Hepatocellular carcinoma (HCC) ranks as the sixth most prevalent malignancy and the third leading cause of cancer-induced mortality worldwide, with approximately 900,000 new cases and 830,000 fatalities in 2020 (1,2). Effective treatments for advanced HCC did not emerge until 2007 (3). Prior to the advent of tyrosine kinase inhibitors, numerous clinical trials for drug therapies for unresectable HCC were conducted without any chemotherapeutic drugs demonstrating a survival benefit, as indicated by a meta-analysis by Mathurin et al. (4). Following research into tumor cell proliferation and angiogenesis mechanisms, the tyrosine kinase inhibitor Sorafenib (SOR) was developed (5). In the "SHARP trial" for unresectable HCC, SOR significantly improved survival over placebo, establishing it as the standard treatment for unresectable HCC in 2007 (6). In 2018, Lenvatinib (LEN) was found to be "non-inferior" to SOR in the "REFLECT trial", offering a choice between SOR or LEN as first-line therapy (7). The emergence of immune checkpoint inhibitors introduced a combination therapy of atezolizumab and bevacizumab (ATZ+BEV) in 2020, which outperformed SOR in clinical trials (8). Consequently, numerous agents, including immune checkpoint inhibitors, have become available for treating unresectable HCC. Hepatic arterial infusion chemotherapy (HAIC) is a localized treatment delivering cytotoxic chemotherapy directly into the hepatic artery via an implanted catheter port system, aiming to maximize HCC exposure to the chemotherapy while minimizing systemic side effects. HAIC has shown promising results in unresectable HCC patients (9). Liver function is assessed with the Child–Pugh classification, which ranges from Child–Pugh class A, indicative of compensated cirrhosis, to Child–Pugh class B and C, which signify decompensated cirrhosis (10). Most systemic chemotherapy trials have enrolled patients with Child–Pugh class A status, often excluding or enrolling few patients with Child–Pugh class B due to the competing risk of mortality from liver dysfunction rather than HCC progression. Thus, systemic chemotherapy is generally recommended for HCC patients with Child-Pugh class A, while optimal treatments for patients with Child–Pugh class B have not been established. Here, our analysis evaluated the outcomes of Child–Pugh class B patients who received systemic chemotherapies (SOR, LEN, ATZ+BEV, HAIC) for HCC, aiming to identify prognostic survival factors.

Patients. In this prospective, single-center study at Aso Iizuka Hospital, the efficacy of systemic chemotherapy was evaluated from December 2003 to June 2023. Systemic chemotherapy treatments (SOR, LEN, ATZ+BEV, HAIC) were administered to 136 patients classified under Child-Pugh class B. The study, adhering to the Declaration of Helsinki, was sanctioned by the Ethics Committee of Aso Iizuka Hospital. Patient consent was obtained using the opt-out approach (approval code: 23139).

Assessment of liver function. Liver function was measured with the ALBI score, derived from the following formula: ALBI score = (log10(T-Bil[mg/dl]×17.1)×0.66) + ((ALB[g/dl]×10)×−0.085). In this formula, T-Bil signifies total bilirubin, and ALB represents the serum albumin level (11).

Chemotherapy regimens. SOR. SOR in doses ranging from 200-800 mg (Bayer Health Care Pharmaceuticals, West Haven, CT, USA), was administered according to package guidelines. Generally, the standard dose is 800mg/day, but initial dosing was tailored based on patient age, body weight, performance status, and liver function.

LEN. LEN was prescribed based on patient body weight (8 mg/day for those under 60 kg, or 12 mg/day for those weighing 60 kg or more) (Eisai Co., Ltd., Tokyo, Japan). We adjusted LEN dosages in response to LEN-induced adverse events, following the Common Terminology Criteria for Adverse Events, version 4.0. Doses were either reduced or temporarily discontinued until adverse event symptoms reduced to grade 1 or 2.

