Abstract

Cholangiocarcinoma (CCA) is a heterogeneous group of neoplasms burdened by a dismal prognosis. Several studies have investigated the genomic profile of CCA and identified numerous druggable genetic alterations, including FGFR2 fusions/rearrangements. Approximately 5-7% of CCAs and 10-20% of intrahepatic iCCAs harbor FGFR2 fusions. With the recent advent of FGFR-targeting therapies into clinical practice, a standardization of molecular testing for FGFR2 alterations in CCA will be necessary. In this review, we describe the technical aspects and challenges related to FGFR2 testing in routine practice, focusing on the comparison between Next-Generation Sequencing (NGS) and FISH assays, the best timing to perform the test, and on the role of liquid biopsy.

Introduction

Cholangiocarcinoma (CCA) is a heterogeneous group of invasive adenocarcinomas arising from different locations within the biliary tree. CCA represents the second most common primary hepato-biliary malignancy and comprises 3% of all gastrointestinal cancers ¹.

CCAs occurring within ductules or segmental ducts are classified as intrahepatic cholangiocarcinoma (iCCA), and those arising from the perihilar (pCCA) or distal portions of the biliary tract (dCCA) are classified as extrahepatic cholangiocarcinoma (eCCA). pCCA is the most common CCA, comprising 50-60% of cases, while iCCA is the least common, comprising 10-20% of CCA ².

Risk factors for CCA include primary sclerosing cholangitis, Caroli’s disease, hepatolithiasis, hepatobiliary fluke infections (Opisthorchis viverrini and Clonorchis sinensis), and comorbid hepatic disorders including chronic hepatitis B or C, non-alcoholic steatohepatitis (NASH), and non-alcoholic fatty liver disease (NAFLD), cirrhosis. However, most patients with cholangiocarcinoma have no identifiable risk factors ³.

Incidence and etiological factors vary between geographical regions. While CCA is a rare cancer in high-income countries, it is much more common in China and Thailand, due to the high prevalence of hepatobiliary flukes and hepatolithiasis. Different from other cancer types, mortality has been increasing in patients with CCA in the last decades. Because CCA is often asymptomatic in its early stages, a significant proportion of patients present with locally advanced and unresectable disease. Patients presenting with resectable disease usually undergo potentially curative surgery, followed by adjuvant chemotherapy. However, due to early relapse rates after surgery, the median post-operative survival is 3 years ^4,5.

Regarding iCCA, three different macroscopic growth patterns have been identified, namely mass forming (MF type), periductal infiltrating (PI type), and intraductal growing (IG type), with the MF type being the most common one. MF iCCA arises from peripheral small bile ducts, while PI and IG iCCA originate from large intrahepatic bile ducts ^2,6.

Microscopically, iCCA shows several histological variants (conventional, cholangiolocarcinoma and rare variants) characterized by a different cells of origin and pathogenesis have been recognized. Conventional iCCA may be further classified into large duct iCCA and small duct iCCA. Large duct iCCA may arise from precancerous lesions, such as biliary intraepithelial neoplasia or intraductal neoplasms and are localized in the large intrahepatic bile ducts near the hepatic hilus; small duct iCCA mainly occurs in the peripheral hepatic parenchyma ^2,6.

The molecular landscape of CCA

Numerous studies have investigated the genomic profile of CCA and have shown substantial molecular heterogeneity within this group of neoplasms (Fig. 1).

In the largest genomic study reported in the literature, Javle et al. profiled 4371 CCA and found the most commonly altered genes to be TP53, CDKN2A/B, KRAS, ARID1A, IDH1, BAP1, PBRM1, and FGFR2 (mostly fusions) ⁷.

Despite being limited by the relatively small number of eCCAs, many studies have revealed important differences between the molecular landscapes of iCCA and eCCA (Fig. 1). The genomic profile of iCCA and eCCA are different, with FGFR fusions, mutations, or amplifications and IDH mutations being much more common in iCCA than in eCCA, while KRAS mutations and ERBB2 amplification and overexpression are more prevalent in eCCA ^8-11.

The morphologic heterogeneity of iCCA reflects a substantial molecular heterogeneity. Small duct iCCA often harbors IDH1 and IDH2 mutations and FGFR2 alterations. On the other hand, large duct iCCA, similar to eCCA, is frequently mutated in KRAS and/or TP53 ^12-14.

Based on the current evidence, the European Society for Medical Oncology (ESMO) and United States National Comprehensive Cancer Network (NCCN) guidelines recommend routine use of Next-Generation Sequencing (NGS) multigene panels on advanced CCAs to identify druggable genetic alterations ^15,16. ESMO recommends NGS testing for level I (i.e., improved outcomes in clinical trials) genetic alterations according to the ESMO Scale for Clinical Actionability of molecular Targets (ESCAT), including IDH1 mutations, FGFR2 fusions, and NTRK fusions. The following level II and III actionable genetic alterations have available targeted therapies that do not have any indication for CCAs: microsatellite instability (MSI)/mismatch repair deficiency (dMMR), BRAF mutations, ERBB2 amplifications or mutations, PIK3CA mutations, BRCA1 and BRCA2 mutations, MET amplifications.

Mutations in IDH1/2 have been reported in 10-30% of iCCAs and 7% of eCCAs, with a higher prevalence of IDH2 mutations in comparison with IDH1 mutations (7-20% vs 3%) ¹⁵.

Ivosidenib is an anti-IDH1 targeted therapy that was approved by the United States Food and Drug Administration (FDA) in August 2021 for patients with previously treated, locally advanced, or metastatic CCA harboring an IDH1 mutation ¹⁷. According to the results of phase III ClarIDHy trial, ivosidenib significantly prolonged progression-free survival (PFS) (which was the primary endpoint of the study) in previously treated advanced or metastatic CCA ¹⁸. At final overall survival (OS) analysis, the improvement in OS resulted not significant when comparing ivosidenib versus placebo group: however, when adjusted for crossover (as 70.5% of the patients assigned to placebo received ivosidenib after progression), the analysis revealed a significant OS benefit for ivosidenib ¹⁹.

At present, therapies (i.e., larotrectinib, entrectinib) are available for patients with advanced or metastatic solid tumors, including CCA, harboring NTRK fusions and progressing after standard therapy ^20,21. However, NTRK fusion genes are a rare finding in CCA (approximately 4%) and limited data are available ²².

MSI is the molecular fingerprint of MMRd. Microsatellites are repetitive sequences distributed throughout the human genome, which are prone to the accumulation of mutations. MSI/MMRd can be assessed by immunohistochemistry (IHC) or molecular assays. CCA can be MSI/MMRd but to a lower extent in comparison with other gastrointestinal neoplasms (i.e., colorectal or gastroesophageal cancer). In fact, previous works have reported rates of MSI/MMRd ranging from 1% to 3%. Pembrolizumab is currently indicated for the treatment of MSI/MMRd unresectable or metastatic pre-treated biliary cancer ^7,23,24.

