Abstract
AIMS: The aim of this study is to develop a prognostic model for hepatocellular carcinoma (HCC) using stemness-related genes (SRGs), while also pinpointing and validating pivotal genes associated with this process. METHODS: Utilizing the TCGA and ICGC database, a prognostic stemness-related scores (SRS) for HCC through a combination of WGCNA and machine learning. Bioinformatics analysis evaluated tumor immune infiltration characteristics and drug sensitivity in different SRS subgroups, identifying the key gene TOMM40L. qRT-PCR and IHC were employed to detect the expression level of TOMM40 L. Kaplan-Meier survival analysis assessed the prognostic value of TOMM40L in HCC. In vitro cell experiments explored the influence of TOMM40L on HCC cell progression and stemness. RESULTS: The prognostic model SRS for HCC was developed and validated, incorporating four SRGs: EIF2B4, CDCA8, TCOF1, and TOMM40L. Distinct variations in tumor immune infiltration profiles and drug sensitivity were noted across different SRS subgroups. Elevated TOMM40L levels are notably detected in malignant tissues in contrast to adjacent tissues, with heightened TOMM40L expression correlating with unfavorable prognostic outcomes. In addition, knockdown of TOMM40L significantly inhibited cell progression and stemness. Conclusion: The newly constructed SRS model is a potential biomarker for assessing HCC prognosis, and the key gene TOMM40L exhibits oncogenic properties.