Apply for CE
On January 15, 2021, experts in the management of patients with cholangiocarcinoma (CCA) convened for a virtual accredited continuing education satellite symposium held during the 2021 annual meeting of the American Society of Clinical Oncology Gastrointestinal Cancers Symposium. The goal was to educate healthcare providers on various aspects of CCA, including epidemiology, current standards of care, unmet clinical needs, the safety and efficacy of fibroblast growth factor receptor (FGFR) inhibitors as second-line therapy, and practical approaches to incorporating FGFR inhibitors into the treatment paradigm for the disease.
Faculty included Milind Javle, MD, Professor, Department of Gastrointestinal Medical Oncology, Division of Cancer Medicine, The University of Texas M.D. Anderson Cancer Center, Houston; Lewis R. Roberts, MB, ChB, PhD, Peter and Frances Georgeson Professor, Gastroenterology Cancer Research, and Director, Hepatobiliary Neoplasia Clinic, Mayo Clinic, Rochester, MN; Mitesh J. Borad, MD, Director, Precision Cancer Therapeutics Program, and Associate Professor, Medicine, Mayo Clinic College of Medicine and Science, Phoenix, AZ, and Michael Morse, MD, FACP, MHS, Professor of Medicine, Gastrointestinal Oncology, Duke University Cancer Center, Durham, NC. This monograph provides key highlights from the symposium.
CCA, which is the second most common primary hepatobiliary cancer, develops when bile duct cholangiocytes undergo neoplastic transformation into intrahepatic, perihilar, or distal extrahepatic tumors.1-3 Intrahepatic CCA, which may develop from hepatic progenitor cells, mature hepatocytes, or cholangiocytes,4 begins at the second-order divisions of the bile ducts more peripherally in the liver and extends to the smallest bile ducts.3 Perihilar CCA extends from the second-order divisions of the bile ducts down to the confluence of the bile ducts and the common hepatic duct. Distal CCA is seen from the insertion of the gallbladder cystic ducts through to the ampulla of Vater.3 Extrahepatic CCA may develop from either biliary progenitor cells or mature cholangiocytes.4 The different subtypes of CCA are clinically heterogeneous and have distinct presentations with varying epidemiology, management approaches, and outcomes.5-7
Over the past 30 years, the incidence of intrahepatic CCA has increased across most industrialized countries.8 In fact, the incidence of CCA reveals substantial regional geographic variation worldwide (Figure 1).8 In most countries, CCA is considered a rare (or uncommon) cancer, with an estimated incidence of approximately <6 cases per 100,000 individuals per year. In the United States, the estimated incidence of CCA is approximately 1.67 cases per 100,000 individuals per year.8
Data from the US Surveillance, Epidemiology, and End Results Program showed an approximately 5.5% increase in the incidence of intrahepatic CCA in the United States between 1995 and 2014.9 The reported incidence was higher among individuals aged ≥45 years compared with those aged <45 years (1.71 per 100,000 individuals vs 0.07 per 100,000 individuals, respectively). CCA has a higher incidence in Asian countries—particularly Thailand, China, and Korea.8 Opistorchis viverrini, which is endemic in certain regions of Asia (ie, Thailand, Vietnam, Cambodia, and Laos), accounts for the relatively high prevalence of liver fluke infestation that causes chronic cholangitis and chronic biliary inflammation and appears to predispose individuals to CCA.10
Key risk factors for CCA include disorders such as biliary tract diseases—most notably primary sclerosing cholangitis, which is an immune-related condition.7,8 Other risk factors in terms of biliary tract disorders are gallstone disease, which causes infections in the bile ducts, as well as non-primary sclerosing cholangitis cirrhosis, often associated with alcohol use, fatty liver disease, hepatitis B infection, or hepatitis C infection, and with an increased risk for CCA.8 These risk factors are relatively uncommon, however, and do not occur in large numbers of individuals with CCA. Taking into account adjusted risk, other risk factors, such as type 2 diabetes, smoking, obesity, and inflammatory bowel disease, which are common among Western populations, are thought to be major contributors to the increased incidence of CCA.11 In particular, these factors have been attributed to the rise in intrahepatic CCA in industrialized nations.