Tumor mutational burden (TMB) is defined as the number of mutations per megabase of coding DNA ²⁵. High TMB (TMB-h), defined as TMB ≥ 10 mutations per megabase, is associated with MSI in some cancer types. In CCA the association between MSI and TMB-h is still debated ²⁶. Javle et al. ⁷ found that 1% of the CCAs profiled had a TMB > 20 mutations per megabase and 3% had a TMB > 10 mutations per megabase.

Following the promising results of the KEYNOTE-158²⁷, the anti-PD-1 pembrolizumab has been approved for the treatment of patients with solid tumors with MSI or TMB ≥ 10 mutations per megabase ²⁸.

BRAF mutations and rearrangements occur in between 1% and 7% of CCAs, with a higher prevalence in iCCAs ^9,11,24. The phase II Rare Oncology Agnostic Research study reported encouraging results in CCA patients with BRAF^V600E mutations treated with dabrafenib in combination with trametinib. However, at present, there is no indication for targeted anti-BRAF therapy in CCA ²⁹.

Furthermore, some of the targeted therapies currently in use in other cancer types are being investigated for use in CCA with BRCA1/2 mutations (olaparib, NCT04042831), ERBB2 amplifications and mutations (trastuzumab, NCT00478140; trastuzumab emtansine, NCT02999672) and PIK3CA hotspot mutations (copanlisib, NCT02631590) ³⁰.

Different testing modalities can be used to identify the large spectrum of genetic alterations in CCAs in tissue samples. Some alterations (i.e., BRAF^V600 mutations, ERBB2 amplifications, EGFR and KRAS mutations, and ALK, ROS1, and EGFR rearrangements) have been well characterized in other cancer types and may be identified with established conventional tests. Conventional tests are based on IHC, fluorescent in situ hybridization (FISH) and PCR-based RNA or DNA sequencing. Despite being rapid and less expensive than NGS, these methods are not suitable to identify multiple genetic alterations in samples with limited amount of tissue ³¹.

Targeted NGS approaches allow molecular profiling of selected gene panels with improved coverage of relevant tumor-specific genes. Due to the lower costs, shorter turnaround time, and simplified data analysis, targeted NGS is more suitable in routine molecular diagnostics than whole genome, whole exome, or transcriptome sequencing ³².

The role of FGFR2 in CCA

The fibroblast growth factor receptor (FGFR) family is a family of tyrosine kinase receptors that include FGFR1, FGFR2, FGFR3, and FGFR4 ³³. Following the binding of growth factors, FGFRs dimerize and activate intracellular signaling pathways responsible for cellular proliferation, survival, and angiogenesis ³⁴. The FGFR2 gene contains at least 24 exonic sequences, however, only subgroups of these are used for different isoforms through alternative splicing. To the present day, more than 25 isoforms of FGFR2 have been described ³⁴.

The binding of an FGF and heparin/heparan sulfates as co-factors to FGFR2 results in dimerization and subsequently trans autophosphorylation of the receptor at its cytoplasmic component. The activated intracellular kinase domain phosphorylates downstream targets, leading to the activation of numerous signaling pathways, including JAK-STAT, MAPK, and PI3K-AKT ³⁵.

FGFR2 gene is located on chromosome 10, and approximately half of FGFR2 fusions evolve through intrachromosomal rearrangements ³⁴. Over 150 different FGFR2 fusion partners have been observed in CCA, with the most common partner being BICC1 ^36,37. All FGFR2 fusions in CCA are “type 2” fusions with transmembrane-type FGFRs with C-terminal substitution to the region of fusion partners ³⁸. FGFR2 fusions are mutually exclusive with FGFR2 mutations and commonly co-occur with BAP1 alterations ³⁹. Of note, no association between FGFR2 rearrangement and high TMB or MSI/dMMR has been reported ^24,40.

In a study by Helsten et al. ⁴¹, 4,853 solid tumors were profiled by NGS and FGFR aberrations were found in 7.1% of all cancers, with the majority (66%) being gene amplifications, followed by single-nucleotide variants (26%), and fusions (8%). Among the 115 CCAs included in the study, 7% harbored FGFR aberrations, mostly in the FGFR2 gene. Alterations in the FGFR1 and FGFR3 genes have also been described in CCA ⁴⁰. Various studies aimed at profiling CCAs estimated the frequency of FGFR2 fusions to be approximately 5-7% in patients with any CCA and in 10-20% of patients with iCCA ⁴². The prevalence of CCA harboring FGFR2 fusions may vary in different geographical areas and is influenced by etiological factors (i.e., FGFR2 fusions are rare in liver fluke-associated CCAs) ⁴³.

Some studies reported a positive association between the presence of FGFR2 fusions and a better prognosis. Graham et al. ¹⁴ evaluated 152 CCAs and 4 intraductal papillary biliary neoplasms of the bile duct for the presence of FGFR2 fusions by FISH. According to the results of the study, the median cancer-specific survival for the 30 patients whose tumors harbored FGFR2 translocations was 123 months compared to 37 months for negative cases without FGFR2 translocations (p = 0.039). In an additional study by Jain et al. ³⁹, out of 377 CCAs, 95 harbored FGFR genetic alterations, including 63 fusions. The presence of FGFR2 aberrations was associated with longer OS compared with patients without FGFR aberrations (37 vs 20 months, respectively; p < 0.001). Furthermore, the rate of FGFR2 genetic alterations was higher among younger patients (≤ 40 years; 20%). Rizzato et al. ²⁴ detected FGFR2 fusions and FGFR3 aberrations in 15/286 (5.2%) and 5/286 (1.5%), respectively, locally advanced or metastatic biliary tract cancers included in the study. At multivariate analysis, FGFR2/3 altered patients had a median OS of 29.2 months compared to 14.4 months for wild-type patients. In the same study, PFS following the start of second-line therapy, was relatively longer in patients with FGFR2 rearrangements (5.0 vs 3.0 months). Another work by Abou-Alfa et al. reported a longer OS and PFS among patients with iCCAs harboring FGFR2 fusions on second-line, but not first-line systemic therapy ⁴⁴.

Targeting FGFR2 in CCA

Several candidate drugs are currently under development and investigations in randomized clinical trials in patients with CCA harboring FGFR pathway alterations, including non-selective and selective FGFR tyrosine kinase inhibitors (TKIs), anti-FGF/FGFR monoclonal antibodies, and FGF traps ⁴⁵. However, the use of non-selective FGFR TKIs comes with various complications in clinical practice, including off-target side effects ⁴⁵. For this reason, a plethora of selective FGFR inhibitors have been evaluated in early-phase clinical trials in patients with refractory iCCA harboring FGFR2 gene fusions, either in randomized clinical trials for iCCA patients or in basket trials ⁴⁶.

Pemigatinib is a small molecule inhibitor of FGFR1, FGFR2, and FGFR3 and represents the first targeted treatment to be approved in the CCA setting. Pemigatinib has received accelerated approval as second-line treatment in April 2020 by the FDA in CCA patients harboring FGFR2 gene fusions or other rearrangements, following the results of the seminal clinical trial FIGHT-202 ⁴⁷. The drug has been subsequently approved by EMA ⁴⁸ and is currently reimbursed by the Italian National Health Service (SSN) for locally advanced or metastatic CCA patients harboring FGFR2 fusions or rearrangements after at least one line of systemic therapy.

