Hepatocellular carcinoma (HCC) complicated by portal vein tumor thrombus (PVTT) is associated with a very poor prognosis, limited treatment options, and substantial clinical challenges. Immunotherapy combined with targeted therapy has become an important first-line treatment approach for unresectable HCC. Although the triplet strategy of hepatic arterial infusion chemotherapy (HAIC) combined with immunotherapy and targeted therapy has demonstrated promising antitumor activity in small phase II studies, large-scale real-world evidence confirming its survival benefits has been lacking.Recently, the CHANCE2504 study, led by Academician Gaojun Teng of Zhongda Hospital, Southeast University, was published online in Clinical and Molecular Hepatology. This nationwide multicenter retrospective cohort study, conducted using a target trial emulation framework, aimed to evaluate the efficacy and safety of immune checkpoint inhibitors (ICIs) combined with molecular targeted agents (MTAs), with or without HAIC, as first-line treatment for HCC with PVTT. The study provides real-world evidence to inform treatment decisions for this difficult-to-treat population.
The results showed that HAIC combined with ICIs and MTAs significantly improved survival compared with ICIs plus MTAs alone, with an acceptable safety profile. These findings suggest a promising treatment strategy for patients with HCC and PVTT. This article provides a detailed interpretation of the study.
Study Design
This nationwide retrospective cohort study employed a target trial emulation framework and the clone-censor-weight (CCW) method. It was conducted across 36 centers in China and aimed to compare clinical outcomes between patients who received HAIC and those who did not, using an 8-week treatment-assignment window.
Patients with HCC and PVTT treated between June 2018 and December 2024 were included and divided into two groups:
- HAIC-ICIs-MTA group: HAIC combined with ICIs and MTAs.
- ICIs-MTA group: ICIs combined with MTAs.
The primary endpoint was overall survival (OS). Secondary endpoints included progression-free survival (PFS), objective response rate (ORR), and safety.
The inclusion criteria were as follows:
- HCC confirmed histologically or clinically.
- Age ≥18 years.
- Presence of PVTT.
- At least one measurable intrahepatic lesion according to RECIST version 1.1.
- Child-Pugh score ≤7.
- Eastern Cooperative Oncology Group (ECOG) performance status score <2.
Study Results
A total of 1,649 patients with HCC were included: 614 in the HAIC-ICIs-MTA group and 1,035 in the ICIs-MTA group.
In the HAIC-ICIs-MTA group, the median age was 55 years (interquartile range [IQR], 48–61 years), and 538 patients (87.6%) were male. In the ICIs-MTA group, the median age was 56 years (IQR, 49–64 years), and 898 patients (86.8%) were male.
Compared with ICIs Plus MTAs Alone, HAIC Combined with ICIs and MTAs Significantly Prolonged OS and PFS and Increased ORR
OS: At the data cutoff, the median follow-up was 21.7 months (95% confidence interval [CI], 20.3–23.1). A total of 511 patients (49.4%) in the ICIs-MTA group and 263 patients (42.8%) in the HAIC-ICIs-MTA group had died.
After adjustment using CCW weighting, median OS was 23.4 months in the HAIC-ICIs-MTA group (95% CI, 21.5–29.3) and 16.5 months in the ICIs-MTA group (95% CI, 15.7–18.3). The difference in restricted mean survival time (RMST) for OS was 7.4 months (95% CI, 3.5–11.7; P<0.001), and the adjusted hazard ratio (HR) was 0.62 (95% CI, 0.52–0.71).
PFS: According to RECIST version 1.1, median PFS was 8.7 months in the HAIC-ICIs-MTA group (95% CI, 7.9–10.0) and 6.2 months in the ICIs-MTA group (95% CI, 5.8–6.7). The difference in RMST for PFS between the groups was 2.6 months (95% CI, 0.9–4.5; P=0.002), with an adjusted HR of 0.76 (95% CI, 0.67–0.83).
ORR: The ORR was 38.9% in the HAIC-ICIs-MTA group (95% CI, 35.1%–42.9%), compared with 22.7% in the ICIs-MTA group (95% CI, 20.2%–25.4%; P<0.001).
