Original article. Introduction

Transcription

1 Original article Annals of Oncology 14: , 2003 DOI: /annonc/mdg288 A randomized phase II trial of irinotecan in combination with infusional or two different bolus 5-fluorouracil and folinic acid regimens as first-line therapy for advanced colorectal cancer K. Bouzid 1 *, S. Khalfallah 2, J. Tujakowski 3, B. Piko 4, G. Purkalne 5, S. Plate 5, P. Padrik 6, M. Serafy 7, E. M. Pshevloutsky 8 & B. Boussard 9 On behalf of the Irinotecan Study Group 1 EHS Centre Pierre et Marie Curie, Algiers, Algeria; 2 Institue Salah Azaiz Bab Saadoun, Tunis, Tunisia; 3 Regionalne Centrum Oncologii, Bydgoszcz, Poland; 4 Pandy K Bekescountry County Hospital, Gyula, Hungary; 5 Oncology Center of Latvia, Riga, Latvia; 6 Clinicum of the University of Tartu, Tartu, Estonia; 7 National Cancer Institute, Cairo, Egypt; 8 Oncology Dispanser, Omsk, Russia; 9 Global Medical Affairs Oncology, Aventis Pharma, Antony, France Received 18 October 2002; revised 13 February 2003; accepted 17 March 2003 Introduction Background: Three different therapeutic regimens of irinotecan (CPT-11) in combination with 5-fluorouracil (5-FU) and folinic acid (FA) were evaluated for efficacy and safety in the first-line therapy of advanced colorectal cancer. Patients and methods: Patients were randomly assigned to receive intravenously either: CPT mg/m 2, FA 20 mg/m 2 followed by 5-FU 500 mg/m 2 bolus, weekly for 4 weeks (arm A, Saltz regimen); or CPT mg/m 2 day 1 then FA 200 mg/m 2 over 2 h and 5-FU 400 mg/m 2 bolus and 5-FU 600 mg/m 2 22-h infusion on days 1 and 2, every 2 weeks (arm B, Douillard regimen); or CPT mg/m 2 (days 1 and 43) alternating with FA 20 mg/m 2 /day followed by 5-FU bolus 425 mg/m 2 /day during 5 days (days 22 26) (arm C, Mayo Clinic regimen). Results: A total of 154 patients were included in the study (arm A, 51 patients; arm B, 53; arm C, 50). Overall response rates for the intention-to-treat populations were 33% [95% confidence interval (CI) 21% to 48%], 42% (95% CI 28% to 56%) and 30% (95% CI 18% to 45%) for arms A, B and C, respectively. Median times to progression were 6, 8 and 7 months for arms A, B and C, respectively. Median survival times were 15, 12 and 17 months for arms A, B and C, respectively. Overall response rates for the evaluable patient populations were 40% (95% CI 24% to 58%) in arm A, 44% (95% CI 29% to 60%) in arm B and 31% (95% CI 17% to 47%) in arm C. Neutropenia was the main serious adverse event in arms A (30% of patients) and C (22% of patients) but occurred in only 8% of patients in arm B. Delayed diarrhea was the main severe adverse event for the three regimens, from 15% to 22%. Conclusion: All three regimens were highly active. The biweekly combination of CPT-11 and 5-FU/FA (arm B) was notable for its low incidence of grade 3/4 neutropenia. The incidence of grade 3/4 delayed diarrhea was equivalent for the three treatment arms. Key words: colorectal cancer, first-line, 5-fluorouracil/folinic acid, irinotecan 5-Fluorouracil (5-FU)-based chemotherapy has been the mainstay of treatment for advanced colorectal cancer for the last three decades [1]. When given alone, as an intravenous (i.v.) bolus, once weekly or for five consecutive days every 4 5 weeks, 5-FU produces response rates of between 11% and 17% and a median survival time of approximately 1 year [2]. A number of biochemical modulators have been used in combination with 5-FU in an attempt to improve its efficacy while maintaining acceptable *Correspondence to: Professor. K. Bouzid, EHS Centre Pierre et Marie Curie, Avenue Salem Bouzenad, Algiers 16005, Algeria. Tel: ; Fax: ; kamelbouzidz@yahoo.fr Additional members of the Irinotecan Study Group are listed in the Acknowledgements. toxicity. The most successful of these to date is folinic acid (FA), and regimens based on 5-FU and FA are standard in the majority of institutions worldwide [3]. Irinotecan (CPT-11) (7-ethyl-10-[4(-1-piperidino)-1-piperidino] carbonyloxy camptothecin), a semisynthetic derivative of the natural alkaloid camptothecin, was introduced into the clinic in the late 1980s. CPT-11 belongs to a new