The open-label phase II study FIGHT-202 study evaluated the efficacy and safety of pemigatinib in 146 patients with locally advanced or metastatic previously treated iCCA, including 107 with FGFR2 fusions or rearrangements, and found a marked difference in overall response rate (ORR) between patients with FGFR2 fusions or rearrangements (35.5%), and those with other FGF or FGFR alterations or no FGF/FGFR alterations (0% in both groups). Moreover, 82% of patients harboring FGFR2 fusions or rearrangements achieved disease control (i.e., objective response or disease stabilization as best response). Patients with FGFR2 fusions or rearrangements had a PFS of 6.9 months; PFS was 2.1 months for patients with other FGF/FGFR alterations, and 1.7 months for patients without FGFR alterations ⁴⁹. Data from FIGHT-202 compare favorably with those achieved with cytotoxic chemotherapy among unselected patients ⁵⁰, even when a potential prognostic impact of FGFR alterations is considered ⁴⁴. Final data from FIGHT-202 (Vogel A, et al – data presented at ESMO GI 2022) pinpointed an ORR of 37% (95% CI: 28-47) a disease control rate (DCR) of 82% (95% CI: 74-89) and a median duration of response (DOR) of 9.1 months (95% CI: 6-14.5) for tumors with FGFR2 fusions or rearrangements, followed for a median follow-up of 42.9 months. Taken together, these results confirm that therapeutic targeting of FGFR fusions and rearrangements is an opportunity that should not be missed in CCA patients.

Promising results have also been reported with other anti-FGFR agents among CCA patients with FGFR2 fusions or rearrangements. In a multicenter, open-label, phase II study (NCT02150967), the FGFR inhibitor infigratinib demonstrated an ORR of 23.1% in a series of 108 previously treated, locally advanced or metastatic patients ^51,52. Derazantinib, a multikinase pan-FGFR inhibitor, was associated with an ORR of 20.7% and a DCR of 82.8% in a phase I/II study (NCT01752920) among 29 patients with unresectable iCCA with FGFR2 fusion, who experienced disease progression or were intolerant or not eligible to first-line chemotherapy ⁵³. More recently, futibatinib (TAS-120, a highly selective irreversible FGFR1-4 inhibitor) confirmed the value of FGFR-targeting in CCA in the phase II FOENIX-CCA2 study (NCT02052778). The trial design was similar to that of FIGHT-202 and enrolled 103 patients with pretreated unresectable or metastatic iCCA with FGFR2 fusions or rearrangements. Futibatinib demonstrated an ORR of 42% and a DCR of 83%, reporting a median PFS of 9.0 months ⁵⁴. On the basis of these data, futibatinib received FDA approval as salvage treatment of molecularly selected iCCA patients. Notably, futibatinib also demonstrated potent activity against some FGFR2 kinase domain mutations associated with resistance to ATP-competitive FGFR-inhibitors, and preliminary reports suggest potential activity after progression to previous FGFR-inhibitors ⁵⁵. If confirmed, these data open the way towards a continuum-of-care with anti-FGFR agents in this small molecularly-defined subset of CCA patients.

Moving from the activity observed among pretreated patients, FGFR inhibitors are currently being tested as first-line therapy in CCA patients with FGFR2 fusions or rearrangements. The ongoing phase III study FIGHT-302 (NCT03656536) is assessing the efficacy and safety of pemigatinib versus standard-of-care gemcitabine plus cisplatin in the first-line treatment of patients with metastatic CCA harboring FGFR2 rearrangements ⁵⁶. Similarly designed trials are ongoing in order to compare infigratinib (NCT03773302) and futibatinib (NCT04093362) with first-line chemotherapy.

FGFR2 fusion testing in CCA: a practical approach

The development of targeted therapies is significantly impacting the diagnostic and therapeutic decision-making process of CCA patients. With the advent of personalized medicine, modern pathology has gone way beyond traditional morphological evaluations of tissue specimens. The pathologist has become a central figure who is responsible for the delivery of a morpho-molecular report ^57,58. In this new era, the delivery of personalized medicine and oncology strongly relies upon personalized diagnostics. The selection of the most appropriate sample, diagnostic technology and test are crucial factors when detecting patient-to-patient variations in genes or protein expression levels, which act as prognostic or predictive biomarkers ^57,58.

WHAT IS THE BEST SAMPLE?

CCA is a relatively rare neoplasm with a dismal prognosis and is highly heterogeneous from a molecular standpoint. A substantial proportion of CCAs, especially iCCAs, harbor somatic alterations with therapeutic implications. For this reason, treatment guidelines recommend using molecular profiling for metastatic and unresectable advanced CCA, allowing patients to receive biomarker-directed therapy or clinical trial enrollment ³⁰. In the diagnostic scenario, a non-neglible percentage of routine practice samples are classified as “scant samples” (including small biopsy or cytological specimens) due to low quality and quantity of nucleic acids available for molecular testing. In this regard, an optimized workflow based on harmonized pre-analytical and analytical procedures play a pivotal role in improving successful rate of molecular techniques ^59,60. Bekaii-Saab et al. suggested that preoperative biopsy for molecular profiling should be encouraged without delay even in patients with advanced resectable CCA, due to the high rates of relapse after surgery ³¹.

Collecting adequate tissue samples for molecular profiling in advanced or metastatic CCA patients is often challenging. According to Lamarca et al. ⁶¹, one in four archived tissue samples may have insufficient neoplastic cell content for molecular profiling, resulting in the failure of molecular analysis regardless of the platform employed. This may be attributed to the unique location of the tumors and the desmoplastic nature of CCA. The involvement of a pathologist can minimize the failure rate of molecular analysis, but the main obstacle remains the low amount of material collected during endoscopic retrograde cholangiopancreatography (ECRP) or biliary brush citology ⁶¹.

Additionally, tumor location and pattern of growth may influence the amount of material to be dedicated to molecular studies. For example, in periductal infiltrating iCCA and pCCA, surgery is rarely performed, and molecular profiling relies more frequently on biopsy or brushing samples ².

Since the majority of the tumor samples of CCA are small biopsy or cytological specimens, an important aspect that the pathologist must take into account is the suitability of the sample to analyze. The number of cells required for successful DNA/RNA extraction for molecular mutation screening is not defined in CCA, but for other neoplasms (i.e., lung cancer) a range of at least 200-400 cells is desirable ⁶². Moreover, the percentage of tumor cells in a given specimen is a crucial parameter to consider and should be correlated with the sensitivity of the downstream molecular test performed ⁶³. Nevertheless, the quality of DNA/RNA extraction should be assessed before molecular testing to determine the DNA fragmentation index and RNA integrity.

