Hepatocellular carcinoma (HCC) ranks the 6^th place among cancers with regard to its morbidity and the 3^rd place among factors inducing cancer-associated mortality worldwide in 2022, and the harm is even greater in China (Bray et al., 2024). The main reason is that the diagnosis in over 70% cases is made at the advanced stage in China, when radical treatment is unfeasible, resulting in extremely poor prognosis (Younossi et al., 2023). In the past decades, lots of novel therapy methods have been developed for cancer treatment, including nanotechnology, targeted therapy, immunotherapy and so on (Zhu et al., 2024; Liu et al., 2023; Dai et al., 2024). As recommended by the Barcelona Clinic Liver Cancer (BCLC) classification system, systemic treatment with immunotherapy and targeted therapy can be applied in advanced HCC, and the life expectancy has obviously improved (Reig et al., 2022).

Sorafenib has been recommended to be a first-line treatment for advanced HCC from 2008, which is effective in the SHARP and Oriental clinical trial (Llovet et al., 2008; Cheng et al., 2009). Until 2017, lenvatinib shows comparable overall survival (OS) and significantly improved progression-free survival (PFS), objective response, time to progression (TTP), as well as postponed life quality decline to sorafenib among untreated, non-resectable HCC cases (Kudo et al., 2018). Based on the result, lenvatinib has become a novel preferred option for advanced HCC. Besides, multiple programmed cell death protein 1 (PD-1) inhibitors gain approval for HCC, including first-line or second-line therapy options. Among them, tislelizumab is promising in treating HCC, and exhibited durable antitumor effect and favorable OS among advanced HCC cases from the front-line cohort in RATIONALE 301 study, comparing to sorafenib (Qin et al., 2023).

Next, with the positive result of the IMbrave150 study, targeted therapy combined with immunotherapy is becoming a preferred option for advanced hepatocellular carcinoma (Finn et al., 2020a). Except for bevacizumab, tyrosine kinase inhibitors (TKIs) combined with immunotherapy have also undergone extensive clinical trials and practice. Lenvatinib combined with pembrolizumab as the first-line therapy displays good antitumor effect in the KEYNOTE-524 phase 1b study, the median OS was 22.0 months, and the median PFS was 8.6 months, accompanied by controllable safety (Finn et al., 2020b). Unfortunately, though LEAP-002 phase III study demonstrated that lenvatinib combined with pembrolizumab could significantly improve prognosis, it could not reach specific significance in terms of prolonged OS and PFS compared with lenvatinib plus placebo, indicating that the therapeutic effect of combination therapy needs to be further enhanced (Llovet et al., 2023).

FOLFOX-based hepatic artery infusion chemotherapy (HAIC) can effectively treat advanced HCC, which can substantially increase OS in comparison with sorafenib (Lyu et al., 2022). HAIC can transfer sustained high drug concentration to the tumors, and induce the great local anticancer efficacy, resulting in effectively shrinking intrahepatic lesions, especially for those with high tumor burden (Chen et al., 2023). In addition, it has reported that chemotherapy can exert synergistic anti-tumor effects with immunotherapy and targeted therapy in various types of cancer (Ye et al., 2023; Tubridy et al., 2024), and many studies have also demonstrated that HAIC combined with PD-1 inhibitors and TKIs can improve survival outcomes for advanced HCC (Zhang et al., 2023; Lin et al., 2023). However, it remains unclear whether the triple combination therapy is safe and effective on advanced HCC with high tumor burden.

Therefore, the present work focused on investigating whether HAIC combined with lenvatinib and tislelizumab versus lenvatinib plus tislelizumab was effective and safe in the treatment of advanced HCC with high tumor burden.

The present multicenter clinical study is the first to examine whether HAIC plus lenvatinib and tislelizumab was as effective and safe as lenvatinib plus tislelizumab for advanced HCC with high tumor burden, and the triple combination therapy significantly improved survival and tumor response compared with the dual combination therapy. Besides, PSM was conducted for eliminating group heterogeneities, a relatively large sample size was included among diverse centers, and long-time follow-ups were performed. At last, no matter in PSM cohort or the whole cohort, HTP group showed significantly prolonged OS and PFS comparing to TP group. Noteworthily, triple combination therapy induced the increased AEs rate, while efficient measures were adopted for mitigating AEs.

According to current guidelines, locoregional therapy, mainly referring to TACE, has been recommended to be the first-line therapy for immediate-stage HCC, while systemic treatment remains the preferred option in advanced stage HCC (Singal et al., 2023; Vogel and Martinelli, 2021). Although the dual immunotherapies atezolizumab-bevacizumab and durvalumab-tremelimumab have been shown to result in better OS and PFS than sorafenib, lenvatinib remained the first-line choice for PVTT patients in China, giving the rather high cost of the dual immunotherapies and the prevalent risk of gastric bleeding due to cirrhosis. However, in the phase 3 randomized controlled study of LEAP 002, lenvatinib plus pembrolizumab failed to meet the pre-specified statistical significance on OS, even though the combination therapy achieved the longest median OS time ever reported in first-line HCC studies. Hence, it is worth noting that locoregional treatment and systemic treatment should complement each other rather than be mutually exclusive, especially for those with high-risk factors, including high tumor burden or extrahepatic metastasis. Locoregional therapy can reduce the tumor burden in a relatively short time, while systemic therapy can effectively control the intrahepatic and extrahepatic lesions in a relatively longer period, resulting in an enhanced synergistic anti-tumor effect (Cappuyns et al., 2024; Brown et al., 2023). As reported, an increasing number of clinical trials have demonstrated that locoregional therapy, including TACE or HAIC, combined with TKIs and/or PD-1 inhibitors possessed favorable outcomes in advanced HCC patients, comparing to monotherapy or dual combination therapy (Kulik and El-Serag, 2019). As reported in a phase II clinical study (TRIPLET), HAIC combined with camrelizumab and apatinib for HCC in BCLC C stage demonstrated encouraging results and manageable safety. Over 50% of the enrolled patients had high risk factors, including Vp3/4 or extrahepatic metastasis. The median PFS was 10.38c months, and the ORR and DCR were 77.1% and 97.1%, respectively (Chen et al., 2023), consistent with our study.

