ABSTRACT
Cholangiocarcinoma (CCA) is an epithelial cell cancer of the biliary tract. CCA can be further classified into intrahepatic cholangiocarcinoma (iCCA), perihilar cholangiocarcinoma (pCCA), and distal cholangiocarcinoma (dCCA) depending on the anatomical location. Until recently, the treatment for advanced CCA has remained highly reliant on chemotherapy, with gemcitabine plus cisplatin used in first-line treatment. Recent developments have led to the addition of immune checkpoint blockade (ICB) to the chemotherapy regimen, highlighting the promising potential of immunotherapies for CCA treatment. Despite these developments, most patients still do not benefit from current treatments, and response rates to ICB monotherapy remain modest. This underscores the need to develop more effective immunotherapeutic strategies. A major obstacle to this is the highly heterogenous nature of the disease. CCA tumors exhibit high inter-tumor heterogeneity in terms of anatomical locations, driver mutations, etiologies, and tumor microenvironment (TME) composition, making each patient immunologically distinct and difficult to benefit from a one-size-fits-all approach. There is a need to stratify patients according to individual disease status to identify immunotherapies and combination therapies that are most beneficial to them. Here we describe the different ways inter-tumor heterogeneity may arise in CCA, including stromal cell abundance, anatomical location, driver mutations, etiologies, TME profile, and tertiary lymphoid structure (TLS) presence. We also discuss what these factors mean to the immune microenvironment and their potential to be used as biomarkers. Careful stratification of patients is crucial in designing personalized medicine to improve survival outcomes and treatment efficacy for CCA patients.
-
Keywords: Cholangiocarcinoma; Liver cancer; Immunotherapy; Immune checkpoint blockade; Molecular targeted therapy
INTRODUCTION
Cholangiocarcinoma (CCA) is an epithelial cell malignancy originating anywhere along the biliary tract. Together with gallbladder cancer (GBC), CCA is referred to as biliary tract cancer (BTC) [
1]. CCA can also be further classified depending on its anatomical site. Intrahepatic cholangiocarcinoma (iCCA) arises from the second-order bile duct located within the liver and represents the second most common form of primary liver cancer, only after hepatocellular carcinoma (HCC). Perihilar cholangiocarcinoma (pCCA) and distal cholangiocarcinoma (dCCA) are collectively referred to as extrahepatic cholangiocarcinoma (eCCA). pCCA arises between the second-order bile duct and the cystic duct insertion, while dCCA originates from the common bile duct [
2].
CCA is a relatively rare cancer, with an incidence of fewer than 6 cases per 100,000 people in most countries. However, in certain parts of Asia, where risk factors for CCA prevail, cases are 10–20 times higher [
2,
3]. The most well-known risk factor for CCA is infection by liver flukes [
2,
4]. Other risk factors include underlying biliary tract diseases, such as primary sclerosing cholangitis (PSC), as well as conditions involving obstruction to the biliary tree, such as choledochal cyst. Liver conditions such as liver cirrhosis, non-alcoholic fatty liver disease (NAFLD) and hepatitis B virus (HBV), and hepatitis C virus (HCV) infections are also associated with CCA development [
5]. A common feature among these risk factors is the development of an inflammatory condition in the bile duct or the liver.
The prognosis for CCA patients remains poor, with a 5-year overall survival (OS) rate remaining at 7–20% [
5]. This is largely attributed to the limited treatment options for CCA. Currently, the only curative treatment for CCA is surgical resection. However, tumor relapses are common, and 5-year OS following surgical resection remains at 20–40% [
3]. Moreover, due to its lack of symptoms at an early stage, the majority of CCA patients are diagnosed at an advanced stage, where surgical resection is not feasible. Until recently, the standard first-line treatment for advanced CCA remained highly reliant on chemotherapy [
1].
Immune checkpoint blockade (ICB) has revolutionized cancer treatment in the past decade, with immunotherapy now a leading field of cancer treatment. However, the efficacy of ICB remains low in CCA. The response rate is between 5–10% in all patients and remains to be around 10% even in PD-L1
+ patients [
6,
7]. The poor response of CCA to ICB underscores the immunosuppressive nature of the disease. Critical questions remain: what makes CCA unresponsive to immunotherapy, and what can be done to make it responsive?
The key to discovering effective immunotherapy lies in understanding the tumor microenvironment (TME) to identify barriers to anti-tumor immunity. One factor that makes this difficult in CCA is its inter-tumoral heterogeneity (
Fig. 1). Not only do CCA tumors exhibit inter-tumoral heterogeneity in terms of anatomical location and etiologies, but CCA tumors are also mutationally distinct with specific driver mutations. The composition of the TME contributes an additional layer of difference between each tumor. The diversity of CCA tumors makes them immunologically distinct and challenging to target with a single agent. Consequently, rather than a one-fit-all approach, there is a need to stratify CCA patients based on their characteristics to identify an immunotherapy or combination therapy that is most effective.
This review highlights the various ways inter-tumoral heterogeneity occurs in CCA patients, focusing on immune cell abundance, anatomical location, driver mutations, etiologies, and TME subtype; what these differences mean to the immune microenvironment and how they can be used to stratify patients for the most effective immunotherapy or combination therapy.
CURRENT FOOD AND DRUG ADMINISTRATION (FDA) APPROVED TREATMENTS
For the last decade, the standard of care treatment for advanced CCA has remained to be systemic chemotherapy treatment of gemcitabine plus cisplatin [
8]. In recent years, novel indications involving immunotherapies have been FDA-approved as alternative first-line treatments. In September 2022, the combination of gemcitabine plus cisplatin together with durvalumab received FDA approval as a first-line treatment based on the TOPAZ-I trial. The addition of durvalumab to the traditional chemotherapy treatment expanded its objective response rate (ORR) from 18.7% to 26.7% [
9]. This combination became the first immunotherapy to be FDA-approved for CCA treatment. In 2023, the combination of gemcitabine, cisplatin, plus pembrolizumab was likewise approved based on the KEYNOTE-966 trial [
10].
Until recently, very little evidence suggested that CCA patients benefitted from second-line treatments due to the rapid exacerbation of the disease upon progression [
11,
12]. However, growing evidence suggests that additional chemotherapy offers survival benefits to patients. In the ABC-06 trial, FOLFOX (folinic acid, fluorouracil, and oxaliplatin) was able to extend the median OS to 6.2 months compared to 5.3 months with active symptom control alone [
12]. For patients with targetable mutations such as FGFR2 fusion, IDH1 mutation, and BRAF mutation, there has been a wave of targeted therapies that received FDA approval available as their option. In April 2020, the FGFR inhibitor pemigatinib became the first targeted therapy to receive FDA approval for CCA treatment. Based on the Phase 2 FIGHT-202 trial, pemigatinib achieved an ORR of 35.5% in patients with FGFR2 fusions or rearrangements previously treated and progressed on at least one round of treatment [
13]. In August 2021, ivosidenib was approved for patients with IDH1 mutation based on the ClarIDHy trial [
14]. Meanwhile, the combination of debrafenib (BRAF
V600E inhibitor) plus trametinib (MEK inhibitor) was approved in June 2022 for patients with BRAF
V600E mutation in advanced solid tumors, including BTC [
15]. Most recently, an FGFR inhibitor, futigatinib, was approved in September 2022, having achieved an ORR of 42% in the TAS-120-101 trial [
16]. Recent developments in CCA treatment illustrate targeted therapies and immunotherapies to be crucial for CCA treatment in the upcoming decade.
STRATIFICATION BASED ON STROMAL CELL TYPE ABUNDANCE
Multiple evidence suggests that the type of cells present in the TME can correlate with patient survival and immunotherapy efficacy [
17-
37]. Here we illustrate diverse cell types that occupy the TME of CCA tumors and the current findings on what their presence could mean to the TME (
Fig. 2). Stratifying patients based on the abundance of a particular cell type may represent a potential method to identify the key population to be targeted using immunotherapy.
TAMs are one of the most abundant immune cell types in CCA [
18]. Macrophages, though to a different extent, can be polarized to a pro-tumorigenic (sometimes referred to as alternatively activated, M2-like) or an anti-tumorigenic (also known as classically activated, M1-like) phenotype. However, in the TME, TAMs are generally pro-tumorigenic [
38-
40]. In the context of the liver, TAMs can either be monocyte-derived macrophages originating from the bone marrow, or liver-resident Kupffer cells originating from yolk-sac progenitors [
41]. Current evidence suggests that the high presence of TAMs in CCA is correlated with negative patient prognosis [
18-
21]. However, there are discrepancies between studies regarding which type of macrophage is responsible for this association. High CD68
+ total macrophage has been reported to be associated with worse OS and recurrence-free survival (RFS) in iCCA, while high CD163
+ macrophage has been associated with poor OS and disease-free survival (DFS) in eCCA [
18-
21]. There may be a disease subtype-specific difference in which group of macrophages could predict prognosis, which warrants further investigation.
Malignant CCA cells harbor various mechanisms to recruit TAMs to the tumor immune microenvironment (TIME). The myeloid cell recruiting chemokine GM-CSF is upregulated in both murine and human CCA, enabling the mobilization of monocytes from the bone marrow to the tumor site [
18]. Interestingly, the metabolism of CCA can also contribute to TAM recruitment. Fatty acid oxidizing enzyme ALOX5 is overexpressed in CCA. The metabolic product of ALOX5, LTB4 was shown to bind to M2-like macrophages, activate the PI3K-AKT pathway, and promote their migration [
40]. Meanwhile, the presence of TAMs can further promote their infiltration in a positive feedback loop. TWEAK expressed by TAMs was shown to bind to FN14 on CCA cells. The binding promoted secretion of the myeloid chemoattractant, MCP-1, in an NF-κB signaling-dependent manner [
42]. Blocking the mediators of myeloid recruitment was shown to be a promising therapeutic option in the above studies [
18,
40,
42].
The recruited TAMs are polarized to an immunoregulatory phenotype, which can hinder the activity of effector cell types through multiple means. First, TAMs can upregulate receptors that are detrimental to the activity and survival of effector cells. Macrophages co-cultured with CCA cells upregulate immunoregulatory markers such as ARG1 and CD206 [
18,
42]. Alternatively, TAMs can secrete and saturate the TME with anti-inflammatory cytokines. The hypoxic environment of CCA activated Sonic Hedgehog (SHH) signaling in the macrophages, which not only promoted an M2-like phenotype in the macrophages but also promoted their TGF-β1 secretion [
43]. Finally, TAMs can also recruit other immunosuppressive cell types. A single-cell RNA sequencing study on patient CCA samples identified activation of HIF-1α in APOE
+ TAMs. HIF-1α promoted a metabolic shift of TAMs towards fatty acid oxidation, which is the preferred mode of energy generation in immunosuppressive macrophages as opposed to pro-inflammatory macrophages that prefer glycolysis [
44]. Interestingly, these APOE
+ TAMs expressed CCL3 and recruited CCR5
+ Tregs, further exacerbating the immunosuppressive condition [
45].
The PD-1/PD-L1 immune checkpoint is perhaps one of the most well-known immune checkpoints. Many cancer cells upregulate PD-L1 to hinder effector T cell activity [
46]. Interestingly, in the context of CCA, the predominant source of PD-L1 expressions is TAMs rather than the cancer cells themselves [
39,
47-
49]. Moreover, in mouse models of CCA, PD-L1 expression on TAMs, rather than the cancer cells is more crucial in hindering anti-tumor immunity. Despite the high expression of PD-L1 on CCA cell line SB1 itself, orthotopic implantation of SB1 into PD-L1
-/- mice results in significantly reduced tumor burden compared to PD-L1 wild-type mice [
47]. CCA cells also harbor mechanisms to upregulate PD-L1 on surrounding TAMs. iCCA cell-derived exosomes contain increased levels of micro-RNA, miR-183-5p, which were shown to be endocytosed by macrophages to promote their PD-L1 expression through AKT signaling [
50]. CD47/SIRPα is another immune checkpoint that is frequently upregulated in CCA [
21,
51]. CD47, also known as the “don’t eat me signal”, is a receptor normally expressed on self-cells to prevent phagocytic cells, such as SIRPα expressing macrophages, from phagocytosing self-cells. However, many cancers upregulate CD47 to escape immune surveillance from TAMs. A high level of SIRPα
+ macrophages has been correlated with poor OS and DFS in CCA [
51]. Targeting TAMs expressing these checkpoints has been a promising therapeutic option for CCA. Although depletion of TAMs alone could not decrease tumor burden due to a compensatory increase in the granulocyte-like myeloid-derived suppressor cell (G-MDSC) population, the dual inhibition of TAMs and G-MDSCs was able to potentiate anti-PD-1 treatment in an otherwise anti-PD-1 resistant mouse CCA tumor [
47]. Meanwhile, CD47 or SIRPα-blocking antibodies have been shown to promote macrophage phagocytosis and decrease tumor burden in mouse CCA models [
51].
Other than altering the immune cells in the TME, TAMs can also directly promote malignancy through modulation of cancer cell invasiveness, stemness and proliferation [
38,
52-
54]. TAMs were shown to secrete high levels of WNT ligands, which were bound to FZD receptors on CCA cells to activate WNT signaling. Activation of this pathway promoted CCA cell proliferation via CTNNB1 downstream targets [
54]. The high level of oxidative stress in the CCA environment further contributes to the proliferative signal. Mechanistically, the high level of reactive oxygen species (ROS) in CCA tumors was associated with DNA damage, liver damage and TNF secretion by the liver resident macrophage, Kupffer cells. TNF induced CCA cell proliferation in a JNK signaling-dependent manner [
53]. Depleting the TAMs or inhibiting the downstream signaling were both effective in terminating the oncogenic proliferation and ameliorating CCA tumors [
53,
54].
Based on the above preclinical studies, for patients with high TAM presence, strategies to reduce TAMs may be considered. TAMs can be directly depleted through agents such as anti-CSF-1R or through inhibition of TAM recruiting chemokines such as anti-GM-CSF [
18,
40,
42]. The TAM depletion can be further combined with anti-PD-1 treatment, and patients should be further stratified for additional G-MDSC targeting in case of a compensatory increase in G-MDSC [
47]. Additionally, targeting other TAM-specific mechanisms, such as a CD47 inhibitor, or strategies to reprogram, such as SHH inhibition, represent promising avenues depending on the suppressive mechanism present in the patient [
43,
51].
Neutrophils are one of the most abundant leukocytes in the body, accounting for more than 50% of the leukocyte population [
55]. Physiologically, neutrophils are known to be first responders to infection. Due to their sheer abundance, neutrophils can migrate promptly to sites of infection to enact an inflammatory response. Key components of neutrophil functionality involve phagocytosis, release of cytotoxic granules and neutrophil extracellular traps (NETs) to eliminate the source of infection [
56]. In the cancer context, neutrophils have been reported to play both anti-tumorigenic and pro-tumorigenic roles. Neutrophils contribute to the inflammation needed for cancer initiation, as well as angiogenesis, metastasis and immune suppression needed for cancer expansion [
55]. Alternatively, neutrophils have also been linked to anti-tumorigenic effects through their cytotoxic function and antigen-presenting capacities [
55,
57]. Certain subsets of suppressive myeloid cells such as polymorphonuclear myeloid-derived suppressor cells (PMN-MDSCs) or G-MDSCs, are known to share surface markers (CD11B
+ CD15
+ CD14
- in human and CD11B
+ Ly6G
+ Ly6C
- in mice) with neutrophils and can only be distinguished by their suppressive functionality. Given the on-going discourse on whether neutrophils and MDSCs are distinct populations, they will all be included in this section. The naming in the original papers will be used to refer to the population.
