Efficacy of transarterial chemoembolization combined with PD-1 versus PD-L1 inhibitors in mass-forming intrahepatic cholangiocarcinoma: a multicenter retrospective study

Introduction

Intrahepatic cholangiocarcinoma (ICC) is an aggressive malignancy and the second most common primary liver cancer after hepatocellular carcinoma, with an increasing incidence worldwide (1). ICC has a poor prognosis, with a 5-year survival rate of less than 10% (1,2). Surgical resection is currently the only potentially curative treatment for ICC. Large population-based studies indicate that most patients present with unresectable or distant metastatic disease, which precludes surgical resection of their cancer (3,4). As for systemic chemotherapy, regimens containing gemcitabine and cisplatin are effective in patients with unresectable ICC. The transarterial chemoembolization (TACE) technique selectively occludes tumor-feeding arteries via transcatheter injection of lipiodol and chemotherapeutic agents, inducing ischemic necrosis of target tumors (5,6). In addition, embolic agents act as drug carriers for the local release of chemotherapeutic agents, thereby increasing the dose of drug exposure at the tumor site and expanding the extent of tumor necrosis (7). Moreover, TACE can induce necrosis of tumor cells, induce the release of tumor antigens, enhance the immune response of tumor-specific CD8⁺ T cells, regulate the proliferation of Treg cells, and transform “cold tumors” into “hot tumors” (8). This provides theoretical support for TACE combined with immune checkpoint inhibitor (ICI).

Meanwhile, the systemic treatment landscape for biliary tract cancers (BTCs), including ICC, has evolved considerably over the past decade. For patients with resected ICC, adjuvant capecitabine has been widely adopted based on phase III trial data (9). In the advanced or unresectable setting, phase III trials such as TOPAZ-1 and KEYNOTE-966 have shown that adding PD-1/PD-L1 inhibitors to gemcitabine-cisplatin yields modest but clinically meaningful improvements in overall survival (OS) compared with chemotherapy alone, without major increases in toxicity, and chemo-immunotherapy has become a new standard of care in many practice guidelines (10,11). Durvalumab, a PD-L1 inhibitor, in combination with gemcitabine and cisplatin significantly prolonged survival in patients with advanced BTC in TOPAZ-1, leading to its adoption as a first-line option (10). More recently, pembrolizumab, a PD-1 inhibitor, also demonstrated an OS benefit when combined with gemcitabine-cisplatin in KEYNOTE-966 (11). In parallel, recent reviews and perspective articles have highlighted the immunosuppressive tumour microenvironment of ICC, the potential synergy between ICIs and targeted agents, and the need to optimize patient selection and combination strategies for cancer immunotherapy (12-15). These developments provide important context for exploring how locoregional therapies such as TACE can be integrated with systemic immunotherapy in ICC.

Although PD-1 and PD-L1 inhibitors target the same inhibitory pathway, they have distinct mechanisms of action. PD-1 antibodies bind to PD-1 and prevent its interaction with both PD-L1 and PD-L2, thereby broadly blocking PD-1-mediated inhibitory signals on T cells (16). In contrast, PD-L1 antibodies bind to PD-L1, disrupting its interaction not only with PD-1 but also with CD80 (B7-1), while leaving the PD-1/PD-L2 interaction intact (17,18). In addition, several PD-L1 antibodies possess Fc regions capable of engaging Fcγ receptors, which may mediate antibody-dependent cellular cytotoxicity (ADCC) against PD-L1-expressing tumour or immune cells in the tumour microenvironment (19,20). These mechanistic differences suggest that PD-1 and PD-L1 blockade could have non-identical immunologic and clinical effects, particularly in tumours with complex immune microenvironments such as ICC. However, head-to-head data comparing PD-1- versus PD-L1-based regimens in ICC are lacking.

According to the Liver Cancer Study Group of Japan classification, based on the macroscopic growth pattern, ICC can be divided into three types: mass-forming (MF) type, periductal-infiltrating (PI) type, and intraductal-growth (IG) type (21). Among them, the MF type is the most common, accounting for 57.1–83.6% of ICCs (22,23). The clinical characteristics and biological behavior differ substantially among macroscopic subtypes, and the prognosis of the MF type is generally worse than that of the PI type (24,25). In recent years, with advances in imaging techniques and biomarkers, prognostic factors for mass-forming ICC (MF-ICC) have been proposed and validated. In a previous study, immunohistochemical staining to analyze the distribution of immune cells in the tumor and surrounding tissues revealed the correlation between CD4, CD8, and CD20 expression and the OS of ICC patients (26). Furthermore, the neutrophil-to-lymphocyte ratio (NLR)-based prognostic model had shown good application. It was found that patients with a preoperative NLR ≥2.36 have a poor prognosis, and the model demonstrates good predictive ability in survival and recurrence prediction (27). However, there is a lack of the risk factors for ICC patients receiving TACE combined with immunotherapy.

In this study, we evaluated the efficacy and safety between the different immunotherapy regimens (PD-1/PD-L1 inhibitors) in MF-ICC patients treated with TACE followed by ICI. Meanwhile, we explored baseline clinical and laboratory factors associated with prognosis in this population and combined flow cytometry with single-cell RNA sequencing (scRNA-seq) to investigate the underlying immune mechanisms of the combined therapy. This study provides clinical data to support the use of TACE combined with ICIs in the treatment of MF-ICC and is expected to offer a basis for therapeutic decision-making and the development of individualized therapy strategy. We present this article in accordance with the TRIPOD reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1917/rc).

