The FGFR pathway is involved in many physiological cellular processes and cancer cells’ proliferation, migration, and survival [10]. The FGFR family consists of four transmembrane receptors, FGFR1-4. Single nucleotide variants, gene fusions, and copy number amplifications were found by NGS analysis, reporting aberrant activation of this signal in 5–10% of human cancers [11]. This percentage increases up to an average of 15% in BTC, mostly in iCCA [12]. Fusions or rearrangements of FGFR2 are the most frequent alteration, which is mutually exclusive with other well-known mutations [KRAS, v-RAF murine sarcoma viral oncogene homologue B1 (BRAF), or IDH1] [13]. Given the poor prognosis of BTCs, targeting this pathway is an attractive therapeutic goal.

Pemigatinib is the first Food and Drug Administration (FDA)-approved drug in advanced previously treated FGFR-altered BTC. Pemigatinib is a selective, reversible, oral inhibitor of FGFR1, 2, and 3. In the phase II Fight-202 study, 35% of patients achieved an objective response (95% CI: 26.5–45.4), with no safety warning or treatment-related deaths [14]. Infigratinib (an irreversible pan-FGFR inhibitor) was recently approved by FDA for previously treated advanced BTC harboring an FGFR2 fusion or rearrangement. In a phase II study, 108 patients were enrolled (77% with FGFR2 fusions), achieving an objective response rate (ORR) of 23.1% in the intention-to-treat population, 34% in the second-line, and 13.8% in later lines [15]. In LUC2001, the selective inhibitor of pan-FGFR erdafitinib was tested on 14 Asian patients with previously treated FGFR altered CCA. Erdafitinib achieved an ORR of 50% and a median duration of response (DoR) of 6.83 months [60% and 100% of ORR and disease control rate (DCR) respectively in FGFR2⁺ restricted patients] [16]. Derazantinib, a multi-kinase inhibitor with potent pan-FGFR activity, was tested on 29 iCCAs with FGFR2 fusions, reporting an ORR of 20.7% and a DCR of 82.8% [17]. The multicenter ongoing phase II FOENIX-CCA2 study is testing futibatinib (an irreversible inhibitor) in BTC with any alteration of the FGFR family. The planned accrual is over 800 patients. In the interim analysis, an ORR of 37.3%, a DCR of 82.1%, and a median DoR of 8.3 months were reported [18].

According to our search, 16 ongoing trials (25%) are investigating FGFR-inhibitors in BTCs (as described in Table 1). Three (18.7%) are open-label randomized multicenter phase III trials, 8 (50%) are single-arm phase II trials and 4 (25%) are phases I studies. A further study, testing the association of anti-FGFR and anti-IDH treatment, will be discussed in the section dedicated to IDH (NCT04088188).

Three are phase III trials testing FGFR-inhibitors versus Gem-Cis doublet in the first line. These trials involve untreated FGFR2 rearranged BTC, with PFS as the primary endpoint. FIGHT-302 trial is an active-controlled study that enrolls patients with BTCs in Europe and the United States, testing the efficacy of pemigatinib. Accrual is expected to be completed in 2026 (NCT03656536). FOENIX-CCA3 is a trial that aims to evaluate futibatinib only in patients with iCCA. Results are awaited in 2022 (NCT04093362). PROOF 301 trial randomizes patients with advanced CCA with FGFR2 gene fusions/translocations to receive infigratinib versus standard therapy. Data are expected in 2024 (NCT03773302). The results of these trials could potentially define a new standard targeted therapy in the front-line setting for molecularly selected advanced BTC.

