Development of molecularly targeted therapies in biliary tract cancers: Reassessing the challenges and opportunities

Authors

  • Andrew X. Zhu,

    Corresponding author
    1. Massachusetts General Hospital Cancer Center, Harvard Medical School, Boston, MA
    • Tucker Gosnell Center for Gastrointestinal Cancers, Massachusetts General Hospital Cancer Center, 55 Fruit Street, Boston MA 02114
    Search for more papers by this author
    • fax: (617) 724-3166

  • Aram F. Hezel

    1. James P. Wilmot Cancer Center, University of Rochester School of Medicine, Rochester, NY
    Search for more papers by this author

  • Potential conflict of interest: Dr. Zhu advises Novartis, Pfizer, and Bayer. Dr. Hezel serves on the speaker's bureau of Augen and Bayer and receives grants from Augen.

Abstract

Biliary tract cancers (BTCs), which encompass intra- and extrahepatic cholangiocarcinomas as well as gallbladder carcinomas, are a genetically diverse collection of cancers. Most patients with BTC will present with unresectable or metastatic disease. Although the standard systemic chemotherapy approaches are emerging, the prognosis remains poor. Development of molecularly targeted therapies in advanced BTC remains challenging. Recent early-stage clinical trials with targeted therapies appear promising, although the relationships between subsets of patients with positive responses to therapy and tumor genetics remain unexplored. Here we summarize the relevant molecular pathogenesis, recent and ongoing clinical trials with targeted agents, and the key issues in clinical trial design in BTC. (HEPATOLOGY 2011;53:695-704)

Biliary tract cancers (BTCs) include a spectrum of invasive adenocarcinomas encompassing both cholangiocarcinoma (CC), used to refer to cancers arising in the intrahepatic, perihilar, or distal biliary tree, as well as carcinoma arising from the gallbladder (GBC). BTC is characterized by early lymph node and distant metastases and, as a result, only 10% of patients present with early-stage disease and are considered candidates for surgical resection, which offers the only chance for cure. In selective centers, liver transplantation may be considered an option for early-stage disease as well. The prognosis is poor for patients with locally advanced or metastatic BTC, with a median survival of approximately 1 year. BTC afflicts approximately 12,000 people in the United States annually and the incidence of CC is rising.1

Although molecularly targeted therapies have emerged as the standard treatments either alone or in combination with chemotherapy or radiation, in many cancer types, the development of targeted agents in BTC has encountered many challenges (Fig. 1).2, 3 First, the molecular mechanism underlying the pathogenesis of BTC remains elusive; the early stages of the disease are poorly understood and the genetics of advanced disease have yet to be fully defined. Second, BTC is a heterogeneous group of related but distinct diseases with various clinical features and likely different genetics changes. Third, BTC is relatively rare and has been traditionally listed as an “orphan” disease. Conducting clinical trials in this disease has many practical concerns. Fourth, the pharmaceutical industry has not considered this disease for drug development due to the relatively small market. However, these challenges have also created great opportunities in reassessing the role of targeted agents in BTC. In recent years there is renewed interest in developing molecularly targeted therapies in BTC for several reasons. First, there is an unmet need in this disease and the incidence of intrahepatic cholangiocarcinoma (IHCC) is rising. Second, the recently completed phase III trial demonstrating improved survival of gemcitabine/cisplatin versus gemcitabine alone has reminded the oncology community that practice changing clinical trials can be done through collaboration in this rare tumor. Third, early trials with targeted agents have demonstrated intriguing results. Finally, the registration pathway for drug approval in this disease is relatively straightforward. This provides input to examine some of the newly developed targeted agents in BTC.

Figure 1.

Molecularly targeted agents under development in biliary tract cancers.

Here we review the molecular pathogenesis of the disease as it is relevant to recent, ongoing, and, hopefully, future clinical trials with targeted agents. We also discuss key issues in clinical trial design in BTC.