Different environmental toxins or industrial toxins may also contribute to an individual’s risk for being diagnosed with CCA. General evidence of increased risk associated with specific toxins appears to be lacking, with the exception of an increased risk for CCA reported among workers with exposure to certain chemical compounds (eg, those used in the offset printing industry).12 Recent studies have also demonstrated protective and/or preventive effects of aspirin (for all subtypes of CCA: adjusted odds ratio, 0.34; 95% confidence interval [CI], 0.30-0.39; P <.001) and of statins (for all subtypes of extrahepatic CCA: adjusted odds ratio, 0.22; 95% CI, 0.16-0.29; P <.001) on the incidence of CCA.13,14
Notably, CCA is among cancers with the worst survival. In individuals aged ≥45 years, the 5-year survival is approximately 10%.9 In patients aged <45 years, the 5-year survival rate is approximately 25%.9 Although CCA is relatively rare, disease-related morbidity and mortality rates are high. Global data from 1995 to 2016 reveal an overall increasing trend in mortality due to a rise in intrahepatic CCA.15 In contrast, a decreasing trend in mortality from extrahepatic CCA has been observed in most countries, which is attributed largely to the increased use of laparoscopic cholecystectomy.15
Although a relatively uncommon disease, CCA is characterized by substantial heterogeneity, as shown in the integrated subclassifications demonstrated by multiomics analyses.16 As presented by The Cancer Genome Atlas project for CCA, different genomic methods, including using copy number aberrations, genome-wide methylation, genome-wide ribonucleic acid sequencing, and exome sequencing for mutations, can reveal important information regarding CCA pathogenesis and help to identify genomic subgroups that are candidates for molecular-targeted therapy.16 An integrated subclassification from this analysis revealed 4 main subgroups. The first subgroup included extrahepatic CCA, which is more frequently associated with SMAD4 mutations.16 Intrahepatic CCA—the largest part of this group—accounted for the other 3 subgroups, including isocitrate dehydrogenase (IDH) mutation-positive CCA, which formed a subgroup with mutations in chromatin-modifying genes; METH2 CCA, which showed a different pattern of methylation (amplification of cyclin D1) from that of the IDH subgroup; and METH3 CCA, which tended to have FGFR2-fusion aberrations, as well as loss of function of BAP1.16
Heterogeneous genetic changes within CCA and an increased understanding of the functional consequences of these genetic modifications form the foundation for targeted therapies. Comprehensive genomic profiling has identified multiple potentially actionable oncogenic alterations in patients with CCA.17 Mutations that are more frequently observed in intrahepatic CCA include IDH1, BAP1, TP53, and FGFR2 fusions.17 Rearrangements or fusions in the gene encoding FGFR2 are found in approximately 10% to 15% of patients with CCA and result in constitutive activation of the FGFR2 receptor, driving tumor growth by leading to overactive signaling and contributing to a variety of tumorigenic processes.18,19 Alterations in IDH1/2 (hotspot gain-of-function mutations), which are found in approximately 19% of patients with CCA, are responsible for an accumulation of the oncometabolite 2-hydroxyglutarate, as well as the constitutive active gene fusion event between FGFR2 and various partners.18,20 Mutations that are more frequently observed in individuals with extrahepatic CCA include KRAS, SMAD4, and STK11 alterations.17 Microsatellite instability-high status is associated with an increased number of neoepitopes, CD8+ T-cell infiltration, and improved responses to immune checkpoint blockade.20
The intrahepatic CCA subgroups are distinguished in a histogenetic fashion between small duct (or cholangiolar) and large duct (or bile duct) subtypes.21 In terms of histologic features, the small duct subtype of intrahepatic CCA has cuboidal to low columnar tumor cells with scanty eosinophilic or amphophilic cytoplasm arranged in small monotonous glands and an “antler-like” pattern, whereas the large duct subtype exhibits more of a glandular pattern, with tall columnar tumor cells arranged in a large glandular pattern and abundant mucin.21,22 Individuals with the small duct subtype have a different gene expression pattern than do those with the large duct subtype.23 Elevated carbohydrate antigen 19-9 and KRAS mutations are more likely to be found in the bile duct subtype.23 Both IDH1 mutations and FGFR2 fusions, on the other hand, are more likely to be detected in the small duct subtype.23 Survival data show improved disease-free and overall survival (OS) in individuals who have the small duct subtype compared with those with the large duct subtype.23 More collaborative studies are warranted to determine whether these histogenetic features can be involved in patient selection for the treatment of CCA.
CCA is treated as a multidisciplinary disease; currently available first-line approaches include surgery, cytotoxic chemotherapy, periadjuvant therapy, and liver transplantation.2,24 Surgery is the only curative treatment for CCA; however, relapse rates following surgery are high.2,25 Moreover, most new cases of CCA (approximately 70%) are diagnosed at an advanced or metastatic stage, with only approximately one-third of patients being eligible for surgical resection.2,26,27 CCA is frequently diagnosed at an advanced, unresectable stage because of the late presentation of nonspecific clinical symptoms of the disease and the lack of effective screening modalities.7,27 Thus, systemic chemotherapy and palliative care with an interprofessional healthcare team are the main first-line treatment options for patients with advanced CCA.
