Targeted therapies and immunotherapies for unresectable cholangiocarcinoma

Abstract

Cholangiocarcinoma (CCA) is a fatal malignancy with steadily increasing incidence and poor prognosis. Since most CCA cases are diagnosed at an advanced stage, systemic therapies, including chemotherapy, radiotherapy, targeted therapy, and immunotherapy, play a crucial role in the management of unresectable CCA. The recent advances in targeted therapies and immunotherapies brought more options in the clinical management of unresectable CCA. This review depicts the advances of targeted therapies and immunotherapies for unresectable CCA, summarizes crucial clinical trials, and describes the efficacy and safety of different drugs, which may help further develop precision and individualization in the clinical treatment of unresectable CCA.

Introduction

Biliary tract cancer (BTC), including cholangiocarcinoma (CCA) and gallbladder cancer, represents one of the most malignant tumors with poor prognosis.^[1] Based on anatomical location, CCA can be divided into distal, perihilar, and intrahepatic types. Among them, intrahepatic cholangiocarcinoma (ICC), defined as CCA occurring in hepatic periphery and proximal second-degree bile ducts, is the second commonest primary liver cancer after hepatocellular carcinoma, with increasing incidence globally.^[2] The epidemiology of CCA varies greatly geographically, with a particularly high incidence in China and Southeast Asia.^[3,4] Due to its characteristics, including an insidious onset, rapid progression, and limited response to treatment, the five-year overall survival (OS) of CCA is below 10%.^[5] Surgery is the only option with a potential curative effect in CCA patients.^[6] However, most CCA tumors are non-resectable at the time of diagnosis. Therefore, systemic treatment is crucial for unresectable CCA. Chemotherapy is paramount in unresectable CCA. According to the phase III ABC-02 trial,^[7] gemcitabine plus cisplatin has been recommended as the first-line therapy for unresectable CCA in National Comprehensive Cancer Network (NCCN) guidelines. Based on the ABC-06 study, a randomized phase III trial, combination of folinic acid, fluorouracil, and oxaliplatin (FOLFOX) could serve as the standard second-line therapy for unresectable CCA.^[8] However, the cumulative toxicity and drug resistance associated with chemotherapy are considered significant challenges in cancer treatment.

In recent years, with the improved understanding of the molecular subtyping in CCA, targeted therapy has offered more accurate and personalized treatments with fewer side effects. Several valuable targets like fibroblast growth factor receptor (FGFR), isocitrate dehydrogenase-1 (IDH1), and human epidermal growth factor receptor 2 (HER2) identified in CCA have provided patients with novel options for targeted therapy. Besides, immunotherapy, treating malignancies mostly by activating human immune system and generating immune responses, has also shown promising prospective in CCA. Combinations of immunotherapy with chemotherapy have achieved huge successes and offered more benefits to patients. With rapid development, targeted therapy and immunotherapy have become significant treatment options, alongside surgery, chemotherapy, and radiotherapy, in CCA. The treatment of unresectable CCA has entered an era of precision. In this review, we aimed to summarize advances in targeted therapy and immunotherapy for unresectable CCA. We describe key clinical trials, evaluate the efficacy and safety of significant drugs, and provide a reference to support the clinical management of CCA.

Targeted Therapy

Several key molecular alterations have been identified as being associated with distinct anatomical locations of biliary tract cancer [Figure 1].^[1,9] Alterations of FGFR and IDH1 are almost exclusively found in ICC, while amplified HER2 is commonly found in extrahepatic CCA and gallbladder cancer.^[1] This distinctive association between anatomical and molecular heterogeneity contributes to the development of genetic tests and thus brings differentiation into clinical decision making at the time of BTC diagnosis. Because of the molecular heterogeneity of CCA, precise treatment is particularly required. Continuous improvement in the identification and description of the molecular profiles of CCA patients has enabled the rapid development of targeted therapy [Figures 2–4].

Figure 1.

Specific molecular alterations have been identified as being associated with distinct anatomical locations of biliary tract cancer. ARID1A: AT-rich interaction domain 1A; BAP1: BRCA1-associated protein 1; BRAF: V-raf murine sarcoma viral oncogene homolog B; EGFR: Epidermal growth factor receptor; ERBB3: Erb-b2 receptor tyrosine kinase 3; FGFR2: Fibroblast growth factor receptor 2; HER2: Human epidermal growth factor receptor 2; IDH1: Isocitrate dehydrogenase-1; PRKACA: Protein kinase cAMP-activated catalytic subunit alpha; PRKACB: Protein kinase cAMP-activated catalytic subunit beta; PTEN: Phosphatase and tensin homolog. This figure was created inhttps://BioRender.com.

Figure 2.

Recommended therapy for CCA patients with different molecular and immune subtypes. BRAF: V-raf murine sarcoma viral oncogene homolog B; CCA: Cholangiocarcinoma; dMMR: Deficient mismatch repair; FGFR2: Fibroblast growth factor receptor 2; HER2: Human epidermal growth factor receptor 2; IDH1: Isocitrate dehydrogenase-1; KRAS: Kirsten rat sarcoma viral oncogene homolog; MSI-H: Microsatellite instability-high; NTRK: Neurotrophic tropomyosin receptor kinase; RET: Rearrange during transfection; VEGF: Vascular endothelial growth factor; VEGFR: Vascular endothelial growth factor receptor.

Figure 4.

Crucial targets and targeted drugs in CCA. AKT: A-related kinase; α-KG: α-ketoglutarate; BRAF: V-raf murine sarcoma viral oncogene homolog B; CCA: Cholangiocarcinoma; CTLA-4: Cytotoxic T lymphocyte-associated antigen-4; EGF: Epidermal growth factor; EGFR: Epidermal growth factor receptor; ERK: Extracellular signal-regulated kinase; FGF: Fibroblast growth factor; FGFR: Fibroblast growth factor receptor; HER2: Human epidermal growth factor receptor 2; IDH1: Isocitrate dehydrogenase-1; MEK: Mitogen-activated protein kinase; mTOR: Mechanistic target of rapamycin; PD-1: Programmed cell death 1 protein; PD-L1: Programmed death-ligand 1; PI3K: Phosphoinositide 3-kinase; RAF: Rapidly accelerated fibrosarcoma; RAS: Rat Sarcoma virus oncogene; RET: Rearrange during transfection; TRK: Tropomyosin receptor kinase; VEGF: Vascular endothelial growth factor; VEGFR: Vascular endothelial growth factor receptor. This figure was created inhttps://BioRender.com.

Figure 3.

Development of targeted therapies, immunotherapy and their combinations in cholangiocarcinoma. BRAF: V-raf murine sarcoma viral oncogene homolog B; dMMR: Deficient mismatch repair; FGFR2: Fibroblast growth factor receptor 2; GEMOX: Gemcitabine plus oxaliplatin; HER2: Human epidermal growth factor receptor 2; IDH1: Isocitrate dehydrogenase-1; KRAS: Kirsten rat sarcoma viral oncogene homolog; MSI-H: Microsatellite instability-high; NTRK: Neurotrophic tropomyosin receptor kinase; RET: Rearrange during transfection; VEGF: Vascular endothelial growth factor; VEGFR: Vascular endothelial growth factor receptor.

FGFR2

The FGFR family, belonging to the receptor tyrosine kinase (RTK) family, is composed of FGFR1–4. After binding to its ligand FGF, downstream pathways, including RAS/RAF/MAPK, PI3K/AKT, JAK/STAT, and PLCγ, are activated, contributing to multiple cell processes.^[10] The FGFR pathway plays crucial roles in metabolism, endocrine function, tissue repair, and tumor development. Among FGFR family members, FGFR2 aberrations, including fusion, amplification, and mutation, occur in multiple solid tumors, and FGFR2 is considered a clinical therapeutic target.^[11]FGFR2 fusions and translocations are detected in approximately 10–15% of ICC cases, while fusions are almost found in ICC exclusively compared to distal and perihilar CCAs.^[12,13] Ras-extracellular signal-regulated kinase (RAS-ERK) has been considered necessary for tumorigenesis of ICC caused by FGFR2 fusions. Through the phosphorylation of fibroblast growth factor receptor substrate 2 (FRS2) located on membrane by FGFR2 fusions, src homology 2 domain-containing protein tyrosine phosphatase 2 (SHP2) is then phosphorylated, thereby promotes the recruitment of growth factor receptor bound 2 (GRB2) to load guanosine tri-phosphate (GTP) onto RAS and then triggers activation of downstream kinase cascade.^[14]FGFR2 was one of the most crucial therapeutic targets in CCA, which promoted the rapid development of FGFR inhibitors.