Safety
Treatment-related adverse events (TRAEs) of any grade occurred in 581 patients (94.6%) in the HAIC-ICIs-MTA group and 866 patients (83.7%) in the ICIs-MTA group. Grade ≥3 TRAEs occurred in 243 patients (39.6%) and 335 patients (32.4%), respectively (P<0.001).
Grade 5 TRAEs were reported in one patient (0.2%) in the HAIC-ICIs-MTA group and two patients (0.2%) in the ICIs-MTA group.
The incidences of grade ≥3 TRAEs, including abdominal pain, elevated aspartate aminotransferase (AST)/alanine aminotransferase (ALT), thrombocytopenia, and decreased white blood cell counts, were significantly higher in the HAIC-ICIs-MTA group than in the ICIs-MTA group (all P<0.001).
Study Conclusions
This nationwide retrospective cohort study, conducted using a target trial emulation framework, provides real-world evidence that HAIC combined with ICIs and MTAs improves survival outcomes in patients with HCC and PVTT compared with ICIs plus MTAs alone, with an acceptable safety profile.
These findings suggest that the combination regimen may offer a promising treatment strategy for this difficult-to-treat population. However, given the inherent limitations of retrospective studies, the results require further validation in prospective multicenter randomized controlled trials. Identifying biomarkers that can predict treatment response will also be important.
Discussion
Several previous phase II clinical trials have demonstrated promising efficacy with HAIC combined with ICIs and MTAs. This large-scale real-world study extends those findings by providing comparative evidence supporting the survival benefits of the HAIC-ICIs-MTA regimen in patients with HCC and PVTT.
The TRIPLET trial evaluated HAIC combined with apatinib and camrelizumab in advanced HCC. Among the enrolled patients, 71.8% had PVTT. Median PFS was 10.4 months, and the ORR was as high as 77.1%.
Similarly, the DoHAICS study demonstrated encouraging efficacy with HAIC combined with donafenib and sintilimab in patients with intermediate- or advanced-stage HCC. Median PFS was 15.3 months, and the ORR was 58.3%; 38.9% of patients in that study had PVTT.
Despite a greater tumor burden in the present study population—with 66.3% of patients having a SIX-T12 score >12—all patients having PVTT, and 21% having extrahepatic metastases—the HAIC-ICIs-MTA group still demonstrated substantial efficacy. Median OS reached 23.4 months, median PFS was 9.0 months, and the ORR was 49.2%, broadly consistent with findings from previous clinical trials.
These benefits may be partly attributable to the effective locoregional tumor control achieved with HAIC. However, the aforementioned phase II trials were limited by immature follow-up and the lack of control groups, leaving them unable to establish whether the HAIC-ICIs-MTA strategy improves OS. The present target trial emulation study provides complementary real-world evidence, demonstrating a significant OS benefit with the combination regimen in patients with HCC and PVTT.
The HAIC-ICIs-MTA combination may beneficially modulate the tumor microenvironment, potentially transforming “cold” tumors into “hot” tumors, restoring immune surveillance, and enhancing the antitumor effects of ICIs. In addition, combining ICIs and MTAs with HAIC may help maintain or improve tumor response while allowing reductions in the dose or frequency of chemotherapy administration.
The present study found that grade 3–4 TRAEs were more frequent with HAIC-ICIs-MTA than with ICIs-MTA alone (39.6% vs. 32.4%; P<0.001). The types and incidence of grade ≥3 TRAEs observed were broadly consistent with previous reports. The most common severe TRAEs included elevated serum AST/ALT, abdominal pain, thrombocytopenia, and decreased white blood cell counts. These adverse events were primarily attributable to the chemotherapy agents used in HAIC and could generally be managed with symptomatic treatment or dose adjustments before the next HAIC cycle.
Furthermore, the two groups had similar rates of temporary treatment interruption, dose adjustment, and treatment discontinuation. Overall, the HAIC-ICIs-MTA combination appears manageable and generally well tolerated.
