Evolution of the Targeted Therapy Landscape for Cholangiocarcinoma: Is Cholangiocarcinoma the ‘NSCLC’ of GI Oncology?

Simple Summary In the past 20 years, the development of targeted therapies that can be matched to a tumor’s molecular and immune abnormalities has resulted in the improvement of outcomes for patients suffering from advanced aggressive malignancies. Remarkably, non-small cell lung cancer (NSCLC) has become the poster child for a lethal malignancy in which numerous molecular aberrations have become druggable. Similar to NSCLC, there are limited responses in cholangiocarcinoma (CCA) to conventional chemotherapy. Next-generation sequencing has identified novel genomic alterations in CCA that vary between patients. Gene- and immune-targeted therapies are leading to a new era of precision/personalized medicine for patients with CCA. Herein, we review the current status of molecularly matched precision-targeted therapy for CCA. Abstract In the past two decades, molecular targeted therapy has revolutionized the treatment landscape of several malignancies. Lethal malignancies such as non-small cell lung cancer (NSCLC) have become a model for precision-matched immune- and gene-targeted therapies. Multiple small subgroups of NSCLC defined by their genomic aberrations are now recognized; remarkably, taken together, almost 70% of NSCLCs now have a druggable anomaly. Cholangiocarcinoma (CCA) is a rare tumor with a poor prognosis. Novel molecular alterations have been recently identified in patients with CCA, and the potential for targeted therapy is being realized. In 2019, a fibroblast growth factor receptor 2 (FGFR2) inhibitor, pemigatinib, was the first approved targeted therapy for patients with locally advanced or metastatic intrahepatic CCA who had FGFR2 gene fusions or rearrangement. More regulatory approvals for matched targeted therapies as second-line or subsequent treatments in advanced CCA followed, including additional drugs that target FGFR2 gene fusion/rearrangement. Recent tumor-agnostic approvals include (but are not limited to) drugs that target mutations/rearrangements in the following genes and are hence applicable to CCA: isocitrate dehydrogenase 1 (IDH1); neurotrophic tropomyosin-receptor kinase (NTRK); the V600E mutation of the BRAF gene (BRAFV600E); and high tumor mutational burden, high microsatellite instability, and gene mismatch repair-deficient (TMB-H/MSI-H/dMMR) tumors. Ongoing trials investigate HER2, RET, and non-BRAFV600E mutations in CCA and improvements in the efficacy and safety of new targeted treatments. This review aims to present the current status of molecularly matched targeted therapy for advanced CCA.


Introduction
In the past two decades, molecular targeted and immune therapy has revolutionized the treatment landscape of several malignancies. Recent findings highlighted that matched targeted therapy improved the response rate and prolonged the survival of patients with advanced cancers [1,2]. Some of the most notable achievements are in non-small cell lung cancer (NSCLC), a disease for which the number of actionable biomarkers has increased rapidly. Indeed, NSCLC is now almost a poster child for the benefits of precision medicine, with almost 70% of NSCLCs having a biomarker-based therapy (including but not limited to EGFR and ERBB2 alterations, mismatch repair gene defects, high tumor mutational burden, BRAF V600E, NTRK fusions, ALK fusions, ROS1 fusions, and RET alterations) [3][4][5][6][7][8][9][10]. Moreover, molecular diagnostics have supported the development of drugs and therapeutic antibodies targeting specific receptors, antigens, or molecular pathways crucial for tumor cell proliferation and invasion, tumor growth, immunity, and metastases across various malignancies [11].
Cholangiocarcinoma (CCA) is a promising candidate for targeted therapy due to its diverse molecular features [12]. The incidence of CCA is low in the Western world, with between 0.35 and 2 cases per a 100,000 population per year [13]. The global incidence of CCA has steadily increased over the last 30 years, from 0.1 to 0.6 cases per a 100,000 population [13]. CCA is a highly aggressive malignancy, with a 5-year OS for locally advanced or metastatic disease of less than 10% [14]. CCA arises from the intrahepatic and extrahepatic biliary epithelium [15]. Anatomically, 60-70% of cases are classified as perihilar CCA (pCCA); 20-30% of cases are distal CCA (dCCA); and 5-10% of cases are intrahepatic CCA (iCAA) [15]. Current population statistics show an increased prevalence of CCA [15]. In early-stage CCA, the primary treatment includes surgical resection and adjuvant chemotherapy, while systemic chemotherapy is the standard treatment for advanced-stage CCA [16]. Patients with CCA often present with late-stage disease, and the prognosis is poor [16].