Pemigatinib

Pemigatinib is a potent and selective inhibitor of FGFR1-3.^[15] It is the first approved targeted drug for CCA in several regions, which obtained approval from the US Food and Drug Administration (FDA) for previously treated, unresectable CCA with FGFR2 fusion.^[16] Several phase I studies have previously demonstrated the efficacy and safety of pemigatinib in solid tumors, including CCA with FGF/FGFR alterations.^[17,18]

The FIGHT-202 study was a phase II trial of pemigatinib in patients with CCA who progressed after receiving at least one previous systemic therapy.^[16] Patients with FGFR2 fusions or rearrangements (n = 107, 98% ICC), other FGF/FGFR alterations (n = 20, 65% ICC), or no FGF/FGFR alterations (n = 18, 61% ICC) were enrolled. The objective response rates (ORR) and disease control rates (DCR) in patients with FGFR2 fusions or rearrangements were 35.5% and 82.2%, respectively, while the median progression-free survival (mPFS) and overall survival (mOS) were 6.9 months and 21.1 months, respectively. In addition, a post-hoc analysis demonstrated that second-line treatment with pemigatinib may be correlated with extended PFS.^[19] Moreover, the phase III Fight-302 trial assessing pemigatinib as the first-line therapy for unresectable CCA with FGFR2 rearrangements was underway.^[20] Since none of the patients enrolled in FIGHT-202 were from China, a bridging study was then performed in China, and found that pemigatinib also had a favorable therapeutic efficacy in Chinese CCA patients with FGFR2 fusions or rearrangements (n = 31, 96.8% ICC).^[21] Among 30 patients available for efficacy evaluation, the DCR was surprisingly 100% (partial response [PR], n = 15, ORR 50%; stable disease [SD], n = 15, 50%). To sum up, pemigatinib is a safe and effective therapeutic option for CCA patients, especially those with FGFR2 fusions.

Infigratinib

Infigratinib (BGJ398) is a potent, orally bioavailable and ATP-competitive inhibitor of pan-FGFR^[22,23] that has shown antitumor activity and an acceptable safety profile in several advanced solid tumors with FGFR alterations.^[24] Infigratinib was approved by the FDA in 2021 for the treatment of CCA.^[25] According to a study by Javel et al^[26], infigratinib had promising clinical activity and safety for pretreated unresectable CCA patients with FGFR2 fusions or rearrangements. The ORR in the 108 CCA patients treated with infigratinib was 23.1%. This phase II study supported the application of infigratinib in treatment of CCA. Therefore, the phase III PROOF trial comparing infigratinib with gemcitabine/cisplatin as first-line therapy for unresectable CCA patients with FGFR fusions/translocations was previously planned to be conducted.^[27] However, it is worth noting that the new drug application of infigratinib has been withdrawn in the United States for business reasons, and the approval of infigratinib has also been withdrawn by FDA in 2024. The PROOF trial also set to be closed.

Futibatinib

Futibatinib, also known as TAS-120, is a novel, third-generation, irreversible, and highly selective FGFR1–4 inhibitor. Due to its unique characteristic of irreversible binding, the risk of resistance to futibatinib is low. Moreover, Futibatinib had demonstrated more potent activity in cases with FGFR mutations that are resistant to reversible ATP-competitive inhibitors, compared to pemigatinib and infigratinib.^[28,29] A first-in-human phase I trial was started in 2020, and partial response (PR) was detected in three ICC patients with FGFR2 fusion.^[30] Besides, in a previous phase I trial, the ORR of futibatinib was 13.7% in 197 patients with advanced solid tumors.^[31] It was remarkable that FGFR2 fusion/rearrangement-positive ICC cases achieved the greatest benefit (ORR, 25.3%). These trials provided a basis for subsequent application of futibatinib in ICC. In an open-label phase II study, previously treated ICC patients with FGFR2 rearrangement achieved clinical benefits from futibatinib treatment.^[32] Of the 103 enrolled patients, 42% achieved a response, including one who had a complete response (CR), and the mPFS and mOS were 9.0 months and 21.7 months, respectively. Overall, futibatinib had a favorable prospect in clinical treatment of ICC and has been approved for this indication by FDA.^[33]

Erdafitinib

Erdafitinib is the first and only FDA-approved FGFR targeted therapy for metastatic urothelial carcinoma so far.^[34] As a potent tyrosine kinase inhibitor of FGFR1–4,^[35] its clinical activity in advanced solid tumors, including CCA, has been assessed in several clinical trials.^{[34,36,37,38,39]} In a phase I trial, erdafitinib demonstrated promising activity in CCA patients (ORR = 27.3%), second only to urothelial carcinoma (ORR = 46.2%).^[37] According to the expansion cohort results of the RAGNAR trial, the ORR and DCR of erdafitinib for 35 CCA cases with FGFR alterations were 60% and 100%, respectively, which demonstrated a favorable antitumor activity in CCA.^[38] In Asian CCA cases with FGFR alterations, the LUC2001 study reported that erdafitinib achieved an ORR of 41% and a DCR of 82%.^[40] Taken together, erdafitinib has shown preliminary efficacy and manageable safety in CCA, although further validation is required specifically in the ICC subgroup.

Derazantinib

Derazantinib (ARQ 087) is an ATP-competitive multi-kinase inhibitor that exhibits potent inhibitory activity against FGFR1-3.^[41] It has shown preliminary therapeutic activity in several solid tumors, including ICC.^[42] A phase I/II trial assessed the efficacy and safety of derazantinib in patients with unresectable ICC harboring FGFR2 fusions who progressed after chemotherapy.^[43] The ORR was 20.7% and the DCR was 80.8%. A pivotal trial of derazantinib in inoperable or advanced ICC cases with FGFR2 mutations, fusions, or amplifications is also ongoing (FIDES-01).^[44] Further studies are needed to confirm the role of derazantinib in CCA.

IDH1

IDH1, a member of the IDH family, which also includes IDH2 and IDH3, is involved in cellular aerobic respiration. IDH1 encodes NADP⁺-dependent isocitrate dehydrogenase, which plays a crucial role in the tricarboxylic acid cycle. It is located in the cytoplasm and catalyzes the oxidative decarboxylation of isocitrate to produce α-ketoglutarate (α-KG). However, mutant IDH1 loses its original function while gaining a new activity to catalyze α-KG into 2-hydroxyglutarate (2-HG), an oncometabolite involved in tumor metabolism and progression.^[45,46,47] High level of 2-HG suppresses dioxygenases including histone and DNA demethylases dependent on α-KG, which play an important role in regulation of cellular epigenetic status. IDH1 mutations have been detected in various types of tumors, including approximately 10–20% of ICC patients,^{[48,49,50,51]} making it a potential therapeutic target of cancer.^[52]

Ivosidenib

Ivosidenib (AG-120) is an oral, potent, and highly selective inhibitor of mutant IDH1, and was firstly approved by the FDA for the treatment of acute myeloid leukemia with IDH1 mutations.^[53,54] In a phase I study, ivosidenib showed favorable efficacy and safety in CCA patients with IDH1 mutations.^[55] Of the 73 CCA patients (89% ICC), 4 achieved PR and 41 had SD. The mPFS and mOS were 3.8 and 13.8 months, respectively. The ClarIDHy trial was a multicenter, randomized, double-blind, placebo-controlled, phase III study, which aimed to assess the efficacy and safety of ivosidenib in patients with unresectable or metastatic CCA cases harboring IDH1-mutation who have progressed following chemotherapy.^[56] Patients (n = 187, 90% ICC) were randomly divided (2:1 ratio) into the ivosidenib (n = 126) and placebo (n = 61) groups. Results showed that patients administered ivosidenib achieved significantly enhanced survival benefits compared to the placebo group (mPFS, 2.7 months vs. 1.4 months, one side P <0.0001; mOS, 10.3 months vs. 5.1 months; one side P <0.001).^[57] Patients administered ivosidenib generally had good tolerability. According to the ClarIDHy trial, ivosidenib was approved by the FDA and European Medicines Agency (EMA) for unresectable CCA with IDH1 mutations.^[58] Furthermore, an Italian real-world study involving 11 patients with locally advanced or metastatic IDH1-mutated CCA, who were treated with ivosidenib in clinic, further validated the efficacy of ivosidenib and the mOS and mPFS were 15 months and 4.4 months, respectively.^[59] The DCR was 63%, with 2 patients achieving a PR (18%). Overall, ivosidenib has encouraging efficacy in CCA patients and has been recommended as second-line regimen for advanced BTC with IDH1 mutations by the Chinese Society of Clinical Oncology (CSCO) and NCCN guidelines.

BAY1436032

BAY1436032 is a pan-mutant IDH1 inhibitor with efficacy in several preclinical models.^[60,61] In a phase I study, 42% of the 12 ICC patients with IDH1 mutations treated with BAY1436032 had SD (no OR), and the PFS rate at 3 months was only 10%.^[62] Therefore, the role of BAY1436032 in ICC needs further exploration.

Overall, ivosidenib is only suitable option for IDH1-mutated CCA cases so far. Other IDH1 inhibitors are still in clinical trials and the therapeutic efficacy remains to be further observed.

V-raf murine sarcoma viral oncogene homolog B (BRAF)

BRAF is a serine-threonine kinase and MAPK signaling cascade activator regulated by RAS.^[63] The BRAF V600E mutation, one type of BRAF variations which increases kinase activity in a RAS-independent manner, has been widely tested in tumors and is considered a significant predictive, diagnostic, and prognostic biomarker. In BTC, the BRAF V600E mutation is rare, with only about 3% found in ICC cases.^[64] But, the BRAF V600E mutation is associated with worse clinical outcomes in ICC.^[65] Therefore, the BRAF V600E mutation is a valuable therapeutic target in CCA, especially ICC.

Dabrafenib

Compared with monotherapy, combination of dabrafenib (a BRAF inhibitor) and trametinib (a mitogen-activated protein kinase kinase [MEK] inhibitor) showed favorable efficacy and acceptable safety in several BRAF V600E-mutation tumors. The ROAR study was a single-arm, multicenter basket phase II trial evaluating the efficacy and safety of dabrafenib plus trametinib for previously treated unresectable BTC patients with the BRAF V600E mutation (n = 43, 91% ICC).^[66] The investigator-evaluated ORR was 51%, while 47% by independent assessment. Overall, dabrafenib plus trametinib has promising efficacy for CCA patients and has been recommended as second-line regimen for advanced BTC patients with BRAF V600E mutation in the NCCN and CSCO guidelines.