class of antineoplastic agents, the topoisomerase I-interactive compounds, which interfere with DNA replication and cell division [4]. Clinical studies with CPT-11 have demonstrated tolerable side-effects and a broad spectrum of efficacy against solid tumors, including colorectal cancer [5]. As first-line chemotherapy in metastatic colorectal cancer, single-agent CPT-11 produced a combined response rate of 26% [95% confidence interval (CI) 20% to 32%)], a median time to progression (TTP) of 8 9 months and a median survival time of 12 months [6 9]. Thus, single-agent CPT-11 demonstrates 2003 European Society for Medical Oncology

2 1107 an antitumor efficacy comparable to that achievable with standard FA-modulated 5-FU-based regimens. Furthermore, CPT-11 has demonstrated promising antitumor activity in patients with 5-FU-refractory colon cancer, producing response rates in the range of 13 23% and median times to progression of 6 8 months [9 11]. In randomized phase III trials, second-line single-agent CPT-11 extended survival significantly compared with either best supportive care [12] or infusional 5-FU and FA [13]. The most frequent adverse events associated with CPT-11 include neutropenia, delayed diarrhea, acute cholinergic syndrome, alopecia, fatigue and nausea/vomiting [9]. Based on the promising single-agent activity of CPT-11 in the treatment of colorectal cancer, it has been investigated in combination with different 5-FU-based regimens. Interesting results in terms of both efficacy and safety have been reported for CPT-11 in combination with either standard bolus or an alternating bolus 5-FU/FA schedule, or with a continuous infusional high-dose 5-FU/FA schedule as first-line therapy [14 18]. Two international (North American and European) randomized phase III trials have confirmed the efficacy of CPT-11 combined with the most frequently used 5-FU/FA bolus and infusional regimens compared with the corresponding 5-FU/FA alone [19, 20]. The combination arms of both trials demonstrated a significant superiority in terms of efficacy (response rate, median TTP and median survival time), compared with 5-FU/FA alone [19, 20]. As a result, CPT-11 in combination with either bolus 5-FU/FA or an regimen has been approved for the first-line treatment of patients with advanced or metastatic colorectal cancer in both the United States and Europe. Several studies are now ongoing in the United States and Europe to assess the efficacy and safety of the different CPT-11/5-FU/FA combinations. The present phase II randomized study of CPT-11 combined with either an infusional or bolus 5-FU/FA or alternating bolus 5-FU/FA regimen, was designed to assess the response rates of three different, but established, therapeutic CPT-11/5-FU/FA regimens in the same trial. The safety profile, median TTP and survival data were also evaluated. Patients and methods Patient eligibility Patients with histologically proven, measurable, advanced adenocarcinoma of the colon or rectum, with no potentially resectable metastases were eligible for the study. The other inclusion criteria were: age between 18 and 75 years; World Health Organization (WHO) performance status (PS) 0 2; a life expectancy >3 months; adequate hematological function (hemoglobin 10 g/dl, neutrophils /l and platelets /l); satisfactory hepatic and renal function [total bilirubin 1.25 upper normal limit (UNL), creatinine 1.25 UNL, transaminases 3 UNL; or in case of liver metastases: total bilirubin 1.5 UNL, aspartate and alanine aminotransferases 5 UNL]. Patients had to have received no prior chemotherapy, or only (neo)adjuvant chemotherapy which had ceased >6 months before randomization, and no radiotherapy within the 4 weeks before study entry. Specific exclusion criteria included a tumor type other than metastatic colorectal adenocarcinoma, except for curatively treated non-melanoma skin cancer or in situ carcinoma of the cervix, evidence of brain metastasis, current infection, unresolved bowel obstruction or subobstruction, uncontrolled Crohn s disease or ulcerative colitis, or a current history of chronic diarrhea. This