Molecular techniques, including next-generation sequencing (NGS), have identified novel mutations in tumors from patients with CCA ( Figure 1) [17]. Patients with CCA show substantial variations in their molecular profiling and genetic aberrations according to their anatomic locations (Table 1). Promising therapeutic molecular targets for CCA have been identified and include isocitrate dehydrogenases 1 and 2 (IDH1 and 2) and products of fusions of the fibroblast growth factor receptor 2 (FGFR2) gene [17]. In 2019, the US Food and Drug Administration (FDA) granted accelerated approval for pemigatinib, an FGFR2 inhibitor, as the first targeted therapy for locally advanced or metastatic iCCA with FGFR2 fusions or rearrangement [18]. Another approval for a targeted therapy specifically for CCA was on 28 May 2021 [19]. The FDA granted accelerated approval to infigratinib, a kinase inhibitor approved for adults with previously treated, locally advanced, unresectable, or metastatic CCA with an FGFR2 gene fusion or rearrangement [19]. This specific approval for CCA required confirmation that an FDA-approved diagnostic test detected FGFR2 gene fusion or rearrangement in CCA [19]. More recently, the FDA granted accelerated approval to futibatinib, an irreversible FGFR1-4 Inhibitor, for previously treated, locally advanced, unresectable, or metastatic CCA with an FGFR2 gene fusion or rearrangement on 30 September 2022 [20]. In support of the recent developments in targeted therapies, in August 2021, the FDA approved ivosidenib as a targeted therapy for adult patients with unresectable locally advanced or metastatic CCA with a mutation in the IDH1 gene detected by an FDA-approved diagnostic test [21]. Finally, ALK and ROS1 mutations occur in between 3% and 9% of patients with CCA, and ALK-positive and ROS1-positive CCA may also be treated with ALK inhibitors [22].    There are also several tumor-agnostic approvals that encompass CCA. For example, in May 2017, the FDA granted accelerated approval for the therapeutic monoclonal antibody, pembrolizumab, for microsatellite instability-high (MSI-H) or mismatch repair deficient (dMMR) unresectable or metastatic solid tumors that have progressed despite prior treatment [36]. In 2020, pembrolizumab received FDA approval for adults and children with high tumor mutational burden (TMB-H) solid tumors [37]. In August 2021, accelerated approval was granted for dostarlimab for adult patients with recurrent or advanced solid tumors identified as mismatch repair deficient (dMMR) as determined by an FDA-approved diagnostic test [38]. In 2018 and 2019, two therapies targeting neurotrophic tyrosine receptor kinase (NTRK) gene fusion, entrectinib and larotrectinib, were approved for locally advanced or metastatic solid tumors [39,40]. In June 2022, the FDA approved dabrafenib combined with trametinib to treat unresectable or metastatic solid tumors with a BRAF V600E mutation [41].
This review aims to present and discuss the advances in molecular targeted therapy for patients with advanced CCA.

