Biliary tract cancers include adenocarcinomas of the gallbladder and bile ducts (cholangiocarcinoma, both intra- and extrahepatic). Cancers of the biliary tree are relatively uncommon in North America, with approximately 8,000 new cases diagnosed in 2003. While surgical resection is potentially curative, only about 25% of patients are resectable at the time of presentation. Furthermore, there is a high rate of both local and systemic relapse after a potentially curative resection. Unresectable or metastatic biliary tract carcinoma carries a poor prognosis due to the lack of effective therapy. Patients with advanced gallbladder cancer have a median survival of approximately 6 months, while those with cholangiocarcinoma may have a more indolent course with a median survival duration of about 1 year. Several chemotherapy regimens have been explored in patients with advanced biliary cancers. However, the published series are small, and they often include a heterogeneous population of patients with both pancreatic and hepatocellular cancers as well as those with biliary cancers. Use of fluorouracil(Drug information on fluorouracil) (5-FU)-based regimens result in a response rates ranging from 10% to 30% but with limited response duration and short median survival (9 to 12 months). In a study using continuous infusion of 5-FU and recombinant human interferon alfa-2b(Drug information on interferon alfa-2b) (n = 32), Patt et al reported an objective response rate of 34% with median survival of 12 months. However, there was substantial toxicity with this regimen. In a subsequent study by the same group (n = 41), the addition of cisplatin(Drug information on cisplatin) and doxorubicin(Drug information on doxorubicin) to 5-FU and recombinant human interferon alfa- 2b resulted in a lower overall response rate (21%) with median survival of 14 months. Both gemcitabine(Drug information on gemcitabine) (Gemzar) and irinotecan(Drug information on irinotecan) (CPT-11, Camptosar) possess single-agent activity in biliary tract cancers. Although the available studies are small (Table 1), response rates with single-agent gemcitabine have consistently been in the range of 16% to 36% with median survival ranging from 6 to 11 months. Several patients treated with gemcitabine experienced prolonged periods of disease stabilization. Two studies using single-agent irinotecan have been reported in the literature. In one study (n = 25) using irinotecan at 125 mg/ m2 once a week for 4 consecutive weeks followed by a 2-week break, Sanz-Altamira et al reported an 8% partial response rate and 40% stable disease lasting at least 2 months. An ongoing study by the North Central Cancer Treatment Group (n = 13) showed substantial toxicity with irinotecan in patients with biliary cancers, while response rates were not reported. Preclinical studies show a dosedependent interaction between gemcitabine and irinotecan, with synergism at clinically relevant concentrations.[ 7,8] Based on the singleagent activity of both agents in biliary cancers, preclinical evidence of synergy, and the lack of overlapping toxicity, we tested the combination of gemcitabine and irinotecan in patients with advanced biliary tract cancers. The dose and schedule used in this study were based on a prior phase I study conducted by Rocha Lima et al. Patients And Methods Patients with previously untreated, unresectable histologically or cytologically documented cancer of the biliary tract (intrahepatic or extrahepatic) or gallbladder were eligible for participation. Patients were required to have measurable or evaluable disease and Eastern Cooperative Oncology Group (ECOG) performance status of 0 to 2. Adequate bone marrow function (granulocyte > 1,500/μL, platelets > 100,000/μL), hepatic function (serum bilirubin < 2.0 mg/dL), and renal function (serum creatinine < 2.0 mg/dL) were also required. Patients with an active second malignancy (other than nonmelanoma skin cancer) or serious medical or psychiatric disease were excluded. The study was approved by the local institutional review board at each participating institution, and informed consent was obtained from all patients. Treatment Plan Gemcitabine at 1,000 mg/m2 was given intravenously (IV) over 30 minutes followed by irinotecan at 100 mg/m2 IV over 90 minutes on days 1 and 8 every 21 days. In the absence of grade 2 or higher nonhematologic toxicity (except for nausea/ vomiting, alopecia, anorexia, and fever), or grade 3 or higher hematologic toxicity during cycle 1, the dose of irinotecan was escalated to 115 mg/ m2 for subsequent cycles. Standard premedication included dexamethasone(Drug information on dexamethasone) and a 5-HT3 receptor antagonist, while atropine was