Real-world outcomes of lenvatinib versus sorafenib as the second-line therapy after atezolizumab plus bevacizumab in patients with hepatocellular carcinoma: a large-scale multicenter study in Korea

INTRODUCTION: After first‑line atezolizumab plus bevacizumab (ATE + BEV) in hepatocellular carcinoma (HCC), optimal second‑line (2L) therapy remains unclear. This study compared the real‑world effectiveness and hepatotoxicity of lenvatinib versus sorafenib as 2L treatment after ATE + BEV. METHODS: This retrospective, multicenter study used electronic medical record data from the Liver Cancer IN Korea (LINK) database. Patients who received 2L lenvatinib or sorafenib after ATE + BEV were included. The primary endpoint was overall survival (OS). Secondary endpoints included time to treatment discontinuation (TTD), time to next treatment (TTNT), hepatotoxicity, and tumor marker dynamics. Analyses were conducted in the overall cohort, with additional propensity score (PS)-matched analyses performed. RESULTS: A total of 295 patients were included in the analysis (2L lenvatinib, n = 132; 2L sorafenib, n = 163). Median OS was significantly longer with lenvatinib (not reached) than with sorafenib (6.67 months [95% CI 5.32-17.31]; p = 0.001). Lenvatinib also showed significantly longer TTD (5.49 months [3.75-6.04] vs. 1.84 months [1.68-2.20]; p < 0.001) and TTNT (5.72 months [4.07-7.85] vs. 3.06 months [2.56-3.48]; p = 0.001). These findings were generally consistent after PS-matching. Incidence rates of laboratory-based hepatotoxicity were lower with lenvatinib. From 2L initiation to 3 months after 2L therapy, both AFP and PIVKA-II generally declined in both groups. CONCLUSION: These findings suggest that lenvatinib may be a promising 2L option following ATE + BEV in HCC, as evidenced by improved OS, longer treatment durability, and lower rates of laboratory-based hepatotoxicity compared with sorafenib. However, given the retrospective observational design, further prospective studies are warranted to confirm these findings.