Pancreatic ductal adenocarcinoma (PDAC) is an extremely malignant tumor. The immune profile of PDAC and the immunologic milieu of its tumor microenvironment (TME) are unique; however, the mechanism of how the TME engineers the carcinogenesis of PDAC is not fully understood. This study is aimed at better understanding the relationship between the immune infiltration of the TME and gene expression and identifying potential prognostic and immunotherapeutic biomarkers for PDAC. Analysis of data from The Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus (GEO) databases identified differentially expressed genes (DEGs), including 159 upregulated and 53 downregulated genes. Gene Ontology analysis and Kyoto Encyclopedia of Genes and Genomes enrichment were performed and showed that the DEGs were mainly enriched for the PI3K-Akt signaling pathway and extracellular matrix organization. We used the cytoHubba plugin of Cytoscape to screen out the most significant ten hub genes by four different models (Degree, MCC, DMNC, and MNC). The expression and clinical relevance of these ten hub genes were validated using Gene Expression Profiling Interactive Analysis (GEPIA) and the Human Protein Atlas, respectively. High expression of nine of the hub genes was positively correlated with poor prognosis. Finally, the relationship between these hub genes and tumor immunity was analyzed using the Tumor Immune Estimation Resource. We found that the expression of SPARC, COL6A3, and FBN1 correlated positively with infiltration levels of six immune cells in the tumors. In addition, these three genes had a strong coexpression relationship with the immune checkpoints. In conclusion, our results suggest that nine upregulated biomarkers are related to poor prognosis in PDAC and may serve as potential prognostic biomarkers for PDAC therapy. Furthermore, SPARC, COL6A3, and FBN1 play an important role in tumor-related immune infiltration and may be ideal targets for immune therapy against PDAC.
Pancreatic ductal adenocarcinoma (PDAC) is an extremely malignant tumor of the digestive system, with a 5-year survival rate of only 8% [
In recent years, cancer immunotherapy has become an active area of cancer research. This approach is designed to improve the immunogenicity of tumor cells, stimulate and enhance the antitumor immune response, and inhibit tumor growth and progression. Since the FDA approval of the first targeted drug ipilimumab (humanized anti-CTLA-4 IgG1 monoclonal antibody) in 2011, six more immune checkpoint inhibitors have been approved for cancer therapy [
In the current study, we analyzed pancreatic cancer gene expression data from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases to identify differentially expressed genes (DEGs). Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment and protein-protein interaction (PPI) network analyses were performed to reveal the interactive relationships between the DEGs to explore the underlying molecular mechanisms involved in the carcinogenesis and progression of PDAC. Subsequently, we used the cytoHubba plugin of Cytoscape to search for hub genes and identify the most significant hub genes using four different models (Degree, Maximal Clique Centrality (MCC), Density of Maximum Neighborhood Component (DMNC), and Maximum Neighborhood Component (MNC)). We further validated the expression levels of the hub genes and their clinical relevance using Gene Expression Profiling Interactive Analysis (GEPIA) and the Human Protein Atlas, respectively. Finally, we evaluated the associations between the prognosis-related genes and several immune cell types based on the Tumor Immune Estimation Resource (TIMER). In summary, our study identified potential prognostic biomarkers and immunotherapeutic targets for PDAC.
The mRNA expression profiles of PDAC and adjacent normal tissues were obtained from GEPIA (TCGA Data Online Analysis Tool;
Gene Ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment were performed on the DEGs, and the results were visualized using the R package. A
The STRING database (
We analyzed the expression of the hub gene using GEPIA (
We analyzed the association of patient prognosis with the hub genes using the human protein atlas website. Based on the fragments per kilobase million (FPKM) value of each gene, pancreatic cancer patients were classified into two expression groups (high expression and low expression), and the correlation between the expression level and patient survival was examined. The prognosis of each group of patients was examined using Kaplan-Meier survival estimators, and the survival outcomes of the two groups were compared by the log-rank test.
TIMER is an online website that uses RNA-Seq expression profiling data to detect immune cell infiltration in more than 30 cancer types [
We identified the differentially expressed genes in PDAC in the TCGA database using GEPIA (Figure
Identification of DEGs in pancreatic cancer using three databases. (a) The differentially expressed genes from the TCGA database were determined by GEPIA. (b) The volcano map for the GSE15471 dataset. (c) A Venn diagram showing 159 upregulated genes in PDAC. (d) A Venn diagram showing 53 downregulated genes in PDAC.
To analyze the underlying interplay of the DEGs, GO analysis and KEGG pathway enrichment were performed using the R package. Based on the GO analysis, the upregulated DEGs mainly participated in the extracellular matrix organization, extracellular structure organization, positive regulation of cell migration, skeletal system development, and cell-substrate adhesion (Figure
Functional enrichment analysis of DEGs. (a) Top ten enriched biological processes for the DEGs. (b) Top ten enriched KEGG pathways for the DEGs. (c) Hierarchical clustering of gene expression profiles in each KEGG pathway. (d) Chord plots showing the relationship between the hub genes and the KEGG pathway.
To explore the potential interactions between the DEGs, we used the STRING database to study the relationship between the various DEGs. We identified ten hub genes based on the Degree (Figure
Identification of hub genes. (a–d) The hub genes were identified using four models (Degree, MCC, DMNC, and MNC) with the Cytoscape plugin cytoHubba. (e) A Venn diagram was used to identify the ten hub genes in PDAC.
