Evolving therapeutic landscape of advanced biliary tract cancer: from chemotherapy to molecular targets

Abstract

Biliary tract cancer, the second most common type of liver cancer, remains a therapeutic challenge due to its late diagnosis and poor prognosis. In recent years, it has become evident that classical chemotherapy might not be the optimal treatment for patients with biliary tract cancer, especially after failure of first-line therapy. Finding new treatment options and strategies to improve the survival of these patients is therefore crucial. With the rise and increasing availability of genetic testing in patients with tumor, novel treatment approaches targeting specific genetic alterations have recently been proposed and have demonstrated their safety and efficacy in numerous clinical trials. In this review, we will first consider chemotherapy options and the new possibility of combining chemotherapy with immune checkpoint inhibitors in first-line treatment. We will then provide an overview of genomic alterations and their potential for targeted therapy especially in second-line therapy. In addition to the most common alterations such as isocitrate dehydrogenase 1 or 2 (IDH1/2) mutations, fibroblast growth factor receptor 2 (FGFR2) fusions, and alterations, we will also discuss less frequently encountered alterations such as BRAF V600E mutation and neurotrophic tyrosine kinase receptor gene (NTRK) fusion. We highlight the importance of molecular profiling in guiding therapeutic decisions and emphasize the need for continued research to optimize and expand targeted treatment strategies for this aggressive malignancy.

Highlights

• •Biliary tract cancer is often associated with genetic alterations and is an excellent model for precision oncology.
• •The most common targetable alterations are IDH1 and 2 mutations as well as FGFR2 fusions or rearrangements.
• •This review summarizes novel therapeutic options for targeted therapy in advanced biliary tract cancer.

Introduction

Biliary tract cancer (BTC) is the second most common type of liver cancer and accounts for ∼3% of all adult cancers worldwide.¹ Recently, an increase in the prevalence of BTC has been reported, mainly due to an increase in the prevalence of intrahepatic cholangiocarcinoma (iCCA).² BTC is an umbrella term that encompasses a heterogeneous group of tumors with different demographic characteristics, risk factors, localizations, and molecular signatures, and therefore often adapted treatment approaches.³ In daily practice, BTCs are classified according to the anatomical tumor localization within the hepatobiliary system. Among BTCs, gallbladder carcinoma (GBC) must be distinguished from CCA. The latter is further subdivided into iCCA and extrahepatic CCA (eCCA). eCCA itself can be further subclassified as perihilar eCCA (pCCA) and distal eCCA (dCCA).⁴ Among CCAs, iCCA accounts for 20%-30% of all CCAs and originates from the epithelium of intrahepatic bile ducts and segmental ductules.⁴ pCCA has its origin in the extrahepatic bile ducts located in the hepatic hilus above the outlet of the cystic duct and accounts for 50% of all CCAs.⁴ dCCA develops in the part of the extrahepatic bile duct that is below the junction with the cystic duct.⁴ Unfortunately, due to the absence of early symptoms, BTC is often diagnosed at advanced tumor stages that are associated with poor prognosis.⁵ For patients with advanced tumor stages, curative treatment by surgical resection is not feasible and systemic therapy with palliative intent is generally recommended.^6,7 In addition to patients with advanced disease at initial diagnosis, many patients experience tumor recurrence after resection and also require systemic therapy.^8,9 In recent years, it has become increasingly clear that a classification system only based of tumor localization alone is not sufficient, but that BTCs exhibit significant heterogeneity at the molecular level.³ This is particularly important with regard to possible individualized therapeutic options in advanced tumor stages.¹⁰ The aim of this narrative review is to provide an overview of current systemic treatment options with a focus on targeted therapy.

Systemic therapies in advanced tumor stages

For over a decade, a combination of gemcitabine and cisplatin (so-called GemCis regimen) has been the standard of care for first-line therapy in patients with advanced diseases and adequate performance status. This recommendation was based on the results of the ABC-02 study, which compared GemCis with gemcitabine as a monotherapy. The combination achieved a significantly higher disease control rate (DCR; 81.4% versus 71.8%; P = 0.049), a longer progression-free survival [PFS; 8.0 versus 5.0 months; hazard ratio (HR) 0.63, 95% confidence interval (CI) 0.51 to 0.77; P < 0.001] as well as a longer overall survival (OS; 11.7 versus 8.1 months; HR 0.64, 95% CI 0.52-0.80; P < 0.001).¹¹ However, in subsequent studies, addition of further classic cytostatic drugs failed to achieve an additional survival benefit. The PRODIGE 38 AMEBICA-trial, comparing the triplet chemotherapy regimen modified FOLFIRINOX (mFOLFIRINOX) consisting of fluorouracil, leucovorin, irinotecan, and oxaliplatin with the standard-of-care regimen GemCis, did not show improvement in 6-month PFS.¹² The addition of nab-paclitaxel to the GemCis scheme also failed to show a significant improvement in OS compared with GemCis alone (NCT03768414).

Strikingly, it has recently been shown that the addition of immunotherapy to the GemCis regimen may for the first time provide a notable survival benefit. The TOPAZ-1 trial, published by Oh and colleagues,¹³ compared the addition of the programmed cell death protein 1 (PD-1) inhibitor durvalumab to the established GemCis regimen with placebo. The combination of GemCis and durvalumab followed by a durvalumab maintenance after eight cycles achieved a significantly longer OS (12.8 versus 11.5 months; HR 0.80, 95% CI 0.66-0.97; P = 0.021) and PFS (7.2 versus 5.7 months; HR 0.75, 95% CI 0.63-0.89; P = 0.001) as well as an improved tumor response (26.7% versus 18.7%, odds ratio 1.60; 95% CI 1.11-2.31). With regard to OS, there was a particular benefit for long-term survival. After 24 months, 24.9% were still alive in the durvalumab group compared with 10.4% in the placebo group. The survival benefit was independent of the localization of the tumor as well as programmed death-ligand 1 (PD-L1) expression. Of note, no significant toxicities beyond those known from the established GemCis regimen could be observed.¹³ The fact that a survival benefit can also be achieved with the addition of other immune checkpoint inhibitors has been underscored by the KEYNOTE 966 trial. In this randomized controlled phase III trial, Kelley and colleagues¹⁴ observed that an addition of pembrolizumab to the established GemCis regimen likely provides a survival benefit compared with the addition of placebo (12.7 months versus 10.9 months; HR 0.83, 95% CI 0.72-0.95). Overall, the results were very similar in the aforementioned two studies. However, it should be noted that in the TOPAZ-1 trial, chemotherapy was stopped after eight cycles, whereas in the KEYNOTE 966 study, pembrolizumab was continued together with gemcitabine. Based on the results of the TOPAZ-1 trial, the combination GemCis plus durvalumab has been approved by both the Food and Drug Administration (FDA) and the European Medicines Agency (EMA) in 2022 and is now the recommended first-line treatment for patients with advanced BTC.^6,7 The aforementioned studies are summarized in Table 1.