Methods

Patients characteristics

This study retrospectively evaluated the ICC patients who received TACE followed by ICI therapy at Beijing Friendship Hospital and Beijing Ditan Hospital between May 2020 and December 2024. Because this was a retrospective cohort study, we included all consecutive patients who met the predefined inclusion and exclusion criteria and received TACE combined with ICIs at the two centers during the study period. The sample size (n=50) was therefore determined by the number of eligible cases available, and no a priori sample size calculation was performed. The inclusion criteria were as follows: (I) imaging findings suggest a MF type; (II) received TACE followed by ICI therapy; (III) age ≤75 years; (IV) Child-Pugh classification was A or B; (V) Eastern Cooperative Oncology Group (ECOG) 0–2. The exclusion criteria were as follows: (I) history of other malignant tumors; (II) serious complications involving the heart, lungs, or brain, or severe impairment of renal function; (III) lost to follow-up or the data were incomplete; (IV) expected survival was less than 3 months. All ICC patients were confirmed by histologically. The enrolled patients received first-line treatment. All ICC diagnoses were confirmed histologically, and all enrolled patients received TACE plus ICI as first-line systemic treatment. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Beijing Friendship Hospital, Capital Medical University (No. 2023-P2-176-02). Beijing Ditan Hospital was also informed of and agreed to the study. Individual consent for this retrospective analysis was waived.

Clinical and pathological data

Demographic and clinical data were collected and used for analysis within 7 days before the initiation of combined treatment. The demographic data included age and gender. Clinical data encompassed liver disease history, ECOG performance status, Child-Pugh classification, alpha-fetoprotein (AFP), carbohydrate antigen 19-9 (CA19-9), carcinoembryonic antigen (CEA), alanine aminotransferase (ALT), aspartate aminotransferase (AST), NLR, monocyte-to-lymphocyte ratio (MLR), platelet-to-lymphocyte ratio (PLR), gamma-glutamyl transferase (GGT), alkaline phosphatase (ALP), total bilirubin (TBIL), and direct bilirubin (DBIL).

Treatment received

TACE procedure

TACE treatment was performed by two board-certified interventional radiologists. Under local anesthesia, a 5-French catheter (TERUMO, Tokyo, Japan) was inserted after puncturing the common femoral artery using the modified Seldinger technique. Arteriography was conducted via an arterial catheter to confirm portal vein patency and delineate the tumor-feeding arteries. Under appropriate conditions, a microcatheter was advanced into the tumor-feeding artery, followed by injection of a doxorubicin-lipiodol emulsion and subsequent embolization using agents such as gelatin gelfoam sponge particles or polyvinyl alcohol particles. The blood flow was monitored until the complete vessel occlusion occurred.

ICI procedure

The choice of ICI (PD-1 vs. PD-L1 inhibitor) was not randomized but was made by the treating physicians based on drug availability, reimbursement status, and their clinical judgment at the time of treatment. The first inoculation was performed 3–5 days after the post-TACE syndrome had abated. The types of PD-1 inhibitors included camrelizumab or sintilimab. The PD-L1 inhibitor was durvalumab. All patients were treated with ICI administered as an intravenous (IV) infusion every 3 weeks. Treatment was maintained until disease progression, as determined by the modified Response Evaluation Criteria in Solid Tumors (mRECIST), the occurrence of intolerable adverse events (AEs), patient withdrawal, or clinical judgment by the attending physician.

Follow-up

Patients were regularly followed up until mortality or the study end date (1 May 2025). After receiving the combined treatment, an assessment of tumor response (magnetic resonance imaging and/or computed tomography scan) was performed. Patients were examined every three months during the first year and every 6 months thereafter. The follow-up included physical examinations, blood tests, and imaging examinations to monitor for tumor progression. The primary endpoint was OS, defined as the time interval from the combined treatment to the date of death or last follow-up. Secondary endpoints comprised progression-free survival (PFS), objective response rate (ORR), and disease control rate (DCR). PFS was defined as the interval from the start of the combined treatment to disease progression or death due to any cause. The tumor response was assessed using the mRECIST, including complete response (CR), partial response (PR), progressive disease (PD), and stable disease (SD). The ORR was the sum of CR and PR, while the DCR was the sum of CR, PR, and SD. Patients who were alive and without documented disease progression at the end of follow-up were censored at the date of last contact. No patients were lost to follow-up after enrolment; all surviving patients were successfully followed until the study end date or last contact.

Clustering and dimension reduction analysis of single-cell data

We downloaded the GSE284205 dataset from the Gene Expression Omnibus (GEO), which included six ICC tumor samples. The raw data contained 11,560 cells and 36,601 genes. For scRNA-seq data, quality control was performed by the Seurat package, where each cell expressed in at least 250, and each gene must occur in at least three cells. Mitochondria and ribosomal RNA (rRNA) quantities were calculated by the PercentageFeatureSet function. Filtering was applied in order to remove low-quality cells, excluding those with a minimum of 500 unique molecular identifiers (UMIs), or unique genes fewer than 200 or greater than 8,000, or greater than 25% mitochondrial gene expression. After filtering, 23,743 genes and 3,050 cells remained. The “NormalizeData” function via the “LogNormalize” method and 2,000 highly variable genes were identified by the “FindVariableGenes” function via the “vst” method. The hypervariable genes were selected as inputs for principal component analysis (PCA). The cells were clustered by FindNeighbors and FindClusters functions under conditions of dim =50 and resolution =0.1. UMAP dimension reduction was used for the dimension reduction. Differential gene expression between clusters was defined using the Findallmarker function as adjusted P value <0.05, log₂fold change >0.25, and a minimum percentage expression greater than 0.1. Cell clusters were annotated according to the CellMarker database and the marker genes obtained from the literature.

