Exploring Key Biomarkers and Common Pathogenesis of Seven Digestive System Cancers and Their Correlation with COVID-19

Abstract

Digestive system cancer and COVID-19 significantly affect the digestive system, but the mechanism of interaction between COVID-19 and the digestive system cancers has not been fully elucidated. We downloaded the gene expression of COVID-19 and seven digestive system cancers (oral, esophageal, gastric, colorectal, hepatocellular, bile duct, pancreatic) from GEO and identified hub differentially expressed genes. Multiple verifications, diagnostic efficacy, prognostic analysis, functional enrichment and related transcription factors of hub genes were explored. We identified 23 common DEGs for subsequent analysis. CytoHubba identified nine hub genes (CCNA2, CCNB1, CDKN3, ECT2, KIF14, KIF20A, KIF4A, NEK2, TTK). TCGA and GEO data validated the expression and excellent diagnostic and prognostic ability of hub genes. Functional analysis revealed that the processes of cell division and the cell cycle were essential in COVID-19 and digestive system cancers. Furthermore, six related transcription factors (E2F1, E2F3, E2F4, MYC, TP53, YBX1) were involved in hub gene regulation. Via in vitro experiments, CCNA2, CCNB1, and MYC expression was verified in 25 colorectal cancer tissue pairs. Our study revealed the key biomarks and common pathogenesis of digestive system cancers and COVID-19. These may provide new ideas for further mechanistic research.

Keywords: COVID-19; common biomarkers; diagnosis; digestive system cancers; pathogenesis; prognosis.

Figure 1

Research design flow chart.

Figure 2

Venn diagram of upregulated and downregulated common DEGs in the eight diseases. (A) Twenty-two overlapping upregulated DEGs. (B) One overlapping downregulated DEG.

Figure 3

GO, KEGG, and PPI network of common DEGs. (A,B) KEGG and GO pathway enrichment analysis results. p < 0.05 was considered significant. (C) PPI network. Red indicates upregulated genes.

Figure 4

Hub gene analysis. (A) Network of PPI of the top nine hub genes from cytoHubba. (B) Hub genes and their co-expression genes were analyzed via GeneMANIA. (C,D) Analysis by GO and KEGG of hub gene enrichment. (* p < 0.05; ** p < 0.01; *** p < 0.001).

Figure 5

Validation of hub genes in TCGA. (A–G) Hub gene expression levels of OC (A), ESC (B), GC (C), CRC (D), HCC (E), BDC (F), and PC (G) in TCGA cohorts. (H–N) Diagnostic AUCs in seven TCGA cohorts. (*** p < 0.001).

Figure 6

Hub gene expression levels in GSE37991 (A), GSE96668 (B), GSE66229 (C), GSE44076 (D), GSE112790 (E), GSE60979 (F), GSE15471 (G). (* p < 0.05; ** p < 0.01; *** p < 0.001).

Figure 7

Differential expression, independent diagnostic efficiency, and receiver operating characteristic (ROC) diagnostic model efficiency of hub genes in the blood of patients with HCC and COVID-19. (A–C) Differential expression, diagnostic efficiency, and ROC diagnostic model between normal tissue and HCC (GSE114564). (D–F) Differential expression, diagnostic efficiency, and ROC diagnostic model between hepatitis and HCC (GSE114564). (G–I) Differential expression, diagnostic efficiency, and ROC diagnostic model between normal tissue and severe COVID-19 (GSE171110). (J–L) Differential expression, diagnostic efficiency, and ROC diagnostic model between normal tissue and COVID-19 (GSE152418). (** p < 0.01; *** p < 0.001).

Figure 8

Survival curves of CCNA2, CCNB1, CDKN3, ECT2, KIF14, KIF20A, KIF4A, NEK2 and TTK in TCGA cohorts. (A) Survival curves of hub genes in OC. (B) Survival curves of hub genes in HCC. (C) Survival curves of hub genes in PC.

Figure 9

Transcription factor regulation of hub genes and transcription factor expression level in TCGA. (A) The regulatory network between the transcription factors and hub genes. Yellow indicates transcription factors, red indicates the hub genes. (B–H) The transcription factor expression levels between tumor and normal tissue in TCGA OC (B), ESC (C), GC (D), CRC (E), HCC (F), BDC (G), and PC (H) cohorts. (* p < 0.05; ** p < 0.01; *** p < 0.001).

Figure 10

qPCR expression level of CCNA2, CCNB1, and MYC in 25 pairs of colorectal tissues. Each dot represents the expression of normal or CRC sample. (** p < 0.01; *** p < 0.001).