Table 1.

Selected trials of chemotherapy in the first-line treatment for patients with advanced BTC

First-line chemotherapy ‘all comers’
Clinical trial, year, reference	Treatment	Study population	Primary endpoint	Secondary endpoint
ABC-02 (randomized phase III), 201016	Gem plus Cis versus Gem alone	410 chemotherapy-naive patients (1 : 1) with nonresectable, recurrent, or metastatic BTC	OS 11.7 months (GemCis) versus 8.1 months (Gem), HR 0.64 (95% CI 0.52-0.80), P < 0.001	PFS 8.0 months (GemCis) versus 5.0 months (Gem), HR 0.63 (95% CI 0.51-0.77), P < 0.001 DCR 81.4% (GemCis) versus 71.8% (Gem), P = 0.049 AEs were similar
BT22 (randomized phase II), 2010104	Gem plus Cis versus Gem alone	84 chemotherapy-naive patients (1 : 1) with nonresectable, recurrent or metastatic BTC	OS 11.2 months (GemCis) versus 7.7 months (Gem), HR 0.69 (95% CI 0.42-1.13), P = 0.139 1-year survival rate 39.0% (GemCis) versus 31.0% (Gem)	PFS 5.8 months (GemCis) versus 3.7 months (Gem), HR 0.66 (95% CI 0.41-1.05), P = 0.077 6-month PFS rate 47.4% (GemCis) versus 27.7% (Gem)
TOPAZ-1 (randomized phase III), 202213	GemCis plus durvalumab versus GemCis plus placebo	685 chemotherapy-naive patients (1 : 1) with nonresectable, metastatic, or recurrent BTC	OS 12.8 months (GemCis + durvalumab) versus 11.5 months (GemCis + placebo), HR 0.80 (95% CI 0.66-0.97), P = 0.021 1-year survival rate 54.1% (GemCis + durvalumab) versus 48% (GemCis + placebo) 2-year survival rate 24.9% (GemCis + durvalumab) versus 10.4% (GemCis + placebo)	PFS 7.2 months (GemCis + durvalumab) versus 5.7 months (GemCis + placebo), HR 0.75 (95% CI 0.63-0.89), P = 0.001 ORR 26.7% versus 18.7% (odds ratio 1.60; 95% CI 1.11-2.31) DOR 6.4 months (GemCis + durvalumab) versus 6.2 months (GemCis + placebo)
Keynote 966 (randomized phase III), 202314	GemCis plus pembrolizumab versus GemCis plus placebo	1069 chemotherapy-naive patients (1 : 1) with unresectable, locally advanced or metastatic BTC	OS 12.7 months (GemCis + pembrolizumab) versus 10.9 months (GemCis + placebo), HR 0.83 (95% CI 0.72-0.95), P = 0.0034	Safety (AE grades 3-4) was similar in both arms
SWOG 1815 (randomized phase III)105	GemCis plus nabpaclitaxel versus GemCis	441 chemotherapy-naive patients (2 : 1) with unresectable, locally advanced or metastatic BTC	OS 14.0 months (GemCis + nabpaclitaxel) versus 12.7 months (GemCis), HR 0.93 (95% CI, 0.74–1.19), P = 0.58	ORR 34% (GemCis + nabpaclitaxel) versus 25% (GemCis), P = 0.11 PFS 8.2 months (GemCis + nabpaclitaxel) versus 6.4 months (GemCis), HR 0.92, 95% CI 0.72-1.16, P = 0.47 Hematologic AE ≥ grade 3: 60% (GemCis + nabpaclitaxel) versus 45% (GemCis), P = 0.003
PRODIGE 38 AMEBICA (randomized phase II-III), 202112	mFOLFIRINOX versus GemCis	191 (1 : 1) with unresectable, locally advanced or metastatic BTC	6-month PFS rate (expected rate 59%) 44.6% (mFOLFIRINOX) versus 47.3% (GemCis)	PFS 6.2 months (mFOLFIRINOX) versus 7.4 months (GemCis) ORR 25% (mFOLFIRINOX) versus 19.4% (GemCis) OS 11.7 months (mFOLFIRINOX) versus 13.8 months (GemCis)
FUGA-BT (JCOG1113) (randomized phase II noninferiority), 2019106	Gem plus S-1 versus GemCis	354 chemotherapy-naive patients (1 : 1) with unresectable or recurrent BTC	OS 15.1 months (Gem–S-1) versus 13.4 months (GemCis), HR 0.945 (90% CI 0.777-1.149), P for noninferiority = 0.0046	PFS 6.8 months (Gem–S-1) versus 5.8 months (GemCis), HR 0.864 (95% CI 0.697-1.070)O RR 29.8% (Gem–S-1) versus 32.4% (GemCis) Clinically significant AE 29.9% (Gem–S-1) versus 35.1% (GemCis)
KHBO1401-MITSUBA (randomized phase III), 2022107	GemCis plus S1 versus GemCis	246 chemotherapy-naive patients (1 : 1) with recurrent or unresectable BTC	OS 13.5 months (GemCis + S-1) versus 12.6 months (GemCis), HR 0.791 (90% CI 0.628-0.996), P = 0.046 1-year survival rate 59.4% (GemCis + S-1) versus 53.7% (GemCis)	PFS 7.4 months (GemCis + S-1) versus 5.5 months (GemCis), HR 0.748 (95% CI 0.577-0.970), P = 0.015 ORR 51.5% (GemCis + S-1) versus 15% (GemCis), P < 0.001
NIFE (AIO-YMO HEP-0315) (randomized phase II), 2021108	Nal-IRI, 5-FU, LV versus GemCis	93 patients with advanced CCA (1 : 1)	4-month PFS rate 51% (Nal-IRI/5-FU/LV) versus 59.5% (GemCis) 4-month PFS rate (iCCA) 41.2% (Nal-IRI/5-FU/LV) versus 71.9% (GemCis) 4-month PFS rate (eCCA) 73.3% (Nal-IRI/5-FU/LV) versus 20.0% (GemCis)	PFS 5.98 months (Nal-IRI/5-FU/LV) versus 6.87 months (GemCis) PFS (eCCA) 9.59 months (Nal-IRI/5-FU/LV) versus 1.76 months (GemCis) OS 15.9 months (Nal-IRI/5-FU/LV) versus 13.63 months (GemCis) OS (eCCA) 18.23 months (Nal-IRI/5-FU/LV) versus 6.34 months (GemCis)

5-FU, 5-fluorouracil; AE, adverse event; BTC, biliary tract cancer; CCA, cholangiocarcinoma; CI, confidence interval; Cis, cisplatin; DCR, disease control rate; DOR, duration of response; eCCA, extrahepatic CCA; Gem, gemcitabine; HR, hazard ratio; iCCA, intrahepatic CCA; LV, leucovorin; mFOLFIRINOX, modified FOLFIRINOX; Nal-IRI, nanoliposomal–irinotecan; ORR, objective response rate; OS, overall survival; PFS, progression-free survival.

