Phase II study of hepatic intraarterial epirubicin and cisplatin, with systemic 5-fluorouracil in patients with unresectable biliary tract tumors




Patients with unresectable biliary tract carcinomas have a very poor prognosis. To improve the efficacy and tolerance of the ECF regimen (epirubicin at a dose of 50 mg/m2, cisplatin at a dose of 60 mg/m2, and 5-fluorouracil [5-FU] at a dose of 200 mg/m2 per day by continuous infusion), the authors designed a novel approach that combined locoregional and systemic chemotherapy with the same agents at the same dosages.


Thirty consecutive patients with advanced or metastatic biliary tumors were treated with epirubicin at a dose of 50 mg/m2 and cisplatin at a dose of 60 mg/m2 administered as a bolus in the hepatic artery on Day 1, combined with systemic continuous infusion of 5-FU at a dose of 200 mg/m2 per day, from Day 1 to Day 14, every 3 weeks.


Tumor sites were the intrahepatic bile ducts in 25 patients and the gallbladder in 5 patients. The overall response rate was 40% (12 of 30 patients), including 1 complete response and 11 partial responses. Stable disease was observed in 12 of 30 patients (40%) and progressive disease in 6 of 30 patients (20%). The median progression-free and overall survival periods were 7.1 and 13.2 months, respectively, and the 1-year and 2-year survival rates were 54% and 20%, respectively. Performance status improved in 9 of 30 patients (30%) and a weight gain of > 7% was observed in 4 of 30 patients (13%). The treatment was well tolerated with minimal hematologic toxicity. The major clinical problem was the deep venous thrombosis related to the central venous catheter, which occurred in 5 patients (17%).


This novel combined locoregional and systemic chemotherapeutic regimen was found to be active and safe for patients with advanced biliary tract carcinoma. Cancer 2005. © 2005 American Cancer Society.

Adenocarcinomas of the biliary tree are relatively rare tumors, with an extremely poor prognosis.1 They can be classified according to location. For example, gallbladder carcinoma is an aggressive disease with a median survival of 6 months, whereas cholangiocarcinoma is more indolent with a median survival of approximately 1 year.2–4 Radical resection, the only potentially curative treatment, is technically feasible in < 20% of patients. Moreover, even when surgery is performed, the 5-year survival rate is still only 20%.5, 6 The activity of systemic chemotherapy is very low.7 The most commonly used single agent in biliary tree tumors is 5-fluorouracil (5-FU), with response rates of 10–20%.8 When combined with 5-FU and doxorubicin, mitomicin C yielded response rates of 20–30%, with median survivals of 8–11 months.9 Cisplatin is efficient when administered in combination with continuous infusion and/or with a bolus of 5-FU, with objective response rates ranging from 10% to 35%.10 This combination has been tested in the ECF schedule, with epirubicin and continuous infusion of 5-FU. The response rate and the median duration of response were 40% and 10 months, respectively.11 Because these neoplasms are usually confined to the liver, hepatic arterial infusion might be a rational approach. The rationale of our study is strengthened by the high hepatic extraction on the first pass of some drugs that reach the bile canaliculi at high concentration and by the finding that the blood supply of the upper biliary tree and gallbladder derives from the hepatic artery.12–14 Our investigation combined systemic and intraarterial approaches. Consequently, we evaluated the activity of epirubicin and cisplatin administered through the hepatic artery, combined with systemic continuous infusion of 5-FU, according to the ECF regimen reported by Ellis et al.15


Eligibility Criteria

Between January 2000 and March 2004, 30 patients were enrolled. Patients were required to have histologically confirmed unresectable and measurable adenocarcinoma of the biliary tree or the gallbladder. Patients with intrahepatic and extrahepatic biliary tree carcinoma were referred to as having cholangiocarcinoma and gallbladder carcinoma, respectively. Other eligibility criteria were age ≥ 18 years, an Eastern Cooperative Oncology Group (ECOG) performance status of 0–2, absence of brain metastasis, adequate bone marrow function (leukocyte count > 3500 cells/μL, neutrophil count > 1500 cells/μL, and platelet count > 100,000/μL), hepatic function (serum bilirubin level < 3.0 mg/dL and aspartate aminotransferase level < 5 times the upper limit of normal), renal function (serum creatinine concentration < 2.0 mg/dL), and normal coagulation profile. The study was approved by an internal review board and written informed consent was obtained from all participating patients.