Neurotrophic Tyrosine Receptor Kinase (NTRK) Gene Fusion-Positive Cholangiocarcinoma (CCA)
Molecular profiling of solid tumors has identified clinically actionable fusions of the NTRK1, NTRK2, and NTRK3 genes, which encode neurotrophic tropomyosin receptor kinase (NTRK) [42]. NTRK fusion products activate the TRK gene and, subsequently, the downstream signaling pathways, PI3K and MAPK, leading to tumor cell proliferation and invasion [43,44]. Therefore, NTRK inhibitors are promising targeted therapies for patients with NTRK fusion-positive cancers and have shown antitumor responses in NCSLC, melanoma, and other tumors [43,44]. NTRK gene fusions have been identified in 1-3% of patients with CCA [45]. Table 2 shows the results of pivotal clinical trials that assessed the outcomes of NTRK inhibitors in NTRK fusion-positive metastatic or unresectable locally advanced solid tumors. Larotrectinib is a first-generation, highly selective pan-NTRK competitive inhibitor that suppresses cancer cell proliferation [62]. It has shown immediate, robust, and longlasting anticancer efficacy in pediatric and adult patients with solid tumors harboring TRK fusions [49]. Drilon and colleagues conducted a phase 1 trial in adults (NCT02122913), a phase 1/2 trial in children (NCT02637687), and a phase 2 trial in children and adults (NCT02576431) with locally advanced or metastatic solid tumors who had received previous standard systemic therapy and were then given larotrectinib (100 mg twice daily), see Table 1 [49]. A total of 55 patients with 17 TRK fusion-positive tumor types included two patients (4%) with CCA [49]. The objective response rate (ORR) was high, at 80% (95% CI, 67-90) [49]. At one year, 55% of patients were still progression-free [49]. The median duration of response (MDR) and the progression-free survival (PFS) remain unmet [49]. The most common toxicities ≥ grade 3 included a raised ALT or AST level (9%), anemia (3.6%), reduced neutrophil count (3.6%), and nausea (3.6%) [49]. Based on these findings, larotrectinib was granted accelerated approval by the FDA in November 2018 for adult and pediatric patients with NTRK-positive solid malignant tumors, either metastatic or where surgical resection is unfeasible due to severe morbidity, who have progressed on systematic therapy [39]. Larotrectinib is also approved for patients with no satisfactory alternative treatments [39]. Currently, the MD Anderson Cancer Center and the National Cancer Institute (NCI) are conducting a phase 2 trial to investigate the efficacy of larotrectinib in previously treated patients with locally advanced or metastatic solid tumors and NTRK gene amplification (NCT04879121). Another ongoing phase 2 trial aims to assess the efficacy of larotrectinib in pediatric patients with relapsed or refractory advanced solid tumors with NTRK gene fusion (NCT03213704), Table 3.
The approval of entrectinib provides an additional treatment option for patients with advanced cancer, potentially creating the opportunity for patients and their physicians to choose between different therapies [65]. Larotrectinib is available in a liquid formulation approved for children younger than 12 years old, whereas entrectinib is not. However, entrectinib may be effective for children with brain tumors, while the efficacy of larotrectinib for primary and metastatic brain tumors is currently being evaluated [65]. The two drugs also have different side effect profiles. The adverse events seen most frequently in the larotrectinib trials include increased ALT or AST, fatigue, and vomiting. Warnings and precautions for larotrectinib include neurotoxicity, hepatotoxicity, and embryo-fetal toxicity. Entrectinib has additional warnings and precautions, including congestive heart failure, CNS effects, and skeletal fractures [65]. The National Comprehensive Cancer Network (NCCN) guidelines recommend larotrectinib and entrectinib as first-line or subsequent-line (following disease progression) treatment options for unresectable or metastatic iCCA and eCCA with NTRK gene fusions. Both entrectinib and larotrectinib are approved in the United States and Europe for the treatment of unresectable or metastatic solid tumors with NTRK gene fusion and progression after previous therapy [66,67]. However, there are currently limited data for patients with CCA.

Cholangiocarcinoma (CCA) with BRAF V600E Mutations
Mitogen-activated protein kinase (MAPK) signaling is essential for cell growth and survival through the RAS/RAF/MEK/ERK pathway [68]. The BRAF gene is an oncogene whose protein product upregulates the RAS/RAR/MEK pathway [68]. BRAF mutations have been identified in several solid malignancies, including colorectal cancer, NSCLC, and melanoma [69,70]. More than 50 different BRAF mutations have been reported, with the V600E point mutation being the most common mutation (BRAF V600E ) [71]. In BRAF V600E , valine (V) is substituted by glutamic acid (E) at amino acid 600, resulting in activating BRAF with subsequent tumor growth and spread [72]. In CCA, BRAF mutations are uncommon, occurring almost exclusively in iCCAs, with a prevalence ranging between 5% and 7%, and with the BRAF V600E mutation in 1.5% of patients with iCCA [26,73].

