Hepatocellular carcinoma microenvironment analysis using single-cell RNA-seq to search for immunotherapy-sensitive subtypes

[Objective]Angiogenesis inhibitors and immune checkpoint inhibitors targeting the tumor microenvironment (TME) have recently become the mainstay of therapy for hepatocellular carcinoma (HCC), but the response rate is still insufficient. Therefore, we performed TME analysis by single-cell RNA-seq(scRNA-seq) to predict the efficacy of immunotherapy in HCC. [Methods]Flex scRNA-seq analysis was performed on 81 frozen tissue samples (63 HCC samples undergoing hepatectomy, 10 noncancerous liver tissue samples, and 8 donor liver tissue samples) to examine the association between cell type frequency, gene expression, and clinical characteristics. We also performed TME classification by hierarchical clustering based on cell type frequency of noncancerous cells and cancer cell classification based on Chiang molecular classification. [Results]The 63 cases consisted of 69.8% male and 31.7% positive for portal vein invasion (VP). VP-positive cases had a significantly higher frequency of regulatory T cells (Treg), TREM2+ macrophages and anti-tumor immunity was suppressed in TME. The 63 cases were stratified into 4 groups according to TME classification, with about 30% in the CD8-positive T-cell-rich group. In a classification using approximately 250,000 cancer cells, PFS was significantly shorter in the PROLIFERAION group, which correlated with Treg infiltration.[Conclusion]scRNA-seq analysis allows detailed TME classification of HCC and suggests the possibility of identifying subtypes that may be associated with the efficacy of immunotherapy.