DNA/RNA could be heavily affected by degradation and fragmentation during pre-analytical phases of tissue handling. For example, studies conducted on the preservation status of nucleic acids in FFPE tissues generally agree on the relatively good (though not optimal) preservation of DNA ⁶⁴. On the other, RNA has been found to be heavily degraded and fragmented so that only short sequences (approximately 100-200 nucleotides) can be recognized ⁵. The main effect of formaldehyde in tissues is linked to the formation of methylol groups on amino groups first, followed by the establishment of cross-linking methylene groups that lead to proper fixation ⁶⁶. Bases of nucleic acids are involved in this process, resulting in cross-linking with side-chain amino groups of proteins.

Bussolati et al. demonstrated that RNA degradation can be inhibited by maintaining a low temperature through the entire fixation process (so-called “cold fixation” [CF]) ⁶⁷. The CF procedure is linked to a lower significantly lower degree of nucleic acid fragmentation, especially of mRNA, while keeping the basic advantages that make formalin the fixative of choice in diagnostic histopathology.

WHEN IS THE BEST TIMING?

At present, there is no consensus about whether and when NGS-based genomic testing should be carried out in patients with CCA.

In general, NGS profiling is not recommended in early-stage cancer patients undergoing potentially curative treatment, because it is unlikely to yield actionable alterations beyond those that can be detected using conventional approaches (i.e., IHC, FISH and PCR-based molecular assays) ^68-70. However, because only a small percentage of CCA patients are candidates for definitive treatment, there is a high rate of relapse after receiving surgical treatment and CCA are molecularly heterogeneous with a large number of potentially actionable genetic alterations, patients with early-stage CCA might benefit from NGS profiling after diagnosis ^7,71.

Additionally, due to the few standard-of-care treatment options available, early molecular profiling for locally advanced or metastatic CCA is recommended for matching patients to basket trials recruiting for a specific genetic alteration ^15,72. Thus, the authors recommend performing NGS profiling as reflex testing on all CCAs, regardless of stage at presentation.

NGS testing is becoming a crucial step in the diagnostic and therapeutic decision-making process of oncologic patients. The routine use of NGS in CCA and other solid tumors requires efforts from governmental institutions to allocate resources toward the delivery of NGS testing and the creation of a functional laboratory network ⁷³. On 22^nd December 2022, the Italian Government allocated 600,000 euros (2023-2025) to guarantee access to early NGS testing to all CCA patients ⁷⁴.

WHICH IS THE BEST METHOD?

FISH uses fluorescence-labeled DNA probes to target specific chromosomal locations within the nucleus to detect and quantify gene amplifications and known rearrangements, including gene fusions ⁷⁵.

At present, two FISH approaches broadly in use are the break-apart probes and dual fusion-specific probes. Break-apart FISH requires the use of two differently labeled DNA probes (red and green fluorescence) near the FGFR2 locus. The absence of alterations within the FGFR2 locus creates a combined fluorescence signal (yellow), while the presence of a DNA break caused by rearrangements results in separate fluorescence signals for each probe (green and red) ⁷⁵.

In the dual fusion probe approach, two gene-specific probes are used. The wild-type situation results in separate fluorescence signals (red and green), conversely, if that specific gene fusion is present the fluorescence signals overlap and result in a yellow fusion signal ⁷⁵.

Break-apart FISH is able to identify rearrangements in a partner-agnostic manner, however, it lacks the detailedness of sequencing-based methods and provides no data on fusion gene partners nor on the expression of the fusion protein. Of note, the break-apart assay may lead to false negative results. In fact, in order to order to ensure the visibility of the red and green fluorescence signals, the two probes have to be separated far enough from each other, resulting in an inadequate detection of intrachromosomal rearrangements, which represent 50% of all FGFR2 rearrangements. Conversely, dual-FISH approaches can only detect fusion involving a specific fusion gene partner and cannot assess the expression of the fusion protein ³⁴.

In the era of precision oncology, highly selective single-gene testing has been outdated by NGS and other multiplexed platforms. At present, targeted NGS panels find application in routine diagnostics because they target genes of clinical significance, and have greater sensitivity, faster turnaround time and lower cost ⁷⁶.

Different types of genetic sources of material (i.e., DNA or RNA) can lead to the detection of different types of genetic alterations through NGS analysis. DNA-based NGS can detect any type of genomic alteration, including single-nucleotide variants (SNVs), indels, rearrangements, amplifications, TMB, and MSI. However, the performance of a specific test is influenced by the size of the gene panel and the type of sequences targeted. RNA-based NGS can determine exons and transcribed rearrangements, including alternative splicing events and complex gene fusions, which often go undetected by DNA-based NGS, and can also quantify gene expression levels. On the contrary, when using DNA-based NGS, the transcripts resulting from rearrangements and gene expression levels have to be predicted using computational levels. With RNA-based NGS, mutations with low variant allele frequency and heterozygous loss-of-function mutations can be missed ^77-79.

Different NGS approaches can be used on both DNA and RNA level. The imbalance assay is fusion partner-agnostic. RNA molecules are quantified by looking for an imbalance between the 5’ and 3’ region of the FGFR2 mRNA (i.e., the two ends of the transcript). Amplicon-based approaches are regarded as “closed” because they can only find gene fusions for which a specific primer pair is included in the panel. The single-primer extension approach is another partner-agnostic approach and can identify gene fusion transcripts through the ligation of an adaptor. Hybrid capture-based assays can be used both on RNA and DNA level and rely on target enrichment before sequencing by using sequence-specific hybridization probes ³⁴.

Overall, NGS may represent a more impacting diagnostic method than FISH in the detection of FGFR2 fusions and allows the detection of multiple genetic alterations in tissue samples with scarce material (Fig. 2) ^59,80.

ARE ALL NGS PANELS THE SAME?

DNA-based NGS is preferred for the detection of exonic mutations, while RNA-based NGS can interrogate directly the fusion transcript and is preferred for the detection of fusion genes, including FGFR2 fusions ³¹. Additionally, RNA-based assays are more sensitive because transcription leads to a signal augmentation due to larger numbers of RNA molecules compared to DNA molecules ³⁴.

However, RNA extracted from FFPE is more unstable and prone to degradation, resulting in higher rates of failure of NGS analyses in comparison with DNA-based assays ⁸¹.

The theoretical performances of the previously mentioned NGS assays for specific fusion events are different. Thus, false negative and discordant results can be encountered when applying all the different assays. The single-primer extension and hybrid capture-based assays (DNA and RNA) have the broadest spectrum of detection of FGFR2 fusion events ³⁴.

When choosing an RNA-based NGS panel, only fusion partner-agnostic assays should be considered. In fact, amplicon-based panels should not be used for the detection of FGFR2 fusions because they rely on gene-specific primer pairs and can only detect a predefined set of fusions ³⁴.

Another significant aspect to consider when choosing the most suitable FGFR2 fusion detection method is the amount of sufficient tumor DNA/RNA (i.e., detection limits) for the different assays ³⁴.