Pretreatment evaluation included a clinical examination (with survey of body weight, performance status, and clinical symptoms), an abdominal computed tomographic (CT) scan, a thoracic CT scan if lung and/or pleural metastases were suspected, and monitoring of tumor carbohydrate antigen (CA 19-9) level.

Toxicity and response were assessed in all patients.

Treatment Schedule

All patients received epirubicin at a dose of 50 mg/m2 in 100 mL of normal saline and cisplatin at a dose of 60 mg/m2 in 120 mL of normal saline (both infused as a bolus into the hepatic artery by an angiographic catheter inserted into the femoral artery with the Seldinger method), and 5-FU at a dose of 200 mg/m2 per day by intravenous continuous infusion through a central venous catheter from Day 1 to Day 14. The cisplatin infusion was preceded and followed by intravenous hydration and all patients were given prophylactic antiemetic treatment comprised of 5-hydroxytryptamine type 3 receptor antagonists and dexamethasone. Treatment was repeated every 3 weeks if the neutrophil count was > 1500 cells/μL, the platelet count was > 100,000/μL, and the serum creatinine concentration was < 2.0 mg/dL. Every cycle required 2 days of hospitalization. Response to treatment was evaluated every three cycles and patients who responded or had stable disease (SD) received three additional cycles. When disease progression occurred, treatment was discontinued and second-line chemotherapy was permitted.

Assessment of Response

Anticancer effects were evaluated by tumor size and CA 19-9 changes. All patients had bidimensionally measurable lesions for response evaluation. The tumor size was evaluated by CT scan every three cycles. Response was assessed by the World Health Organization criteria.16 Bone metastases and pleural and peritoneal effusions were not considered to be evaluable metastatic sites, but they were taken into account for the determination of disease progression. In addition, this trial also evaluated toxicities, response duration, and overall survival. Body weight, performance status, pain, and CA 19-9 serum levels were recorded at each cycle.

Toxicity was recorded according to the National Cancer Institute-Common Toxicity Criteria 2.0. Complete blood and platelet counts were obtained weekly to determine the level of myelosuppression. A complete chemistry panel, full blood count, and physical examination were performed before each cycle. Toxicity related to the angiographic procedure was also evaluated and recorded.

Statistical Analysis

The study was conducted employing Simon's two-stage design.17 The sample size was calculated to reject a 10% response rate in favor of a target response of 30%, with a significant level of 0.05 and a power of 80%. In the initial stage, 10 patients had to be entered and evaluated for response. If there was less than one response, accrual was stopped. If more than 1 response was observed in the first stage, then 19 additional patients had to be entered in the second stage to achieve a total of 29 evaluable patients. Response, progression-free survival, and survival were calculated from the date of the first treatment, using the Kaplan–Meier standard life table method.


Patients' characteristics are summarized in Table 1. The study sample included 25 patients with intrahepatic cholangiocarcinoma and 5 patients with gallbladder adenocarcinoma. The population consisted of 15 men and 15 women. The median age was 65.3 years and the median ECOG performance status was 1. Six patients underwent radical surgery, one of whom had received radiotherapy in an adjuvant setting. After 6 months, he developed local disease recurrence and demonstrated liver progression. No patient had received a previous chemotherapic treatment. The median value of CA 19-9 was 130 IU/mL.

Table 1. Patients Characteristics
CharacteristicsNo. of patients (%)
  • ECOG: Eastern Cooperative Oncology Group.

  • a

    “Others” included bone, pleural, and adrenal gland.

