Patients included in the study were required to be 18~75 years old with a confirmed histologically, metastatic or recurrent colorectal cancer, and who had not received any prior chemotherapy. Adjuvant treatment, if administered, should have been completed at least six months before being enrolled. Other inclusion criteria were as follows: (i) a performance status of 0-2 on the Eastern Cooperative Oncology Group (ECOG); (ii) a bi-dimensionally measurable lesion, with a metastatic lesion or lymphadenopathy greater than 10 mm; (iii) an adequate bone marrow reserve (WBC count >4,000/µL, platelet count >100,000/µL); and (iv) a normal cardiac function. (v) normal renal function (serum creatinine <1.5 times the upper normal limit (UNL)); (vi) an adequate hepatic function (AST and ALT <2.5 times UNL and total bilirubin <1.5 times UNL; patients with obstructive jaundice should have undertaken internal or external biliary drainage before enrollment); and (vii) an estimated life expectancy of at least 12 weeks at the time of enrollment. All approved patients or legal representatives signed informed consent forms before treatment began.

All patients received chemotherapy under the following schedule: 180 mg/m² of irinotecan intravenously (i.v) on Day 1, 400 mg/m² of 5-FU via i.v bolus with 600 mg/m² of continuous infusion for 22 hrs on Day 1 and 2, and 20 mg/m² of leucovorin on Day 1 and 2, every two weeks. Treatment was continued until the progressive disease, unacceptable toxicity, or withdrawal of the patients' consent.

Prior to each treatment, we recorded the patient's history and performed physical examinations, CBC (complete blood counts), blood chemistry, and conducted a urine analysis. If the absolute neutrophil count was less than 1,500/µL or the platelet count was less than 100,000/µL, treatment was delayed for one week or until myelosuppression was resolved. If a patient was not fully recovered after seven days, the dose was reduced by 25%. If Grade 3/4 non-hematologic toxicity developed, the dose was decreased by 25%. If there was a second episode of Grade 3/4 non-hematologic toxicity, treatment was stopped.

Physical examinations, peripheral blood count, blood chemistry, and chest X-ray were performed during each patient visit. CT scans were performed at the baseline and after four treatment courses. Response to treatment was assessed according to World Health Organization (WHO) criteria. Before each cycle, patients were assessed for toxicity according to the National Cancer Institute common toxicity criteria.

Time to progression (TTP) was defined from the first day of chemotherapy to the day of confirmed progression. Overall survival (OS) was calculated from the first day of chemotherapy to death or final follow-up. All data were analyzed using the Kaplan-Meier method using the SPSS program (version 11.0).

A total of 44 patients (32 men and 12 women) diagnosed between January 2002 and October 2004 at Kangnam St. Mary's Hospital and St. Mary's Hospital were enrolled in this study. The median age was 58 years (range: 27-72 years). Twenty-two patients had recurrent cancer after curative surgery, while 22 patients had metastatic diseases at the initial diagnosis. Nineteen patients had received adjuvant treatment with 5-FU and LV or 5-FU chemoradiation. At baseline, all patients had and ECOG performance between 0 and 2. Patients' characteristics are summarized in Table 1. Most of the patients received chemotherapy with oxaliplatin containing a regimen after diagnosis of progressive disease.

Of the 44 patients enrolled, 4 patients were lost to follow-up after the first cycle. One patient refused all anti-cancer therapy, and two patients could not withstand the irinotecan-containing chemotherapy. An additional patient, who experienced Grade 4 neutropenic infection during the 1^st chemotherapy cycle, changed their chemotherapy regimen.

The total number of cycles was 288, and the median number of cycles was six (range, 1-16). The overall response rate was 47.8% (95% Confidence interval; 32.5~63.2%) (Table 2). Of the 44 patients, one had CR (2.3%), while 20 had PR (45.5%). Of the 40 evaluated patients, the response rate was 52.5% (95% Confidence interval; 37.2~67.8%). There was no significant difference in response rates associated with previous chemotherapy history (Table 3) (p=0.054). The median follow-up for surviving patients was 53 weeks (23-121 weeks). The median time to progression was 23 weeks (7-57 weeks) while median survival was 47 weeks (18-121 weeks) (Figure 1).

A total of 288 cycles for 44 patients were evaluated for toxicity, and the toxicity incidence is summarized in Table 4. One patient required a 25% dose reduction due to delayed bone marrow suppression, and two patients required a delay of treatment for one week due to non-recovery from bone marrow suppression. Three patients experienced a total of 23 episodes of Grade 3/4 leukopenia, but no Grade 3/4 thrombocytopenia was observed (Table 4). Two patients experienced Grade 3/4 neutropenic infections without septic shock. No treatment-related deaths occurred.

The most common non-hematologic toxicities were diarrhea, nausea and vomiting. But only one patient developed Grade 3/4 diarrhea, and one patient developed Grade 3 vomiting. Three episodes of Grade 4 mucositis were observed in one patient. No treatment-related deaths occurred during any of the chemotherapy cycles.