Other BRAF Inhibitors Currently Undergoing Clinical Trials
In a phase 1 study, patients with BRAF V600 -mutated solid tumors (including CCA) are currently being evaluated for a response to ABM-1310, a selective inhibitor of BRAF V600E mutation tumors (NCT04190628). Patients with advanced solid tumors, including biliary tract cancers, with BRAF mutations, are the focus of a phase 1 study of BGB-3245, a secondgeneration BRAF inhibitor (NCT04249843), Table 3. Combining the selective ERK1/2 inhibitor JSI-1187 with a BRAF inhibitor is another potential study strategy.
Despite the significant advances in managing patients with BRAF V600 mutations, further studies are required. For example, in studying the efficacy of dabrafenib and trametinib and concurrent mutations of TP53 and BRAF V600E , early studies reported that this was associated with a more aggressive disease, resulting in less clinical benefits from dabrafenib and trametinib [78]. In addition, patients with BRAF V600E /TP53 mutations were associated with reduced PFS and OS [79].

Cholangiocarcinoma (CCA) with Fibroblast Growth Factor Receptor 2 (FGFR2) Gene Fusion or Rearrangement
Alterations in the FGFR gene and dysregulated FGFR signaling play a role in the development and progression of several types of cancer, including CCA. Four receptors belong to the FGFR family, including FGFR1, 2, 3, and 4, which share a cytoplasmic tyrosine kinase domain [80]. FGFR2 alterations include rearrangements, amplifications, and mutations, present in 10-16% of patients with iCCA [33]. These alterations activate mitogen-activated protein kinases (MAPKs), triggering constitutive signaling cascades that prompt tumor cell proliferation, survival, migration, and angiogenesis [81,82]. Therefore, FGFR inhibitors are promising targeted therapies that can potentially improve the survival of patients with CCA. Initially, non-selective tyrosine kinase inhibitors (TKIs) were investigated in phase 1/2 clinical trials and showed low antitumor activity and a limited survival benefit [83][84][85]. More recently, selective FGFR inhibitors were introduced for patients with FGFR2 fusionpositive iCCA and resulted in a significant clinical response, prompting phase 2/3 trials and accelerated FDA approvals [18].
Infigratinib is another highly selective ATP-competitive FGFR1-3 inhibitor that showed promising antitumor activity in patients with FGFR2 fusions in early-phase trials [87]. Patients with advanced iCCA resistant to chemotherapy were enrolled in a phase 2 study of infigratinib in a multicenter, open-label, phase 2 trial [53]. Of the 122 enrolled patients, 108 had positive FGFR2 fusions or FGFR2 rearrangements [53]. Patients received 25 mg once daily infigratinib for 21 days in a 28-day cycle [53]. In patients with FGFR2 fusions or FGFR2 rearrangements, the ORR was 23.1% (95% CI, 15.6-32.3%), indicating clinically significant activity after treatment [53]. In this trial, one patient had a complete response (CR) and 24 patients had partial responses [53]. The median duration of response (DOR) was five months (95% CI, 3.7-9.3), and eight patients had a maintained response for more than six months [53]. Two-thirds of the patients (66.2%) had grade ≥ 3 toxicities, with hypophosphatemia (14.1%) and hyperphosphatemia (12.7%) being the most common adverse events, Table 2 [53]. A phase 1 dose-escalation study showed that the risk of hyperphosphatemia in patients receiving infigratinib increased with higher drug exposure and was associated with higher antitumor activity [88]. On May 2021, the FDA granted accelerated approval to infigratinib for previously treated locally advanced CCA or for patients with metastatic CCA with FGFR2 fusions or FGFR2 rearrangements [19].
Infigratinib is also a potential targeted therapy for patients with untreated CCA with FGFR2 fusions or FGFR2 rearrangements. PROOF-301 is an ongoing phase 3 trial that recruited patients with untreated locally advanced CCA or patients with metastatic CCA with FGFR2 fusions or FGFR2 rearrangements [89]. In PROOF-301, patients received either infigratinib or a standard chemotherapy regimen of gemcitabine and cisplatin [89]. Infigratinib may potentiate the apoptotic activities of chemotherapies in multidrug-resistant tumor cells [90]. Therefore, there is a potential future role for infigratinib in combination therapy for patients with advanced CCA, and the results of further clinical trials are awaited with interest.