not given routinely unless patients experienced an episode of early diarrhea. All patients were instructed to take loperamide(Drug information on loperamide) at the earliest signs of diarrhea and/or abdominal cramping that occurred more than 24 hours after receiving irinotecan. Dose adjustments were made independently for gemcitabine and irinotecan depending on the types of toxicity observed (Table 2). Tumor measurements were performed every two cycles, and standard response criteria were used. Results Sixteen patients have been enrolled in this study (10 females and 6 males). Six patients had cholangiocarcinoma, and 10 patients had gall bladder cancer. The median age was 60 years (range: 45-76 years) and all patients had an ECOG performance status of 0 or 1. Extrahepatic metastases were present in five patients (Table 3). A total of 65 cycles of chemotherapy have been administered, with an average of 4.5 cycles per patient (range: 1-11 cycles). The median treatment duration of treatment was 3 months (range: 0.75-8 months). An objective partial response was determined radiographically in two patients (14%), while stable disease for periods ranging from 4 to 11.5 months was noted in six patients (43%). The median time to progression was 1.5 months (range: 1-11.5 mo). Patient accrual is ongoing, with a target of 25 patients. The major toxicity was gastrointestinal and myelosuppression. Two patients (14%) had grade 3 diarrhea and one patient had grade 3 nausea. Grade 3/4 neutropenia occurred in 7 patients (50%) with no episodes of febrile neutropenia, and grade 3/4 thrombocytopenia occurred in four patients (28%). There were 2 deaths during treatment one attributed to bowel obstruction secondary to disease progression, and one due to pneumonia without neutropenia. Discussion and Conclusion The combination of gemcitabine and irinotecan represents a new option in the treatment of biliary tract cancers. The activity of gemcitabine in biliary tract cancers was suggested in case reports several years ago. Castro reported a single patient with hepatic and intraperitoneal metastases from gall bladder cancer who had failed two other chemotherapeutic regimens. He had dramatic response to single-agent gemcitabine with complete reversal of small bowel obstruction and the disappearance of hepatic metastases. In one of the earliest trials, Mezger et al treated 13 patients with gallbladder and biliary tract with gemcitabine at 1,000 mg/m2 weekly for 7 weeks, then weekly for 3 out of 4 weeks until disease progression. Although only one patient had a partial response, 11 (85%) patients had stable disease with a median time to progression of 7 months and median overall survival of 11 months. Several subsequent studies and case reports have demonstrated consistent activity of gemcitabine in biliary tract cancers. Single-agent irinotecan has modest clinical activity in biliary tract cancers. Preclinical studies evaluating combinations of anticancer drugs report a dose-dependent interaction between gemcitabine and irinotecan. In vitro studies show antagonism at low concentrations (< 0.1 μM) , but synergism at concentrations of gemcitabine above 0.1 μM and irinotecan above 3.2μM in the SCOG small-cell lung cancer cell line. Absolute, marked synergism was evident in the HL-60 acute myeloid leukemia cell line. Synergism at concentrations of 0.1 to 2 μM gemcitabine and 0.2 to 10 μM irinotecan, but antagonism at high concentrations (ie, concentrations > 2 μM gemcitabine and 20 μM irinotecan), was seen in MCF7 breast cancer cells. Phase I studies evaluating the combination of gemcitabine and irinotecan on a day 1 and 8 schedule reported a maximum tolerated dose of 1,000 mg/m2 gemcitabine and 100 mg/m2 irinotecan. Escalation of irinotecan to 115 mg/m2 was recommended for subsequent cycles in patients with minimal or no toxicity during the first cycle. The sequence of administration of gemcitabine followed by irinotecan was empirical. No preclinical data were available to suggest sequence- related differences in toxicity or efficacy. This combination was subsequently tested in pancreatic and other cancers and was demonstrated to be clinically active with manageable toxicity. While a previous phase II study showed improvement in response rate with the combination of gemcitabine and irinotecan in pancreatic cancer, a subsequent phase III randomized study failed to show any improvement in survival with this combination compared with singleagent gemcitabine. In summary, our phase II trial suggests that the combination of gemcitabine and irinotecan has modest clinical activity and is well tolerated in patients with advanced or metastatic biliary cancer. Patient accrual is ongoing to this study.