Efficacy and safety of surufatinib and anti-PD-L1 therapy in combination with hepatic arterial infusion chemotherapy for unresectable intrahepatic cholangiocarcinoma

Abstract

BACKGROUND

The prognosis for unresectable intrahepatic cholangiocarcinoma (ICC) is poor, and the efficacy of conventional chemotherapy remains unsatisfactory. Hepatic arterial infusion chemotherapy (HAIC) with oxaliplatin, leucovorin, and 5-fluorouracil (FOLFOX) has demonstrated activity in patients with unresectable ICC. Surufatinib is a small-molecule antiangiogenic tyrosine kinase inhibitor approved in China for second-line treatment of advanced cholangiocarcinoma.

AIM

To evaluate the preliminary clinical efficacy and safety of surufatinib plus anti- programmed death-ligand 1 (PD-L1) therapy in conjunction with FOLFOX-HAIC among individuals diagnosed with treatment-naive, surgically inoperable ICC.

METHODS

Between August 2023 and July 2025, patients with unresectable ICC who initially received surufatinib plus anti-PD-L1 therapy in conjunction with FOLFOX-HAIC were reviewed for study inclusion. Efficacy was determined based on tumor response and survival outcomes, while the safety profile was evaluated based on the incidence of principal adverse events (AEs).

RESULTS

Forty eligible patients were enrolled in this study. Among the 36 patients evaluable for efficacy, the objective response rate based on Response Evaluation Criteria in Solid Tumors version 1.1 criteria was 47.2%. The 6-month, 12-month, and 24-month overall survival rates were 82.5%, 63.3%, and 55.4%, respectively. The median progression-free survival was 6.3 months (95% confidence interval: 5.7-6.9). Notably, 22.5% (9/40) of patients underwent radical surgical resection following successful conversion. All patients (100.0%) experienced AEs, while 30.0% (12/40) developed grade 3 or 4 AEs.

CONCLUSION

The combination of FOLFOX-HAIC, surufatinib, and anti-PD-L1 therapy demonstrated substantial clinical efficacy alongside a favorable safety profile among patients with previously untreated and unresectable ICC. This therapeutic strategy could represent an alternative treatment option for selected patients.

Key Words: Unresectable intrahepatic cholangiocarcinoma; Surufatinib; Anti-programmed cell death-ligand 1; Hepatic arterial infusion chemotherapy; 5-Fluorouracil

Core Tip: Combining surufatinib and anti-programmed death-ligand 1 therapy with hepatic arterial infusion chemotherapy yields promising antitumor activity and survival benefits for unresectable intrahepatic cholangiocarcinoma. This innovative regimen demonstrates a highly favorable safety profile, characterized by a low incidence of grade 3 or 4 adverse events. Notably, patients who respond to this combination therapy achieve significantly better prognostic outcomes compared to non-responders. These findings highlight a potent, well-tolerated therapeutic strategy that could substantially improve clinical management for unresectable intrahepatic cholangiocarcinoma.

INTRODUCTION

As malignant tumors originating from biliary epithelial cells, cholangiocarcinomas exhibit substantial heterogeneity and aggressive behavior, typically categorized into intrahepatic cholangiocarcinoma (ICC) and extrahepatic cholangiocarcinoma[1]. As the second most prevalent primary liver tumor, ICC is responsible for up to 20% of all cases of hepatic malignancies. Notably, a greater than 140% increase in the incidence of ICC has been observed during the past four decades[2]. Although surgical resection remains the only curative option, most patients present with advanced disease at symptom onset, thereby precluding surgical candidacy. Without therapeutic intervention, median overall survival (OS) for surgically ineligible ICC ranges from 2.5 to 7.5 months, reflecting a highly unfavorable prognosis[3].