Approximately 25% of patients who receive first-line therapy continue with second-line therapy (Table 2) after tumor progression.¹⁵ For those patients, whose tumors do not carry targetable molecular alterations, the mFOLFOX regimen is the recommended standard of care, even though mFOLFOX only provides a small survival benefit with a low response rate (5%). This recommendation is based on the results of the phase III ABC-06 trial, which compared mFOLFOX with active symptom control alone. Patients in the experimental arm showed slightly longer OS (6.2 versus 5.3 months; adjusted HR 0.69, 95% CI 0.50-0.97; P = 0.031) and improved 6- (50.6% versus 35.5%) and 12- (25.9% versus 11.4%) month OS rate.¹⁶ An alternative to mFOLFOX was investigated in the Korean NIFTY trial, which compared liposomal irinotecan plus fluorouracil and folinic acid (FOLFIRI) with fluorouracil and folinic acid alone. PFS and OS were significantly prolonged in the experimental arm.¹⁷ As the study cohort was a Korean-only patient group, further studies with a Western study population were needed. Accordingly, the addition of nanoliposomal irinotecan to fluorouracil and leucovorin was investigated in a multicenter, open-label, randomized phase II trial (AIO NALIRICC) at 17 centers in Germany. The triplet combination did not show an advantage over the combination of fluorouracil and leucovorin alone in terms of PFS and OS.¹⁸ Another combination consisting of trifluridine/tipiracil plus irinotecan is currently under investigation in the German TRITICC-1 trial.¹⁹

Table 2.

Selected trials of chemotherapy in the second-line treatment for patients with advanced BTC

Second-line chemotherapy: “all comers”
ABC-06 (randomized phase III), 202116	FOLFOX plus ASC versus ASC alone	162 patients (1 : 1) with locally advanced or metastastic BTC with disease progression to previous first-line gemcitabine plus cisplatin	OS 6.2 months (FOLFOX + ASC) versus 5.3 months (ASC), HR 0.69 (95% CI 0.50-0.97), P = 0.031 12-month OS rate 25.9% (FOLFOX + ASC) versus 11.4 months (ASC)	PFS (FOLFOX + ASC) 4.0 months (95% CI 3.2-5.0 months) ORR (FOLFOX + ASC) 5% DCR (FOLFOX + ASC) 33%
NIFTY (randomized phase IIb), 2021,17 updated results 2023,109	Nanoliposomal–irinotecan (Nal–IRI) plus 5-FU and LV versus 5-FU and LV	174 patients (1 : 1) with metastatic BTC with disease progression to first-line gemcitabine plus cisplatin	PFS 4.2 months (Nal-IRI/-5-FU/LV) versus 1.7 months (5-FU/LV) HR 0.61 (95% CI 0.44-0.86), P = 0.004
NCT03464968 (randomized phase II), 2021110	mFOLIRI versus mFOLFOX	118 patients (1: 1) with locally advanced or metastatic BTC and failure to first-line gemcitabine plus cisplatin	6-month OS rate 44.1% (mFOLFIRI) versus 54.1% (mFOLFOX) OS 5.7 months (mFOLFIRI) versus 6.3 months (mFOLFOX), P = 0.677	ORR 4.0% (mFOLFIRI) versus 5.9% (mFOLFOX), P = 0.663 DCR 64% (mFOLFIRI) versus 66.7% (mFOLFOX), P = 0.778 PFS 2.1 months (mFOLFIRI) versus 2.8 months (mFOLFOX), P = 0.974
TRITICC (single arm phase IIa), 202319	Trifluridine/tipiracil plus irinotecan	28 patients with locally advanced or metastatic BTC with disease progression to first-line gemcitabine-based chemotherapy	PFS: ongoing, recruitment completed	OS: ongoing, recruitment completedO RR: ongoing, recruitment completed Safety: ongoing, recruitment completed

5-FU, 5-fluorouracil; ASC, active symptom control; BTC, biliary tract cancer; CI, confidence interval; DCR, disease control rate; HR, hazard ratio; LV, leucovorin; mFOLFOX, modified FOLFOX; mFOLIRI, modified FOLIRI; Nal-IRI, nanoliposomal–irinotecan; ORR, objective response rate; OS, overall survival; PFS, progression-free survival.

Especially, after the failure of first-line therapy, the prognosis of patients with advanced BTC remains very poor. Accordingly, BTC is often referred to as a chemotherapy-refractory carcinoma. For patients with a viable target, the hope for the treatment of BTC lies in precision medicine.

The genomic landscape and targeted therapies in BTC

As a result of an increasing use of next-generation sequencing (NGS) analysis, it has been shown that ∼40%-50% of patients with BTC carry genetic alterations that constitute potential targets for individualized therapeutic approaches.20, 21, 22 An overview of the different genetic alterations is shown in Figure 1. Compared with other entities of gastrointestinal cancer, this is a relatively high proportion. Given the striking lack of well-established and effective second-line treatment options, the identification of potential targets for targeted therapy is of great significance. Therefore, both European and German guidelines recommend molecular pathological analysis to identify potential targets at the time of diagnosis or at the latest during first-line therapy in patients with advanced tumor stages and suitable for systemic therapy.^6,7

Figure 1.

The genomic landscape of BTC and the corresponding available drugs. Numbers in brackets indicate the frequency of the corresponding targetable and nontargetable genetic alterations in the different BTC subtypes (iCCA, eCCA, and GBC). The percentages are based on Lamarca et al.¹⁰ For the druggable genetic alterations (IDH mutations, FGFR2 alterations, NTRK fusions, BRAF mutation, and HER2 amplification/overexpression), the substances that are currently available and described in the review article are listed. Some of them have already been approved by the European Medicines Agency (EMA) and the Food and Drug Administration (FDA) (created with BioRender). ARID 1A, AT-rich interaction domain 1A; BAP1, BRCA1 associated deubiquitinase 1; BRAF, B-Raf proto-oncogene, serine/threonine kinase; BTC, biliary tract cancer; CDKN2A, cyclin-dependent kinase inhibitor 2A/B; eCCA, extrahepatic CCA; FGFR2, fibroblast growth factor receptor 2; GBC, gallbladder carcinoma; HER2, human epidermal growth factor receptor 2; iCCA, intrahepatic cholangiocarcinoma; IDH, isocitrate dehydrogenase; KRAS, Kirsten rat sarcoma virus; NTRK, neurotrophic tyrosine kinase receptor gene; PIK3CA, phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha; SMAD 4, SMAD family member 4.