Abbreviations

AJPBD, anomalous junction of the pancreaticobiliary duct; BTC, biliary tract cancer; CC, cholangiocarcinoma; EGFR, epidermal growth factor receptor; GBC, gallbladder carcinoma; VEGF, vascular endothelial growth factor.

Molecular Features of BTC

The development of targeted agents in cancer is increasingly guided by an individual tumor's genetic profile. Examples of this include the use of HER2 blocking antibodies in beast cancer, epidermal growth factor receptor (EGFR) and anaplastic lymphoma kinase (ALK) inhibitors in lung cancer,4, 5 specific v-raf murine sarcoma viral oncogene homolog B1 (BRAF) inhibitors in melanoma,6 and the selection of EGFR blocking antibodies for the treatments of v-Ki-ras Kirsten rat sarcoma viral oncogene homolog (KRAS) wildtype colon cancers.7 As tumor genetics have come to shape the paradigm of personalized medicine among a range of other tumor types, such approaches have lagged in BTC for reasons described above.

One key question that remains outstanding is to determine the precise genetic landscape of these tumors. We know that biliary tract cancers have a spectrum of mutations in established oncogenes and tumor suppressors (listed in Table 1). It has been hard to pinpoint the frequency of these events due to the differing methodologies used to detect mutations and the high degree of tumor infiltrating stromal cells that can dilute cancer-specific nucleic acids and limit analysis. Beyond the genes and associated pathways that harbor mutations in BTC, a number of other molecules have demonstrated importance in the disease and provided additional potential targets. The mutational analysis of BTC has been reviewed elsewhere.8 Below we review key pathways of importance in the disease, particularly those in which drug development efforts are under way.

Table 1. Mutational Spectrum of Oncogenes and Tumor Suppressor Genes
GeneGallbladder Carcinoma %Cholangiocarcinoma %MethodRefs.
EHCCIHCC
  1. Abbreviations: SEQ; sequencing, PCR-SSCP; polymerase chain reaction single-strand confirmation polymorphism, RFLP; restriction fragment length polymorphism, GLCR; gap ligase chain reaction.

CTNNB1/- CATENIN5  SEQ75
90 SEQ76
0  SEQ77
KRAS38  PCR-SSCP13
20  PCR-RFLP78
19  PCR-RFLP79
315 SEQ11
 10 PCR-SSCP40
  54SEQ14
  45SEQ10
  48SEQ80
BRAF33  SEQ15
  22SEQ10
000SEQ & GLCR17
EGFR91820SEQ20
 6 SEQ21
12510IHC and FISH22
PIK3CA409SEQ29
ERBB2/HER-21650IHC and FISH22
P16INK4A31  SSCP78
6255 numerous38
  88numerous14
TP5336  SSCP78
 33 PCR-SSCP40
  37SEQ41
SMAD4 16 PCR-SSCP42
 5513IHC43
STK11/ LKB1 6 SEQ45

Mutations identified within the KRAS pathway and a number of studies functionally linking RAS/MAPK (mitogen-activated protein kinase) pathway activation with growth of BTC have brought this axis to the forefront of therapeutic efforts. Activation of KRAS, a member of the RAS/RAC family that propagates growth signals via downstream effectors such as RAF and phosphoinosotide 3-kinase (PI3K), is found in subsets of both GBC and CCs.9 Mutation rates range from as low as 3% in some series to 100% among patients with GBC in the setting of anomalous junction of the pancreaticobiliary duct (AJPBD) (see Table 1).10-14 The mutation rate of BRAF, which sits at the upstream of MEK/MAPK signaling pathways, and is a key effector of the oncogenic activity of KRAS, is unclear. Two European BTC collections including both GBCs and IHCC identified mutations in ≈20% of cases.10, 15, 16 However, no mutations were identified in GBC and CC from North America and Chile despite the use of three methods to detect mutations.17 Ongoing mutational analysis of BRAF among larger cohorts with newer sensitive technologies could provide better assessment of the true mutation frequency. Downstream of BRAF, the MAPK pathway appears active in ≈77% of BTC, as evidenced by P-MAPK immunostaining.16 Expression profiling across a panel of seven human BTC cell lines demonstrated expression of a number of RAS/MAPK pathway components and treatment of these cell lines with a MAPK inhibitor CI-1040 was synergistic when used in combination with an EGFR inhibitor.18 Other BTC lines have also demonstrated sensitivity to MEK pathway inhibitors.19

