A Mechanism-Derived Four-Gene Prognostic Signature for Hepatocellular Carcinoma Based on the Multi-Target Anti-Tumor Activity of Icaritin

Background: To systematically dissect the multi-target molecular mechanisms of icaritin against hepatocellular carcinoma (HCC), and construct a mechanism-based gene prognostic signature for HCC risk stratification.

Methods: Transcriptomic profiles and clinical data from the TCGA-LIHC cohort, icaritin potential targets, and HCC pathogenic genes were integrated for Venn intersection screening. Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment and protein-protein interaction (PPI) network analyses were performed. Univariate Cox regression followed by LASSO penalized Cox regression was applied to eliminate multicollinearity and screen stable prognostic genes. The signature’s predictive performance was validated via Kaplan-Meier survival curves, time-dependent receiver operating characteristic (ROC) curves, and multivariate Cox regression in TCGA and independent GEO external cohorts.

Results: A total of 37 intersecting genes linking icaritin to HCC progression were identified, enriched in cell cycle regulation, oxidative stress metabolism and p53 signaling pathways. After LASSO regularization to resolve severe multicollinearity between PLK1 and NEK2, a four-gene signature (CYP19A1, NQO1, MMP3, PLK1) was established. Patients were stratified into high- and low-risk subgroups using a median risk score cutoff of 3.02. In the TCGA cohort, high-risk patients exhibited significantly worse overall survival (OS, P<0.001). Time-dependent ROC analysis revealed 1-, 3-, and 5-year AUC values of 0.827, 0.709, and 0.662, respectively, indicating robust short-term predictive efficacy and limited long-term discrimination capacity. The four-gene risk score remained an independent prognostic factor after adjusting for conventional clinical covariates (HR=2.450, 95%CI:1.569–3.830, P<0.001). Reliable survival stratification was further validated in the external GSE27150 cohort (1-year AUC=0.786, 3-year AUC=0.758, log-rank P=0.010).

Conclusions: This study systematically elucidates the synergistic multi-pathway anti-HCC mechanisms of icaritin and constructs a statistically robust four-gene prognostic signature. The signature provides a translational tool for individualized HCC risk evaluation and offers candidate biomarkers for subsequent preclinical validation of icaritin therapy.