Hepatic Arterial Infusion Chemotherapy Plus Systemic Chemotherapy,Lenvatinib and PD-(L)1 Inhibitors Versus Systemic Chemotherapy Plus PD-(L)1 Inhibitors in Advanced Intrahepatic Cholangiocarcinoma

Abstract

Introduction

The prognosis for unresectable intrahepatic cholangiocarcinoma (ICC) remains poor. This retrospective study aims to evaluate the efficacy and safety of HAIC combined therapy in advanced ICC.

Methods

Between April 2021 and July 2024, patients with advanced ICC treated with either HAIC of oxaliplatin plus systemic chemotherapy of gemcitabine, lenvatinib and PD-(L)1 inhibitors (HSLP) or systemic chemotherapy plus PD-(L)1 inhibitors (SP) were consecutively enrolled. Propensity score matching (PSM) was used to minimize selection bias. Primary endpoint was objective response rate (ORR) and secondary endpoints included overall survival (OS), progression-free survival (PFS) and safety.

Results

A total of 110 patients were recruited in this study with 53 in HSLP group and 57 in SP group. Prior to PSM, ORR was significantly higher (47.2% vs. 24.6%; P = 0.023) and both median OS and PFS were prolonged in HSLP group compared to SP group (OS: 15.2 vs. 12.3 months, P = 0.002; PFS: 11.3 vs. 7.3 months, P = 0.004). After 1:1 PSM, outcomes remained superior in the HSLP group, with a significantly higher ORR (43.2% vs. 18.9%; P=0.032) and longer median OS (15.4 vs. 11.8 months; P=0.018) and PFS (11.9 vs. 6.9 months; P=0.019). Multivariate analysis identified the treatment regimen as an independent prognostic factor for OS and PFS. All events were manageable with supportive care, and no additional toxicities were observed in the HSLP group.

Conclusion

HAIC plus systemic chemotherapy, lenvatinib and PD-(L)1 inhibitors represents a promising therapeutic regimen for advanced ICC patients, with tolerable toxicity.

Keywords

unresectable intrahepatic cholangiocarcinoma hepatic arterial infusion chemotherapy targeted therapy immune checkpoint inhibitor systemic chemotherapy

Introduction

As the second most common primary liver cancer, intrahepatic cholangiocarcinoma (ICC) accounts for 10–20% of primary hepatic malignancies, with a rising global incidence.^1,2 Risk factors for ICC include primary sclerosing cholangitis (PSC), parasitic infection, viral hepatitis and non-alcoholic fatty liver disease (NAFLD).³ Most patients are diagnosed at an advanced stage, and untreated median survival is less than 5 months.⁴ Although surgical resection offers the only potential cure, only about 22% of patients are eligible, and recurrence rates remain high.^5-7 For unresectable disease, systemic chemotherapy constitutes the primary therapeutic regimen.

The phase III ABC-02 trial has established gemcitabine and cisplatin (GemCis) as the standard first-line regimen for advanced biliary tract cancer,⁸ with a median overall survival (OS) of 11.7 months and gemcitabine plus oxaliplatin (GemOx) shows comparable efficacy.⁹ For platinum-refractory disease, the ABC-06 study confirmed the efficacy of second-line FOLFOX chemotherapy.¹⁰ Current guidelines recommend PD-(L)1 inhibitors combined with chemotherapy as first-line therapy for advanced ICC, with acceptable safety and survival benefit (median OS 12.7 months, ORR 26.7%) in clinical trials.^10-12 Nevertheless, the modest response rates and limited options after progression highlight an unmet clinical need, driving the search for more effective strategies.

Loco-regional therapies, such as hepatic arterial infusion chemotherapy (HAIC) and transcatheter arterial chemoembolization (TACE), provide potent local control for liver-confined disease. Emerging evidence suggests loco-regional therapies achieve favorable objective response rates in locally advanced ICC compared with standard systemic chemotherapy.^13,14 By continuously delivering chemotherapy drugs (e.g., oxaliplatin, fluorouracil) via arterial catheter, HAIC achieves higher intratumoral drug concentration compared with systemic chemotherapy while decreasing extrahepatic toxicity. A prior comparative study demonstrated significantly prolonged mOS in the HAIC group (GemCis-HAIC plus 5-Fu) versus systemic chemotherapy alone,¹⁵ supporting HAIC as a compelling first-line option that may increase conversion surgery rates and survival. Mechanistically, HAIC can induce immunogenic cell death (ICD), releasing tumor-associated antigens and damage-associated molecular patterns (DAMPs), which promote the activation and maturation of antigen-presenting cells and enhance tumor antigen presentation, thereby improve tumor-specific T cell responses. Lenvatinib, through inhibition of VEGFR and FGFR, promotes vascular normalization and alleviates the immunosuppressive tumor microenvironment. These mechanisms may synergize with PD-(L)1 blockade to enhance antitumor immunity and improve clinical outcomes.

Concurrently, targeted therapy has reshaped the treatment landscape. Lenvatinib, a multi-target tyrosine kinase inhibitor (TKI) mainly targeting vascular endothelial growth factor receptor (VEGFR) and fibroblast growth factor receptor (FGFR), exhibits synergistic effects when combined with PD-(L)1 inhibitors,^16,17 showing promising efficacy in advanced ICC.^18,19 Building on this, triple-combination therapy (chemotherapy + lenvatinib + PD-(L)1 inhibitor) has achieved ORRs up to 80% and median OS of 22.5 months in phase II trials.²⁰ Integration of HAIC in the combinations has been explored, with reported ORRs ranging from 43% to 65.2% for regimens combining HAIC, lenvatinib, and PD-(L)1 inhibitors.^21-26

Integration of HAIC and systemic chemotherapy may potentially enhance tumor response rates and survival outcomes. However, evidence supporting this approach remains limited to date. A retrospective analysis and a subsequent phase II trial preliminary suggest encouraging clinical efficacy and safety for the combined regimen,^27,28 with the phase II cohort reporting an ORR of 58% and a median OS of 30.5 months. Comparable outcomes were observed in another multicenter phase II study published in 2024 ASCO Annual Meeting (ORR: 54%; median OS: 22.1 months).²⁹ Another small-scale study reported favorable outcomes with lenvatinib plus PD-1 inhibitors in patients with tumors >5 cm.³⁰ The superiority of these intensive regimens requires validation in comparative studies.

Therefore, we conducted this retrospective cohort study to directly compare the efficacy and safety of HAIC plus systemic chemotherapy, lenvatinib, and PD-(L)1 inhibitors against systemic chemotherapy plus PD-(L)1 inhibitors in patients with advanced intrahepatic cholangiocarcinoma.

Materials and Methods

Study Design and Patients

This retrospective study consecutively enrolled patients with histologically confirmed intrahepatic cholangiocarcinoma who were treated with either the oxaliplatin-HAIC plus systemic chemotherapy of gemcitabine, lenvatinib and immunotherapy, or systemic chemotherapy plus immunotherapy between April 2021 and July 2024. Eligible criteria were as follows: age ≥18 years; histologically proven ICC with locally advanced or metastatic disease deemed unresectable by multidisciplinary evaluation; at least one measurable disease according to the Response Evaluation Criteria in Solid Tumors(RECIST v1.1); Child-Pugh class A or B; ECOG performance status 0–1; adequate bone marrow function(white blood cell count ≥2,000cells/mm³, hemoglobin > 9.0 g/dL, Platelet count ≥75,000/mm³); adequate hepatic and renal function (albumin ≥ 30 g/L, total bilirubin < 3 times the upper limit of normal, aspartate and alanine transaminases < 5 times the upper limit of normal, serum creatinine of ≤ 1.5 times the upper limit of the normal) and coagulation function(PT < 18s, INR < 1.26); written informed consent.