Flow cytometry

Peripheral blood samples were collected from 32 ICC patients both before and after treatment. Peripheral blood mononuclear cells (PBMCs) were isolated using density gradient centrifugation with Ficoll at 700 ×g for 20 minutes at room temperature. Mononuclear cells at the Ficoll-plasma interface were carefully collected, diluted in phosphate-buffered saline (PBS), and centrifuged at 500 ×g for 8 minutes to remove residual platelets and plasma components. The resulting cell pellet was resuspended in staining buffer (PBS supplemented with 2% fetal bovine serum) and processed for surface marker staining according to standard protocols. Cells were incubated with fluorochrome-conjugated monoclonal antibodies targeting CD3, CD4, CD8, CD16+56, CD19, and CD45 for 30 minutes at 4 ℃ in the dark. After incubation, cells were washed twice with a staining buffer and resuspended in PBS for flow cytometric analysis. Data acquisition was performed on a BD FACSC anto II flow cytometer (BD Biosciences, San Jose, CA, USA), and results were analyzed using FlowJo software (version 10.8.1). Immune cell subsets were identified based on established gating strategies, enabling quantitative assessment of dynamic changes in lymphocyte populations before and after treatment.

Statistical analysis

All data processing and analyses were carried out in R 4.1.2. Categorical variables were compared using the Chi-squared test or Fisher’s exact test, while continuous variables were analyzed using the t-test or the Mann-Whitney U test. Survival curves for OS and PFS were estimated using the Kaplan-Meier method, and differences between the TACE-PD-1 and TACE-PD-L1 groups were compared using the log-rank test. To quantify the association between treatment regimen and survival outcomes, Cox proportional hazards models were fitted to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). Least absolute shrinkage and selection operator (LASSO) regression was used to screen the risk factors of OS. All statistical tests were two-sided, and a P<0.05 was considered statistically significant.

Results

Patient characteristics

Between May 2020 and December 2024, a total of 50 patients with unresectable nodular ICC who underwent TACE followed by immunotherapy at Beijing Friendship Hospital and Beijing Ditan Hospital were enrolled in our study, including 22 patients who received PD-1 therapy and 28 who received PD-L1 therapy. The follow-up period ended March 15, 2025, and the median follow-up time was 12 months.

In this study, 29 (58%) subjects were males, and 21 (42%) were females, showing an average age of 64.84±9.85 years. There were 37 Child-Pugh A patients (74%) and 13 Child-Pugh B patients (26%). Nine patients (18%) were diagnosed with hepatitis. The mean levels of serum AFP, ALT, AST, GGT, and TBIL were 73.33±423.36 ng/mL, 41.28±42.96 U/L, 49.44±56.92 U/L, 251.57±236.19 U/L, and 24.12±20.33 µmol/L (Table 1).

Efficacy

The median OS (mOS) for patients receiving TACE combined with immunotherapy was 16 months, and the median PFS (mPFS) was 11 months. There was 1 CR patient, 16 PR patients, 21 SD patients, and 12 PD patients. The ORR was 34%, and the DCR was 76%. The mOS and mPFS from the start of TACE-PD-L1 sequential therapy were 18 and 13 months, significantly longer than TACE-PD-1 sequential therapy (mOS: 12 months, HR: 0.42, 95% CI: 0.17–1.03, P=0.047; mPFS: 8 months, HR: 0.29, 95% CI: 0.12–0.73, P=0.006, Figure 1). The ORR for TACE combined with PD-L1 was 25%, and the DCR was 60.7%. The ORR for TACE combined with PD-1 was 9%, and the DCR was 54.5% (Table 2).

Figure 1 Kaplan-Meier curves of OS and PFS for different immune checkpoint inhibitor treatments. (A) Kaplan-Meier of OS; (B) Kaplan-Meier of PFS. OS, overall survival; PD-1, programmed cell death protein-1; PD-L1, programmed death ligand-1; PFS, progression-free survival.

Exploratory analysis of prognosis-related factors

LASSO regression was used to screen risk factors for OS, and the changes in variable coefficients are shown in Figure 2A. The model performance was excellent when the λ was 0.13733 (Figure 2B). Four variables—pre-treatment MLR, Child-Pugh classification, TBIL, and ALT—were retained by the LASSO procedure as candidates associated with OS.

Figure 2 LASSO regression. (A) The variation characteristics of the coefficient of variables; (B) the selection process of the optimum value of the parameter λ in the LASSO regression model by cross-validation method. LASSO, least absolute shrinkage and selection operator.

Single-cell analysis to investigate the impact of immunotherapy on systemic immune cells

We found that the pre-treatment MLR was the risk factor for TACE combined with immunotherapy. ICC patients in our study were divided into high-MLR and low-MLR groups according to their baseline MLR values. Kaplan-Meier revealed significant differences in OS (P=0.02) and PFS (P=0.01) between the two groups, with lower MLR levels associated with a more favorable prognosis (Figure 3A,3B). In addition, one week after combination therapy, the patient’s MLR level significantly increased (P=0.03, Figure 3C), further supporting the potential role of MLR in the efficacy of immunotherapy.