Neurotrophic tropomyosin receptor kinase (NTRK)

NTRK is the gene encoding for tropomyosin receptor kinase (TRK), which is involved in a series of physiological functions of the human body such as movement, memory, and proprioception. NTRK fusion is the commonest activator of oncogenic TRK.^[67] Once binding with neurotrophin, TRK fusion proteins can active the same downstream pathways as activated by full-length TRK proteins and then promote the tumor development. However, it has been rarely detected in tumors, including BTC. Nevertheless, basket trials of the NTRK fusion have achieved particular success, which brings hope to patients.

Larotrectinib and entrectinib

Both larotrectinib and entrectinib are TRK inhibitors. To date, no clinical trial of NTRK fusion BTC cases was conducted. However, the efficacy and safety of larotrectinib and entrectinib was favorable in several basket trials of solid tumors. In a study by Drilon et al^[68], 75% of NTRK fusion patients treated with larotrectinib, including one CCA case, achieved a therapeutic response, including 71% with ongoing response at one year. One CCA patient had a PR with a PFS of 12 months based on an integrated analysis of clinical trials of entrectinib.^[69] Overall, larotrectinib and entrectinib are potential options for CCA patients with NTRK fusion.

Rearrange during transfection (RET)

RET is a proto-oncogene encoding a transmembrane receptor tyrosine kinase. After binding with ligand, the RET-bound complex is incorporated into membrane and actives distinct oncogenic signal pathways through various signaling proteins binding to RET.^[70]RET fusions accelerate oncogenic development and are detected mostly in lung and thyroid cancers, but rarely in other solid tumors. Similar to the NTRK fusion, the clinical trials of the RET fusion have been generally conducted as basket trials in solid tumors.

Selpercatinib and pralsetinib

Selpercatinib and pralsetinib are highly selective RET kinase inhibitors. In the LIBRETTO-001 trial, the ORR of advanced solid tumor patients with RET fusions administered selpercatinib was 43.9%, and one CCA patient had a PR as examined by an independent review committee assessment.^[71] The ARROW trial aimed to assess the efficacy and safety of pralsetinib in advanced solid tumor patients with RET fusions.^[72] The ORR was 57%, and PRs were found in two of the three CCA patients examined. Overall, in unresectable CCA patients with RET fusions, selpercatinib and pralsetinib are considered to be beneficial.

Kirsten rat sarcoma viral oncogene homolog glycine 12 to cysteine mutation (KRAS G12C)

Known as one of the most important oncogenes, kirsten rat sarcoma viral oncogene homolog (KRAS) has attracted increasing attentions. Once its upstream receptors, such as EGFR, are activated, the guanosine diphosphate (GDP) bound to KRAS is replaced by GTP, KRAS is then transferred from a resting state to an activated state. This activates various downstream signaling pathways involved in proliferation, metabolism, immune escape, drug resistance, and as so on.^[73] For CCA, KRAS was the second most frequently mutant gene, especially in ICC.^[74] For a long term, KRAS was recognized as an undruggable target. Recently, novel inhibitor of KRAS G12C proved new prospects of targeting KRAS.

Adagrasib

In the phase II cohort of the KRYSTAL-1 study, the efficacy and safety of adagrasib, a KRAS G12C inhibitor, were evaluated in patients with KRAS G12C-mutated advanced solid tumors.^[75] Among 8 patients with CCA, ORR was 50%, and mOS and mPFS reached 15.1 months and 11.3 months, respectively. Overall, adagrasib was a favorable choice for CCA patients with KRAS G12C mutation.

Epidermal growth factor receptor (EGFR)/HER2

EGFR, also termed HER1 or ERBB1, is a tyrosine kinase of 170 kDa.^[76] As a growth factor receptor, EGFR is involved in cell differentiation and proliferation as well as tumor development and drug resistance upon activation by binding to its ligands and inducing downstream signaling cascades.^[77] EGFR is overexpressed in various solid tumors, which promotes tumor development. It is also activated and overexpressed in CCA cells.^[78,79] Besides, EGFR overexpression was found in 27% of 236 CCA cases and showed a correlation with tumor progression.^[80] Therefore, targeting EGFR may present a favorable approach for the treatment of CCA.

HER2, also known as ERBB2, is a transmembrane tyrosine kinase belonging to the EGFR family. HER2 possesses a strong ligand binding ability since it exists in an open manner, making it the better choice as the dimerization partner compared with other family members and generating more potent oncogenic signals.^[81] It is overexpressed in various tumors and considered a prognostic and predictive biomarker in gastric and breast cancers. In patients with BTC, overexpression or amplification of HER2 was detected in up to 20% of cases, varying according to location.^[82,83,84] Therefore, HER2 may also be a promising therapeutic target in CCA.

Erlotinib

Erlotinib, a small-molecule tyrosine kinase inhibitor of EGFR, has exhibited antitumor efficacy and synergistic effect with gemcitabine in non-small-cell lung and pancreatic cancers, respectively.^[85,86] Recently, combination of erlotinib and gemcitabine showed great preclinical efficacy in ICC cells.^[87] Besides, a phase I trial evaluated erlotinib plus gemcitabine in pancreatic cancer and other solid tumors, including one CCA patient.^[88] While, a PR was found in only one patient with pancreatic cancer and one patient with CCA. Several phase II trials have assessed the efficacy of the combination of erlotinib with vascular endothelial growth factor (VEGF) inhibitors in BTC. Of 53 unresectable BTC cases (81% CCA) treated with erlotinib plus bevacizumab, 6 had a PR and 25 had SD, and the mOS was 9.9 months.^[89] Erlotinib was also combined with sorafenib for the treatment of advanced BTC patients.^[90] However, there were only two unconfirmed responses, and mPFS and mOS were only 2 and 6 months, respectively. Currently, only one phase III trial of erlotinib for unresectable BTC patients has been reported.^[91] Patients were randomized to the gemcitabine plus oxaliplatin (GEMOX) (n = 133, 63% CCA) and GEMOX plus erlotinib (n = 135, 71% CCA) groups. Although there were no significant differences in mPFS (5.8 months vs. 4.2 months, P = 0.087) and mOS (9.5 months vs. 9.5 months, P = 0.611) between the GEMOX plus erlotinib and GEMOX groups, more patients treated with GEMOX plus erlotinib achieved an objective response (30% vs. 16%, P = 0.005). Specially, subgroup analysis demonstrated that CCA patients achieved significant benefits from GEMOX plus erlotinib (mPFS: 5.9 months vs. 3.0 months, P = 0.049). Taken together, erlotinib in combination may be a potential therapeutic option in CCA, but further investigation is needed in the patients with ICC.

Cetuximab

Cetuximab, a chimeric monoclonal antibody targeting EGFR, seems to be more efficient in combination rather than monotherapy, like erlotinib. In a previous study, GEMOX plus cetuximab was shown to be a well-tolerated treatment for advanced CCA patients.^[92] In a phase II study performed by Gruenberge et al^[93], GEMOX plus cetuximab also had encouraging antitumor activity as first-line treatment for unresectable BTC patients (n = 30, 90% ICC), with an ORR of 63% and three patients who achieved a CR. However, in BINGO, a randomized phase II trial, although showed a good tolerability, unresectable BTC patients (n = 150, 82% CCA) had no significant survival benefit from GEMOX plus cetuximab compared with GEMOX (mPFS, 6.1 months vs. 5.5 months; mOS, 11.0 months vs. 12.4 months),^[94] which is consistent with Chen et al^[95], although a trend of improved PFS was found (ORR, 27% vs. 15%, P = 0.12; PFS, 6.7 months vs. 4.1 months, P = 0.05), the OS between two groups is not significant difference. Nevertheless, a post-hoc analysis found that GEMOX plus cetuximab significantly improved DCR (68% vs. 41%, P = 0.044) and mPFS (7.3 months vs. 4.9 months, P = 0.026), and marginally prolonged mOS (14.1 months vs. 9.6 months, P = 0.056) compared with GEMOX in ROS1, ALK and c-MET (RAM) low-expressed unresectable ICC cases.^[96] Besides, cetuximab also showed clinical activity in combination with gemcitabine or gemcitabine plus capecitabine.^[97,98] To sum up, although currently there is no evidence to support the addition of cetuximab to the first-line therapy for advanced CCA cases, specific patients identified by molecular subtyping may achieve certain benefits.

Panitumumab

Panitumumab is a recombinant, fully humanized, monoclonal antibody targeting EGFR. Combination of panitumumab with GEMOX as first-line therapy has previously shown efficacy in unresectable KRAS wild-type BTC cases (n = 46, 82.6% CCA).^[99] Nevertheless, according to the Vecti-BIL study, a phase II randomized trial, GEMOX plus panitumumab has marginal advantages compared with GEMOX in advanced KRAS wild-type BTC cases.^[100] It is worth mentioning that ICC patients administered GEMOX plus panitumumab compared with GEMOX may have improved survival benefits based on subgroup analysis (OS, 15.1 months vs. 11.8 months, P = 0.13). A study by Amin et al^[101] also demonstrated that adding panitumumab to chemotherapy provides no benefit in advanced CCA patients. But, the combination of gemcitabine, irinotecan, and panitumumab exhibited encouraging efficacy in advanced BTC patients (n = 35, 83% CCA).^[102] Overall, more exploration is required for the application of panitumumab in combination with chemotherapy in unresectable CCA cases.