Patients meeting the criteria followed were excluded:life expectancy less than 2 months; concurrent other malignancy; prior anticancer therapy (including chemotherapy, radiotherapy, surgery, or interventional treatments); known allergy to chemotherapy drugs, lenvatinib or PD-(L)1 inhibitors; patients with uncontrolled comorbidities, active infections; pregnancy or lactation; refuse to comply with study and/or follow-up procedures; incomplete clinical or follow-up data; gastrointestinal bleeding of any grade within 4 weeks prior to the integrated treatment.

This study was conducted in accordance with the principles of Declaration of Helsinki and was approved by the Ethics Committee. Written informed consent was obtained from all participants prior to treatment. All patient data were de-identified prior to analysis to ensure that no individual could be identified in any way. The reporting of this study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.³¹

Treatment Protocol

Eligible patients in HSLP group received oxaliplatin 85mg/m² via HAIC pump for 3 hours on day 1, plus gemcitabine 1g/m² intravenously on day 1 and day 8 of a 3 weeks cycle. PD-(L)1 inhibitors (including toripalimab, tislelizumab or sintilimab) were administered intravenously every 3 weeks at doses specified in their respective prescribing information. The selection of these PD-(L)1 inhibitors was based on real-world clinical practice and drug accessibility in China during the study period, reflecting the routine treatment landscape. Lenvatinib was administered orally once daily depending on weight (8 mg for body weight < 60 kg; 12 mg for ≥ 60 kg). Hepatic artery infusion catheter placement was performed by experienced doctors. Following femoral artery cannulation using seldinger technique, the infusion catheter was placed into primary tumor-feeding artery via arterial angiography.³² The decision to discontinue HAIC was based on an intraprocedural angiographic assessment during the scheduled intervention. Patients in the HSLP group were transitioned to systemic GemOx chemotherapy combined with maintaining targeted and immunotherapy if radiographic assessment by surgery indicated an absence of significant tumor blood supply. This protocol was designed to mitigate potential overtreatment and hepatotoxicity. Patients in SP group received gemcitabine 1g/m² on day 1 and 8 and oxaliplatin 85mg/m² intravenously on day 1 every 3 weeks. PD-(L)1 inhibitors were administered at identical doses as those in the HSLP group.

The HSLP regimen was administered for a maximum of six cycles of HAIC combined with systemic chemotherapy, followed by maintenance therapy with targeted therapy and immunotherapy. The SP regimen was administered for a maximum of eight cycles of systemic chemotherapy, followed by immunotherapy alone. Dose reduction or discontinuation of chemotherapy drug due to adverse events were implemented in accordance with relative literature and established guidelines.^8,33

Data Collection

Demographic data, clinical characteristics, laboratory and radiology data were collected from medical system. Pretreatment tumor burden parameters included: maximum tumor diameter, tumor number (single or multiple), presence or absence of vascular invasion, lymph node metastasis, and distant organ metastasis. All patient data were de-identified prior to analysis to protect patient confidentiality.

Patients received contrast-enhanced CT or MRI scans of chest and abdomen every two treatment courses and clinical tumor response was assessed according to RECIST v1.1 criteria. All images were independently reviewed by two radiologists who were blinded to treatment allocation and clinical outcomes. Discrepancies resolved by consensus or adjudication by a third senior radiologist. For patients who missed a scheduled scan, the following rules were applied: if a subsequent scan was available within a reasonable timeframe, it was used for response assessment; if no follow-up imaging was available, the patient was classified as having progressive disease.

Objective response rate (ORR) was the primary endpoint, and was defined as the proportion of patients of completely response (CR) or partial response (PR) according to RECISTv1.1 criteria. Secondary endpoints include OS, which was defined as the time from initiation of therapy to the last follow-up or death caused by any reason, progression free survival (PFS, defined as the interval from initiation of therapy to disease progression, the last follow-up or death), disease control rate (DCR, defined as the proportion of patients of CR, PR or stable disease (SD)), time to response (TTR, defined as the time from initiation of therapy to the first documented objective tumor response) and duration of response (DOR, defined as the time from initiation of objective response to progression or death due to any reason). Treatment-related adverse events and toxicity were graded according to the National Cancer Institute Common Toxicity Criteria version 5.0.

Statistics Analysis

Normally distributed continuous variables were compared using Student's t test, while non-normally distributed variables were analyzed with the Mann-Whitney U test. Categorical variables were expressed as number (percentage) and were analyzed by continuity-corrected chi-squared test, or Fisher’s exact test, as propriate. Continuous variables were summarized as medians with ranges. Survival outcomes (OS, PFS) were compared with Kaplan–Meier curves and log-rank tests. Prognostic factors were identified using univariable and multivariable Cox models.

To minimize confounding, we performed 1:1 propensity score matching (PSM) using nearest-neighbor matching without replacement, with a caliper of 0.2 times the standard deviation of the logit of the propensity score. Propensity scores were derived from a logistic regression model that included tumor burden (TNM stage, tumor diameter, tumor number, vascular invasion, lymph node metastasis, extrahepatic metastasis), age, gender, ECOG performance status, etiology, Child-Pugh score, and key laboratory values (ALT, AST, ALB, TBIL, CA-199, CEA). The selected variables were clinical measurable and associated with patient prognosis, Balance was assessed using standardized mean differences (SMD <0.1 indicating adequate balance). P values are also provided for reference. A Love plot displays the SMD for each covariate, all post matching SMDs are below the conventional threshold of 0.1.

To assess the robustness of our findings to potential confounding, we performed inverse probability of treatment weighting (IPTW) as a sensitivity analysis. Propensity scores were estimated using a logistic regression model including the same covariates as those used in the PSM. Weights were calculated as the inverse probability of the actual treatment received and were trimmed to mitigate the influence of extreme weights. Weighted Cox proportional hazards models were used to estimate the treatment effect on OS and PFS, and weighted logistic regression was used for ORR. A two-tailed P-value < 0.05 was considered statistically significant. All analyses were performed using SPSS 25.0 and R 4.4.2.

Result

Baseline Characteristics

Between April 2021 and July 2024, a total of 110 consecutive patients with pathologically confirmed unresectable ICC were enrolled in this study. We performed propensity score matching to create 1:1 paired cohorts, with 37 patients were allocated to the HSLP group, and 37 to the SP group (Figure 1). The baseline characteristics of two groups are shown in Table 1 and a love plot (Supplementary Figure 1) illustrates the SMD changes of covariates before and after matching. All baseline covariates included were complete with no missing data. In the entire cohort, approximately half of the patients were diagnosed with stage III disease and most patients presented with multiple intrahepatic tumors and experienced lymph node metastasis. Prior to PSM, patients in the HSLP group had significantly larger tumor diameters (SMD = 0.381, p = 0.041) and a higher incidence of vascular invasion (SMD = 0.444, p = 0.037) compared to the SP group, with most covariates exhibiting an SMD greater than 0.1. After matching, the baseline characteristics were well-balanced between the two groups, with most key tumor-related parameters achieving an SMD below 0.1. Minor residual imbalances persisted in hemoglobin levels (SMD = 0.256) and CA-199 (SMD = 0.218).

