Opioid signaling-related genes and their prognostic role in hepatocellular carcinoma: insights from bioinformatics and functional studies.

Hepatocellular carcinoma (HCC) is a highly lethal malignancy with limited effective treatment options. Opioid signaling related genes (including lncRNAs) (OSRGs) have been implicated in HCC progression, yet their prognostic significance and roles in tumor biology remain poorly understood. This study aimed to investigate the relationship between OSRGs and biological characteristics of HCC, providing insights for targeted therapies. Gene expression data from TCGA-LIHC were analyzed using DESeq2. Gene Set Enrichment Analysis (GSEA) of OSRGs was performed with cluster Profiler. Kaplan-Meier (KM) analysis assessed survival differences between high- and low-expression groups. Unsupervised clustering using Consensus ClusterPlus identified two clusters based on 18 OSRGs. A risk model was constructed using 5 OSRGs (G6PD, KIF20A, MSC-AS1, TFF3, ZFPM2-AS1) through LASSO and Cox regression and validated in training and test datasets. Immune cell infiltration and drug sensitivity were analyzed using CIBERSORT and oncoPredict, respectively. qRT-PCR was used to assess gene expression levels, while Transwell assays evaluated the functions of the 5 OSRGs in HCC cells. Additionally, an orthotopic model was employed to investigate the role of KIF20A. We observed that cluster 1 had an unfavorable prognosis. The risk model revealed significant tumor microenvironment and drug sensitivity differences between high- and low-risk groups. Silencing the 5 OSRGs inhibited HCC cell migration and invasion, with KIF20A suppression significantly reducing tumor growth in vivo. This study highlights the strong predictive ability of the risk model based on 5 OSRGs and the oncogenic role of these genes in HCC, potentially providing a foundation for prognosis and therapeutic target.