Background: Hepatocellular carcinoma (HCC) is a major contributor to cancer-related mortality worldwide, with its progression significantly influenced by immune evasion mechanisms. This research aimed to identify key genes associated with immune evasion and assess their clinical significance in HCC.

Methods: Gene expression data from 342 HCC patients were obtained from The Cancer Genome Atlas (TCGA) repository. The Estimation of Stromal and Immune cells in Malignant Tumor tissues using Expression data (ESTIMATE) algorithm was employed to calculate immune and stromal scores. Differential gene expression analysis was conducted between high and low immune score groups, as well as between high and low stromal score groups. The resulting differentially expressed genes (DEGs) underwent functional enrichment analysis. A prognostic risk score model for HCC was constructed utilizing Least Absolute Shrinkage and Selection Operator (LASSO)-Cox, univariate, and multivariate Cox regression analyses. The prognostic performance of the model was evaluated through Kaplan-Meier survival analysis and receiver operating characteristic (ROC) curve analysis. Tumor and adjacent normal tissues from HCC patients were collected, and immune cell infiltration was quantified utilizing the Tumor Immune Estimation Resource (TIMER) database. Immunohistochemistry was performed to validate the expression of immune evasion-related proteins in clinical specimens.

Results: A total of 999 DEGs were identified from the immune and stromal score-based groups. These genes were predominantly enriched in immune-related pathways, including cytokine-cytokine receptor interactions, chemokine signaling, Wnt signaling, and the Hippo signaling pathway. Using univariate and multivariate Cox regression analysis, a prognostic model comprising six genes (Myocardin (MYOCD), Matrix Metallopeptidase 7 (MMP7), Matrix Metallopeptidase 16 (MMP16), PR/SET Domain 9 (PRDM9), Caudal Type Homeobox 2 (CDX2), and T Cell Receptor Beta Variable 10-3 (TRBV10-3)) was developed and demonstrated a significant association with overall survival (OS) in HCC patients (p < 0.05). The model yielded a concordance index (C-index) of 0.764, indicating robust prognostic performance. MYOCD was associated with a protective effect, showing an inverse correlation with the risk score (p < 0.001). Conversely, MMP7, PRDM9, CDX2, and MMP16 were risk-associated, with their expression levels positively correlating with the risk score (p < 0.001), whereas TRBV10-3 showed no significant difference between high- and low-risk groups (p > 0.05). Immunohistochemical validation confirmed the downregulation of MYOCD and upregulation of MMP7, MMP16, PRDM9, and CDX2 in HCC tissues (p < 0.001). Analysis of immune cell infiltration revealed a significant negative correlation between MYOCD expression and both stromal and ESTIMATE scores (p < 0.001), while MMP7 and MMP16 exhibited significant positive correlations with stromal, immune, and ESTIMATE scores (p < 0.001). PRDM9 and CDX2 also showed significant positive correlations with stromal scores (p < 0.05).

Conclusion: This study presents the first comprehensive identification of immune evasion-related genes (IERGs) within the tumor microenvironment (TME) of HCC and introduces a novel prognostic model based on immune evasion with strong clinical applicability. These findings offer valuable insights into the immunobiology of HCC and highlight potential therapeutic targets for precision immunotherapy.