Efficacy and safety of sorafenib in patients with advanced hepatocellular carcinoma according to baseline status: Subset analyses of the phase III Sorafenib Asia-Pacific trial

Background: The phase III Sorafenib Asia-Pacific (AP) trial-conducted in China, Taiwan and South Korea - confirmed that sorafenib improves overall survival (OS) and is safe for patients with advanced hepatocellular carcinoma (HCC). We performed a series of exploratory subset analyses to determine whether baseline status affected response to sorafenib. Methods: In the Sorafenib AP trial, 226 patients with well-preserved liver function (>95% Child-Pugh A) were randomised 2: 1 to sorafenib 400 mg bid or matching placebo. Subanalyses were based on aetiology (hepatitis B virus present/absent); tumour burden (macroscopic vascular invasion and/or extrahepatic spread present/absent); presence or absence of either lung or lymph node metastasis at baseline, Eastern Cooperative Oncology Group performance status (0, 1-2); serum concentrations of alanine aminotransferase/aspartate aminotransferase (normal, mildly elevated, moderately elevated), alpha-fetoprotein (normal/elevated) and total bilirubin (normal/elevated); and whether or not there was a history of hepatectomy or transarterial chemoembolisation/embolisation. Subgroup assessments included OS, time to progression (TTP), disease control rate and safety. Findings: Sorafenib consistently improved both median OS and median TTP, compared with placebo (range of hazard ratios (HR), 0.32-0.87 and 0.31-0.75, respectively). The most common grade 3/4 adverse events were hand-foot skin reaction, diarrhoea and fatigue, the incidence of which was similar between subgroups. Interpretation: The efficacy and safety profiles of sorafenib in the subpopulations described were comparable with those in the overall study population. These exploratory analyses suggest that sorafenib is effective for patients from the AP region with advanced HCC, irrespective of baseline status. (C) 2011 Elsevier Ltd. All rights reserved.