Afatinib

Afatinib is an irreversible inhibitor targeting ERBB family proteins.^[103] According to a phase I trial, eight advanced solid tumor cases, including one CCA, treated with a combination of afatinib with paclitaxel achieved a PR.^[104] However, in a phase I trial of BTC, gemcitabine/cisplatin plus afatinib failed to confer survival benefits in advanced BTC cases.^[105] More clinical evidence is required to demonstrate the utility of afatinib in CCA.

Other EGFR pathway inhibitors

Upon ligand binding with EGFR, signaling cascades are initialized and several downstream effectors are activated, including MAPK, RAS/RAF/MEK/ERK, and PI3K/AKT/mTOR, which are also considered therapeutic targets in cancer.^[106] Sirolimus, an inhibitor of mTOR, has been investigated in advanced CCA cases.^[107] However, no objective response was observed. A study of everolimus, another mTOR inhibitor, also failed to meet the primary objective.^[108] Besides, the potent MEK1/2 inhibitor binimetinib showed promising activity in advance BTC patients.^[109] Combination of gemcitabine/cisplatin and copanlisib, a PI3K inhibitor, as first-line therapy may be more effective and prolong survival in phosphatase and tensin homolog (PTEN) low-expression advanced BTC cases.^[110] In general, there is no overt evidence supporting clinical applications of these EGFR pathway inhibitors in CCA, which needs to be further examined in larger clinical trials.

Trastuzumab

Trastuzumab is an oral, humanized, monoclonal anti-HER2 antibody. Combination of chemotherapy and trastuzumab has served as standard treatment for HER2-positive gastric and breast cancers.^[111,112] The KCSG-HB19-14 study aimed to assess the clinical activity of FOLFOX plus trastuzumab as second- or third-line therapy in HER2-positive (immunohistochemistry [IHC] 3+ or IHC 2+ and in-situ hybridization [ISH]+) advanced BTC cases refractory to gemcitabine and cisplatin.^[113] Among 34 patients (47% CCA), ORR was 29.4% and DCR was 79.4%, while mPFS and mOS were 5.1 and 10.7 months, respectively. Combination of trastuzumab and tucatinib was also effective and well-tolerated in previously treated metastatic BTC cases with HER2 overexpression or amplification (n = 30, 50% CCA; ORR, 46.7%; DCR, 76.7%).^[114] Besides, a dual anti-HER2 regimen (pertuzumab plus trastuzumab) achieved an ORR of 23.3% in previously treated HER2-positive advanced BTC cases (n = 39, 36% CCA).^[115] Summarily, trastuzumab has exhibited promising efficacy in BTC, which supports the development of phase III trials.

Zanidatamab

Zanidatamab is a humanized, bispecific monoclonal antibody targeting two distinct HER2 epitopes. It is tolerated and active in several advanced HER2-positive solid tumors, including BTC.^[116] The HERIZON-BTC-01 study enrolled 87 unresectable HER2-amplified BTC cases (48.3% ICC), who progressed after gemcitabine treatment.^[117] The patients received zanidatamab treatment, and the ORR was 41.3% in 80 patients in cohort 1 (IHC 2+ or 3+). The encouraging clinical effect supports zanidatamab as a potential option for CCA patients in the future.

DS-8201

Trastuzumab deruxtecan (DS-8201) is an antibody-drug conjugate comprising an anti-HER2 antibody, a linker, and a potent topoisomerase I inhibitor. In the HERB study, ORR was 36.4% and DCR was 81.8% in 22 HER2-positive advanced BTC cases (41% CCA) treated with DS-8201.^[118] Besides, in the DESTINY-PanTumor02 phase II trial, the ORR in 41 HER2-overexpressing (IHC 2+ or 3+) unresectable BTC cases treated with DS-8201 was 22%, while that of IHC 3+ patients was 56.3%.^[119] Therefore, DS-8201 may also be a potential choice for HER2-positive CCA patients.

VEGF/vascular endothelial growth factor receptor (VEGFR)

The VEGF family is composed of VEGF-A, B, C, D, and E and placental growth factor (PlGF).^[120] It is one of the most important inducers of angiogenesis, which plays a crucial role in homeostasis. Nevertheless, altered angiogenesis can accelerate and promote the development of diseases, including tumors. New blood vessels are formatted by proliferated endothelial cells stimulated by VEGF secreted from tumor cells, which may be structurally abnormal and promote tumorigenesis.^[121] VEGF was found to be overexpressed in multiple tumors, and inhibiting VEGF is considered an efficient antitumor strategy. Besides, VEGF is also overexpressed in BTC and regarded as a poor prognostic factor.^[122] VEGFR is the cognate receptor of VEGF, including VEGFR1–3. Targeting VEGF/VEGFR is an important therapeutic option for CCA.

Bevacizumab

Bevacizumab is a recombinant, humanized, monoclonal antibody targeting VEGF and suppressing angiogenesis. It is the most widely used anti-angiogenesis therapy for tumors. Besides targeting angiogenesis, bevacizumab may also enhance chemotherapy.^[123] Therefore, it is generally applied in combination with chemotherapy. In a phase II study, GEMOX plus bevacizumab showed anti-tumor activity and acceptable safety in advanced BTC cases (n = 35, 71.4% CCA).^[124] However, other phase II studies have reported that combination of bevacizumab with gemcitabine, capecitabine and combination of bevacizumab with gemcitabine, capecitabine, and irinotecan confer no survival advantages to advanced BTC cases compared with historical controls.^[125,126] Overall, although the addition of bevacizumab to chemotherapy is generally well-tolerated, its potential benefits in CCA remains unclear.

Ramucirumab

Ramucirumab is a humanized, direct inhibitor of VEGFR-2. Although it is well-tolerated in previously treated unresectable BTC cases, it does not prolong PFS compared with cases administered other chemotherapies historically (n = 60, 78.3% CCA).^[127] Besides, there was also no significant difference in PFS between advanced BTC patients treated with GEMOX plus ramucirumab compared with GEMOX plus placebo as first-line therapy (n = 207, 69.1% CCA).^[128] Therefore, further studies are needed to determine the efficacy of ramucirumab.

Apatinib

Apatinib is an oral VEGFR2 inhibitor. In advanced BTC patients refractory to chemotherapy, apatinib showed clinical efficacy and safety in two phase II trials (59.1% and 100% CCA, respectively).^[129,130] The role of apatinib in the treatment of ICC deserves further examination.

Sorafenib

Sorafenib is an oral antiangiogenic agent targeting multiple proteins, including VEGF. The activity of sorafenib as a single agent is limited in BTC patients.^[131,132] The DCR of the combination of sorafenib with best supportive care in unresectable ICC patients was also low, i.e., only 15.9%.^[133] Besides, combination of sorafenib with chemotherapy did not improve efficacy in advanced BTC patients compared to chemotherapy.^[134,135] Sorafenib may not be a good therapeutic option for unresectable CCA cases.

Sunitinib

Sunitinib is an oral inhibitor of multiple receptor tyrosine kinases, including but not limited to VEGF. Sunitinib monotherapy only showed marginal efficacy in BTC patients (n = 56, 73.2% CCA) in a phase II trial.^[136] Besides, combination of sunitinib with capecitabine conferred a PR in several patients with solid tumors, including one CCA case.^[137] However, the role of sunitinib in the treatment of CCA still needs to be examined.

Surufatinib

Surufatinib is an inhibitor of VEGFR1–3, FGFR1, and colony stimulating factor 1 receptor (CSF1R). In a study by Xu et al^[138], surufatinib as second-line monotherapy provided moderate clinical efficacy in unresectable BTC cases (n = 39, 87.2% CCA). Intriguingly, according to a subgroup analysis, ICC patients had enhanced therapeutic benefits from surufatinib than other patients with extrahepatic cholangiocarcinoma or gallbladder cancer. Therefore, surufatinib may be a potential option as anti-angiogenesis for CCA and has been included in CSCO guidelines.

Regorafenib

Regorafenib is a multi-kinase inhibitor targeting VEGF, FGFR, and others. In two single-arm phase II studies, regorafenib showed initial efficacy in unresectable BTC patients refractory to chemotherapy (88% and 84.6% CCA).^[139,140] REACHIN was a randomized phase II trial aiming to compare regorafenib and placebo in unresectable BTC patients who had progressed after first-line therapy (n = 66, 86.4% CCA).^[141] Encouragingly, regorafenib treatment prolonged mPFS (3 months vs. 1.5 months, P = 0.004) and improved the SD rate (74% vs. 34%, P = 0.002). Overall, regorafenib may as a viable option offering survival benefits in unresectable CCA patients refractory to chemotherapy, and it has also been incorporated into the CSCO guidelines.

Lenvatinib

Lenvatinib is an inhibitor of VEGFR1–3, FGFR1–4, and others. As a second-line monotherapy, lenvatinib had approximately an ORR of 12% in previously treated unresectable BTC patients.^[142,143] Lenvatinib seems to have better efficacy in combination with immunotherapy.

