Prognostic factors and predictors of sorafenib benefit in patients with hepatocellular carcinoma: Analysis of two phase III studies

To identify potential prognostic factors for OS, univariate and multivariate (MV) analyses were performed for baseline variables by Cox proportional hazards model. Hazard ratios (HRs) and median OS were evaluated across pooled subgroups. To assess factors predictive of sorafenib benefit, the interaction term between treatment for each subgroup was evaluated by Cox proportional hazard model.

In 827 patients (448 sorafenib; 379 placebo) analyzed, strong prognostic factors for poorer OS identified from MV analysis in both treatment arms were presence of macroscopic vascular invasion (MVI), high alpha-fetoprotein (AFP), and high neutrophil-to-lymphocyte ratio (NLR; ⩽ vs. >median [3.1]). Sorafenib OS benefit was consistently observed across all subgroups. Significantly greater OS sorafenib benefit vs. placebo was observed in patients without extrahepatic spread (EHS; HR, 0.55 vs. 0.84), with hepatitis C virus (HCV) (HR, 0.47 vs. 0.81), and a low NLR (HR, 0.59 vs. 0.84).