Analysis of clinical significance and prospective molecular mechanism of main elements of the JAK/STAT pathway in hepatocellular carcinoma

Abstract

Hepatocellular carcinoma (HCC) is one the most common malignancies and has poor prognosis in patients. The aim of the present study is to explore the clinical significance of the main genes involved in the Janus kinase (JAK)‑signal transducer and activator of transcription (STAT) pathway in HCC. GSE14520, a training cohort containing 212 hepatitis B virus‑infected HCC patients from the Gene Expression Omnibus database, and data from The Cancer Genome Atlas as a validation cohort containing 370 HCC patients, were used to analyze the diagnostic and prognostic significance for HCC. Joint‑effect analyses were performed to determine diagnostic and prognostic significance. Nomograms and risk score models were constructed to predict HCC prognosis using the two cohorts. Additionally, molecular mechanism analysis was performed for the two cohorts. Prognosis‑associated genes in the two cohorts were further validated for differential expression using reverse transcription‑quantitative polymerase chain reaction of 21 pairs of hepatitis B virus‑infected HCC samples. JAK2, TYK2, STAT3, STAT4 and STAT5B had diagnostic significance in the two cohorts (all area under curves >0.5; P≤0.05). In addition, JAK2, STAT5A, STAT6 exhibited prognostic significance in both cohorts (all adjusted P≤0.05). Furthermore, joint‑effect analysis had advantages over using one gene alone. Molecular mechanism analyses confirmed that STAT6 was enriched in pathways and terms associated with the cell cycle, cell division and lipid metabolism. Nomograms and risk score models had advantages for HCC prognosis prediction. When validated in 21 pairs of HCC and non‑tumor tissue, STAT6 was differentially expressed, whereas JAK2 was not differentially expressed. In conclusion, JAK2, STAT5A and STAT6 may be potential prognostic biomarkers for HCC. JAK2, TYK2, STAT3, STAT4 and STAT5B may be potential diagnostic biomarkers for HCC. STAT6 has a role in HCC that may be mediated via effects on the cell cycle, cell division and lipid metabolism.

Figure 1

Relative mRNA expressions of key genes of JAK/STAT pathway. Relative mRNA expressions of key genes of JAK/STAT pathway in tumor and non-tumor tissue of the (A) GSE14520 cohort and (B) TCGA cohort. Analysis of low and high expressions of key genes of JAK/STAT pathway in the (C) GSE14520 cohort and (D) TCGA cohort. ^*P≤0.05, ^**P≤0.01, ^***P≤0.001, ^****P≤0.0001, ^#P>0.05. TCGA, The Cancer Genome Atlas; JAK, Janus kinase; TYK2, tyrosine kinase 2; STAT, signal transducer and activator of transcription.

Figure 2

Diagnostic ROC curves of key genes of JAK/STAT pathway using the GSE14520 dataset. ROC curves of JAK/STAT pathway members: (A) JAK1, (B) JAK2, (C) JAK3, (D) TYK2, (E) STAT1, (F) STAT2, (G) STAT3, (H) STAT4, (I) STAT5A, (J) STAT5B and (K) STAT6. ROC curves of combinations: (L) STAT2 + 4, (M) STAT2 + 6, (N) STAT4 + 6 and (O) STAT2 + 4 + 6, respectively. ROC, receiver operating characteristic; AUC, area under the curve; CI, confidence interval; JAK, Janus kinase; TYK2, tyrosine kinase 2; STAT, signal transducer and activator of transcription.

