Single-cell sequencing combined with bulk RNA seq reveals the roles of Natural Killer Cell in prognosis and immunotherapy of Hepatocellular carcinoma.

Background Hepatocellular carcinoma (HCC) is a type of highly heterogeneous tumor characterized by a high mortality rate and poor prognosis. Natural Killer cells (NK cells) are important immune cells that play a role in anti-tumor activities, antiviral responses, and immune regulation. The relationship between NK cells and HCC remains unclear. It would be valuable to identify a NK-related prognostic signature for HCC. Methods WGCNA and single-cell sequencing RNA were performed to identify NK cell related genes. Gene Enrichment Analysis were used to identify the potential signal pathway. After combing genes from WGCNA and scRNA, Unicox, LASSO + StepCox and Multicox analysis were used to filter prognostic-related gene and construct a prognostic model. Then we performed Proposed time analysis to identify the developmental trajectories of NK cells. Finally, ssGSEA and estimate methods were used to evaluate the immune microenvironment and immune sensitivity drugs. Results Using the scRNA-seq data, we identified 1396 genes with high NK cell scores. Based on the results of scRNA-seq, 250 NK-related genes were identified from WGCNA. We identified 223 intersecting genes between the scRNA-seq and WGCNA. After integrating clinical data with the bulk RNA-seq data of these intersecting genes, we constructed a prognostic model to accurately predict the prognosis of HCC patients. Eventually, we found that high-risk HCC patients exhibited worse survival outcomes and lower sensitivity to immunotherapy. Conclusion We constructed a risk model based on NK cell-related genes that can predict the prognosis of HCC patients accurately. This model can also predict the immunotherapy response of HCC effectively.