Figure 1.

Flowchart for patients enrollment. Abbreviation: ICC, intrahepatic cholangiocarcinoma; HAIC, hepatic arterial infusion chemotherapy; HSLP, HAIC plus systemic chemotherapy, lenvatinib and PD-(L)1 inhibitors; SP, systemic chemotherapy plus PD-(L)1 inhibitors

Table 1.

Baseline Characteristics of Patients Before and After PSM

Characteristics	Before PSM				After PSM
Characteristics	SP group (N=57)	HSLP group (N=53)	SMD	P-value	SP group (N=37)	HSLP group (N=37)	SMD	P-value
Age (n, %)			0.105	0.731			0.058	1.000
≥60, years	20 (35.1%)	16 (30.2%)			12 (32.4%)	11 (29.7%)
<60, years	37 (64.9%)	37 (69.8%)			25 (67.6%)	26 (70.3%)
Sex (n, %)			0.192	0.418			<0.001	1.000
Female	28 (49.1%)	21 (39.6%)			18 (48.6%)	18 (48.6%)
Male	29 (50.9%)	32 (60.4%)			19 (51.4%)	19 (51.4%)
TNM stage (n, %)			0.034	1.000			0.055	1.000
III	30 (52.6%)	27 (50.9%)			16 (43.2%)	15 (40.5%)
IV	27 (47.4%)	26 (49.1%)			21 (56.8%)	22 (59.5%)
Tumor Diameter, mm (median, [IQR])	80.0 [62.0;104]	88.0 [76.0;114]	0.381	0.041	80.0 [70.0;104]	87.0 [72.0;113]	0.008	0.585
Tumor Number (n, %)			0.084	0.828			0.062	1.000
single	13 (22.8%)	14 (26.4%)			27 (73.0%)	28 (75.7%)
multiple	44 (77.2%)	39 (73.6%)			10 (27.0%)	9 (24.3%)
Vascular Invasion (n, %)			0.444	0.037			0.108	0.816
Absent	36 (63.2%)	22 (41.5%)			19 (51.4%)	17 (45.9%)
Present	21 (36.8%)	31 (58.5%)			18 (48.6%)	20 (54.1%)
Lymph Node Metastasis (n, %)			0.188	0.479			0.083	1.000
Absent	6 (10.5%)	9 (17.0%)			4 (10.8%)	5 (13.5%)
Present	51 (89.5%)	44 (83.0%)			33 (89.2%)	32 (86.5%
Extrahepatic Metastasis (n, %)			0.034	1.000			0.055	1.000
Absent	30 (52.6%)	27 (50.9%)			16 (43.2%)	15 (40.5%)
Present	27 (47.4%)	26 (49.1%)			21 (56.8%)	22 (59.5%)
Child-Pugh Score (n, %)			0.131	0.686			<0.001	1.000
A	48 (84.2%)	47 (88.7%)			32 (86.5%)	32 (86.5%)
B	9 (15.8%)	6 (11.3%)			5 (13.5%)	5 (13.5%)
ECOG (n,%)			0.076	0.832			0.108	0.814
0	29 (50.9%)	29 (54.7%)			22 (59.5%)	20 (54.1%)
1	28 (49.1%)	24 (45.3%)			15 (40.5%)	17 (45.9%)
Etiology (n,%)			0.207	0.379			0.056	1.000
HBV	17 (29.8%)	21 (39.6%)			13 (35.1%)	14 (37.8%)
Others	40 (70.2%)	32 (60.4%)			24 (64.9%)	23 (62.2%)
WBC, x10⁹/L (median, [IQR])	8.79 [8.04;9.33]	8.31 [7.49;9.21]	0.067	0.601	8.84 [8.32;9.30]	8.27 [7.32;9.07]	0.060	0.272
PLT, x10⁹/L (median, [IQR])	279 [240;365]	275 [234;325]	0.097	0.540	270 [201;328]	284 [235;325]	0.073	0.496
Hb, g/L(median, [IQR])	129 [124;140]	135 [123;144]	0.121	0.474	128 [123;138]	130 [123;145]	0.256	0.338
ALT, U/L(median, [IQR])	31.6 [15.2;57.8]	24.6 [17.3;57.3]	0.195	0.839	23.3 [15.2;43.5]	22.6 [16.7;51.6]	0.163	0.596
AST, U/L(median, [IQR])	36.5 [23.0;59.1]	34.8 [22.8;61.9]	0.058	0.715	34.0 [23.0;40.5]	29.7 [21.5;53.1]	0.104	0.893
ALB, g/L(median, [IQR])	41.9 [40.9;44.1]	43.3 [41.1;45.1]	0.022	0.365	42.2 [40.7;44.1]	43.0 [41.3;45.1]	0.118	0.364
TBIL, umol/L(median, [IQR])	9.80 [7.82;14.7]	9.20 [6.78;15.2]	0.176	0.407	9.10 [7.30;12.1]	9.40 [7.00;14.8]	0.045	0.918
CA199 (n,%)			0.107	0.713			0.218	0.485
<100, U/mL	31 (54.4%)	26 (49.1%)			20 (54.1%)	16 (43.2%)
≥100, U/mL	26 (45.6%)	27 (50.9%)			17 (45.9%)	21 (56.8%)
CEA (n, %)			0.173	0.482			0.056	1.000
<5, ng/mL	34 (59.6%)	36 (67.9%)			23 (62.2%)	24 (64.9%)
≥5, ng/mL	23 (40.4%)	17 (32.1%)			14 (37.8%)	13 (35.1%)

Values are presented as the median (IQR) or n (%).

Abbreviations: PSM, propensity score matching; SMD, standard mean difference; HAIC, hepatic arterial infusion chemotherapy; ECOG, Eastern Cooperative Oncology Group; WBC, white blood cell; PLT, platelet count; Hb, hemoglobin; ALT, alanine transaminase; AST, aspartate transaminase; ALB, albumin; TBIL, total bilirubin; CA19–9, carbohydrate antigen 19–9; CEA, carcinoembryonic antigen; CRE, creatinine; TNM, tumor–node–metastasis.

Treatment and Efficacy

The median treatment cycle was 4.0 (IQR 3.0-6.0) in the HSLP group and 5.0 (IQR 4.0-6.0) in the SP group. In the HSLP group, patients received up to six cycles of HAIC combined with systemic chemotherapy, followed by maintenance therapy with lenvatinib and PD-(L)1 inhibitor. In the SP group, patients received up to eight cycles of systemic GemOx chemotherapy, followed by PD-(L)1 inhibitor monotherapy. Post-progression regimens in the SP group versus the HSLP group was as follows: FOLFIRINOX (8 vs. 3 patients), nab-paclitaxel plus S-1 (5 vs. 1), FOLFOX (4 vs. 3), and radiotherapy/chemoradiotherapy (2 vs. 3). Other regimens, including NALIRIFOX, trastuzumab plus pertuzumab, and capecitabine plus anlotinib, were administered to only a few patients.