Liver transplantation may be an option in select patients with CCA, particularly young patients with fewer tumors or those with underlying liver disease, in whom the transplant would not only help with the tumor, but also with the liver disease.28 The Mayo Clinic protocol, which is used in patients with perihilar CCA, involves treatment with external beam radiation, chemotherapy, and, in some cases, brachytherapy, followed by liver transplantation (mostly from deceased donors, but in rare instances, from living donors).28 A retrospective study that included 136 patients who underwent liver transplantation according to the Mayo Clinic protocol reported a 74% OS rate at 5 years following the surgery.28 In comparison, outcomes from a large, multi-institutional database study were slightly less favorable, with a 5-year OS rate of 64% reported in 49 patients who underwent liver transplantation.29
The Houston Methodist-M.D. Anderson protocol involves liver transplantation for intrahepatic CCA in the neoadjuvant setting, with adjuvant chemotherapy administered based on explant pathology.30 In a prospective case-series study that included 6 patients with intrahepatic CCA who had liver transplantation, positive outcomes have been reported in 5 of these 6 patients, showing survival at 2.5, 3, and 5 years post-liver transplantation.30
In the BILCAP study conducted in the United Kingdom, patients with macroscopically resected biliary tract cancer were randomized to observation or adjuvant capecitabine for 6 months.31,32 The study population had an Eastern Cooperative Oncology Group (ECOG) performance status (PS) of ≥2, and the primary end point was OS.31 The capecitabine dose (8 cycles, 1250 mg/m2 twice daily) was slightly higher than might be tolerated in patients in the United States, because of regional differences in folate supplementation.31,33 In the United States, patients are more likely to receive folate supplementation, and to have increased toxicity from drug−drug interactions between folic acid and capecitabine-based therapy.33
A total of 753 patients were assessed for eligibility, with 447 of these individuals ultimately randomized in a 1:1 ratio to either capecitabine (N = 223) or observation (N = 224).31 With a median follow-up of 60 months, 51% of patients who received capecitabine had died, compared with 58% in the observation group.32 The median OS was 51.1 months (95% CI, 34.6-59.1) in the capecitabine group versus 36.4 months (95% CI, 29.7-44.5) in the observation group, although the survival difference in the intention-to-treat analysis was not statistically significant (P = .097).32 The per-protocol analysis, however, was statistically significant (53 months vs 36 months, respectively; P = .028), offering some evidence of benefit and supporting the use of capecitabine in practice, given the lack of other options for adjuvant therapy in the current treatment paradigm.32
Toxicities and quality of life (QOL) should also be considered when administering adjuvant therapy. Adverse events (AEs) frequently associated with capecitabine treatment include fatigue, hand-foot syndrome, diarrhea, mucositis, and, in rare instances, cardiac events.32 Adjuvant therapy may be avoided or used with caution in certain patients, particularly those with cardiac risk factors.
First-line chemotherapy is appropriate in patients with unresectable disease. In the phase 3 ABC-02 clinical trial that recruited a total of 400 patients, gemcitabine plus cisplatin was established as a standard first-line systemic therapy for advanced biliary cancers.34 Patients with locally advanced or metastatic CCA, gallbladder cancer, or ampullary carcinoma, with an ECOG PS of ≥2, were randomized to gemcitabine plus cisplatin or gemcitabine alone. The primary study end point was OS, with key secondary end points including progression-free survival (PFS), tumor response, and safety. Early improvement in OS was observed and remained for up to 3 years of follow-up.34 The median OS was 11.7 months in the gemcitabine plus cisplatin group compared with 8.1 months in the gemcitabine alone group (hazard ratio, 0.64; 95% CI, 0.52-0.80; P <.001).34
A single-arm, phase 2 study (N = 74) evaluated gemcitabine plus nanoparticle albumin-bound (nab)-paclitaxel as first-line chemotherapy in 74 patients with advanced CCA.35 The primary end point was improvement in 6-month PFS rate, with key secondary end points including median OS, PFS, time to progression rate, and best overall response rate (ORR) and disease control rate (DCR).35 Efficacy outcomes were comparable to those in the ABC-02 trial, with a median PFS of 7.7 months (95% CI, 5.4-13.1) and a median OS of 12.4 months (95% CI, 9.2-15.9). The best ORR in this study was 30%.35
The single-arm, phase 2 GAP clinical trial, which was conducted at The University of Texas M.D. Anderson Cancer Center, Houston, and the Mayo Clinic, Phoenix, AZ, evaluated the addition of nab-paclitaxel to gemcitabine plus cisplatin for the treatment of patients with advanced biliary tract cancers.36 The study population of 60 patients consisted predominantly of individuals with intrahepatic CCA (63%) and metastatic CCA (78%), making it comparable to the ABC-02 study population.36 The median PFS was 11.8 months and the median OS was 19.2 months.36 The 12-month rate of PFS was 45%.36 Grade 3 or higher AEs occurred in 58% of patients, and 9 patients (16%) withdrew owing to AEs. Neutropenia was the most common grade 3 or higher AE, occurring in 19 patients (33%) overall. Although higher AE rates were observed in patients treated with the gemcitabine-cisplatin plus nab-paclitaxel regimen versus those treated with gemcitabine plus cisplatin, these were more common in the high-dose group. Higher AEs rates must be balanced against the survival and quality-of-life improvements associated with a more complex regimen.36 Therefore, additional research is warranted to compare the nab-paclitaxel triplet regimen with standard gemcitabine plus cisplatin therapy.
