Background/aims: In both SHARP and Asia-Pacific Study, sorafenib was proved to improve the overall survival of the patients with hepatocellular carcinoma. However, factors contributing to the improvement of overall survival of the patients treated by sorafenib have not been fully evaluated. In this study, patient-derived, background liver disease-derived and tumor-derived factors before treatment were evaluated whether they have contributed to the improvement of the overall survival.
Methodology: Forty-seven cases with HCC treated by sorafenib between Sept 2009 and Feb 2011 were included in this analysis. The survival of these cases was analyzed by Kaplan-Meier Method. Factors used for univariate analysis were two patient-derived parameters, two background liver disease-derived, five tumor-derived. Factors related to the over-all survival were analyzed by multivariate analysis using Cox regression model.
Results: In the multivariate analysis, only background liver disease-derived parameter Child-Pugh class A vs. B, (p=0.007, HR=0.21 (0.07-0.65)) was significant. No other parameters including tumor-derived factors were statistically significant by multivariate analysis.
Conclusions: We undertook the statistical analysis on the three categories. Surprisingly, no tumor derived parameter contributed to the overall survival. Background liver disease-derived parameter rather than tumor-derived parameter was found to define the prognosis of patients with advanced HCC treated by sorafenib.