Drug Resistance Analysis of Pancreatic Cancer Based on Universally Differentially Expressed Genes

Abstract

The high heterogeneity between patients can complicate the diagnosis and treatment of pancreatic ductal adenocarcinoma (PDAC). Here, we explored the association of universally differentially expressed genes (UDEGs) with resistance to chemotherapy and immunotherapy in the context of pancreatic cancer. In this work, sixteen up-regulated and three down-regulated genes that were dysregulated in more than 85% of 102 paired and 5% of 521 unpaired PDAC samples were identified and defined as UDEGs. A single-cell level analysis further validated the high expression levels of the up-UDEGs and the low levels of the down-UDEGs in cancer-related ductal cells, which could represent the malignant changes seen in pancreatic cancer. Based on a drug sensitivity analysis, we found that ANLN, GPRC5A and SERPINB5 are closely related to the resistance mechanism of PDAC, and their high expression predicted worse survival for PDAC patients. This suggests that targeting these genes could be a potential way to reduce drug resistance and improve survival. Based on the immune infiltration analysis, the abnormal expression of the UDEGs was found to be related to the formation of an immunosuppressive tumor microenvironment. In conclusion, these UDEGs are common features of PDAC and could be involved in the resistance of pancreatic cancer and might serve as novel drug targets to guide research into drug repurposing.

Keywords: drug resistance; immunity; pancreatic ductal adenocarcinoma; universally differentially expressed genes.

Figure A1

MSI counts and TMB in PDAC. (A) MSI statistics for TCGA PDAC patients. (B) Comparison of TMB between TCGA PDAC, LUSC and SKCM patients. MSI-H: high microsatellite instability; MSI-L: low microsatellite instability; MSS: microsatellite stability; PDAC: pancreatic ductal adenocarcinoma; SKCM: skin cutaneous melanoma; LUSC: lung squamous cell carcinoma.

Figure A2

Relationship between resistance to gemcitabine and expression of ANLN and GPRC5A. (A) Expression of ANLN and GPRC5A in TCGA PDAC patient groups with different responses to gemcitabine. (B) Number of significant correlations between drug resistance and UDEG expression.

Figure 1

Identification of UDEGs in PDAC. (A) Workflow for identifying UDEGs in PDAC. (B) Overlapping genes in two DEG lists were defined as UDEGs. (C) Dysregulation frequencies of nineteen UDEGs. (D) Brief information on UDEGs.

Figure 2

Functional analysis of UDEGs. (A) UDEGs and their direct neighbors in PPI network. Light blue represents neighbor genes, red represents upregulated genes, and green represents downregulated genes. (B) KEGG enrichment analysis of UDEGs and their direct neighbor genes. (C) Pathways with different functions that are hierarchically enriched with UDEGs according to Reactome.

Figure 3

Single-cell analysis of 24 PDAC and 11 control samples. t-SNE plot of clustering and cell-type annotation of tumor (A) and normal (B) samples. Cell numbers for each cell type in tumor (C) and normal (D) samples. Expression of UDEGs in different cell types in tumor (E) and normal (F) samples.

Figure 4

Correlation analysis between expression of UDEGs and resistance to anticancer drugs. (A) Overview of correlations between IC50 of drug in pancreatic cells and UDEG expression. (B) Overview of significant correlations between drug resistance and expression of UDEGs. (C) Survival analysis of ANLN, GPRC5A and SERPINB5 in TCGA PDAC patients. (D) DGI network of UDEGs and their PPI partners. Light blue represents neighbor genes, red represents upregulated genes, green represents downregulated genes, and purple represents drugs.

Figure 5

Roles of UDEGs in response of PDAC patients to immunotherapy. (A). Overview of immune cell proportions in TCGA PDAC patients. (B). Correlation between UDEGs and immune cell proportions. “×” represents a non-significant correlation. Genes labeled in green are down-regulated, and those in red are up-regulated.