Neutrophils are highly abundant in the CCA immune landscape compared to the surrounding normal tissue [
22,
58]. iCCA cells have been shown to recruit PMN-MDSCs through upregulation of CXCL8 in humans and CXCL5 in mice. Mechanistically, iCCA cells upregulated the N
7-methylguanosine tRNA methyltransferase, METTL1. METTL1-mediated tRNA modification upregulated CXCL8/CXCL5, which bound to CXCR2 on PMN-MDSCs to mediate their recruitment [
58]. The presence of neutrophils has been shown to be a negative prognostic marker in CCA patients in multiple studies. High CD15
+ neutrophil cell count in resected tumor sections has been associated with poor OS and DFS in CCA patients who have undergone curative resection [
22]. Moreover, a high neutrophil-to-lymphocyte ratio (NLR) in the circulation has been associated as a risk factor for OS independently in patients who underwent curative resection for iCCA, pCCA and dCCA [
23-
25]. A high NLR was also shown to be positively correlated with the presence of PD-1
+ exhausted T cells and negatively correlated with the presence of IFN-γ
+ effector T cells [
59]. Suggesting that the presence of neutrophils is not only associated with poor prognosis but also associated with an immunosuppressive TME. In a large cohort single-cell RNA sequencing study of primary liver cancer samples, five types of TIME were identified. iCCA was not only highly enriched in the immune suppressive myeloid type (TIME-ISM) but also significantly associated with high levels of neutrophi [
60]. The tumor-associated neutrophils (TANs) in this TIME type was associated with increased expression of PD-L1 and greater suppression of T cells [
60]. Targeting neutrophil infiltration into the tumor has been shown to not only reduce tumor burden but also potentiate the efficacy of immunotherapy. Total TAN depletion using anti-Ly6G alone was sufficient to attenuate tumor progression in CCA mouse models.60 Meanwhile, blocking neutrophil recruitment using CXCR2 blocking antibody was able to potentiate anti-PD-1 in mouse CCA tumors [
58].
Other than generating an immunosuppressive TME, neutrophils can also directly alter the malignant features of cancer cells. One factor that may contribute to the tumor-promoting effects of neutrophils is the NETs. NETs are web-like materials composed of decondensed chromatin DNA associated with antimicrobial proteins. Normally, NETs function to entrap infectious organisms and promote their clearing through proteases and granules [
61]. However, NETs can also act as a net to trap circulating cancer cells, promote their attachment to distant sites, and foster metastasis [
62]. In CCA, NETs have been shown to be highly abundant in tumor tissues compared to surrounding tissues [
63,
64]. iCCA cells promoted NET formation in neutrophils through a platelet dependent manner in both
in vitro and
in vivo models [
63]. On the other hand, NETs, more specifically, NET DNA promoted epithelial to mesenchymal transition (EMT) and angiogenic signature in CCA cells. Mechanistically, NET DNA interacted with ITGAV overexpressed on CCA cells and activated downstream NFκB signaling [
64]. A positive feedback loop exists where NET secretion is induced by CCA cells, and in turn, NETs promote the malignant features of the CCA. High NET score has been associated with worse OS and DFS in a Chinese cohort of patients; however, this association was not found in a Japanese cohort [
63,
64]. Meanwhile, targeting NET formation was able to reduce primary and metastatic CCA burden in mouse models [
63,
6].4 More evidence is needed to confirm whether NET presence can be used as a prognostic factor and a way to stratify patients for NET targeting therapies.
Cooperativity with TAMs is another key factor in the protumor effect of neutrophils. TAM inhibition alone has been shown to be unsuccessful in potentiating anti-PD1 treatment efficacy in CCA due to a compensatory increase in G-MDSC populations [
47]. Another study showed that conditioned medium from TANs and TAMs co-culture can enhance the proliferation, invasion and colony-forming ability of iCCA cells more robustly compared to conditioned medium from TANs or TAMs alone [
52]. Mechanistically, TANs and TAMs secreted oncostatin-M and IL-11, respectively, upon co-culture with iCCA cells. Both cytokines activated STAT3 signaling and enhanced the tumorigenicity of iCCA cells. The above studies highlight the potential synergistic effect between TANs and TAMs [
47,
52].
Given the preclinical result, similarly to TAM-rich CCAs, neutrophil-rich patients should be treated with neutrophil-depleting strategies, such as anti-LY6C or anti-CXCR2 antibody, combined with anti-PD1 [
58,
60]. Alternatively, their protumorigenic effects can be targeted through inhibiting NET formation [
63,
64]. Due to the cooperation between TAMs and neutrophils, strategies that target both populations, such as dual blockade of the two populations or targeting the common downstream pathway, appear promising [
47,
52].
Tregs are a group of immunosuppressive CD4
+ T lymphocytes characterized by their expression of the transcription factor FOXP3. Physiologically, Tregs maintain peripheral tolerance by preventing overactivation of effector T cells. The regulation can occur via targeting T cells directly through the secretion of inhibitory molecules, such as IL-10, or depletion of key effector molecules, such as IL-2, in the TME [
65]. Alternatively, they can also hinder dendritic cell (DC) function through the expression of inhibitory checkpoints, such as CTLA-4, which binds to co-stimulatory molecules such as CD80/CD86 to prevent effective antigen presentation [
65]. In the cancer context, such regulatory effects of Tregs can be exploited by the cancer cells to hinder anti-tumor immunity and support immune evasion.
In CCA, Tregs have been reported to be abundantly infiltrated in the tumor, while the extent of accumulation has been correlated with disease progression [
26-
28]. Interestingly, the extent of Treg accumulation is more significant in iCCA, with iCCA tumors showing greater infiltration of Tregs than eCCA tumors [
66]. Evidence suggests Treg accumulation in the CCA TME to be malignant cell dependent. The extent of Treg infiltration has been shown to be highly linked to the level of TGF-β1 expression by the cancer cells [
26]. Moreover, MUC1, a transmembrane glycoprotein overexpressed in CCA cells, was shown to interact with EGFR to activate the EGFR/PI3K/AKT signaling pathway. The overexpression of MUC1 and activation of the EGFR pathway led to increased polarization of Tregs [
27]. However, further investigation is needed to determine how exactly the EGFR pathway leads to Treg accumulation.
The infiltrated Tregs are highly suppressive with elevated expression of inhibitory molecules such as CTLA-4, TIGIT, TIM-3, and ENTPD1 [
28,
67]. The Tregs in the TME harbor a tumor-specific transcriptional program that drives this highly suppressive state. In a single-cell RNA sequencing study of six iCCA patient samples, tumor Tregs were shown to have increased activity of transcription factors needed for mediating immunosuppression such as IKZF2, IRF4 and BATF [
28]. One such transcription factor, MEOX1, was shown to be highly enriched in tumor Tregs and the overexpression of MEOX1 was sufficient to transform peripheral Tregs to acquire the epigenetic and transcriptional features of tumor infiltrating Tregs. Another study highlighted an enrichment of TIGIT-PVR-dependent communication between Tregs and iCCA cells, which represents another tumor-specific program driving this highly suppressive state [
67]. Interestingly, the presence of these highly suppressive Tregs was accompanied by lower infiltration of CD39
+ tumor-specific CD8
+ T cells and reduced activity of transcription factors needed for cytotoxic functions, thus highlighting the detrimental effect of Tregs on anti-tumor immunity [
28].
Consistently with the well-known tumor-promoting effect of Tregs, numerous evidences suggest a correlation between Treg infiltration into CCA tumor and patient prognosis. High levels of FOXP3
+ signal in resected tumor sections were associated with poor OS and progression-free survival (PFS) in multiple studies including iCCA, pCCA and dCCA [
26,
29,
30].
For patients with Treg-rich CCA, therapeutic strategies should focus on countering Treg-mediated immunosuppression. Immediate approaches include Treg-targeting ICB, such as anti-CTLA-4 or anti-TIGIT [
28,
67]. However, research into how Treg abundance predicts patient response to ICB is lacking and warrants further investigation. Complementary strategies include targeting upstream drivers of Treg accumulation, such as TGF-β signaling [
26]. Prospectively, inhibiting novel, tumor-specific regulators of Treg, such as MEOX1, could offer a more targeted method to reprogram the Treg-rich TME [
28].
T cells are the main effector cells of the anti-tumor immune response; their presence and cytotoxic function are critical for effective immunotherapy. Yet, T cells are also the cell type most often impacted by the harsh conditions of the TME. Understanding how and why the T cells are altered in the TME remains a key challenge in identifying effective immunotherapy.
Current evidence suggests CD8
+ T cells, particularly cytotoxic CD8
+ T cells, are decreased in CCA tumors compared to surrounding normal tissues [
66,
68-
71]. The scarce CD8
+ T cells that are present in the tumor core are exhausted with high expression of immune checkpoints [
66,
69,
70]. Multiple pieces of evidence suggest that the majority of cytotoxic CD8
+ T cells in CCA are sequestered at the tumor margin, illustrating CCA to be an immune-excluded type of immunologically cold tumor [
71,
72]. This exclusion may occur at the physiological level, through the formation of a dense ECM or through the generation of an immunosuppressive TME that impedes T cell survival and functionality. Unsurprisingly, the exclusion of CD8
+ T cells from the epithelial region of the tumor has been associated with worse OS in iCCA patients [
72]. Meanwhile, high CD8
+ T cell infiltration, particularly GZMB
+ cytotoxic CD8
+ T cells, is generally associated with better OS in CCA, illustrating how indispensable functional T cells are in maintaining anti-tumor immunity [
30-
33].
Contrary to CD8
+ T cells, the extent to which CD4
+ T cell infiltration differs between CCA tumors and non-tumor tissues remains a topic of debate that compels further investigation. In a Cytometry Time of Flight (CyTOF) based study on thirteen resected iCCA tumors with six paired non-tumor samples, the authors found a general increase in CD4
+ T cells in iCCA tumor tissue compared to the normal liver [
68]. Conversely, single-cell RNA sequencing conducted on two treatment naïve eCCA patients found a progressive decrease in CD4
+ T cell proportion from peripheral blood, para-tumor, to the tumor core [
69]. The different trends observed may be attributed to the difference in the cancer subtype. iCCA tumors have been reported to be more greatly infiltrated with CD4
+ Tregs as opposed to eCCA tumors; thus the increase in CD4
+ T cells in the first study may be attributed to the increase in Tregs [
66].
Expression of immune checkpoints remains a hindrance to antitumor immunity in multiple cancer types. Likewise, the upregulation of multiple immune checkpoints including PD-1, CTLA-4 and TIM-3 in tumor CD8
+ T cells compared to para-tumor CD8
+ T cells has been reported in CCA [
71,
73]. Interestingly, the expression of immune checkpoints has been reported to be higher in eCCA patients compared to iCCA patients, illustrating a tumor subtype-specific difference in immune checkpoint expression that needs to be considered when stratifying patients for ICB [
66]. The high expression of PD-1 in T cells has been associated with poor OS and advanced tumor stage in multiple cohorts of CCA; similar associations have also been reported in TIGIT and LAG3 expression [
30,
74-
76]. This indicates that T cell exhaustion is a significant negative prognostic marker for patient outcome. Targeting the upregulated immune checkpoints has been shown to be a promising therapeutic target in preclinical models.
Ex vivo treatment of nivolumab (anti-PD1) and ipilimumab (anti-CTLA-4) on patient-derived tumor infiltrating lymphocytes (TILs) was shown to enhance proliferation and cytotoxic activity of the TILs [
71]. In a patient-derived xenograft (PDX) model of eCCA, anti-TIGIT treatment was able to reduce tumor burden and increase IFNγ
+ GZMB
+ cytotoxic CD8
+ T cells and TCF1
+ progenitor exhausted T cells [
76]. Despite the tumor-specific enrichment of immune checkpoints and the success of ICB in preclinical models, the efficacy of single-treatment ICB in clinical trials has not been as promising [
6,
7]. The poor response is indicative of ICB resistance mechanisms that need to be further elucidated. One source of ICB resistance is the immunosuppressive TME. Irrespective of whether the inhibitory signals from the immune checkpoints are removed, the T cells are challenged with other extrinsic factors that impede their survival, functionality and anti-tumor effect.
The CCA cells themselves harbor mechanisms to contribute to the suppressive TME and dysfunctionality of tumor-infiltrating T cells. Interestingly, the intratumoral heterogeneity of CCA cells has been shown to be unfavorable to T cell functionality. In two transcriptomic studies of HCC and iCCA patient samples, iCCA tumor samples demonstrated a more heterogenous intratumoral transcriptome profile compared to HCC tumors [
77,
78]. Not only was higher tumor diversity associated with poor OS and PFS, but the T cells from the highly diverse tumors were also less inflammatory and cytotoxic. Notably, tumor samples from the high clonality group demonstrated greater interaction between the malignant cells and T cells [
78]. Although a more mechanistic investigation into this observation is needed, the high intratumoral heterogeneity of CCA tumors is a hinderance to T cell functionality.
Furthermore, CCA cells can also alter T cell abundance in the tumor by hindering T cell proliferation, promoting T cell apoptosis, and impeding T cell infiltration [
78-
81]. In the above-mentioned transcriptomic study of primary liver cancer patient samples, one of the most enriched interactions between cancer cells and T cells was the SPP1-CD44 pair [
78]. SPP1 encoding osteopontin is a ligand for CD44 expressed on T cells. In a single-cell RNA sequencing study of publicly available iCCA patient data, SPP1 was found to be overexpressed in tumor epithelium and associated with poor OS [
79]. Moreover, both tumor SPP1 level and T cell CD44 level were negatively correlated with T cell proliferation score [
79]. Alternatively, CCA cells can also directly induce T cell apoptosis. In an iCCA cell- peripheral blood mononuclear cells (PBMCs) co-culture system, iCCA cells were shown to increase expression of Fas and FasL to induce T cell apoptosis. Meanwhile, the iCCA cells themselves upregulated the anti-apoptotic regulator, c-FLIP, to evade apoptosis induced by the death signaling [
80]. Finally, T cell abundance could also be altered by hindering T cell infiltration into the tumor. iCCA cells were shown to reduce the expression of inflammatory cytokine, IFN-β, through downregulation of the cGAS-STING pathway [
81]. Mechanistically, malignant cells overexpressed the cholesterol biosynthesis enzyme NSDHL, which interacted with STING and promoted their degradation. The expression of NSDHL was negatively correlated with IFN-β expression and T cell infiltration [
81].
Other than altering T abundance, CCAs can also deter the inflammatory function of T cells. CD73 was shown to be upregulated in iCCA tumor tissues compared to peri-tumor [
82,
83]. CD73 converts adenosine monophosphate into adenosine in the extracellular space, which can bind to adenosine receptors on T cells to hinder their cytotoxic function [
84,
85]. High CD73 expression was associated with lower CD8
+ T cell infiltration, shorter OS and DFS [
82,
83,
86]. Mechanistically, iCCA cells upregulated circHMGC1016, which enhanced CD73 expression via regulating miR-1236-3p [
87]. Interestingly, CD73 was upregulated upon ICB treatment in both anti-PD1 resistant iCCA mouse models and patient samples through secretion of TNF-α and activation of NF-κB signaling [
83,
86]. CD73 inhibitor potentiated anti-PD1 treatment in anti-PD1 resistant mouse iCCA via increasing effector lymphocyte infiltration and reducing myeloid cell infiltration [
86]. Dedicating greater efforts to investigate changes in the TME following ICB will be crucial in combating ICB resistance and advancing the field of CCA tumor immunology to the next stage.
Other than conventional T cells, the infiltration of two specific subtypes of T cells has been associated with patient prognosis in CCA. Mucosal-associated invariant T (MAIT) cells are innate-like T cells that harbor semi-invariant TCR that can account for up to 45% of liver-infiltrating T cells in humans [
88,
89]. Physiologically, MAIT cells participate in anti-microbial immunity by recognizing microbial vitamin B metabolite presented on major histocompatibility complex (MHC) class I-related protein 1 (MR1) and secreting cytotoxic molecules such as IFN-γ, TNFα and granzyme B [
34,
88]. The role of MAIT cells in the cancer context remains contended as they exhibit both pro-tumorigenic and antitumorigenic activities in the TME [
90]. However, in CCA, high infiltration of MAIT cells has been linked with favorable antitumor gene signatures and increased OS [
34]. Notably, MAIT cells were often found on the tumor side of the tumor margin, which contrasts with conventional T cells which are often sequestered on the non-tumor side, underscoring the promising potential of MAIT cells in contributing to anti-tumor immunity [
70]. Despite this, like conventional T cells, the abundance and cytotoxicity of MAIT cells are gradually lost from peri-tumor to tumor [
34,
70]. Mechanistically, iCCA tumors were shown to contain high levels of bacteria species, resulting in continuous stimulation and depletion of MAIT cells [
34]. Given the high proportion of MAIT cells in the liver and its ability to localize to the tumor side of the invasive front, identifying mechanisms to restore their abundance and cytotoxicity may represent a promising field of research in CCA.
