Which is the best combination of TACE and Sorafenib for advanced hepatocellular carcinoma treatment? A systematic review and network meta-analysis

The aim of this study was to assess the comparative efficacy and safety of combination therapy with transarterial chemoembolization (TACE) and Sorafenib for patients with advanced hepatocellular carcinoma (HCC) through a systematic review and network meta-analysis and identify the best combination of TACE and Sorafenib. We searched databases for publications prior to May 2018. The prespecified efficacy outcomes were the objective response rate, overall survival rate, and time to progression. adverse effects included dermatologic, gastrointestinal, and general disorders. Subgroup analyses, meta-regression, and a network meta-analysis regarding two types of outcomes by different chemotherapy agents in TACE (5-fluorouracil, Adriamycin, Platinum, mitomycin C, hydroxycamptothecin) were included. The study is registered with PROSPERO (CRD42018098541). For efficacy outcomes, subgroups which included 5-fluorouracil and hydroxycamptothecin ranked higher than other chemotherapy agents, while mitomycin C ranked the lowest. For advanced effects, the use of mitomycin C or 5-fluorouracil as the chemotherapy agent ranked higher, while hydroxycamptothecin ranked the lowest. Therefore, we excluded 5-Fu and Mitomycin C in subsequent studies. Additionally, in the evaluation of primary adverse effects by the network meta-analysis, Platinum ranked the highest while hydroxycamptothecin ranked the lowest. Therefore, we excluded Platinum this time. Furthermore, all types of Adriamycin are not same, and some studies included two types of Adriamycin. The network meta-analysis results showed that the TACE (hydroxycamptothecin + pirarubicin) + Sorafenib arm and TACE (hydroxycamptothecin + epirubicin) + Sorafenib arm had significant efficacy differences. In conclusion, for patients with advanced HCC, combination therapy with HCPT plus THP/EPI in TACE and Sorfenib may be used as a first-line treatment.