ATZ+BEV. Following the Imbrave150 trial guidelines (8), patients received intravenous doses of atezolizumab (1,200 mg) and bevacizumab (7.5 mg/kg) (Chugai Co., Ltd., Tokyo, Japan) every three weeks, continuing until either disease progression or the occurrence of intolerable side effects.

HAIC. A 5-Fr-W-spiral catheter (Piolax, Yokohama, Japan) was inserted via the right femoral artery for HAIC catheter implantation. The catheter’s distal end was placed in the hepatic or gastroduodenal artery, and a subcutaneous port (Sofa Port, Nipro Pharma Co., Ltd., Osaka, Japan) was installed in the front femoral region (12). The HAIC treatment involved a cisplatin-lipiodol mixture, comprising 10-50 mg of finely powdered cisplatin in 5-10 ml of lipiodol, adjusted according to tumor size. On day one, this suspension was injected through the catheter under angiography, followed by a 5-day continuous infusion of 1500 mg 5-FU via an infusion balloon pump (SUREFUSER PUMP; Nipro Pharma Co., Ltd.).

Evaluation of efficacy. Treatment responses were evaluated every 6 to 12 weeks using either computed tomography or magnetic resonance imaging. The assessment of antitumor effects employed the Modified RECIST version 1.1 criteria (13). Indicators of the disease control rate (DCR) included complete response (CR), partial response (PR), and stable disease (SD) for a minimum of 4 months. The objective response rate (ORR) combined PR and CR. Patient evaluations and treatments were conducted every three weeks until either disease progression or unacceptable adverse effects occurred.

Statistical analysis. Statistical analysis was conducted using JMP Pro version 11 (SAS Institute, Cary, NC, USA), with results presented as medians. The Kaplan–Meier method, log-rank test, and Cox hazard analysis were applied for comparative statistical analyses. The chi-squared test or Fisher exact test assessed group differences, considering a p-value below 0.05 as significant.

Patient characteristics. The characteristics of the 137 patients who underwent systemic chemotherapy are presented in Table I (SOR: n=43, LEN: n=16, ATZ+BEV: n=18, HAIC: n=60). Patients who underwent HAIC were younger than those who received other treatments. Patients treated with ATZ+BEV had a higher incidence of microvascular invasion (MVI) positivity, while those treated with HAIC had a higher incidence of extrahepatic spread (EHS) positivity and larger tumor sizes than those receiving other treatments. Patients receiving ATZ+BEV or HAIC had poorer Child–Pugh scores compared to those treated with SOR or LEN. Patients undergoing HAIC treatment were more likely to have BCLC stage C HCC. Serum α-fetoprotein (AFP) levels were lower in patients treated with LEN compared to other therapies. Sex, albumin-bilirubin (ALBI) scores, the number of intrahepatic lesions, and protein induced by vitamin K absence or antagonist-II (PIVKA-II) levels were similar across the different treatments.

Overall survival (OS). No significant differences in OS were observed among the different systemic chemotherapy groups (SOR: 8.3 months, LEN: 10.2 months, ATZ+BEV: 8.5 months, HAIC: 7.3 months) (Figure 1).

Overall response. The ORR among patients who received SOR was 4/43 (12.1%), and the DCR was 6/43 (18.2%). The ORR for patients who received LEN was 2/16 (12.5%) and the DCR was 8/16 (50.0%). For those who received ATZ+BEV, the ORR was 2/18 (15.4%) and the DCR was 8/18 (61.5%). For patients treated with HAIC, the ORR was 17/60 (29.8%) and the DCR was 26/60 (45.6%). There were no significant differences in ORR among the systemic chemotherapy treatments (Table II).

Rate of treatment discontinuation and effects on liver function. Patients receiving HAIC and LEN were less likely to discontinue treatment within 3 months compared to those receiving ATZ+BEV and SOR (Table III). HAIC treatment resulted in fewer ALBI score changes and better preserved liver function (Figure 2).