Another subtype of T cell that has been associated with prognosis in CCA is the tissue-resident memory T cells (Trms). Trm is a subset of memory T cells that reside long-term in non-lymphoid tissues through the expression of retention markers such as CD69 and CD103 [
91]. In iCCA, a high proportion of CD103
+ CD8
+ T cells was associated with increased OS and negatively correlated with tumor size and advanced pathological stage [
31]. Furthermore, in a cohort of iCCA specimens, CD103
+ CD8
+ T cells were shown to be enriched in tumor tissue compared to the stromal area. The extent of CD103
+ CD8
+ T cell infiltration was also associated with an elevated level of conventional CD8
+ T cell infiltration [
73]. Interestingly, Trms not only express high levels of immune checkpoints themselves but the extent of their infiltration was also correlated with the percentage of immune checkpoint-positive CD8
+ T cells [
31,
73]. Consequently, the presence of Trms may be an indicator of immune cell-rich tumors suitable for ICB, which probes further investigation.
CCA patients should be stratified based on the density, location, and functional state of their tumor-infiltrating T cells. Patients with high infiltration of functional tumor-infiltrating CD8
+ T cells are most likely to benefit from ICB [
31-
33]. However, given the low response rate to ICB monotherapy in clinical trials, many patients, such as those exhibiting T cell exclusion and exhaustion, would require additional combination therapies [
6,
7]. As evidenced by preclinical models, strategies to combat the immunosuppressive environment, such as a CD73 inhibitor, should be considered [
86]. Moreover, the presence of specific T cell subsets such as Trm or MAIT cells may identify patients with a more favorable prognosis and potential for ICB response, although further evidence is needed to validate their predictive power [
31,
34,
70,
73].
CCA tumors are known to be highly desmoplastic, with a dense layer of fibrotic stroma composed of ECM, CAFs and immune cells [
17]. This phenomenon is especially true when iCCA tumors are compared to their primary liver cancer counterpart, HCC, making CAFs an indispensable part of the CCA TME [
92]. In the liver, CAFs can either be derived from hepatic stellate cells (HSCs) or pulmonary fibroblasts [
93]. Current evidence in the field suggests that the majority of CAFs in CCA originate from HSCs [
35,
94].
Due to their highly heterogeneous nature, the role of CAFs in cancer is contended. CAFs can promote cancer by stimulating cell proliferation, inducing cancer stemness, and hindering immune cell infiltration into the tumor [
95]. However, CAFs have also been implicated in inhibiting tumor progression by acting as a barrier to tumor spreading [
35,
95]. In the context of CCA, the high presence of CAFs, particularly, FAP
+ or α-SMA
+ CAFs, have been associated with decreased OS and RFS in multiple cohorts [
35-
37,
72]. This association of CAFs with poor patient prognosis suggests a pro-tumorigenic role of CAFs. Pre-clinical studies in mice support this notion. The presence of CAFs was shown to promote cancer stemness, invasion and tumor growth in multiple
in vitro and
in vivo models of CCA [
67,
96,
97]. Meanwhile, the depletion of CAFs through selective targeting of CAFs resulted in reduced tumor burden [
98].
In the literature, CAFs have been proposed to promote tumorigenicity in CCA through two main mechanisms. One is through the modulation of the immune microenvironment. Particularly, CAFs have been reported to work cooperatively with MDSCs to modulate CCA cells. Firstly, the depletion of MDSCs has been shown to dampen the tumor-promoting effects of CAFs [
96,
97,
99]. While FAP
+ CAFs have been shown to enhance MDSC infiltration into the tumor through the CCL2-CCR2 axis in both iCCA patient samples and mouse models [
96,
99]. Mechanistically, FAP expression in CAFs activated FAK-Src-JAK2 signaling through uPAR and facilitated CCL2 secretion, which increased MDSC infiltration and reduced cytotoxic CD8
+ T cell infiltration into the tumor [
99]. Alternatively, CAFs can also modulate the tumor-promoting functions of MDSCs. CAF-derived IL-6 and IL-33 were shown to increase ALOX5 expression in MDSCs. ALOX5 upregulation resulted in the increased production of the metabolite LTB4, which was shown to bind to BLT2 on tumor cells to enhance their stemness properties in a PI3K-AKT-mTOR-dependent manner. Inhibition of BLT2 enhanced gemcitabine efficacy [
97]. Targeting CAFs may represent a novel therapeutic option to combat MDSC-dependent immunosuppression in CCA.
Additionally, CAFs can directly modulate the tumorigenicity of CCA cells. In CCA, CAFs have been reported to be the predominant population interacting with malignant cells [
35,
36,
67]. A single-cell RNA sequencing study on CAFs derived from human and murine CCA identified two major CAF subtypes: inflammatory CAFs (iCAFs) and myofibroblastic CAFs (myCAFs). iCAFs were shown to promote malignant cell proliferation in an ERK-dependent manner through the secretion of hepatocyte growth factor (HGF). Meanwhile, myCAFs were shown to upregulate the hyaluronic acid (HA) producing enzyme hyaluronan synthase 2 (HAS), resulting in the enrichment of HA in both human and mouse CCAs. Targeting HAS was able to reduce iCCA tumor burden but the detailed mechanism of how HA supports CCA progression remains unclear [
35]. Identification of disparate tumor-promoting mechanisms in different subtypes of CAFs highlights the importance of targeting CAF subtypes individually rather than as a single entity. Interestingly, ECM-modulating proteins can also have ECM-independent effects on CCA cells. Lysyl oxidase (LOX), a protein well-known for its ECM cross-linking function, was shown to be upregulated in CAFs. Rather than altering the ECM structure, LOX was shown to modulate mitochondrial assembly and enhance the metabolic fitness of CCA cells to promote their stemness, proliferation and tumor burden, highlighting the diversity of CAF functionality [
100].
IL-6 is a key cytokine of CAF-dependent tumor promotion. IL-6 is specifically upregulated in CAFs compared to normal fibroblasts and its high expression in the fibrotic area of iCCA tumors is correlated with shorter OS [
101,
102]. In a single-cell RNA sequencing study on CAFs derived from two iCCA patients, vascular CAFs (vCAFs), localized in the tumor core, were shown to be the major CAF subtype. vCAFs interacted significantly with malignant cells through IL-6 and IL-6R signaling. Mechanistically, CAF-derived IL-6 upregulated EZH2 and enhanced the stemness of CCA cells. In turn, miR-9-5p present in tumor-derived exosomes promoted IL-6 expression in CAFs, creating a feedback loop [
67]. Moreover, IL-6 has also been implicated in chemotherapy resistance in CCA by hindering autophagy [
101-
103]. Patients with high expression of autophagy flux were shown to have better prognosis and response to adjuvant chemotherapy [
102]. Conversely, high IL-6 expression was associated with lower autophagy flux and lower response to chemotherapy in both CCA patients and in
in vitro models [
101-
103]. The blockade of IL-6 signaling was shown to disrupt CAF-CCA interactions and sensitize CCA cells to chemotherapy, highlighting IL-6 and IL-6R as a potential therapeutic target [
101].
Due to the highly heterogenous nature of the CAF population, broad depletion strategies for CAFs are challenging. Consequently, patients with CAF-rich tumors should be further stratified using biomarkers such as IL-6 and MDSC infiltration for precise downstream targeting. Patients with high MDSC infiltration may benefit from MDSC depletion or CCL2-CCR2 blockade [
96,
97,
99]. IL-6/IL-6R inhibition appears promising in combating the effect of CAF on both immune cells and on tumor cells [
67,
97,
101-
103]. Other strategies should target specific CAF subtypes based on their oncogenic signaling, such as inhibiting HAS in myCAF-rich tumors [
35]. Given CAF’s involvement in immunoregulation and chemotherapy resistance, the CAF combating strategies should be further combined with immunotherapy and chemotherapy [
97,
101].
Emerging evidence has established B cells as critical contributors to anti-tumor immunity. High densities of tumor infiltrating B cells have been associated with positive prognosis and better response to ICB in multiple cancer types [
104]. This prognostic value is also evident in CCA, where increased B cell density and infiltration have been associated with better survival in two independent studies [
30,
105]. Moreover, the functionality of peripheral B cells has prognostic value for predicting immunotherapy response. A single-cell RNA sequencing study of PBMCs from patients treated with gemcitabine, cisplatin, plus durvalumab revealed that responders had a higher proportion of B cells expressing activation markers such as BAFFR and CD21 compared to non-responders [
106].
Like other immune cell types, B cells are also negatively impacted by the immunosuppressive nature of the CCA TME. Single-cell RNA sequencing of tumor, peri-tumor, and blood from iCCA patients demonstrated reduced B cell infiltration in the tumor core. The scarce B cells present exhibited a dysfunctional phenotype, characterized by reduced activation, impaired differentiation, and loss of effector function. Mechanistically, TGF-β and IL-6 secreted by iCCA cells and CAFs hindered B cell maturation, increased the frequency of abnormal B cells subsets, and promoted expression of immunosuppressive markers such as IL-10. Importantly, dual inhibition of TGF-β and IL-6 was shown to restore B cell functionality [
106].
Given the evident role of B cells in predicting immunotherapy response, CCA patients could be stratified based on B cell functionality. Patients with an activated B cell population may be an optimal candidate for treatment with gemcitabine, cisplatin, plus durvalumab. In contrast, patients with a dysfunctional B cell profile may benefit from a combination of TGF-β and IL-6 blockade to restore B cell functionality prior to or concurrent with immunotherapy.
Natural killer (NK) cells
NK cells are a type of innate lymphoid cell specialized for cytotoxicity. Their ability to recognize and eliminate cancer cells that evade T cell surveillance, such as through MHC class I downregulation, makes them an indispensable component of the anti-tumor immune response in complementing T cells [
107]. Accordingly, impaired NK cell function is associated with increased cancer susceptibility and metastasis across many cancer types and preclinical models [
108].
However, in CCA, the potent anti-tumor potential of NK cells is severely compromised. Despite their prominent role in anti-tumor immunity, no prognostic association was found between NK cell abundance and patient outcome, indicating profound suppression of NK cell functionality in the CCA TME [
30]. Multiple factors contribute to this. NK cells are consistently excluded from the tumor and sequestered at the tumor margins [
70,
71]. A single-cell RNA sequencing analysis of eCCA confirmed a progressive decline in NK cell proportion from PBMC to peri-tumor to tumor core [
69]. The few NK cells that do infiltrate the tumor exhibit functional abnormality, characterized by low expression of activation molecules such as NKG2D and effector molecules like perforin [
71,
109].
A mechanism driving this NK cell dysfunction is the dysregulation of MICA/B- NKG2D axis. The NKG2D ligands, MICA/B, were shown to be upregulated in iCCA patient samples. MICA/B are upregulated under cellular stress, such as viral infection or malignant transformation, to enable NK cell recognition. However, cancer cells evade NK recognition by proteolytically shedding MICA/B through metalloproteinases like ADAM10/17. Furthermore, the resulting soluble MICA/B (sMICA/B) acts as a decoy for NKG2D receptors, preventing ligand binding and downregulating NKG2D expression. Consistent with this, iCCA patients exhibited elevated levels of ADAM10/17 in tumor and sMICA/B in serum, which correlated with reduced intratumoral NK cell infiltration and NKG2D expression [
109]. Therapeutic strategies targeting this pathway are promising. A MICA/B targeting antibody 7C6 inhibits MICA/B shedding without blocking NKG2D binding. In ex vivo studies, 7C6 treatment enhanced NK cell degranulation, IFN-γ production, and tumor cell killing, effectively restoring their anti-tumor function [
109].
Given the above evidence, ICB alone may be insufficient to reactivate the deeply dysfunctional NK cell in the CCA TME. Patients with high NK cell infiltration should be grouped for strategies that overcome NK evasion mechanisms, such as soluble MICA/B inhibitors, before being subjected to ICB. Further investigation into the relationship between NK cell infiltration, response to immunotherapy, and the efficacy of NK-targeting agents in combination with immunotherapies is warranted.
DCs
Conventional dendritic cells (cDCs) are the primary antigen-presenting cells responsible for T cell activation. Unsurprisingly, their presence is positively correlated with better patient outcomes and CD8
+ T cell infiltration in CCA [
110,
111]. Moreover, the functional status of the cDCs is tightly linked with the exhaustion status of the CD8
+ T cells. Higher proportion of SIGN
+ immature DCs correlated with an increased frequency of CD8
+ T cells expressing exhaustion markers such as TOX2, EOMES, and LAG3. Similarly, a higher density of PD-L1
+ DCs correlated with more PD-1
+ CD8
+ T cells [
30].
The critical role of functional cDCs is further highlighted in the context of lymph node metastasis (LNM). The presence of CD83
+ mature DCs is negatively correlated with the presence of LNM in CCA [
111]. Consistent with this, a single-cell RNA sequencing analysis of 25 treatment-naïve iCCA patients found that primary tumors of patients with LNM had lower infiltration of both DCs and CD8
+ T cells. The DCs present exhibited reduced antigen-presenting capability and lower expression of class II MHC molecules [
110]. A key mechanism suppressing DC recruitment in CCA is Wnt/β-catenin signaling. Wnt/β-catenin signaling is upregulated in iCCA and transcriptionally represses the DC-recruiting chemokine CXCL12. Therapeutically, strategies to expand and mature DCs through FLT3L and TLR3 agonist, as well as β-catenin inhibition both successfully rescued DC and CD8
+ T cell infiltration and reduced the incidence of LNM in preclinical models [
110].
In contrast to the anti-tumorigenic role of cDCs, plasmacytoid DC (pDC) infiltration in the peri-tumoral region has been associated with worse prognosis in iCCA patients. The higher abundance of pDC also correlated with increased Treg infiltration, likely driven by an inflammatory environment involving pDC-derived IFN-γ [
112].
The DC population is a critical determinant of the immune contexture in CCA. For patients with features of DC dysfunction or LNM, DC-targeting therapies, such as agents that promote expansion and maturation of DCs, including FLT3L and TLR3 agonists, or agents that enhance DC recruitment, such as β-catenin inhibition, represent a promising strategic avenue.
STRATIFICATION BASED ON ANATOMICAL LOCATION
The anatomical origin of CCA represents a significant source of heterogeneity, offering a potential strategy for patient stratification. CCA patients may respond differently to immunotherapies depending on their CCA subtype.
Initial findings from phase II trials of ICB monotherapy suggested a trend towards higher ORR in eCCA compared to iCCA. Specifically, nivolumab monotherapy resulted in ORRs of 40% in eCCA versus 21% in iCCA, while pembrolizumab showed rates of 25% and 5%, respectively [
6,
113]. However, these findings were based on a small cohort of patients, with the total number of iCCA patients significantly larger (n=28 and 20) than eCCA patients (n=5 and 8), necessitating validation in larger cohort studies. Interestingly, this pattern does not hold for combination immunotherapy. A subgroup analysis of a phase II trial of nivolumab and ipilimumab combination therapy found responses only in iCCA and GBC patients but not in eCCAs patients [
114]. This may be attributed to the high infiltration of Tregs in the iCCA TME, which can be targeted by ipilimumab [
66].