Mutational analysis of EGFR, models of biliary tract disease, and early clinical data with EGFR inhibitors point towards this molecule's importance in BTC. EGFR is found to be both mutated (13.6%-15%) and amplified (≈6%) in subsets of BTC.20-22 Balanced polysomy of chromosome 7, which harbors EGFR, is a common event in BTC and accounts for one copy gains in EGFR copy number in a significant 46% of BTC analyzed in one series.23 EGFR is expressed at some level in over half of BTC and blocking antibodies are growth inhibitory to BTC cell lines.24 An indirect, but compelling, line of evidence in support of the pathogenic role of EGFR in BTC come from genetic studies in mice with mutation of nuclear factor 2 (NF2), which encodes the protein Merlin.25 These studies have shown that Merlin, through regulation of EGFR, controls hepatic precursor cell proliferation. Such cell populations are potential cells of origin of BTC. NF2 mutant mice are tumor-prone, developing both hepatocellular carcinoma (HCC) and BTC. EGFR inhibitors have shown early evidence of efficacy in clinical treatment trials as discussed below and remain a key area of therapeutic investigation.

HER-2/NEU (v-erb-b2 erythroblastic leukemia viral oncogene homolog 2, neuro/glioblastoma derived oncogene homolog (avian), also known as HER-2/NEU) receptor tyrosine kinases have been implicated in BTC pathogenesis. ERBB2/HER2NEU overexpression and gene amplification is most commonly seen in GBC (≈15%) compared with low or absent amplification events in extrahepatic cholangiocarcinoma (EHCC) at ≈5% and IHCC at 0%. The biologic relevance is further demonstrated in transgenic mouse models created through the expression of Erbb2 in the gallbladder epithelium.26 GBC and EHCC occur with a 100% penetrance in this model. Tumors from this model exhibit elevated MAPK signaling and have been used to preclinically evaluate the mammalian target of rapamycin (mTOR) inhibitor rapamycin.27 Although the incidence of HER2 amplifications is low in terms of total number of patients diagnosed each year with BTC containing HER2 amplifications, there may be an appreciable benefit to therapy directed at HER2. Recently, similar efforts aimed at identifying a similarly small subset of HER2 amplified cases in gastric cancer proved fruitful. Treatment with an HER2 targeting antibody, trastuzumab, significantly extended the lives of those with the HER2 amplification.28

PIK3CA (phosphoinositide-3-kinase, catalytic, alpha polypeptide) hotspot mutations, commonly found in a number of other cancer types, are rarely found in BTC.29 Despite a low prevalence of activating mutations, immunohistochemical evidence of activation of downstream molecules suggests that additional mechanisms may be at play in positively regulating this pathway. Expression profiling of BTC identified up-regulation of AKT/mTOR signaling components including the potential drug target IGF1-R (insulin like growth factor 1 receptor).30 IGF-R and its ligands are expressed in the majority of primary GBC and metastasis, making this an attractive target.31, 32 Along these lines, treatment of BTC cell lines with a small molecule inhibitor of the IGF-R was growth inhibitory, further supporting such an approach.33 Although mutations in PTEN (phosphatase and tensin homolog) have not been reported, targeted mutation of PTEN was found to contribute to BTC in a mouse model. Compound mutant mice with somatic mutation of Smad4 and Pten in the liver develop IHCC.34 Although deletion of Smad4 alone did not lead to any hepatic phenotype, mutation of Pten did cause biliary hyperplasia and IHCC, albeit with a significantly longer latency, suggesting the pathways role in controlling proliferation and survival in the biliary tract.

