A multicenter cohort study on transarterial chemoembolization with or without sorafenib for intermediatestage hepatocellular carcinoma: Reconsidering combinationtherapy trial design

Background: The proof of the superiority of combination therapy with sorafenib and transarterial chemoembolization (TACE) over TACE alone for hepatocellular carcinoma (HCC) in term of survival is lacking. We conducted this multicenter retrospective study to evaluate the efficacy of combination therapy over TACE alone, and to compare the overall survival (OS) between patients with grade 2 sorafenib-related dermatologic adverse events (AEs) in the combination therapy group and patients treated with TACE alone. Methods: From January 2009 to December 2012, 606 consecutive patients with intermediate stage HCC, Eastern Cooperative Oncology Group performance status 01, and ChildPugh class AB ( 7) were included. Of them 202 received combination therapy and 404 received TACE alone therapy, respectively. Results: There was no significant difference between the two groups in median OS although a trend toward longer survival was observed in the combination therapy group (22.3 vs. 18.1 months, P= 0.281). After propensity score matching the difference in OS was still not different (22.3 vs. 17.9 months, P= 0.343). Of note, in the combination therapy group, 119 patients with grade 1 dermatologic AEs within the first two months of sorafenib initiation, which were defined as non-responders, had increased risk of death compared with 83 patients with grade 2 dermatologic AEs which were defined as responders (HR = 1.85; 95%CI 1.27-2.68; P= 0.001). By using the second propensity score matching to balance the baseline differences between responders subgroup and TACE alone group, a significantly prolonged median OS was observed in the responders subgroup (27.9 vs.18.3 months, P= 0.046). Conclusions: Combination therapy, not in all, but in responders to sorafenib, results in longer overall survival than TACE alone. Sorafenib-related dermatologic AEs may be considered a possible clinical marker to stratify responders from all patients. Before the appearance of any assured biomarkers, the design of prospective comparative studies needs to focus on the responders to treatment who are evaluated by clinical markers.