Although the gemcitabine-cisplatin (GC) regimen serves as the first-line standard of care for surgically ineligible biliary tract cancer, it offers modest efficacy, yielding a median OS of just 11.7 months[4]. By contrast, hepatic arterial infusion chemotherapy (HAIC) provides a locoregional treatment strategy, enabling high concentrations of chemotherapeutic agents to be delivered directly to hepatic tumors while minimizing systemic toxicity[5]. Comparative studies have shown that HAIC offers superior tumor control relative to systemic chemotherapy in advanced ICC[6,7]. Moreover, the oxaliplatin, leucovorin and 5-fluorouracil (FOLFOX) regimen has become a viable palliative option for advanced ICC, demonstrating meaningful clinical activity[7,8]. Previous study further demonstrated that combining FOLFOX-HAIC with lenvatinib and anti-programmed cell death protein 1 (PD-1) therapy yields favorable efficacy and safety in treating ICC[9]. Collectively, these findings highlight the promising clinical value of FOLFOX-HAIC for advanced ICC, supporting its potential as an effective therapeutic strategy.

As a novel oral tyrosine kinase inhibitor, surufatinib exerts its pharmacological action by inhibiting vascular endothelial growth factor receptors 1, 2, and 3, fibroblast growth factor receptor 1, and the colony-stimulating factor 1 receptor. Surufatinib has been approved for the treatment of locally advanced or metastatic neuroendocrine tumors that are not suitable for surgical resection[10]. Recently, owing to its strong efficacy and manageable safety profile, surufatinib has also been approved in China as a second-line treatment for advanced cholangiocarcinoma[11].

Two pivotal phase III trials (TOPAZ-1 and KEYNOTE-966) showed that adding programmed death-ligand 1 (PD-L1) or PD-1 inhibitors to chemotherapy significantly improves OS in patients with unresectable or advanced biliary tract cancer[12,13]. Our prior findings further indicate that anti-PD-L1 therapy yield higher objective response rates (ORR) and more favorable safety profiles than anti-PD-1 therapy when combined with HAIC for unresectable ICC[14]. Therefore, this study aims to investigate the efficacy and safety of HAIC combined with surufatinib and anti-PD-L1 therapy in the treatment of unresectable ICC.

MATERIALS AND METHODS

This study received ethical approval from the institutional ethics committee at Sun Yat-sen University Cancer Center (SYSUCC) in Guangzhou, China, No. B2024-665-Y01, and all procedures were conducted in accordance with the Declaration of Helsinki.

Patients

From August 2023 to July 2025, patients with ICC treated with surufatinib, anti-PD-L1 therapy, and FOLFOX-HAIC in the Liver Surgery Department of SYSUCC were assessed for inclusion in the study. Patients were required to meet the following eligibility prerequisites: (1) Clinically or histopathologically verified ICC; (2) Determination of ineligibility for radical surgery, assessed by at least two hepatobiliary surgeons with more than 10 years of experience (the criteria for unresectability were as follows: Tumor diameter > 5 cm with insufficient future liver remnant or tumors in close proximity to major vascular structures, such as the inferior vena cava, right main portal vein, or middle hepatic vein, precluding a negative margin); (3) Aged between 18 and 75 years; (4) Sufficient liver and kidney function; (5) An Eastern Cooperative Oncology Group performance status of 0-1; and (6) The presence of at least one target lesion considered measurable under Response Evaluation Criteria in Solid Tumors, version 1.1 (RECIST 1.1). Patients were excluded if they: (1) Underwent prior anti-cancer treatments; (2) Possessed a historical record of concurrent or alternative malignancies; (3) Deficient medical documentation; or (4) Were lost to follow-up.