 Measurable disease30
 Evaluable for toxicity30
 Evaluable for response30
Median age (range) (yrs)(49–75 yrs)
ECOG performance status 
 012 (40)
 1 8 (27)
 210 (33)
Tumor diagnosis 
 Intrahepatic cholangiocarcinoma25 (83)
 Gallbladder carcinoma 5 (17)
Previous therapy 
 None23 (77)
 Surgery 6 (20)
 Adjuvant radiotherapy 1 (3)
Sites of metastases 
 Liver29 (97)
  Involvement < 50%18
  Involvement > 50%11
 Lymph nodes 4
 Peritoneum 5
 Local disease recurrence 2
 Othersa 4
Median CA 19-9 level (range) (IU/mL)130 (7–32,450)

The overall response rate was observed in 12 of 30 patients (40%), including 1 complete response (CR) as observed on a positron emission tomographic scan and 11 partial responses. SD was observed in 12 patients (40%) and progressive disease (PD) occurred in 6 patients (20%). The median progression-free survival period was 7.1 months (range, 1.6–19.1 months) and the median overall survival period was 13.2 months. The 1-year and 2-year survival rates were 54% and 20%, respectively. The 25 patients with intrahepatic cholangiocarcinoma had a median survival period of 13.5 months and the 5 patients with gallbladder adenocarcinoma had a median survival period of 8 months. The median survival periods of the responders and nonresponders were 15.5 and 8.5 months, respectively. Efficacy data are shown in Table 2.

Table 2. Best Response and Survival Data
Efficacy parameterPatients (n = 30) (%)
  1. PFS: progression-free survival; OS: overall survival.

 Complete 1 (3.3)
 Partial11 (36.6)
 Stable disease12 (40)
 Progressive disease 6 (20)
Median PFS (mos)7.1
Median OS (mos)13.2
1-yr survival rate54%
2-yr survival rate20%

The patient with a CR developed retroperitoneal lymph node recurrence 25 months after radical surgery and PD after 19 months. The patient was still alive at the time of last follow-up. Three patients underwent posttreatment laparotomy. The first patient received intraoperative radioablation after a major objective response and died after 14 months. The second patient underwent a radical right hepatectomy and demonstrated PD after 12 months. The third patient underwent intraoperative radioablation and died after 6 months. Of 17 patients with abnormal CA 19-9 values, 11 (69%) showed a reduction > 50% from baseline, whereas 3 patients developed PD. Ten patients received second-line chemotherapy. Six patients received oxaliplatin and irinotecan, three patients received capecitabine, and one patient received the systemic ECF regimen. Nine of 18 patients (50%) with an initial ECOG performance status > 0 improved during therapy. Pain improved in 10 of 16 initially symptomatic patients (63%). A weight gain of > 7% was observed in 4 patients (13%).


Overall, this regimen was well tolerated (Table 3), with a cumulative Grade 3 toxicity observed in 11 of 30 patients (37%). A total of 136 cycles of therapy were delivered, with a median of 4 cycles per patient (range, 2–8 cycles). No treatment-related deaths and no side effects related to the angiographic procedure were observed. The major clinical problem was the deep venous thrombosis due to the central venous catheter, which occurred in five patients. No patient required interruption of therapy because of this complication. As regards hematologic toxicity, Grade 3 leukopenia occurred in 1 patient. Grade 3 nausea/emesis occurred in 1 patient and Grade 3 mucositis occurred in 3 patients. After the second cycle, one patient developed acute pancreatitis that required hospitalization. The treatment was stopped because of a rapid progression of disease. Grade 3 alopecia occurred in 5 patients (17%).

Table 3. Treatment-Related Toxicity (n = 30)
ToxicityGrade 1a No. (%)Grade 2 No. (%)Grade 3 No. (%)Grade 4 No. (%)
  • CVC: central venous catheter.

  • a

    Toxicity was graded according to the National Cancer Institute-Common Toxicity Criteria 2.0.

 Neutropenia 6 (20)1 (3)1 (3)
 Thrombocytopenia 3 (10)
 Anemia 3 (10)5 (17)
 Stomatitis 1 (3)1 (3)3 (10)
 Nausea/emesis 8 (27)2 (7)1 (3)
 Diarrhea 1 (3)1 (3)
 Alopecia10 (33)8 (27)5 (17)
 Neurotoxicity 1 (3)1 (3)
 Hand-foot syndrome 2 (7)
 Hepatic 3 (10)2 (7)
 Renal 2 (7)1 (3)
Thrombosis related to CVC 5 (17)  