Cediranib

Cediranib is an orally bioavailable inhibitor of pan-VEGFR. The ABC-03 study was a randomized and placebo-controlled phase II trial that aimed to evaluate the effect of cediranib combined with gemcitabine and cisplatin as first-line treatment on PFS in advanced BTC patients (n = 124, 62.1% CCA).^[144] However, compared with placebo, addition of cediranib failed to prolong PFS (8.0 months vs. 7.4 months, P = 0.72).

Vandetatinib

Vandetatinib is an oral inhibitor targeting VEGFR2, EGFR, and RET. Combination of vandetanib with gemcitabine and capecitabine showed promising efficacy in advanced BTC patients.^[145] However, in a randomized study, vandetatinib failed to improve PFS in advanced BTC cases.^[146] Therefore, the role of vandetatinib in unresectable CCA remains unclear.

Overall, the efficacy of anti-VEGF/VEGFR combined with chemotherapy in unresectable CCA patients remains unclear, which deserves further investigation. Crucial clinical trials of targeted treatments have been listed in Table 1.

Table 1.

Crucial clinical trials of targeted treatments in unresectable cholangiocarcinoma.

Targets	Drugs	Trials	Phase	Subjects	Number of patients	Treatments and groups	Results	References
FGFR2	Pemigatinib	FIGHT-202NCT02924376	II	Locally advanced or metastatic CCA with different FGFR status	146 (89% ICC)	Pemigatinib FGFR2 fusions or rearrangements group (n = 107) vs. Other FGF/FGFR alterations group (n = 20) vs. No FGF/FGFR alterations group (n = 18)	ORR (35.5% vs. 0% vs. 0%) DCR (82% vs. 40% vs. 22%) mPFS (6.9 vs. 2.1 vs. 1.7 months) mOS (21.1 vs. 6.7 vs. 4.0 months)	[16]
		NCT04256980	II	Locally advanced or metastatic Chinese CCA with FGFR2 fusions or rearrangements	31 (96.8% ICC)	Pemigatinib (n = 31)	ORR 50% DCR 100% mPFS 6.3 months mOS NR	[21]
	Infigratinib	NCT02150967	II	Unresectable, locally advanced or metastatic CCA with FGFR2 alterations	108	Infigratinib (n = 108)	ORR 23.1% mPFS 7.3 months mOS 12.4 months	[26]
	Futibatinib	NCT02052778 (FOENIX-CCA2)	II	Unresectable or metastatic ICC with an FGFR2 fusion or rearrangement	103 (100% ICC)	Futibatinib (n = 103)	ORR 42%	[32]
	Erdafitinib	NCT04083976 (RAGNAR)	II	Advanced or metastatic CCA with FGFR alterations	35	Erdafitinib (n = 35)	ORR 60% DCR 100% mPFS 8.4 months mOS 18.7 months	[38]
	Derazantinib	NCT01752920	I/II	Unresectable ICC with FGFR2 fusion	29 (100% ICC)	Derazantinib (n = 29)	ORR 20.7% DCR 82.8% mPFS 5.7 months mOS NR	[43]
IDH1	Ivosidenib	NCT02989857 (ClarIDHy)	III	Advanced, IDH1-mutant CCA	187 (91.4% ICC)	Ivosidenib (n = 126) vs. Placebo (n = 61)	mPFS (2.7 vs. 1.4 months, one side P <0.0001) mOS (10.3 vs. 5.1 months, one side P <0.001)	[56,57]
BRAF V600E	Dabrafenib	NCT02034110 (ROAR)	II	Unresectable, metastatic, locally advanced, or recurrent BTC with BRAF V600E mutation	43 (91% ICC)	Dabrafenib plus trametinib (n = 43)	ORR (51% assessed by investigator; 47% by independent reviewer) mPFS 9 months mOS 14 months	[66]
NTRK	Larotrectinib	NCT02122913, NCT02637687, NCT02576431	I/II	Metastatic or locally advanced NTRK fusion-positive solid tumor	55 (4% CCA)	Larorectinib (n = 55)	ORR (80% assessed by investigator; 75% by independent reviewer) One of two CCA patients got PR	[68]
	Entrectinib	NCT02650401-NG (STARTRK), NCT02097810 (STARTRK-1), NCT02568267 (STARTRK-2), EudraCT, 2012-000148-88 (ALKA-372-001)	I/II	Metastatic or locally advanced NTRK fusion-positive solid tumor	54 (2% CCA)	Entrectinib (n = 54)	ORR 57% One of one CAA got PR	[69]
RET	Selpercatinib	NCT03157128 (LIBRETTO-001)	I/II	Advanced RET fusion-positive solid tumor	45 (4% CCA)	Selpercatinib (n = 45)	ORR 43.9% One of one CAA got PR assessed by independent reviewer in efficacy analysis	[71]
	Pralsetinib	NCT03037385 (ARROW)	I/II	Advanced RET-altered solid tumor	29 (14% CCA)	Pralsetinib (n = 29)	ORR 57% Two of three CCA got PR in efficacy analysis	[72]
KRAS G12C	Adagrasib	NCT03785249 (KRYSTAL-1)	II	Patients with KRAS G12C-mutated advanced solid tumors	64 (12.5% CCA)	Adagrasib (n = 64)	In CCA cases: ORR 50% mPFS 11.3 months mOS 15.1 months	[75]
EGFR	Erlotinib	NCT01149122	III	Metastatic BTC	268 (67.2% CCA)	GEMOX plus erlotinib (n = 135) vs. GEMOX (n = 133)	ORR (30% vs. 16%, P = 0.005) mPFS (5.8 vs. 4.2 months, P = 0.087) mOS (9.5 vs. 9.5 months, P = 0.611) mPFS of CCA (5.9 vs. 3.0 months, P = 0.049)	[91]
	Cetuximab	NCT00552149 (BINGO)	II	Locally advanced or metastatic BTC	150 (82% CCA)	GEMOX plus cetuximab (n = 76) vs. GEMOX (n = 74)	ORR (23.7% vs. 23%) DCR (81.6% vs. 64.9%) mPFS (6.1 vs. 5.5 months) mOS (11 vs. 12.4 months)	[94]
	Panitumumab	NCT01389414 (Vecti-BIL)	II	Unresectable or metastatic BTC with wild-type KRAS	99 (70.7% CCA)	GEMOX plus panitumumab (n = 45) vs. GEMOX (n = 44)	mPFS (5.3 vs. 4.4 months, P = 0.27) mOS (9.9 vs. 10.2 months, P = 0.42) ORR (26.6% vs. 18.1%) DCR (75.5% vs. 68.1%) mOS of ICC (15.1 vs. 11.8 months, P = 0.13)	[100]
HER2	Trastuzumab	NCT04722133 (KCSG-HB19-14)	II	HER2-positive advanced BTC	34 (47% CCA)	FOLFOX plus trastuzumab (n = 34)	ORR 29.4% DCR 79.4% mPFS 5.1 months mOS 10.7 months	[113]
		NCT04579380 (SGNTUC-019)	II	Previously treated HER2-positive metastatic BTC	30 (50% CCA)	Tucatinib plus trastuzumab (n = 30)	ORR 46.7% DCR 76.7%	[114]
		NCT02091141 (MyPathway)	II	HER2-positive metastatic advanced BTC	39 (36% CCA)	Pertuzumab plus trastuzumab (n = 39)	ORR 23% DCR 51.3%	[115]
	Zanidatamab	NCT04466891 (HERIZON-BTC-01)	IIb	HER2-amplified, unresectable, locally advanced, or metastatic BTC	87 (48.3% CCA)	Zanidatamab HER2 IHC 2+ or 3+ (n = 80, cohort1) vs. HER2 IHC 0 or 1+ (n = 7, cohort2)	ORR (33% vs. 0%) DCR 68.8% in cohort1 mPFS (5.5 vs. 1.9 months) mOS (NR vs. 5.5 months)	[117]
	DS-8201	HERB	II	Unresectable or recurrent BTC with HER2 expression	30 (46.7% CCA)	DS-8201 (n = 30)	HER2-positive (n = 22): ORR 36.4% DCR 81.8% HER2-low (n = 8): ORR 12.5% DCR 75%	[118]
VEGF/VEGFR	Bevacizumab	NCT00361231	II	Locally unresectable or metastatic BTC	35 (71.4% CCA)	GEMOX plus bevacizumab (n = 35)	ORR 40% DCR 68.6% mPFS 7 months mOS 12.7 months	[124]
	Ramucirumab	NCT02711553	II	Unresectable, recurrent, or metastatic BTC	207 (69.1% CCA)	Gemcitabine/cisplatin plus ramucirumab (n = 106) vs. Gemcitabine/cisplatin (n = 101)	mPFS (6.5 vs 6.6 months, P = 0.48) mOS (10.5 vs. 13 months, P = 0.087) ORR (31.1% vs. 32.7%, P = 0.88) DCR (75.5% vs. 68.1%)	[128]
	Surufatinib	NCT02966821	II	Unresectable or metastatic BTC	39 (87.2% CCA)	Surufatinib (n = 39)	16-week PFS rate 46.33% ORR 0% DCR 81.5% mPFS 3.7 months mOS 6.9 months	[138]
	Regorafenib	NCT02162914 (REACHIN)	II	Unresectable or metastatic BTC	66 (86.4% CCA)	Regorafenib (n = 33) vs. Placebo (n = 33)	ORR 0% DCR (70% vs. 33%, P = 0.002) mPFS (3 vs. 1.5 months, P = 0.004) mOS (5.3 vs. 5.1 months, P = 0.28)	[141]

BRAF: V-raf murine sarcoma viral oncogene homolog B; BTC: Biliary tract cancer; CCA: Cholangiocarcinoma; DCR: Disease control rate; EGFR: Epidermal growth factor receptor; FGFR2: Fibroblast growth factor receptor 2; FOLFOX: 5-Fluorouracil plus leucovorin calcium and oxaliplatin; GEMOX: Gemcitabine plus oxaliplatin; HER2: Human epidermal growth factor receptor 2; ICC: Intrahepatic cholangiocarcinoma; IDH1: Isocitrate dehydrogenase-1; IHC: Immunohistochemistry; KRAS: Kirsten rat sarcoma viral oncogene homolog; mOS: Median overall survival; mPFS: Median progression free survival; NR: Not reached; NTRK: Neurotrophic tropomyosin receptor kinase; ORR: Objective response rate; OS: Overall survival; PFS: Progression-free survival; PR: Partial response; RET: Rearrange during transfection; VEGF: Vascular endothelial growth factor; VEGFR: Vascular endothelial growth factor receptor.