For FGFR2 rearrangements, the location of the breakpoint has clinical implications. All FGFR2 fusions in iCCA are “type 2” fusions with C-terminal fusion partners. Because the C-terminal end of the kinase domain is encoded by exon 17, only rearrangements involving exon 17, intron 17, or the protein-coding region of exon 18 will maintain the kinase domain and thus act as oncogenic ³⁴.

According to ESMO guidelines, gene fusions involving the FGFR2 gene should preferably be interrogated at the transcriptomic level using a panel that can detect fusion transcripts of known and unknown fusion partners. Ideally, small biopsies of cholangiocarcinoma are well suited for combinatorial DNA and RNA profiling by single-primer extension and hybrid capture-based assays to identify breakpoints involving mainly exons 17 and 18 of FGFR2. If the tumor content is not sufficient for NGS-based analysis, a break-apart FISH can be performed, since it requires a minimum of 50-100 cells ⁸². Considering these aspects, harmonized bioinformatic pipelines are required to successfully interpret clinically meaningful variants. Bioinformatic tools should consider the heterogeneous landscape of fusion partners that promote clinically relevant aberrant transcripts. In particular, several strategies are available to detect unknown fusion partners taking into account variant calling, unbalanced normalized ratio and quality score. Accordingly, false positive results derived from low-quality samples may occur when low stringency analytic pipeline is approached. In these challenging cases, secondary analysis levels may improve success rate for data interpretation and clinical administration of iCCA patients ⁸³.

WHAT IS THE ROLE OF LIQUID BIOPSY?

Liquid biopsy is emerging as a promising minimally invasive tool for biomarker testing in solid tumors. Liquid biopsy approaches may overcome challenges associated with molecular profiling of tissue samples, including insufficient tumor cellularity, molecular intratumor heterogeneity and the impossibility to perform serial biopsy sampling to monitor the onset of resistance to targeted therapy ⁸⁴. However, the low fraction of ctDNA retrieved from blood samples may challenge the sensitivity of liquid biopsy-based assays for solid tumors, including CCA. Interestingly, Lamarca et al. reported a significantly lower failure rate of molecular profiling of ctDNA in comparison with tissue samples, highlighting that ctDNA may be a valid way of accessing molecular analysis for patients with insufficient tissue ⁶¹.

According to ESMO guidelines ⁸⁵, IDH1 mutation and FGFR2 fusion testing in circulating tumor (ctDNA) is recommended when tissue testing is not feasible or when urgent decision-making for fast therapeutic intervention is required. Although repeat biopsy is the gold standard in case of failure of molecular analysis due to inadequate tumor content, this may be not feasible in locally advanced and metastatic patients ⁸⁶.

Guardant360^® CDx (Guardant Health, Redwood city, CA, USA) and FoundationOne^® Liquid CDx (Foundation Medicine, Cambridge, MA, USA) are commercial FDA-approved blood-based companion diagnostics. While the Guardant360^® CDx panel detects fusions exclusively in FGFR2 and FGFR3, the FoundationOne^® Liquid CDx panel detects fusions in FGFR1, FGFR2, FGFR3, FGFR4 ^87,88.

In a study by Berchuck et al., cell-free DNA (cfDNA) samples of 1671 patients with advanced biliary tract cancers were analyzed using Guardant360^® CDx. cfDNA analysis detected IDH1 mutations and BRAF^V600E at similar rates to tissue biopsies, but the concordance rate for FGFR2 fusions detection was only 18%. The sensitivity of cfDNA profiling for FGFR2 fusions was affected by the diversity of FGFR2 fusion partners. In fact, the sensitivity for FGFR2-BICC1 fusions was 58%, but only 2% for non-BICC1 fusions ⁸⁹. According to the authors, the performance of cfDNA-based fusion detection could be improved by incorporating probes that target common fusion breakpoints and/or a broad range of fusion partner genes and using bioinformatics tools to detect non-targeted fusion partners ⁸⁹. To increase the capability of detecting FGFR2 fusions and rearrangements, a novel fusion partner agnostic algorithm was applied to the Guardant360^® CDx test on 73 plasma samples from CCA patients. The novel algorithm reached an agreement of 84% between tissue NGS and cfDNA profiling, with 24 additional FGFR2 fusions detected in comparison with the standard algorithm ⁹⁰.

Conclusion

CCA still remains an aggressive and deadly neoplasm, due to the lack effective of conventional treatments. Several studies have provided a greater understanding of the complex and heterogenous molecular landscape, identifying several druggable genetic alterations, including a large variety of FGFR2 rearrangements. With the recent approval of pemigatinib targeting of FGFR2 fusions in CCA, a standardization of molecular profiling of these tumors will be necessary.

Due to the advancement of sequencing technologies, NGS-based tests have now lower costs, shorter turnaround time, and simplified data analysis. Overall, NGS outperforms alternative conventional methods like FISH in the identification of FGFR2 rearrangements and allows the detection of multiple genetic alterations in CCA biopsies with low tumor content. Ideally, the best test is combinatorial DNA and RNA profiling by hybrid capture-based assays and single-primer extension panels, in order to cover the broadest spectrum of FGFR2 fusion events. If the tumor content is not sufficient for NGS-based analysis, a break-apart FISH can be perforMed In the setting of CCA, liquid biopsy is emerging as a promising minimally invasive tool for biomarker testing as a way of accessing molecular analysis for patients with insufficient tissue. However, FGFR2 rearrangement detection by ctDNA analysis is still suboptimal but this versatile, dynamic and easily-managing source of nucleic acids may reveal other clinical applications for the management of patients ³¹.

Personalized diagnostics (i.e., the selection of the most appropriate sample, diagnostic technology and test when detecting patient-to-patient variations) has become the cornerstone of personalized oncology. Thus, pathologists and oncologists must be equipped to navigate the complexity of the evolving diagnostic scenario of predictive biomarkers, including FGFR2. Moreover, ongoing clinical trials are evaluating the clinical performance of emerging FGFR inhibitors involving FGFR1-2-3 isoforms. Further investigations about this topic will provide new insights for the comprehensive evaluation of molecular hallmarks in iCCA patients ^91,92.

CONFLICTS OF INTEREST

MF reports research funding (to Institution) from QED, Macrophage pharma, Astellas, Diaceutics; personal honoraria as invited speaker from Roche, Astellas, AstraZeneca, Incyte, Bristol Myers Squibb, Merck Serono, Pierre Fabre, GlaxoSmithKline, Novartis, Amgen; participation in advisory board for Amgen, Astellas, Roche, Merck Serono, GlaxoSmithKline, Novartis, Janssen. UM has received personal fees (as consultant and/or speaker bureau) from Boehringer Ingelheim, Roche, MSD, Amgen, Thermo Fisher Scientifics, Eli Lilly, Diaceutics, GSK, Merck and AstraZeneca, Janssen, Diatech, Novartis and Hedera unrelated to the current work. GP reports research funding (to Institution) from Exact Sciences, Astrazeneca, Novartis, Merck Serono, Boehringer Ingelheim; personal honoraria as invited speaker from Roche, Lilly, AstraZeneca, Incyte, Diatech Pharmacogenetics, Veracyte, GlaxoSmithKline, Novartis, Amgen, Exact Sciences, Bio-Optica, Janssen; participation in advisory board for Amgen, Astrazeneca, Novartis, Exact Sciences. LF reports personal honoraria as invited speaker from Incyte, Bristol Myers Squibb, EliLilly; research funding (to Institution) from MSD, Bristol Myers Squibb, AstraZeneca, Incyte, BeiGene, Astellas, Daiichi Sankyo, Roche; participation in advisory board for MSD, AstraZeneca, Incyte, Taiho, Servier, Daiichi Sankyo, EliLilly.