Figure 3

Diagnostic ROC curves of key genes of JAK/STAT pathway using data from The Cancer Genome Atlas. ROC curves of JAK/STAT pathway members: (A) JAK1, (B) JAK2, (C) JAK3, (D) TYK2, (E) STAT1, (F) STAT2, (G) STAT3, (H) STAT4, (I) STAT5A, (J) STAT5B and (K) STAT6. ROC curves of combination of (L) JAK1 + 2, (M) JAK1 + STAT3, (N) JAK1 + TYK2, (O) JAK2 + STAT3, (P) JAK2 + TYK2, (Q) STAT3 + TYK2, (R) JAK1 + 2 + STAT3, (S) JAK1 + 2 + TYK2, (T) JAK1 + STAT3 + TYK2, (U) JAK2 + STAT3 + TYK2 and (V) JAK1 + 2 + STAT3 + TYK2. ROC, receiver operating characteristic; AUC, area under the curve; CI, confidence interval; JAK, Janus kinase; TYK2, tyrosine kinase 2; STAT, signal transducer and activator of transcription.

Figure 4

Kaplan-Meier plots of key genes of JAK/STAT pathway in the GSE14520 cohort. (A) JAK1, (B) JAK2, (C) JAK3, (D) TYK2, (E) STAT1, (F) STAT2, (G) STAT3, (H) STAT4, (I) STAT5A, (J) STAT5B and (K) STAT6. JAK, Janus kinase; TYK2, tyrosine kinase 2; STAT, signal transducer and activator of transcription.

Figure 5

Kaplan-Meier plots of key genes of JAK/STAT pathway in The Cancer Genome Atlas cohort. (A) JAK1, (B) JAK2, (C) JAK3, (D) TYK2, (E) STAT1, (F) STAT2, (G) STAT3, (H) STAT4, (I) STAT5A, (J) STAT5B and (K) STAT6. JAK, Janus kinase; TYK2, tyrosine kinase 2; STAT, signal transducer and activator of transcription.

Figure 6

Joint-effect analysis of prognosis-related genes in two cohorts. Joint-effect analysis of (A) JAK2 + STAT5A, (B) JAK2 + STAT6, (C) STAT5A + 6, (D) JAK2 + STAT5A + 6 in the GSE14520 cohort; Joint-effect analysis of (E) JAK2 + STAT5A, (F) JAK2 + STAT6, (G) STAT5A + 6 and (H) JAK2 + STAT5A+ 6 in The Cancer Genome Atlas cohort, respectively.

Figure 7

GO and KEGG pathway results of STAT6 in the GSE14520 cohort. (A-L) GO term and (M-P) KEGG pathways in gene set enrichment analysis. STAT6, signal transducer and activator of transcription 6; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes.

Figure 8

GO and KEGG pathway results of STAT6 in The Cancer Genome Atlas cohort. (A-L) GO term and (M-P) KEGG pathways in gene set enrichment analysis. STAT6, signal transducer and activator of transcription 6; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes.

Figure 9

Risk score model, Kaplan-Meier plot, and survival ROC curves in GSE14520 cohort. (A) Risk score model including risk score ranking, survival status, and JAK2, STAT5A and STAT6 heat map. (B) Kaplan-Meier plot of low and high risk score groups. (C) Time-dependent ROC curves at 1, 2, 3, 4 and 5 years. JAK, Janus kinase; STAT, signal transducer and activator of transcription; ROC, receiver operating characteristic; AUC, area under the curve.

Figure 10

Risk score model, Kaplan-Meier plot, and survival ROC curves in The Cancer Genome Atlas cohort. (A) Risk score model including risk score ranking, survival status, and JAK2, STAT5A and STAT6 heat map. (B) Kaplan-Meier plot of low and high risk score groups. (C) Time-dependent ROC curves at 1, 2, 3, 4 and 5 years. JAK, Janus kinase; STAT, signal transducer and activator of transcription; ROC, receiver operating characteristic; AUC, area under the curve.

Figure 11

Correlation matrix and interactive networks among JAK/STAT pathway members. Correlation matrix of key genes of JAK/STAT pathway in the (A) GSE14520 and (B) The Cancer Genome Atlas cohorts. (C) Co-expression and pathway interactions of key genes of JAK/STAT pathway. (D) Protein-protein interaction network. ^*P≤0.05, ^**P≤0.01. JAK, Janus kinase; TYK2, tyrosine kinase 2; STAT, signal transducer and activator of transcription.