On the other hand, the effect of anatomical location on first-line chemo-immunotherapy regimens remain unclear. In the TOPAZ-I trial, the addition of durvalumab to chemotherapy provided a greater PFS benefit in eCCA patients (hazard ratio=0.52) than in iCCA patients (hazard ratio=0.79) [
9]. Conversely, in the KEYNOTE-966 trial, the OS benefit from adding pembrolizumab was significant for iCCA (hazard ratio=0.76) but absent in eCCA (hazard ratio=0.99) [
10]. The disparity between these results highlights the complexity of predicting response by anatomical location alone.
These conflicting conclusions suggest that anatomical phenotype per se may be an insufficient predictor of ICB efficacy. Not only is a more systematic investigation is warranted, but additional factors such as underlying molecular phenotypes and etiologies associated with each CCA subtype need to be considered.
STRATIFICATION BASED ON DRIVER MUTATIONS
CCA is a mutationally distinct disease with more than 40% of patients harboring driver mutations that can be targeted [
115]. The most common genomic alterations found in CCA include TP53, IDH1/2, KRAS, BAP1, ARID1A, BRAF, EGFR, FGFR2 fusion and ERBB2 amplification. Interestingly, a significant correlation is found between specific driver mutations and the anatomical location of the tumor. ARID1A, IDH1/2 and BAP1 mutations along with FGFR2 fusions are more frequently found in iCCA, while ERBB2 amplifications are more common in eCCA [
1,
115].
Recent evidence suggests a significant correlation between driver mutations and tumor immune signatures, illustrating the promising prospect of driver mutation-based patient stratification for identifying effective immunotherapies (
Fig. 3) [
116,
117]. In recent years, several small-molecule inhibitors targeting specific driver mutations were FDA-approved for CCA treatment [
13-
16]. Identifying immunotherapies that work synergistically with targeted therapy represents another encouraging area of exploration.
Mutation in the tricarboxylic acid cycle enzyme, IDH1 results in the production of an oncometabolite, R-2-hydroxylglurate (R-2HG) [
118]. Traditionally, R-2HG was known to mediate tumorigenesis by acting as a competitive inhibitor for α-ketoglutarate (α-KG) dependent enzymes and inducing metabolic and epigenetic changes. More recently, R-2HG was reported to be exported to the extracellular space to promote a pro-tumorigenic immune landscape in other cancer types [
119-
122]. In line with this result, IDH1 mutant CCA patient samples have been reported to harbor a colder TME with reduced immune cell infiltration [
123-
125].
IDH1 mutation has been described to have an increased infiltration of M2-like macrophages into the tumor in both patients and mouse models by overexpressing the macrophage-recruiting cytokine CCL2 [
126]. CCL2 neutralization reduced tumor burden by decreasing CD206
+ macrophages and increasing CD8
+ effector T cells, illustrating the critical role of CCL2 in mediating immunosuppression. However, how CCL2 is upregulated in IDH1 mutant tumors remains unclear and warrants further investigation.
Multiple studies in the glioma context have elucidated the role of R-2HG in hindering effector T cell function [
119,
120]. Mechanistically, R-2HG was shown to inhibit ATP synthase and the electron transport chain [
119]. The decrease in intracellular ATP level perturbed calcium signaling, resulting in inhibition of NFAT translocation and attenuation of TCR signaling in an AMPK and PLC-γ dependent manner. In another study, R-2HG was shown to act as an inhibitor of LDH, resulting in a reduction in glycolysis needed for T cell activation and proliferation [
120]. Similarly, in CCA, R-2HG has been reported to directly suppress T cell function by limiting T cell metabolic fitness, resulting in reduced cytokine production and effector function [
127]. Moreover, R-2HG was also shown to limit anti-tumor immunity through tumor-dependent mechanisms [
127,
128]. R-2HG inhibits TET2 DNA demethylases in tumor cells, thereby suppressing interferon response element and transposon element-dependent generation of double-stranded DNA and cGAS-STING signaling. This leads to a dampened immune response, including both tumor-intrinsic type I interferon response and immune cell-dependent type II interferon response. Clinically, high tumor R-2HG levels have also been associated with a lower number of TILs in patient samples from the ClarIDHy trial [
124]. Notably, in mouse models, the increased infiltration and activation of CD8
+ T cells upon mutant IDH1- specific inhibitor AG120 treatment was accompanied by an upregulation of immune checkpoints including PD-1 and CTLA-4 [
127]. Combining AG120 with anti-CTLA-4 treatment has shown synergistic effects in reducing tumor burden, suggesting a promising therapeutic strategy. Furthermore, IDH1 mutant patient samples have also been associated with increased presence of CAFs [
123,
129]. However, more mechanistic studies are needed to elucidate the role of R-2HG in this phenomenon.
FGFR2 fusion is another common genomic alteration found in CCA. FGFR2 is a receptor for fibroblast growth factors [
130]. Upon ligand binding, FGFR2 dimerizes and initiates downstream signaling such as the MAPK pathway and PI3K-AKT-mTOR pathway. In FGFR fusion protein expressing CCA, the C-terminal of the FGFR2 protein is fused with a partner protein [
131]. The partner proteins contain constitutive dimerization motifs allowing the fusion protein to undergo ligand-independent dimerization and activation, resulting in the overactivation of downstream signaling, cell proliferation and malignant transformation. Multiple partner proteins have been reported, including BICC1, AHCYL1, PPHLN1, and more, amongst which fusion with the BICC1 protein is the most common [
130].
FGFR2 fusion expressing tumors have been associated with reduced immune infiltration, particularly in T cells and macrophages, and downregulation of immune-related gene signatures such as genes related to chemotaxis, T cell cytotoxicity and exhaustion [
116,
129]. The expression of FGFR2: BICC fusion protein was sufficient to recapitulate the reduction of immune cell infiltration in hydrodynamic tail vein injection-based mouse iCCA models [
116]. Additionally, FGFR2 fusion-positive patient samples were demonstrated to have lower expression of PD-L1 on tumor cells [
132]. However, how FGFR2 fusion proteins elicit such effects calls for additional investigation.
Evidence suggests that the neoantigen generated from the FGFR2 fusion protein can be recognized by TILs. T cells cultured under the presence of antigen-presenting cells presenting the fusion peptide from FGFR2: BICC and FGFR2: TDRD1 proteins were shown to upregulate the expression of effector and activation molecules, suggesting a peptide-dependent activation of the T cells [
116,
133,
134]. T cell-based therapy using FGFR2 fusion neoantigen-specific T cells appears to be a promising therapeutic option. However, why FGFR2 fusion tumors remain poorly infiltrated with effector T cells despite the presence of neoantigen remains an important question. T cells in FGFR2 tumors are likely to be excluded from the TME rather than inactive. Understanding the mechanisms behind this exclusion and developing effective T cell-based therapies are key research areas in FGFR2 fusion tumors.
KRAS is one of the most frequently mutated genes in cancer with nearly 20% of all solid cancers harboring this mutation [
135]. The mutation occurs in multiple types of cancer including pancreatic cancer, colorectal cancer, and lung cancer, as well as CCA. This mutation in the small GTPase leads to the overactivation of the MAPK signaling, promoting cancer cell proliferation and tumorigenesis. Despite its prevalence, KRAS mutation was considered “undruggable” for many years due to the difficulty in identifying a binding pocket for small molecule inhibitors. Consequently, KRAS mutant tumors were often targeted at the downstream level through RAF, MEK and ERK inhibitors [
136]. Recently, multiple pan-KRAS inhibitors and KRAS
G12C and KRAS
G12D specific inhibitors were identified and revolutionized the field of cancer therapy [
135].
Unlike IDH1 mutant and FGFR2 fused CCA tumors, KRAS mutated tumors are highly inflammatory due to the overactivation of the MAPK signaling pathway [
136]. This proinflammatory signature results in the secretion of inflammatory molecules such as IL-1B and CXCL3, the accumulation of myeloid cell types and reduced infiltration and exhaustion of lymphocytes [
116,
117,
137-
139]. KRAS mutant tumors have been demonstrated to harbor mechanisms to downregulate this inflammation [
137]. In a multi-omics study of iCCA patient samples, KRAS mutant tumors were shown to have an increased alternative splicing event of IL-1 receptor antagonist (IL1RN), resulting in the upregulation of IL1RN-201 and IL1RN-203. IL1RN-201/203 counteracted ERK1/2-CXCL3- dependent neutrophil recruitment. Based on this endogenous negative feedback, an IL-1R antagonist (anakinra) was applied to KRAS mutant mouse iCCA. Anakinra reduced tumor burden by reducing neutrophil infiltration and increasing cytotoxic T cell infiltration, illustrating the beneficial role of myeloid cell targeting in KRAS mutant tumors.
Increased MAPK signaling in KRAS mutant cells has been associated with an increased expression of PD-L1. Mechanistically, ERK was shown to stabilize PD-L1 expression by inhibiting autophagy and preventing autophagy-mediated degradation of PD-L1. ERK inhibitor increased the cytotoxicity of T cells co-cultured with iCCA cells with endogenous KRAS mutation [
140]. In accordance with this, treatment of KRAS-SOS1 inhibitor potentiated anti-PD1 efficacy in a hydrodynamic tail vein injection-based KRAS mutant iCCA model and patient-derived T cell-organoid coculture system [
139]. However, the responsiveness of KRAS mutant tumors to PD-1 monotherapy remains contentious. In an orthotopic KRAS mutant mouse model, anti-PD-1 was effective in reducing tumor size but not anti-PD-L1 [
141]. KRAS mutant iCCA patients have also been shown to have an increased response to anti-PD-1 treatment. In the previously mentioned study on increased IL1RN alternative splicing in KRAS mutant patients, iCCA patients harboring high levels of IL1RN were associated with improved response to anti-PD-1 [
137]. More systematic studies should be conducted to investigate whether stratifying patients based on KRAS mutation status would be beneficial in identifying those who are responsive to ICB.
EGFR mutation is another common mutation found in ~15% of CCA patients [
142]. Other than having a mutation on the receptor itself, the EGFR pathway is activated in 20–30% of CCA patients through increased release of EGFR ligands [
143]. EGFR targeting therapies have been FDA-approved for treatment in other types of cancer such as non-small cell lung cancer and colorectal cancer [
144,
145]. However, the effect of EGFR targeting in CCA is subpar. Multiple clinical trials of EGFR inhibitors such as panitumumab and erlotinib in combination with chemotherapies show little to no survival benefit over chemotherapy alone [
146-
148]. Although the underlying resistance mechanism remains unclear and multiple factors are likely to play a role, the TIME could be a contributor to this resistance. Currently, the TIME of EGFR signaling activated CCAs is understudied. Exploration into the immune landscape of EGFR-altered tumors and whether the efficacy of EGFR inhibitors can be enhanced with immunotherapy remains a potential field of research.
STRATIFICATION BASED ON ETIOLOGIES
The etiologies of CCA are diverse; the risk factors vary largely depending on the geographical location. For instance, CCAs related to infectious agents such as liver fluke, HBV or HCV are more commonly found in Asia, while factors such as PSC and NAFLD are more prevalent in the West [
149,
150]. Moreover, the etiologies of CCA subtype also differ; cirrhosis and HBV infections are more linked with iCCA, while PSC is more correlated with eCCA [
151,
152].
Certain etiologies are also associated with specific driver mutations. IDH1/2 mutation and FGFR2 fusion are predominantly found in non-fluke-infected CCAs, while TP53 mutation and ERBB2 amplification are more commonly found in liver-fluke-infected samples [
2,
153]. Genomic alterations in PSC-related CCAs more closely resemble those in eCCA, with an enrichment of alterations in TP53, KRAS and ERBB2 [
154].
Despite the diverse etiologies, a common feature among CCA is chronic inflammation of the bile duct. Prolonged inflammatory signaling and immune cell activation have significant impacts on the immune landscape. In fact, the various etiological factors associated with CCA have been linked to distinct TIME (
Fig. 3). Identifying the individual patient’s underlying etiology may represent another means of stratifying patients for tailored immunotherapies.
Liver fluke infection is a common cause of CCA in Asian populations. Infection with platyhelminths such as
Opisthorchis viverrine and
Clonorchis sinensis typically occurs through consumption of undercooked freshwater fish and is believed to mediate carcinogenesis through three primary mechanisms: physical damage, excreted/secreted products (ESPs) and chronic inflammation [
149].
Liver fluke infection has been linked to poor OS as well as reduced PFS in patients undergoing anti-PD1 treatment, suggesting that liver fluke infection induces changes in the immune landscape that decrease the efficacy of immunotherapy [
153,
155]. Notably, liver fluke infected tumors have been shown to exhibit increased tumor mutational burden (TMB), likely due to genomic instability caused by chronic inflammation. The high TMB is accompanied by increased expression of immune-associated genes, T cell clonal expansion, and exhaustion [
153,
155]. The resistance to immunotherapy despite the presence of T cells suggests that tumor immune surveillance is impaired, potentially through suppression of T cell cytotoxicity. An example of this is demonstrated in a single-cell RNA sequencing study of liver fluke-infected iCCA samples, where fluke infection was associated with TAM-dependent induction of T cell exhaustion. Mechanistically, liver fluke ESPs were found to induce polarization of TAMs to a pro-tumorigenic phenotype via FASN-dependent upregulation of fatty acid synthesis in malignant cells. These TAMs interacted with T cells through TIM-3 and PD-1, triggering an inhibitory signal. Inhibition of FASN potentiated anti-PD-1 treatment by reducing the tumorigenic phenotype of TAMs and increasing T cell cytotoxicity [
155]. Addressing the immunosuppressive TME alongside anti-PD-1 therapy appears to be critical to achieve effective treatment outcomes in liver fluke-infected CCA patients.
HBV infection is another predisposition to CCA, especially in Asia. Like HCC, HBV infection is believed to mediate CCA pathogenesis through genomic alterations induced by viral DNA integration and the establishment of liver cirrhosis via chronic inflammation [
3,
156]. Interestingly, HBV-infected iCCA patients have shown better OS and RFS following curative resection and increased immune cell infiltration, particularly CD8
+ T cells, into the tumor [
116,
117,
157,
158]. However, CD8
+ T cells in HBV-infected tumors are often exhausted, expressing higher levels of PD-1 and TIGIT compared to CD8
+ T cells from non-HBV-infected tumors [
158,
159].
In the context of HCC, it was recently identified that viral-HCCs were more responsive to immunotherapy compared to non-viral HCCs [
160]. This has generated significant interest in determining whether HBV-positive CCAs are also more responsive to ICB. However, such studies are lacking for CCA. A multi-omics study revealed that the gene signature of HBV-infected iCCA is more similar to that of ICB-responsive tumors [
117]. Moreover, both PD-1 and PD-L1 expressions have been correlated with HBV infection in iCCA patient samples [
159]. Together illustrating the potential for HBV-infected iCCA samples to respond to ICB. However, within the HBV-infected patient cohort, those with lower PD-1 expression responded better to ICB, creating ambiguity about whether the higher PD-1/PD-L1 positivity in HBV-positive iCCA patients is indicative of a favorable response to ICB [
159]. Therefore, a systematic study assessing the efficacy of ICB in HBV-positive versus HBV-negative CCA patients is warranted.
In contrast to liver fluke and HBV infections, PSC-related CCAs are more commonly found in Western countries. PSC is a chronic disease that results in progressive damage and scarring of the bile duct. Persistent inflammation and liver fibrosis associated with PSC have been implicated in contributing to CCA pathogenesis [
150]. Accordingly, mouse models of PSC have demonstrated an increase in both tumor initiation and tumor burden of CCA [
161,
162].
PSC has frequently been linked with intestinal barrier dysfunction with 60–70% of PSC patients also suffering from colitis or Crohn’s disease [
150]. This connection has raised interest in the role of the gut microbiome in the TIME of PSC-related CCAs. In PSC mouse models, PSC harboring mice were shown to have an increased presence of gram-negative bacteria and LPS in the liver [
161]. The gram-negative bacteria-derived LPS promoted PMN-MDSC infiltration into the tumor in a TLR4-CXCL1-dependent mechanism. Meanwhile, targeting MDSCs or TLR4 activation reduced CCA tumor burden. Modulation of the gut-liver axis appears to be a crucial component in identifying potential therapeutics for PSC-related CCAs.
