Targeted therapies in advanced cholangiocarcinoma: new targets and strategies
- Open Access
- 18.11.2025
- Gallenblasen- und Gallengangskarzinom
- short review
Erschienen in:
Summary
Standard first-line therapy
For over 10 years, the systemic standard therapy for locally advanced or metastatic cholangiocarcinoma (CCA) has been a doublet chemotherapy of cisplatin and gemcitabine (cis/gem), with a median overall survival (mOS) of 11.7 months [1]. A recent meta-analysis summarizes the unsuccessful efforts of recent years to improve first-line therapy through substitution or addition of chemotherapeutic agents [2]. One example is the randomized phase 2 NIFE study, which showed no benefit of nanoliposomal irinotecan in combination with 5‑fluorouracil and leucovorin (nal-IRI/FU/LV) compared to cis/gem [3]. Even a triple-chemotherapy approach, as tested in the phase 3 SWOG S1815 study, failed to show an advantage of adding nab-paclitaxel to cis/gem [4].
A new standard was only established with the addition of checkpoint inhibitors to the cisplatin/gemcitabine doublet, leading to a 14% reduction in the risk of death [2]. In registration trials, both the anti-PD-L1 antibody durvalumab (mOS 12.8 months—TOPAZ-1) [5] and the anti-PD‑1 antibody pembrolizumab (mOS 12.7 months—KEYNOTE-966) [6] in combination with cis/gem demonstrated prolonged overall survival. Of note, progression-free survival (PFS) was significantly prolonged with durvalumab, whereas the improvement with pembrolizumab was only numerical and did not reach statistical significance [5, 6].
Anzeige
Immunotherapy
As cholangiocarcinomas are relatively resistant to chemotherapy, and only the addition of checkpoint inhibitors has shown an OS benefit so far, new immunotherapy strategies in CCA include bispecific antibodies, alternative checkpoint inhibitors, or combinations with vascular endothelial growth factor (VEGF) inhibitors.
Promising results from a phase 2 study of the bispecific PD‑1 and VEGF antibody ivonescimab with cis/gem were presented at ASCO 2024: median OS was 16.8 months, with an objective response rate (ORR) of 63.6% and disease stabilization in all patients [7].
The randomized phase 2 IMbrave151 study evaluated adding the VEGF antibody bevacizumab to chemotherapy plus PD-L1 antibody atezolizumab [8]. Despite improvements in PFS and duration of response (DOR), overall survival was comparable (14.9 vs. 14.6 months) [8].
The phase 2 AIO-HEP-0123_BATALLION study is currently being prepared in Germany to investigate the addition of dual checkpoint blockade with botensilimab/balstilimab (BOT/BAL) to cis/gem.
Anzeige
Currently enrolling is the GEMINI-Hepatobiliary phase II study, which in a substudy is evaluating cis/gem with the bispecific anti-PD-1/TIGIT antibody rilvegostomig or anti-PD-1/CTLA4 antibody volrustomig as first-line therapy in CCA.
Targeted therapy
Up to 50% of advanced CCA harbor driver mutations treatable with molecularly targeted therapies (Table 1; [9]). Therefore, current ESMO guidelines recommend next-generation sequencing (NGS) for all patients at the start of first-line therapy [9]. The most common and relevant targets are IDH1 mutations, FGFR2 fusions, and HER2 amplification or mutation, with the number of targetable alterations steadily increasing [10]. Besides reaching new targets, future concepts aim to integrate targeted therapy earlier in the disease course.
Molecular-guided maintenance therapy
Since approved targeted therapies are currently only recommended after failure of first-line therapy, the randomized phase 3 umbrella study SAFIR-ABC10 was designed to assess the impact of molecularly guided maintenance therapy after four cycles of standard first-line therapy [11]. Depending on the molecular alteration, maintenance therapy consists of ivosidenib (IDH1 mutation), futibatinib (FGFR2 fusion/rearrangement), zanidatamab (HER2 amplification), neratinib/trastuzumab (HER2 mutation), or encorafenib/binimetinib (BRAF V600E mutation) [11]. The results will show whether personalized treatment can already be effectively implemented in a first-line strategy in order to improve outcome and reduce chemotherapy-related toxicities.