Studies such as the phase 3, randomized S1815 clinical trial of gemcitabine plus cisplatin with or without nab-paclitaxel, are evaluating a potentially practice-changing standard-of-care in the first-line treatment of patients with advanced CCA and gallbladder cancer.37 Results from this trial are eagerly awaited and may have a major impact on the standard of care, adding a first-line option for appropriate patients who are able to tolerate a cytotoxic triplet regimen.
Relapse occurs in as many as two-thirds of patients following surgery, and patients with CCA often continue to have poor prognosis with adjuvant therapy.24 There is no established standard of care after failure of first-line chemotherapy in patients with advanced biliary cancers, and the efficacy of second-line chemotherapy regimens remains low, resulting in persistent unmet clinical needs.38,39 A median OS of 6.2 months has been observed with the leucovorin (folinic acid), 5-fluorouracil, and oxaliplatin (FOLFOX) regimen in the second-line treatment setting, but high rates of toxicity often preclude additional cytotoxic therapy in patients who have disease progression and worsening PS following first-line therapy.24
Since most patients with CCA have disease progression, a number of retrospective analyses have been conducted to evaluate the efficacy of second-line treatments.38,40 The results of a systematic review that included 23 studies of second-line chemotherapy in advanced biliary cancer showed a mean OS of 7.2 months (95% CI, 6.2-8.2); a mean PFS of 3.2 months (95% CI, 2.7-3.7); and a mean response rate of 7.7% (95% CI, 4.6%-10.9%).40 A retrospective chart review study of 56 patients with advanced CCA, the majority of whom had intrahepatic CCA, was conducted at The University of Texas M.D. Anderson Cancer Center.41 The median PFS was 2.7 months (95% CI, 2.3-3.8) and the median OS was 13.8 months (95% CI, 12.0-19.3).41 The DCR was 50%.41 A multicenter, retrospective analysis of second-line therapy in 198 patients with advanced biliary cancers reported the shortest OS of 6.8 months in those with extrahepatic CCA and the longest OS of 13.4 months in those with intrahepatic CCA.38
The results of the randomized, phase 3 ABC-06 study that enrolled 162 patients support the use of a FOLFOX regimen as a new second-line standard of care in advanced biliary cancers.42 Patients with advanced biliary tract cancer who had disease progression on first-line gemcitabine plus cisplatin were randomized either to FOLFOX plus active symptom control (ASC) or ASC control alone. The primary end point was OS.42 The difference in median OS between FOLFOX plus ASC and ASC alone (6.2 months vs 5.3 months, respectively) showed a modest, although statistically significant, improvement (P = .031; Figure 2).42 The differences in OS rate at 6 months (50.6% vs 35.5%, respectively) and at 12 months (25.9% vs 11.4%, respectively) were clinically meaningful.42 The subgroup analysis showed consistent benefit across exploratory subgroups, with the greatest benefit reported in subgroups with poorer prognosis, including platinum-resistant patients, those with low albumin levels, and those with metastatic disease.42
A recent study from Korea showed equivalent efficacy using modified regimens of either FOLFOX (mFOLFOX) or irinotecan, leucovorin, and 5-fluorouracil (mFOLFIRI) in the second-line setting.43 The median PFS was 2.8 months (95% CI, 2.3-3.3) with mFOLFOX and 2.1 months (95% CI, 1.3-2.9) with mFOLFIRI (P = .682).43 The median OS was 6.6 months (95% CI, 5.6-7.6) with mFOLFOX and 5.9 months (95% CI, 4.3-7.5) with mFOLFIRI (P = .887).43 The safety profiles differed between the 2 regimens: peripheral neuropathy (36.8%) and thrombocytopenia (35.1%) were more common with mFOLFOX, whereas vomiting (19.3%) and cholangitis (10.5%) were more common with mFOLFIRI.43 In light of similar efficacy profiles, safety concerns may influence the treatment choice between these 2 options.
