A multicenter observational study of lenvatinib for unresectable hepatocellular carcinoma in Japan.

e16118 Background: Lenvatinib is an oral tyrosine kinase inhibitor that has been available for treatment of unresectable hepatocellular carcinoma (uHCC) in Japan since March 2018. We conducted a multicenter prospective observational study to evaluate the safety and efficacy of lenvatinib in clinical practice. Methods: This study conducted as a prospective study, and enrolled patients with uHCC initially receiving lenvatinib as systemic therapy who gave informed consent from July 2018 through January 2019. The observation period per patient was 12 months and e-CRF was used. Results: 713 patients were registered at 133 clinical sites. The safety analysis set and efficacy analysis set were 703 patients. Patient baseline characteristics included median age:73.0 years (25 to 94), male/ female: 564/139, bodyweight: < 60kg/≥60kg: 323/380, HBV/HCV/alcohol/NAFLD or NASH: 137/287/167/88, Child-Pugh classification A/B or higher: 624/75, BCLC stage A/B/C/D: 57/291/332/11, Vp 0/1/2/3/4: 520/31/52/60/22, with/without extrahepatic metastasis: 236/461. Previous treatment (local and systemic therapy) for hepatocellular carcinoma is with/without TACE: 513/190, median number of TACE treatment: 3.0 times (1 to 26), with/without HAIC: 74/629, with/without TKI pretreatment: 131/572. The initial dose was 8 mg (80.5%) and 12 mg (68.2%) in patients with < 60 kg and ≥60 kg, respectively. Median relative dose intensity was 64.2% and the treatment continuation rate at 3 and 12 months after administration of lenvatinib were 67.3% and 24.8%. Regarding safety, the incidence of treatment-related adverse events (TRAEs) was 84.9% (grade 3 or higher: 42.5%). Common TRAEs (incidence rate ≥10%) were decreased appetite (23.9%), malaise (21.8%), hypertension (21.3%), proteinuria (18.3%), palmar-plantar erythrodysesthesia syndrome (16.5%), hypothyroidism (15.8%), and diarrhea (13.9%). TRAEs of grade 3 or higher （incidence rate ≥2.5%） were proteinuria (6.7%), hypertension (6.4%), malaise (3.8%), decreased appetite (3.6%), platelet count decreased (3.3%), palmar-plantar erythrodysesthesia syndrome (2.7%), diarrhea (2.7%) and hepatic encephalopathy (2.6%). The incidence of TRAEs leading to discontinuation of lenvatinib was 32.9％, and main TRAEs leading to discontinuation of lenvatinib were (incidence rate ≥2%) decreased appetite (6.3%), malaise (4.7%), proteinuria (3.3%) and hepatic encephalopathy (2.4%). Regarding efficacy, the ORR and DCR was 39.5% (95% CI: 35.1, 43.9) and 78.9% (75.1, 82.5) in 494 patients who were evaluated for the best response by mRECIST. The median OS was 498.0 days (95% CI: 439.0, -) in 703 patients. Conclusions: The incidence of TRAEs leading to discontinuation of lenvatinib tended to be higher than in phase 3 REFLECT trials (19.7%), but the safety profile and effectiveness findings for lenvatinib in this study were consistent with previous reports.