study was approved by the local ethical committees of each center and was conducted in accordance with the Declaration of Helsinki. All patients provided written informed consent. Pretreatment baseline evaluations included a complete medical history and physical examination, complete blood cell count and blood chemistry, an ECG (electrocardiography), complete tumor imaging (X-ray, computed tomography, ultrasound) and an assessment of tumor markers. Treatment CPT-11 (Campto ) was supplied by Aventis Pharma (Antony, France) as a sterile solution of 20 mg/ml in 5 ml vials. Patients were randomly assigned to receive one of three regimens: arm A, consisting of CPT mg/m 2 administered i.v. over min, FA 20 mg/m 2 administered i.v. over 15 min immediately after completion of the CPT-11 infusion and 5-FU 500 mg/m 2 administered as an i.v. bolus immediately following FA, weekly for 4 weeks on days 1, 8, 15 and 22, every 6 weeks (Saltz regimen); arm B, consisting of CPT mg/m 2 administered i.v. over min on day 1, followed immediately by FA 200 mg/m 2 administered i.v. over 2 h followed immediately by 5-FU 400 mg/m 2 administered as an i.v. bolus and 5-FU 600 mg/m 2 as a 22-h i.v. infusion on days 1 and 2, every 2 weeks (Douillard regimen); or arm C, consisting of CPT mg/m 2 administered i.v. over min on day 1 and day 43 (6 weeks) alternating with FA 20 mg/m 2 /day followed by 5-FU 425 mg/m 2 /day administered as an i.v. bolus daily for 5 days (days 22 26) (Mayo Clinic regimen). One cycle corresponded to 6 weeks for each treatment arm. Treatment was continued until disease progression, unacceptable toxicity or withdrawal of patient consent. A cycle was considered to be delayed if the cycle duration was >6 weeks. If this duration was >8 weeks the patient was taken off study. A 20% reduction in the dose of CPT-11 and 5-FU was made in the case of grade 3/4 hematological and non-hematological toxicities, for the three arms. Concomitant medication included subcutaneous atropine (0.25 mg) as a curative treatment for severe cholinergic symptoms, including early diarrhea, which could be continued in subsequent cycles with prophylactic intent, and oral loperamide, 2 mg every 2 h, for at least 12 h as soon as the first liquid stool occurred and up to 12 h after the last liquid stool without exceeding a total treatment duration of 48 h, and oral rehydration for the curative treatment of delayed diarrhea. If diarrhea persisted for >48 h despite loperamide treatment, or in the case of severe diarrhea or diarrhea associated with vomiting, fever or severe neutropenia, prophylactic broad-spectrum oral antibiotics, with fluoroquinolone or cotrimoxazole, had to be administered for 7 days and the patient had to be hospitalized for rehydration. Patients with febrile neutropenia had to be hospitalized to receive i.v. antibiotics. No prophylactic antiemetics were allowed on day 1 of treatment, but could be used at the discretion of the investigator in subsequent cycles. Growth factor support was administered at the discretion of the investigator. Response and toxicity evaluation Tumor response was radiologically assessed every 12 weeks (or two cycles of treatment) according to WHO criteria [complete response (CR), partial response (PR), minor response, stable disease (SD) and progressive disease (PD)]. The overall response rate determined by the investigators was defined as the percentage of patients with a CR or PR. The period for CR lasted from the date the CR was achieved to the date thereafter on which PD was first noted. The period of overall response lasted from the first day of treatment to the date of first observation of PD. The duration of SD was taken from the date that SD was first recorded until the date thereafter when PD was first recorded. TTP was the time measured from the day of randomization to the first progression or death due to malignant disease. Toxicity, graded according to National Cancer Institute (NCI) Common Toxicity Criteria, was assessed before each infusion by clinical and biological examinations (before each infusion for hematology and every 6 weeks for biochemistry).