Figure 3 The relationship between MLR and combination therapy. (A) Kaplan-Meier curves of OS for MLR groups; (B) Kaplan-Meier curves of PFS for MLR groups; (C) changes in MLR before and after combination therapy. MLR, monocyte-to-lymphocyte ratio; OS, overall survival; PFS, progression-free survival.

To investigate the systemic immune response to immunotherapy, we analyzed the dynamic changes in immune cell populations before and after ICC treatment using the single-cell dataset GSE284204. The single-cell analysis was performed on six samples, including three collected before treatment and three collected after treatment. After stringent quality control, we retained 3,050 cells and 23,743 genes for downstream analysis. The sample distribution before and after remove batch is presented in Figure S1. After log-normalization and dimensionality reduction, we applied a resolution of 0.2 and identified seven distinct cell clusters. The cell types of the different clusters were further annotated based on the expression patterns of the marker genes. We applied CD4 to label CD4⁺ T, CD8A and CD8B to label CD8⁺ T, CD14 and C1QC to label monocyte, COL1A2 and COL1A1 to label fibroblasts, TMPRSS2 and OLFM4 to label circulating tumor cell, CD79A and CD79B to label B cell, VWF and SELE to label vascular endothelial cell, and NKG7 to label natural killer (NK) cell. A statistical analysis was performed of the proportions of different cell types. Our results indicate that CD4⁺ T cells and CD8⁺ T cells were significantly increased after treatment while circulating tumor cells and vascular endothelial cells were markedly decreased (Figure 4).

Figure 4 Single-cell analysis. (A) UMAP of 8 cell subgroups; (B) UMAP of 8 cell types; (C) the expression of major marker genes in 8 cell clusters; (D) cell proportions per patient; (E) proportional distribution of cell types across patients before and after treatment; (F-H) UMAP visualization of CD4, CD8, and CD19 before and after treatment. NK, natural killer; UMAP, uniform manifold approximation and projection.

Validate of changes in lymphocyte populations before and after treatment

To further validate the changes in lymphocyte populations, we collected peripheral blood samples from 32 patients before and after treatment and performed flow cytometry analysis. Treatment led to a notable increase in CD8⁺ T cells and a reduction in CD19⁺ B cells. Besides, regardless of treatment status, ICC patients in the PD-L1 groups exhibited higher levels of CD4⁺ and CD8⁺ T cells compared to the PD-1 group, whereas B cells were lower in the PD-L1 group than in the PD-1 group (Figure 5).

Figure 5 Changes in CD4⁺, CD8⁺, and CD19⁺ lymphocyte subsets before and after ICI treatment and between PD-1 and PD-L1 groups. (A) Paired box plots showing the proportion of CD4⁺ T cells; (B) paired box plots showing the proportion of CD8⁺ T cells; (C) paired box plots showing the proportion of CD19⁺ B cells; (D,E) box plots comparing the proportions of CD4⁺, CD8⁺, and CD19⁺ lymphocytes between patients receiving PD-1 versus PD-L1 blockade before (D) and after (E) treatment. ICI, immune checkpoint inhibitor; PD-1, programmed cell death protein-1; PD-L1, programmed death ligand-1.

Discussion

ICC is a highly aggressive malignancy with limited responsiveness to conventional radiotherapy and chemotherapy. It is most commonly observed in its MF subtype, which accounts for approximately 57.1% to 83.6% of cases (22,23,28,29). Although substantial progress has been made in molecular targeted therapy in the past five years, patient outcomes remain dismal. Immunotherapy has revolutionized the therapeutic landscape across different cancer types, including ICI. However, many previous studies have shown that immunotherapy has primarily been used as a monotherapy in ICC patients, and their clinical efficacy as monotherapies is limited (30,31). Exploring combination immunotherapeutic strategies is crucial for enhancing therapeutic outcomes. In this retrospective study, we analyzed the clinical outcomes of ICC patients treated with TACE in combination with PD-1 or PD-L1 inhibitors. Our findings revealed that the PD-L1-based combination therapy conferred greater therapeutic benefit than the PD-1-based regimen. Notably, immune cells appeared to play a pivotal role in mediating the efficacy of combination treatment. Furthermore, we identified prognostic factors associated with OS, thereby providing a potential tool to aid clinical decision-making. From a clinical perspective, these results support the use of TACE followed by ICIs as a potentially valuable option for patients with unresectable MF-ICC.

From a clinical perspective, our findings have several practical implications for the management of unresectable MF-ICC. In our centers, TACE followed by ICIs has gradually become a preferred option for patients with liver-dominant disease, preserved liver function (mainly Child-Pugh class A), and good performance status, in whom an aggressive locoregional-immunotherapy approach is considered acceptable. The mOS of 16 months and mPFS of 11 months observed in our cohort appear to be numerically higher than most historical series of TACE alone or systemic chemotherapy in similar populations, suggesting that the TACE-ICI strategy may provide meaningful additional benefit in carefully selected patients. Although cross-trial comparisons should be interpreted with caution, these real-world data may help multidisciplinary teams to consider TACE-PD-L1 sequences as a rational option when curative resection or transplantation is not feasible.