FUNDING

Matteo Fassan is supported by grants from the Italian Health Ministry/Veneto region research program NET-2016-02363853 and AIRC 5 per mille 2019 (ID. 22759 program). The funding agencies had no role in the design and performance of the study.

AUTHORS’ CONTRIBUTIONS

Conceptualization, MF, GP and UM; methodology LF, FP and GP; data curation, AV; writing-original draft preparation, AV, LF, SMR, and FP; writing-review and editing, MF, GP and UM. All authors have read and agreed to the published version of the manuscript.

Figures and tables

Figure 1.Common genetic alterations in intrahepatic and extrahepatic cholangiocarcinoma.

Figure 2.Comparison of different methods for the detection of FGFR2 spectrum of translocations/fusions.

References

1. Rizvi S, Khan SA, Hallemeier CL, Kelley RK, Gores GJ. Cholangiocarcinoma - evolving concepts and therapeutic strategies. Nat Rev Clin Oncol. 2018; 15:95-111. DOI
2. Sarcognato S, Sacchi D, Fassan M. Cholangiocarcinoma. Pathologica. 2021; 113:158-169. DOI
3. Khan SA, Tavolari S, Brandi G. Cholangiocarcinoma: Epidemiology and risk factors. Liver Int Off J Int Assoc Study Liver. 2019; 39:19-31. DOI
4. Gad MM, Saad AM, Faisaluddin M. Epidemiology of Cholangiocarcinoma; United States Incidence and Mortality Trends. Clin Res Hepatol Gastroenterol. 2020; 44:885-893. DOI
5. Mukkamalla SKR, Naseri HM, Kim BM, Katz SC, Armenio VA. Trends in Incidence and Factors Affecting Survival of Patients With Cholangiocarcinoma in the United States. J Natl Compr Canc Netw. 2018; 16:370-376. DOI
6. Nagtegaal I, Arends MJ, Odze D, Lam A. WHO Classification of Tumours Editorial Board. Digestive System Tumours.. 2019; 1Publisher Full Text
7. Javle M, Bekaii-Saab T, Jain A. Biliary cancer: Utility of next-generation sequencing for clinical management. Cancer. 2016; 122:3838-3847. DOI
8. Nakamura H, Arai Y, Totoki Y. Genomic spectra of biliary tract cancer. Nat Genet. 2015; 47:1003-1010. DOI
9. Weinberg BA, Xiu J, Lindberg MR. Molecular profiling of biliary cancers reveals distinct molecular alterations and potential therapeutic targets. J Gastrointest Oncol. 2019; 10:652-662. DOI
10. Jusakul A, Cutcutache I, Yong CH. Whole-Genome and Epigenomic Landscapes of Etiologically Distinct Subtypes of Cholangiocarcinoma. Cancer Discov. 2017; 7:1116-1135. DOI
11. Simbolo M, Fassan M, Ruzzenente A. Multigene mutational profiling of cholangiocarcinomas identifies actionable molecular subgroups. Oncotarget. 2014; 5:2839-2852. DOI
12. Lowery MA, Ptashkin R, Jordan E. Comprehensive Molecular Profiling of Intrahepatic and Extrahepatic Cholangiocarcinomas: Potential Targets for Intervention. Clin cancer Res an Off J Am Assoc Cancer Res. 2018; 24:4154-4161. DOI
13. Kipp BR, Voss JS, Kerr SE. Isocitrate dehydrogenase 1 and 2 mutations in cholangiocarcinoma. Hum Pathol. 2012; 43:1552-1558. DOI
14. Graham RP, Barr Fritcher EG, Pestova E. Fibroblast growth factor receptor 2 translocations in intrahepatic cholangiocarcinoma. Hum Pathol. 2014; 45:1630-1638. DOI
15. Mosele F, Remon J, Mateo J. Recommendations for the use of next-generation sequencing (NGS) for patients with metastatic cancers: a report from the ESMO Precision Medicine Working Group. Ann Oncol Off J Eur Soc Med Oncol. 2020; 31:1491-1505. DOI
16. National Comprehensive Cancer Network. NCCN clinical practice guidelines in oncology: hepatobiliary cancer, version 5.2021 e September 21, 2021. NCCN Clinical Practice Guidelines in Oncology; 2021. 2022. Publisher Full Text
17. Publisher Full Text
18. Abou-Alfa GK, Macarulla T, Javle MM. Ivosidenib in IDH1-mutant, chemotherapy-refractory cholangiocarcinoma (ClarIDHy): a multicentre, randomised, double-blind, placebo-controlled, phase 3 study. Lancet Oncol. 2020; 21:796-807. DOI
19. Zhu AX, Macarulla T, Javle MM. Final Overall Survival Efficacy Results of Ivosidenib for Patients With Advanced Cholangiocarcinoma With IDH1 Mutation: The Phase 3 Randomized Clinical ClarIDHy Trial. JAMA Oncol. 2021; 7:1669-1677. DOI
20. Drilon A, Laetsch TW, Kummar S. Efficacy of Larotrectinib in TRK Fusion-Positive Cancers in Adults and Children. N Engl J Med. 2018; 378:731-739. DOI
21. Doebele RC, Drilon A, Paz-Ares L. Entrectinib in patients with advanced or metastatic NTRK fusion-positive solid tumours: integrated analysis of three phase 1-2 trials. Lancet Oncol. 2020; 21:271-282. DOI
22. Boilève A, Verlingue L, Hollebecque A, Boige V, Ducreux M, Malka D. Rare cancer, rare alteration: the case of NTRK fusions in biliary tract cancers. Expert Opin Investig Drugs. 2021; 30:401-409. DOI
23. Goeppert B, Roessler S, Renner M. Mismatch repair deficiency is a rare but putative therapeutically relevant finding in non-liver fluke associated cholangiocarcinoma. Br J Cancer. 2019; 120:109-114. DOI
24. Rizzato M, Brignola S, Munari G. Prognostic impact of FGFR2/3 alterations in patients with biliary tract cancers receiving systemic chemotherapy: the BITCOIN study. Eur J Cancer. 2022; 166:165-175. DOI
25. Galuppini F, Dal Pozzo CA, Deckert J, Loupakis F, Fassan M, Baffa R. Tumor mutation burden: from comprehensive mutational screening to the clinic. Cancer Cell Int. 2019; 19:209. DOI
26. Zhao Z, Li W, Zhang X, Ge M, Song C. Correlation between TMB and MSI in patients with solid tumors. J Clin Oncol. 2020; 38(15_suppl):e15169-e15169. DOI
27. Marabelle A, Le DT Ascierto PA. Efficacy of Pembrolizumab in Patients With Noncolorectal High Microsatellite Instability/Mismatch Repair-Deficient Cancer: Results From the Phase II KEYNOTE-158 Study. J Clin Oncol. 2019; 38:1-10. DOI