STRATIFICATION BASED ON TERTIARY LYMPHOID STRUCTURES (TLS)
An additional factor that can be considered while stratifying patients is the presence of TLS. TLS are aggregates of immune cells in non-lymphoid tissues that allow local antigen presentation. TLS formation occurs in the presence of persistent inflammation such as autoimmunity, chronic infection, or cancer [
163,
164]. Thus, the presence of TLS could be an indicator for immune cell infiltration, effective anti-tumor immunity and overall TME status. In line with this, in the context of cancer, TLS has generally been associated with a positive prognosis and enhanced response to immunotherapy [
163].
Like many other cancer types, the presence of TLS in CCA is associated with better prognosis in multiple cohorts [
66,
164-
168]. However, additional factors play critical roles in this association. Two separate studies highlighted the prognostic ability of TLS to be largely dependent on the position of the TLS within the tumor. A high number of TLS in the intra-tumor region of CCA (represented by the T-score) was associated with favorable prognosis and positive clinicopathological features. Conversely, the high presence of TLS in the peri-tumor region (represented by the P-score) was associated with worse prognosis, LNM, and satellite lesions [
164,
165]. A likely explanation for this association is that the presence of TLS in the intratumoral region is indicative of immune cell infiltration into the tumor, while TLS in the peri-tumoral region suggests a state of immune exclusion. Supporting this hypothesis, the number of Tregs in intratumoral TLS was highly correlated with the patient’s P-score, indicating a link between the suppressiveness of the TME and the presence of TLS in the peri-tumor region [
164].
The maturation state of the TLS also significantly impacts its prognostic value. TLS can vary in maturation status depending on their structural resemblance to secondary lymphoid organs. Lymphoid aggregates represent the least mature TLS, lacking B and T cell zone organization and follicular DCs. More mature, follicle-like TLS harbor follicular DCs and B and T cell zone segregation; these can be further divided into primary and secondary follicle-like TLS depending on the presence of germinal center reactions [
163]. When iCCA tumors were classified based on the maturation states of their TLS, samples with more follicle-like TLS had better prognoses as opposed to those with lymphocyte aggregates [
164]. Additionally, in a cohort of pCCA patients, only the presence of intratumoral secondary follicle-like TLS was shown to be associated with better OS and RFS [
168]. As the maturation status of TLS is reflective of their functional capacity, mature TLS are more capable of antigen presentation and positively impact anti-tumor immunity and thus be associated with better prognosis.
Beyond predicting patient survival, the presence of TLS has also been implicated in immunotherapy response. In CCA patients undergone chemotherapy plus immunotherapy, TLS-positive patients exhibited improved treatment response and survival [
165,
166]. Moreover, TLS-positive tumors displayed greater vein density, TIL infiltration and TIL exhaustion, suggesting that TLS is a promising indicator for tumors with functional tumor surveillance upon ICB [
166,
167]. However, it remains unclear whether TLS presence contributes to functional anti-tumor immunity or if TLSs are formed because of it. More mechanistic studies on the relationship between TLS, anti-tumor immunity, and ICB response are warranted.
There are subtype-specific differences in the prevalence of TLS in CCA. In a multiomics study of BTC patient samples, TLS was enriched in eCCA compared to iCCA or GBC, most likely due to the high expression of CXCL13, a chemokine known to mediate TLS formation [
66]. Accordingly, eCCA patients with high CXCL13 expression were associated with increased levels of TLS, improved response to ICB and extended OS. Stratifying patients according to their TLS status represents a promising method for identifying those who may benefit from immunotherapies. However, careful stratification considering TLS position, maturation and tumor subtype is essential.
STRATIFICATION BASED ON TIME PROFILE
The microenvironment-based classification of iCCA has suggested that immune contexture is associated with specific driver gene mutations, etiologies and immune cell infiltration [
139]. Stratifying patients based on their TIME profile represents a method to combine the above findings regarding cell type abundance, driver mutations and etiologies to evaluate a patient’s TIME more holistically to identify effective treatment options.
Based on the level of tumor-infiltrated immune cells, iCCA could be classified into two broader classes defined by high immune infiltrates namely “inflamed iCCA” (35% of all iCCA) and low immune infiltrates namely “non-inflamed iCCA” (65% of all iCCA) [
139]. Inflamed iCCA is composed of “immune classical class” and “inflammatory stroma class” while non-inflamed iCCA is composed of “hepatic stem-like class”, “tumor classical class”, and “desert-like class”. “Hepatic stem-like class” refers to the most frequently found subcluster of iCCA exhibiting stem-cell-like features. “Tumor classical class” represents iCCA tumors that demonstrate high expression of cholangiocyte markers and enrichment of cell cycle pathways, while “desert-like class” represents immune desert tumors characterized by the least amount of immune cell enrichment among the subclusters. Generally, “inflamed class” presents higher number of CD8
+ T cells than “non-inflamed class”. FOXP3 staining, indicating FOXP3
+ Treg cells, is found to be higher in “desert-like class”. Immune checkpoints are found to be highly expressed in inflamed classes, specifically the “inflammatory stroma class”. Interestingly, it was found that iCCA with only KRAS mutations are in the “inflammatory stroma class”. Meanwhile, iCCA with BAP1 mutations, FGFR2 fusion genes, and IDH1/2 mutations are found to be in the “hepatic stem-like class”. iCCA with co-occurrence of TP53 and KRAS mutations is the “tumor classical class” featured by increased Treg cells. Interestingly, based on this classification, it is expected that “inflamed class” iCCA is more suitable for ICB treatment. Specifically, “immune classical class” should be treated with anti-PD-1/PDL1 alone while “inflammatory stroma” which is enriched with KRAS mutation alone should be treated with KRAS inhibitor and anti-PD-1. Meanwhile, the non-inflamed class should be targeted to overcome underlying immunosuppressive mechanisms using tailored combination strategies [
139].
NOVEL THERAPEUTIC OPTIONS
Due to extensive preclinical research and numerous multi-omics studies on patient samples, multiple novel treatment options for CCA have emerged. Here we highlight these novel treatments, which can be used alone or in combination with currently available therapies for CCA (
Table 1).
Multiple studies have highlighted that the myeloid compartment of CCA is a key mediator of immunosuppression and immunotherapy resistance. Indeed, preclinical studies have demonstrated that perturbation of TAM and TAN infiltration through inhibition of myeloid recruiting chemokines and receptors reduces tumor burden and enhances ICB efficacy [
18,
42,
47,
54,
58,
161]. Examples of targetable chemokines and receptors include GM-CSF, CCL2 and CSF-1R for TAMs and LY6G, CXCL1 and CXCR2 for TANs. Beyond depletion, reprogramming the myeloid cells into an anti-tumorigenic phenotype through activation of pro-inflammatory signals represents another promising option. CD40 agonists, for instance, synergize with anti-PD-1 in ICB-resistant CCA models by enhancing antigen presentation and T cell activation through increasing CD86
+ macrophages and MHC-II
+ DCs [
169]. This effect is further amplified when combined with gemcitabine plus cisplatin, suggesting its potential in first-line therapy.
Despite the limited efficacy of ICB monotherapy in CCA, novel combinatorial approaches show promise [
6,
7]. Anti-VEGF is an inhibitor for angiogenesis commonly used in combination with ICB in other types of cancers including HCC [
170]. Recent evidence shows that anti-VEGF treatment combined with anti-CTLA4 and anti-PD-L1 overcame ICB resistance in murine models of CCA [
171]. Mechanistically, the triple therapy induced a BAFF-dependent activation of B cells and IL-12 secretion. IL-12 signaling and reduced VEGF-VEGFR2 signaling promoted Treg fragility, loss of Treg suppressive function and reduced tumor burden. An open-label phase II clinical trial confirmed the superior outcome of the triple therapy compared to a previous trial with anti-CTLA4 and anti-PD-L1 alone, making the combination a promising option in clinical settings.
Moreover, multiple evidence has suggested additional CTLA-4 blockade may further augment current approved treatments. In orthotopic mouse iCCA models, adding CTLA-4 blockade to the current first-line treatment of gemcitabine, cisplatin, plus anti-PD1 resulted in greater treatment efficacy compared to gemcitabine, cisplatin, plus anti-PD1 alone [
172]. The efficacy of the combination was dependent on CXCR3
+ CD8
+ T cells, with the treatment only showing efficacy in immunocompromised mice that were adoptively transferred with CD8
+ T cells from CXCR3 WT mice but not from CXCR3
-/- mice. In IDH1 mutant iCCA, mutant IDH1 specific inhibitor, AG120 treatment reverted the R-2HG mediated suppression of T cell activity and downregulation of interferon response, resulting in PD-1 and CTLA-4 upregulation. Additional CTLA-4 blockade, but not PD-1 blockade, worked synergistically with AG120 [
127]. CTLA-4 expression was also associated with treatment resistance clinically. In a Chinese cohort of patients that underwent combination treatment of gemicitabine, oxaliplatin, lenvatinib, plus anti-PD1 (GOLP) therapy, the treatment efficacy was found to be correlated with the level of CD5L
+ macrophage infiltration. CD5L secreted by these macrophages hindered the transition of CD8
+ T cells into the anti-tumorigenic GZMK
+ CD8
+ T cells and induced CTLA-4 expression. Anti-CTLA-4 treatment restored treatment sensitivity to GOLP [
173-
175].
CONCLUSION
CCA is an aggressive malignancy with limited effective treatment. CCA is confounded by its anatomical subtypes, etiologies, and unique mutational profiles. Recently, it is evident that CCA also has diverse TME. Systemic chemotherapy, targeted therapy specific to the mutations, and immunotherapy have begun to emerge as new treatment options for CCA.
TME plays a critical role in determining response for treatments. While TAMs, neutrophils, MDSCs, and Treg cells, as well as CAFs mostly contribute to immune suppression, effector T cells and TLS are more associated with better clinical and treatment outcome for immunotherapy.
Stratification of patients based on TME features, driver mutations, and etiological factors is important for the design of personalized treatment. This allows the identification of patients who are more likely to respond to immunotherapies and combination treatments, especially targeted therapies.
In summary, to overcome the immunosuppressive barriers in CCA remains a formidable challenge that requires a more thorough understanding of inter-tumor heterogeneity and TME features. Future efforts should be spent on integrating molecular, cellular, and clinical information to more precisely stratify CCA patients for the design of innovative, effective, combination-based personalized treatments to improve outcome of CCA patients.
FOOTNOTES
-
Authors’ contributions
Cindy Xinqi Liu conducted literature review, conceptualized, drafted and reviewed the manuscript. Carmen Chak- Lui Wong drafted and reviewed the manuscript and provided overall supervision and funding.
-
Acknowledgements
This study is supported by National Natural Science Foundation of China – the Distinguished Young Scholars Fund (Project# 82425043), Research Grant Council Collaborative Research Fund (RGC CRF) (C7008-22G), and Research Grant Council Research Fellow Scheme (RGC RFS) (RFS2425-7S01). This work is supported by the Centre for Oncology and Immunology under the Health@Inno-HK initiative funded by the Innovation and Technology Commission, the Government of Hong Kong SAR, China.
All figures were created using BioRender.
-
Conflicts of Interest
The authors have no conflicts to disclose.
Figure 1.Source of inter-tumor heterogeneity in CCA. iCCA, intrahepatic cholangiocarcinoma; pCCA, perihilar cholangiocarcinoma; dCCA, distal cholangiocarcinoma; TAMs, tumor associated macrophages; Tregs, regulatory T cells; CAFs, cancer associated fibroblasts; HBV, hepatitis B virus; HCV, hepatitis C virus; CCA, cholangiocarcinoma as abbreviation. Created in Biorender.
Figure 2.Impact of stromal cell on the immune microenvironment of CCA. CCA, cholangiocarcinoma; TME, tumor microenviroment; EMT, epithelial-mesenchymal transition; NETs, neutrophil extracellular traps; iCCA, intrahepatic cholangiocarcinoma; eCCA, extrahepatic cholangiocarcinoma; MAIT, mucosal associated invariant T; Trm, resident memory T cell; MDSCs, myeloid-derived suppressor cells; ECM, extracellular matrix. Created in Biorender.
Figure 3.Impact of driver mutations and etiologies on the immune microenvironment of CCA. (A) IDH1 mutation is associated with 1) CCL2-dependent infiltration of M2-like macrophages 2) paracrine suppression of CD8+ T cells by oncometabolite R-2HG 3) TET2 inhibition induced downregulation of type I and type II interferon response. (B) FGFR2 fusion is associated with 1) reduced immune cell infiltration 2) lower expression of PD-L1 on cancer cells 3) recognition of FGFR2 fusion neoantigen by T cells. (C) KRAS mutation and overactivation of MAPK signaling are associated with 1) increased myeloid cell infiltration, reduced lymphocyte infiltration and high levels of T cell exhaustion 2) high PD-L1 expression on cancer cells due to ERK-dependent inhibition of autophagy and PD-L1 recycling. (D) In liver fluke infection-related CCA, 1) inflammation increases TMB, T cell clonal expansion and exhaustion 2) ESPs increase M2-like polarization of TAM which hinders T cell cytotoxicity via expression of PD-L1 and TIM-3. (E) HBV-infected tumors are associated with 1) greater immune cell infiltration 2) exhaustion of CD8+ T cells 3) high PD-L1 expression on cancer cells. (F) Gut barrier dysfunction results in gram-negative bacteria presence in the liver. Gram-negative bacteria-derived LPS recruits PMN-MDSCs in a TLR4-CXCL1-dependent manner. TEs, transposable elements; ISGs, interferon-stimulated genes; TAMs, tumor associated macrophages; DCs, dendritic cells; ESPs, excreted/ secreted products; TMB, tumor mutational burden; FA, fatty acid; HBV, hepatitis B virus; LPS, lipopolysaccharide; PMN-MDSCs, polymorphonuclear myeloid-derived suppressor cells; CCA, cholangiocarcinoma. Created in Biorender.