Patients with advanced biliary tract malignancies have a poor prognosis, with a median survival period ≤ 6 months in patients with gallbladder carcinoma and approximately 1 year in patients with cholangiocarcinoma.1 Surgical resection or liver transplantation represents the only curative treatment, but the median survival period of patients treated with these radical therapies is approximately 2 years only in most of the series.19, 20 Systemic chemotherapy can increase the quantity and improve the quality of life in patients with advanced biliary tract carcinomas.21 Regimens combining 5-FU and platin analogs are the most effective, with response rates ranging from 21% to 40%. The combination of cisplatin, epirubicin, and continuous infusion of 5-FU is considered one of the standard therapies in the treatment of biliary tract carcinoma.11 Even new drugs, such as paclitaxel and gemcitabine used as single agents, have shown response rates ranging from 0% to 16%, with a median survival period of approximately 6 months.22

A single experience with a combination of oxaliplatin and gemcitabine, given as a fixed-rate infusion, has shown encouraging response rates and median survival periods according to tumor location (e.g., gallbladder, extrahepatic bile ducts, ampulla of Vater, and intrahepatic bile ducts). When treated with this combination as a first-–line therapy, patients with performance status 0–2 and a bilirubin level < 2.5 times normal had a response rate of 36% with a median survival period of 15.4 months and a favorable toxicity profile.23

The rationale for the use of intraarterial chemotherapy can be summarized as follows. First, biliary tract carcinomas are usually confined to the liver and patients mainly die of liver failure. Second, some drugs have a high hepatic extraction after the first pass. Thus, they can reach, by intraarterial administration, bile canaliculi at high concentration with minimal systemic toxicity. Third, the blood supply of the upper biliary tree and gallbladder derives from the hepatic artery.24, 25 Administration of cisplatin through the hepatic artery provides a high drug concentration in the perfusing blood, whereas the systemic concentration is much lower.26 Then, the highest epirubicin concentration within the liver is obtained after bolus-arterial infusion, with an extraction rate of 60%.27 To our knowledge, there are only a few reports published to date concerning the effectiveness of hepatic arterial chemotherapy in these rare tumors. 5-FU has been the most frequently used agent. In addition, anthracycline derivatives and mitomycin C were provided either alone or in combination with 5-FU.14, 28, 29 Melichar et al.30 treated 32 patients with biliary tract carcinomas (i.e., 17 patients with cholangiocarcinoma and 15 patients with gallbladder carcinoma) with the combination of 5-FU, cisplatin, and folinic acid, administered through the hepatic artery. In comparison to patients treated with supportive measures or with systemic chemotherapy, this regimen showed good activity and improved survival.

In the current study, the combined systemic and intraarterial regimen shows an overall response rate of 40%, including 1 CR and 4 unresectable patients who were downstaged and underwent surgery and radioablation. Very mild toxicity (no Grade 4 toxicity was observed), and objective clinical response rates regarding pain, ECOG performance status, and weight of 63%, 56%, and 13%, respectively, were observed.

Comparing our ECF regimen with the original regimen reported by Ellis et al.11 and given systemically, we noted a very different toxicity setting with a more favorable profile for the intraarterial route. For example, in the study by Ellis et al., the rate of Grade 3–4 neutropenia was 16% with 13% of patients developing neutropenic infection compared with 3% of the current study patients without infections. In addition, the rate of Grade 3–4 thrombocytopenia was 19% versus 0% in the current study, the rate of Grade 3–4 diarrhea was 13% versus 3% in the current study, the rate of Grade 3–4 emesis was 6% versus 3% in the current study, and the rate of Grade 3–4 alopecia was 31% versus 17% in the current study.

The results of the current Phase II study confirms the feasibility and efficacy of combined systemic and intraarterial treatment in patients with unresectable biliary tract carcinomas. In addition, these results highlight the need for further trials and underscores the utility of guidelines in the management of a systemic central line because this represents the most common cause of morbidity.

Future development of a new generation of drugs is needed. Capecitabine might be a substitute for continuous infusion 5-FU to improve tolerability. Oxaliplatin might replace cisplatin in regimens utilizing intraarterial administration. Phase I and II studies of oxaliplatin administered through the hepatic artery have demonstrated feasibility and activity in patients with liver metastases from colorectal carcinoma.31, 32

The results of the current study are noteworthy and we believe they require further investigation with new drugs before the initiation of a Phase III trial