Abstract
BACKGROUND: Hepatocellular carcinoma (HCC) presents challenges and opportunities for immunotherapy due to its intricate immune microenvironment. Exhausted CD8 T cells (CD8Tex) are pivotal in this context but are inadequately characterized in HCC. METHODS: We conducted single-cell analysis on the GSE140228 dataset to identify key genes associated with CD8Tex in HCC. Cell communication analysis elucidated strong interactions of CD8Tex with CD8 T cells, dendritic cells (DCs), and monocytes/macrophages. RESULTS: Pearson correlation analysis using the TCGA-LIHC dataset identified CD8Tex-associated long non-coding RNAs (lncRNAs). Utilizing univariate and multivariate Cox regression analyses, along with LASSO regression to prevent overfitting, we developed a prognostic model incorporating 5 lncRNAs. This model exhibited strong prognostic performance, and the derived risk score was validated as an independent predictor of overall survival in HCC patients. Among these lncRNAs, AL158166.1 showed the strongest correlation with CD8⁺ T cell exhaustion and was significantly associated with poor prognosis, highlighting its potential as both a biomarker and therapeutic target in HCC. CONCLUSION: Our study not only elucidates the role of CD8Tex cells in HCC but also proposes a novel molecular classification of the disease. This classification holds promise for guiding clinical immunotherapy and precision treatments tailored to different molecular subtypes of HCC, as identified through drug sensitivity analysis. This work provides a foundational framework for advancing clinical strategies in HCC treatment and the development of targeted therapies.