Table 1.Current available therapies for CCA
Table 1.
|
Molecular target/Pharmacological mechanism |
Level of supporting evidence |
|
Targeted therapy |
FGFR inhibitor |
Clinical trial [13] |
|
Pemigatinib |
FGFR inhibitor |
Clinical trial [13] |
|
Ivosidenib |
Mutant IDH1 inhibitor |
Clinical trial [14] |
|
Debrafenib+trametinib |
BRAFV600E inhibitor+MEK inhibitor |
Clinical trial [15] |
|
Futigatinib |
FGFR inhibitor |
Clinical trial [16] |
|
Chemotherapy |
|
|
|
Gemcitabine+cisplatin |
Chemotherapy |
Clinical trial [8] |
|
Folinic acid+fluorouracil+oxaliplatin (FOLFOX) |
Chemotherapy |
Clinical trial [12] |
|
Immunotherapy |
|
|
|
GM-CSF Blockade |
TAM recruitment inhibitor |
Pre-clinical [18] |
|
CCL2 Blockade |
TAM recruitment inhibitor |
Pre-clinical [42,126] |
|
CSF-1R Blockade |
TAM depletion |
Pre-clinical [40,47,54] |
|
CD47 Blockade |
CD47/SIRPα inhibitor |
Pre-clinical [51] |
|
LY6G Blockade |
Neutrophil depletion |
Pre-clinical [47,58,60,161] |
|
CXCL1 Blockade |
Neutrophil recruitment inhibitor |
Pre-clinical [161] |
|
CXCR2 Blockade |
Neutrophil recruitment inhibitor |
Pre-clinical [58,161] |
|
TIGIT Blockade |
Anti-TIGIT |
Pre-clinical [76] |
|
CD73 Blockade+PD-1 Blockade |
CD73 inhibitor+Anti-PD1 |
Pre-clinical [86] |
|
Combination therapy |
|
|
|
Gemcitabine+cisplatin+durvalumab |
Chemotherapy+Anti-PD1 |
Clinical trial [9] |
|
Gemcitabine+cisplatin+pembrolizumab |
Chemotherapy+Anti-PD1 |
Clinical trial [10] |
|
Gemcitabine+oxaliplatin+lenvatinib+toripalimab (GOLP) |
Chemotherapy+Tyrosine Kinase inhibitor+Anti-PD1 |
Clinical trial [174,175] |
|
CTLA-4 Blockade+PD1 Blockade+gemcitabine+cisplatin |
Anti-CTLA-4+Anti-PD1+Chemotherapy |
Pre-clinical [172] |
|
VEGF Blockade+CTLA-4 Blockade+PD1 Blockade |
Angiogenesis inhibitor+Anti-CTLA-4+Anti-PD1 |
Pre-clinical+Clinical trial [171] |
|
CD40 Agonist+PD1 Blockade+gemcitabine+cisplatin |
Antigen presentation enhancer+Anti-PD1+Chemotherapy |
Pre-clinical [169] |
|
CTLA-4 Blockade+ivosidenib |
Anti-CTLA-4+Mutant IDH1 inhibitor |
Pre-clinical [127] |
|
CTLA-4 Blockade+gemcitabine+oxaliplatin+lenvatinib+ PD-1 Blockade (GOLP) |
Anti-CTLA-4+Chemotherapy+Tyrosine Kinase inhibitor+Anti-PD1 |
Pre-clinical [173] |
Abbreviations
cancer-associated fibroblasts
conventional dendritic cells
distal cholangiocarcinoma
extrahepatic cholangiocarcinoma
excreted/secreted products
Food and Drug Administration
fibroblast growth factor receptor 2
granulocyte-like myeloid-derived suppressor cell
immune checkpoint blockade
intrahepatic cholangiocarcinoma
isocitrate dehydrogenase 1
mucosal-associated invariant T
major histocompatibility complex
non-alcoholic fatty liver disease
neutrophil extracellular traps
neutrophil-to-lymphocyte ratio
peripheral blood mononuclear cells
perihilar cholangiocarcinoma
progression-free survival
polymorphonuclear myeloid-derived suppressor cells
primary sclerosing cholangitis
tumor-associated neutrophils
tumor associated macrophages
tumor infiltrating lymphocytes
tumor immune microenvironment
tertiary lymphoid structures
tissue-resident memory T cells
REFERENCES
- 1. Valle JW, Kelley RK, Nervi B, Oh DY, Zhu AX. Biliary tract cancer. Lancet 2021;397:428-444.
- 2. Brindley PJ, Bachini M, Ilyas SI, Khan SA, Loukas A, Sirica AE, et al. Cholangiocarcinoma. Nat Rev Dis Primers 2021;7:65.
- 3. Banales JM, Cardinale V, Carpino G, Marzioni M, Andersen JB, Invernizzi P, et al. Expert consensus document: cholangiocarcinoma: current knowledge and future perspectives consensus statement from the European Network for the Study of Cholangiocarcinoma (ENS-CCA). Nat Rev Gastroenterol Hepatol 2016;13:261-280.
- 4. Sithithaworn P, Yongvanit P, Duenngai K, Kiatsopit N, Pairojkul C. Roles of liver fluke infection as risk factor for cholangiocarcinoma. J Hepatobiliary Pancreat Sci 2014;21:301-308.
- 5. Banales JM, Marin JJG, Lamarca A, Rodrigues PM, Khan SA, Roberts LR, et al. Cholangiocarcinoma 2020: the next horizon in mechanisms and management. Nat Rev Gastroenterol Hepatol 2020;17:557-588.
- 6. Kim RD, Chung V, Alese OB, El-Rayes BF, Li D, Al-Toubah TE, et al. A phase 2 multi-institutional study of nivolumab for patients with advanced refractory biliary tract cancer. JAMA Oncol 2020;6:888-894.
- 7. Piha-Paul SA, Oh DY, Ueno M, Malka D, Chung HC, Nagrial A, et al. Efficacy and safety of pembrolizumab for the treatment of advanced biliary cancer: results from the KEYNOTE-158 and KEYNOTE-028 studies. Int J Cancer 2020;147:2190-2198.
- 8. Valle J, Wasan H, Palmer DH, Cunningham D, Anthoney A, Maraveyas A, et al.; ABC-02 Trial Investigators. Cisplatin plus gemcitabine versus gemcitabine for biliary tract cancer. N Engl J Med 2010;362:1273-1281.
- 9. Oh DY, Ruth He A, Qin S, Chen LT, Okusaka T, Vogel A, et al. Durvalumab plus gemcitabine and cisplatin in advanced biliary tract cancer. NEJM Evid 2022;1:EVIDoa2200015.
- 10. Kelley RK, Ueno M, Yoo C, Finn RS, Furuse J, Ren Z, et al.; KEYNOTE-966 Investigators. 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;401:1853-1865 Erratum in: Lancet 2023;402:964. Erratum in: Lancet 2024;403:1140.
- 11. Lamarca A, Hubner RA, David Ryder W, Valle JW. Second-line chemotherapy in advanced biliary cancer: a systematic review. Ann Oncol 2014;25:2328-2338.
- 12. Lamarca A, Palmer DH, Wasan HS, Ross PJ, Ma YT, Arora A, et al.; Advanced Biliary Cancer Working Group. Second-line FOLFOX chemotherapy versus active symptom control for advanced biliary tract cancer (ABC-06): a phase 3, openlabel, randomised, controlled trial. Lancet Oncol 2021;22:690-701.
- 13. Abou-Alfa GK, Sahai V, Hollebecque A, Vaccaro G, Melisi D, Al-Rajabi R, et al. Pemigatinib for previously treated, locally advanced or metastatic cholangiocarcinoma: a multicentre, open-label, phase 2 study. Lancet Oncol 2020;21:671-684 Erratum in: Lancet Oncol 2024;25:e3.
- 14. Abou-Alfa GK, Macarulla T, Javle MM, Kelley RK, Lubner SJ, Adeva J, et al. Ivosidenib in IDH1-mutant, chemotherapy-refractory cholangiocarcinoma (ClarIDHy): a multicentre, randomised, double-blind, placebo-controlled, phase 3 study. Lancet Oncol 2020;21:e462 Erratum in: Lancet Oncol 2024;25:e61.
- 15. Subbiah V, Kreitman RJ, Wainberg ZA, Gazzah A, Lassen U, Stein A, et al. Dabrafenib plus trametinib in BRAFV600E-mutated rare cancers: the phase 2 ROAR trial. Nat Med 2023;29:1103-1112.
- 16. Goyal L, Meric-Bernstam F, Hollebecque A, Valle JW, Morizane C, Karasic TB, et al.; FOENIX-CCA2 Study Investigators. Futibatinib for FGFR2-rearranged intrahepatic cholangiocarcinoma. N Engl J Med 2023;388:228-239.
- 17. Høgdall D, Lewinska M, Andersen JB. Desmoplastic tumor microenvironment and immunotherapy in cholangiocarcinoma. Trends Cancer 2018;4:239-255.
- 18. Ruffolo LI, Jackson KM, Kuhlers PC, Dale BS, Figueroa Guilliani NM, Ullman NA, et al. GM-CSF drives myelopoiesis, recruitment and polarisation of tumour-associated macrophages in cholangiocarcinoma and systemic blockade facilitates antitumour immunity. Gut 2022;71:1386-1398.
- 19. Atanasov G, Hau HM, Dietel C, Benzing C, Krenzien F, Brandl A, et al. Prognostic significance of macrophage invasion in hilar cholangiocarcinoma. BMC Cancer 2015;15:790.
- 20. Hasita H, Komohara Y, Okabe H, Masuda T, Ohnishi K, Lei XF, et al. Significance of alternatively activated macrophages in patients with intrahepatic cholangiocarcinoma. Cancer Sci 2010;101:1913-1919.
- 21. Yang H, Yan M, Li W, Xu L. SIRPα and PD1 expression on tumor-associated macrophage predict prognosis of intrahepatic cholangiocarcinoma. J Transl Med 2022;20:140.
- 22. Mao ZY, Zhu GQ, Xiong M, Ren L, Bai L. Prognostic value of neutrophil distribution in cholangiocarcinoma. World J Gastroenterol 2015;21:4961-4968.
- 23. Buettner S, Spolverato G, Kimbrough CW, Alexandrescu S, Marques HP, Lamelas J, et al. The impact of neutrophilto-lymphocyte ratio and platelet-to-lymphocyte ratio among patients with intrahepatic cholangiocarcinoma. Surgery 2018;164:411-418.
- 24. Ji F, Kang Q, Wang L, Liu L, Ke Y, Zhu Y, et al. Prognostic significance of the neutrophil-to-lymphocyte ratio with distal cholangiocarcinoma patients. Medicine (Baltimore) 2020;99:e22827.
- 25. Ge MY, Liu ZP, Pan Y, Wang JY, Wang X, Dai HS, et al. Assessment of the prognostic value of the neutrophilto-lymphocyte ratio and platelet-to-lymphocyte ratio in perihilar cholangiocarcinoma patients following curative resection: a multicenter study of 333 patients. Front Oncol 2023;12:1104810.
- 26. Kinoshita M, Kobayashi S, Gotoh K, Kubo M, Hayashi K, Iwagami Y, et al. Heterogeneity of Treg/Th17 according to cancer progression and modification in biliary tract cancers via self-producing cytokines. Dig Dis Sci 2020;65:2937-2948.
- 27. Zhang G, Zheng G, Zhang H, Qiu L. MUC1 induces the accumulation of Foxp3+ Treg cells in the tumor microenvironment to promote the growth and metastasis of cholangiocarcinoma through the EGFR/PI3K/Akt signaling pathway. Int Immunopharmacol 2023;118:110091.
- 28. Alvisi G, Termanini A, Soldani C, Portale F, Carriero R, Pilipow K, et al. Multimodal single-cell profiling of intrahepatic cholangiocarcinoma defines hyperactivated Tregs as a potential therapeutic target. J Hepatol 2022;77:1359-1372.
- 29. Konishi D, Umeda Y, Yoshida K, Shigeyasu K, Yano S, Toji T, et al. Regulatory T cells induce a suppressive immune milieu and promote lymph node metastasis in intrahepatic cholangiocarcinoma. Br J Cancer 2022;127:757-765.
- 30. Xia T, Li K, Niu N, Shao Y, Ding D, Thomas DL, et al. Immune cell atlas of cholangiocarcinomas reveals distinct tumor microenvironments and associated prognoses. J Hematol Oncol 2022;15:37.
- 31. Chen L, Huang H, Huang Z, Chen J, Liu Y, Wu Y, et al. Prognostic values of tissue-resident CD8+T cells in human hepatocellular carcinoma and intrahepatic cholangiocarcinoma. World J Surg Oncol 2023;21:124.
- 32. Kitano Y, Okabe H, Yamashita YI, Nakagawa S, Saito Y, Umezaki N, et al. Tumour-infiltrating inflammatory and immune cells in patients with extrahepatic cholangiocarcinoma. Br J Cancer 2018;118:171-180.
- 33. Wirta EV, Szeto S, Koppatz H, Nordin A, Mäkisalo H, Arola J, et al. High immune cell infiltration predicts improved survival in cholangiocarcinoma. Front Oncol 2024;14:1333926.
- 34. Zimmer CL, Filipovic I, Cornillet M, O’Rourke CJ, Berglin L, Jansson H, et al. Mucosal-associated invariant T-cell tumor infiltration predicts long-term survival in cholangiocarcinoma. Hepatology 2022;75:1154-1168.
- 35. Affo S, Nair A, Brundu F, Ravichandra A, Bhattacharjee S, Matsuda M, et al. Promotion of cholangiocarcinoma growth by diverse cancer-associated fibroblast subpopulations. Cancer Cell 2021;39:866-882.e11. Erratum in: Cancer Cell 2021;39:883.
- 36. Zhong YJ, Luo XM, Liu F, He ZQ, Yang SQ, Ma WJ, et al. Integrative analyses of bulk and single-cell transcriptomics reveals the infiltration and crosstalk of cancer-associated fibroblasts as a novel predictor for prognosis and microenvironment remodeling in intrahepatic cholangiocarcinoma. J Transl Med 2024;22:422.
- 37. Waki Y, Morine Y, Noma T, Takasu C, Teraoku H, Yamada S, et al. Association between high expression of intratumoral fibroblast activation protein and survival in patients with intrahepatic cholangiocarcinoma. BMC Gastroenterol 2023;23:415.
- 38. Noy R, Pollard JW. Tumor-associated macrophages: from mechanisms to therapy. Immunity 2014;41:49-61. Erratum in: Immunity 2014;41:866.
- 39. Yu F, Gong L, Mo Z, Wang W, Wu M, Yang J, et al. Programmed death ligand-1, tumor infiltrating lymphocytes and HLA expression in Chinese extrahepatic cholangiocarcinoma patients: possible immunotherapy implications. Biosci Trends 2019;13:58-69.
- 40. Chen J, Tang Y, Qin D, Yu X, Tong H, Tang C, et al. ALOX5 acts as a key role in regulating the immune microenvironment in intrahepatic cholangiocarcinoma, recruiting tumor-associated macrophages through PI3K pathway. J Transl Med 2023;21:923.
- 41. Krenkel O, Tacke F. Liver macrophages in tissue homeostasis and disease. Nat Rev Immunol 2017;17:306-321.
- 42. Dwyer BJ, Jarman EJ, Gogoi-Tiwari J, Ferreira-Gonzalez S, Boulter L, Guest RV, et al. TWEAK/Fn14 signalling promotes cholangiocarcinoma niche formation and progression. J Hepatol 2021;74:860-872.
- 43. Chen Z, Li H, Li Z, Chen S, Huang X, Zheng Z, et al. SHH/GLI2-TGF-β1 feedback loop between cancer cells and tumor-associated macrophages maintains epithelial-mesenchymal transition and endoplasmic reticulum homeostasis in cholangiocarcinoma. Pharmacol Res 2023;187:106564.
- 44. Van den Bossche J, O’Neill LA, Menon D. Macrophage immunometabolism: where are we (going)? Trends Immunol 2017;38:395-406.
- 45. Liang Y, Bu Q, You W, Zhang R, Xu Z, Gan X, et al. Single-cell analysis reveals hypoxia-induced immunosuppressive microenvironment in intrahepatic cholangiocarcinoma. Biochim Biophys Acta Mol Basis Dis 2024;1870:167276.
- 46. Yamaguchi H, Hsu JM, Yang WH, Hung MC. Mechanisms regulating PD-L1 expression in cancers and associated opportunities for novel small-molecule therapeutics. Nat Rev Clin Oncol 2022;19:287-305.
- 47. Loeuillard E, Yang J, Buckarma E, Wang J, Liu Y, Conboy C, et al. Targeting tumor-associated macrophages and granulocytic myeloid-derived suppressor cells augments PD-1 blockade in cholangiocarcinoma. J Clin Invest 2020;130:5380-5396.
- 48. Mocan LP, Craciun R, Grapa C, Melincovici CS, Rusu I, Al Hajjar N, et al. PD-L1 expression on immune cells, but not on tumor cells, is a favorable prognostic factor for patients with intrahepatic cholangiocarcinoma. Cancer Immunol Immunother 2023;72:1003-1014.
- 49. Fontugne J, Augustin J, Pujals A, Compagnon P, Rousseau B, Luciani A, et al. PD-L1 expression in perihilar and intrahepatic cholangiocarcinoma. Oncotarget 2017;8:24644-24651.
- 50. Luo C, Xin H, Zhou Z, Hu Z, Sun R, Yao N, et al. Tumor-derived exosomes induce immunosuppressive macrophages to foster intrahepatic cholangiocarcinoma progression. Hepatology 2022;76:982-999.