Recent studies have shown that the observed genetic alterations not only differ according to tumor location, but also seem to be influenced by the underlying etiology (e.g. liver fluke) and there is an association regarding the histological subtype.²³ Here, the correct distinction between small-duct-type iCCA and large-duct-type iCCA is particularly relevant.²⁴ Genetic alterations are most common in iCCA. Isocitrate dehydrogenase 1 or 2 (IDH1/2) mutations and fibroblast growth factor receptor 2 (FGFR2) fusions or rearrangements are particularly specific for small-duct-type iCCA and occur in ∼20% and 10%-15% of cases, respectively.²⁵ These are currently the best studied genetic alterations in all BTCs.^10,26 Other promising, although less common, mutations in iCCA are neurotrophic tyrosine kinase receptor gene (NTRK) fusions (4%) and BRAF V600E mutation (3%-6%).^27,28 While human epidermal growth factor receptor 2 (HER2) amplification is rare in iCCA, this alteration is more commonly encountered in GBC (20%) and eCCA (15%) and is usually the only good option for targeted therapy in GBC and eCCA.²⁹ Overall, genetically addressable alterations occur less frequently in eCCA as well as in GBC.¹⁰ Interestingly, while some mutations seem to be mutually exclusive (e.g. IDH1 mutation and FGFR2 fusion/rearrangement), others appear to be clustered (e.g. FGFR2 fusion/rearrangement and mutation in tumor suppressor gene BAP1).²² Moreover, some mutations seem to be associated with the prognosis of the respective tumor, while others have no impact.^22,25,30 The fact that a significant survival benefit can be achieved through improved patient selection and the use of targeted therapy has already been demonstrated in the MOSCATO-01 trial.³¹ In the following sections, we will look in more detail the alterations that can already be therapeutically targeted. The drugs described below are used as second line after failure of first-line chemotherapy, when a suitable alteration has been identified. Some of them have already been approved by the EMA and the FDA. An overview of the trials mentioned in the following sections is given in Table 3.

Table 3.

Selected trials of targeted therapies in the second-line treatment for patients with advanced BTC

Second-line targeted therapies
Trial	Treatment	Population	Primary endpoint	Secondary endpoint
IDH1 mutation
ClarIDHy (randomized phase III), 202046,58	Ivosidenib versus placebo	187 patients (2 : 1) with unresectable, locally advanced or metastatic IDH-mutant CCA and with disease progression after at least one but no more than two previous systemic therapies	PFS (IRC assessment) 2.7 months (ivosidenib) versus 1.4 months (placebo), HR 0.37 (95% CI 0.25-0.54), P < 0.0001	OS (adjusted for crossover) 10.3 months (ivosidenib) versus 5.1 months (placebo), HR 0.49 (95% CI 0.34-0.70), P < 0.001 ORR (IRC assessment) 2% (ivosidenib) versus 0% (placebo)
FGFR2 fusion/rearrangement
FIGHT-202 (single arm phase II), 202035	Pemigatinib (FGFR1-3 inhibitor)	locally advanced or metastatic CCA with at least one previous systemic therapy (107 FGFR2 fusion/rearrangement, 20 other FGF/FGFR alterations, 18 no FGF/FGFR alterations)	ORR (FGFR2 fusion/rearrangement) 35.5% (95% CI 26.5% to 45.4%)	ORR (other FGF/FGFR alterations) 0% ORR (no FGF/FGFR alterations) 0% PFS (FGFR2 fusion/rearrangement) 6.9 months (95% CI 6.2-9.6 months) OS (FGFR2 fusion/rearrangement) 21.1 months (14.8 months-NE)
CBGJ398X2204 (single-arm phase II), 201737	Infigratinib (BGJ398, FGFR1-3 inhibitor)	108 patients with unresectable or metastatic CCA with FGFR2 fusion or rearrangement	ORR 23.1% (95% CI 15.6% to 32.2%)	DOR 5.0 months (IQR 3.7-9.3 months) PFS 7.3 months (95% CI 5.6-7.6 months) OS 12.2 months (95% CI 10.7-14.9 months)
FOENIX-CCA2 (single-arm phase II), 202336	Futibatinib (FGFR1-4 inhibitor)	103 patients with unresectable or metastatic iCCA with at least one previous systemic therapy and an FGFR2 fusion/rearrangement	ORR 42% (95% CI 32% to 52%)	DOR 9.7 months (95% CI 7.6-17 months) DCR 83% (95% CI 74% to 89%) PFS 9.0 months (95% CI 6.9-13.1 months) OS 21.7 months (95% CI 14.5 months-not reached)
ReFocus (single-arm phase I-II), 2022 (NCT04526106)	Lirafugratinib (RLY-4008, highly selective FGFR2 inhibitor)	38 patients with FGFR fusion/rearrangement, FGFR-inhibitor-naive CCA	ORR 63.2 (95% CI 46-78.2)	DOR: not yet mature
BRAFV600Emutation
ROAR basket trial (single-arm phase II), 202080	Dabrafenib plus trametinib	43 patients (39 iCCA) with BRAFV600E-mutated, unresectable, metastatic, locally advanced, or recurrent BTC and with disease progression under previous systemic therapy	ORR 22 of 43 (51%; 95% CI 36% to 67%)	PFS 9 months (95% CI 5.0-10.0 months) OS 14 months (95% CI 10.0-33.0 months)
HER2overexpression/amplification
MyPathway basket trial (nonrandomized phase II), 202192	Pertuzumab plus trastuzumab	39 patients with metastatic BTC with HER2 amplification, Her2 overexpression, or both	ORR 9 of 39 (23%; 95% CI 11% to 39%)
HERB trial (single-arm, phase II), 2022 93	Trastuzumab deruxtecan	22 patients with HER2 positive BTC (IHC3+ or IHC2+/ISH+)	ORR 36.4% (90% CI 19.6% to 56.1%)	DCR 81.8% (95% CI 59.7% to 94.8%) PFS 4.4 months (95% CI 2.8-8.3 months) OS 7.1 months (95% CI 4.7-14.6 months)
Destiny-PanTumor02 (open-label phase II), 202394	Trastuzumab deruxtecan	41 patients with BTC (16 IHC3+, 14 IHC2+, 3 IHC1+, 7 IHC0, 1 unknown), previous HER2-targeted therapy was allowed	ORR (IHC3+) 56.3% (95% CI 29.9% to 80.2%)	PFS (IHC3+) 7.4 months (95% CI 2.8-12.5 months) OS (IHC3+) 12.4 months (95% CI 2.8-NE months)
HERIZON-BTC-01 (single-arm, phase IIb), 202396	Zanidatamab	80 patients with advanced HER2 positive (IHC3+/2+) BTC and progression on previous gemcitabine-based therapy	ORR (IHC3+/2+) 41.3% (95% CI 30.4% to 52.8%)	DOR (HC 3+/2+) 12.9 months (95% CI, 5.95 months-NE)
MSI-H/dMMR
Keynote 158 (multicohort, phase II), 201890	Pembrolizumab	22 patients with unresectable or metastatic BTC and disease progression under at least one previous systemic therapy	ORR 40.9% (95% CI 20.7% to 63.6%; 7 PR, 2 CR)	DOR: not reached PFS 4.2 months (95% CI 2.1 months-NE) OS 24.3 months (95% CI 6.5 months-NE)
NTRK fusion
STARTRK-1 (phase I), STARTRK-2 (basket phase II), and ALKA-372-002 (phase I)101	Entrectinib (RXDX-101, inhibitor of TRKA, TRKB and TRKC, ROS1, ALK, JAK2, and ACK1)	121 patients with NTRK fusion-positive tumors	ORR 61.2% (95% CI 51.9% to 69.9%) with CR 15.7% (19 patients) and PR 45.5% (55 patients)	PFS 13.8 months (95% CI 10.1-19.9 months) OS 33.8 months (95% CI 23.4-46.4 months)
NAVIGATE NCT02576431 (multibasket, phase II adults and adolescents), SCOUT NCT02637687 (phase I/II in children), NCT02122913 (phase I in adults)100,111	Larotrectinib (LOXO-101, inhibitor of TRKA, TRKB, and TRKC)	159 patients (adults and children) with NTRK fusion-positive, non-CNS solid tumors	ORR 79% (95% CI 72% to 85%) with CR 16% (24 patients) and PR 63% (97 patients)	DOR 35.2 months (95% CI 22.8 months-NE) PFS 28.3 months (95% CI 22.1 months-NE) OS 44.4 months (95% CI 36.5 months-NE)

