A novel four-gene of iron metabolism-related and methylated for prognosis prediction of hepatocellular carcinoma

Figures & data

Figure 1.. Identification of genes that act as prognostic factors for HCC

(a) Volcano map describes the distribution of upregulated and downregulated differentially expressed genes (DEGs) involved in iron metabolism genes. Red, green and blue dots represent downregulated, no change and upregulated DEGs. (b-c) LASSO Cox regression analysis was used to identify prognostic factors for HCC from TCGA. (d) Univariate Cox regression analysis of prognostic factors. (e) Multivariate Cox regression analysis of prognostic factors.

Figure 2. Construction of the prognostic model

(a-b) High-risk patients with HCC were correlated with a higher death rate and shorter survival time. (c) Heatmap depicting the expression levels of the four genes in HCC tissue. (d) The KM curve depicting the OS of the patient cohort from TCGA. (e) A time-dependent receiver operating characteristic curve depicting the 1-, 3-, and 5-year OS events of the patient cohort from TCGA.

Table 1. Multivariate Cox regression analysis of the gene signature in HCC patients

Variables		Univariate analysis			Multivariate analysis
	HR (95%Cl)	Coefficient	P-value	HR (95%Cl)	Coefficient	P-value
Age Gender Gender male Histologic grade Pathologic T Pathologic N Pathologic stage Risk score	1(0.99–1) 0.73(0.47–1.1) NA 1.1(0.79–1.4) 1.7(1.4–2.1) 2(0.48–8) 1.7(1.4–2.2) 1.8(1.5–2.2)	0.0081 –0.31 NA 0.056 0.52 0.67 0.55 0.59	0.35 0.16 NA 0.71 2.3e-06 2.3e-09 3.7e-06 2.3e-09	1(1–1) NA 0.88(0.55–1.4) 0.99(0.73–1.4) 1.6(0.65–3.9) 1.8(0.27–12) 0.97(0.37–2.6) 1.7(1.4–2.1)	0.014 NA -0.13 –0.0053 0.46 0.57 –0.026 0.52	0.13 NA 0.59 0.97 0.31 0.55 0.96 2.8e-06

The risk score calculated using the gene signature was found to be an independent variable correlated with the prognosis of HCC (P = 2.8e-06).

HR, hazard ratio; CI, confidence interval. Values in bold indicate significant P-value < 0.05.

Figure 3. Validation of the gene signature in a patient cohort from the ICGC

(a-b) High-risk patients with HCC were correlated with a higher death rate and shorter survival time. (c) Heatmap depicting the expression levels of the four genes in HCC tissue. (d) The KM curve depicting the OS of the patient cohort from the ICGC. (e) A time-dependent receiver operating characteristic curve depicting the 1- and 3-year OS events of the patient cohort from the ICGC.

Figure 4. Co-expression network of the four iron metabolism-related and methylated genes

(a)The co-expression network of the four genes that are a part of the identified gene signature. The rose red nodes denote the key genes, and the green nodes denote genes, which co-expressed with key genes. (b) Functional enrichment analysis with gene set enrichment analysis tool.

Figure 5. Immunohistochemical analysis of RRM2, FTCD, CYP2C9, and ATP6V1C1 in adjacent non-cancerous and cancerous tissues from HCC patients (original magnification, ×200)

(a) RRM2 expression in adjacent non-cancerous tissue. (b) RRM2 expression in cancerous tissue. (c) FTCD expression in adjacent non-cancerous tissue. (d) FTCD expression in cancerous tissue. (e) CYP2C9 expression in adjacent non-cancerous tissue. (f) CYP2C9 expression in cancerous tissue. (g) ATP6V1C1 expression in adjacent non-cancerous tissue. (h) ATP6V1C1 expression in cancerous tissue.

Data availability statement

The data that support the findings of the study are available in The Cancer Genome Atlas (https://cancergenome.nih.gov/), the Molecular Signatures Database (https://www.gsea-msigdb.org/gsea/msigdb), and the International Cancer Genome Consortium (https://dcc.icgc.org/).