- 51. Vaeteewoottacharn K, Kariya R, Pothipan P, Fujikawa S, Pairojkul C, Waraasawapati S, et al. Attenuation of CD47-SIRPα signal in cholangiocarcinoma potentiates tumor-associated macrophage-mediated phagocytosis and suppresses intrahepatic metastasis. Transl Oncol 2019;12:217-225.
- 52. Zhou Z, Wang P, Sun R, Li J, Hu Z, Xin H, et al. Tumor-associated neutrophils and macrophages interaction contributes to intrahepatic cholangiocarcinoma progression by activating STAT3. J Immunother Cancer 2021;9:e001946.
- 53. Yuan D, Huang S, Berger E, Liu L, Gross N, Heinzmann F, et al. Kupffer cell-derived tnf triggers cholangiocellular tumorigenesis through JNK due to chronic mitochondrial dysfunction and ROS. Cancer Cell 2017;31:771-789.e6.
- 54. Boulter L, Guest RV, Kendall TJ, Wilson DH, Wojtacha D, Robson AJ, et al. WNT signaling drives cholangiocarcinoma growth and can be pharmacologically inhibited. J Clin Invest 2015;125:1269-1285.
- 55. Coffelt SB, Wellenstein MD, de Visser KE. Neutrophils in cancer: neutral no more. Nat Rev Cancer 2016;16:431-446.
- 56. Burn GL, Foti A, Marsman G, Patel DF, Zychlinsky A. The neutrophil. Immunity 2021;54:1377-1391.
- 57. Wu Y, Ma J, Yang X, Nan F, Zhang T, Ji S, et al. Neutrophil profiling illuminates anti-tumor antigen-presenting potency. Cell 2024;187:1422-1439.e24.
- 58. Liu H, Zeng X, Ren X, Zhang Y, Huang M, Tan L, et al. Targeting tumour-intrinsic N7-methylguanosine tRNA modification inhibits MDSC recruitment and improves anti-PD-1 efficacy. Gut 2023;72:1555-1567.
- 59. Lin G, Liu Y, Li S, Mao Y, Wang J, Shuang Z, et al. Elevated neutrophil-to-lymphocyte ratio is an independent poor prognostic factor in patients with intrahepatic cholangiocarcinoma. Oncotarget 2016;7:50963-50971.
- 60. Xue R, Zhang Q, Cao Q, Kong R, Xiang X, Liu H, et al. Liver tumour immune microenvironment subtypes and neutrophil heterogeneity. Nature 2022;612:141-147.
- 61. Papayannopoulos V. Neutrophil extracellular traps in immunity and disease. Nat Rev Immunol 2018;18:134-147.
- 62. Cools-Lartigue J, Spicer J, McDonald B, Gowing S, Chow S, Giannias B, et al. Neutrophil extracellular traps sequester circulating tumor cells and promote metastasis. J Clin Invest 2013;123:3446-3458.
- 63. Yoshimoto M, Kagawa S, Kajioka H, Taniguchi A, Kuroda S, Kikuchi S, et al. Dual antiplatelet therapy inhibits neutrophil extracellular traps to reduce liver micrometastases of intrahepatic cholangiocarcinoma. Cancer Lett 2023;567:216260.
- 64. Zhang C, Wu D, Dong B, Liao G, Yu Y, Huang S, et al. The scaffold of neutrophil extracellular traps promotes CCA progression and modulates angiogenesis via ITGAV/NFκB. Cell Commun Signal 2024;22:103.
- 65. Vignali DA, Collison LW, Workman CJ. How regulatory T cells work. Nat Rev Immunol 2008;8:523-532.
- 66. Nie J, Zhang S, Guo Y, Liu C, Shi J, Wu H, et al. Mapping of the T-cell landscape of biliary tract cancer unravels anatomic subtype-specific heterogeneity. Cancer Res 2025;85:704-722.
- 67. Zhang M, Yang H, Wan L, Wang Z, Wang H, Ge C, et al. Single-cell transcriptomic architecture and intercellular crosstalk of human intrahepatic cholangiocarcinoma. J Hepatol 2020;73:1118-1130.
- 68. Zhang QW, Zhu MX, Liu WF, Rui WW, Chen Y, Ding XY, et al. Identification of clinically relevant subsets CD39+PD-1+CD8+ T cells and CD39+ regulatory T cells in intrahepatic cholangiocarcinoma using single-cell CyTOF. Transl Oncol 2024;44:101954.
- 69. Xu L, Lu Y, Deng Z, Li X, Shi Y, Zhao K, et al. Single-cell landscape of immunocytes in patients with extrahepatic cholangiocarcinoma. J Transl Med 2022;20:210.
- 70. Zuyin L, Zhao L, Qian C, Changkun Z, Delin M, Jialing H, et al. Single-cell and spatial transcriptomics delineate the microstructure and immune landscape of intrahepatic cholangiocarcinoma in the leading-edge area. Adv Sci (Weinh) 2025;1:e2412740.
- 71. Zhou G, Sprengers D, Mancham S, Erkens R, Boor PPC, van Beek AA, et al. Reduction of immunosuppressive tumor microenvironment in cholangiocarcinoma by ex vivo targeting immune checkpoint molecules. J Hepatol 2019;71:753-762.
- 72. Lubuela G, Beaufrère A, Albuquerque M, Pignollet C, Nicolle R, Lesurtel M, et al. Prognostic impact of the tumour microenvironment in intrahepatic cholangiocarcinoma: identification of a peritumoural fibro-immune interface. Virchows Arch 2024;485:901-911.
- 73. Kim HD, Jeong S, Park S, Lee YJ, Ju YS, Kim D, et al. Implication of CD69+ CD103+ tissue-resident-like CD8+ T cells as a potential immunotherapeutic target for cholangiocarcinoma. Liver Int 2021;41:764-776.
- 74. Tian L, Ma J, Ma L, Zheng B, Liu L, Song D, et al. PD-1/PDL1 expression profiles within intrahepatic cholangiocarcinoma predict clinical outcome. World J Surg Oncol 2020;18:303.
- 75. Carapeto F, Bozorgui B, Shroff RT, Chagani S, Solis Soto L, Foo WC, et al. The immunogenomic landscape of resected intrahepatic cholangiocarcinoma. Hepatology 2022;75:297-308.
- 76. Tang T, Wang W, Gan L, Bai J, Tan D, Jiang Y, et al. TIGIT expression in extrahepatic cholangiocarcinoma and its impact on CD8 + T cell exhaustion: implications for immunotherapy. Cell Death Dis 2025;16:90.
- 77. Ma L, Hernandez MO, Zhao Y, Mehta M, Tran B, Kelly M, et al. Tumor cell biodiversity drives microenvironmental reprogramming in liver cancer. Cancer Cell 2019;36:418-430.e6.
- 78. Ma L, Wang L, Khatib SA, Chang CW, Heinrich S, Dominguez DA, et al. Single-cell atlas of tumor cell evolution in response to therapy in hepatocellular carcinoma and intrahepatic cholangiocarcinoma. J Hepatol 2021;75:1397-1408.
- 79. Cheng M, Liang G, Yin Z, Lin X, Sun Q, Liu Y. Immunosuppressive role of SPP1-CD44 in the tumor microenvironment of intrahepatic cholangiocarcinoma assessed by single-cell RNA sequencing. J Cancer Res Clin Oncol 2023;149:5497-5512.
- 80. Carnevale G, Carpino G, Cardinale V, Pisciotta A, Riccio M, Bertoni L, et al. Activation of Fas/FasL pathway and the role of c-FLIP in primary culture of human cholangiocarcinoma cells. Sci Rep 2017;7:14419.
- 81. Yu W, Chen J, Tong Y, Zhu L, Deng Y, He J, et al. NSDHL promotes the degradation of sting in cholangiocarcinoma. Heliyon 2024;10:e37592.
- 82. Xu YP, Zhou YQ, Zhao YJ, Zhao Y, Wang F, Huang XY, et al. High level of CD73 predicts poor prognosis of intrahepatic cholangiocarcinoma. J Cancer 2021;12:4655-4660.
- 83. Sun BY, Yang ZF, Wang ZT, Liu G, Zhou C, Zhou J, et al. Integrative analyses identify CD73 as a prognostic biomarker and immunotherapeutic target in intrahepatic cholangiocarcinoma. World J Surg Oncol 2023;21:90.
- 84. Deaglio S, Dwyer KM, Gao W, Friedman D, Usheva A, Erat A, et al. Adenosine generation catalyzed by CD39 and CD73 expressed on regulatory T cells mediates immune suppression. J Exp Med 2007;204:1257-1265.
- 85. Stagg J, Divisekera U, Duret H, Sparwasser T, Teng MW, Darcy PK, et al. CD73-deficient mice have increased antitumor immunity and are resistant to experimental metastasis. Cancer Res 2011;71:2892-2900.
- 86. Sun BY, Zhang D, Gan W, Wu JF, Wang ZT, Sun GQ, et al. Targeting CD73 limits tumor progression and enhances antitumor activity of anti-PD-1 therapy in intrahepatic cholangiocarcinoma. J Cancer Res Clin Oncol 2024;150:348.
- 87. Xu YP, Dong ZN, Wang SW, Zheng YM, Zhang C, Zhou YQ, et al. circHMGCS1-016 reshapes immune environment by sponging miR-1236-3p to regulate CD73 and GAL-8 expression in intrahepatic cholangiocarcinoma. J Exp Clin Cancer Res 2021;40:290.
- 88. Godfrey DI, Koay HF, McCluskey J, Gherardin NA. The biology and functional importance of MAIT cells. Nat Immunol 2019;20:1110-1128.
- 89. Dusseaux M, Martin E, Serriari N, Péguillet I, Premel V, Louis D, et al. Human MAIT cells are xenobiotic-resistant, tissue-targeted, CD161hi IL-17-secreting T cells. Blood 2011;117:1250-1259.
- 90. Li YR, Zhou K, Wilson M, Kramer A, Zhu Y, Dawson N, et al. Mucosal-associated invariant T cells for cancer immunotherapy. Mol Ther 2023;31:631-646.
- 91. Okła K, Farber DL, Zou W. Tissue-resident memory T cells in tumor immunity and immunotherapy. J Exp Med 2021;218:e20201605.
- 92. Zhou PY, Zhou C, Gan W, Tang Z, Sun BY, Huang JL, et al. Single-cell and spatial architecture of primary liver cancer. Commun Biol 2023;6:1181.
- 93. Sirica AE. The role of cancer-associated myofibroblasts in intrahepatic cholangiocarcinoma. Nat Rev Gastroenterol Hepatol 2011;9:44-54.
- 94. Mederacke I, Hsu CC, Troeger JS, Huebener P, Mu X, Dapito DH, et al. Fate tracing reveals hepatic stellate cells as dominant contributors to liver fibrosis independent of its aetiology. Nat Commun 2013;4:2823.
- 95. Ping Q, Yan R, Cheng X, Wang W, Zhong Y, Hou Z, et al. Cancer-associated fibroblasts: overview, progress, challenges, and directions. Cancer Gene Ther 2021;28:984-999. Erratum in: Cancer Gene Ther 2021;28:1074.
- 96. Lin Y, Li B, Yang X, Cai Q, Liu W, Tian M, et al. Fibroblastic FAP promotes intrahepatic cholangiocarcinoma growth via MDSCs recruitment. Neoplasia 2019;21:1133-1142.
- 97. Lin Y, Cai Q, Chen Y, Shi T, Liu W, Mao L, et al. CAFs shape myeloid-derived suppressor cells to promote stemness of intrahepatic cholangiocarcinoma through 5-lipoxygenase. Hepatology 2022;75:28-42.
- 98. Mertens JC, Fingas CD, Christensen JD, Smoot RL, Bronk SF, Werneburg NW, et al. Therapeutic effects of deleting cancer-associated fibroblasts in cholangiocarcinoma. Cancer Res 2013;73:897-907.
- 99. Yang X, Lin Y, Shi Y, Li B, Liu W, Yin W, et al. FAP promotes immunosuppression by cancer-associated fibroblasts in the tumor microenvironment via STAT3-CCL2 signaling. Cancer Res 2016;76:4124-4135.
- 100. Lewinska M, Zhuravleva E, Satriano L, Martinez MB, Bhatt DK, Oliveira DVNP, et al. Fibroblast-derived lysyl oxidase increases oxidative phosphorylation and stemness in cholangiocarcinoma. Gastroenterology 2024;166:886-901 e7.
- 101. Kittirat Y, Suksawat M, Thongchot S, Padthaisong S, Phetcharaburanin J, Wangwiwatsin A, et al. Interleukin-6-derived cancer-associated fibroblasts activate STAT3 pathway contributing to gemcitabine resistance in cholangiocarcinoma. Front Pharmacol 2022;13:897368.
- 102. Thongchot S, Vidoni C, Ferraresi A, Loilome W, Khuntikeo N, Sangkhamanon S, et al. Cancer-associated fibroblast-derived IL-6 determines unfavorable prognosis in cholangiocarcinoma by affecting autophagy-associated chemoresponse. Cancers (Basel) 2021;13:2134.
- 103. Li Z, Chen Z, Li S, Qian X, Zhang L, Long G, et al. Circ_0020256 induces fibroblast activation to drive cholangiocarcinoma development via recruitment of EIF4A3 protein to stabilize KLF4 mRNA. Cell Death Discov 2023;9:161.
- 104. Laumont CM, Banville AC, Gilardi M, Hollern DP, Nelson BH. Tumour-infiltrating B cells: immunological mechanisms, clinical impact and therapeutic opportunities. Nat Rev Cancer 2022;22:414-430.
- 105. Chen Z, Yu M, Yan J, Guo L, Zhang B, Liu S, et al. PNOC expressed by B cells in cholangiocarcinoma was survival related and LAIR2 could be a T cell exhaustion biomarker in tumor microenvironment: characterization of immune microenvironment combining single-cell and bulk sequencing technology. Front Immunol 2021;12:647209.
- 106. Milardi G, Franceschini B, Camisaschi C, Puccio S, Costa G, Soldani C, et al. Immunosuppressive contribution of tumour-infiltrating B cells in human intrahepatic cholangiocarcinoma and their role in chemoimmunotherapy outcome. Gut 2025;gutjnl-2025-334861.
- 107. Wolf NK, Kissiov DU, Raulet DH. Roles of natural killer cells in immunity to cancer, and applications to immunotherapy. Nat Rev Immunol 2023;23:90-105.
- 108. Laskowski TJ, Biederstädt A, Rezvani K. Natural killer cells in antitumour adoptive cell immunotherapy. Nat Rev Cancer 2022;22:557-575.
- 109. Oliviero B, Varchetta S, Mele D, Pessino G, Maiello R, Falleni M, et al. MICA/B-targeted antibody promotes NK cell-driven tumor immunity in patients with intrahepatic cholangiocarcinoma. Oncoimmunology 2022;11:2035919.
- 110. Sun BY, Wang ZT, Chen KZ, Song Y, Wu JF, Zhang D, et al. Mobilization and activation of tumor-infiltrating dendritic cells inhibits lymph node metastasis in intrahepatic cholangiocarcinoma. Cell Death Discov 2024;10:304.
- 111. Takagi S, Miyagawa S, Ichikawa E, Soeda J, Miwa S, Miyagawa Y, et al. Dendritic cells, T-cell infiltration, and Grp94 expression in cholangiocellular carcinoma. Hum Pathol 2004;35:881-886.
- 112. Hu ZQ, Zhou ZJ, Luo CB, Xin HY, Li J, Yu SY, et al. Peritumoral plasmacytoid dendritic cells predict a poor prognosis for intrahepatic cholangiocarcinoma after curative resection. Cancer Cell Int 2020;20:582.
- 113. Kang J, Jeong JH, Hwang HS, Lee SS, Park DH, Oh DW, et al. Efficacy and safety of pembrolizumab in patients with refractory advanced biliary tract cancer: tumor proportion score as a potential biomarker for response. Cancer Res Treat 2020;52:594-603.