Futibatinib is an irreversible FGFR1-4 inhibitor that binds to the conserved cysteine residues of the P-loop of the kinase domain [91]. Previous studies showed that futibatinib exhibited highly selective antitumor activities against tumor cells harboring FGFR mutations, particularly against mutations commonly associated with resistance to ATPdependent FGFR inhibitors [91]. In addition, the oral futibatinib molecule was associated with a comparatively lower number of drug-resistant clones than ATP-dependent FGFR inhibitors [91,92]. Futibatinib was studied in a phase I trial that recruited 197 patients with previously-treated advanced solid tumors, 83 of which had CCA with a mutation, fusion, or amplification of FGFR2 (NCT02052778) [92]. In this trial, 64 patients were treated with futibatinib at 20 mg and 19 patients at 16 mg [92]. The results showed that futibatinib at 20 mg daily resulted in an ORR of 15.6%, a disease control rate (CDR) of 71.9%, a median DOR of 5.3 months, and a median PFS of 5.1 months [92]. The updated analysis of the singlearm FOENIX-CCA2 phase 2 trial included 103 patients with previously-treated advanced or metastatic iCCA with FGFR2 fusions or FGFR2 rearrangements (NCT02052778) [54]. Futibatinib 20 mg daily led to an ORR of 41.7% at a median follow-up period of 25.0 months [54]. The DC was 82.5%, the median PFS was 8.9 months, and the median OS was 20.0 months, Table 2 [54]. Based on these findings, on September 30, 2022, the FDA granted accelerated approval for futibatinib for adult patients with previously treated locally advanced or metastatic iCCA with FGFR2 gene fusion or rearrangement [20].
The phase 1/2 ReFocus trial evaluated RLY-4008, a selective FGFR2 inhibitor that can target FGFR resistance mutations, in CCA patients with FGFR2 fusions or FGFR2 rearrangements who did not receive FGFR inhibitors before. The preliminary analysis of 38 patients showed an ORR of 63.2% (95% CI 46.0-78.2) and a DCR of 94.7%. There was no observation of grade 4/5 adverse events [94].
Despite the promising findings of selective FGFR inhibitors in patients with CCA and FGFR2 fusions or FGFR2 rearrangements, several issues remained unanswered. More than 150 fusion partners are associated with FGFR2 gene rearrangement, which results in significant molecular diversity in patients with FGFR2 fusions and rearrangements [95,96]. Furthermore, nearly 50% of the gene fusion and rearrangement partners are present within the same chromosome of the FGFR2 gene [95,96].
It is still unclear which patients might respond to FGFR2-targeted therapies and whether fusion partners can affect the response and survival with FGFR2-targeted therapies [96]. Therefore, future research should focus on the effect of combined genetic alterations on the responses to FGFR2 inhibitors and their role in the development of acquired resistance, as well as identifying reliable response biomarkers for FGFR2-targeted therapies [96,97].

High Tumor Mutational Burden (TMB-H) as a Predictive and Prognostic Biomarker
TMB is a recently identified biomarker of the response to immune checkpoint inhibitors (ICIs) in several types of cancer [97]. The TMB is the number of somatic mutations per megabase (Mb) of the genomic sequence of a tumor [97]. A TMB score of ≥10 mutations/Mb has been proposed as a threshold with a high likelihood of neoantigen formation and represents TMB-H status [97]. In patients with several tumor types, including melanoma, NSCLC, and bladder cancer, patients with TMB-H had better outcomes when treated with programmed death protein-1 and programmed cell death ligand-1 (PD-1/PD-L1) checkpoint inhibitors, or a cytotoxic T lymphocyte antigen 4 (CTLA-4) blockade [98][99][100]. TMB-H has been detected in 27.3% of patients with iCCA [101].
Pembrolizumab is a humanized antibody that inhibits the PD-1 receptors on lymphocytes by inhibiting the ligands that would block the receptor and inhibit an immune response [102][103][104]. In a subgroup analysis of 102 patients with TMB-H, who were enrolled in the phase 2 KEYNOTE-158 trial and received pembrolizumab, the ORR was 29% (95% CI, 21-39%), the median OS was 11.7 months (95% CI, 9.1-19.1), and the median PFS was 2.1 months (95% CI, 2.1-4.1) [105]. Notably, there were no patients with CCA in the TMB-H subgroup of this trial [105]. However, based on these findings, in 2017, the FDA approved pembrolizumab to treat adult and pediatric patients with unresectable or metastatic solid tumors (including CCA) [36]. The indications for this approval also required that the tumor tissue was TMB-H (≥10 mutations/megabase), and that the patients had progressed following prior therapy and for whom there were no satisfactory alternative treatment options [36].