Immunotherapy

The main purpose of immunotherapy is to activate the human immune system to attack tumor cells. Compared with chemotherapy, immunotherapy increases antitumor immune responses with less off-target effects. Therefore, immunotherapy is considered a powerful strategy in cancer treatment.

Immune checkpoint inhibitors

Immune checkpoints, usually existing in ligand-receptor forms, play important roles in suppressing immune responses to self-antigens. However, they are also crucial for immune escape in cancer. Activated T cells express several surface inhibitory receptors, including cytotoxic T lymphocyte-associated antigen-4 (CTLA-4), programmed cell death 1 protein (PD-1), and others. To evade attack by the immune system, tumor cells abnormally express the ligands of these receptors, e.g., programmed cell death 1 ligand 1 (PD-L1), thereby inhibiting the function of cytotoxic T cells through binding with inhibitory receptors.^[147] Immune checkpoint inhibitors (ICIs), used to interfere these receptor-ligand interactions, are considered one of the most crucial types of immunotherapies in cancer. ICIs showed favorable efficacy in several tumors, highlighting their huge potential in tumor treatment, including CCA [Figures 2–4].

PD-1/PD-L1

Durvalumab

Durvalumab is a humanized, highly selective IgG1 monoclonal antibody that inhibits PD-L1 binding to PD-1. Previously, durvalumab monotherapy or durvalumab plus tremelimumab (an inhibitor of CTLA-4) showed initial efficacy in Asian advanced BTC cases.^[148] Besides, gemcitabine and cisplatin plus durvalumab with or without tremelimumab also showed promising activity and acceptable safety as first-line therapy in advanced BTC cases.^[149] In the TOPAZ-1 study, a randomized, placebo-controlled phase III trial, adding durvalumab to gemcitabine and cisplatin as first-line therapy significantly improved the survival outcomes of advanced BTC cases (n = 685, 75% CCA) compared with gemcitabine and cisplatin plus placebo (OS, 12.8 months vs. 11.5 months, P = 0.021; PFS, 7.2 months vs. 5.7 months, P = 0.001).^[150] Besides, according to the updated survival data of TOPAZ-1, durvalumab plus gemcitabine and cisplatin improved the OS rate at 12 months (54.3% vs. 47.1%), 18 months (34.8% vs. 24.1%), and 24 months (23.6% vs. 11.5%) and prolonged the mOS (12.9 months vs. 11.3 months) versus placebo.^[151] Based on the results of TOPAZ-1, combination of gemcitabine/cisplatin with durvalumab has been recommended by NCCN and CSCO guidelines as first-line therapy for unresectable and metastatic BTC patients.

Pembrolizumab

Pembrolizumab is a humanized monoclonal antibody against PD-1. Combination of pembrolizumab with capecitabine and oxaliplatin showed potential efficacy in advanced BTC patients (n = 11, 73% CCA cases) in a study by Monge et al^[152]. KEYNOTE-966 was a randomized, placebo-controlled phase III trial, which aimed to examine whether adding pembrolizumab to gemcitabine and cisplatin could improve prognosis in advanced BTC patients compared with gemcitabine and cisplatin alone.^[153] This study enrolled 1069 patients (78.2% CCA), who were divided into the pembrolizumab and placebo groups. The results demonstrated significantly prolonged OS in patients administered gemcitabine/cisplatin plus pembrolizumab compared with those treated with gemcitabine/cisplatin plus placebo (12.7 months vs. 10.9 months, P = 0.0034), with no new safety signals. Besides, monotherapy of pembrolizumab achieved an ORR of 34.3% in previously treated patients with unresectable non-colorectal solid tumors with high microsatellite instability (MSI-H)/DNA mismatch repair (dMMR), including BTC.^[154] Overall, chemotherapy plus pembrolizumab provides survival benefits in unresectable ICC cases, and monotherapy of pembrolizumab may be a potential option for cases with MSI-H/dMMR. Combination of gemcitabine/cisplatin with pembrolizumab has also been recommended by NCCN and CSCO guidelines as first-line therapy for unresectable and metastatic BTC patients.

Nivolumab

Nivolumab is a humanized IgG4 monoclonal anti-PD-1 antibody with high affinity and selectivity. Nivolumab monotherapy showed modest efficacy in previously treated advanced BTC patients.^[155] In Japanese advanced BTC cases, nivolumab also showed clinical efficacy and acceptable safety.^[156] In a single-arm phase II study, combination of nivolumab with gemcitabine and cisplatin provided encouraging efficacy in advanced BTC cases (n = 32, 81.3% CCA), with an ORR of 55.6% and a DCR of 92.6%.^[157] However, in the BilT-01 trial, addition of nivolumab to gemcitabine and cisplatin or ipilimumab (a CTLA-4 inhibitor) failed to improve 6-month PFS in advanced BTC cases.^[158] A phase II trial combining nivolumab with stereotactic body radiation therapy in advanced CCA patients was also carried out.^[159] Taken together, although nivolumab demonstrated initial clinical efficacy, it offered no survival benefits. Nivolumab is therefore a potential choice for unresectable CCA cases but needs further investigation for validation.

Camrelizumab

Camrelizumab is a fully humanized IgG4 monoclonal antibody targeting PD-1. In 37 evaluable advanced BTC cases enrolled in a phase II trial administered the combination of GEMOX with camrelizumab, the ORR was 54%, and mPFS and mOS were 6.1 and 11.8 months, respectively.^[160] Specifically, the ORR of patients with PD-L1 tumor proportion score (TPS) ≥1% was 80%, while that of PD-L1 TPS <1% was 53.8%. Another phase II trial conducted in China aimed to assess the efficacy and safety of camrelizumab combined with GEMOX or FOLFOX4 as first-line therapy in advanced BTC cases.^[161] A total of 92 patients were enrolled, of whom 29 (100% ICC) were administered FOLFOX4 plus camrelizumab while the remaining (82.5% CCA) were treated with GEMOX plus camrelizumab. The ORRs were 10.3% and 19%, while the DCRs were 75% and 89.7% in both treatment groups, respectively. Overall, combination of camrelizumab with chemotherapy is effective and tolerable for unresectable CCA patients and should be further examined in phase III trials.

Penpulimab

Penpulimab is a humanized, high-affinity IgG1 anti-PD-1 monoclonal antibody with high specificity. It is currently in the experimental stage, with promising efficacy and acceptable safety in advanced upper gastrointestinal cancers, including CCA.^[162] Of nine CCA patients, one had a PR and three had SD. Therefore, penpulimab warrants further examination in large cohorts.

CTLA-4

Tremelimumab

Tremelimumab is a blocking antibody targeting CTLA-4. Previous studies have demonstrated that combination of tremelimumab with durvalumab and chemotherapy has clinical efficacy in advanced BTC cases.^[148,149] Besides, tremelimumab in combination with microwave ablation increased circulating activated CD8⁺ T cells in refractory BTC patients (n = 20, 90% CCA).^[163] Overall, tremelimumab may be a potential strategy in the treatment of unresectable CCA cases.

Ipilimumab

Ipilimumab, a humanized anti-CTLA-4 monoclonal antibody, is commonly investigated in combination with nivolumab in clinical trials of BTC. Although nivolumab plus ipilimumab conferred no survival benefits,^[158] it achieved an ORR of 31% in 16 advanced ICC patients.^[164] However, none patients with extrahepatic cholangiocarcinoma achieved objective response. Therefore, combination of nivolumab with ipilimumab may be a latent option for the clinical treatment of unresectable ICC.

Adoptive cell therapy

Adoptive cell therapy (ACT) has emerged as a potential and promising therapeutic option for several tumors. Through infusion of engineered T cells in vivo, ACT generates a powerful immune response to suppress the tumor. Depending on the source of T cells, ACT could be divided into the autologous and allogeneic types. Autologous T cells include specific T cells isolated from naturally occurring tumor tissues and genetically modified T cells derived from blood.^[165]

Chimeric antigen receptor–modified T (CAR-T) cells

Application of CAR-T cells, one type of genetic engineered T cells, has achieved huge success in hematologic malignancies and has been assessed in several solid tumors. In EGFR-positive unresectable BTC patients (n = 19, 73.7% CCA), the efficacy and safety of EGFR-specific CAR-T were evaluated in a phase I trial.^[166] The results showed that one patient had a CR while ten cases had SD, and the enrichment of central memory T cells improved the clinical outcomes. Besides, HER2-specific CAR-T has also shown significant clinical efficacy in HER2-positive unresectable BTC cases.^[167] Taken together, CAR-T is a novel promising therapeutic option for unresectable CCA patients, which warrants further investigation.