Patients underwent periodic contrast-enhanced CT/MR scans every two treatment cycles with tumor response evaluated according to RECIST v1.1 criteria (Table 2). The HSLP regimen showed superior efficacy both before and after matching. In the matched cohort, the HSLP group achieved a significantly higher objective response rate (ORR: 43.2% vs. 18.9%; conditional logistic regression: OR=4.00, 95%CI 1.13–14.17, P=0.032). The disease control rate (DCR) was numerically higher in the HSLP group compared with the SP group (86.5% vs. 67.6%), but the difference did not reach statistical significance (conditional logistic regression P=0.097). These findings were consistent with the pre-matched cohort, where the HSLP group also demonstrated superior ORR (47.2% vs. 24.6%; P=0.023; adjusted OR= 2.78, 95%CI: 1.14-7.14) and DCR (88.7% vs. 70.2%; P=0.032), despite the higher tumor burden. Inverse probability of treatment weighting (IPTW) was performed as a sensitivity analysis. After weighting, all covariates achieved good balance with standardized mean differences <0.1. Weighted logistic regression confirmed that the ORR remained significantly higher in the HSLP group compared with the SP group (OR 2.483, 95% CI 1.010–6.101, P=0.047), consistent with the primary analysis.

Table 2.

Tumor Response Evaluated According to RECIST v1.1 Criteria

Tumor response	Primary cohort			Psm cohort
Tumor response	HSLP group (n=53)	SP group (n=57)	P-value	HSLP group (n=37)	SP group (n=37)	P-value
CR	0	0	-	0	0	-
PR	25	14	-	16	7	-
SD	22	26	-	16	18	-
PD	6	17	-	5	12	-
ORR	47.20%	24.60%	0.023	43.2%	18.9%	0.032
DCR	88.7%	70.2%	0.032	86.5%	67.6%	0.097

Abbreviations: CR, complete response; PR, partial response; SD, stable disease; PD, progressive disease; ORR, objective response rate; DCR, disease control rate.

In the primary cohort, the HSLP regimen showed longer DOR than SP group (median DOR: 12.0 vs 7.8 months; P = 0.048), with a comparable TTR (3.3 vs 3.8 months; P = 0.163). This trend persisted in the matched cohort, though statistical significance was attenuated likely due to reduced sample size (Supplementary Figure 2; Supplementary Table 3). Notably, two patients in the HSLP group achieved sufficient downstaging to undergo conversion surgery, an outcome not observed in the SP group.

The median follow-up time was 24.0 months (95% CI: 21.6 months to not reached) for the entire cohort. During follow-up, 90 patients died, and majority of patients experienced disease progression. Patients in the HSLP group had prolonged overall survival and progression-free survival than those in the SP group (Figure 2). In the matched cohort, the median OS was 15. 4 (95% CI: 13.8-16.9) versus 11.8 months (95% CI: 9.5-14.1) for the HSLP and SP groups, respectively (P=0.018). The median PFS was 11.9 months (95% CI: 9.7-14.1) versus 6.9 months (95% CI: 5.6-8.2) (P=0.019). Specifically, a consistent survival advantage was observed in the pre-matched cohort, with the mOS of 15.2 months (95% CI: 13.9-16.5) versus 12.3 months (95% CI: 11.0-13.5) (P=0.002), and mPFS was 11.3months (95% CI: 9.8-12.8) vs. 7.3 months (95% CI: 6.5-8.1) (P=0.004). Following IPTW weighting, survival curve analysis similarly demonstrated improved OS (weighted log-rank P=0.0065) and PFS (weighted log-rank P=0.0077) in the HSLP group versus the SP group (Supplementary Figure 3).

Figure 2.

Kaplan-Meier curves for overall survival (OS) and progression-free survival (PFS) before (A and B) and after (C and D) propensity score matching (PSM) between the two cohorts

Exploratory subgroup analyses were performed to evaluate the consistency of treatment effects in the matched cohort (Figure 3). HSLP regimen was associated with improved OS in several subgroups, including patients with non-metastatic disease (stage III; HR 0.364, 95% CI 0.15–0.88), tumor diameter <100 mm (HR 0.340, 95% CI 0.16–0.71), Child-Pugh A (HR 0.455, 95% CI 0.24–0.85), and ECOG 0 (HR 0.460, 95% CI 0.22–0.95). Interaction tests revealed significant heterogeneity in treatment effects for OS by Child-Pugh grade (interaction P=0.025) and CA19-9 level (interaction P=0.032). For PFS, significant interactions were observed for TNM stage (interaction P=0.036), Child-Pugh grade (interaction P=0.013), and CA19-9 level (interaction P=0.029). Interaction tests for other variables did not reach statistical significance, suggesting benefit of the HSLP regimen was consistent across subgroups. Nevertheless, the lack of statistical significance may due to limited subgroup sample sizes and observed trends warrant further investigation. Similar patterns were observed in the pre-PSM cohort (Supplementary Figure 4).

Figure 3.

Forest plots of survival (A) and progression-free survival (B) in different subgroups of PSM cohort

Univariate and Multivariate Cox Regression Analyses

Prognostic analysis of the matched cohort is summarized in Table 3. In the univariate Cox analysis, assignment to the HSLP group was associated with significantly longer OS and PFS. Conversely, presence of vascular invasion, Child-Pugh score B, and an ECOG PS score 1 were significant risk factors for poorer survival, and Child-Pugh score B and ECOG PS score of 1 were associated with shorter PFS. To account for the matched design, we performed Cox proportional hazards models with robust standard errors clustered on matched pairs for survival outcomes. Univariable analysis accounting for the matched design showed that the HSLP group was associated with significantly improved OS (HR 0.519, 95% CI 0.309–0.872, P=0.013) and PFS (HR 0.554, 95% CI 0.345–0.887, P=0.014) compared with the SP group. In the analysis of all patients, treatment regimen was also identified as an independent prognostic factor for both OS and PFS (Supplementary Table 1).

Table 3.

Univariate and Multivariate Analyses of Prognostic Factors Associated With Overall Survival and Progression-Free Survival for PSM Cohort

	Overall survival						Progression-free survival
Characteristics	Univariate analysis			Multivariate analysis			Univariate analysis			Multivariate analysis
Characteristics	HR	95%CI	p value	HR	95%CI	p value	HR	95%CI	p value	HR	95%CI	p value
Treatment (HSLP/SP)	0.519	0.300-0.900	0.019	0.373	0.199-0.700	0.002	0.554	0.335-0.914	0.021	0.353	0.198-0.630	<0.001
Age (≥60/<60,years)	0.600	0.35-1.031	0.064				0.577	0.345-0.967	0.037	0.677	0.386-1.189	0.175
Sex (male/female)	1.000	0.601-1.664	0.999				1.169	0.717-1.906	0.531
TNM stage (IV/III)	1.311	0.782-2.198	0.305				1.123	0.682-1.847	0.649
Tumor Diameter (≥100/<100,mm)	1.151	0.656-2.02	0.623				1.045	0.609-1.792	0.873
Tumor Number (single/multiple)	0.609	0.33-1.122	0.112				0.654	0.364-1.177	0.157
Vascular Invasion (Present/Absent)	1.727	1.027-2.906	0.039	1.762	0.998-3.109	0.051	1.568	0.954-2.577	0.076
Lymph Node Metastasis (Present/Absent)	1.168	0.56-2.436	0.678				0.848	0.418-1.723	0.649
Extrahepatic Metastasis (Present/Absent)	1.311	0.782-2.198	0.305				1.123	0.682-1.847	0.649
Child-Pugh Score (B/A)	2.734	1.239-6.035	0.013	1.996	0.854-4.664	0.110	3.986	1.9-8.363	<0.001	4.053	1.876-8.758	<0.001
ECOG PS Score (1/0)	1.883	1.119-3.171	0.017	2.138	1.23-3.718	0.007	2.278	1.374-3.777	0.001	2.253	1.265-4.011	0.006
Etiology (Others/HBV)	1.399	0.825-2.372	0.213				1.480	0.88-2.487	0.139
WBC (≥10/<10 ×10^9/L)	2.355	0.989-5.605	0.053				1.798	0.759-4.257	0.182
PLT (≥300/<300 ×10^9/L)	0.710	0.405-1.245	0.232				0.676	0.393-1.163	0.157
Hb (≥120/<120, g/L)	0.762	0.393-1.477	0.421				0.709	0.364-1.383	0.313
ALT (≥40/<40,U/L)	1.059	0.605-1.852	0.841				0.918	0.536-1.573	0.757
AST (≥40/<40,U/L)	1.702	0.989-2.929	0.055				1.538	0.904-2.616	0.112
ALB (≥40/<40,g/L)	1.288	0.551-3.01	0.559				1.851	0.796-4.307	0.153
TBIL (≥17.1/<17.1,umol/L)	0.864	0.418-1.786	0.694				0.999	0.49-2.039	0.999
CA19-9 (≥100/<100,U/mL)	1.188	0.709-1.992	0.513				1.109	0.679-1.809	0.680
CEA (≥5/<5,ng/mL)	1.165	0.695-1.956	0.562				1.067	0.646-1.763	0.800