FGFR2 fusion
FGFR2 fusions can be detected in up to 15% of intrahepatic CCA [12]. Approved second-line therapy options for FGFR2 fusions include futibatinib and pemigatinib [10]. In the respective phase 2 registration trials, pemigatinib achieved an ORR of 32.7% with a median DOR of 7.5 months (FIGHT-202) [13], while futibatinib demonstrated an ORR of 42% and a median DOR of 9.7 months (FOENIX-CCA2) [14].
Resistance to FGFR2 inhibitors typically arises through secondary “on-target” mutations. Some mutations, acting as “gatekeepers” or “molecular brakes”, have already been identified and show varied resistance to different FGFR2 inhibitors [15].
To address resistance mutations specifically, the irreversible third-generation FGFR1‑3 inhibitor tinengotinib was developed. A phase 2 study showed clinical efficacy in FGFR2 altered CCA even after progression on prior FGFR2 inhibitors [16]. The phase 3 FIRST-308 trial is investigating tinengotinib as a third-line treatment and is currently enrolling patients at multiple international sites, including centers in Austria (Ordensklinikum Linz, Universitätsklinikum Wiener Neustadt). In the future, sequential and liquid biopsy-guided therapeutic strategies to overcome resistance mutations for FGFR2 fusions could thus represent a new therapeutic standard.
IDH1 mutation
Mutations in IDH1 are detected in approximately 10–20% of patients with intrahepatic cholangiocarcinoma [9]. Based on the phase 3 ClarIDHy study [17], the IDH1 inhibitor ivosidenib is recommended as a second-line treatment in patients with IDH1 mutated CCA [9]. In a cohort of previously treated patients, PFS at 6 and 12 months was 32 and 22%, respectively, in those receiving ivosidenib, whereas all patients in the placebo group experienced disease progression within 6 months of follow-up [17].
HER2 alteration
HER2 overexpression and HER2 amplification or mutation occurs more frequently in gallbladder cancer and extrahepatic CCA than in intrahepatic CCA, with a prevalence of up to 20% [18, 19]. Trastuzumab deruxtecan (T-DXd; DESTINY-PanTumor 02: ORR 22.0%, mPFS 4.6 months) [20] and the bispecific HER2 antibody zanidatamab (HERIZON-BTC-01: ORR 41.3%, mPFS 5.5 months) [21] are options for HER2 3+ or 2+/IHC-positive or HER2 mutant patients after first-line therapy failure [9]. Although the effect is less pronounced, clinical benefit is still observed in CCA patients with low HER2 expression, as demonstrated in the HERB trial (DCR 75%, ORR 12.5%) [22].
Anzeige
Due to the possibly worse prognosis of HER2-positive CCA under chemoimmunotherapy [23], current efforts aim to address HER2 already in the first-line setting. Trastuzumab deruxtecan combined with rilvegostomig is being compared to cis/gem/durvalumab in the phase 3 DESTINY-BTC01 trial. Other recruiting trials include the phase 2 TRAP-BTC (cis/gem/pembrolizumab + trastuzumab) and phase 3 HERIZON-BTC-302 (cis/gem ± checkpoint inhibitor + zanidatamab) study.
Rare fusions
Tumor-agnostic basket trials have demonstrated efficacy of targeted therapies for rare gene fusions in CCA. Treatment with selpercatinib achieved disease stabilization for 5.6 months in a patient with a RET fusion [24] and is already recommended by the European Medicines Agency (EMA) as a tumor-agnostic therapy. For NTRK fusions, ESMO guidelines recommend entrectinib, larotrectinib, or repotrectinib as second-line therapies [10]. A promising new target is the NRG1 fusion, where the HER3 targeting antibody zenocutuzumab demonstrated an ORR of 20% and median DOR of 8.3 months in patients with cholangiocarcinoma [25].