The spectrum of tumor suppressor gene (TSG) mutations seen in BTC largely mirrors that of other gastrointestinal malignancies (see Table 1). Loss of function of CDKN2A (cyclin-dependent kinase inhibitor 2A, commonly known as P16INK4A),35 TP53 (tumor protein p53)(36), and SMAD4 (SMAD family member 4) are all commonly found.37 CDKN2A, which is located at the 9p21 locus, encodes two proteins through splicing of an alternative first exon, may be silenced in BTC through hypermethylation, homozygous deletion, and inactivating point mutation.14, 38, 39 TP53 inactivation, through either overexpression or mutation, is identified across all subtypes of BTC.11, 40, 41 SMAD4 mutations are identified in both GBC and CC.42, 43 Engineered mutants of both p53 and Smad4 mutations have been used in concert to model IHCC in mice.34, 44 With regular toxin (carbon tetrachloride, CCL4) exposure causing liver injury and fibrosis, p53 mutant mice develop IHCC with a number of molecular similarities to the human disease, such as elevated expression of c-Met, Cox2, and Erbb2.44 Finally, one homozygous deletion of STK11 (serine/threonine kinase 11, also known as LKB1), a key regulator of metabolism and cellular polarity, was identified among 16 tumors of the common bile duct.45 STK11 is established as a key TSG in nonsmall-cell lung carcinoma (NSCLC) and to a lesser extent pancreatic cancer.46 Additional molecular alterations found in BTC that have been generally linked with cancer are reviewed elsewhere.47, 48

Beyond targeting the tumor cell itself, successful strategies in cancer treatment have included disruption of tumor vasculature, stromal cells such as fibroblasts, and immune modulation.49 VEGFs (vascular endothelial growth factors) and other angiogenic molecules have been used prognostically in BTC. VEGF is expressed by most tumors50, 51 and high VEGF expression appears to correlate with hematogenous metastasis in IHCC,50 nodal metastases, and poor survival.52, 53 The larger tumor microenvironment, including activated fibroblasts, macrophages, and T-regulatory cells, may also influence tumor growth. Hepatic stellate cells (resident myofibroblasts within the hepatic parenchyma) are associated with a poor prognosis in IHCC and enhance the growth of BTC cell lines in vitro.54, 55 Transforming growth factor beta (TGF-β) ligands are increased in IHCC, serving to both drive stromal proliferation and dampen immune response.56 TGF-β may also positively modulate angiogenesis through the transcriptional activation of VEGF.57 The role of tumor stroma and review of additional factors involved are discussed in further depth elsewhere.58

Targeted Therapy

Chemotherapy has been the main therapeutic modality in locally advanced or metastatic BTC. This has historically been loosely defined based on reports of activity of single and combinations of agents such as gemcitabine, cisplatin, oxaliplatin, capecitabine, and 5-fluorouracil (5-FU) in phase 2 trials.59 However, a recently published phase III study has established the combination of gemcitabine and cisplatin as the new global standard for locally advanced or metastatic BTCs. In a randomized phase III trial (ABC-02), Valle et al.60 demonstrated improved overall survival (OS) with gemcitabine and cisplatin compared to gemcitabine alone. The trial, which included 410 patients with both CC and GBC, demonstrated that the addition of cisplatin to gemcitabine afforded a significant progression-free survival (PFS) (median of 8.0 versus 5.0 months, P < 0.001) and OS benefit (median of 11.7 versus 8.1 months, hazard ratio, 0.64; 95% confidence interval [CI] 0.52 to 0.80; P < 0.001). However, the benefits of gemcitabine/cisplatin in perihilar CC remain to be defined.

Due to the practical reasons outlined above, development of targeted agents in BTCs has lagged behind other tumor types (Fig. 1). There are only a few completed phase II studies examining the early experience of efficacy and safety of a few classes of targeted agents in BTC (Table 2). These trials were conducted either as single agents, combined targeted agents, or in combination with chemotherapy regimens. All of these early-stage trials include patients with locally advanced or metastatic disease, both GBC and CC.

