Primary liver cancer ranks as the sixth most common malignancy globally and the third leading cause of cancer-related mortality. HCC constitutes the predominant histopathological subtype, accounting for 75%–85% of cases (1). Early-stage HCC may be managed through surgical resection, ablation therapies, or liver transplantation. However, due to the insidious onset of HCC and stringent eligibility criteria for curative interventions, only 30%–40% of patients ultimately receive definitive treatment at early disease stages (2, 3). In clinical practice, a proportion of patients with intermediate-stage HCC (BCLC-B stage) face unresectable hepatocellular carcinoma (uHCC) due to multifactorial constraints, thereby falling into the same therapeutic category as those diagnosed with advanced-stage disease (BCLC-C stage). For this patient population where curative interventions are not feasible, systemic therapy emerges as an evidence-based treatment strategy (4).

TACE and HAIC remain established locoregional options for uHCC (5). Recent evidence from the LEAP-012 trial indicates that TACE combined with systemic therapy yields improved clinical outcomes (6). In patients with uHCC refractory to multiple TACE sessions, hypoxia induced by repetitive embolization triggers vascular endothelial growth factor (VEGF) upregulation, potentially driving compensatory neovascularization and diminishing TACE efficacy (7, 8). Concurrently, TACE induces immunomodulatory effects within tumor lesions, which may synergistically potentiate the therapeutic efficacy of ICIs. TKIs exhibit dual antiangiogenic activity and immunomodulatory mechanisms that synergize with ICIs (9, 10). For uHCC patients with high tumor burden, conventional TACE often fails to achieve complete embolization due to heterogeneous tumor vasculature. HAIC utilizing oxaliplatin, calcium folinate, and 5-fluorouracil (FOLFOX regimen) delivered via sustained infusion into tumor-feeding arteries prolongs drug exposure, offering an alternative therapeutic strategy (11). Furthermore, HAIC has demonstrated significant survival benefits in HCC patients with PVTT (12, 13). The TRIPLET study further validated the clinical acceptability of HAIC combined with systemic therapy in advanced-stage HCC (14). This retrospective investigation aims to evaluate the efficacy and safety of combined TACE+HAIC with ICI and TKI therapy in BCLC stage B-C HCC patients.

This retrospective study demonstrated that the T+H+T+I regimen conferred improved OS and PFS compared with the H+T+I regimen. Higher survival rates were observed in subgroups with a tumor diameter ≥5 cm, PVTT, or VP-4 thrombus classification, despite comparable tumor response rates between the two regimens.

The strengths of this study are threefold. This study compares the efficacy between combination therapeutic regimens, a less explored area in current clinical research. While existing studies predominantly focus on evaluating combination therapies (TACE+TKI+ICI or HAIC+TKI+ICI) against monotherapies (e.g., TACE alone), limited evidence exists regarding head-to-head comparisons of distinct combination strategies. Second, the inclusion of a real-world, heterogeneous uHCC population—encompassing multifocal tumors, high tumor burden, main portal trunk/first-branch tumor thrombosis, and extrahepatic metastasis—ensures clinical relevance and informs personalized therapeutic strategies for advanced HCC. Third, methodologically, we implemented refined 1:1 PSM across demographic characteristics, tumor profiles, baseline hepatic/renal function, and nutritional/coagulation parameters, systematically mitigating confounding biases and enhancing result reliability.

For the results obtained in this study, we will conduct the following analysis: the current real-world study CHANCE-2201 suggests that TACE combined with systemic therapy is superior to TACE treatment alone (15); The median OS reached 22.6 months. Meanwhile, in the subgroup analysis of OS in CHANCE-2201, combination therapy was found to be superior to TACE alone in subgroups with HBV infection, large vessel invasion, and no extrahepatic metastasis. The forest plot of subgroup analysis in this study suggests that only in some subgroups, the survival benefit of the T+H+T+I group is better than that of the H+T+I group, especially in subgroups with tumor diameter ≥ 5CM, presence of portal vein cancer thrombus and VP4 type cancer thrombus. However, in other subgroups, there is no significant statistical difference because this study compared the efficacy of combination therapy and did not show similar results in terms of survival benefit as other studies.

Notably, the median PFS of 10.60 months in our H+T+I group exceeds the 9.53 months reported by Zhang et al. (14) for HAIC combined with camrelizumab plus apatinib. Similarly, Yuan et al. (16) observed a median PFS of 14.8 months with combination therapy versus TACE alone in PVTT-positive HCC, slightly lower than our T+H+T+I cohort’s 15.43 months. This discrepancy may reflect our inclusion of BCLC stage B patients, potentially contributing to marginally improved survival metrics. However, the lack of OS benefit in BCLC B subgroups—likely attributable to limited sample size—warrants cautious interpretation. Furthermore, a meta-analysis by Haber et al. (17) suggested that ICI therapy demonstrated enhanced efficacy in patients with virus-related hepatitis-induced HCC, whereas TKI efficacy showed no significant association with etiology. This may partially account for the superior survival benefits observed with the T+H+T+I combination regimen in the hepatitis-positive subgroup.