An ongoing phase 3 trial is evaluating the addition of NUC-1031 (a phosphoramidate transformation of gemcitabine) to cisplatin (ClinicalTrials.gov identifier NCT04163900)44 as a prototype doublet chemotherapy regimen versus standard of care. In addition, combinations of gemcitabine plus cisplatin plus nab-paclitaxel (ClinicalTrials.gov identifier NCT02632305)45; nan-irinotecan plus fluorouracil plus leucovorin (ClinicalTrials.gov identifier NCT04005339)46; and trifluridine and tipiracil (Lonsurf) plus irinotecan (ClinicalTrials.gov Identifier: NCT04072445)47 are being investigated in the phase 2 setting (Table 1).
FGFRs are involved in the activation of multiple signaling pathways that are critical in tumorigenesis, leading to cellular proliferation, differentiation, survival, antiapoptosis, and angiogenesis.48 In particular, FGFR2 fusions or rearrangements have been implicated in the pathogenesis of intrahepatic CCA.23 With FGFR2, the extracellular, transmembrane, and kinase domains remain predominantly intact. However, a breakpoint occurs in exons 17, 18, or 19, resulting in rearrangements including fusions with one of a variety of other genes.48 FGFR fusion partners contribute specific domains that favor the dimerization of FGFR2 and then subsequent phosphorylation of tyrosine kinase domains, triggering downstream signaling (including phospholipase C gamma, and the RAS-MAPK, PI3K-AKT, and JAK-STAT pathways).48
Multiple nonselective inhibitors, such as dovitinib, lucitanib, nintedanib, ponatinib (Iclusig), and derazantinib, target FGFR (typically FGFR1-4), in addition to a number of other tyrosine kinases, such as vascular endothelial growth factor receptors (VEGFRs), rearranged during transfection (RET), KIT, and platelet-derived growth factor receptors (PDGFRs);49 however, these nonselective inhibitors are limited by their lower specificity and potency for FGFR1-4. Thus, the therapeutic doses needed to achieve adequate FGFR inhibition with many of these agents often lead to unacceptable toxicities.
Unlike other nonspecific inhibitors, derazantinib has demonstrated strong activity against FGFR2, as well as FGFR1 and FGFR3. Other targets of derazantinib are colony-stimulating factor 1 receptor (CSF1R), RET, KIT, and PDGFRs.50,51 The half maximal inhibitory concentration (IC50) value of derazantinib, which is the concentration required in a particular in vitro assay to achieve 50% inhibition of the molecule, is lower for FGFR2 compared with many of the other molecules that are targeted by this agent.52
In a phase 1/2, multicenter, open-label clinical trial that enrolled 29 participants, derazantinib showed benefit in patients with unresectable intrahepatic CCA with FGFR2 fusions who had progressed on previous therapy, or were intolerant of or ineligible for first-line treatment.50 The ORR was 20.7% (all partial responses), with a median duration of response of 4.6 months; the rate of stable disease was 62.1%.50 The median duration of drug exposure was 5.6 months (range, 1.5-18.2 months); the median OS was not reached.50 AEs observed with derazantinib use were similar to those associated with other FGFR-targeting agents. The most common treatment-related AEs included dry mouth (44.8%), nausea (44.8%) asthenia/fatigue (34.5%), dysgeusia (31%), and vomiting (31%). Grade ≥3 AEs were reported in 8 (27.6%) patients.50
The topline results from cohort 1 of the phase 2 FIDES-01 trial were updated recently, in February 2021.53 A total of 103 patients with intrahepatic CCA and FGFR2 gene fusions who had received ≥1 previous chemotherapy regimen demonstrated an objective response rate of 20.4%, based on 21 patients with a confirmed partial response.53 The DCR, which reflects the proportion of patients with a partial response or with stable disease, was 72.8%. The median PFS was 6.6 months.52 The results are not yet fully mature, as evaluations of 12 patients are still ongoing, including 3 individuals who had a partial response.53
Consistent with previous data, derazantinib demonstrated a well-manageable safety profile. The most common drug-related AEs reported with once-daily oral administration of derazantinib 300 mg were hyperphosphatemia, asthenia/fatigue, increased liver enzymes, nausea, diarrhea, dysgeusia, dry mouth, and dry eye.53 The proportion of patients who had drug-related AEs of nail toxicities was low, at 6%, and events of retinopathy, stomatitis, or hand-foot syndrome each were reported in only 1% of patients.53
The selective FGFR inhibitors include the pan-FGFR type I inhibitors infigratinib, pemigatinib (Pemazyre), and futibatinib, all of which target primarily FGFR1-3 and inhibit FGFR4, but to a lesser extent (Table 2).54 The IC50 values for these drugs show that their FGFR1-3 inhibition is much more potent than their FGFR4 inhibition. Although pemigatinib may have a slightly lower affinity for FGFR2 than do infigratinib and futibatinib, the selective FGFR inhibitors generally show comparable FGFR2 inhibition in vitro. Infigratinib and pemigatinib are reversible FGFR inhibitors (ie, they bind reversibly to the adenosine triphosphate (ATP) binding pocket), whereas futibatinib covalently binds to the ATP binding pocket in an irreversible manner.