Furthermore, there is a huge number of phase II studies with a more complex landscape. All these studies enroll patients after progression to first-line Gem-platinum-based chemotherapy, have a single-arm design and ORR is the primary endpoint. A phase II study with pemigatinib is ongoing in China, in BTC characterized by FGFR2 rearrangement, to evaluate its efficacy in the Eastern population (NCT04256980). Erdafitinib and derazantinib are investigated also in phase II trials with a different design from the studies mentioned above. Notably in erdafitinib study, BTC is included if FGFR fusions or mutations are present (NCT04083976). FIDES-01 with derazantinib enrolls the only iCCA with FGFR amplification, mutations, and fusions (NCT03230318). The basket trial Foenix 101 consists of three parts, including a phase II to evaluate infigratinib in iCCA harboring FGFR2 gene fusions. Interestingly, in one cohort, infigratinib has been tested in patients previously treated with FGFR inhibitor (NCT04233567). Finally, 3 further phase II studies, similar in design and endpoint, are ongoing. The drugs tested are HMPL-453, E7090, and ICP-192. The first enroll only iCCA (NCT04353375; NCT04238715; NCT04565275). Another phase II trial with famitinib has been withdrawn for slow accrual (NCT04479904).

Four studies are phase I basket trials enrolling previously treated BTC with safety as the primary endpoint (NCT03144661; NCT04149691; NCT03583125; NCT04526106).

In our opinion, although FGFR alterations affect only a small percentage of iCCA, targeting this pathway could be a valuable option. In particular, data from the FIGHT-302 trial are strongly awaited because, if positive, we could spare our patients systemic chemotherapy in favor of a tailored and less toxic treatment. In addition, we reckon that an effort should be made to early detect the mechanisms of resistance to FGFR inhibitors as a proactive research strategy. Likewise, in a phase II trial of infigratinib, a cohort of FGFR-treated patients was included.

The IDH proteins are critical metabolic enzymes, members of the oxidoreductase family. Cancer-associated mutations have been identified in two of the three existing isoforms: IDH1 and IDH2, which are mutually exclusive. They occur in approximately 10–12% of iCCA, while their incidence in eCCA and GBC is rare [19]. Almost all IDH mutations are missense, confined to a single residue (R132) for IDH1, in the active site of the enzyme [20]. These mutations play a role in the pathogenesis of CCA by blocking hepatocyte differentiation and promoting cellular proliferation [21].

Most IDH inhibitors target isoform 1. The first-in-class IDH1 inhibitor is ivosidenib (AG-120), an oral small molecule, already approved by FDA for patients with IDH-mutant acute myeloid leukemia. In the multicenter, randomized, double-blind, placebo-controlled phase III ClarIDHy study, ivosidenib significantly improved PFS (2.7 vs. 1.4 months, HR = 0.37; 95% CI: 0.25–0.54; one-sided P < 0.0001) in IDH1-mutant, chemo-refractory patients [22]. Researchers presented final OS data at the 2021 Gastrointestinal Cancers Symposium, showing a difference that becomes statistically significant after adjusting for crossover (HR = 0.49, P < 0.0001) [23].

We found nine (14.1%) clinical trials with IDH 1/2 targeting therapy (see Table 2). Ivosidenib, in combination with Cis and Gem, is currently investigated in a multicenter phase I trial (NCT04088188) in treatment-naive patients. The primary objective is to evaluate safety, tolerability, MTD, and RP2D.

Three other promising IDH1 inhibitors, able to cross the blood-brain barrier, are currently under investigation in patients with IDH1 mutant advanced malignancies, including CCA. LY3410738 is evaluated in a multicenter, phase I trial (NCT04521686) actively recruiting patients (estimated enrollment 180), to define the RP2D. Other two trials have already completed the accrual: a phase I trial (NCT02381886), that enrolled 166 patients to estimate the MTD and RP2D of IDH305; a phase I/II trial (NCT03684811), whose primary objective is to evaluate DLT, dose recommended for future studies and ORR of olutasidenib (FT-2102). The latter study is divided into two parts, to test olutasidenib single agent and the combination with nivolumab (hepatobiliary tumors) and the combination with Cis plus Gem (iCCA).

The development of selective IDH2 inhibitors has given less favorable results. Enasidenib (AG-221) has been investigated in a phase I trial (NCT02273739), consisting of a dose-escalation phase to determine MTD followed by expansion cohorts to further evaluate safety and tolerability. However, enrollment was closed at the end of the dose-escalation phase, and the expansion phase was not conducted.