TACE induces the release of tumor antigens and proinflammatory cytokines, and locally triggers immunogenic cell death in HCC to further enhance immune responses (18); The chemotherapeutic agents in HAIC modulate the immune microenvironment by enhancing CD8+ T cell infiltration into tumors, which may synergize with ICIs to achieve better outcomes (19, 20);Lu et al. (21) performed a meta-analysis comparing HAIC combined with systemic therapy versus systemic therapy alone, indicating that the integration of HAIC with systemic treatment confers favorable survival benefits in advanced HCC. Both TACE and HAIC elicit antitumor immune responses through microenvironment remodeling, thereby synergizing with immunotherapy. Hypoxia induced by TACE promotes angiogenesis; TKIs counter this effect through their multi-kinase inhibitory activity, which confers anti-proliferative and anti-angiogenic properties (22);Fan et al.’s (23) randomized clinical trial demonstrated that for patients with recurrent intermediate-stage HCC and microvascular invasion (MVI) positivity after hepatectomy, TACE combined with sorafenib was superior to TACE alone. Pro-angiogenic factors such as VEGF are frequently upregulated in HCC patients. VEGF impairs leukocyte-endothelial interactions, limiting immune cell infiltration. Consequently, VEGF blockade may increase effector T-cell populations and potentiate immunotherapy (10);Zhang et al.’s meta-analysis revealed that the TACE plus lenvatinib and ICIs group demonstrated superior efficacy compared to the TACE plus lenvatinib group, with an acceptable safety profile. HAIC demonstrates superior efficacy in patients with large tumors or portal vein invasion. In contrast, TACE monotherapy often fails to achieve complete tumor devascularization in high-burden HCC, adjunctive administration of particulate embolic agents not only fails to confer prognostic improvement (24), but also elevates risks of non-target embolization and hepatic decompensation. Studies by Deng et al. (25) and He et al. (26) indicate that HAIC achieves superior tumor control rates compared to TACE in patients with massive HCC lacking vascular invasion or distant metastasis, HAIC also provides higher intratumoral drug concentration, reduced systemic toxicity. Patients with PVTT, particularly VP3/4 subtypes, exhibit poorer prognosis (27), and are ineligible for surgical resection. While combined local and systemic therapies are essential, no standardized treatment protocols exist for this subgroup. Local treatment options include radioactive seed implantation (28) and radiotherapy (29). TACE monotherapy shows limited efficacy in these patients, with studies reporting increased ischemic liver injury risk in PVTT cases, especially those with compromised hepatic reserve or insufficient portal collateral circulation (30). HAIC is prioritized as a treatment option in Japan and Taiwan (31, 32). Hu et al. (33) demonstrated superior survival benefits of HAIC over TACE/TAE in HCC patients with PVTT. Thus, HAIC addresses the limitations of TACE monotherapy in managing large tumors and PVTT, which may explain the improved OS and PFS observed in the T+H+T+I cohort of this study.

The T+H+T+I cohort demonstrated marginally superior tumor response rates compared to the H+T+I group. In this trial, the tumor response rate of the T+H+T+I cohort was comparable to that of the TACE combined with sorafenib plus ICIS regimen in Zheng et al.’s study (34) (ORR: 54.0% vs. 54.5%). Potential explanations include this study’s larger sample size and poorer baseline characteristics (particularly tumor size) in the enrolled population. Zhou et al. (35) reported a 6-month DCR (61.4% vs. 64%) similar to this study in their HAIC combined with lenvatinib and immune checkpoint inhibitor cohort, though with a higher ORR. This discrepancy may stem from differences in evaluation timing and the relatively smaller sample size of the H+T+I subgroup. No statistically significant differences in tumor response were observed between groups either before or after PSM. This may reflect the comprehensive baseline heterogeneity in the study population. Subgroup stratification based on tumor burden or vascular invasion patterns might reveal statistically significant positive outcomes regarding tumor control.

Regarding safety profiles, the T+H+T+I cohort exhibited a marginally higher incidence of AEs. Treatment-related pain and hepatic impairment were primarily attributable to TACE and HAIC procedures. Compared to HAIC monotherapy, the combined TACE+HAIC approach imposed greater hepatic burden and resulted in a higher incidence of postprocedural pain. Cutaneous reactions (rash, hand-foot syndrome) and gingival bleeding occurred at comparable frequencies in both cohorts, likely owing to their shared systemic therapy components. Notably, the incidence of grade ≥3 AEs showed no statistically significant intergroup difference, aligning with historical data from prior combination therapy studies, thus supporting the acceptable safety profile of the T+H+T+I regimen.

This retrospective study has several inherent limitations. Although PSM mitigated baseline disparities, residual selection bias persists. The single-center design constrained post-matching sample sizes, particularly in subgroup analyses, resulting in inconclusive findings for certain subgroups. Furthermore, single-center data limit generalizability due to the absence of multicenter validation with standardized protocols. To address these limitations, multicenter prospective trials are warranted. Such trials should implement stratified designs based on treatment center, hepatic function, and vascular invasion status, while establishing standardized certification systems for interventional procedures. Uniform imaging assessment criteria and quality control of therapeutic processes should be implemented to ensure sufficient subgroup power and robust outcome evaluation for validating the efficacy and safety of combination therapies in BCLC stage B-C patients. Additionally, this study did not utilize drug-eluting beads as embolic materials; future investigations could conduct subgroup analyses regarding embolic agent selection.

The TACE+HAIC+TKI+ICI combination therapy demonstrates prolonged survival, favorable tumor response rates, and acceptable safety-particularly in patients with tumor diameter ≥5 cm or PVTT. This multimodal regimen represents a potential therapeutic option for advanced HCC with high tumor burden or vascular invasion, pending validation in prospective multicenter trials.