A phase 2, multicenter, open-label study evaluated the use of pemigatinib in 146 previously treated patients with locally advanced or metastatic CCA.55 The study included 3 cohorts of patients: (1) those with FGFR2 fusions or rearrangements; (2) those with other types of fibroblast growth factor (FGF) or FGFR genetic alterations; and (3) those with no FGF or FGFR genetic alterations. The majority of patients in the study had FGFR2 fusions or rearrangements.55 Patients received pemigatinib 13.5 mg orally, in a 2-weeks-on, 1-week-off schedule—the dosing regimen that is now approved by the US Food and Drug Administration.55,56 Among patients with FGFR2 fusions or rearrangements, the objective response rate was 35.5% (partial response, 32.7%), and the median duration of response was 7.5 months (range, 5.7-14.5 months). The median PFS was 6.9 months (range, 6.2-9.6 months), and the median OS was 21.1 months (range, 14.8 months-not estimable).55
Based on cross-study comparisons, the data for pemigatinib demonstrate improved survival over the use of chemotherapeutic agents and/or regimens (eg, FOLFIRI, FOLFOX, and gemcitabine-based chemotherapy) in patients with FGFR2 fusions or rearrangements.37,38 This survival benefit, however, was not observed in patients with other or with no FGFR2 alterations. The median OS was only 6.7 months (range, 2.1-10.6 months) in patients with other FGF or FGFR2 alterations and 4.0 months (range, 2.3-6.5 months) in those with no FGF or FGFR2 alterations.55
Pemigatinib has a similar safety profile to that of other agents from the selective FGFR2 inhibitor class because they all target FGFR1-3.55 Hyperphosphatemia is the most common treatment-related AE.55 Hypophosphatemia can also occur as a result of aggressive management of hyperphosphatemia. Other treatment-related AEs observed with pemigatinib treatment include alopecia, dysgeusia, diarrhea, nausea, ocular effects, joint complaints, hypercalcemia, and skin and nail changes. Most AEs are grade 1 or 2.55 Pemigatinib should be withheld until resolution of grade 3 or 4 AEs, which are relatively uncommon.
In a single-arm, open-label, phase 2 clinical trial, infigratinib was evaluated in 108 patients with previously treated, advanced CCA, including 81% (88 of 108) of patients with FGFR2 fusions.57 The study included patients who had progressed on or were intolerant to gemcitabine-based chemotherapy.57 Patients received infigratinib 125 mg orally for 21 days of a 28-day cycle.57 Overall, 53.7% of the patients had received ≥2 previous lines of therapy; all but 1 patient had stage IV disease.57
At a median follow-up of 11.3 months (range, 0.03-20.90+ months), the objective response rate was 23.1% (95% CI, 15.6%-32.2%), with 1 complete response and 24 partial responses reported.57 The median duration of response was 5.0 months (range, 0.9-19.1 months). The median PFS was 7.3 months (95% CI, 5.6-7.6).57
Patients who were less heavily pretreated achieved a better response, although there were still some responders among those who had received ≥2 previous lines of therapy. In a prespecified subgroup analysis of infigratinib in 50 patients who had received ≤1 line of previous therapy compared with 58 patients who had received ≥2 lines of previous therapy, the objective response rate was 34.0% and 13.8%, respectively (Table 3).57
The AE profile of infigratinib was similar to that of other FGFR2-targeted inhibitors. The most common grade 3 or 4 AEs included stomatitis (14.8%), hyponatremia (13.0%), and hypophosphatemia (13.0%).57,58
In preclinical trials, fewer resistant clones were reported with treatment using the irreversible FGFR inhibitor futibatinib compared with the reversible FGFR inhibitors.59 Of the mutations that subsequently were observed in preclinical trials with futibatinib, none occurred in the kinase domain. Futibatinib seems to be effective even if mutations develop in FGFR2, with one of the reasons for the continued activity of the agent attributed to its covalent binding, which creates stable complexes and may allow for continued inhibition of the molecule. The site where futibatinib binds is thought to be less prone to mutations compared with where the reversible inhibitors bind, leading to fewer mutations being observed with futibatinib.59
The phase 2 FOENIX-CCA2 clinical trial evaluated the use of futibatinib in 67 patients with locally advanced or metastatic intrahepatic CCA who harbored FGFR2 fusions or rearrangements.60 Eligible patients who had measurable disease after receiving previous gemcitabine plus platinum-based therapy were treated with futibatinib 20 mg orally.60 The objective response rate was 37.3% (partial response, 35.8%), and the median duration of response was 8.3 months.60 A higher likelihood of response to futibatinib was observed in patients with increased phosphate levels.60 Response was also observed in patients with mutations other than FGFR2, such as IDH1 comutations. Patients with a PDRM1 mutation, however, had stable disease or progressive disease.60 The AE profile of futibatinib was similar to that of other FGFR2-targeted inhibitors; grade ≥3 AEs included hyperphosphatemia (28%), nail toxicities (1%), hand-foot syndrome (1%), and ocular toxicities (1%).