As the most widely employed intra-arterial treatment for ICC, TACE has demonstrated a significant survival benefit in clinical practice (32). In comparison to systemic chemotherapy, TACE enables the localized delivery of higher doses of cytotoxic agents within the tumor, with relatively reduced systemic exposure and toxicity (33,34). Moreover, Scheuermann et al. found that the survival outcomes of patients with unresectable ICC treated with either conventional TACE (cTACE) or drug-eluting bead TACE (DEB-TACE), with a mOS of 11 months, were similar to those of patients who underwent surgical resection with positive margins (11 months) or lymph node involvement (9 months) (35). TACE has been shown to facilitate the release of tumor antigens and proinflammatory cytokines, promote immunogenic cell death, and convert non-immunogenic tumors into immunogenic phenotypes, thereby potentiating the therapeutic effects of immunotherapy (36,37). Currently, PD-1/PD-L1 is one of the most studied and clinically successful ICI drug targets. Emerging evidence indicates that the PD-1/PD-L1 signaling pathway is hyperactivated in tumor tissues of ICC, with its heightened expression positively associated with aggressive pathological features such as lymph node invasion and higher tumor-node-metastasis (TNM) staging (38). In our study, the combination of TACE with PD-1/PD-L1 inhibitors achieved a mPFS of 11 months and a mOS of 16 months, supporting the biological rationale that post-TACE antigen release and local inflammation may create a window of opportunity for PD-1/PD-L1 blockade in MF-ICC.

Our findings suggested that PD-L1 inhibitors may offer greater clinical benefit than PD-1 inhibitors in ICC patients, particularly in the context of a post-TACE microenvironment. Cancer cells can escape immune detection by upregulating PD-L1, which engages PD-1 on T cells and triggers SHP-2-mediated dephosphorylation events that attenuate T-cell receptor (TCR) signaling, ultimately leading to impaired T cell activation and function (39,40). In addition to tumor cells, PD-L1 is upregulated on antigen-presenting cells (APCs), including dendritic cells (DCs) and macrophages, on which PD-L1 expression has been shown to correlate closely with the efficacy of PD-1/PD-L1 blockade therapy (41). TACE induces tumor necrosis and the release of danger signals and tumor antigens, which are taken up by APCs and can drive their activation and PD-L1 upregulation. In such a setting, PD-L1 blockade may simultaneously target PD-L1 expressed on both tumor cells and APCs, thereby relieving inhibitory signals at multiple checkpoints of the cancer-immunity cycle after TACE. Studies have shown that PD-L1 expression on DCs can suppress T cell activation, an effect that can be reversed upon PD-L1 blockade, leading to the functional restoration of tumor-infiltrating T cells (42). Besides PD1, PD-L1 can also bind to the costimulatory CD80 receptor (B7-1). This alternative immunosuppressive pathway is not targeted by PD-1 inhibitors but can be effectively disrupted by PD-L1 blockade, providing a potential explanation for the enhanced efficacy of PD-L1-targeted therapies (43). Moreover, some PD-L1 antibodies retain Fc-dependent effector functions and may mediate ADCC against PD-L1-expressing cells in the tumor microenvironment, further reshaping the balance between effector and suppressive immune cells. Taken together, although our retrospective data cannot establish causality, these mechanistic considerations provide a plausible framework to understand why TACE-PD-L1 might be associated with more favorable outcomes than TACE-PD-1 in MF-ICC.

Tao et al. found that MLR was an independent predictor of OS following radical resection for ICC. Patients with a high MLR exhibited poorer disease-free survival (DFS) rates at 1, 3, and 5 years compared to those with a low MLR. Compared with the low MLR group, patients with high MLR had more aggressive tumor characteristics, including larger tumor size, increased microvascular invasion, higher incidence of lymph node metastasis, and a greater number of satellite lesions (44). Zhang et al. employed LASSO regression analysis and identified MLR as a prognostic factor in ICC patients (45). The serum levels of Child-Pugh classification, TBIL, and ALT indicate more severe liver cirrhosis, which may lead to decreased liver regeneration and increased risk of recurrence (46). Studies have shown that patients with Child-Pugh class B have a higher risk of recurrence and lower survival rates after undergoing TACE combined with ablation (47). Additionally, changes in TBIL are indicative of greater tumor aggressiveness and metastatic capacity and may serve as a marker of a more invasive tumor phenotype and unfavorable prognosis (46).

