001; (C): The difference of cancer-associated fibroblasts infiltration level between immunotherapy responder and non-responder patients based on the TCGA-LUSC cohort, ns = p > 0.05, * = p < 0.05, ** = p < 0.01, *** = p < 0.001; (D): The difference of hub gene expression between immunotherapy responder and non-responder patients based on the TCGA-LUSC cohort, ns = p > 0.05, * = p < 0.05, ** = p < 0.01, *** = p < 0.001; (E): Correlations between cancer associated fibroblasts, five hub genes and response to cisplatin in NSCLC patients; (F): Correlations between cancer associated fibroblasts, five hub genes and response to paclitaxel in NSCLC patients.