Based on patient-reported outcomes data in the FOENIX-CCA2 study, QOL was maintained during treatment with futibatinib.61 Mean scores for European Organization for Research and Treatment of Cancer QOL Questionnaire (EORTC-QLQ-C30; 5 functional and 9 physical measures) and EuroQOL-5 Dimension (EQ-5D-3L; utility index and 5 dimensions: anxiety/depression, mobility, pain/discomfort, self-care, and usual activity) revealed no adverse effect on QOL.61 The CI values, however, were very wide. Patients who had significant antitumor activity would be expected to have some improvement in patient-reported outcome measures, but those who did not respond to futibatinib may have had diminished QOL because of toxicities. Individual components of quality-of-life metrics, such as pain, also showed wide CI values. These findings demonstrate that the clinical benefit of futibatinib resulted in sufficient QOL gain overall as well as for individual components, to offset any toxicities associated with treatment.61
The potential adverse effects associated with the use of FGFR2 inhibitors are similar, although the percentages may vary among the different agents.62 It is important to recognize that hyperphosphatemia is a renal effect linked to the targeting of FGFR1. In the future, hyperphosphatemia may not occur with the use of FGFR2-specific drugs. Under normal circumstances, FGF23 binds to FGFR in the renal proximal tubule, which inhibits renal phosphate reabsorption.62 Therefore, an inhibitor with FGFR1 inhibition activity leads to resorption of phosphate and causes phosphate levels to become elevated. When phosphate levels are >5.5 mg/dL, patients should be started on a low phosphorus diet.62
When phosphate levels are between 7 and 9.9 mg/dL, phosphate-lowering therapy (eg, sevelamer) should be initiated. If phosphate levels remain >7 mg/dL after 2 weeks of phosphate-lowering therapy, FGFR2 inhibitor therapy should be withheld until the levels decrease to >7 mg/dL and then resumed at the same dose. The dose may need to be lowered with persistent hyperphosphatemia; phosphate levels of >10 mg/dL require FGFR2 inhibitor therapy to be withheld, as well as a dose reduction to the next lower dose level, in addition to phosphate-binding therapy.61
Dry eye is a common AE associated with the use of FGFR2 inhibitors, and ocular demulcents or artificial tears should be used, as needed. Corneal effects can also occur and should be managed by an ophthalmologist. In addition, ocular toxicities consisting predominantly of retinal pigment epithelial detachment may also occur in some patients, which can be asymptomatic or can manifest as blurred vision or floaters.62 It is recommended that patients see an ophthalmologist for optical coherence tomography before initiating FGFR inhibitor therapy, then every 2 months for the first 6 months and every 3 months thereafter during treatment. If patients have symptomatic retinal pigment epithelial detachment, the FGFR inhibitor should be withheld until symptoms resolve and then resumed at a lower dose. If symptoms persist, the FGFR inhibitor should be discontinued. If the patient is asymptomatic, then the agent can generally be continued.
Patients should also be warned about skin and nail toxicities, which can sometimes be alarming, particularly if the nails detach from the nail bed.62 Skin and nail toxicities can be treated with moisturizers prescribed by a dermatologist. In cases of infection, cephalexin or mupirocin ointment may be used. In patients with hand-foot syndrome, urea-containing creams are helpful. To prevent infection in patients with nail toxicities, daily soaking in vinegar is recommended. Urethane-type lotions can also be applied to the nail beds at night. Topical antibacterials or topical steroids may be needed in some cases. If the area of the nail has thinned, clear nail polish can be used to protect that area of the nail.
“Magic mouthwashes” (commonly consisting of diphenhydramine, viscous lidocaine, antacid, nystatin, and occasionally corticosteroids)63 or dexamethasone elixir (Decadron) can help in cases of stomatitis. For dry mouth, artificial saliva or drugs that stimulate our cholinergic agonists (eg, pilocarpine or cevimeline [Evoxac]) can be used. It is important to remember the risks and side effects associated with FGFR2 inhibitors, to heighten patients’ awareness of possible side effects, and to be proactive in monitoring and treating AEs.
