Abstract
BACKGROUND & AIMS: A plethora of second-line therapies have been recently introduced for hepatocellular carcinoma (HCC) treatment with promising results. A meta-analysis of second-line treatments for HCC has been performed to better tailor their use based on improved patient stratification and to identify the best available option. METHODS: Pubmed, Scopus, Web of Science, and ClinicalTrials.gov were searched for randomized controlled trials evaluating second-line treatment for advanced HCC in patients already treated with sorafenib. The primary outcome was overall survival (OS). Secondary outcomes were progression-free survival (PFS) and drug withdrawal due to adverse events. Network meta-analyses were performed considering placebo as the basis for comparison in efficacy and safety analyses. Subgroup stratification considered gender, age, sorafenib-responsiveness and drug tolerability, viral infection, macrovascular invasion, HCC extrahepatic spread, performance status, and alpha-fetoprotein levels. RESULTS: Fourteen phase II or III randomized controlled trials, involving 5,488 patients and 12 regimens, were included in the analysis. Regorafenib (hazard ratio (HR) = 0.63, 95% confidence interval (CI) = 0.50-0.79), cabozantinib (HR = 0.76, 95% CI = 0.63-0.92), and ramucirumab (HR = 0.82, 95% CI = 0.70-0.76) significantly prolonged OS compared with placebo. Cabozantinib (HR = 0.44, 95% CI = 0.36-0.52), regorafenib (HR = 0.46, 95% CI = 0.37-0.56), ramucirumab (HR = 0.54, 95% CI = 0.43-0.68), brivanib (HR = 0.56, 95% CI = 0.42-0.76), S-1 (HR = 0.60, 95% CI = 0.46-0.77), axitinib (HR = 0.62, 95% CI = 0.44-0.87), and pembrolizumab (HR = 0.72, 95% CI = 0.57-0.90) significantly improved PFS compared with placebo. None of the compared drugs deemed undoubtedly superior after having performed a patients' stratification. CONCLUSIONS: The results of this network meta-analysis suggest the use of regorafenib and cabozantinib as second-line treatments in HCC.