Allogeneic ACT

Transfer of gamma delta (γδ) T cells, a T cell type with T cell receptors composed of γ and δ chains and natural killer cell receptors, is safe and effective for lung and liver cancers. In a study by Zhang et al^[168], locoregional therapy plus allogeneic γδ T cell therapy significantly prolonged mPFS in unresectable ICC patients compared with locoregional therapy (8 months vs. 4 months, P = 0.021). Therefore, allogeneic γδ T cell therapy is a potential option for unresectable CCA cases.

Combination of Targeted Therapy with Immunotherapy

Since targeted therapy and immunotherapy have shown encouraging efficacy in unresectable CCA patients, their combination has attracted increasing attention. A series of clinical trials have been conducted subsequently.

Toripalimab is a humanized IgG4 monoclonal antibody targeting PD-1. Combination of toripalimab with lenvatinib and GEMOX as first-line therapy showed favorable efficacy and tolerated safety in advanced ICC cases (n = 30).^[169] The ORR of the unresectable ICC patients was 80%, and mPFS and mOS were 10.2 and 22.5 months, respectively. A randomized phase III trial was performed by the National Medical Products Administration to validate these results. In addition, combination of anlotinib with an inhibitor of PD-L1 (TQB2450) in pretreated advanced BTC patients also illustrated promising efficacy and controlled safety.^[170] The ORR reached 21.21% and DCR reached 72.73%, and mPFS and mOS were 6.24 and 15.77 months, respectively. Besides, avelumab, an anti-PD-L1 antibody, in combination with regorafenib has also showed anti-tumor activity in unresectable BTC cases (n = 34, 97.1% CCA).^[171] Last but not least, adding bevacizumab to chemotherapy plus atezolizumab, an inhibitor of PD-L1, may modestly improve PFS but not OS in advanced BTC patients (n = 162, 73.5% CCA) versus placebo.^[172] Particularly, higher expression of VEGFA was related to improved PFS in bevacizumab arm, which made VEGFA a potential predictive biomarker.

In a phase Ib/II study, camrelizumab in combination with apatinib showed favorable efficacy and acceptable safety in patients with advanced primary liver cancer (n = 28, 25% ICC).^[173] In addition, although ramucirumab plus pembrolizumab exhibits limited clinical activity in advanced BTC patients, those with PD-L1-positive tumors have longer OS than those with PD-L1-negative tumors.^[174]

In conclusion, combination of targeted therapy with immunotherapy, especially combining anti-angiogenesis with an ICI, has a favorable prospect in the treatment of unresectable CCA. Crucial clinical trials of immunotherapy and combined therapy have been listed in Table 2. Further combination regimens are being explored, and the results are worth looking forward to.

Table 2.

Crucial clinical trials of immunotherapy and combined therapy in unresectable cholangiocarcinoma.

Types	Drugs	Trials	Phase	Subjects	Number of patients	Treatments and groups	Results	References
ICI
PD-L1	Durvalumab	NCT03875235 (TOPAZ-1)	III	Previously untreated, unresectable, locally advanced or metastatic BTC	685 (75% CCA)	Gemcitabine /cisplatin plus durvalumab (n = 341) vs. Gemcitabine /cisplatin plus placebo (n = 344)	mOS (12.8 vs. 11.5 months, P = 0.021) mPFS (7.2 vs. 5.7 months, P = 0.001) Updated mOS (12.9 vs. 11.3 months)	[150,151]
PD-1	Pembrolizumab	NCT04003636 (KEYNOTE-966)	III	Previously untreated, unresectable, locally advanced or metastatic BTC	1069 (78.2% CCA)	Gemcitabine /cisplatin plus pembrolizumab (n = 533) vs. Gemcitabine/cisplatin plus placebo (n = 536)	mOS (12.7 vs. 10.9 months, P = 0.0034)	[153]
	Nivolumab	NCT03101566 (BilT-01)	II	Unresectable, locally advanced or metastatic BTC	68 (79.4% CCA)	Gemcitabine /cisplatin plus nivolumab (n = 35) vs. Nivolumab plus ipilimumab (n = 33)	6-month PFS rate (59.4% vs. 21.2%) mPFS (6.4 vs. 3.9 months, P = 0.077) mOS (10.6 vs.8.2 months, P = 0.61) ORR (22.9% vs. 3%)	[158]
	Camrelizumab	NCT03486678	II	Advanced BTC	38	GEMOX plus camrelizumab (n = 37)	6-month PFS 50% ORR 54% mPFS 6.1 months mOS 11.8 months	[160]
CTLA-4	Tremelimumab	NCT03046862	II	Unresectable or recurrent BTC	138	Gemcitabine /cisplatin followed by gemcitabine /cisplatin plus durvalumab and tremelimumab (n = 32) vs. Gemcitabine /cisplatin plus durvalumab (n = 49) vs. Gemcitabine /cisplatin plus durvalumab and tremelimumab (n = 47)	ORR 50% vs. 72% vs. 70%	[149]
	Ipilimumab	NCT02923934 (CA209-538)	II	Advanced BTC	39 (67% CCA)	Nivolumab plus ipilimumab (n = 39)	ORR 23% DCR 44% mPFS 2.9 months mOS 5.7 months	[164]
ACT
Allogeneic ACT	γδ T cell	NCT03183219	I/II	Unresectable ICC	29 (100% ICC)	Locoregional therapy (n = 15) vs. Locoregional therapy plus γδ T cell therapy (n = 14)	Distant mPFS (4 vs. 8 months, P = 0.021)	[168]
Combination with targeted therapy
PD-1/PD-L1 and VEGF	Toripalimab plus lenvatinib	NCT03951597	II	Advanced ICC	30 (100% ICC)	GEMOX plus toripalimab and lenvatinib (n = 30)	ORR 80% DCR 93.3% mPFS 10.2 months mOS 22.5 months	[169]
	Anlotinib plus TQB2450	NCT03825705 NCT03996408	Ib	Pretreated advanced BTC	66	Anlotinib plus TQB2450 (n = 66)	ORR 21.21% DCR 72.73% mPFS 6.24 months mOS 15.77 months	[170]
	Avelumab plus regorafenib	NCT03475953 (REGOMUNE)	II	Advanced BTC	34 (97.1% CCA)	Avelumab plus regorafenib (n = 34)	ORR 13.8% DCR 51.7% mPFS 2.5 months mOS 11.9 months	[171]
	Atezolizumab plus bevacizumab	NCT04677504 (IMbrave151)	II	Advanced BTC	162 (73.5% CCA)	Gemcitabine /cisplatin plus atezolizumab and bevacizumab (n = 79) vs. Gemcitabine /cisplatin plus atezolizumab and placebo (n = 83)	mPFS (8.3 vs. 7.9 months) mOS (14.9 vs. 14.6 months) ORR (26.6% vs. 26.5%) DCR (79.7% vs. 75.9%)	[172]

ACT: Adoptive cell therapy; BTC: Biliary tract cancer; CCA: Cholangiocarcinoma; CTLA-4: Cytotoxic T lymphocyte-associated antigen-4; DCR: Disease control rate; GEMOX: Gemcitabine plus oxaliplatin; ICC: Intrahepatic cholangiocarcinoma; ICI: Immune checkpoint inhibitor; mOS: Median overall survival; mPFS: Median progression free survival; ORR: Objective response rate; OS: Overall survival; PD-1: Programmed cell death 1 protein; PD-L1: Programmed death-ligand 1; PFS: Progression-free survival; VEGF: Vascular endothelial growth factor.

Discussion and Prospects

Over the past few decades, the incidence of CCA, especially ICC, has increased in most countries, with an increase of more than 140% in the United States particularly.^[175] Despite the rapid development of treatment tools, CCA prognosis remains dismal. The absence of early diagnosis leads to most CCA cases being unresectable. Based on randomized phase III clinical trials, the standard treatment regimens for unresectable CCA have been established.^[7,8] In recent years, targeted therapy and immunotherapy have developed rapidly, providing additional options for unresectable CCA patients.

