Disease Control Biomarker to Predict Sorafenib Susceptibility in Advanced Stage Hepatocellular Carcinoma

Abstract

Background: Hepatocellular carcinoma (HCC) has two first-line treatments, those are sorafenib and lenvatinb. But because of predictive biomarker's absence, these two agents have poor response rates and overall survival period. Disease control biomarker could be a method that potentially improves the effectiveness of sorafenib, one of two agents. Here, we aimed to develop a clinical useful biomarker that predicts disease control of sorafenib. Methods: Using nanostring nCounter, we analyzed expression levels of 770 genes in 73 advanced-stage HCC patients with sorafenib trearment. With the 770 genes expression levels of 73 patients, we identified differentially expressed genes (DEGs) and computed combination of weighted gene expression for disease control biomarker. To validate gene signature, we analyzed cross validation and meta-analysis. For predicted poor responders, we listed up recommended medicine, analyzing individual DEGs by meta-analysis. Results: 8-gene signature showed 0.90 of area under the curves (AUC), 91.78% of accuracy. In cross validation, 8-gene signature showed well-performance with 83.67% of cross validation accuracy. Also, when classification with 8-gene signature, median overall survival (median OS) was improved 27.3 months from 11.3months. In promising alternative agents for predicted poor responders, recommended agents were listed up individually based on individual gene expression. Conclusions: 8-gene signature provides a best compromise between sorafenib effectiveness and coverage of sorafenib treatment patients. In perspective of precision medicine, our process of precision medicine recommendation can be drive the precision medicine one step forward.