Fig. 7: Highly proliferative malignant hepatocytes in HBV-related HCC foster more immunosuppressive TME than non-HBV-HCC.

A UMAP plot of re-clustered malignant cells derived from HBV and non-HBV liver tumor. Cluster number annotations are provided in the Figure. B UMAP plot showing malignant cells, color-coded according to patient origin (left panel). The UMAP plot showing tumor type by color, HBV, or non-HBV HCC (right panel). C UMAP colored by viral status in malignant cells, HBV-infected (green), and non-infected cells (red). D Sankey diagram showing the percentages of malignant cell clusters among HCC patients and vice versa. E Expression of marker genes correlated with malignant hepatocytes (ALB, TTR, APOA2, GPC3, SERPINA1, and AFP). F Boxplot showing the fractions of malignant cells in HBV (blue) and non-HBV tumor patients (green). Error bars represent mean ± standard deviation (SD). G Based on the HALLMARK gene set of upregulated genes, the bar chart showing the enrichment of specific pathways in HBV (upper panel) and non-HBV (lower panel) malignant cells. H Violin plot indicating the average score of metastatic, immune escape, immune surveillance, and proliferation signals in malignant cells across HBV (blue) and non-HBV (green) tumor samples in discovery (left and middle panel) and validation cohort 2 (right panel). I Violin plot showing selected immune escape genes (MDK, CD47, and TGFB1) in tumor cells from HBV (blue) and non-HBV (green) HCC patients in discovery (upper panel) and validation cohort 2 (lower panel). The p-values in (H and I) were calculated using Student’s t-test, and in (F) by the unpaired Wilcoxon test. All statistical comparisons were performed on biologically independent samples (n = 5 HBV-HCC and n = 3 non-HBV-HCC) from both the discovery and validation cohort 2. See also Supplementary Fig. 13.