Table 2. Clinical Trials with Molecularly Targeted Therapies
Multiple Agents (First Line)
TreatmentTarget#RR (%) Refs.
  1. Abbreviations: GEMOX gemcitabine and oxaliplatin; RR: response rate; PFS: progression free survival; TTP: time to tumor progression; NA: not available.

    4-mo PFS 
GEMOX 50NA44%67
Vs.EGFR    
GEMOX- cetuximab 51 61% 
GEMOX- bevacizumabVEGF3540PFS 7 mo68
Single Agents (First and Second Line)
TreatmentTarget#RR (%)PFS (mo)Refs.
AZD6244MEK1/222145.472
ErlotinibEGFR437NA63
LapatinibEGFR/ HER2170NA64
SorafenibBRAF/ VEGFR366271
  4622.570
Combined Targeted Therapy
TreatmentTarget#RR (%)PFSRefs.
Erlotinib/ BevacizumabEGFR/ VEGF49124.4 mo73

Anti-EGFR Agents.

Emerging evidence has implicated EGFR as a potential therapeutic target in BTC. Amplification events and mutations leading to increased EGFR activity are found in subsets of BTC and inhibition of EGFR expression can modulate the growth BTC in animal models.20-22 Both small molecule inhibitors of the kinase domain and blocking antibodies of EGFR have been tested in BTC.22, 61, 62

Philip et al.63 conducted a phase II trial of 42 patients with BTC treated with single-agent erlotinib. This study demonstrated a 17% 6-month PFS rate, and three patients had partial responses (PRs) as determined by the Response Evaluation Criteria in Solid Tumors (RECIST). Of these patients, 57% had received first-line chemotherapy. In this study, EGFR mutation status was not tested, and therefore it is unknown if the response correlated with EGFR mutation status in BTC. The combination of erlotinib with gemcitabine and oxaliplatin (GEMOX) will soon be explored. Lapatinib, a dual EGFR1 and ERBB2 (HER-2) inhibitor, has been tested in 17 patients with BTCs.64 No responses were observed, although five had stable disease (SD). PFS and OS were only 1.8 (95% CI: 1.7-5.2) and 5.2 (95% CI 3.3-infinity) months, respectively.

The experience of cetuximab in BTCs has been reported from retrospective case series and phase II studies. Single-agent cetuximab activity was reported in a small case series of five patients in which four achieved radiologic responses.65 Paule et al.61 reported their early experience of adding cetuximab in patients with advanced IHCC who failed a GEMOX-based chemoregimen. Of the nine patients treated, one patient each achieved complete response (CR), PR, and SD, respectively. A single arm phase II study of combining cetuximab and GEMOX in the first-line therapy was initially reported by Gruenberger et al.66 Patients received cetuximab at 500 mg/m2 followed by 1,000 mg/m2 gemcitabine on day 1 and 100 mg/m2 oxaliplatin on day 2 biweekly as first-line therapy. Of the 30 patients enrolled, a response rate of 63% was reported, including three patients with CR. The median PFS was 8.3 months (95% CI 5.85-10.81) and median OS was 12.7 months (95 CI 7.96-17.37). Interestingly, the K-ras mutation status did not affect efficacy, although only three patients had confirmed K-ras mutation in this cohort. A randomized phase II study evaluating the impact of cetuximab when added to GEMOX among patients who did not receive prior chemotherapy suggests a benefit in terms of PFS rate at 4 months.67 In all, 101 randomized patients were stratified according to tumor stage, location, and prior treatments; 36 patients were included in an interim analysis. The 4-month PFS rate was 44% (95% CI, 20-70) and 61% (95% CI, 36-83) among those who received GEMOX and GEMOX-cetuximab, respectively. Toxicity was balanced between the arms with the exception of rash, which was more common among patients on cetuximab. The difference in PFS rate at 4 months observed at this early point appears promising. In a multi-institutional study we are currently examining the combination of panitumumab with GEMOX in patients with advanced BTC whose tumors harbor wildtype K-ras/B-raf genotypes (Table 3).