Some rationales were proposed for combining HAIC rather than TACE as the locoregional treatment in the study. First, previous studies have demonstrated that comparing with TACE, FOLFOX-HAIC results in favorable tumor response and superior outcomes for advanced HCC due to the direct delivery high-dose chemotherapeutic drugs into tumor via hepatic artery (Li et al., 2021). As a result, Chinese guidelines have recommended HAIC rather TACE as one of first-line options for advanced HCC, particularly for patients with high tumor burden (Xie et al., 2023). Second, TACE can induce hypoxia in the tumor microenvironment, which may activate hypoxia-inducible factors (HIFs) and promote tumor survival, angiogenesis, and subsequent recurrence (Gai et al., 2020). Third, many embolization particles are required for embolizing large-sized HCC, probably leading to an increased risk of hepatic functional reserve deterioration, nontarget embolization, and postembolization syndrome (Roehlen et al., 2023). Finally, total embolization is not easy for large-sized HCC due to extrahepatic collateral arteries (Li et al., 2022).

We also carried out subgroup analyses on OS and PFS according to different factors. HTP group showed the clinical benefits for almost all the subgroups. Nonetheless, broad 95% CI ranges were observed in females, patients of Child-Pugh class B, and those without HBV infection, probably associated with the small sample size in each subgroup.

When analyzing various endpoints to PFS, HTP group had obviously lower proportion of intrahepatic metastasis and local lesion progression comparing to TP group. These findings suggested that HAIC, as a locoregional approach, could play an important role in controlling intrahepatic and local recurrence, and the triple combination therapy could exert the synergistic anticancer efficacy. Expectedly, intrahepatic and local recurrence mainly restricted the survival benefits of systemic therapy. Therefore, the combination with HAIC probably further improved prognosis (Yuan et al., 2023).

The ration of subsequent therapy was higher in HTP group (82.1% versus 70.5%), suggesting that treatment following progression might be conducted at a higher frequency among patients receiving HAIC plus lenvatinib and tislelizumab. This may be explained from two aspects. First, triple combination therapy provided superior efficacy, improving the compliance of patients. Second, triple combination therapy significantly prolonged survival, providing more opportunities for subsequent treatments.

In addition to favorable outcomes, HAIC combined with lenvatinib and tislelizumab increased AEs rate to some extent. The higher HAIC-related AEs incidence took place, consistent with those in previous trials of HAIC. However, those AEs could be controllable using corresponding supporting medications, without aggravating the condition or treatment discontinuation. HTP group had increased AEs incidence relative to locoregional therapy or systemic therapy in the previous studies (Kudo et al., 2018; Qin et al., 2023; Llovet et al., 2023; Lyu et al., 2018a), the reasons of which may be that the enrolled patients in the study were with worse baseline levels, and the systemic therapy aggravated the toxicity of HAIC. Abdominal pain resulting from arteria vasospasm in oxaliplatin infusion was also a common HAIC-specific AE. So far, abdominal pain cannot be effectively avoided, with the exception of the prescription of pain/spasm relieving agents or slowing down oxaliplatin infusion (Wu et al., 2021; Proietti et al., 2007). Collectively, HAIC combined with lenvatinib and tislelizumab showed acceptable safety and tolerability.

Certain limitations should be noted in the present work. First, due to the retrospective nature, selection bias could not be avoided. Though PSM was conducted for minimizing between-group heterogeneities, there were still endogenous differences due to unmeasured confounders, population differences, temporal bias, exclusion of non-matched cases, and information bias. Consequently, more prospective randomized controlled studies should be conducted for validating our results. Second, since the study primarily included HBV-positive individuals, it is important to note that HCC in patients with HBV infection can differ biologically and clinically from HCC in individuals with other etiologies, such as hepatitis C virus (HCV) or non-alcoholic fatty liver disease (NAFLD). The treatment responses, prognosis, and underlying mechanisms of HCC may vary across these different patient populations. It should acknowledge that the findings may not be fully applicable to non-HBV patients, especially in regions where HCV or NAFLD-related HCC is more common. Third, the patient cohort being sourced mainly from Chinese medical centers is another important limitation. The genetic, environmental, and healthcare-related factors in China may differ from those in other countries or regions, such as Europe or North America. These differences could impact treatment response and outcomes. It is important to state that the study’s results may not be directly applicable to populations in other parts of the world, particularly where the prevalence of HBV and other risk factors for HCC may differ significantly. The study should call for further validation in more diverse cohorts from different geographic regions to confirm the external validity of the findings. Finally, we just enrolled cases undergoing tislelizumab treatment, and further exploration is warranted for determining if the same results are attained using other PD-1 inhibitors.

In conclusion, comparing to lenvatinib plus tislelizumab, HAIC plus lenvatinib and tislelizumab achieves markedly improved OS, PFS, and ORR for advanced HCC with high tumor burden, and this combination treatment is safe. Before prospective results are revealed, our results support applying the triple combination treatment for advanced HCC with high tumor burden.