Due to the molecular heterogeneity based on different tumor locations in BTC, identifying molecular subtypes and developing precise treatment are crucial for CCA management. With advances in the next-generation sequencing technology, additional targetable genetic alterations have been identified in unresectable CCA patients. The FGFR2 fusion has been identified in over 10% of ICC patients. In the FIGHT-202 trial, pemigatinib had improved ORR, PFS and OS in FGFR2-fused CCA patients compared with those harboring other FGFR2 mutations or wild-type cases.^[16] Besides, other FGFR inhibitors, including infigratinib, futibatinib, erdafitinib, and derazantinib, have also shown promising efficacy in CCA.^{[26,32,38,43]} Therefore, FGFR inhibitor has been recommended as second-line therapy for previously treated unresectable CCA patients with FGFR2 alterations. Pemigatinib and futibatinib have received approval from the FDA and have been included in NCCN guidelines. Particularly, as the first irreversible FGFR inhibitor, futibatinib possesses promising prospective in the treatment of CCA. In addition, the IDH1 inhibitor ivosidenib significantly prolongs OS and PFS in unresectable CCA patients with IDH1 mutations, which made it approved as second-line therapy for these patients by the FDA and EMA and has been included in NCCN and CSCO guidelines.^[56]

Based on the ROAR study, dabrafenib plus trametinib was recommended for CCA patients with BRAF V600E mutations.^[66] Besides, although clinical trials examining CCA patients with NTRK or RET fusions are rare, multiple basket trials have been highly successful.^{[68,69,71,72]} Therefore, the NTRK inhibitors, larotrectinib and entrectinib, are recommended as first-line therapies for NTRK-fused ICC, while the RET inhibitors, selpercatinib and pralsetinib, are considered suitable second-line therapies for cases with RET fusions. Overall, targeting altered FGFR2, IDH1, BRAF V600E, NTRK, and RET has huge value in the treatment of unresectable CCA and related drugs have been included in CSCO guidelines, except for infigratinib. Last but not least, due to the success of the KRAS G12C inhibitor adagrasib, unresectable CCA patients with KRAS G12C mutation may benefit from it,^[75] and adagrasib has been recommended in NCCN guidelines for these patients.

In addition to targeting mutant genes, several amplified or overexpressed molecules also possess targetable value. As a potential target, EGFR is overexpressed in CCA. Although not improving survival outcomes, EGFR inhibitors in combination with chemotherapy may provide several benefits in specific patients, especially ICC cases.^[91,96,100] Intriguingly, according to subgroup analysis, CCA and ICC patients had enhanced benefits from erlotinib and panitumumab treatment, respectively, which may offer a reference for the clinical treatment of CCA. In HER2-positive CCA patients, the SGNTUC-019 and MyPathway studies showed that trastuzumab in combination with tucatinib or pertuzumab has encouraging efficacy and acceptable safety.^[114,115] Taken together, ERBB1/2 may become promising and valuable therapeutic targets for unresectable CCA. A series of clinical trials of ERBB pathway inhibitors are underway.

Angiogenesis plays a crucial role in tumor development, and VEGF is overexpressed in over 50% of CCA cases. Therefore, anti-angiogenesis therapy has attracted substantial attention. However, the role of anti-angiogenesis in BTC remains unclear. Current clinical trials have not yielded satisfactory results, possibly due to drug resistance caused by tumor and stromal components in the tumor microenvironment and the lack of selection of suitable patients to undergo anti-angiogenesis treatment. Besides, the current evidence for choosing anti-VEGF/VEGFR monotherapy or combined with chemotherapy in CCA is insufficient. Most data were generated from small cohort studies, and the role of anti-VEGF/VEGFR in CCA remains unclear, which warrants validation by large cohort studies in the future.

Use of ICIs is becoming a crucial strategy in cancer treatment. However, the efficacy of ICI monotherapy is limited in CCA, except in patients with MSI-H/dMMR.^[154] Therefore, ICIs are usually applied in combination regimens for unresectable CCA treatment. Based on the TOPAZ-1 and KEYNOTE-966 trials, addition of durvalumab or pembrolizumab to chemotherapy was recommended as first-line therapy in unresectable CCA patients since it significantly improved survival outcomes.^[150,153] Specifically, high PD-L1 expression was not associated with improved outcome with chemotherapy plus immunotherapy in these two studies. Therefore, immunotherapy should generally be added to first-line chemotherapy of unresectable CCA. Besides, combination of nivolumab or camrelizumab with chemotherapy also showed encouraging efficacy in unresectable CCA cases.^[157,160] Nevertheless, application of an ICI in second-line therapy for unresectable CCA patients still lacks high-quality evidence, which requires further investigations. Particularly, combination of an ICI with anti-angiogenesis demonstrated promising efficacy in unresectable ICC.^{[169,171,172,173,174]} Actually, the efficacy of ICIs is largely determined by the amount of infiltrated antitumor immune cells. However, abnormal angiogenesis results in reduced infiltration of antitumor immune cells, while promoting the infiltration of immunosuppressive cells.^[176] Therefore, anti-angiogenesis plus ICI administration may become a promising option for unresectable CCA patients, warranting further exploration.

Notably, several patients were initially non-response to treatment, and several others possessed limited PFS, which indicated the presence of primary and secondary resistance. Therefore, it is crucial to identify sensitive biomarkers or genetic status for the prediction of therapeutic efficacy. In the FIGHT-202 trial, BAP1 was identified as the most frequently co-mutant gene with FGFR2, and patients harboring BAP1 co-alterations exhibited significantly shorter mPFS compared to those with unaltered BAP1 (6.9 months vs. 9.1 months, P = 0.06). Besides, patients with CDKN2A/B or PBRM1 mutations also possessed a shorter mPFS compared with those with unaltered CDKN2A/B or PBRM1 (CDKN2A/B, 6.4 months vs. 9.0 months, P = 0.03; PBRM1, 4.7 months vs. 7.0 months, P = 0.05), and those with TP53 mutations illustrated non-response to pemigatinib and possessed shorter mPFS compared with patients with unaltered TP53 (2.8 months vs. 9.0 months, P = 0.0003).^[177,178] When it comes to ICC patients with IDH1 mutations, 2-HG was recognized associated with tumor burden.^[45] However, the result of biomarker analysis of ROAR study was inconclusive.^[66] The mechanism of resistance of dabrafenib plus trametinib was still unclear.

For patients treated with EGFR inhibitors, tumors that occur in different locations seem to respond differently to treatment. In subgroup analyses, CCA patients benefited more from GEMOX plus erlotinib, while ICC patients achieved more survival benefits from GEMOX plus panitumumab compared with GEMOX.^[91,100] Besides, it was also reported that ICC patients possessed enhanced therapeutic benefits from surufatinib than other BTC patients.^[138] Therefore, tumor locations may partly predict the efficacy of several treatments.

In addition, the expression of several key proteins may also be correlated to treatment efficacy. For instance, the ORR of HER2-overexpressing (IHC 3+) unresectable BTC cases treated with DS-8201 was higher.^[119] Besides, for patients treated with GEMOX with camrelizumab, the ORR of patients with high PD-L1 expression achieved higher ORR compared with those with low expressions.^[160] Taken together, genetic status, tumor locations, and protein expression levels are all potentially associated with drug resistance or treatment efficacy. Treatment decision should be more individual and precise based on patients’ own conditions.

Except for these targets discussed before, various other targets are also being developed and researched. According to the proteogenomic characterization of Chinese ICC patients performed by Dong et al^[179], TP53 and KRAS were most frequently mutant genes. However, drug development against these two targets is still difficult. Several inhibitors such as adagrasib^[75] and sotorasib^[180] targeting KRAS G12C have succeeded, but their functions in CCA were still needed to be further explored. Therefore, more attentions should be paid to treatments targeting these most frequently mutant genes in CCA.

Besides, according to the study of Dong et al^[179], KRAS alteration was mutually exclusive with IDH1/2, BAP1, and FGFR2 mutations, while FGFR2 mutation mutually excluded TP53, KRAS, and IDH1/2 alterations, which offered several references to clinical treatment decision after genetic sequencing. Therefore, it is also urgent to better understand the relationships between alterative genes in CCA treatment exploration.

Futhermore, several gene alterations such as FGFR2 fusion may generate peptide correlated to T cell response,^[179] which indicated its potential therapeutic value combined with immunotherapy. Actually, most combination of targeted therapy with immunotherapy in CCA was ICI plus anti-angiogenesis treatment. Overall, new combination treatments should also be explored based on changes in signaling pathways and microenvironment caused by gene alterations, and the researches should be focus on the detailed mechanism of different combined therapy.

In conclusion, targeted therapy and immunotherapy are novel treatment approaches, which promote the update and development of clinical guidelines for unresectable CCA. Targeted therapy and immunotherapy provide additional options and benefits for unresectable CCA patients, especially those with alterations or expression changes in specific genes. Therefore, developing novel targets and biomarkers should be accelerated to determine the specific population who may benefit from targeted therapy and immunotherapy. Furthermore, resistance to targeted therapy and adverse events related to immunotherapy require further attention from the scientific community. With improved understanding of the molecular subtypes and immune microenvironments of unresectable CCA cases, as well as the progressive transformation from basic experiments to clinical applications, the development of targeted therapy and immunotherapy will be accelerated continuously, improving patient outcomes.

Funding

This study was funded by grants from the National Natural Science Foundation of China (Nos. 82171824, 82272906, 82103087, 82002625, and 82373133), the Scientific and Technological Innovation Project of Science and Technology Commission of Shanghai Municipality (No. 21JC1404300), the Innovation Group Project of Shanghai Municipal Health Commission (No. 2019CXJQ03), the Shanghai Municipal Commission of Health and Family Planning (No. 2018ZHYL0223), Shanghai Municipal Education Commission-Gao Feng Clinical Medicine Grant Support (No. 20161312), Shanghai Key Clinical Specialty (Oncology), Shanghai Leading Talents Project, Clinical Research Plan of SHDC (No. SHDC2020CR1035B), Shanghai Sailing Program (No. 20YF1446400), the National Key R&D Program of China (No. 2019YFC1315900), Project from CSCO Clinical Oncology Research Foundation (No. Y-2019AZZD-0513), and the Innovative Research Team of High-Level Local Universities in Shanghai (No. SHSMU-ZDCX20210802).

Conflicts of interest

None.