However, it is unclear if the TMB levels for predicting the response to the PD-1 blockade are consistent throughout the spectrum of solid tumors. There are situations in which a high TMB does not indicate a response. Therefore, novel biomarkers are required that reflect the complexity of the tumor immune microenvironment and consider the effects of tumor mutations on the immune response.

High Microsatellite Instability and Mismatch Repair Deficient (MSI-H/dMMR) Cholangiocarcinoma (CCA)
The production of neoantigens and CD8+ T cell infiltrations into the tumor microenvironment are both increased in tumors with dMMR or high levels of MSI [106]. MSI-H/dMMR makes the errors produced during DNA replication difficult to repair, which leads to mutations [106]. The prevalence of MSI-H/dMMR in patients with iCCA ranges from 4.7-18.2% [35].
Dostarlimab is another anti-PD-1 monoclonal antibody [107]. The phase 1 GARNET study was a non-randomized, multicenter, open-label trial to evaluate the safety and efficacy of dostarlimab (500 mg every 3 weeks for four cycles, then 1000 mg every 6 weeks) in 106 patients with advanced solid tumors (two of whom had CCA) with confirmed MSI-H/dMMR, one of whom had a biliary tract cancer [57]. The ORR was 38.7% (95% CI, 29.4-48.6) [57]. With a median duration of follow-up of 12.4 months, the median DOR was not reached [57]. At 12 months, the Kaplan-Meier estimate for the chance of response preservation was 91%, and at 18 months it was 80% [57]. Approximately 8.3% of patients reported at least one grade 3 adverse event, including anemia (3.9%), elevated lipase (2.3%), elevated ALT (1.6%), and diarrhea (1.6%), Table 2 [57]. Based on the findings from this study, the FDA approved dostarlimab in adult patients with MSI-H/dMMR recurrent or advanced solid tumors that progressed on or following prior treatment and were not candidates for satisfactory alternative treatment options [38].
Given the limited single-agent activity and the limited number of patients with CCA included in these clinical trials, further studies are required to investigate novel immunotherapy combinations to enhance the treatment efficacy in patients with CCA.

Isocitrate Dehydrogenase Isoenzyme (IDH1) Gene Mutations in Cholangiocarcinoma (CCA)
IDH1 gene mutations commonly occur in CCA [108]. Missense mutations in the IDH1 R132 codon leads to the overproduction of the oncometabolite R-2-hydroxyglutarate (R-2HG) [109]. In tumor progenitor cells, an increase in the R-2HG levels inhibits cellular differentiation and drives oncogenesis by promoting histone methylation and DNA methylation [110]. The prevalence of an IDH1 mutation in iCCA has been estimated at 13% [34].
PARP inhibitors are also being studied in this subgroup of patients as they exhibit dysregulated homologous recombination repair. The National Cancer Institute (NCI) is conducting a phase 2 clinical trial that aims to investigate the safety and efficacy of olaparib as a subsequent line therapy for patients with advanced solid tumors (including CCA) with IDH1 or IDH2 mutations (NCT03212274). Another ongoing study combined ceralasertib and olaparib in patients with refractory CCA and advanced solid tumors with IDH1/2 (NCT03878095), Table 3.
The use of IDH1 inhibitors as single treatments has shown promising results in targeted therapy of malignancies harboring IDH1 mutations in preclinical and clinical settings.
However, these studies are preliminary and currently include small numbers of patients with CCA.