To further control confounding, the primary multivariable Cox proportional hazards models adjusted for clinically relevant potential confounders (e.g., age, gender, tumor burden), identified a significant survival benefit associated with the HSLP group (Table 3). The adjusted hazard ratios for the HSLP regimen were 0.373 (95% CI, 0.199-0.700; p=0.002) for OS and 0.353 (95% CI, 0.198-0.630; p<0.001) for PFS. After adjusting for variables with P < 0.05 of univariable analysis in Cox models with robust standard errors clustered on matched pairs, the HSLP regimen remained independently associated with improved OS (HR 0.417, 95% CI 0.237–0.734, P=0.002) and PFS (HR 0.424, 95% CI 0.274–0.655, P<0.001). To further evaluate the robustness of the findings, IPTW-weighted multivariable Cox regression was performed. The results showed that the HSLP regimen remained independently associated with improved OS (HR 0.520, 95% CI 0.315–0.860, P=0.011) and PFS (HR 0.534, 95% CI 0.334–0.854, P=0.009), consistent with both the pre- and post-PSM analyses, confirming the robustness of the treatment effect.

Safety

Treatment-related adverse events (TRAEs) were observed in the majority of patients, but no treatment-related deaths occurred (Table 4 and Supplementary Table 2 for primary cohort). In the matched cohort, seven patients in the SP group required treatment modification (two dose reductions, five discontinuations), primarily due to myelosuppression. In the HSLP group, five patients suspended treatment due to hepatic dysfunction (elevated ALT/AST, refractory ascites, or jaundice), two discontinued due to gastrointestinal hemorrhage, and three required dose reduction. Two other patients discontinued due to poor general condition, and one due to immune-mediated dermatitis.

Table 4.

Treatment Related Adverse Events (TRAEs) Based on Common Terminology Criteria for Adverse Events (CTCAE) Version 5.0

Adverse events	Any grade AEs (%)		p-value	Grade 3-4 AEs (%)		p-value
Adverse events	HSLP group (n=37)	SP group (n=37)	p-value	HSLP group (n=37)	SP group (n=37)	p-value
Neutropenia	23 (62.2%)	16 (43.2%)	0.162	11 (29.7%)	6 (16.2%)	0.269
Hypoalbuminemia	18 (48.6%)	6 (16.2%)	0.006	0 (0%)	0 (0%)	-
Elevated AST	13 (35.1%)	10 (27.0%)	0.615	9 (24.3%)	4 (10.8%)	0.222
Abdominal pain	13 (35.1%)	13 (35.1%)	1.000	0 (0%)	0 (0%)	-
Nausea	13 (35.1%)	18 (48.6%)	0.346	0 (0%)	0 (0%)	-
Elevated ALT	11 (29.7%)	4 (10.8%)	0.083	3 (8.11%)	3 (8.11%)	1.000
Anemia	11 (29.7%)	24 (64.9%)	0.005	2 (5.41%)	6 (16.2%)	0.261
Thrombocytopenia	8 (21.6%)	12 (32.4%)	0.432	4 (10.8%)	3 (8.11%)	1.000
Elevated Bilirubin	10 (27.0%)	8 (21.6%)	0.786	5 (13.5%)	2 (5.41%)	0.430
Elevated GGT	9 (24.3%)	8 (21.6%)	1.000	5 (13.5%)	2 (5.41%)	0.430
Abdominal distension	8 (21.6%)	7 (18.9%)	1.000	0 (0%)	0 (0%)	-
Rash	8 (21.6%)	6 (16.2%)	0.767	0 (0%)	0 (0%)	-
Vomiting	5 (13.5%)	2 (5.41%)	0.430	0 (0%)	0 (0%)	-
Fatigue	3 (8.11%)	5 (13.5%)	0.711	0 (0%)	0 (0%)	-
Diarrhea	3 (8.11%)	3 (8.11%)	1.000	0 (0%)	0 (0%)	-
Elevated creatinine	2 (5.41%)	1 (2.70%)	1.000	0 (0%)	0 (0%)	-
Hand-foot Syndrome	2 (5.41%)	0 (0.00%)	0.493	0 (0%)	0 (0%)	-
Gastrointestinal Hemorrhage	2 (5.41%)	0 (0.00%)	0.493	0 (0%)	0 (0%)	-
Fever	1 (2.70%)	1 (2.70%)	1.000	0 (0%)	0 (0%)	-
Immune-related Adverse Events	1 (2.70%)	2 (5.41%)	0.493	0 (0%)	0 (0%)	-

Abbreviations: AEs, adverse events; AST, Aspartate aminotransferase; ALT, Alanine aminotransferase; GGT, γ-Glutamyl transferase.

Common TRAEs included neutropenia, hypoalbuminemia, elevated AST/ALT, abdominal pain, nausea, anemia, and thrombocytopenia. Among any-grade TRAEs, neutropenia was the most frequent any-grade TRAE in the HSLP group (62.2%), while anemia predominated in the SP group (64.9%). The HSLP group had significantly higher rates of hypoalbuminemia (48.6% vs. 16.2%; p=0.006) and a trend toward more frequent ALT elevation (29.7% vs. 10.8%; p=0.083). Anemia was more common in the SP group (64.9% vs. 29.7%; p=0.005). Grade 3-4 TRAEs were comparable between groups, including neutropenia, elevated AST/ALT level, anemia, thrombocytopenia, and hyperbilirubinemia. All adverse events were manageable by symptomatic treatment, dose modification, or treatment suspension.

Discussion

The aggressive nature of ICC often precludes curative resection at diagnosis. The current standard treatment—platinum-based systemic chemotherapy combined with immune checkpoint inhibitors—still yields unsatisfactory outcomes. Although clinical trials for biliary tract cancers typically enroll all subtypes, the marked heterogeneity of cholangiocarcinoma necessitates subtype-specific therapeutic strategies. To our knowledge, this is the first comparative study evaluating the regimen of HAIC of oxaliplatin plus systemic chemotherapy of oxaliplatin, lenvatinib and PD-(L)1 inhibitors versus standard therapy in advanced unresectable ICC.