In our study, the prognosis of the high-MLR group was significantly worse than that of the low-MLR group. As a prototypical inflammation-related cancer, approximately 90% of the liver cancer burden results from prolonged hepatitis due to viral hepatitis, excessive alcohol intake, nonalcoholic fatty liver disease (NAFLD), or nonalcoholic steatohepatitis (NASH) (48). The immune microenvironment and inflammatory markers are integral components of the systemic inflammatory response and play critical roles in tumor initiation, progression, invasion, and metastasis (49,50). MLR is the ratio of monocyte to lymphocyte. Monocyte gathers at the inflammatory site and differentiate into M1 and M2 macrophages when inflammation occurs. M2 macrophages have the effect of promoting tumor formation, and activated circulating monocytes can secrete a variety of pro-inflammatory factors that accelerate the development of cancer (51-53). Lymphocytes are negatively correlated with tumor stage and play a key role in anti-tumor immune response (54). Our single-cell analysis revealed an increase in CD4⁺ and CD8⁺ T cell levels following ICI treatment, which was further validated by flow cytometry results. Plasma cells release anti-tumor antibodies to induce antibody-dependent cell cytotoxicity. In contrast, regulatory B-cells (Bregs) can suppress the immune response and enhance angiogenesis through the secretion of vascular endothelial growth factor (VEGF) and interleukin-10 (IL-10) (55). Zhang et al. (56) revealed that CD19⁺CD73⁺ B cells were linked to a poor prognosis and invalid therapeutic response to immunotherapy of gastric cancer. In line with these concepts, our flow cytometry results showed that CD4⁺ and CD8⁺ T cell levels were higher in the PD-L1 group compared to the PD-1 group both before and after treatment, whereas CD19⁺ B cells were lower in the PD-L1 group. This pattern is consistent with a more favorable effector/suppressor balance in patients receiving PD-L1 blockade and may partly underlie the better clinical outcomes observed in this group. Importantly, the scRNA-seq dataset analyzed in our study was derived from an independent ICC cohort, and our flow-cytometric analyses were performed on peripheral blood rather than paired tumor tissue. Therefore, these immune findings should be viewed as complementary and hypothesis-generating, providing biological context for the clinical observations rather than direct mechanistic proof.

There are certain limitations in this study. First, the study was a retrospective, non-randomized analysis conducted at two centers in China, so selection bias and residual confounding cannot be completely excluded, and the generalizability of our findings to other populations and practice settings may be limited. Second, the overall sample size and number of events were still modest. Therefore, some effect estimates should be interpreted with caution. Third, the scRNA-seq dataset was derived from an independent ICC cohort and our flow-cytometry validation was performed in a relatively small subset of patients using peripheral blood rather than paired tumor tissue, so the proposed immune mechanisms remain exploratory and hypothesis-generating.

Conclusions

TACE combined with a PD-L1 inhibitor was associated with longer OS and PFS compared with TACE combined with a PD-1 inhibitor. The immune system, particularly lymphocytes, plays a critical role in the efficacy of combination therapy, and our single-cell and flow cytometry analyses suggest that enhanced CD4⁺/CD8⁺ T-cell responses and a lower B-cell compartment may be linked to the observed clinical benefit with PD-L1 blockade.

Acknowledgments

We would like to thank all patients who participated in the study.

Footnote

Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1917/rc

Data Sharing Statement: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1917/dss

