Abstract
BACKGROUND: T cells are essential mediators of antitumor immune responses. While T cell proliferation-related genes (TRGs) have been identified, how they relate to prognostic outcomes in hepatocellular carcinoma (HCC) patients remains uncertain. This study attempts to analyze the relationship between TRGs and the prognosis of patients with HCC from multiple perspectives. METHODS: The expression of 33 TRGs and associated clinical data from HCC patients were obtained from several datasets and used to conduct bioinformatics analyses. T cell proliferation-related molecular subtypes were established via consensus clustering, while a least absolute shrinkage and selection operator (LASSO)-Cox regression approach was used to develop a predictive gene signature that was validated in an external cohort. These results were also used to guide the design of a nomogram with high levels of accuracy to improve the clinical value of TRG scores. In vitro analyses were also used to establish and validate candidate biomarkers. RESULTS: Patients with HCC were divided into two TRG-based clusters and differentially expressed genes (DEGs) between the clusters were identified. Three gene clusters were separated according to the DEGs and the associations among the two TRG subtypes, the three gene clusters, patient survival, and immune cells were investigated. Multiple datasets were used to successfully validate the utility of this risk score to gauge survival outcomes in patients. Key TRGs identified through these analyses included SOX12, LPCAT, MMP1, CLEC3B, and KRT17. HCC tumor tissue samples exhibited high levels of keratin 17 (KRT17) expression, and the knockdown of this gene hampered HCC cell proliferation and invasivity while promoting the upregulation of apoptosis-related markers and suppressing Akt/mTOR signaling activity. CONCLUSIONS: The use of T cell proliferation-based molecular subtyping and related predictive signatures can enable the effective prognostic assessment of HCC patients. KRT17 may also hold promise as a target for therapeutic interventions in the cancer type.