A Multi-Gene Signature Associated with 1-Year Survival in Patients with Stage I Liver Cancer: Integration of Preclinical and TCGA Data

Approximately 50% of individuals diagnosed with Stage I liver cancer live beyond four years; however, a small subset of Stage I patients die within the first year. A prognostic biomarker panel that can identify high-risk Stage I patients may be extremely valuable. In this study, we used the Long-Evans Cinnamon (LEC) rat model of Wilson's Disease and hepatocellular carcinoma (HCC), along with data from The Cancer Genome Atlas (TCGA) human database, to create a novel biomarker panel. We generated and analyzed a rat microarray gene expression profile by comparing liver tumor tissues with adjacent normal tissues from the same animals, covering approximately 30,000 genes. The microarray results were translated into a five-gene panel associated with 1-year survival in Stage I liver cancer patients based on TCGA data, in combination with machine learning and bioinformatics approaches. The panel was internally validated following the "REporting recommendations for Tumor MARKer prognostic studies (REMARK)" guidelines. With no existing Stage-I-specific prognostic tools, a biomarker panel associated with 1-year survival in patients with Stage I liver cancer is a potential candidate for rigorous external validation.