Stage-Specific Transcriptome Landscape of Hepatocellular Carcinoma: Insights From Super and Poor Survivors With Prognostic Signature Identification

Abstract

Objectives

Patient prognosis for hepatocellular carcinoma (HCC) varies significantly even when they share the same clinical stage. We aimed to characterize the stage-specific transcriptomic landscape in super survivors with HCC and develop a prognostic gene signature for the prediction of patient survival.

Methods

Data from The Cancer Genome Atlas (TCGA) among 76 age- and sex-matched super (alive at 5 years) and poor survivors (deceased within 1 year) with HCC were analyzed. Gene set enrichment analysis stratified by tumor stage was conducted, and a key prognostic gene signature was developed. The gene signature was then validated by using the whole TCGA cohort and the independent International Cancer Genome Consortium (ICGC) cohort.

Results

Stage-specific transcriptomic profiling revealed that stages I and II HCC cases showed positive enrichment in immune response pathways, while stage III tumor exhibited enhanced catabolic activities but reduced glycolysis. Across all tumor stages, cell cycle biological processes were less active in super survivors. A 19-gene signature, incorporating immune-, metabolism-, and cell cycle-related genes, accurately distinguished super survivors from poor survivors with 90.8% accuracy. The gene signature reliably predicted overall survival in both the verification cohort (area under the receiver operating characteristic curve [AUROC] for 1-, 3-, and 5-year survival: 0.82, 0.80, and 0.78) and the independent validation cohort (AUROC for 1- and 3-year survival: 0.80 and 0.83). Consistent AUROC was observed across tumor stages.

Conclusion

The 19-gene signature, considering the dynamic shift during HCC progression, may accurately predict survival outcomes in HCC patients as a potential tool for personalized prognosis.