Table 3. Molecularly Targeted Therapies in Development
TreatmentTargetPhaseSponsoring InstitutionsClinicaltrials.gov ID & link
Oxaliplatin, Capecitabine & SorafenibBRAF/ VEGFRI/IIUniversity of Wisconsin, MadisonNCT00634751
Gemcitabine, Oxaliplatin (GEMOX)- SorafenibBRAF/ VEGFRI/IIUniversity of Miami Sylvester Comprehensive Cancer CenterNCT00955721
Erlotinib & DocetaxelEGFRIIHoosier Oncology GroupNCT00532441
BIBW 2992 in Cancers With EGFR and/or HER2 Gene AmplificationEGFR/ HER2IIMGH Cancer CenterNCT00748709
FOLFOX6 & BevacizumabVEGFIIGeorgetown UniversityNCT00881504
Gemcitabine, Cisplatin & SorafenibBRAF/ VEGFRIIMemorial Sloan-Kettering Cancer CenterNCT00919061
Gemcitabine, Capecitabine, & ZD6474VEGFR/EGFRI/IIUniversity of Colorado at Denver and Health Sciences CenterNCT00551096
ARRY-438162MEKI/IISarah Cannon Research Institute Nashville, TennesseeNCT00959127
Gemcitabine/ Cisplatin in combination with placebo or CediranibVEGFRII/IIIUniversity College, LondonNCT00939848
Gemcitabine,Oxaliplatin w/ ErlotinibEGFRIINew Mexico Cancer Care AllianceNCT00832637
Gemcitabine,Irinotecan w/ panitumumabEGFRIIUniversity of PennsylvaniaNCT00948935
Gemcitabine,Oxaliplatin w/ panitumumabEGFRIIUniversity of Rochester/ MGH Cancer Center/ DFCI 

Antiangiogenic Agents.

Numerous antiangiogenic agents are presently in use (sorafenib, sunitinib, bevacizumab) or in development. Most of these agents target VEGF signaling either at the level of ligand or receptor and come in the form of blocking antibodies and small molecule inhibitors. Several have been tested in BTC.

In a single-arm phase II study, Zhu et al.68 examined the efficacy and safety of bevacizumab, a humanized monoclonal antibody against VEGF, in combination with modified GEMOX in patients with advanced BTC. Patients received treatment intravenously on days 1 and 15 of a 28-day cycle: bevacizumab at 10 mg/kg, followed by gemcitabine 1,000 mg/m2 (10 mg/m2/min) and oxaliplatin 85 mg/m2. Of the 35 patients enrolled, 14 (40%) had a PR and an additional 10 (29%) had SD. The median OS was 12.7 months, and the median PFS was 7.0 months. Treatment was generally well tolerated and grade 3 and 4 toxicities were predictable and expected. Despite the encouraging results, due to the single-arm study design, patient selection bias, and well-known efficacy of GEMOX alone, the relative contribution of bevacizumab remains to be defined.

Sorafenib is an oral multikinase inhibitor that targets tumor-cell proliferation and tumor angiogenesis by inhibiting the serine-threonine kinases Raf-1 and B-Raf through the canonical Raf/mitogen-activated protein kinase/extracellular signal-regulated kinase (RAF/MEK/ERK) signaling pathway and the receptor tyrosine kinases (RTK) of VEGF receptor (VEGFR)-1, VEGFR-2, VEGFR-3), and platelet-derived growth factor receptor (PDGFR)-α and β and stem-cell factor receptor (KIT).69 It has been evaluated in two different phase II studies. A single-agent trial of sorafenib in advanced BTC was reported by Bengala et al.70 This study included 46 patients including 26 (56%) who had received prior chemotherapy. Sorafenib was administered at a dose of 400 mg twice a day. One patient had a PR (2%). The median PFS and OS were 2.3 and 4.4 months, respectively. The most common toxicities were skin rash (35%) and fatigue (33%) and 22% of patients required a dose reduction. Another phase II study of sorafenib as a single agent in 36 treatment-naïve patients reported a RR of 6%, PFS 2 months, and a median OS of 6 months.71 Despite the limited single-agent activity, sorafenib combinations, including a number of chemotherapy regimens (gemcitabine/cisplatin, GEMOX, gemcitabine/capecitabine), are currently under investigation (Table 3). Building on the success of ABC-02 trial, the British investigators are conducting a randomized phase II/III trial of combining either cediranib (AZD2171), a potent VEGFR inhibitor, or placebo, with gemcitabine/cisplatin for patients with advanced BTC (ABC-03).