Treatment procedure

HAIC was administered at 3-week intervals, following established protocols[9]. The procedure involved positioning a microcatheter into the primary feeding hepatic artery, through which the following regimen was delivered: Oxaliplatin (135 mg/m²) from hour 0 to 3 on day 1; leucovorin (400 mg/m²) from hour 3 to 4.5 on day 1; FOLFOX (400 mg/m²) as a bolus from hour 4.5 to 6.5 on day 1; followed by continuous FOLFOX infusion (2400 mg/m²) over 46 hours from day 1 to day 3. Concurrently, patients received surufatinib orally at a daily dose of 200 mg or 250 mg according to body weight, and anti-PD-L1 therapy was given intravenously or subcutaneously every 3 weeks. The selection of specific PD-L1 inhibitors depended on patients’ financial situation, drug availability, and hospital pharmaceutical policies. The types and dosages of PD-L1 agents are listed in Supplementary Table 1. The initial administration of surufatinib and anti-PD-L1 occurred within seven days of starting HAIC. Surufatinib and anti-PD-L1 therapy were continued after HAIC discontinuation until disease progression or the onset of intolerable adverse events (AEs).

Data collection

We conducted a comprehensive review of the SYSUCC electronic medical records to compile the pretreatment characteristics of the selected participants. The extracted dataset captured demographic details (age and sex), liver disease background (Child-Pugh score and hepatitis infection history), and detailed tumor metrics (number, size, localization, vascular invasion, lymph node/distant metastatic status, and TNM staging). Furthermore, routine baseline blood tests were collated, specifically targeting carbohydrate antigen 19-9, hematological indices (white blood cells, neutrophils, hemoglobin, platelets), hepatic function parameters [alanine transaminase (ALT), albumin, total bilirubin], prothrombin time, and renal function (creatinine).

Assessment of follow-up and survival

During every follow-up appointment, patients underwent medical history taking, clinical examinations, biochemical testing, and radiological imaging (contrast-enhanced computed tomography, magnetic resonance imaging, or both). We scheduled the primary clinical assessment at six to eight weeks post-treatment onset, which corresponds to the completion of two HAIC cycles. The observation period ended on February 13, 2026.

OS was defined as the duration between the first administration of HAIC and death of any etiology. Furthermore, progression-free survival (PFS) was determined as the interval starting from the primary HAIC therapy until either radiographic tumor progression or mortality, whichever happened first. ORR was determined based on the percentage of patients with a complete or partial response that remained evident for a minimum of four weeks following the first imaging-based confirmation. Disease control rate was defined as the percentage of patients whose best response was complete response, partial response, or stable disease. Treatment response was evaluated according to RECIST v1.1[15,16]. AEs were graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events version 4.03.

Statistical analysis

Normally distributed variables are reported as the mean ± SD, whereas nonnormally distributed variables are presented as the median and interquartile range. Categorical variables were compared using the Pearson χ² test or Fisher’s exact test. Continuous parametric variables were analyzed using the unpaired Student’s t-test, while the Mann-Whitney U test was applied for continuous nonparametric variables. Survival analyses were performed using the Kaplan-Meier method, and differences between survival curves were evaluated with the log-rank test. We deployed a forward log-rank approach within univariable and multivariable Cox proportional hazards models to isolate independent prognostic significance. A two-sided P value below 0.05 denoted statistical significance in this study. All analytical procedures and computations were carried out with SPSS version 24.0 (SPSS Inc., Chicago, IL, United States).

RESULTS

Patient characteristics

Between August 2023 and July 2025, a total of 40 eligible patients who consented to the combined regimen of surufatinib, an anti-PD-L1 agent, and FOLFOX-HAIC were included in the final study cohort. Baseline demographics and clinical features are detailed in Table 1. The cohort included 23 male and 17 female patients, ranging from 30 to 72 years of age. Among them, 97.5% (39/40) had Child-Pugh class A liver function; 42.5% (17/40) had hepatitis B infection, and one had hepatitis C infection. The maximum tumor diameter ranged from 2.4 cm to 14.4 cm; 26 patients (65.0%) had multiple lesions; 11 (27.5%) had macrovascular invasion; 18 (45.0%) had hilar lymph node metastasis; and 7 (17.5%) had distant metastasis. Additionally, 75% of patients had a TNM stage of III or higher.

Table 1 Baseline clinical characteristics of included patients (n = 40), median (interquartile range)/mean ± SD.

