In this multicenter retrospective analysis from Korea and Japan, we demonstrated that lenvatinib was associated with better median PFS and ORR compared with sorafenib as second-line therapy. These findings were consistent even after PSM. Although trends favored lenvatinib in terms of OS, the differences were not statistically significant. Our results suggest that lenvatinib may be a more favorable option than sorafenib for second-line treatment in patients previously treated with atezolizumab–bevacizumab.

Most clinical research tends to focus on first-line treatments, leaving subsequent therapy for patients who progress after immune checkpoint inhibitor (ICI) combination therapy less studied. Optimizing post-immunotherapy treatment is important, as subsequent therapies have been associated with improved survival outcomes after progression on immunotherapy (Lee et al. 2024; Persano et al. 2023). Given the lack of a globally established standard for second-line therapy in patients who progress on atezolizumab–bevacizumab, treatment decisions are largely influenced by physician preference, regulatory approval, and reimbursement policy. MKIs, such as lenvatinib and sorafenib, remain widely used in daily practice (Lee et al. 2024).

Our findings are consistent with those reported in previous retrospective studies(Chon et al. 2024; Lee et al. 2024; Persano et al. 2023; Yoo et al. 2021). In the current study, lenvatinib and sorafenib achieved median PFS of 4.9 months and 3.3 months, respectively, which aligns with previous studies reporting median PFS of 3.5–6.1 months for lenvatinib and 1.8–2.5 months for sorafenib (Chon et al. 2024; Lee et al. 2024; Yoo et al. 2021). Some previous studies, including ours, did not show a significant OS benefit with lenvatinib compared with sorafenib after progression on atezolizumab–bevacizumab (Chon et al. 2024; Persano et al. 2023; Yoo et al. 2021). However, recent large multinational studies demonstrated significantly better OS with lenvatinib compared with sorafenib as subsequent therapy after progression on atezolizumab–bevacizumab (Lee et al. 2024; Persano et al. 2024). This suggests that the lack of significant differences in our study may be attributed to the smaller sample size.

In addition to lenvatinib and sorafenib, other anti-VEGFR-MKIs, such as cabozantinib or regorafenib—both approved for use in patients previously treated with sorafenib have not been widely investigated in the setting of post-immunotherapy (Bang et al. 2022; Yoo et al. 2020). However, recent single-arm prospective phase 2 studies have shown that cabozantinib and regorafenib may be effective in patients who progress after immunotherapy. These studies reported median PFS and OS of 4.3 and 9.9 months, respectively, with cabozantinib, and 3.5 and 10.5 months, respectively, with regorafenib as second-line treatment after progression on atezolizumab–bevacizumab (Chan et al. 2024; Cheon et al. 2023). Although our data, along with others, suggest that lenvatinib may have a more favorable efficacy profile compared with sorafenib, there is insufficient evidence to directly compare lenvatinib with cabozantinib or regorafenib. Further investigations are required to explore these comparisons.

Both lenvatinib and sorafenib demonstrated manageable safety profiles as second-line treatments after atezolizumab–bevacizumab. Most adverse events were effectively managed through supportive care and appropriate dose modifications. No new safety signals were observed for either lenvatinib or sorafenib in this study.

This study has certain limitations. As a retrospective study, it may have been subject to unintentional biases. We employed various statistical methods, including PSM, to adjust for baseline characteristics and mitigate these biases. Studies based on retrospective assessment of information in medical records may underestimate the true rate of AEs. While the study was conducted across multiple international institutions, the relatively small sample size may limit the generalizability of our findings on a global scale.