Background and Aims: Tumor microenvironment (TME) heterogeneity leads to a discrepancy in survival prognosis and clinical treatment response for hepatocellular carcinoma (HCC) patients. The clinical applications of documented molecular subtypes are constrained by several issues. Approach and Results: We integrated three single-cell datasets to describe the TME landscape and identified six prognosis-related cell subclusters. Unsupervised clustering of subcluster-specific markers was performed to generate transcriptomic subtypes. The predictive value of these molecular subtypes for prognosis and treatment response was explored in multiple external HCC cohorts and the Xiangya HCC cohort. TME features were estimated using single-cell immune repertoire sequencing, mass cytometry and multiplex immunofluorescence. The prognosis-related score (PRS) was constructed based on machine learning algorithm. Comprehensive single-cell analysis described TME heterogeneity in HCC. The five transcriptomic subtypes possessed different clinical prognoses, stemness characteristics, immune landscapes and therapeutic responses. Class 1 exhibited an inflamed phenotype with better clinical outcomes, while Classes 2 and 4 were characterized by a lack of T cell infiltration. Classes 5 and 3 indicated an inhibitory tumor immune microenvironment. Analysis of multiple therapeutic cohorts suggested that Classes 5 and 3 were sensitive to ICB and targeted therapy, whereas Classes 1 and 2 were more responsive to transcatheter arterial chemoembolization treatment. Class 4 displayed resistant to all conventional HCC therapies. Three potential therapeutic agents and four targets were further identified for high-PRS HCC patients. Conclusion: Our study generated a clinically valid molecular classification to guide precision medicine in patients with HCC. Export

Background: Nonalcoholic steatohepatitis (NASH)-driven hepatocellular carcinoma (HCC) is becoming a major health-related problem. The exploration of NASH-related prognostic biomarkers and therapeutic targets is necessary. Methods: Data were downloaded from the GEO database. The “glmnet” package was used to identify differentially expressed genes (DEGs). The prognostic model was constructed by the univariate Cox and LASSO regression analyses. Validation of the expression and prognosis by immunohistochemistry (IHC) in vitro. Drug sensitivity and immune cell infiltration were analyzed by CTR-DB and ImmuCellAI. Results: We constructed a prognostic model that identified the NASH-related gene set (DLAT, IDH3B, and MAP3K4), which was validated in a real-world cohort. Next, seven prognostic transcription factors (TFs) were identified. The prognostic ceRNA network included three mRNAs, four miRNAs, and seven lncRNAs. Finally, we found that the gene set was associated with drug response which was validated in six clinical trial cohorts. Moreover, the expression level of the gene set was inversely correlated with CD8 T cell infiltration in HCC. Conclusions: We established a NASH-related prognostic model. Upstream transcriptome analysis and the ceRNA network provided clues for mechanism exploration. The mutant profile, drug sensitivity, and immune infiltration analysis further guided precise diagnosis and treatment strategies.