Characteristics	Patients
Gender
Male	23
Female	17
Age	55.5 (30-72)/55.4 ± 10.5
Child-Pugh grade
A	39
B	1
Hepatitis virus infection
HBV	17
HCV	1
None	22
Tumor size, cm	7.8 (2.4-14.4)/7.6 ± 2.7
≤ 5	7
> 5	33
≤ 7	16
> 7	24
Tumor number
Single	14
Multiple	26
Tumor distribution
Uni-lobe	23
Bi-lobe	17
Vascular invasion
Absence	29
Presence	11
LN metastasis
Absence	22
Presence	18
Distant metastasis
Absence	33
Presence	7
TNM stage
I-II	10
IIIb	23
IV	7
CA19-9, U/mL
≤ 40	14
> 40	26
WBC, × 109/L	8.8 (6.8-9.7)
NE, × 109/L	6.0 (4.4-6.9)
Hgb, g/L	137.0 (126.5-147.0)
PLT, × 109/L	271.0 (207.8-299.8)
ALT, U/L	21.8 (17.1-33.9)
ALB, g/L	44.9 (42.7-46.7)
TBil, μmol/L	11.6 (9.8-14.4)
PT, seconds	11.5 (10.8-12.2)
CRE, μmol/L	68.2 (55.1-81.4)

HBV: Hepatitis B virus; HCV: Hepatitis C virus; LN: Lymph node; CA19-9: Carbohydrate antigen 19-9; WBC: White blood cell; NE: Neutrophil count; Hgb: Hemoglobin; PLT: Platelet count; ALT: Alanine transaminase; ALB: Albumin; TBil: Total bilirubin; PT: Prothrombin time; CRE: Creatinine.

Tumor response and patient survival

A total of 113 HAIC cycles were administered, with a median of 3 cycles (range, 1-4 cycles). The median follow-up duration for the entire cohort was 18.5 months (range, 14.2-22.8 months). At the end of follow-up, 16 patients had died (16/40, 40.0%), and 31 had experienced disease progression (31/40, 77.5%). As shown in Figure 1, the median OS was not reached; accordingly, the 6-month, 12-month, and 24-month OS rates were 82.5%, 63.3%, and 55.4%, respectively. The median PFS was 6.3 months [95% confidence interval (CI): 5.7-6.9], and the 6-month and 12-month PFS rates were 57.5% and 35.0%, respectively.

Figure 1 Kaplan-Meier curves depicting overall survival and progression-free survival. A: Overall survival; B: Progression-free survival. OS: Overall survival; PFS: Progression-free survival; CI: Confidence interval.

Efficacy outcomes are outlined in Table 2. Among the total cohort (n = 40), 36 patients were assessable for treatment response; the remaining 4 patients were excluded owing to incomplete or missing imaging follow-up data. The ORR based on RECIST 1.1 criteria was 47.2%, and the disease control rate was 94.4%. The maximum percentage reduction in intrahepatic target lesions is graphically visualized in the waterfall plot (Figure 2). After treatment, nine individuals exhibited radiographic downstaging or a reduction in tumor burden, subsequently undergoing radical hepatic resection. A median response time of 2.1 months was observed across the cohort, ranging from 1.0 to 9.3 months. As of the follow-up deadline, the follow-up period for these nine converted patients was 12.8-30.4 months (median: 21.3 months), with no deaths (100% survival rate).

Figure 2 Best percentage changes from baseline in intrahepatic target lesion size in patients receiving surufatinib plus anti-programmed death-ligand 1 combined with 5-fluorouracil-hepatic arterial infusion chemotherapy based on Response Evaluation Criteria in Solid Tumors 1.1 criteria. RECIST: Response Evaluation Criteria in Solid Tumors; PR: Partial response; SD: Stable disease; PD: Progressive disease.

Table 2 Tumor response (n = 36), n (%).

Characteristics	RECIST
Complete response	0
Partial response	17 (47.2)
Stable disease	17 (47.2)
Progressive disease	2 (5.6)
Objective response rate, %	47.2
Disease control rate, %	94.4

RECIST: Response Evaluation Criteria in Solid Tumors.