Accumulating studies on targeted and loco-regional therapy for ICC show have shown inspiring efficacy. Notably, relative study indicates ICC patients are prone to develop locally progressive or multifocal disease that limits radical surgery, while HAIC may offer superior tumor control compared to systemic treatment alone.^34,35 This is corroborated by the post-analysis of ABC trial, revealing limited survival benefit from systemic chemotherapy of GemCis in locally advanced disease.³⁶ In this retrospective analysis of 110 patients, 37 matched pairs were established via 1:1 PSM. The HSLP group showed a superior ORR compared to the SP group, both in the primary cohort (47.2% vs. 24.6%, P=0.023) and the matched cohort (43.2% vs. 18.9%, P=0.045). Notable tumor control was still achieved with HSLP therapy, even though the HSLP group presented with a greater baseline tumor burden before matching. A significant PFS benefit was also observed in the HSLP group, with a median PFS of 11.9 months versus 6.9 months in the SP group (P=0.019). The marked tumor shrinkage, reflected in the superior ORR, suggests that effective control of intrahepatic disease, which is a common site of progression in ICC, may be a major contributing factor to the observed PFS benefit. The efficacy of HAIC in multifocal tumors, stems from its ability to ensure homogeneous drug distribution across disseminated or micrometastatic intrahepatic lesions, independent of anatomical constraints—unlike ablation or embolization, which require macroscopic targets. This fundamental difference from other locoregional therapies may explains how HAIC exerts its comprehensive control and contributes to the PFS benefit. HAIC is well-established in liver cancer and achieves a high objective response rate by enabling continuous drug infusion directly into intrahepatic lesions, maintaining high local drug concentrations, and reducing first-pass metabolism. This results in prolonged tumor cell exposure to chemotherapy drug, enhancing cytotoxicity for all cell cycle phases. Furthermore, HAIC concurrently induces immunogenic cell death, which releases tumor-associated antigens and remodels the tumor immune microenvironment (TME),³⁷ thereby potentiating the synergistic efficacy when combined with immune checkpoint inhibitors. Collectively, these mechanisms provide a strong rationale for incorporating HAIC into combination therapy regimens to achieve superior local tumor control, as clearly evidenced by the compelling local tumor response in our clinical data.

Critically, the dismal prognosis in advanced ICC is primarily driven by intrahepatic tumor progression and its subsequent complications, including liver failure, portal hypertension, gastroesophageal variceal bleeding, thrombocytopenia, refractory ascites, and hepatic encephalopathy.³⁸ It implies that mortality is more closely tied to intrahepatic tumor burden than to distant metastases, underscoring the competing risks. Achieving superior local control, as manifested by the higher ORR and prolonged PFS with loco-regional therapy of HAIC, directly prevents or delays the progression of intrahepatic lesions and the onset of liver-related complications. This cumulative treatment effect ultimately translates into the observed overall survival benefit, as the definitive clinical benefit for patients, evidenced in the matched cohort by a median OS of 15.4 versus 11.8 months (adjusted HR: 0.373, 95% CI, 0.199-0.700; P = 0.018). Furthermore, successful conversion to R0 resection was achieved in two initially unresectable patients from the HSLP group after treatment. These patients subsequently demonstrated significant extensions in both PFS and OS, highlighting the role of the HSLP regime as a viable conversion strategy for locally advanced intrahepatic disease. The promising outcome warrants further exploration to definitively establish its translational potential.

Beyond the locoregional control offered by HAIC, the systemic component of lenvatinib also plays a critical role. While current guidelines endorse targeted therapy with ivosidenib or pemigatinib for eligible patients.^39,40 lenvatinib is not yet part of standard regimens. Nonetheless, emerging evidence indicates that lenvatinib may enhance the efficacy of immunotherapy by remodeling the immunosuppressive TME of ICC. By primarily inhibiting VEGFR and FGFR, it promotes vascular normalization, thereby disrupting the hypoxia, acidosis, and immunosuppression environment driven by pathological angiogenesis that impair anti-tumor immunity.⁴¹ It also modulates immune cell populations, reducing immunosuppressive cells and facilitating immune infiltration.^42,43 Given that systemic chemotherapy is still necessary to manage micrometastatic and extrahepatic disease in this commonly disseminated malignancy, the addition of a targeted therapy like lenvatinib may synergize to overcome resistance and improve outcomes. This mechanistic insight supports its inclusion in combination therapy, which may ultimately contribute to patient survival benefits.

Exploratory subgroup analyses suggested that patients with locally advanced disease and preserved liver function might derive particular benefit from the HSLP regimen. These observations, while hypothesis-generating, provide preliminary insights for patient selection and warrant further validation in prospective studies.

Safety analysis revealed no additional toxicity burden in adding HAIC and targeted therapy. The majority of AEs were moderate and manageable through supportive care, dose modification or treatment interruption. The most common AEs in both cohorts were systemic toxicities (neutropenia and thrombocytopenia). Notably, the HSLP group exhibits expected higher incidence of hypoalbuminemia and elevated ALT/AST level, consistent with the hepatic toxicity of HAIC compared with systemic chemotherapy. Nonetheless, the hepatic adverse events were predominantly low-grade and manageable, highlighting the importance of proactive management, including hepatoprotective agents, aggressive nutritional support, and close monitoring of liver function. Therefore, rather than absolute exclusion, compromised hepatic function should prompt intensified supportive care to enable patients to derive the pronounced survival benefit from the HSLP regimen.

Our results advance the field of HSLP therapy for ICC. Earlier studies preliminary established HAIC with systemic chemotherapy, while a recent pilot study showed the efficacy in a small, selected cohort when combined with targeted-immunotherapy.^27-30 Our study builds on this by validating a HSLP triplet regimen in a larger, comparative cohort. We confirm the superior efficacy of this approach in the broader patient population and solidify its role as a key therapeutic strategy.

This study has several limitations. First, the retrospective study carries a risk of inherent selection bias, including incomplete follow-up data, which led to the exclusion of some patients and reflects a typical limitation of real-world data. Although propensity score matching reduced measurable baseline discrepancies, residual confounding due to unmeasured factors cannot be entirely ruled out. Moreover, variation in clinical practice could have introduced confounding, as physicians might have been more inclined to employ the HSLP regimen for cases with features such as a higher burden of intrahepatic disease. Second, because the study was designed to compare the HSLP regimen against the current standard of care, the SP group did not receive lenvatinib, which limits the ability to independently assess the effect of HAIC. Theoretically, the survival benefit observed in the HSLP group may arise from both HAIC-enhanced local efficacy and the addition of lenvatinib, and their individual contributions cannot be fully disentangled. Third, rarity of ICC limits the sample size, thus larger prospective randomized phase II trials are warranted to establish high-level evidence. Furthermore, the inclusion of different classes of PD-(L)1 inhibitor may have introduced confounding bias. Although all are approved and widely used in clinical practice in China, potential differences between these agents may have introduced heterogeneity, representing an inherent limitation of this retrospective analysis.

Conclusion

In conclusion, our retrospective study provides preliminary evidence that hepatic arterial infusion chemotherapy with systemic chemotherapy, lenvatinib, and PD-(L)1 inhibitors presents clinically antitumor efficacy in advanced unresectable ICC, with marked locoregional control, and showing no additional toxicity.