Several ongoing studies aim to investigate targeted HER2 agents in treating patients with CCA with ERBB2 mutations (Table 3). In the front-line setting, gemcitabine combined with cisplatin and trastuzumab and the combination of gemcitabine, cisplatin, and varlitinib are currently being studied in two clinical trials (NCT03613168 and NCT02992340). In addition, HER2-targeting agents are currently being studied for subsequent lines of therapy, either as monotherapy or in combination with standard chemotherapy agents. Ongoing trials include: the oral HER2 covalent inhibitor, TAS0728, (NCT03410927); the HER2 antibody-drug conjugate, trastuzumab deruxtecan (NCT04482309 and JMA-IIA00423); the antibody-drug conjugate, RC48-ADC, (NCT04329429); capecitabine plus the dual HER2 and EGFR inhibitor, varlitinib (NCT03093870); chemotherapy (5-FU or IRI or Cape) plus trastuzumab (NCT03185988); and chemotherapy plus the HER2-targeted bispecific antibody, zanidatamab (NCT02892123).
The results of these ongoing trials may provide multiple options for targeted treatments for patients with CCA in the molecular subgroup and improve patient survival.

RET Gene Fusion-Positive Cholangiocarcinoma
More than three decades have passed since the discovery of the gene encoding the receptor tyrosine kinase, RET [117]. RET rearrangements and mutations are recognized as treatable drivers of oncogenesis [117]. Certain RET fusion proteins and activating point mutations can drive oncogenesis and tumor progression by activating downstream signaling pathways, leading to uncontrolled cell proliferation [117]. RET gene fusions are present in between 1% and 2% of NSCLC and thyroid cancers and represent potential targets for therapeutic inhibition of RET kinase [118]. However, RET fusions seem rare in CCA, and data regarding their exact prevalence in CCA are limited [119].
Selpercatinib is another highly selective inhibitor of the RET receptor tyrosine kinase, with CNS activity [121]. In the phase 1/2, open-label LIBRETTO-001 trial, 45 patients with RET fusion-positive solid tumors other than lung or thyroid tumors were included, with two patients with CCA [61]. Of the 45 patients, 43 received a starting recommended dose of 160 mg BID. In the 41 patients who were evaluated for efficacy, the ORR was 43.9% (95% CI, 28.5-60.3), the median OS was 18.0 months (95% CI, 10.7-NE), the median PFS was 13.2 months (95% CI, 7.4-26.2), and the median duration of response was 24.5 months (95% CI: 9.2-NE). One patient with CCA was evaluated for efficacy; the ORR was 100% and the duration of response was 5.6 months [61]. Altogether, 49% of patients had grade ≥ 3 adverse events that included hypertension (22%), increased ALT (16%), and increased AST (13%), Table 2 [61]. Selpercatinib is currently approved for locally advanced or metastatic RET-fusion-positive solid tumors [122].
Again, further studies are needed to focus on CCA to validate these findings in this group of patients.

Liquid Biopsy for Assessment of Circulating Tumor DNA (ctDNA) and Cholangiocarcinoma (CCA)
Liquid biopsies refer to blood sampling to detect circulating tumor DNA (ctDNA), circulating cell-free RNA (ccfRNA), and cell-free DNA (cfDNA) [123,124]. Liquid biopsy as an adjunctive diagnostic method has gained popularity over the last decade due to its potential benefits for cancer patients [123]. The most commonly interrogated element in liquid biopsies is ctDNA or DNA fragments produced from tumors [125]. Treatment resistance tracking, response monitoring, target selection, relapse detection, and early diagnosis are the potential benefits of identifying tumor-derived material in liquid biopsies [126].
Although liquid biopsy is a potentially useful diagnostic tool for patients with CCA, this modality remains underexplored in CCA. Advances in this diagnostic area for patients with CCA have been limited by several practical factors, including the small amounts of ctDNA shed into the bloodstream in patients with localized tumors [127]. On the other hand, CCA is an internal malignancy and it is often difficult to obtain a tissue biopsy. It is also challenging to acquire sufficient tumor cells for diagnosis on aspiration cytology, and these challenges can prevent adequate molecular tumor profiling for CCA [128]. Therefore, blood-derived ctDNA may play an important role in identifying molecular alterations in patients with CCA who may not have an adequate biopsy or cytology sample available for analysis [129]. Consistent mutation findings between tumor tissue and ctDNA were reported in 2015 by Zill et al. in a prospective study of 26 pancreaticobiliary cancers, including eight patients with CCA [130]. In this preliminary study, 93% of mutations found in tissue samples were also identified by cfDNA [130]. In 2019, Mody et al. analyzed ctDNA from 138 patients with biliary tract cancers and identified genetic alterations in 89% of the samples [131].