Peer Review File: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1917/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1917/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Beijing Friendship Hospital, Capital Medical University (No. 2023-P2-176-02). Beijing Ditan Hospital was also informed of and agreed to the study. Individual consent for this retrospective analysis was waived.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Beal EW, Tumin D, Moris D, et al. Cohort contributions to trends in the incidence and mortality of intrahepatic cholangiocarcinoma. Hepatobiliary Surg Nutr 2018;7:270-6. [Crossref] [PubMed]
2. Owen ML, Beal EW. Minimally Invasive Surgery for Intrahepatic Cholangiocarcinoma: Patient Selection and Special Considerations. Hepat Med 2021;13:137-43. [Crossref] [PubMed]
3. Sulpice L, Rayar M, Boucher E, et al. Treatment of recurrent intrahepatic cholangiocarcinoma. Br J Surg 2012;99:1711-7. [Crossref] [PubMed]
4. Nathan H, Pawlik TM, Wolfgang CL, et al. Trends in survival after surgery for cholangiocarcinoma: a 30-year population-based SEER database analysis. J Gastrointest Surg 2007;11:1488-96; discussion 1496-7. [Crossref] [PubMed]
5. Savic LJ, Chapiro J, Geschwind JH. Intra-arterial embolotherapy for intrahepatic cholangiocarcinoma: update and future prospects. Hepatobiliary Surg Nutr 2017;6:7-21. [Crossref] [PubMed]
6. Song JE, Kim DY. Conventional vs drug-eluting beads transarterial chemoembolization for hepatocellular carcinoma. World J Hepatol 2017;9:808-14. [Crossref] [PubMed]
7. Jiang N, Zhang Z, Yin X, et al. Systemic chemotherapy plus transarterial chemoembolization versus systemic chemotherapy alone for unresectable intrahepatic cholangiocarcinoma: a multicenter retrospective cohort study. Radiol Med 2024;129:631-42. [Crossref] [PubMed]
8. Wu Y, Lin H, You X, et al. Immune Checkpoint Blockade in Chinese Patients With Hepatocellular Carcinoma: Characteristics and Particularity. Front Oncol 2022;12:764923. [Crossref] [PubMed]
9. Rizzo A, Brandi G. BILCAP trial and adjuvant capecitabine in resectable biliary tract cancer: reflections on a standard of care. Expert Rev Gastroenterol Hepatol 2021;15:483-5. [Crossref] [PubMed]
10. Oh DY, He AR, Bouattour M, et al. Durvalumab or placebo plus gemcitabine and cisplatin in participants with advanced biliary tract cancer (TOPAZ-1): updated overall survival from a randomised phase 3 study. Lancet Gastroenterol Hepatol 2024;9:694-704. [Crossref] [PubMed]
11. Kelley RK, Ueno M, Yoo C, 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;401:1853-65. [Crossref] [PubMed]
12. Brandi G, Rizzo A. IDH Inhibitors and Immunotherapy for Biliary Tract Cancer: A Marriage of Convenience? Int J Mol Sci 2022;23:10869. [Crossref] [PubMed]
13. Vitale E, Rizzo A, Santa K, et al. Associations between "Cancer Risk", "Inflammation" and "Metabolic Syndrome": A Scoping Review. Biology (Basel) 2024;13:352. [Crossref] [PubMed]
14. Bas O, Sahin TK, Karahan L, et al. Prognostic significance of the cachexia index (CXI) in patients with cancer: A systematic review and meta-analysis. Clin Nutr ESPEN 2025;68:240-7. [Crossref] [PubMed]
15. Ricci AD, Rizzo A, Lotesoriere C. Immunotherapy in biliary tract cancer: are we finally on the right path? Lancet Gastroenterol Hepatol 2024;9:678-9. [Crossref] [PubMed]
16. Munari E, Mariotti FR, Quatrini L, et al. PD-1/PD-L1 in Cancer: Pathophysiological, Diagnostic and Therapeutic Aspects. Int J Mol Sci 2021;22:5123. [Crossref] [PubMed]
17. Chen M, Zhou Y, Bao K, et al. Multispecific Antibodies Targeting PD-1/PD-L1 in Cancer. BioDrugs 2025;39:427-44. [Crossref] [PubMed]
18. Zhang Y, Song Q, Cassady K, et al. Blockade of trans PD-L1 interaction with CD80 augments antitumor immunity. Proc Natl Acad Sci U S A 2023;120:e2205085120. [Crossref] [PubMed]
19. Martineau R, Susini S, Marabelle A. Fc Effector Function of Immune Checkpoint Blocking Antibodies in Oncology. Immunol Rev 2024;328:334-49. [Crossref] [PubMed]
20. Cohen Saban N, Yalin A, Landsberger T, et al. Fc glycoengineering of a PD-L1 antibody harnesses Fcγ receptors for increased antitumor efficacy. Sci Immunol 2023;8:eadd8005. [Crossref] [PubMed]
21. Ma CH, Hwang DW, Song KB, et al. Prognostic factors predicting survival rate over 10 years of patients with intrahepatic cholangiocarcinoma after hepatic resection. Ann Surg Treat Res 2020;98:116-23. [Crossref] [PubMed]
22. Bagante F, Weiss M, Alexandrescu S, et al. Long-term outcomes of patients with intraductal growth sub-type of intrahepatic cholangiocarcinoma. HPB (Oxford) 2018;20:1189-97. [Crossref] [PubMed]
23. Dover LL, Jacob R, Wang TN, et al. Improved Postoperative Survival for Intraductal-Growth Subtype of Intrahepatic Cholangiocarcinoma. Am Surg 2016;82:1133-9.
24. Imai K, Yamamoto M, Ariizumi S. Surgery for periductal infiltrating type intrahepatic cholangiocarcinoma without hilar invasion provides a better outcome than for mass-forming type intrahepatic cholangiocarcinoma without hilar invasion. Hepatogastroenterology 2010;57:1333-6.
25. Vijgen S, Terris B, Rubbia-Brandt L. Pathology of intrahepatic cholangiocarcinoma. Hepatobiliary Surg Nutr 2017;6:22-34. [Crossref] [PubMed]
26. Wang ZR, Zhang CZ, Ding Z, et al. Establishing prognostic models for intrahepatic cholangiocarcinoma based on immune cells. World J Gastrointest Oncol 2024;16:4092-103. [Crossref] [PubMed]
27. Qi S, Ma Z, Shen L, et al. Application of preoperative NLR-based prognostic model in predicting prognosis of intrahepatic cholangiocarcinoma following radical surgery. Front Nutr 2024;11:1492358. [Crossref] [PubMed]