In the univariate analysis of clinical variables, tumor number demonstrated a significant association with both OS [hazard ratio (HR) = 5.632, 95%CI: 1.271-24.946, P = 0.023] and PFS (HR = 2.646, 95%CI: 1.178-5.947, P = 0.018). Distant metastasis (HR = 7.033, 95%CI: 2.425-20.394, P < 0.001), total bilirubin (HR = 3.153, 95%CI: 1.162-8.555, P = 0.024), and tumor response (HR = 3.611, 95%CI: 1.599-8.157, P = 0.002) were significant predictors of PFS. Multivariate Cox proportional analysis identified that distant metastasis, tumor number, total bilirubin, and tumor response were all significant and independent prognostic factors for PFS (Supplementary Table 2). The OS of responders was longer than that of nonresponders, albeit without statistical significance (HR = 2.80, 95%CI: 0.87-8.96, P = 0.083). In contrast, PFS was significantly better in responders compared to nonresponders (HR = 3.61, 95%CI: 1.60-8.16, P = 0.002; Supplementary Figure 1).

AEs and safety

No treatment-associated mortality was recorded. In response to hepatic toxicity, the HAIC regimen was modified for three patients during their first cycle, with no further AE-related dose adjustments made thereafter. Any-grade treatment-related AEs were observed in the entire cohort (n = 40; Table 3). Among grade 1-2 AEs, hypoalbuminemia was the most frequently observed (97.5%), followed by abdominal pain (55.0%), elevated ALT (47.5%), and elevated creatinine (47.5%). The most frequent grade 3-4 AEs were hypertension (10%), abdominal pain (5.0%), and hyperbilirubinemia (5%). Any-grade liver-related AEs (e.g., ALT elevation, hyperbilirubinemia, and decreased albumin) were predominantly grade 1-2 and resolved with appropriate clinical intervention. Additionally, abdominal pain specifically associated with oxaliplatin-based HAIC occurred in 24 (60.0%) patients; most episodes were mild and tolerable. The few cases of severe, acute pain were rapidly alleviated by pausing oxaliplatin infusion or administering spasmolytics, after which treatment was continued.

Table 3 Summary of adverse events by severity, n (%).

Adverse event	Any grade	Grade 1-2	Grade 3	Grade 4	Treatment-related grade ≥ 3
Total	40 (100.0)	28 (70.0)	11 (27.5)	1 (2.5)	12 (30.0)
Hypoalbuminemia	39 (97.5)	39 (97.5)	0	0	0
Abdominal pain	24 (60.0)	22 (55.0)	2 (5.0)	0	2 (5.0)
ALT level elevated	19 (47.5)	19 (47.5)	0	0	0
CRE level elevated	19 (47.5)	19 (47.5)	0	0	0
Anemia	15 (37.5)	14 (35.0)	1 (2.5)	0	1 (2.5)
Hyperbilirubinemia	15 (37.5)	13 (32.5)	2 (5.0)	0	2 (5.0)
Neutropenia	14 (35.0)	13 (32.5)	1 (2.5)	0	1 (2.5)
Thrombocytopenia	14 (35.0)	13 (32.5)	1 (2.5)	0	1 (2.5)
Leukocytopenia	13 (32.5)	12 (30.0)	1 (2.5)	0	1 (2.5)
Vomiting	9 (22.5)	9 (22.5)	0	0	0
Dysuria	6 (15.0)	6 (15.0)	0	0	0
Hypertension	5 (12.5)	1 (2.5)	4 (10.0)	0	4 (10.0)
Nausea	3 (7.5)	3 (7.5)	0	0	0
Chest distress	3 (7.5)	3 (7.5)	0	0	0
Diarrhea	3 (7.5)	3 (7.5)	0	0	0
Allergy	3 (7.5)	2 (5.0)	1 (2.5)	0	1 (2.5)
Ascites	2 (5.0)	2 (5.0)	0	0	0
Constipation	2 (5.0)	2 (5.0)	0	0	0
Insomnia	2 (5.0)	2 (5.0)	0	0	0
Rash	2 (5.0)	2 (5.0)	0	0	0
Poor appetite	1 (2.5)	1 (2.5)	0	0	0
Bloating	1 (2.5)	1 (2.5)	0	0	0
Fever	1 (2.5)	1 (2.5)	0	0	0
HE	1 (2.5)	0	0	1 (2.5)	1 (2.5)

ALT: Alanine transaminase; CRE: Creatinine; HE: Hepatic encephalopathy.