The challenge for oncologists who manage individuals with intrahepatic CCA is how to best incorporate FGFR inhibitors into the treatment paradigm for their patients. Findings from a multicenter, retrospective analysis that included 95 patients with FGFR-altered CCA showed a predominance of young patients (aged <40 years) who were more likely to have FGFR alteration than FGFR wild-type.64 Interestingly, FGFR alterations are associated with a favorable prognosis, with a median OS of 37 months (95% CI, 24-65) compared with 20 months (95% CI, 17-26) in patients without this alteration (Figure 3).64
The following example illustrates how FGFR inhibitors may be incorporated into clinical practice.
A patient diagnosed with CCA was treated with gemcitabine plus cisplatin, then with proton therapy following disease progression. The patient had a recurrence of cancer in his liver and received a transarterial chemoembolization procedure, along with capecitabine. A repeat next-generation sequencing panel was performed, which revealed an FGFR fusion that was not detected earlier on a 50-gene panel. The patient was treated in a clinical trial with derazantinib for approximately 7 months, and received an orthotopic liver transplantation and is still doing well 2 years after diagnosis.
Tumor genomic alterations were assessed in a phase 2 study of infigratinib. BAP1 was the most commonly occurring co-mutation with FGFR. Whereas FGFR is a driver mutation, BAP1 has no prognostic significance in CCA. Other genes (eg, PIK3CA, CDKN2A, PBRM1) may be associated with poor outcomes, even in the presence of FGFR genetic aberrations. IDH1 and KRAS occur rarely with FGFR fusions.65 Additional studies are warranted to understand how outcomes with FGFR inhibitors are influenced by concurrent genetic alterations and their resulting effects on downstream signaling, which may inform sequencing of therapies and the combining of FGFR inhibitors with other agents.
Gatekeeper mutations, which limit the accessibility of the kinase ATP-binding pocket to small-molecule inhibitors, have been identified as a common mechanism for acquired resistance to targeted therapies.19 In addition to gatekeeper mutations that induce resistance to FGFR inhibition in the ATP binding pocket, other secondary gatekeeper mutations have been identified that drive acquired resistance to infigratinib and pemigatinib, and innate resistance to futibatinib, including FGFR2 p.V564F and FGFR2 p.N549.66-68 Other mutations involved in the development of resistance include alternative receptor tyrosine kinases in the human epidermal growth factor receptor 2 (HER2) family, KRAS, RAS, and MEK.19 Since mechanisms of innate and acquired resistance appear to be somewhat similar across all FGFR inhibitors evaluated, a new class of inhibitors is needed to circumvent at least some of the acquired gatekeeper mutations.
FGFR2 inhibitors have relatively similar ORRs, DCRs, and PFS rates. Although OS may differ according to the stage at which patients were enrolled or their lines of therapy, these agents appear to have relatively similar efficacy profiles. Thus, treatment selection can be guided by increased knowledge of the molecular and biological characterization of CCA.20 A combination of clinical, radiologic, histologic, genomic, and molecular features need to be taken into account for the management of CCA.20
As part of the characterization of CCA, molecular considerations, such as genetic alterations, kinase activity, and mechanisms of resistance, should be factored into treatment decisions. A greater understanding of how resistance to signaling mechanisms affects treatment outcomes will help inform optimal combination strategies in the first-line and the refractory settings. A study by Krook and colleagues showed that combination therapy with FGFR and mechanistic target of rapamycin inhibitors may be used to overcome acquired resistance to infigratinib.69 Combination strategies have also been explored with infigratinib in hepatocellular carcinoma, with the programmed cell death 1 inhibitor erdafitinib (Balversa) in lung cancer, and in an unpublished phase 1 study with pemigatinib added to gemcitabine plus cisplatin.70-72
FGFR inhibitors hold promise for future strategies in CCA and serve as a prime example of what we can achieve with precision medicine. Rapid sequencing and liquid biopsies to test for resistance will alter the current paradigm, potentially decreasing the use of chemotherapy. Understanding primary resistance to FGFR inhibitors and concurrent mutations will inform us about combination approaches and will help guide the sequential use of treatments. Moreover, the need exists to develop new FGFR inhibitors whose potency is not affected by kinase domain mutations (including gatekeeper mutations), which will offer improved outcomes for patients who have had limited options with currently available treatments.
Recent clinical advances have altered the treatment paradigm for biliary cancers, with FGFR inhibitors garnering interest as potential therapeutics for CCA.24 Promising results have been shown in studies of these agents in patients with FGFR alterations, expanding treatment options for those who have disease progression on first-line chemotherapy.24 Future research should be directed toward the selection of novel targeted therapies to overcome resistance and the use of combination therapies to inhibit additional pathways of cellular proliferation.24