Comparative effectiveness of traditional chemoembolization with or without sorafenib for hepatocellular carcinoma

To compare the overall survival (OS) and progression-free survival (PFS) with associated adverse events (AE) in patients with unresectable hepatocellular carcinoma (HCC) treated with transarterial chemoembolization (TACE) + sorafenib vs TACE alone.

In this retrospective cohort study we collected data on all consecutive patients with a diagnosis of unresectable HCC between 2007 and 2011 who had been treated with TACE + sorafenib or TACE alone. We hypothesized that the combination therapy is superior to TACE alone in improving the survival in these patients. Data extracted included patient's demographics, etiology of liver disease, histology of HCC, stage of liver disease with respect to model of end stage liver disease score and Child-Turcotte-Pugh (CTP) classification and Barcelona Clinic Liver Cancer (BCLC) staging for HCC. Computed tomography scan findings, alpha fetoprotein levels, number of treatments and related AE were also recorded and analyzed.

Of the 43 patients who met inclusion criteria, 13 were treated with TACE + sorafenib and 30 with TACE alone. There was no significant difference in median survival: 20.6 mo (95%CI: 13.4-38.4) for the TACE + sorafenib and 18.3 mo (95%CI: 11.8-32.9) for the TACE alone (P = 0.72). There were also no statistically significant differences between groups in OS (HR = 0.82, 95%CI: 0.38-1.77; P = 0.61), PFS (HR = 0.93, 95%CI: 0.45-1.89; P = 0.83), and treatment-related toxicities (P = 0.554). CTP classification and BCLC staging for HCC were statistically significant (P = 0.001, P = 0.04 respectively) in predicting the survival in patients with HCC. The common AE observed were abdominal pain, nausea, vomiting and mild elevation of liver enzymes.

Combination therapy with TACE + sorafenib is safe and equally effective as TACE alone in patients with unresectable HCC. CTP classification and BCLC staging were the significant predictors of survival. Future trials with large number of patients are needed to further validate this observation.