Supplemental Material

Supplemental Material - Hepatic Arterial Infusion Chemotherapy Plus Systemic Chemotherapy, Lenvatinib and PD-(L)1 Inhibitors Versus Systemic Chemotherapy Plus PD-(L)1 Inhibitors in Advanced Intrahepatic Cholangiocarcinoma

Supplemental Material for Hepatic Arterial Infusion Chemotherapy Plus Systemic Chemotherapy, Lenvatinib and PD-(L)1 Inhibitors Versus Systemic Chemotherapy Plus PD-(L)1 Inhibitors in Advanced Intrahepatic Cholangiocarcinoma by Huazhong Guo, MD, Yexing Huang, MD, Na Liu, MD, Minke He, MD, Zhicheng Lai, MD, Ming Shi, MD and Qijiong Li, MD in Technology in Cancer Research & Treatment

Footnotes

Acknowledgements

The authors declare that there are no persons or institutions that need to be acknowledged for their contributions to this study.

ORCID iDs

Huazhong Guo

Qijiong Li

Ethical Considerations

This study was conducted in accordance with Declaration of Helsinki and was approved by the Ethics Committee of Sun Yat-sen University Cancer Center (B2025-625-01). Written informed consent was provided by all patients before treatment.

Author Contributions

Huazhong Guo: Conceptualization, Data curation, Methodology, Software, Formal Analysis, Investigation, Visualization, Writing of the original draft. Yexing Huang: Conceptualization, Data curation, Methodology, Software, Formal Analysis, Investigation, Visualization, Writing of the original draft. Na Liu: Conceptualization, Data curation, Formal Analysis, Investigation, Visualization, Writing of the original draft. Zhicheng Lai: Conceptualization, Data curation, Methodology, Methodology, Investigation, Writing of the original draft. Minke He: Conceptualization, Data curation, funding acquisition, Methodology, Formal Analysis, Investigation, Writing, reviewing, and editing. Ming Shi: Conceptualization, Funding acquisition, Investigation, Supervision, Validation, Writing review, and editing. Qijiong Li: Conceptualization, Data curation, funding acquisition, Methodology, Project administration, Supervision, Writing, review, and editing.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by National Natural Science Foundation of China (No. 82203126), Young Talents Program of Sun Yat-sen University Cancer Center (No. YTP-SYSUCC-0071).

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The raw data supporting the conclusion will be available on request from the authors, without undue reservation.*

Supplemental Material

Supplemental Material for this article is available online.

Appendix

References

Bray

Laversanne

Sung

, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2024;74(3):229-263. doi:10.3322/caac.21834.

Valle

Kelley

Nervi

Zhu

. Biliary tract cancer. Lancet. 2021;397(10272):428-444. doi:10.1016/S0140-6736(21)00153-7.

Petrick

Yang

Altekruse

, et al. Risk factors for intrahepatic and extrahepatic cholangiocarcinoma in the United States: A population-based study in SEER-Medicare. PLoS One. 2017;12(10):e0186643. doi:10.1371/journal.pone.0186643. Published 2017 Oct 19.

Park

Kim

, et al. Natural History and Prognostic Factors of Advanced Cholangiocarcinoma without Surgery, Chemotherapy, or Radiotherapy: A Large-Scale Observational Study. Gut Liver. 2009;3(4):298-305. doi:10.5009/gnl.2009.3.4.298.

Bertuccio

Malvezzi

Carioli

, et al. Global trends in mortality from intrahepatic and extrahepatic cholangiocarcinoma. J Hepatol. 2019;71(1):104-114. doi:10.1016/j.jhep.2019.03.013.

Gad

Saad

Faisaluddin

, et al. Epidemiology of Cholangiocarcinoma; United States Incidence and Mortality Trends. Clin Res Hepatol Gastroenterol. 2020;44(6):885-893. doi:10.1016/j.clinre.2020.03.024.

Ruff

Pawlik

. Clinical management of intrahepatic cholangiocarcinoma: surgical approaches and systemic therapies. Front Oncol. 2024;14:1321683. doi:10.3389/fonc.2024.1321683. Published 2024 Jan 24.

Valle

Wasan

Palmer

, et al. Cisplatin plus gemcitabine versus gemcitabine for biliary tract cancer. N Engl J Med. 2010;362(14):1273-1281. doi:10.1056/NEJMoa0908721.

André

Tournigand

Rosmorduc

, et al. Gemcitabine combined with oxaliplatin (GEMOX) in advanced biliary tract adenocarcinoma: a GERCOR study. Ann Oncol. 2004;15(9):1339-1343. doi:10.1093/annonc/mdh351.

10.

Lamarca

Palmer

Wasan

, et al. Second-line FOLFOX chemotherapy versus active symptom control for advanced biliary tract cancer (ABC-06): a phase 3, open-label, randomised, controlled trial. Lancet Oncol. 2021;22(5):690-701. doi:10.1016/S1470-2045(21)00027-9.

11.

Burris

3rd Okusaka

Vogel

, et al. Durvalumab plus gemcitabine and cisplatin in advanced biliary tract cancer (TOPAZ-1): patient-reported outcomes from a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Oncol. 2024;25(5):626-635. doi:10.1016/S1470-2045(24)00082-2.

12.

Kelley

Ueno

Yoo

, et al. Pembrolizumab in combination with gemcitabine and cisplatin compared with gemcitabine and cisplatin alone for patients with advanced biliary tract cancer (KEYNOTE-966): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet. 2023;401(10391):1853-1865. doi:10.1016/S0140-6736(23)00727-4.

13.

Edeline

Touchefeu

Guiu

, et al. Radioembolization plus chemotherapy for first-line treatment of locally advanced intrahepatic cholangiocarcinoma: a phase 2 clinical trial. JAMA Oncol. 2020;6(1):51-59. doi:10.1001/jamaoncol.2019.3702.

14.

Phelip

Vendrely

Rostain

, et al. Gemcitabine plus cisplatin versus chemoradiotherapy in locally advanced biliary tract cancer: Fédération Francophone de Cancérologie Digestive 9902 phase II randomised study. European Journal of Cancer. 2014;50(17):2975-2982. doi:10.1016/j.ejca.2014.08.013.

15.

Ishii

Itano

Morinaga

Shirakawa

Itano

. Potential efficacy of hepatic arterial infusion chemotherapy using gemcitabine, cisplatin, and 5-fluorouracil for intrahepatic cholangiocarcinoma. PLoS One. 2022;17(4):e0266707. doi:10.1371/journal.pone.0266707. Published 2022 Apr 22.

16.

Kudo

Finn

Qin

, et al. Lenvatinib versus sorafenib in first-line treatment of patients with unresectable hepatocellular carcinoma: a randomised phase 3 non-inferiority trial. Lancet. 2018;391(10126):1163-1173. doi:10.1016/S0140-6736(18)30207-1.

17.

Kato

Tabata

Kimura

, et al. Lenvatinib plus anti-PD-1 antibody combination treatment activates CD8+ T cells through reduction of tumor-associated macrophage and activation of the interferon pathway. PLoS One. 2019;14(2):e0212513. doi:10.1371/journal.pone.0212513. Published 2019 Feb 27.

18.

Lin

Yang

Long

, et al. Pembrolizumab combined with lenvatinib as non-first-line therapy in patients with refractory biliary tract carcinoma. Hepatobiliary Surg Nutr. 2020;9(4):414-424. doi:10.21037/hbsn-20-338.

19.

Jian

Fan

Shi

G-M

, et al. Lenvatinib plus toripalimab as first-line treatment for advanced intrahepatic cholangiocarcinoma: a single-arm, phase 2 trial. J Clin Oncol. 2021;39:4099.

20.