Factors associated with OS. Univariate analysis revealed that a tumor number >5, Child-Pugh score, and serum AFP >400 ng/ml were associated with OS. The type of systemic chemotherapy regimen did not affect the OS of advanced HCC patients with Child-Pugh class B. Multivariate analysis identified serum AFP >400 ng/ml [hazard ratio (HR)=1.94; p=0.001], tumor number >5 (HR=1.55; p=0.043), and Child-Pugh score (HR=2.53; p=0.002) as independent factors associated with OS.

The effectiveness of systemic chemotherapies in patients with Child-Pugh class B remains uncertain, as these patients have been historically excluded from clinical trials of anticancer drugs. Currently, there is no established treatment for advanced HCC patients with Child–Pugh class B. This study is the first to compare SOR, LEN, ATZ+BEV, and HAIC in unresectable HCC patients with Child-Pugh class B. The GIDEON study was a prospective, observational registry designed to evaluate the real-world safety of SOR in patients with HCC. It included data from 669 treatment-naive patients with Child-Pugh class B. The median OS was significantly shorter for patients with Child-Pugh class B compared to those with Child-Pugh class A. Similarly, higher adverse event rates were observed in patients with Child-Pugh class B (14). Studies have demonstrated lower response rates and shorter OS in patients with HCC treated with LEN who had Child-Pugh class B compared to those with Child-Pugh A class (15,16). Several studies reported that patients with HCC treated with atezolizumab and bevacizumab showed a significant difference in median OS between Child-Pugh class A and B (17,18). It has been reported that OS did not significantly differ between patients with Child-Pugh class A and B who received HAIC (19). This study also indicated that HAIC treatment resulted in lower treatment discontinuation rates and fewer ALBI score changes. Most patients in our study with HCC who received HAIC had active hepatitis C virus (HCV) infection before the introduction of direct-acting antivirals. Previous reports have highlighted the crucial role of HCV eradication in the survival outcomes of advanced HCC patients treated with SOR (20). Given the historical context, HAIC may represent an important treatment option for HCC in patients with Child-Pugh class B. There are few studies comparing various systemic chemotherapy treatments. Ohama et al. (21) found no significant differences in OS between Child-Pugh class B patients treated with ATZ+BEV and LEN. Kikugawa et al. (22) reported similar findings, adding that Child-Pugh score was an OS-associated factor, as observed in our study. This study’s limitations include a small sample size and its single-center design. It encompasses unresectable HCC patients across different stages and eras. Ideally, the groups would be matched by liver function, HCC stage, and chemotherapy line, but this is challenging with a small cohort. The absence of a control group not receiving treatment also obscures the potential benefit of systemic chemotherapy for patients with Child-Pugh class B. Recent advances include the STRIDE regimen, combining Durvalumab and Tremelimumab, which surpassed SOR in the HIMALAYA trial, and Durvalumab monotherapy, which demonstrated non-inferiority to SOR (23). Kudo et al. noted the effectiveness and safety of Nivolumab, suggesting its suitability for patients with Child-Pugh class B (24). The use of Durvalumab and Tremelimumab promises efficacy and safety for such patients.

There were no significant differences in OS and response rates among the systemic chemotherapies. The prognosis for HCC patients with Child-Pugh class B was linked to liver function. Further research is required to determine optimal treatments for HCC in patients with Child-Pugh class B.

All Authors declare no competing interests in relation to this study.

A.K., M.Y., A.M., and K.M. designed the study. A.K., Y.K, S.N., K.T., and M.Y. assisted with the data analyses. A.K. wrote the initial draft of the manuscript. M.Y. contributed to the analysis and interpretation of the data. M.Y., A.M., and K.M. assisted in the preparation and critical review of the manuscript. All Authors approved the final version of the manuscript and agreed to be accountable for all aspects of the work.

The Authors are grateful to Y. Ishibashi for his assistance with manuscript preparation.