Although the ClarIDHy study was the first positive trial of targeted therapy in BTC, the clinical relevance of the results obtained with this drug is still questioned and the approval by regulatory authorities is still pending. We think that clinical research with IDH-inhibitors should not focus on monotherapy in chemo-refractory disease, but in our opinion, a more challenging setting will be their early use in combination with chemotherapy or other targeted therapies. As an example, preclinical studies showed that IDH mutations could change cell metabolism with a predominance of oxidative metabolism [24]. A phase I/II trial (NCT02496741) is testing the combination of the oral antidiabetic metformin and the oral antimalarial drug chloroquine, to specifically inhibit these metabolic processes in patients with IDH1/2 mutated iCCA. The primary endpoint is the identification of MTD of both drugs, and the trial completed the accrual, enrolling 15 participants.

DDR pathway is related to genomic instability and cancer susceptibility. Homologous recombination (HR), mismatch repair (MMR), base excision repair (BER), and other mechanisms have been associated with CCA, but correlation to actionable target and precision medicine remains still unclear. The presence of DDR alteration could also confer a different sensitivity to platinum-based chemotherapy [25, 26]. Other recent research showed elevated DNA damage in IDH-mutant CCA, and treatment strategies using a synthetic lethality approach with PARP inhibitors are in development [27]. Three phase II trials adopting this strategy are currently recruiting patients: these studies are testing olaparib alone (NCT03212274) or in combination with ceralasertib (AZD6738-NCT03878095) or durvalumab (NCT03991832) in both IDH1 and IDH2 mutant recurrent CCA. Interest in PARP inhibitors is growing, and three phase II trials (4.5%) are testing their use in patients with molecular alterations other than IDH1/2. The NCT03207347 trial is evaluating the activity of niraparib in tumors with BRCA1 associated protein-1 (BAP1) and other selected DNA damage response pathway genes. The combination of olaparib and pembrolizumab is investigated in patients who have progressed to the first-line therapy in the trial NCT04306367. We also report a phase I trial (NCT01282333) testing the combination of veliparib with Cis and Gem in treatment-naive, advanced eCCA with known or suspected BRCA1/2 germline mutation. The trial is terminated, but the results are still unpublished.

The epidermal growth factor receptor (EGFR) kinase family consists of four receptors [ErbB1/EGFR, ErbB2/human epidermal growth factor receptor (HER) 2, ErbB3, and ErbB4] that play a crucial role in cell proliferation, differentiation, and motility and may lead to cancer development. The HER2 overexpression, gene amplification, or mutation have been described in breast, gastric, lung, and colorectal cancer. HER2 is a predictive biomarker for anti-HER2 target therapies and could be tested by IHC, FISH, and NGS.

A systematic review and meta-analysis showed a HER2 expression rate of 26.5% in BTCs, with different detection rates according to the tumor site and the testing procedures [28]. Considering the studies in which HER overexpression was defined by moderate/strong IHC expression, HER2 positive rate was 20% in eCCC and 5% in iCCA. In a cohort of 1,863 patients evaluated by NGS, Jacobi et al. [29] reported HER2 alteration in 10% of eCCA and 4% of iCCA. Despite the not negligible number of HER2 molecular alterations in BTCs, only a few clinical prospective data are available in this setting.

In a pilot study, 4 patients with HER2-positive BTCs were treated with trastuzumab plus Cis and Gem. Two patients experienced a partial response (PR, 50%), and 2 obtained stable disease (SD, 50%), with a median PFS of 6.1 months. Despite the small sample size, this pilot study showed encouraging data regarding HER2-target therapy plus standard chemotherapy, without safety warning [30].

The MyPathway trial, a multi-basket phase IIA trial, investigated pertuzumab + trastuzumab in 11 patients with HER2-positive refractory advanced BTCs. Among the 7 patients with HER2 overexpression or amplification, 2 achieved PR (29 %) and 3 SD for more than 4 months (38%) [31]. According to these promising results, pertuzumab plus trastuzumab could be further explored in HER-positive BTCs.

Promising results come from other new drugs, such as the antibody-drug conjugate trastuzumab-deruxtecan [32], the anti-HER2 antibodies mergetuximab [33], and zanidatamab, the novel antibody that simultaneously binds 2 distinct epitopes of HER2 [34].