BTC, biliary tract cancer; CCA, cholangiocarcinoma; CI, confidence interval; CNS, central nervous system; CR, complete response; DCR, disease control rate; dMMR, mismatch repair deficiency; DOR, duration of response; FGFR2, fibroblast growth factor receptor 2; HER, human epidermal growth factor receptor; HR, hazard ratio; iCCA, intrahepatic cholangiocarcinoma; IDH, isocitrate dehydrogenase; IHC, immunohistochemistry; IQR, interquartile range; IRC, central independent radiology centre; MSI-H, microsatellite instability-high; NE, not evaluable; ORR, objective response rate; OS, overall survival; PFS, progression-free survival; PR, progressive response.

FGFR2

FGFR2 functions as a receptor for the hormone fibroblast growth factor (FGF), and is a member of a gene family of four receptor tyrosine kinases (FGFR1-4).³² Under physiological conditions, the FGF/FGFR2 signaling pathway is involved in cellular processes that modulate cell proliferation and migration, as well as cell survival and angiogenesis.³³ Fusion or rearrangement of the FGFR2 gene may lead to tumorigenesis and progression via constitutive activation of the receptor.³⁴ Fusion or rearrangement of the FGFR2 constitutes a therapeutic target and is present in ∼10%-15% of patients with iCCA, as mentioned earlier.^22,25 By contrast, this alteration is almost absent in patients with both eCCA and GBC.¹⁰ As a promising target in precision medicine, inhibitors of FGFR2 have already been investigated in several phase II studies and have shown promising results.35, 36, 37

Pemigatinib is an oral selective inhibitor of FGFR1, 2, and 3. In the single-arm phase II FIGHT 202 study, pemigatinib was evaluated in previously treated patients with advanced BTC.³⁵ Patients were divided into three cohorts based on the status of their genomic alterations: FGFR2 fusions/rearrangements, other FGF/FGFR gene alterations, or no FGF/FGFR gene alterations. Patients with FGFR2 fusions/rearrangements showed a very good response to treatment with an objective response rate (ORR) of 35.5%, DCR of 82%, median PFS (mPFS) of 6.9 months, and OS of 21.1 months. There was no relevant response in the other two groups (other FGF/FGFR alterations or no FGF/FGFR alterations).³⁵ The toxicity profile was favorable, with hyperphosphatemia being the most common adverse event related to the treatment with pemigatinib.³⁵ This common side-effect can be explained by the inhibition of the interaction between FGF23 and FGFR1 and their involvement in phosphate homeostasis.^34,35 The positive results of the FIGHT 202 trial support pemigatinib as a standard of care for patients with FGFR2 fusion or rearrangement and led to its approval by the FDA and the EMA as a monotherapy for patients with advanced BTC and tumor progression after at least one prior treatment.^6,7

In addition to pemigatinib, two other FGFR2 inhibitors have shown remarkable efficacy in the second-line treatment of patients with FGFR2 fusion or rearrangement, infigratinib and futibatinib.^36,37 The selective pan-FGFR kinase inhibitor infigratinib achieved an ORR of 23.1% in a multicenter, single-arm phase II study involving 108 patients who had progressed under chemotherapy or were intolerant to chemotherapy; significantly, two-thirds of the included patients had received at least third-line therapy. The ORR was highest in patients that had only received one prior therapy (34%). Infigratinib was also well tolerated. The most common treatment-related adverse events were hyperphosphatemia, constipation, decreased appetite, and stomatitis.³⁷ Based on these promising data, infigratinib was approved by the FDA for patients with advanced BTC and the presence of an FGFR2 fusion or other rearrangement. However, an application for approval by the EMA has been withdrawn by the manufacturer, so this agent is unlikely to play a major role in Europe (https://www.esmo.org/oncology-news/withdrawal-of-application-for-the-ema-marketing-authorisation-of-infigratinib).