KRAS mutation
KRAS mutations occur in approximately 15% of CCAs, with a high heterogeneity among the different entities. KRAS G12D mutations most currently arise in intra- and extrahepatic cholangiocarcinomas, whereas G13D mutations typically are detected in gallbladder cancers [26].
In the rare G12C mutation subgroup, the selective KRAS G12C inhibitor adagrasib, which is already approved for the treatment of other tumors, showed clinical efficacy in 12 CCA patients (ORR 41.7%, mPFS 8.6 months) [27]. KRAS G12D and pan-RAS inhibitors are currently under investigation in pancreatic carcinoma, with daraxonrasib (RMC-6236) already in a phase 3 trial.
Anzeige
Claudin 18.2 expression
Claudin 18.2 (CLDN18.2) is a potentially tumor-agnostic biomarker. A recent study found expression in 29.5% (70/237) of patients of a mixed CCA cohort. CLDN18.2 positivity (moderate to strong membrane staining in ≥ 75% of tumor cells) was most frequently observed in gallbladder cancer (15.6%), followed by extrahepatic (8.6%) and intrahepatic (2.0%) CCA [28].
Interestingly, chimeric antigen receptor T cell therapy (CAR-T), which has revolutionized hematology in recent years, is also being intensively studied in solid tumors. A phase 1 trial of the CAR‑T construct CT041/satricabtagene autoleucel (satri-cel), which targets CLDN18.2, showed promising early results in gastrointestinal cancers. In a cohort predominantly composed of gastric cancer patients, ORR and DCR were 38.8 and 91.8%, respectively, with a median PFS of 4.4 months and a median OS of 8.8 months. Clinical efficacy was also observed in a small cohort of CCA patients (n = 4) [29].
MTAP loss
Methylthioadenosine phosphorylase (MTAP) plays a key role in the adenosine monophosphate (AMP) salvage pathway and has emerged as a promising molecular target in gastrointestinal cancers [30].
With a prevalence of approximately 15%, MTAP loss represents one of the most common potentially targetable alterations in intrahepatic CCA and is almost always accompanied with CDKN2A/2B loss [30, 31]. Notably, patients with MTAP loss have a significantly poorer outcome with a median OS of less than 6 months, as Lyon et al. observed in a recent retrospective cohort study [31].
Anzeige
Homozygous MTAP deletion results in intracellular accumulation of 5‑methylthioadenosine (MTA), a natural inhibitor of protein arginine methyltransferase 5 (PRMT5). Consequently, MTAP deficient cells exhibit enhanced susceptibility to PRMT5 inhibition due to the pre-existing partial attenuation of PRMT5 enzymatic activity. Novel MTA-cooperative PRMT5 inhibitors exploit this vulnerability by inducing synthetic lethality [32].
AMG 193 demonstrated clinical activity in a phase 1 dose-escalation study across multiple tumor types, including cholangiocarcinoma [33].
Table 1
Molecular targets in cholangiocarcinomas
Gene/alteration | Frequency (%) | Tumor type | Targeted therapy |
|---|---|---|---|
IDH1 mutation | 10–20 [9] | iCCA | Ivosidenib |
FGFR2 fusion | 10–15 [12] | iCCA | Pemigatinib, futibatinib, tinengotinib |
HER2 overexpression, HER2 amplification/mutation | GBC, eCCA | Trastuzumab deruxtecan, zanidatamab | |
Rare fusions (NRG1, NTRK, RET) | < 5 | Rare in all types | Selpercatinib (RET), larotrectinib, entrectinib, repotrectinib (NTRK), zenocutuzumab (NRG1) |
KRAS mutation | 15 [26] | eCCA | Adagrasib (KRAS G12C), (Pan-RAS inhibitors? G12D inhibitors?) |
Claudin 18.2 expression | 5–15 [28] | GBC, eCCA | (Zolbetuximab? CAR-T?) |
MTAP loss | 15 [30] | iCCA | (AMG 193?) |
Take home message
Around 50% of advanced cholangiocarcinomas harbor actionable mutations. Molecular profiling at diagnosis is essential to guide targeted therapy and improve patient outcomes.
Conflict of interest
H. Windhager and B. Doleschal declare that they have no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.