Uncovering gene expression signatures and diagnostic – Biomarkers in hepatocellular carcinoma through multinomial logistic regression analysis

Hepatocellular carcinoma (HCC) is one of the leading causes of cancer death worldwide, and classifying the developmental stages of HCC can help with early prognosis and treatment. This study aimed to investigate diagnostic and prognostic molecular signatures underlying the progression of HCC, including tumor initiation and growth, and to classify its developmental stages based on gene expression levels. We integrated data from two cancer systems, including 78 patients with Edmondson-Steiner (ES) grade and 417 patients with TNM stage cancer. Functional profiling was performed using identified signatures. Using a multinomial logistic regression model (MLR), we classified controls, early-stage HCC, and advanced-stage HCC. The model was validated in three independent cohorts comprising 45 patients (neoplastic stage), 394 patients (ES grade), and 466 patients (TNM stage). Multivariate Cox regression was employed for HCC prognosis prediction. We identified 35 genes with gradual upregulation or downregulation in both ES grade and TNM stage patients during HCC progression. These genes are involved in cell division, chromosome segregation, and mitotic cytokinesis, promoting tumor cell proliferation through the mitotic cell cycle. The MLR model accurately differentiated controls, early-stage HCC, and advanced-stage HCC across multiple cancer systems, which was further validated in various independent cohorts. Survival analysis revealed a subset of five genes from TNM stage (HR: 3.27, p < 0.0001) and three genes from ES grade (HR: 7.56, p < 0.0001) that showed significant association with HCC prognosis. The identified molecular signature not only initiates tumorigenesis but also promotes HCC development. It has the potential to improve clinical diagnosis, prognosis, and therapeutic interventions for HCC. This study enhances our understanding of HCC progression and provides valuable insights for precision medicine approaches.