A major barrier with FGFR2 inhibitor treatment is the possible development of resistance of the tumor to this therapeutical approach, which is responsible for limiting the duration of response to a few months in some patients.³⁴ A possible mechanism of resistance is the occurrence of secondary FGFR2 kinase domain mutations during treatment. This has been well demonstrated by the work of Goyal et al.,³⁸ in which serial analysis of cell-free circulating tumor DNA demonstrated the presence of point mutations coincident with tumor progression in three patients receiving infigratinib who experienced tumor progression during treatment. Futibatinib is a highly selective and irreversible FGFR1-4 inhibitor that seems to be effective even after resistance to other FGFR2 inhibitors has developed. This is due to the irreversible nature of futibatinib binding and the targeting of the ATP-binding pocket of FGFR2.^36,39 The multinational phase II FOENIX-CCA1 study evaluated the efficacy of futibatinib in patients with iCCA harboring a FGFR2 fusion or other rearrangement and whose tumor had progressed after at least one systemic therapy, but for whom prior treatment with FGFR2 inhibitors was not allowed. The authors reported an ORR of 42% (95% CI 32% to 52%) with a median duration of response of 9.7 months (95% CI 7.6-17 months). A significantly longer median OS (mOS) of 21.7 months (95% CI 14.5 months-not reached) was also achieved.³⁶ In addition, as mentioned earlier, futibatinib has been shown in both preclinical and early clinical studies to be able to induce responses in patients who have already progressed on a FGFR2 inhibitor due to the development of resistance.³⁹ Based on the results of the FOENIX-CCA1 trial, futibatinib was granted accelerated approval by the FDA, making it a promising agent for sequential therapy. In July 2023, futibatinib was also approved by the EMA for patients with advanced BTC and the presence of FGFR2 fusion or rearrangement who have experienced tumor progression after at least one prior systemic therapy. Another potentially viable treatment option for tackling resistance is lirafugratinib (RLY 4008), an orally available, highly selective FGFR2 inhibitor that has shown promising results in the phase I/II ReFocus study (NCT04526106). Several phase III trials are now also evaluating the efficacy of FGFR2 inhibitors in first-line therapy. The FOENIX-CCA3 trial, a multinational, randomized phase III trial, is evaluating the efficacy of futibatinib versus gemcitabine-cisplatin as first-line treatment for patients with advanced iCCA and an FGFR2 gene rearrangement (NCT04093362). The FIGHT-302 trial, that is currently actively recruiting, is a phase III, randomized, active-controlled trial evaluating the efficacy of pemigatinib compared with the GemCis regimen as first-line therapy in the same patient population (NCT03656536). The PROOF trial evaluating infigratinib as a first-line treatment was halted because infigratinib is no longer in development (NCT03773302).

Overall, FGFR inhibitors showed significantly better tumor response as well as significantly better tolerability compared with conventional chemotherapy for selected patients. Thus FGFR2 inhibitors represent a promising therapeutic option for patients with advanced BTC and a fusion or rearrangement in the FGFR2 gene.

IDH

IDH1/2⁴⁰ are the most commonly mutated metabolic genes in human cancers⁴¹ and are promising candidates for targeted therapies in BTC.^24,42 The guidelines of the European Society for Medical Oncology (ESMO) recommend testing for IDH using NGS at diagnosis with concomitant presentation to a tumor board.⁷ According to the German S3 guidelines for diagnosis and therapy of hepatocellular and biliary carcinoma, it is recommended that this test should be performed no later than after the failure of initial treatment and if the patient’s ECOG status is not worse than 2.⁶

An essential factor is the enzyme IDH1, which is present in particularly high concentrations in hepatocytes, where it plays a key role in lipogenesis and maintenance of RedOx homeostasis.⁴¹ According to current understanding, the IDH1 mutation occurs mainly in codon R132.^43,44 This mutation causes IDH to produce abnormally high levels of the oncometabolite R-2-hydroxyglutarate (R-2HG), competitively inhibiting enzymes that regulate epigenetics, DNA repair, metabolism, and other processes.⁴¹ Changes in epigenetic processes in particular have profound effects on the transcription process, which consequently trigger changes in cellular differentiation⁴⁵ and may promote the development and progression of BTC.^1,44 Whether the IDH1 mutation has any prognostic significance in BTC currently remains unclear.⁴⁶ In eight publications from 2013 to 2021 that have examined this question, no association has been found between mutant IDH1 (mIDH1) and PFS, OS, time to progression, or time to recurrence.^20,40,47, 48, 49, 50, 51, 52 In particular, for IDH2 mutations, Wang et al.⁴⁰ described that the time to recurrence was longer and the probability of recurrence was significantly lower at 1, 4, and 7 years compared with wild-type iCCA. For the first time, Rimini et al.⁵³ demonstrated a negative prognostic factor associated with IDH1 mutations in second-line patients who did not receive targeted treatment. Of 119 patients with iCCA, 56 had an IDH1 mutation. After failure of first-line therapy, patients received either FOLFOX or best supportive care. When considering the overall population, the presence of an IDH1 mutation was associated with worse OS (8.2 versus 14.1 months; HR 1.9, 95% CI 1.2-3.0, P = 0.0047) and a worse response rate (ORR 3.6% versus 15.9%; P = 0.033). The OS results did not translate to the FOLFOX-treated subgroup. No differences were observed between IDH1 mutant and IDH1 wild-type tumors; only response (ORR 0% versus 19%; P < 0.0001) and DCR (17% versus 55%; P < 0.0001) were significantly worse in IDH1-mutant tumors. The results should be confirmed with higher case numbers. Similar to other mutations, the occurrence of IDH mutations seems to be correlated with tumor location.⁵⁴ While somatic mutations in IDH1/2 are present in ∼20% of iCCA cases, with IDH1 mutations comprising 13.1% of those cases, they are rare in eCCA.^26,55

Available targeted therapies against mIDH1 or mIDH2 in BTC are limited. Ivosidenib (AG-120) is an oral mIDH1 inhibitor approved by the FDA in August 2021 for pretreated patients with advanced or metastatic BTC with a proven IDH1 mutation.⁵⁶ The approval was based on data from the phase III ClarIDHy study. In this clinical trial, 187 patients with BTC and confirmed IDH1 mutation after previous therapy failure were compared with a placebo group in a 2 : 1 randomization. PFS was significantly improved in the experimental arm versus placebo (mPFS 2.7 months, 95% CI 1.6-4.2 months versus 1.4 months, 95% CI 1.4-1.6 months; HR 0.37, 95% CI 0.25-0.54; one-sided P < 0.0001].⁴⁶ The mOS was 10.3 months for patients treated with ivosidenib, compared with 7.5 months for those who received the placebo. To account for the high crossover rate of >70%, the mOS was adjusted using the rank-preserving structural failure time (RPSFT)⁵⁷ model. According to these data, the mOS for the placebo group is 5.1 months (95% CI 3.8-7.6 months; HR 0.49; 95% CI 0.34-0.70; one-sided P < 0.001).⁵⁸ EMA approval was given in May 2023.