3 1108 Statistical analysis According to the single-stage Fleming design, 45 patients needed to be recruited into each arm and the sample size was not calculated for comparative analysis. The analyses were performed with SAS Software version 8.0 for Windows and S-plus V 4.5. The primary end point of the study was to evaluate the response rate of each combination, calculated with the 95% CI. The secondary efficacy criteria were the duration of response and of stabilization, TTP and survival. The times to event were estimated by the Kaplan Meier method. Efficacy analyses were performed for both the intention-to-treat (ITT) and evaluable populations (eligible patients having received at least one complete cycle of treatment, except for patients whose disease had progressed before the end of the cycle, and same method of radiological assessment during the study). The survival was evaluated for the ITT populations only. Safety analyses were performed for all patients who had received one complete cycle of CPT-11/5-FU/FA. Results Patient characteristics Between April 1998 and July 2000, 154 patients (51 patients in arm A, 53 in arm B and 50 in arm C) were randomly assigned and treated in the study in 28 centers from 16 countries (Eastern Europe, Middle East, Asia and North Africa). Sixteen patients (10%) were ineligible due to the absence of a measurable lesion (n = 3), no adequate hematological/biochemical function (n = 5), no complete initial work-up (n = 6) or first treatment not being within 8 days of randomization (n = 2). Table 1 shows the baseline characteristics of the patients, all of which were balanced across the treatment groups except for the proportion of men, which appeared greater in arm B (64% versus 43% in arm A and 52% in arm C) and for prior radiotherapy and/or adjuvant chemotherapy, which appeared more frequently in arm C (prior radiotherapy in 26% of patients versus 18% in arm A and 15% in arm B, prior adjuvant chemotherapy in 36% of patients versus 22% in arm A and 13% in arm B). However, the median number of cycles of previous adjuvant chemotherapy was similar for the three treatment groups (median of six, range from one to nine in arm A, from three to six in arm B, and from two to 26 in arm C). Approximately 80% of patients had undergone surgery before inclusion. The median age was >50 years for all treatment arms. More than 86% of the patients had a WHO PS of 0 1 at baseline. The majority of patients had one organ involved, with the liver being the most common site of metastatic disease. Dose intensity The median number of cycles administered was similar for the three treatment arms (Table 2). A total of 203 cycles were administered in arm A (weekly schedule), 222 cycles in arm B (biweekly schedule) and 208 cycles in arm C (alternating schedule). The median relative dose intensities (ratio of actual dose intensity to planned dose intensity) of CPT-11 and 5-FU were higher in the alternating schedule (94.8% and 93.2%, respectively) than in either the weekly combination of CPT-11 with bolus 5-FU/FA (83.2% and 87.5%, respectively) or the biweekly combination of CPT-11 with bolus and (85.7% and 72.7%, respectively). Approximately one-quarter of the cycles had to be delayed in each treatment arm. The most frequent reason for this was hematological toxicity in arm A, and non-study-drug-related adverse events in arms B and C. The dose had to be reduced in 8% of cycles in arm A, 6% of cycles in arm B and 13% of cycles in arm C. The most frequent reasons for this were hematological toxicity in arm A and non-hematological toxicity related to study drug in arms B and C (Table 2). Efficacy results Of the 138 eligible patients, 18 could not be assessed for response due either to deviation from the method of measurement used at baseline (17 patients) or to failure to administer a full cycle of treatment (one patient), leading to a total of 120 evaluable