Although there are only a few retrospective reports on ICI-based treatment after ATE/BEV failure, the most frequently studied regimen is a combination of anti-cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) inhibitor (ipilimumab) with anti-PD-1 monoclonal antibody (nivolumab). Data of 47 hepatocellular carcinoma (HCC) patients who were treated with nivolumab plus ipilimumab after ATE/BEV failure from our institute showed overall response rate (ORR) of 25.5% and disease control rate (DCR) of 42.6%, median progression-free survival (PFS) of 1.4 months (95% confidence interval, 1.1–1.7); and median overall survival (OS) was not reached [2]. Wong et al. [3] reported ORR of 16% and median OS of 10.9 months among 25 HCC patients treated with ipilimumab and nivolumab/pembrolizumab after progression on prior anti-PD-1/PD-L1 (prior ATE/BEV 14%), and three (12%) patients achieved complete response. In a study by Roessler et al. [4], among 10 patients who were treated with nivolumab plus ipilimumab after progression from ICI-based treatment (prior ATE/BEV 70%), median PFS and OS were 2.9 months and 7.4 months, respectively. Combination of CTLA-4 and PD-1 blockade works synergistically with longer-lasting immune effects. Additionally, CTLA-4/PD-1 blockade combination has been shown to increase the CD8+ T cell to myeloid-derived suppressor cells, and decrease the fraction of regulatory T cells which can reverse the tumor immune escape in patients with primary resistance to PD-1 inhibitors [3,5].

The combination of ICI plus TKI or antiangiogenic agents after ATE/BEV failure (ICI failure) has been also investigated. In a phase II study by Li et al. [6], tislelizumab (anti-PD-1 monoclonal antibody) plus regorafenib treatment achieved 28.6% ORR and 71.4% DCR, with a median PFS of 6.8 months among 28 pretreated patients (21.4% had prior PD1-inhibitor and TKI combination). Another combination of suvemcitug (VEGF-A inhibitor) plus envafolimab (PD-L1 inhibitor) for 18 HCC patients (40.0% had prior anti-PD-1/PD-L1 antibody) showed 11.1% ORR and 72.2% DCR, with median PFS of 4.3 months [7].

In fact, several studies have demonstrated that combination therapies involving ICIs and TKIs offer significant advantages over monotherapies. The combination of lenvatinib, a multi-kinase inhibitor, with pembrolizumab, an anti-PD-1 antibody, demonstrated promising efficacy in advanced HCC. The phase Ib KEYNOTE-524 study reported 36% ORR, which is significantly higher than the ORR observed with pembrolizumab alone in the KEYNOTE-224 study [8,9]. Similarly, the combination of cabozantinib, another multi-kinase inhibitor, with atezolizumab has been investigated. The phase III COSMIC-312 trial revealed a statistically significant improvement in PFS compared to sorafenib, although the OS benefit did not reach statistical significance [10]. Another promising combination involves the use of sintilimab (anti-PD-1) with the bevacizumab biosimilar IBI305, which showed significant benefits in overall survival and PFS in the ORIENT-32 trial conducted in China [11]. Furthermore, research into the combination of ICIs with other novel agents, such as fibroblast growth factor receptor 4 (FGFR4) and MET inhibitors, is ongoing. These inhibitors have shown potential for modulating the tumor microenvironment and enhancing immune responses. For example, targeting FGFR4 was found to decrease PD-L1 expression and reduce regulatory T cell infiltration, thereby potentiating the immune-mediated killing of tumor cells [12].

Overall, while these combination strategies represent significant advancements in the treatment landscape of HCC, their specific efficacy after ATE/BEV failure needs further detailed investigation. Continued research into these combinations and their underlying mechanisms will be crucial in optimizing therapeutic outcomes and expanding treatment options for advanced HCC after ATE/BEV failure.

Radiation therapy, converting tumor microenvironments into more immunogenic condition by releasing tumor antigen, effectively activates circulating cytotoxic T-cells, which can enhance immunogenic cell death through modulated immunity in HCC [13,14]. Radiotherapy can be considered for oligoprogression or newly developed vascular invasion during ICI therapy in advanced HCC patients. Sindhu et al. [15] analyzed 16 patients with HCC who received stereotactic body radiation therapy (SBRT) after oligoprogression during ICI therapy. PFS after SBRT was 7.1 months, with 1-year OS of 96.3%. There is an ongoing phase II study of sintilimab plus bevacizumab combined with radiotherapy as first-line treatment for HCC with malignant portal vein thrombi. ORR and DRR were 58.7% and 100%, and the median OS and PFS were 24.0 months and 13.8 months, respectively [16]. Theoretically, SBRT can also be extrapolated to cases with vascular invasion after progression with prior ICI-based therapy.

Because the relative dose intensity of lenvatinib has a significant impact on its effectiveness, authors from the Editorial concerned about “a sufficient dose would not be administered when using lenvatinib as a second-line treatment where many of Child-Pugh class B patients may be involved in the real world”. According to the study by Singal et al. [17], among patients with Child-Pugh class B who were treated with lenvatinib in a second- or higher-line setting showed ORR of 50.6% and median PFS and OS of 11.9 and 13.5 months, respectively. However, another study by Muto et al. [18] demonstrated the numerically lower clinical outcomes (OS, 6.7 months; PFS, 3.6 months) in Child-Pugh class B patients (n=4) treated with second-line lenvatinib. Our previous study showed lower ORR (7.5% in unmatched cohort and 5.6% in propensity score-matched cohort), OS (10.3 months) and PFS (3.5 months) than expected in the lenvatinib second-line cohort [19]. The OS (6.5 months) and PFS (2.2 months) were lower in Child-Pugh class B patients in our study, but the patient number (n=3) was too small to draw any conclusions regarding clinical outcomes. Considering that Child-Pugh class B patients receiving second or higher-line systemic therapy will be common in clinical practice, lenvatinib can be a good alternative if liver function and adverse events are well monitored and managed as there are few treatment options. Future studies with real-world data on Child-Pugh Class B patients are anticipated.

In conclusion, there are several options for second-line treatment after ATE/BEV failure, such as TKI, ICI-based treatment, locoregional therapy such as SBRT in case of oligoprogression or newly developed vessel invasion. Currently, selection is based on consideration of serious adverse events or contraindications from previous treatment. We will be able to predict and adjust customized treatment for each individual patient by accumulating real world data from multiple cases. Since liver function is the most important predictor of OS in second-line treatment or higher, it will be necessary to accumulate data on the outcomes of lenvatinib and other regimens especially in patients with Child-Pugh class B patients, preferably by dividing groups into Child-Pugh score 7 and 8.