DISCUSSION

Our team has long focused on preoperative conversion and postoperative adjuvant strategies for hepatobiliary malignancies using HAIC[17,18]. We previously reported that for ICC with predominantly intrahepatic tumor burden, HAIC combined with lenvatinib and durvalumab demonstrated better efficacy than the standard first-line regimen of gemcitabine, cisplatin, and durvalumab[9]. Building on these findings, we are now investigating the efficacy and safety of HAIC combined with the novel antiangiogenic agent surufatinib and anti-PD-L1 therapy in unresectable ICC. According to the inclusion criteria, patients enrolled in this study had good hepatic and renal function, an Eastern Cooperative Oncology Group performance status of 0-1, and limited tumor burden. The healthier overall condition of the study population may partly explain the positive outcomes observed. Additionally, these clinical conversion decisions were supported by our department’s long-established multidisciplinary team framework. At our center, hepatobiliary surgeons routinely manage surgical and nonsurgical treatments, including interventions and systemic therapy. For complex cases, regular monthly multidisciplinary team meetings involving radiologists, pathologists, and oncologists were held to optimize perioperative strategies.

Our retrospective analysis demonstrates that the triplet combination of surufatinib, an anti-PD-L1 agent, and FOLFOX-HAIC yields a robust ORR of 47.2% in unresectable ICC. Furthermore, this treatment regimen provided promising survival benefits, with a median PFS of 6.3 months and 6-, 12-, and 24-month OS rates of 82.5%, 63.3%, and 55.4%, respectively, while maintaining a manageable safety profile. Compared with nonresponders, patients who responded to therapy experienced significantly longer PFS. Although responders tended to have better OS, the difference was not statistically significant (P = 0.083), possibly due to the small sample size and limited follow-up duration. Collectively, these findings indicate that this innovative triplet combination offers substantial clinical benefits and a clear survival advantage for patients with treatment-naive, unresectable ICC.

By primarily targeting vascular endothelial growth factor receptor, tyrosine kinase inhibitors can potentiate the antitumor activity of anti-PD-L1/PD-1 therapy in hepatic malignancies[19,20]. The robust tumor responses and prolonged survival achieved in this study likely stem from the synergistic interplay among FOLFOX-HAIC, surufatinib, and anti-PD-L1 therapy. HAIC decreases tumor burden by sustaining high intrahepatic concentrations of chemotherapeutic agents. Additionally, chemotherapy-induced immunogenic cell death enhances the antitumor activity of immune checkpoint inhibitors (ICIs)[21,22]. Surufatinib may further potentiate immunotherapy by dampening immunosuppressive subsets like regulatory T cells, while promoting the recruitment of effector T cells to the tumor microenvironment[23,24]. Additionally, surufatinib combined with ICIs could potentially help overcome FOLFOX resistance by normalizing tumor vasculature and reducing hypoxia within the tumor microenvironment[25]. Thus, triple therapy with surufatinib, anti-PD-L1, and HAIC may promptly reduce tumor burden, maintain systemic efficacy, and ultimately improve long-term survival. Notably, surufatinib was administered at a lower dose (200 mg or 250 mg) instead of the standard 300-mg single-agent dose. Prior research in hepatocellular carcinoma has shown that dose reductions of targeted therapies are feasible when combined with immunotherapy, lowering AE risks and improving treatment tolerance without compromising efficacy[26,27]. We adjusted the surufatinib dose accordingly, aiming to achieve an optimal balance between antitumor activity and safety. Importantly, surufatinib exerted minimal effects on hepatic function, consistent with its known safety profile. This low hepatotoxicity risk is particularly beneficial for maintaining an adequate liver functional reserve, thereby preserving eligibility for subsequent radical surgical resection.