Shi

Huang

, et al. Toripalimab combined with lenvatinib and GEMOX is a promising regimen as first-line treatment for advanced intrahepatic cholangiocarcinoma: a single-center, single-arm, phase 2 study. Signal Transduct Target Ther. 2023;8(1):106. doi:10.1038/s41392-023-01317-7. Published 2023 Mar 17.

21.

Zheng

Wang

, et al. Hepatic arterial infusion chemotherapy plus targeted therapy and immunotherapy versus systemic chemotherapy for advanced intrahepatic cholangiocarcinoma: a retrospective cohort study. Int J Surg. 2025;111(1):1552-1557. doi:10.1097/JS9.0000000000002013. Published 2025 Jan 1.

22.

Huang

Kan

, et al. Clinical and biomarker analyses of hepatic arterial infusion chemotherapy plus lenvatinib and PD-1 inhibitor for patients with advanced intrahepatic cholangiocarcinoma. Front Immunol. 2024;15:1260191. doi:10.3389/fimmu.2024.1260191. Published 2024 Feb 7.

23.

Zhang

Wang

, et al. Clinical outcomes of hepatic arterial infusion chemotherapy combined with tyrosine kinase inhibitors and anti-PD-1 immunotherapy for unresectable intrahepatic cholangiocarcinoma. J Dig Dis. 2022;23(8-9):535-545. doi:10.1111/1751-2980.13127.

24.

Zhao

Zhou

Miao

, et al. Efficacy and safety of lenvatinib plus durvalumab combined with hepatic arterial infusion chemotherapy for unresectable intrahepatic cholangiocarcinoma. Front Immunol. 2024;15:1397827. doi:10.3389/fimmu.2024.1397827. Published 2024 May 10.

25.

Lin

Zou

, et al. Efficacy analysis of HAIC combined with lenvatinib plus PD1 inhibitor vs. first-line systemic chemotherapy for advanced intrahepatic cholangiocarcinoma. Sci Rep. 2024;14(1):23961. doi:10.1038/s41598-024-75102-z. Published 2024 Oct 14.

26.

Zhao

Zhou

Xiong

, et al. Hepatic arterial infusion chemotherapy in combination with lenvatinib and durvalumab versus standard first-line treatment gemcitabine and cisplatin plus durvalumab in advanced intrahepatic cholangiocarcinoma. Am J Cancer Res. 2024;14(10):4922-4934. doi:10.62347/HVOF5644. Published 2024 Oct 15.

27.

Konstantinidis

Groot Koerkamp

, et al. Unresectable intrahepatic cholangiocarcinoma: Systemic plus hepatic arterial infusion chemotherapy is associated with longer survival in comparison with systemic chemotherapy alone. Cancer. 2016;122(5):758-765. doi:10.1002/cncr.29824.

28.

Cercek

Boerner

Tan

, et al. Assessment of Hepatic Arterial Infusion of Floxuridine in Combination With Systemic Gemcitabine and Oxaliplatin in Patients With Unresectable Intrahepatic Cholangiocarcinoma: A Phase 2 Clinical Trial. JAMA Oncol. 2020;6(1):60-67. doi:10.1001/jamaoncol.2019.3718.

29.

Franssen

Rousian

Nooijen

, et al. Hepatic arterial infusion pump chemotherapy in patients with advanced intrahepatic cholangiocarcinoma confined to the liver: A multicenter phase II trial. JCO. 2024;42:433.

30.

Sun

Guo

Zhou

Wei

. Hepatic arterial infusion of GEMOX plus systemic gemcitabine chemotherapy combined with lenvatinib and PD-1 inhibitor in large unresectable intrahepatic cholangiocarcinoma. Int Immunopharmacol. 2024;140:112872. doi:10.1016/j.intimp.2024.112872.

31.

von Elm

Altman

Egger

, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med. 2007;147(8):573-577.

32.

Zhao

Guo

Zou

, et al. Arterial chemotherapy for hepatocellular carcinoma in China: consensus recommendations. Hepatol Int. 2024;18(1):4-31. doi:10.1007/s12072-023-10599-6.

33.

Verderame

Russo

Di Leo

, et al. Gemcitabine and oxaliplatin combination chemotherapy in advanced biliary tract cancers. Ann Oncol. 2006;17(Suppl 7):vii68-vii72. doi:10.1093/annonc/mdl955.

34.

Hyder

Marsh

Salem

, et al. Intra-arterial therapy for advanced intrahepatic cholangiocarcinoma: a multi-institutional analysis. Ann Surg Oncol. 2013;20(12):3779-3786. doi:10.1245/s10434-013-3127-y.

35.

Endo

Gonen

Yopp

, et al. Intrahepatic cholangiocarcinoma: rising frequency, improved survival, and determinants of outcome after resection. Ann Surg. 2008;248(1):84-96. doi:10.1097/SLA.0b013e318176c4d3.

36.

Lamarca

Ross

Wasan

, et al. Advanced Intrahepatic Cholangiocarcinoma: Post Hoc Analysis of the ABC-01, -02, and -03 Clinical Trials. J Natl Cancer Inst. 2020;112(2):200-210. doi:10.1093/jnci/djz071.

37.

Huang

Lai

, et al. Single-Nucleus and Spatial Transcriptome Profiling Delineates the Multicellular Ecosystem in Hepatocellular Carcinoma After Hepatic Arterial Infusion Chemotherapy. Adv Sci (Weinh). 2025;12(5):e2405749. doi:10.1002/advs.202405749.

38.

Yamashita

Koay

Passot

, et al. Local therapy reduces the risk of liver failure and improves survival in patients with intrahepatic cholangiocarcinoma: A comprehensive analysis of 362 consecutive patients. Cancer. 2017;123(8):1354-1362. doi:10.1002/cncr.30488.

39.

Abou-Alfa

Macarulla

Javle

, et al. Ivosidenib in IDH1-mutant, chemotherapy-refractory cholangiocarcinoma (ClarIDHy): a multicentre, randomised, double-blind, placebo-controlled, phase 3 study. Lancet Oncol. 2020;21(6):796-807. doi:10.1016/S1470-2045(20)30157-1.

40.

Abou-Alfa

Sahai

Hollebecque

, et al. Pemigatinib for previously treated, locally advanced or metastatic cholangiocarcinoma: a multicentre, open-label, phase 2 study. Lancet Oncol. 2020;21(5):671-684. doi:10.1016/S1470-2045(20)30109-1.

41.

Hack

Zhu

Wang

. Augmenting Anticancer Immunity Through Combined Targeting of Angiogenic and PD-1/PD-L1 Pathways: Challenges and Opportunities. Front Immunol. 2020;11:598877. doi:10.3389/fimmu.2020.598877. Published 2020 Nov 5.

42.

Deng

Kan

Lyu

, et al. Dual Vascular Endothelial Growth Factor Receptor and Fibroblast Growth Factor Receptor Inhibition Elicits Antitumor Immunity and Enhances Programmed Cell Death-1 Checkpoint Blockade in Hepatocellular Carcinoma. Liver Cancer. 2020;9(3):338-357. doi:10.1159/000505695.

43.

Torrens

Montironi

Puigvehí

, et al. Immunomodulatory Effects of Lenvatinib Plus Anti-Programmed Cell Death Protein 1 in Mice and Rationale for Patient Enrichment in Hepatocellular Carcinoma. Hepatology. 2021;74(5):2652-2669. doi:10.1002/hep.32023.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

1.40 MB

0.00 MB