To verify the differential expression of the hub genes between PDAC and normal pancreatic tissue, we analyzed the ten hub genes using the GEPIA-based TCGA database. We found that the mRNA expression levels of the hub genes were significantly increased in PDAC compared to normal pancreatic tissue (Figure
Expression analysis of ten hub genes in PDAC based on GEPIA. The mRNA expression levels in TCGA pancreatic tumors (
Validation of protein expression and genetic alterations of ten hub genes in pancreatic cancer. (a) Representative protein expression of eight genes in pancreatic cancer tumors and normal tissue. Data was obtained from the Human Protein Atlas database. (b) Genetic alterations of ten genes in pancreatic cancer. Data was obtained from the cBioPortal.
To estimate the influence of hub gene expression on the prognosis of PDAC, we performed survival analysis for the hub genes using the Human Protein Atlas online tool for differential analysis. High expression of nine of the hub genes was positively correlated with poor prognosis (COL5A2
Survival analysis of the ten hub genes in PDAC based on the Human Protein Atlas. (a) BGN; (b) COL5A2; (c) COL6A; (d) COL11A1; (e) COL12A1; (f) FBN1; (g) POSTN; (h) SPARC; (i) THBS2; (j) VCAN.
To explore whether there was a correlation between immune cell infiltration into the tumors and hub gene expression, we analyzed the relationships between the ten hub gene signatures and tumor purity and six important immune cell types (CD4+ T cells, CD8+ T cells, B cells, neutrophils, macrophages, and dendritic cells) using TIMER. We observed that most of these prognosis-related genes were positively correlated with the infiltrating levels of the different immune cell types but negatively related to tumor purity (Supplementary Figure
Correlation analysis of prognosis-related genes with tumor-infiltrating immune cell types using TIMER.
Since immunotherapy is currently focused on immunological checkpoint inhibitors (e.g., CTLA4, PDCD1, PDCD1LG2, and CD274), we chose SPARC, COL6A3, and FBN1 as potential target genes and further analyzed the coexpression relationship of these three genes with these immune checkpoint-related genes by TIMER. We found that all three of these potential target genes had strong coexpression relationships with CTLA4, PDCD1, PDCD1LG2, and CD274 (Figure
The association between the expression levels of SPARC (a), COL6A3 (b), and FBN1 (c) with immune checkpoints using TIMER.
Pancreatic cancer is one of the most malignant and aggressive tumors. Traditional treatment methods (e.g., surgery, chemotherapy, radiotherapy, and other locoregional therapies) provide low survival rates. Currently, several clinical studies have focused on immunotherapeutic strategies in pancreatic cancer [
In the current study, we found 212 reliable DEGs in PDAC by the comprehensive analysis of three datasets. These differentially expressed genes were then subjected to GO and KEGG pathway enrichment analysis, which indicated that these DEGs were mainly enriched in pathways involved in the extracellular matrix (ECM) and extracellular structure organization and the PI3K-Akt signaling pathway. The ECM, a major component of the tumor microenvironment, is known to undergo significant changes during angiogenesis and tumor progression [
Immune cell infiltration in the tumor microenvironment has received much attention and has become a promising therapeutic target. To further investigate the relationship between these prognosis-related genes and different immune cell types, we predicted gene-immune cell interactions using TIMER. We found that the expression levels of SPARC, COL6A3, and FBN1 were significantly correlated with six infiltrating immune cell types (CD4+ T cells, CD8+ T cells, B cells, neutrophils, macrophages, and dendritic cells).
SPARC is a multifunctional calcium-binding glycoprotein that is usually secreted into the extracellular matrix and plays a key role in proliferation, migration, adhesion, and differentiation [
COL6A3 is an extracellular matrix protein that is typically found in most connective tissues, including muscle, skin, tendon, and vessels. Many recent studies have demonstrated the important role of COL6A3 in the diagnosis and prognosis of colorectal, lung, and prostate cancer [
FBN1 encodes fibrillin, which is the primary component of microfibrils in the extracellular matrix. A previous study reported that FBN1 overexpression plays a key role in the development of germ cell tumors [
Previous studies have suggested that the tumor microenvironment is one of the significant factors determining the prognosis of PDAC and may even play a role in the resistance to treatment [
Taken together, our results suggest that the nine high expression hub genes (COL5A2, COL6A3, COL11A1, COL12A1, FBN1, POSTN, THBS2, SPARC, and VCAN) may be associated with the development and prognosis of PDAC. All these genes may be potential prognostic biomarkers for PDAC. In addition, three hub genes (SPARC, COL6A3, and FBN1) may also play a vital role in the microenvironment of pancreatic cancer through the regulation of tumor-infiltrating immune cells, suggesting they could serve as potential therapeutic targets for the modulation of the antitumor immune response. Nevertheless, further investigation is needed to confirm the mechanisms underlying the potential roles of these biomarkers in the immune microenvironment.
The data used to support the findings of this study are available from the corresponding author upon request.
The authors declare that they have no conflicts of interest.
LPZ designed and supervised the study. HL and CZ analyzed the data, manufactured the figures, and wrote the manuscript. TZ, YH, YNS, and LPZ reviewed the manuscript. All authors have read and approved the final manuscript.
Supplementary Figure 1: correlation analysis of the expression levels of BGN, COL5A2, COL11A1, COL12A1, POSTN, THBS2, and VCAN with tumor-infiltrating immune cell types using TIMER.
Supplementary Table 1: relationship between 10 hub gene expression and immune infiltration level in PDAC.