28. Uno M, Shimada K, Yamamoto Y, et al. Periductal infiltrating type of intrahepatic cholangiocarcinoma: a rare macroscopic type without any apparent mass. Surg Today 2012;42:1189-94. [Crossref] [PubMed]
29. Lim JH. Cholangiocarcinoma: morphologic classification according to growth pattern and imaging findings. AJR Am J Roentgenol 2003;181:819-27. [Crossref] [PubMed]
30. Ricci AD, Rizzo A, Brandi G. Immunotherapy in Biliary Tract Cancer: Worthy of a Second Look. Cancer Control 2020;27:1073274820948047. [Crossref] [PubMed]
31. Rizzo A, Ricci AD, Brandi G. Recent advances of immunotherapy for biliary tract cancer. Expert Rev Gastroenterol Hepatol 2021;15:527-36. [Crossref] [PubMed]
32. Park SY, Kim JH, Yoon HJ, et al. Transarterial chemoembolization versus supportive therapy in the palliative treatment of unresectable intrahepatic cholangiocarcinoma. Clin Radiol 2011;66:322-8. [Crossref] [PubMed]
33. Raoul JL, Forner A, Bolondi L, et al. Updated use of TACE for hepatocellular carcinoma treatment: How and when to use it based on clinical evidence. Cancer Treat Rev 2019;72:28-36. [Crossref] [PubMed]
34. Chen S, Yu W, Zhang K, et al. Comparison of the efficacy and safety of Transarterial chemoembolization with and without Apatinib for the treatment of BCLC stage C hepatocellular carcinoma. BMC Cancer 2018;18:1131. [Crossref] [PubMed]
35. Scheuermann U, Kaths JM, Heise M, et al. Comparison of resection and transarterial chemoembolisation in the treatment of advanced intrahepatic cholangiocarcinoma--a single-center experience. Eur J Surg Oncol 2013;39:593-600. [Crossref] [PubMed]
36. Chang X, Lu X, Guo J, et al. Interventional therapy combined with immune checkpoint inhibitors: Emerging opportunities for cancer treatment in the era of immunotherapy. Cancer Treat Rev 2019;74:49-60. [Crossref] [PubMed]
37. Jin ZC, Chen JJ, Zhu XL, et al. Immune checkpoint inhibitors and anti-vascular endothelial growth factor antibody/tyrosine kinase inhibitors with or without transarterial chemoembolization as first-line treatment for advanced hepatocellular carcinoma (CHANCE2201): a target trial emulation study. EClinicalMedicine 2024;72:102622. [Crossref] [PubMed]
38. Lu JC, Zeng HY, Sun QM, et al. Distinct PD-L1/PD1 Profiles and Clinical Implications in Intrahepatic Cholangiocarcinoma Patients with Different Risk Factors. Theranostics 2019;9:4678-87. [Crossref] [PubMed]
39. Bardhan K, Anagnostou T, Boussiotis VA. The PD1:PD-L1/2 Pathway from Discovery to Clinical Implementation. Front Immunol 2016;7:550. [Crossref] [PubMed]
40. Yi M, Yu S, Qin S, et al. Gut microbiome modulates efficacy of immune checkpoint inhibitors. J Hematol Oncol 2018;11:47. [Crossref] [PubMed]
41. Lin H, Wei S, Hurt EM, et al. Host expression of PD-L1 determines efficacy of PD-L1 pathway blockade-mediated tumor regression. J Clin Invest 2018;128:805-15. [Crossref] [PubMed]
42. Peng Q, Qiu X, Zhang Z, et al. PD-L1 on dendritic cells attenuates T cell activation and regulates response to immune checkpoint blockade. Nat Commun 2020;11:4835. [Crossref] [PubMed]
43. Sugiura D, Maruhashi T, Okazaki IM, et al. Restriction of PD-1 function by cis-PD-L1/CD80 interactions is required for optimal T cell responses. Science 2019;364:558-66. [Crossref] [PubMed]
44. Tao BF, Zhu HQ, Qi LN, et al. Preoperative monocyte-to-lymphocyte ratio as a prognosis predictor after curative hepatectomy for intrahepatic cholangiocarcinoma. BMC Cancer 2024;24:1179. [Crossref] [PubMed]
45. Zhang H, Huang G, Li Q, et al. Construction and validation of a novel tumor morphology immune inflammatory nutritional score (TIIN score) for intrahepatic cholangiocarcinoma: a multicenter study. BMC Cancer 2024;24:630. [Crossref] [PubMed]
46. Zhang J, Wang X, Zhang L, et al. Radiomics predict postoperative survival of patients with primary liver cancer with different pathological types. Ann Transl Med 2020;8:820. [Crossref] [PubMed]
47. Qiao W, Sheng S, Xiong Y, et al. Nomogram for predicting post-therapy recurrence in BCLC A/B hepatocellular carcinoma with Child-Pugh B cirrhosis. Front Immunol 2024;15:1369988. [Crossref] [PubMed]
48. Llovet JM, Zucman-Rossi J, Pikarsky E, et al. Hepatocellular carcinoma. Nat Rev Dis Primers 2016;2:16018. [Crossref] [PubMed]
49. Coussens LM, Werb Z. Inflammation and cancer. Nature 2002;420:860-7. [Crossref] [PubMed]
50. Mao S, Yu X, Shan Y, et al. Albumin-Bilirubin (ALBI) and Monocyte to Lymphocyte Ratio (MLR)-Based Nomogram Model to Predict Tumor Recurrence of AFP-Negative Hepatocellular Carcinoma. J Hepatocell Carcinoma 2021;8:1355-65. [Crossref] [PubMed]
51. Jiang S, Yang Y, Fang M, et al. Co-evolution of tumor-associated macrophages and tumor neo-vessels during cervical cancer invasion. Oncol Lett 2016;12:2625-31. [Crossref] [PubMed]
52. Lippitz BE, Harris RA. Cytokine patterns in cancer patients: A review of the correlation between interleukin 6 and prognosis. Oncoimmunology 2016;5:e1093722. [Crossref] [PubMed]
53. Chen Y, Zhang S, Wang Q, et al. Tumor-recruited M2 macrophages promote gastric and breast cancer metastasis via M2 macrophage-secreted CHI3L1 protein. J Hematol Oncol 2017;10:36. [Crossref] [PubMed]
54. Wang Q, Qiao W, Liu B, et al. The monocyte to lymphocyte ratio not only at baseline but also at relapse predicts poor outcomes in patients with hepatocellular carcinoma receiving locoregional therapy. BMC Gastroenterol 2022;22:98. [Crossref] [PubMed]
55. Downs-Canner SM, Meier J, Vincent BG, et al. B Cell Function in the Tumor Microenvironment. Annu Rev Immunol 2022;40:169-93. [Crossref] [PubMed]
56. Zhang Y, Wang W, Liu Q, et al. CD19+CD73+ B cells infiltration indicates poor prognosis and unfavorable responses to immunotherapy in gastric cancer. Int Immunopharmacol 2024;141:113002. [Crossref] [PubMed]