28. US Food and Drug Administration. Keytruda (pembrolizumab) injection, for intravenous use. Prescribing information. 2021. Publisher Full Text
29. Subbiah V, Lassen U, Élez E. Dabrafenib plus trametinib in patients with BRAF(V600E)-mutated biliary tract cancer (ROAR): a phase 2, open-label, single-arm, multicentre basket trial. Lancet Oncol. 2020; 21:1234-1243. DOI
30. Normanno N, Martinelli E, Melisi D. Role of molecular genetics in the clinical management of cholangiocarcinoma. ESMO open. 2022; 7:100505. DOI
31. Bekaii-Saab TS, Bridgewater J, Normanno N. Practical considerations in screening for genetic alterations in cholangiocarcinoma. Ann Oncol Off J Eur Soc Med Oncol. 2021; 32:1111-1126. DOI
32. Cappello F, Angerilli V, Munari G. FFPE-Based NGS Approaches into Clinical Practice: The Limits of Glory from a Pathologist Viewpoint. J Pers Med. 2022;12. DOI
33. Katoh M. Fibroblast growth factor receptors as treatment targets in clinical oncology. Nat Rev Clin Oncol. 2019; 16:105-122. DOI
34. Neumann O, Burn TC, Allgäuer M. Genomic architecture of FGFR2 fusions in cholangiocarcinoma and its implication for molecular testing. Br J Cancer. 2022; 127:1540-1549. DOI
35. Xie Y, Su N, Yang J. FGF/FGFR signaling in health and disease. Signal Transduct Target Ther. 2020; 5:181. DOI
36. Arai Y, Totoki Y, Hosoda F. Fibroblast growth factor receptor 2 tyrosine kinase fusions define a unique molecular subtype of cholangiocarcinoma. Hepatology. 2014; 59:1427-1434. DOI
37. Wu Y-M, Su F, Kalyana-Sundaram S. Identification of targetable FGFR gene fusions in diverse cancers. Cancer Discov. 2013; 3:636-647. DOI
38. Chen L, Zhang Y, Yin L. Fibroblast growth factor receptor fusions in cancer: opportunities and challenges. J Exp Clin Cancer Res. 2021; 40:345. DOI
39. Jain A, Borad MJ, Kelley RK. Cholangiocarcinoma With FGFR Genetic Aberrations: A Unique Clinical Phenotype. JCO Precis Oncol. 2018; 2:1-12. DOI
40. Silverman IM, Hollebecque A, Friboulet L. Clinicogenomic Analysis of FGFR2-Rearranged Cholangiocarcinoma Identifies Correlates of Response and Mechanisms of Resistance to Pemigatinib. Cancer Discov. 2021; 11:326-339. DOI
41. Helsten T, Elkin S, Arthur E, Tomson BN, Carter J, Kurzrock R. The FGFR Landscape in Cancer: Analysis of 4,853 Tumors by Next-Generation Sequencing. Clin cancer Res an Off J Am Assoc Cancer Res. 2016; 22:259-267. DOI
42. Lamarca A, Barriuso J, McNamara MG, Valle JW. Molecular targeted therapies: Ready for “prime time” in biliary tract cancer. J Hepatol. 2020; 73:170-185. DOI
43. Kongpetch S, Jusakul A, Lim JQ. Lack of Targetable FGFR2 Fusions in Endemic Fluke-Associated Cholangiocarcinoma. JCO Glob Oncol. 2020; 6:628-638. DOI
44. Abou-Alfa GK, Bibeau K, Schultz N. Effect of FGFR2 Alterations on Overall and Progression-Free Survival in Patients Receiving Systemic Therapy for Intrahepatic Cholangiocarcinoma. Target Oncol. 2022; 17:517-527. DOI
45. Ghedini GC, Ronca R, Presta M, Giacomini A. Future applications of FGF/FGFR inhibitors in cancer. Expert Rev Anticancer Ther. 2018; 18:861-872. DOI
46. Goyal L, Kongpetch S, Crolley VE, Bridgewater J. Targeting FGFR inhibition in cholangiocarcinoma. Cancer Treat Rev. 2021; 95:102170. DOI
47. Publisher Full Text
48. Hoy SM. Pemigatinib: First Approval. Drugs. 2020; 80:923-929. DOI
49. Abou-Alfa GK, Sahai V, Hollebecque A. Pemigatinib for previously treated, locally advanced or metastatic cholangiocarcinoma: a multicentre, open-label, phase 2 study. Lancet Oncol. 2020; 21:671-684. DOI
50. Lamarca A, Palmer DH, Wasan HS. Second-line FOLFOX chemotherapy versus active symptom control for advanced biliary tract cancer (ABC-06): a phase 3, open-label, randomised, controlled trial. Lancet Oncol. 2021; 22:690-701. DOI
51. Javle M, Lowery M, Shroff RT. Phase II Study of BGJ398 in Patients With FGFR-Altered Advanced Cholangiocarcinoma. J Clin Oncol Off J Am Soc Clin Oncol. 2018; 36:276-282. DOI
52. Javle M, Roychowdhury S, Kelley RK. Infigratinib (BGJ398) in previously treated patients with advanced or metastatic cholangiocarcinoma with FGFR2 fusions or rearrangements: mature results from a multicentre, open-label, single-arm, phase 2 study. lancet Gastroenterol Hepatol. 2021; 6:803-815. DOI
53. Mazzaferro V, El-Rayes BF, Droz Dit Busset M. Derazantinib (ARQ 087) in advanced or inoperable FGFR2 gene fusion-positive intrahepatic cholangiocarcinoma. Br J Cancer. 2019; 120:165-171. DOI
54. Goyal L, Meric-Bernstam F, Hollebecque A. Futibatinib for FGFR2-Rearranged Intrahepatic Cholangiocarcinoma. N Engl J Med. 2023; 388:228-239. DOI
55. Goyal L, Shi L, Liu LY. TAS-120 Overcomes Resistance to ATP-Competitive FGFR Inhibitors in Patients with FGFR2 Fusion-Positive Intrahepatic Cholangiocarcinoma. Cancer Discov. 2019; 9:1064-1079. DOI
56. Bekaii-Saab TS, Valle JW, Van Cutsem E. FIGHT-302: first-line pemigatinib vs gemcitabine plus cisplatin for advanced cholangiocarcinoma with FGFR2 rearrangements. Future Oncol. 2020; 16:2385-2399. DOI
57. Angerilli V, Galuppini F, Pagni F, Fusco N, Malapelle U, Fassan M. The Role of the Pathologist in the Next-Generation Era of Tumor Molecular Characterization. Diagnostics (Basel, Switzerland). 2021;11. DOI
58. Pisapia P, L’Imperio V, Galuppini F. The evolving landscape of anatomic pathology. Crit Rev Oncol Hematol. 2022; 178:103776. DOI
59. Pisapia P, Pepe F, Sgariglia R. Next generation sequencing in cytology. Cytopathology. 2021; 32:588-595. DOI
60. Maruki Y, Morizane C, Arai Y. Molecular detection and clinicopathological characteristics of advanced/recurrent biliary tract carcinomas harboring the FGFR2 rearrangements: a prospective observational study (PRELUDE Study). J Gastroenterol. 2021; 56:250-260. DOI