Other IDH inhibitors are currently in development in clinical trials. These include the IDH1-targeted compounds olutasidenib (NCT03684811), LY3410738 (NCT04521686),⁵⁵ IDH305 (NCT02381886),^43,55 and BAY1436032 (NCT02746081), as well as vorasidenib directed against pan-mutated IDH1/2 (NCT02481154)^55,59 and enasidenib, which targets IDH2 mutations (NCT02273739). It is anticipated that the aforementioned substances can also be used after the development of resistance to ivosidenib, because the mechanisms of inhibition are different.⁴¹ Resistance development of mIDH inhibitors has been well explored in acute myeloid leukemia. Secondary acquired IDH mutations or isoform switching of acquired IDH2 mutations into mIDH1 tumors and vice versa have been observed, among others.⁶⁰ Mutations in the Tet-methylcytosine dioxygenase 2 (TET2), Runt-related transcription factor (RUNX), and receptor tyrosine kinase signaling pathways also appear to play a role here.^61,62 Overall, the consistently elevated level of R-2HG seems to be a factor in the development of resistance in this context.⁴¹ Whether these mechanisms can be transferred to the development of resistance to IDH inhibitors in iCCA has not yet been adequately investigated. However, initial studies in a small number of patients show that also in iCCA secondary acquired mutations^60,63,64 and a persistently high R-2HG level may play a significant role.⁴¹ With this background, sequential treatment with the IDH1 inhibitor ivosidenib followed by the IDH2 inhibitor enasidenib or the dual IDH1/2 inhibitor vorasidenib would be conceivable to overcome mechanisms of resistance.⁴¹ Nevertheless, a more complete look at acquired mutations or possibly other factors contributing to ivosidenib resistance is warranted. More recently, the tyrosine kinase inhibitor dasatinib has been identified as a potential and particularly promising candidate also in IDH mutations.^55,65, 66, 67 Results from the phase II trial in patients with IDH-mutated iCCA (NCT02428855) have been published only on ClinicalTrials.gov to date.

BRAF

BRAF plays a fundamental role in the mitogen-activated protein kinase signaling pathway and is thus crucially involved in the control and regulation of cell growth. A BRAF V600E mutation leads to a constitutive activation of kinase activity, which is thus no longer subject to physiological control and can lead to uncontrolled proliferative cell growth.⁶⁸ The frequency of BRAF V600E mutations in BTC is supported by highly controversial data and reported to be 0%-33% for GBC.69, 70, 71, 72 The mutation is predominantly limited to iCCA, occurring with an incidence of 2%-3%.^6,70 Differing from this, 3%-6% are mentioned in the S3 guidelines.^6,70,72, 73, 74 In 2017, Kocsis et al.⁷⁵ reported a very rare case of an extrahepatic BRAF V600E mutation.

A BRAF V600E mutation can be detected both by immunohistochemical (IHC) and by molecular analyses. In an interlaboratory study, it was shown that molecular analyses such as NGS, qPCR, or Sanger sequencing are superior to IHC analysis and should therefore be preferred for a therapy decision.⁶⁸ According to European and German guidelines, the same recommendations apply for molecular analysis as for FGFR2, IDH1, HER2, and NTRK.^6,7

Currently, there is no approved therapy targeting a BRAF V600E mutation in this entity. However, successful single-case reports with the BRAF inhibitor vemurafenib, which has been approved for melanoma, are available.⁷⁶ The MATCH study enrolled four patients with BRAF V600E mutation and iCCA, among other solid tumors, and treated them with the BRAF inhibitor dabrafenib in combination with the MEK inhibitor trametinib, a combination already approved for melanoma with BRAF V600E mutation. All patients were previously treated. Three of these patients had a partial response with an individual PFS of 9.1, 12.8, and 29.4 months, respectively.⁷⁷ Further individual case reports confirm these positive results,^75,78,79 in one case even complete remission was reported.⁷⁹ In the interim analysis of the phase II basket study ROAR published in 2020, 43 patients with BTC and a BRAF V600E mutation were evaluated. In this study, patients were also treated with the combination of dabrafenib and trametinib and achieved an ORR of 51% (95% CI 36% to 67%). The PFS was 9.0 months (95% CI 5-10 months); 30% (95% CI 16% to 45%) of patients were progression-free at 1 year and 8% (95% CI 2% to 22%) at 2 years. The OS was 14.0 months (95% CI 10-33 months).⁸⁰ These data show promising results in that combination and should be further evaluated in this tumor entity. Additional trials for BRAF-mutated patients with BTC are ongoing. Currently, clinical trials are evaluating the PD-L1 inhibitor atezolizumab in combination with the MEK inhibitor cobimetinib (NCT03201458), the ERK1/2 inhibitor ulixertinib (NCT04566393), and the BRAF inhibitor ABM-1310 (NCT05501912 and NCT04190628).

High microsatellite instability/mismatch repair deficiency

DNA mismatch repair deficiency (dMMR) is extremely rare in BTC, comprising <2% of cases.81, 82, 83 The data for iCCA range widely: while Goeppert et al.⁸³ report the frequency at 0.6%, other authors name it at 4.7%-18.2%.^84,85 This last assessment is certainly due to the small number of cases in the total cohorts of only 22 and 23 patients and should be ranked accordingly.

dMMR can be detected well by NGS.⁸⁶ However, multiplex PCR and capillary electrophoresis represent the gold standard.⁸⁷ In cases of dMMR, mutation in any of the four major mismatch repair genes (MLH1, MSH2, MSH6, and PMS2) causes errors such as the addition of missing bases or the replacement of incorrect bases in DNA replication that are not repaired.^88,89 The methylation of the MLH1 promoter is by far the most common mechanism causing sporadic cancers with MMR deficiency.⁸⁹ dMMR implies a high microsatellite instability (MSI-H) and a very high rate of somatic mutations, which carry the potential to produce neoantigens. This results in a high immunogenicity of such tumors, which thus provides a good target for immune checkpoint inhibitor.⁹⁰ The PD-1 inhibitor pembrolizumab was evaluated in the KEYNOTE-158 study of 233 patients in 27 tumor entities with MSI-H/dMMR. Twenty-two of the patients had an advanced BTC. This cohort achieved a PFS of 4.2 months (95% CI 2.1 months-not evaluable [NE]) and an OS of 24.3 months (95% CI 6.5 months-NE) with pembrolizumab. The ORR was 40.9% (95% CI 20.7% to 6.6%).⁹⁰ Based on these data, pembrolizumab as monotherapy is recommended in European guidelines and has already been approved by the EMA for patients with advanced BTC and MMR deficiency or MSI-H after tumor progression on at least one prior systemic therapy.⁷ The FDA has approved pembrolizumab for all MSI-H/dMMR tumors regardless of their entity. The Keynote-158 trial (NCT02628067) and the MOST-CIRCUIT trial (NCT04969887), using the cytotoxic T-lymphocyte associated protein 4 (CTLA-4) inhibitor ipilimumab and the PD-1 inhibitor nivolumab, are currently enrolling participants to address the issue of MSI-H/dMMR in BTC.