patients. The overall response rates were 40% (95% CI 24% to 58%) in arm A (weekly regimen), 44% (95% CI 29% to 60%) in arm B (biweekly regimen) and 31% (95% CI 17% to 47%) in arm C (alternating regimen) for the evaluable patient populations (Table 3). CRs were observed in 9% of arm A patients and 7% of arm B patients, but no CRs were reported for the alternating schedule, arm C. Similar results were obtained for the ITT populations with best overall response rates of 33% (95% CI 21% to 48%), 42% (95% CI 28% to 56%) and 30% (95% CI 18% to 45%) in arms A, B and C, respectively. The median durations of response were 8.5, 11.0 and 12.1 months in arms A, B and C, respectively for the evaluable patient population. The median durations of stabilization were 5.5, 2.7 and 5.7 months for arms A, B and C, respectively, evaluable populations (Table 3). The median TTPs were 5.8, 8.0 and 7.3 months for arms A, B and C, respectively. The Kaplan Meier estimates of TTP for the three patient groups are shown in Figure 1. The median survival times were 15.2, 12.4 and 17.0 months for arms A, B and C, respectively, ITT population (Table 3). The Kaplan Meier survival curves are shown in Figure 2. Safety results All 154 patients were evaluable for hematological and nonhematological toxicity. Hematological toxicity. Hematological toxicity (worst grade) by patient is reported in Table 4. Neutropenia was the main severe adverse event in arm A (weekly CPT-11 plus bolus 5-FU/FA regimen) and arm C (alternating regimen). Grade 3/4 toxicity occurred in 30% of patients for arm A and in 22% of patients for arm C. The incidence of neutropenia was lower in arm B patients (biweekly CPT-11 plus bolus and regimen) with the occurrence of grade 3/4 toxicity in 8% of patients. Febrile neutropenia only occurred in arm A patients (6% of patients). Grade 3/4 anemia occurred in 8% of patients in arm A, 2% of patients in arm B and 4% of patients in arm C. Grade 4 thrombocytopenia was rare, and was only reported in 2% of patients in arms A and C (and for no patients in arm B). Non-hematological toxicities. The most common non-hematological adverse events (all grades) that were possibly or probably related to the treatment were delayed diarrhea (66 78% of

4 1109 Table 1. Patient characteristics (n = 154) Characteristics Arm A, CPT-11 + bolus 5-FU/FA (n = 51) CPT-11 and 5-FU/FA (n = 50) Treated/evaluable (%) 51/35 (69%) 53/43 (81%) 50/42 (84%) Median age, years (range) 57 (20 70) 56 (24 72) 52 (23 69) Sex Male 22 (43%) 34 (64%) 26 (52%) Female 29 (57%) 19 (36%) 24 (48%) WHO performance status 0 17 (33%) 15 (28%) 16 (32%) 1 28 (55%) 31 (58%) 31 (62%) 2 6 (12%) 7 (13%) 3 (6%) Primary tumor site Colon 16 (31%) 21 (40%) 21 (42%) Rectum 21 (41%) 14 (26%) 16 (32%) Colon rectosigmoid 14 (27%) 18 (34%) 13 (26%) Organ involved Liver 38 (75%) 44 (83%) 32 (64%) Lung 12 (24%) 5 (9%) 8 (16%) Lymph node 13 (25%) 13 (25%) 13 (26%) Bone 0 2 (4%) 1 (2%) Skin 1 (2%) 0 0 Other 11 (22%) 4 (8%) 12 (24%) Other soft tissue 7 (14%) 8 (15%) 7 (14%) Median number (range) 1 (1 3) 1 (1 4) 1 (1 3) Median time from first metastasis diagnosis, months (range) 1 (0 9.8) 0.9 (0 14.6) 1.2 (0 19.4) Prior therapy Surgery 40 (78%) 43 (81%) 40 (80%) Radiotherapy 9 (18%) 8 (15%) 13 (26%) Adjuvant therapy 11 (22%) 7 (13%) 18 (36%) Radiotherapy + surgery 8 (16%) 7 (13%) 12 (24%) Radiotherapy + adjuvant therapy 6 (12%) 5 (9%) 9 (18%) Surgery + adjuvant therapy 11 (22%) 7 (13%) 18 (36%) Surgery + adjuvant therapy + radiotherapy 6 (12%) 5 (9%) 9 (18%) patients), alopecia (53 60% of patients), nausea (57 62% of patients), vomiting (45 58% of patients), anorexia (30 39% of patients), stomatitis (20 34% of patients) and cholinergic syndrome (20 32% of patients). Grade 3/4 NCI adverse events per patient (Table 5) were delayed diarrhea (18% versus 15% versus 22% of patients in arms A, B and C, respectively), grade 3/4 nausea/vomiting (equivalent in the three