Due to the highly invasive nature of cholangiocarcinoma, more than 70% of patients experience tumor recurrence after surgery[13]. With advances in first-line therapies for advanced cholangiocarcinoma, these regimens have increasingly been evaluated as neoadjuvant or conversion strategies in potentially resectable disease[28]. We previously reported a multicenter, real-world study of 326 patients with initially unresectable advanced ICC treated with HAIC combined with antiangiogenic agents and ICIs, in which 20% achieved successful conversion and subsequently underwent surgical resection. Patients who underwent conversion surgery achieved a 2-year OS rate approaching 80%, whereas the median OS for nonsurgical patients reached a notable 27.2 months. Similarly, 22.5% of patients (9/40) in the present study successfully converted to radical surgery, consistent with our earlier findings. These results suggest that combining HAIC with antiangiogenic agents and ICIs provides notable conversion efficiency and favorable therapeutic activity in cholangiocarcinoma. In a single-arm phase II trial, Shi et al[29] observed a 22.5-month median OS alongside a 10.2-month median PFS following treatment with GEMOX combined with lenvatinib and toripalimab, indicating highly favorable clinical outcomes. However, the conversion resection rate was only 10% (3/30), which is lower than the 22.5% observed in our study. HAIC delivers chemotherapy directly to the tumor through the hepatic arterial system, maintaining high local drug concentrations and achieving superior local control. This may partly explain why HAIC-based combination regimens yield higher conversion resection rates.

While AEs of any grade occurred in the entire cohort, the rate of severe AEs remained consistent with prior literature[30,31]. It is widely acknowledged that leukopenia is a standard complication associated with chemotherapeutic agents[32]. Here, 30% (12/40) of patients developed leukocytopenia of any grade, with only 2.5% (1/40) experiencing grade ≥ 3 events. The durvalumab-GC combination evaluated in the TOPAZ-1 cohort resulted in a 75.7% rate of high-grade toxicities[33]. In contrast, the grade 3-4 AE incidence in our study was only 30% (12/40), substantially lower than that observed in TOPAZ-1. Several factors may account for this difference. Unlike systemic chemotherapy, HAIC delivers chemotherapeutic agents directly into the hepatic artery, achieving higher intratumoral concentrations while limiting extrahepatic exposure, thereby reducing systemic toxicity and preserving tolerability. Furthermore, this analysis was confined exclusively to ICC cases, avoiding the higher inherent complication risks associated with the extrahepatic and gallbladder malignancies evaluated in the TOPAZ-1 population. Finally, we used the FOLFOX regimen rather than the GC platform; these distinct cytotoxic foundations play a crucial role in shaping the differing toxicity patterns.

Our findings should be interpreted in the context of certain methodological constraints. First, the retrospective, single-institution design carries the risk of unmeasured selection bias, which may limit the generalizability of our results. Second, due to the limited sample size, we were unable to perform formal subgroup analyses to explore variations in efficacy and toxicity across the different anti-PD-L1 therapies used (durvalumab, adebrelimab, and envafolimab). Finally, although the follow-up period was extended, it remained relatively short. Consequently, larger-scale, multicenter, prospective, randomized controlled trials are required to confirm our findings in a wider population and to assess long-term survival benefits.

CONCLUSION

Overall, the combination of FOLFOX-HAIC, surufatinib, and anti-PD-L1 therapy demonstrated substantial clinical efficacy alongside a favorable safety profile among patients with previously untreated and unresectable ICC. This therapeutic strategy could represent an alternative treatment option for selected patients; however, large-scale, prospective, randomized clinical trials are necessary to confirm these findings.

ACKNOWLEDGEMENTS

The authors acknowledge and express their deepest gratitude to the participants of this research.