A CD8+ T cell-related genes prognostic model for hepatocellular carcinoma patients

Immune checkpoint blockade (ICB) therapy is a promising treatment for hepatocellular carcinoma (HCC). However, it is limited by low response rates. Tumour-infiltrating immune cells (TIICs), specifically CD8(+) T cells, in the tumour microenvironment (TME) are related to tumour immune responses and are potential predictors for immunotherapeutic and survival outcomes. Therefore, we established a new CD8(+) T cell-related genes prognostic index (CD8T-RGPI) for evaluating immune responses and prognostic outcomes for HCC. Using the TCGA databases, we evaluated the RNA-seq data of CD8(+) T cell-related genes with gene co-expression network. Then, the CD8T-RGPI model was built by combining univariate Cox regression and the least absolute shrinkage and selection operator (LASSO) Cox regression analyses. The nomogram exhibited a good prediction ability for HCC patients' prognosis in either TCGA or ICGC databases. The CD8T-RGPI was elevated in patients with high T, M and TNM stages. Functional enrichment analyses revealed that in CD8T-RGPI high-risk group, tumour-related and immunosuppressive pathways, such as focal adhesion, aerobic glycolysis, HIF1 transcription factor pathway, IL-4 and IL-13 signalling were significantly enriched. Moreover, for CD8T-RGPI high-risk HCC patients, immune tolerance and immunosurveillance escape was associated with poor therapeutic outcomes for ICB therapy. The CD8T-RGPI low-risk HCC patients with elevated immune infiltration levels were associated with good immunotherapeutic responses to ICB. Our findings provide prospective prognostic biomarkers and elucidate on HCC immunotherapy.