arms at 7 8%), grade 3/4 stomatitis (6% in arms A and C), grade 3/4 infection (low incidence of 6% in arm A, 2% in arm B, none in arm C), and low incidence of grade 3/4 cholinergic syndrome in the three arms. Grade 2 alopecia occurred in 30% versus 40% versus 54% of patients in arms A, B and C, respectively. Hepatic and renal tolerance was good. Patient withdrawal and deaths. Patients were withdrawn from the study due mainly to disease progression (47% versus 60% versus 62% of patients in arms A, B and C, respectively), and withdrawal of patient consent (20% versus 13% versus 14% of patients in arms A, B and C, respectively). At the time of analysis, 58 patients were alive. The majority of deaths were due to malignant disease, but three deaths were related to the study treatment. There were two patients in arm A, one 70-year-old patient who died of septic

5 1110 Table 2. Exposure to study medication shock 6 days after the last infusion of the first cycle and one 42-year-old patient who died of hypokalemia plus hyponatremia 3 days after the last infusion of the first cycle. The third treatmentrelated death, in arm C, was a 42-year-old patient who died 16 days after the last infusion of the fifth cycle of therapy from a cardiac arrest following grade 4 fever with infection. There were six deaths (4%) across all three treatment arms within the first 60 days of therapy. Four early deaths were reported for arm A. Two of these were considered to be treatment related as described above, and two unrelated to treatment due to malignant disease. Two early deaths (cerebral stroke, malignant disease) were also reported for arm B but both were unrelated to treatment (Table 6). Discussion Cycles Arm A, CPT-11 + bolus 5-FU/FA (n = 51) The primary objective of this multicenter open phase II study was to evaluate the efficacy of three different therapeutic regimens in which CPT-11 was combined with concomitant or alternated bolus and/or for the first-line therapy of advanced colorectal cancer. The weekly combination of CPT mg/m 2, FA 20 mg/m 2 and bolus 5-FU 500 mg/m 2 evaluated in arm A was previously shown to be feasible and well-tolerated in a phase I trial reported by Saltz et al. [14]. Neutropenia was the major dose-limiting toxicity and diarrhea was common, but rarely dose-limiting [14]. The efficacy of this regimen was confirmed by Saltz et al. [19] in their pivotal North American phase III trial that Number Median (range) 4 (1 8) 4 (1 10) 4 (1 9) Percent relative dose intensity, (median and range) CPT-11 and 5-FU/FA (n = 50) CPT (48 111) 85.7 (34 102) 94.8 (66 103) 5-FU 87.5 (49 111) 72.7 (31 100) 93.2 (52 107) FA 91.9 (49 124) 70.5 (31 149) 92.2 (52 110) No. of delayed cycles 48 (24%) 60 (27%) 49 (24%) Reason for delay Hematological 23 (48%) 8 (13%) 14 (29%) Non-hematological 7 (15%) 20 (33%) 9 (18%) Non-study-related adverse experience 13 (27%) 25 (42%) 23 (47%) Other 5 (10%) 7 (11%) 3 (6%) No. of cycles with dose modification 16 (8%) 14 (6%) 26 (13%) Reason for dose modification Hematological 9 (56%) 2 (14%) 10 (38%) Non-hematological 6 (38%) 9 (64%) 16 (62%) Hematological and non- hematological 0 2 (14%) 0 Non-study-related adverse experience 1 (6%) 1 (7%) 0 evaluated the present arm A CPT-11/week 5-FU/FA regimen versus the Mayo Clinic 5-FU/FA regimen (bolus for 5 days every 4 weeks) and versus CPT mg/m 2 alone administered weekly for 4 weeks. The overall confirmed response rate for the CPT-11/5-FU/FA arm of this phase III trial (ITT population) was 39% (n = 231), and was significantly higher than the response rate reported for the corresponding control 5-FU/FA regimen of 21% (n = 226). The confirmed response rate for CPT-11 alone was 18% (n = 226). The median duration of response was 9 months for all three arms of the trial. The median TTPs were 7.0 and 4.3 months, and the overall median survivals were 14.8 and 12.6 months for the CPT-11/5-FU/FA and 5-FU/FA