Tailored therapy could be a novel therapeutic strategy for BTCs harboring HER2 molecular alterations, but further prospective data are needed to better understand the right workflow for HER2 testing and treating.

We found 9 (14.1%) clinical trials with HER2 targeting therapy (as shown in Table 3). HER2-positivity is a required inclusion criterion in all studies, defined by the presence of HER2 protein overexpression by IHC or by the presence of HER2 gene amplification documented with in situ hybridization (ISH) or FISH. NGS is also allowed for HER2 testing in 5 studies (55.6 %). HER2 mutation is an inclusion criterion in only one trial (11.1%). BTCs with low HER2 expression, defined by at least 1+ at IHC, could be enrolled in two trials (22.2%).

Seven (77.7%) are basket trials including different malignancies and only two trials (22.2%) are specifically designed for BTCs. Four (44.4%) studies are phase I trials investigating MTD determined according to DLT and incidence of AEs. Two (22.2%) are phase I/II trials evaluating both safety and efficacy. Three (33.3%) phase II trials have response rate (RR) as the primary endpoint and other outcomes of efficacy and tolerability as secondary endpoints. All studies are non-randomized, mostly have single group assignments (66.6 %) and two (22.2%) have sequential assignments. Eight trials (88.8%) are multicentre and are available mostly in United States (88.8%). Only three (33.3%) are not sponsored by a pharmaceutical company.

Advanced BTCs, without subtypes restriction, are included in the majority of these clinical trials: restriction due to the tumor intrahepatic or extrahepatic origin is taken into account only in one study (11.1%). One study (11.1%, NCT03613168) is investigating a first-line combination (trastuzumab plus Cis-Gem). Another study (11.1%) include patients with both treatment-naive and previously treated BTCs to evaluate the combination of trastuzumab plus the farnesyltransferase inhibitor tipirfanib (NCT00005842). Patients with BTCs in further lines of therapy (≥ 2) are included in the remaining trials (77.7 %). Trastuzumab is the main targeted agent used in further line in three trials (42.8 %): plus tucatinib and oxaliplatin-based chemotherapy (NCT04430738), plus IL12 (NCT00004074), or plus tucatinib monotherapy (NCT04579380). Anti-HER2 CAR in engineered macrophages is a potential therapeutic approach in one ongoing study (14.2 %, NCT04660929). Zanidatamab is currently investigated in patients with previously treated BTCs in one phase II trial (14.2%, NCT04466891). ADC targeting HER2 positive tumors are used as an experimental treatment in two (28.5%) clinical trials (NCT03821233, NCT03602079).

So far, anti-HER2 therapies have yielded very limited evidence but promising data of the activity. A major barrier is due to the rarity of this alteration, although is relatively more common in eCCA.

Several clinical trials are currently investigating the role of anti-HER2 target therapy in BTCs but, because of the difficulties we have just discussed, the most popular researching approach involves basket studies. Whenever an alteration is rare in a rare disease, innovative study designs could help the clinicians to match the proper drug to the appropriate patient. Most studies are in early phases, but this could not exclude a breakthrough approval in case of exceptional results.

MAPK cascades are implicated in the survival and proliferation of tumor cells; targeting these signaling cascades has proven to be effective in many types of tumor histology (for example melanoma, lung, and colon cancer) [35, 36]. Anti-EGFR monoclonal antibodies are not as effective in BTC as in other types of gastrointestinal cancer. Both cetuximab and panitumumab, even in a KRAS wild-type (wt) population did not demonstrate an OS advantage [6–8]. One study evaluated erlotinib plus cetuximab in KRAS wt selected patients but until now no data were published (NCT00397384). Conversely, the promising activity of targeting BRAF mutated BTC arise from the phase II ROAR study.

A phase I study with AMB13-10, a BRAF inhibitor, is ongoing in patients with documented BRAF V600 mutation (NCT04190628). No phase III studies are ongoing, probably due to the rarity of this mutation, which accounts for 5% of iCCA.