Recently, Kumari et al. evaluated the diagnostic role of cfDNA in gallbladder carcinoma [132]. Serum was obtained from 34 patients with gallbladder carcinoma and 39 matched controls without malignancy [132]. This study showed that the cfDNA levels were lower in healthy individuals than in patients with gallbladder cancer [132]. In addition, there was a significant correlation between the cfDNA and the presence of jaundice, lymph node metastases, and overall disease TNM stage [132]. Therefore, the quantitative analysis of cfDNA may have the potential to be a unique marker for the molecular detection of targeted therapies in CCA. Furthermore, the analysis of cfDNA may have a role in distinguishing between neoplastic and inflammatory conditions of the gallbladder and biliary tract identified by imaging but without available biopsy material. However, concerns remain about the overall sensitivity of ctDNA mutations for diagnosing early-stage CCA [133].
An additional feature of ctDNA/cfDNA is to track the development of resistance to chemotherapy and targeted treatments [134]. In 2019, Ettrich and colleagues sequenced 15 common gene mutations in ctDNA samples from patients with biliary tract cancer throughout their chemotherapy [135]. In the iCCA cohort (n = 13), there was a 92% agreement between tissue samples and blood-derived ctDNA [135]. The level of agreement for the overall cohort was 74% [135]. A change in the mutational profile was also seen in 63% of chemotherapy-naive individuals after treatment [135]. A study reported in 2017 showed that the integrative genomic analysis of cfDNA could identify acquired treatment resistance due to multiple recurrent point mutations in the FGFR2 kinase domain during tumor progression [136].

Conclusions
Until recently, clinical studies considered CCA a homogeneous entity, which may have led to the limited antitumor activity of conventional treatment regimens. It is now apparent that the behavior and responses of CCA vary substantially according to the underlying molecular profile. These relatively recent findings highlight the crucial role of precision medicine in guiding the treatment selection for patients with CCA.
The number of investigational and approved targeted therapies for CCA has increased exponentially in the past decade. Several targeted agents are now approved as first-line and subsequent treatments for patients with locally advanced or metastatic CCA. The selective FGFR inhibitors pemigatinib and infigratinib and the IDH1 inhibitor ivosidenib are now approved for previously treated patients with FGFR2 fusions or FGFR2 rearrangements and IDH1 mutations, respectively. These gene fusions, rearrangements, and mutations are present in a subset of patients with iCCA. Pemigatinib and infigratinib are also currently being investigated in the first-line setting. Other targeted therapies are available across solid tumors, including CCA, and include: NTRK inhibitors (entrectinib and larotrectinib); a BRAF/MEK inhibitor combination (dabrafenib and trametinib); pembrolizumab (for high ≥ 10 mutations/mb tumor mutational burden or mismatch repair defect cancers); and RET inhibitors (selpercatinib). Targeted therapies for other genomic alterations in CCA and solid tumors in general continue to be developed and explored.
Further therapeutic possibilities for patients with CCA may emerge as our knowledge of the tumor microenvironment and its impact on tumor growth increases. Combined treatments that target both actionable mutations and the tumor microenvironment is another approach that assists with patient selection for the most appropriate molecular targeted therapy.
The value of blood-derived ctDNA in the clinic is currently being explored in CCA, especially since tissue biopsies can be difficult in this type of cancer.

Future Directions
Despite multiple FDA approved targeted therapies for patients with CCA, the overall survival for these patients remains dismal. Utilizing combination approaches with targeted therapies may offer some patients more benefit than single agents [137][138][139]. Mechanisms of resistance to single agent therapies may include RNA silencing [140] as well as multiple gene driven pathogenesis, which could be overcome with 'multiomic' patient diagnostics using genomic, proteomic, transcriptomic, and immunomic data as well as n-of-1 cus-tomized combination therapeutic approaches [141][142][143][144]. Further, using patient-specific immunomic data may allow for a tailored immunotherapy-targeted treatment as well as being informed by specific genomic aberrations and the most logical immunotherapeutic target [145]. A next step in improving outcomes in these patients likely involves identifying more therapeutic targets as well as overcoming secondary resistance mechanisms via targeting as many genomic alterations present within a specific tumor as possible [144,146,147].