- 114. Klein O, Kee D, Nagrial A, Markman B, Underhill C, Michael M, et al. Evaluation of combination nivolumab and ipilimumab immunotherapy in patients with advanced biliary tract cancers: subgroup analysis of a phase 2 nonrandomized clinical trial. JAMA Oncol 2020;6:1405-1409.
- 115. Nakamura H, Arai Y, Totoki Y, Shirota T, Elzawahry A, Kato M, et al. Genomic spectra of biliary tract cancer. Nat Genet 2015;47:1003-1010.
- 116. Lin Y, Peng L, Dong L, Liu D, Ma J, Lin J, et al. Geospatial immune heterogeneity reflects the diverse tumor-immune interactions in intrahepatic cholangiocarcinoma. Cancer Discov 2022;12:2350-2371.
- 117. Lin J, Dai Y, Sang C, Song G, Xiang B, Zhang M, et al. Multimodule characterization of immune subgroups in intrahepatic cholangiocarcinoma reveals distinct therapeutic vulnerabilities. J Immunother Cancer 2022;10:e004892.
- 118. Wu MJ, Shi L, Merritt J, Zhu AX, Bardeesy N. Biology of IDH mutant cholangiocarcinoma. Hepatology 2022;75:1322-1337.
- 119. Bunse L, Pusch S, Bunse T, Sahm F, Sanghvi K, Friedrich M, et al. Suppression of antitumor T cell immunity by the oncometabolite (R)-2-hydroxyglutarate. Nat Med 2018;24:1192-1203.
- 120. Notarangelo G, Spinelli JB, Perez EM, Baker GJ, Kurmi K, Elia I, et al. Oncometabolite d-2HG alters T cell metabolism to impair CD8+ T cell function. Science 2022;377:1519-1529.
- 121. Friedrich M, Sankowski R, Bunse L, Kilian M, Green E, Ramallo Guevara C, et al. Tryptophan metabolism drives dynamic immunosuppressive myeloid states in IDH-mutant gliomas. Nat Cancer 2021;2:723-740.
- 122. Friedrich M, Hahn M, Michel J, Sankowski R, Kilian M, Kehl N, et al. Dysfunctional dendritic cells limit antigen-specific T cell response in glioma. Neuro Oncol 2023;25:263-276.
- 123. Xiang X, Liu Z, Zhang C, Li Z, Gao J, Zhang C, et al. IDH mutation subgroup status associates with intratumor heterogeneity and the tumor microenvironment in intrahepatic cholangiocarcinoma. Adv Sci (Weinh) 2021;8:e2101230.
- 124. Duygu Saatcioglu H, Valle J, Macarulla T, Javle M, Oh D, Goyal L, et al. 552 Characteristics of the tumor microenvironment in IDH1-mutated cholangiocarcinoma patients from ClarIDHy trial. J Immunotherap Cancer 2022;10(Suppl 2):A576-A577.
- 125. Makawita S, Lee S, Kong E, Kwong LN, Abouelfetouh Z, Danner De Armas A, et al. Comprehensive immunogenomic profiling of IDH1-/2-altered cholangiocarcinoma. JCO Precis Oncol 2024;8:e2300544.
- 126. Zabransky DJ, Kartalia E, Lee JW, Leatherman JM, Charmsaz S, Young SE, et al. Tumor-derived CCL2 drives tumor growth and immunosuppression in IDH1- mutant cholangiocarcinoma. Hepatology 2025;82:599-611.
- 127. Wu MJ, Shi L, Dubrot J, Merritt J, Vijay V, Wei TY, et al. Mutant IDH inhibits IFNγ-TET2 signaling to promote immunoevasion and tumor maintenance in cholangiocarcinoma. Cancer Discov 2022;12:812-835.
- 128. Wu MJ, Kondo H, Kammula AV, Shi L, Xiao Y, Dhiab S, et al. Mutant IDH1 inhibition induces dsDNA sensing to activate tumor immunity. Science 2024;385:eadl6173.
- 129. Baretti M, Shekhar S, Sahai V, Shu D, Howe K, Gunchick V, et al. Deep immune profiling of intrahepatic cholangiocarcinoma with CODEX multiplexed imaging. Hepatol Commun 2025;9:e0632.
- 130. Goyal L, Kongpetch S, Crolley VE, Bridgewater J. Targeting FGFR inhibition in cholangiocarcinoma. Cancer Treat Rev 2021;95:102170.
- 131. Chen L, Zhang Y, Yin L, Cai B, Huang P, Li X, et al. Fibroblast growth factor receptor fusions in cancer: opportunities and challenges. J Exp Clin Cancer Res 2021;40:345.
- 132. Sridharan V, Neyaz A, Chogule A, Baiev I, Reyes S, Barr Fritcher EG, et al. FGFR mRNA expression in cholangiocarcinoma and its correlation with FGFR2 fusion status and immune signatures. Clin Cancer Res 2022;28:5431-5439.
- 133. Dong L, Lu D, Chen R, Lin Y, Zhu H, Zhang Z, et al. Proteogenomic characterization identifies clinically relevant subgroups of intrahepatic cholangiocarcinoma. Cancer Cell 2022;40:70-87.e15.
- 134. White BS, Sindiri S, Hill V, Gasmi B, Nah S, Gartner JJ, et al. Specific recognition of an FGFR2 fusion by tumor infiltrating lymphocytes from a patient with metastatic cholangiocarcinoma. J Immunother Cancer 2023;11:e006303.
- 135. Punekar SR, Velcheti V, Neel BG, Wong KK. The current state of the art and future trends in RAS-targeted cancer therapies. Nat Rev Clin Oncol 2022;19:637-655.
- 136. Huang L, Guo Z, Wang F, Fu L. KRAS mutation: from undruggable to druggable in cancer. Signal Transduct Target Ther 2021;6:386.
- 137. Zhang M, Huang Y, Pan J, Sang C, Lin Y, Dong L, et al. An inflammatory checkpoint generated by IL1RN splicing offers therapeutic opportunity for KRAS-mutant intrahepatic cholangiocarcinoma. Cancer Discov 2023;13:2248-2269.
- 138. Colyn L, Alvarez-Sola G, Latasa MU, Uriarte I, Herranz JM, Arechederra M, et al. New molecular mechanisms in cholangiocarcinoma: signals triggering interleukin-6 production in tumor cells and KRAS co-opted epigenetic mediators driving metabolic reprogramming. J Exp Clin Cancer Res 2022;41:183.
- 139. Martin-Serrano MA, Kepecs B, Torres-Martin M, Bramel ER, Haber PK, Merritt E, et al. Novel microenvironment-based classification of intrahepatic cholangiocarcinoma with therapeutic implications. Gut 2023;72:736-748.
- 140. Gao Z, Chen JF, Li XG, Shi YH, Tang Z, Liu WR, et al. KRAS acting through ERK signaling stabilizes PD-L1 via inhibiting autophagy pathway in intrahepatic cholangiocarcinoma. Cancer Cell Int 2022;22:128.
- 141. Tomlinson JL, Li B, Yang J, Loeuillard E, Stumpf HE, Kuipers H, et al. Syngeneic murine models with distinct immune microenvironments represent subsets of human intrahepatic cholangiocarcinoma. J Hepatol 2024;80:892-903.
- 142. Pellat A, Vaquero J, Fouassier L. Role of ErbB/HER family of receptor tyrosine kinases in cholangiocyte biology. Hepatology 2018;67:762-773.
- 143. Pellino A, Loupakis F, Cadamuro M, Dadduzio V, Fassan M, Guido M, et al. Precision medicine in cholangiocarcinoma. Transl Gastroenterol Hepatol 2018;3:40.
- 144. Koch AL, Vellanki PJ, Drezner N, Li X, Mishra-Kalyani PS, Shen YL, et al. FDA approval summary: osimertinib for adjuvant treatment of surgically resected non-small cell lung cancer, a collaborative project orbis review. Clin Cancer Res 2021;27:6638-6643.
- 145. Xie YH, Chen YX, Fang JY. Comprehensive review of targeted therapy for colorectal cancer. Signal Transduct Target Ther 2020;5:22.
- 146. Vogel A, Kasper S, Bitzer M, Block A, Sinn M, Schulze-Bergkamen H, et al. S. PICCA study: panitumumab in combination with cisplatin/gemcitabine chemotherapy in KRAS wild-type patients with biliary cancer-a randomised biomarker-driven clinical phase II AIO study. Eur J Cancer 2018;92:11-19.
- 147. Leone F, Marino D, Cereda S, Filippi R, Belli C, Spadi R, et al. Panitumumab in combination with gemcitabine and oxaliplatin does not prolong survival in wild-type KRAS advanced biliary tract cancer: a randomized phase 2 trial (Vecti-BIL study). Cancer 2016;122:574-581.
- 148. Lee J, Park SH, Chang HM, Kim JS, Choi HJ, Lee MA, et al. Gemcitabine and oxaliplatin with or without erlotinib in advanced biliary-tract cancer: a multicentre, open-label, randomised, phase 3 study. Lancet Oncol 2012;13:181-188 Erratum in: Lancet Oncol 2012;13:e49.
- 149. Zheng S, Zhu Y, Zhao Z, Wu Z, Okanurak K, Lv Z. Liver fluke infection and cholangiocarcinoma: a review. Parasitol Res 2017;116:11-19.
- 150. Song J, Li Y, Bowlus CL, Yang G, Leung PSC, Gershwin ME. Cholangiocarcinoma in patients with primary sclerosing cholangitis (PSC): a comprehensive review. Clin Rev Allergy Immunol 2020;58:134-149.
- 151. Clements O, Eliahoo J, Kim JU, Taylor-Robinson SD, Khan SA. Risk factors for intrahepatic and extrahepatic cholangiocarcinoma: a systematic review and meta-analysis. J Hepatol 2020;72:95-103.
- 152. Izquierdo-Sanchez L, Lamarca A, La Casta A, Buettner S, Utpatel K, Klümpen HJ, et al. Cholangiocarcinoma landscape in Europe: diagnostic, prognostic and therapeutic insights from the ENSCCA Registry. J Hepatol 2022;76:1109-1121.
- 153. Jusakul A, Cutcutache I, Yong CH, Lim JQ, Huang MN, Padmanabhan N, et al. Whole-genome and epigenomic landscapes of etiologically distinct subtypes of cholangiocarcinoma. Cancer Discov 2017;7:1116-1135.
- 154. Goeppert B, Folseraas T, Roessler S, Kloor M, Volckmar AL, Endris V, et al. Genomic characterization of cholangiocarcinoma in primary sclerosing cholangitis reveals therapeutic opportunities. Hepatology 2020;72:1253-1266.
- 155. Xu L, Zhang Y, Lin Z, Deng X, Ren X, Huang M, et al. FASN-mediated fatty acid biosynthesis remodels immune environment in Clonorchis sinensis infection-related intrahepatic cholangiocarcinoma. J Hepatol 2024;81:265-277.
- 156. An J, Kim D, Oh B, Oh YJ, Song J, Park N, et al. Comprehensive characterization of viral integrations and genomic aberrations in HBV-infected intrahepatic cholangiocarcinomas. Hepatology 2022;75:997-1011.
- 157. Li Z, Gao Q, Wu Y, Ma X, Wu F, Luan S, et al. HBV infection effects prognosis and activates the immune response in intrahepatic cholangiocarcinoma. Hepatol Commun 2024;8:e0360.
- 158. Shen Y, Xu S, Ye C, Li Q, Chen R, Wu W, et al. Proteomic and single-cell landscape reveals novel pathogenic mechanisms of HBV-infected intrahepatic cholangiocarcinoma. iScience 2023;26:106003.
- 159. Lu JC, Zeng HY, Sun QM, Meng QN, Huang XY, Zhang PF, et al. Distinct PD-L1/PD1 profiles and clinical implications in intrahepatic cholangiocarcinoma patients with different risk factors. Theranostics 2019;9:4678-4687.
- 160. Pfister D, Núñez NG, Pinyol R, Govaere O, Pinter M, Szydlowska M, et al. NASH limits anti-tumour surveillance in immunotherapy-treated HCC. Nature 2021;592:450-456.
- 161. Zhang Q, Ma C, Duan Y, Heinrich B, Rosato U, Diggs LP, et al. Gut microbiome directs hepatocytes to recruit MDSCs and promote cholangiocarcinoma. Cancer Discov 2021;11:1248-1267.
- 162. Huang P, Wei G, Kirkpatrick JD, Lin Y, Tan L, Matta H, et al. Transposon-based oncogene integration in Abcb4(Mdr2)-/- mice recapitulates high susceptibility to cholangiocarcinoma in primary sclerosing cholangitis. J Hepatol 2025;82:84-96.
- 163. Schumacher TN, Thommen DS. Tertiary lymphoid structures in cancer. Science 2022;375:eabf9419.
- 164. Ding GY, Ma JQ, Yun JP, Chen X, Ling Y, Zhang S, et al. Distribution and density of tertiary lymphoid structures predict clinical outcome in intrahepatic cholangiocarcinoma. J Hepatol 2022;76:608-618.
- 165. Shang T, Jiang T, Lu T, Wang H, Cui X, Pan Y, et al. Tertiary lymphoid structures predict the prognosis and immunotherapy response of cholangiocarcinoma. Front Immunol 2023;14:1166497.
- 166. Chung SY, Yeh YC, Huang CJ, Chiang NJ, Hsu DS, Chan MH, et al. Comparative impact of tertiary lymphoid structures and tumor-infiltrating lymphocytes in cholangiocarcinoma. J Immunother Cancer 2025;13:e010173.
- 167. Doi N, Ino Y, Fuse M, Esaki M, Shimada K, Hiraoka N. Correlation of vein-rich tumor microenvironment of intrahepatic cholangiocarcinoma with tertiary lymphoid structures and patient outcome. Mod Pathol 2024;37:100401.
- 168. Zhang FP, Zhu K, Zhu TF, Liu CQ, Zhang HH, Xu LB, et al. Intra-tumoral secondary follicle-like tertiary lymphoid structures are associated with a superior prognosis of overall survival of perihilar cholangiocarcinoma. Cancers (Basel) 2022;14:6107.
- 169. Diggs LP, Ruf B, Ma C, Heinrich B, Cui L, Zhang Q, et al. CD40-mediated immune cell activation enhances response to anti-PD-1 in murine intrahepatic cholangiocarcinoma. J Hepatol 2021;74:1145-1154.
- 170. Khan KA, Kerbel RS. Improving immunotherapy outcomes with anti-angiogenic treatments and vice versa. Nat Rev Clin Oncol 2018;15:310-324.
- 171. Benmebarek MR, Oguz C, Seifert M, Ruf B, Myojin Y, Bauer KC, et al. Anti-vascular endothelial growth factor treatment potentiates immune checkpoint blockade through a BAFF-and IL-12-dependent reprogramming of the TME. Immunity 2025;58:926-945.e10.
- 172. Chen J, Amoozgar Z, Liu X, Aoki S, Liu Z, Shin SM, et al. Reprogramming the intrahepatic cholangiocarcinoma immune microenvironment by chemotherapy and CTLA-4 blockade enhances Anti-PD-1 therapy. Cancer Immunol Res 2024;12:400-412.
- 173. Lu JC, Wu LL, Sun YN, Huang XY, Gao C, Guo XJ, et al. Macro CD5L+ deteriorates CD8+T cells exhaustion and impairs combination of Gemcitabine-Oxaliplatin-Lenvatinib-anti-PD1 therapy in intrahepatic cholangiocarcinoma. Nat Commun 2024;15:621.
- 174. Jian Z, Fan J, Shi GM, Huang XY, Wu D, Yang GH, et al. Gemox chemotherapy in combination with anti-PD1 antibody toripalimab and lenvatinib as first-line treatment for advanced intrahepatic cholangiocarcinoma: a phase 2 clinical trial. J Clin Oncol 2021;39:4094.
- 175. Shi GM, Huang XY, Wu D, Sun HC, Liang F, Ji Y, 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;8:106.