61. Lamarca A, Kapacee Z, Breeze M. Molecular Profiling in Daily Clinical Practice: Practicalities in Advanced Cholangiocarcinoma and Other Biliary Tract Cancers. J Clin Med. 2020;9. DOI
62. Thunnissen E, Kerr KM, Herth FJF. The challenge of NSCLC diagnosis and predictive analysis on small samples. Practical approach of a working group. Lung Cancer. 2012; 76:1-18. DOI
63. Nibid L, Sabarese G, Righi D. Feasibility of Comprehensive Genomic Profiling (CGP) in Real-Life Clinical Practice. Diagnostics (Basel, Switzerland). 2023;13. DOI
64. Ferrer I, Armstrong J, Capellari S. Effects of formalin fixation, paraffin embedding, and time of storage on DNA preservation in brain tissue: a BrainNet Europe study. Brain Pathol. 2007; 17:297-303. DOI
65. Dotti I, Bonin S, Basili G. Effects of formalin, methacarn, and fineFIX fixatives on RNA preservation. Diagnostic Mol Pathol. 2010; 19:112-122. DOI
66. Masuda N, Ohnishi T, Kawamoto S. Analysis of chemical modification of RNA from formalin-fixed samples and optimization of molecular biology applications for such samples. Nucleic Acids Res. 1999; 27:4436-4443. DOI
67. Bussolati G, Annaratone L, Medico E. Formalin fixation at low temperature better preserves nucleic acid integrity. PLoS One. 2011; 6:e21043. DOI
68. Remon J, Dienstmann R. Precision oncology: separating the wheat from the chaff. ESMO open. 2018; 3:e000446. DOI
69. Schwartzberg L, Kim ES, Liu D. Precision oncology: who, how, what, when, and when not?. Am Soc Clin Oncol Educ B. 2017;160-169. DOI
70. Colomer R, Mondejar R, Romero-Laorden N, Alfranca A, Sanchez-Madrid F, Quintela-Fandino M. When should we order a next generation sequencing test in a patient with cancer?. EClinicalMedicine. 2020; 25:100487. DOI
71. Ross JS, Wang K, Gay L. New routes to targeted therapy of intrahepatic cholangiocarcinomas revealed by next-generation sequencing. Oncologist. 2014; 19:235-242. DOI
72. National Comprehensive Cancer Network. NCCN clinical practice guidelines in oncology: hepatobiliary cancer, version 5.2021. 2023. Publisher Full Text
73. Marchetti A, Barbareschi M, Barberis M. Real-World Data on NGS Diagnostics: a survey from the Italian Society of Pathology (SIAPeC) NGS Network. Pathologica. 2021; 113:262-271. DOI
74. Publisher Full Text
75. Chrzanowska NM, Kowalewski J, Lewandowska MA. Use of Fluorescence In Situ Hybridization (FISH) in Diagnosis and Tailored Therapies in Solid Tumors. Molecules. 2020;25. DOI
76. de Biase D, Fassan M, Malapelle U. Next-Generation Sequencing in Tumor Diagnosis and Treatment. Diagnostics (Basel, Switzerland). 2020; 10:962. DOI
77. Mercer TR, Clark MB, Crawford J. Targeted sequencing for gene discovery and quantification using RNA CaptureSeq. Nat Protoc. 2014; 9:989-1009. DOI
78. Heyer EE, Deveson IW, Wooi D. Diagnosis of fusion genes using targeted RNA sequencing. Nat Commun. 2019; 10:1388. DOI
79. Reeser JW, Martin D, Miya J. Validation of a Targeted RNA Sequencing Assay for Kinase Fusion Detection in Solid Tumors. J Mol Diagn. 2017; 19:682-696. DOI
80. Luca C De, Pepe F, Pisapia P. RNA-based next-generation sequencing in non-small-cell lung cancer in a routine setting: an experience from an Italian referral center. Per Med. 2022; 19:395-401. DOI
81. van Maldegem F, de Wit M, Morsink F. Effects of processing delay, formalin fixation, and immunohistochemistry on RNA Recovery From Formalin-fixed Paraffin-embedded Tissue Sections. Diagnostic Mol Pathol. 2008; 17:51-58. DOI
82. Vogel A, Bridgewater J, Edeline J. Biliary tract cancer: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up. Ann Oncol Off J Eur Soc Med Oncol. 2023; 34:127-140. DOI
83. Haas BJ, Dobin A, Li B, Stransky N, Pochet N, Regev A. Accuracy assessment of fusion transcript detection via read-mapping and de novo fusion transcript assembly-based methods. Genome Biol. 2019; 20:213. DOI
84. Ceccon C, Angerilli V, Rasola C. Microsatellite Instable Colorectal Adenocarcinoma Diagnostics: The Advent of Liquid Biopsy Approaches. Front Oncol. 2022; 12:930108. DOI
85. Pascual J, Attard G, Bidard F-C. ESMO recommendations on the use of circulating tumour DNA assays for patients with cancer: a report from the ESMO Precision Medicine Working Group. Ann Oncol. 2022; 33:750-768. DOI
86. Levit LA, Peppercorn JM, Tam AL. Ethical Framework for Including Research Biopsies in Oncology Clinical Trials: American Society of Clinical Oncology Research Statement. J Clin Oncol. 2019; 37:2368-2377. DOI
87. Publisher Full Text
88. Publisher Full Text
89. Publisher Full Text
90. Berchuck JE, Facchinetti F, DiToro DF. The clinical landscape of cell-free DNA alterations in 1671 patients with advanced biliary tract cancer. Ann Oncol. 2022; 33:1269-1283. DOI
91. Schram A, Borad M, Sahai V. Identifying FGFR2 fusions/rearrangements in cholangiocarcinoma patients using a novel cfDNA algorithm for treatment with RLY-4008, a highly selective irreversible FGFR2 inhibitor. Eur J Cancer. 2022; 174:S116. DOI
92. Javle MM, Fountzilas C, Li D. Phase II study of FGFR1-3 inhibitor tinengotinib as monotherapy in patients with advanced or metastatic cholangiocarcinoma: Interim analysis. J Clin Oncol. 2023; 41(4_suppl):539. DOI
93. Subbiah V, Iannotti NO, Gutierrez M. FIGHT-101, a first-in-human study of potent and selective FGFR 1-3 inhibitor pemigatinib in pan-cancer patients with FGF/FGFR alterations and advanced malignancies. Ann Oncol Off J Eur Soc Med Oncol. 2022; 33:522-533. DOI