MEK Inhibitor.

Selumetinib (AZD6244; ARRY-142886) is a tight-binding, uncompetitive inhibitor of mitogen-activated protein kinase kinases (MEK) 1 and 2 currently in clinical development. It has been tested in 29 patients with advanced BTC including 22 who were evaluable for response at the time of reporting.72 Of all the patients enrolled, 39% had received prior therapy. Three patients had responses (1 CR and 2 PRs) with a RR of 14%. Median PFS was 5.4 months and OS was 8.2 months. Correlation between clinical responses and p-ERK staining was observed. Genetic analysis of BRAF and KRAS mutation status is ongoing. Among single agents tested in a mixed population of pretreated and treatment-naïve patients, the initial results with a relatively high RR and long PFS are encouraging. Other MEK inhibitors (i.e., ARRY-438162) have entered clinical trials for BTC.

Combined Targeted Therapy.

Combining targeted agents that inhibit different pathways critical to cancer growth and survival represents an attractive strategy in oncology. The general principles of combining targeted agents include the selection of two agents with different mechanism of action, favorable pharmacokinetics profiles, and nonoverlapping toxicities. Lubner et al.73 examined bevacizumab in combination with erlotinib as first-line therapy in a phase II consortium study of 53 patients with advanced CC and GBC. Patients received bevacizumab at 5 mg/kg intravenously on days 1 and 15 and erlotinib 150 mg by mouth daily in a 28-day cycle. The primary endpoint of the trial was response rate per RECIST criteria. Of 49 evaluable patients, six (12%; 95% CI, 6%-27%) had a PR and additional 25 patients (51%) had SD. Rash was the most common grade 3 toxicity. Four patients had grade 4 toxicities. Median OS was 9.9 months, and time to tumor progression (TTP) was 4.4 months. Despite the encouraging results from this single-arm phase II study, how this regimen will be further developed in BTC is unknown.

Key Issues in Trial Design in Advanced BTC

As more targeted agents are entering clinical trials in advanced BTC, some relevant issues related to clinical trial design are emerging. For phase III trials the primary endpoint will continue to be overall survival. The trial will aim to provide a sufficient level of evidence of survival advantage over the standard therapy. Despite promising data on GEMOX and gemcitabine/capacitabine from phase II trials, the standard first-line systemic therapy for advanced BTC remains gemcitabine/cisplatin based on the recent randomized phase III ABC-02 study. Examining the efficacy of any targeted agents in the first line should consider adding the tested drugs to the gemcitabine/cisplatin backbone. Currently, there are no validated biomarkers to enrich the patient population to test a specific targeted agent in BTC, although this design has proven useful in other tumor types. As we gain further understanding about the molecular genetics of BTC, this strategy may become more attractive so we can assess the efficacy of targeted agents in an enriched population up front and compare it against standard chemotherapy based on the specific predictive molecular marker(s). For patients who failed gemcitabine/cisplatin, there are no standard treatments in this setting, although many clinicians will use a fluoropyrimidine-based regimen. Therefore, phase III studies with novel agents in the second-line setting should compare the efficacy of tested agents against best supportive care or a fluoropyrimidine-based regimen. Assessing the improvement of quality of life (QOL) would be important in this disease; however, we have no validated instrument to measure QOL in BTC. Due to the heterogeneity of BTC, a phase III trial should have adequate stratification for Eastern Cooperative Oncology Group (ECOG) performance status (0 versus 1-2), tumor burden (locally advanced versus metastatic), and specific tumor types (GBC versus CC). As we start to appreciate the molecular genetics differences between CC and GBC, it would be desirable to design trials specifically for either CC or GBC. In addition, the increased recognition of different histopathology and genetic signatures between intrahepatic, perihilar, and distal CC argues for designing site-specific trials or having stratification for these sites.