HER2

HER2 is a therapeutically targetable alteration found in various tumor diseases. It is well-known to be involved in both breast and gastric cancers.⁹¹ HER2 amplification, overexpression, or both is present in ∼5%-15% of eCCAs and up to 20% of GBCs.¹⁰ In the multiple basket trial, MyPathway, treatment with pertuzumab plus trastuzumab resulted in high response rates (ORR 23%) in well-selected patients with BTC.⁹² The mPFS was 4.0 months and OS was 10.9 months.⁹² However, monoclonal antibodies are not the only treatment option for HER2-expressing BTC. Antibody–drug conjugates (ADCs) are another promising new class of agents. The ADC trastuzumab deruxtecan (T-DXd), consisting of the humanized monoclonal anti-HER2 antibody trastuzumab, a cleavable linker, and the topoisomerase I inhibitor deruxtecan, was recently evaluated in a multicenter, single-arm phase II trial (HERB trial) in patients with an advanced, Her2-expressing BTC.⁹³ The study included 24 patients with HER2-positive [IHC3+ or IHC2+/in situ hybridization (ISH)+] and 8 patients with HER2-low (IHC/ISH status of 0/+, 1+/−, 1+/+ or 2+/−) BTC. A significantly better treatment response was observed in patients with HER2-positive BTC (ORR HER2-positive 36.5% versus HER2-low 12.5%; 90% CI 19.6% to 56.1%; P = 0.01). Although response rates were lower in the HER2-low group an effect of therapy was also seen in these patients with similar PFS and OS.⁹³ The recently published results of the phase II DESTINY-PanTumor02 trial further support the use of T-DXd in patients with HER2-expressing BTC. Patients expressing HER2 IHC3+ achieved an ORR of 56.3% and an OS of 12.4 months. Seven of the 41 patients included had already received HER2-targeted therapy.⁹⁴ A serious side-effect of T-DXd is the possible occurrence of drug-induced interstitial lung disease but its exact mechanism is not yet fully understood. Early detection and treatment of this side-effect is essential to avoid fatal outcome. Other agents (e.g. zanidatamab) are currently tested.⁹⁵ Zanidatamab is a bispecific antibody that binds two distinct HER2 epitopes, resulting in HER2 binding across all HER2 expression levels. Zanidatamab was investigated in the HERIZON-BTC-01, an open-label phase II study, in patients with HER2-amplified, advanced BTC who had received prior gemcitabine-based therapy. Patients with an HER2 positive BTC (IHC3+/2+) achieved an ORR of 41%. The response occurred very fast (time to first response 1.8 months) and was long-lasting (duration of response 12.9 months, 95% CI 5.95 months-NE).⁹⁶ Despite promising results so far, no agent is approved by the FDA or the EMA. Nevertheless, current guidelines already recommend the use of HER2-targeted therapy in patients with HER2-expressing BTC.⁷

NTRK

Neurotrophic tyrosine receptor kinase (NTRK) genes encode neurotrophin receptors (TRK A, B, and C). Following ligand binding, several signaling pathways are stimulated that play important roles in cell proliferation and neuronal development.⁹⁷ However, NTRK alterations, in particular NTRK fusion, are also pivotal in tumor development.⁹⁸ A fusion between the 3′ part of the NTRK gene and the 5′ part of another gene could lead to enduring ligand-independent activation of the receptor and pathways that facilitate tumorigenesis.⁹⁹ In general, NTRK fusion is an infrequent genetic alteration in BTC that impacts only a small fraction of patients with BTC (4%).¹⁰ Nonetheless, with the availability of highly effective and well-tolerated inhibitors, this target should always be considered in the context of NGS analysis. NTRK inhibitors, larotrectinib and entrectinib, manifested highly promising results in early-phase trials.^100,101 In three clinical trials evaluating the effect of larotrectinib in children and adults with NTRK fusion-positive tumors of different origins, an ORR of 79% (95% CI 72% to 85%) was demonstrated. The PFS lasted 28.3 months (95% CI 22.1 months-NE) while OS was 44.4 months (95% CI 36.5 months-NE)¹⁰⁰ Promising results were also achieved with entrectinib, which was also investigated in three phase I/II clinical trials.¹⁰¹ Therefore these NTRK inhibitors were approved by both the EMA and the FDA for the treatment of NTRK fusion-positive cancers, using a tumor-agnostic treatment strategy. Of note, such an approach is even approved for first-line treatment. However, the European guideline recommends the use of NTRK inhibitors in patients who have experienced tumor progression or who are intolerant to prior first-line therapy.⁷ As with FGFR2 inhibitors, acquired mutations may also arise during first-generation NTRK inhibitor therapy.¹⁰² As a result, next-generation inhibitors are currently under development (NCT03093116 and NCT03215511).

Conclusion

In recent years, it has become increasingly clear that BTC, although a rare tumor entity, comprises a broad spectrum of different subtypes in itself. Its molecular complexity, with the identification of various genetic alterations, has made BTC a model for the development of precision oncology approaches. From an all-comers approach with gemcitabine–cisplatin and durvalumab in the first-line setting, the landscape of second-line therapy has changed dramatically with the identification of targeted therapy options for different genetic alterations (e.g. IDH1 mutations, FGFR2 fusions, HER2neu amplification or overexpression). The coming years and studies will certainly bring us more interesting and promising results. While FGFR2 inhibitors and IDH1 inhibitors already offer good options for iCCA, eCCA and GBC remain major challenges, making the identification of new targets and the development of new targeted agents essential. One potential application is the newly developed compound brigimadlin, which targets the mouse double minute 2 homolog (MDM2) and inhibits the interaction between MDM2 and p53, thereby restoring the tumor suppressor function of p53.¹⁰³ In addition, the identification of tumor-resistance mechanisms to targeted drugs and the development of new drugs to counter this resistance will be of particular interest. Furthermore, the application of targeted therapies in the first-line setting, whether as monotherapy or in combination, is a field that still needs substantially more research.