Finally, there is also an interesting phase II study with MAPK kinase (MEK) inhibitor (trametinib) associated with hydroxychloroquine (autophagy inhibitor) in patients with KRAS mutation (NCT04566133) (See Table 4).

Aberrant forms and expression of anaplastic lymphoma kinase (ALK), a receptor tyrosine kinase from the insulin receptor superfamily, are implicated in tumorigenesis. The receptor tyrosine kinase c-ros oncogene1 receptor tyrosine kinase (ROS1) belongs to the insulin receptor family and is an oncogenic driver in different malignancies.

A post-hoc analysis retrospectively evaluated the aberrant expression of ROS1, ALK, and mesenchymal to epithelial transition (MET) in 110 patients with advanced BTCs treated with Gem plus oxaliplatin (GEMOX) with or without cetuximab, using IHC. Eighteen patients (16.3%) had IHC score intensity 3+ for any markers ROS1, ALK, and c-MET (R.A.M. high), and 92 (83.6 %) had IHC score intensity lower than three for any markers (RAM low). All RAM high BTCs were iCCA (100%) with worse OS than RAM low iCCA (median 5.7 vs. 11.7 months, P = 0.021) [37].

Sporadic cases of BTCs with ALK rearrangement or alterations and ROS1 fusion are described in few case reports [38, 39] and preclinical studies [40–42].

ALK and ROS1 molecular alterations in BTCs are a neglected topic in literature, with scarce emerging data regarding incidence and targeted anticancer drugs. Riding the wave of successes of ALK and ROS1 targeted therapy in non-small cell lung cancer (NSCLC), few clinical trials are testing these treatments in ALK and ROS1 positive BTCs population.

Three non-randomized phases II trials (4.7%) have enrolled patients with ALK and ROS1 alterations in advanced BTCs, two are terminated and only one is actually ongoing. An international multicentre basket trial is evaluating the role of entrectinib in first or subsequent lines in patients with activated neurotrophic tyrosine receptor kinase (NTRK) 1/2/3, ALK, or ROS1 pathway in different malignancies (NCT02568267). The primary endpoint is ORR. Entrectinib has already been approved by FDA and European Medical Association (EMA) for the treatment of patients with solid tumors, included BTC, expressing an NTRK gene fusion.

There are other clinical studies that test targeted agents in different pathways not mentioned above, to to briefly describe them, Table 5 has been drawn up.

Multitarget drugs can inhibit several cell signaling pathways involved in cancer development, growth, and spreading. Plenty of multitarget inhibitors, with or without chemotherapy, had been examined in advanced BTCs in different studies with conflicting results. For example, five studies failed to show a survival benefit with sorafenib, a multikinase inhibitor targeting tyrosine kinase receptor involved in tumorigenesis [CD117 and REarranged during Transfection (RET)] and angiogenesis [VEGF receptor (VEGFR) 1, 2, 3 and platelet-derived growth factor receptor (PDGFR) β] [43–47]. In a single-arm phase II study, refractory BTCs received lenvatinib 24 mg orally daily resulting in an ORR of 11.5 % and median overall survival (mOS) of 7.35 months, with a relevant rate of grade ≥ 3 AEs (80 %) [48].

Angiogenesis creates a supportive microenvironment and promotes tumor growth and metastasis. The VEGF is expressed in 54 % and 59% of iCCA and eCCA respectively [49, 50]. Several clinical trials investigated bevacizumab in combination with chemotherapy or erlotinib in untreated and pre-treated BTCs, with modest results [51–53].

Forty-five clinical trials (15.2%) are investigating not oncogenic-driven multitarget therapy (Table 5). Lenvatinib, a multireceptor tyrosine kinase inhibitor, is currently been investigated in 9 monocentre phase II clinical trials (20.9 %) in China. In eight (88.8 %) clinical trials lenvatinib is used as a systemic treatment in first or further line for locally advanced BTCs, as monotherapy, or in association with different drugs (PD-1 monoclonal antibody or chemotherapy). The remaining 36 not oncogenic driven clinical trials are evaluating EGFR tyrosine-kinase inhibitors (TKI) and VEGF TKI.