Most of the clinical trials conducted in advanced BTC are phase II studies. Due to the heterogeneity of BTC, as discussed above, it is very difficult to compare the efficacy signal from single-arm phase II trials. Therefore, randomized phase II trials will be favored. In this setting, the outcomes of the tested arm will be compared to a control arm, usually a standard treatment. Although some phase II studies are still using response rate as the primary endpoint, particularly those that aim to develop more effective preoperative regimens, future phase II studies should use TTP or PFS as the primary endpoints. This has become more relevant for the molecularly targeted agents, as most of these agents are cytostatic as opposed to cytotoxic.

In addition to use conventional imaging methods including computed tomography (CT) scans and magnetic resonance imaging (MRI) scans to assess tumor response with the standard RECIST criteria, we should make efforts identifying any potential surrogate markers to assess treatment response in advanced BTC. Unfortunately, the data in this area are very sparse. CA-19, a serum marker of glycoprotein, has been routinely collected in many trials with BTC. Despite a few small studies assessing the correlation of CA19-9 level change with clinical outcomes, this has not been validated in prospective study. 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) has proven helpful in several tumor types and change in FDG-PET standardized uptake value (SUV) has become a useful surrogate marker in assessing clinical response. In a single-arm trial of advanced BTC, we found that changes in FDG-PET SUV after two cycles of treatment with GEMOX in combination with bevacizumab correlated with PFS and OS.68 Future prospective studies should continue to explore the value of PET scan in BTC.

Along with the development of targeted agents in BTC, perhaps the most pressing need is to develop and validate potential predictive molecular markers. This effort holds promise of selecting the right patient population with the matched targeted agents. The most successful examples in targeted therapy development comes from exploiting “oncogene addiction” theory, i.e., using imatinib in chronic myeloid leukemia (CML) based on the presence of BCR-ABL translocation and in Gastrointestinal stromal tumors (GIST) based on the presence of KIT, using trastuzumab in breast and gastric cancers with HER-2 expression,28 using erlotinib/gefitinib in lung cancer with EGFR mutation,5 and more recently using PLX4032 in melanoma harboring B-RAF mutation (V600E)6 and crizotinib in lung adenocarcinomas based on the presence of ALK translocation.4 In BTCs, we should continue our efforts collecting tissue and blood specimens while conducting future trials in an attempt to identify any potential predictive markers.

Conclusions

Human BTCs represent a collection of distinct diseases that could be classified both by the anatomic location of the tumor's origin: gallbladder, extra- or intrahepatic ducts, and underlying genetics. This genetic heterogeneity can be mirrored by differing clinical behavior and responses to therapy.74 Future research should continue to assess the landscape of genetic changes in BTC and the relevance of these mutations in BTC pathogenesis. Improved animal models may help not only to define the biology and behavior of cancers with specific genetics, but may also provide insights into the optimum strategy for using targeted agents and thus providing the rationale for clinical trials. A newly defined standard of care of gemcitabine in combination with cisplatin in BTC provides a foundation on which targeted agents can be assessed.60 Despite the limited number of trials reported, early experience with targeted agents, particularly those targeting EGFR, angiogenesis, and MEK, has provided encouraging results. Identification of predictive markers including well-characterized genetic mutations holds promise to enrich specific populations of BTC for targeted therapy. Given the smaller number of patients with BTC as compared with other common solid tumors, coordination of trials among institutions and cooperative groups, both nationally and internationally, and optimizing trial design, will be the key to future progress.

Ancillary