arms, respectively [19]. The biweekly Douillard regimen combination of CPT mg/m 2 (5-FU/FA fixed dose: 200 mg/m 2 FA, 5-FU bolus 400 mg/m 2 and 5-FU 600 mg/m 2 over 22 h, on days 1 and 2) evaluated in arm B was previously shown to be safe in pretreated patients without overlapping toxicity in the phase I study reported by Ducreux et al. [18]. These results were confirmed by Douillard et al. [20] in a pivotal European phase III trial which evaluated CPT-11 in combination with either the biweekly de Gramont (LV5FU2) or the weekly German Arbeitsgemeinschaft Internische Onkologie 5-FU/FA infusional regimens (n = 387). The overall confirmed response rates were 35% and 22% (ITT analysis), respectively, for the CPT-11 combination arm and the 5-FU/FA control arm. Again, the median durations of response were approximately 9 months for both the CPT-11/5-FU/FA and

6 1111 Table 3. Analysis of efficacy Evaluable patients a Arm A, CPT-11 + bolus 5-FU/FA a Eighteen patients were non-evaluable for the following reasons: method of measurement different from baseline (17 patients); one did not complete a full cycle of treatment. Number Best response Complete response 3 (9%) 3 (7%) 0 Partial response 11 (31%) 16 (37%) 13 (31%) Minor response/stable disease 12 (34%) 13 (30%) 19 (45%) Progressive disease 9 (26%) 11 (26%) 10 (24%) CPT-11 and 5-FU/FA Overall response rate 40% (24% to 58%) 44% (29% to 60%) 31% (17% to 47%) Tumor growth control rate 74% 74% 76% Median duration of stabilization, months Median response duration, months Median time to progression, months Intention-to treat patients 5.5 ( 7.3) 2.76 ( ) 5.7 ( ) 8.5 ( ) 11.0 ( ) 12.1 ( ) 5.8 ( ) 8.0 ( ) 7.3 ( ) Number Overall response rate 33% (21% to 48%) 42% (28% to 56%) 30% (18% to 45%) Median time to progression, months Median response duration, months Median survival time, months 6.1 ( ) 7.9 ( ) 7.2 ( ) 8.5 ( ) 11.0 ( ) 11.5 ( ) 15.2 ( ) 12.4 ( ) 17.0 ( ) 5-FU/FA arms [20]. The median TTPs were 6.7 and 4.2 months, and median survival times 17.4 and 14.1 months for the CPT-11/ 5-FU/FA and 5-FU/FA arms, respectively [20]. The alternating schedule of CPT-11 (350 mg/m 2 ) day 1 with FA (20 mg/m 2 /day) followed by 5 days bolus of 5-FU (425 mg/m 2 /day) (Mayo Clinic regimen) evaluated in arm C was previously reported by Van Cutsem et al. [16] in a phase II trial, and was shown to have a good tolerability profile, without any cumulative or overlapping toxicity. The present study showed that all three of the CPT-11/5-FU/FA combination regimens were highly active and comparable in terms of overall response rates (from 31% to 44% in the evaluable populations, and 30% to 42% in the ITT populations), median duration of response ( months) and median TTP (6 8 months) for the evaluable population, and median survival times [12 17 months (ITT)]. The study populations were representative of the candidates for first-line chemotherapy seen in clinical practice in both western Europe and the Unites States and the efficacy data were consistent with the efficacy data from the USA and European pivotal phase III trials [19, 20]. In previous phase II trials of the alternating schedules of bolus 5-FU/FA (Mayo Clinic regimen) and CPT-11 given either on a weekly-times-four schedule at 100 mg/m 2 [15] or every 6 weeks at 350 mg/m 2 [16] similar efficacy was shown with overall response rates of approximately 30% and median survival times of between 16 and 18 months. The three combinations of CPT-11 plus 5-FU and FA were well-tolerated and the toxic effects reversible, non-cumulative and manageable. Consistent with the observations made in previous studies, neutropenia and delayed diarrhea were the most common toxic effects. Neutropenia was the main severe adverse event occurring in the weekly CPT-11 plus bolus 5-FU/FA regimen and in the alternating regimen, as already observed in previous trials. Saltz et al. [19] reported grade 3/4 neutropenia in 54% of patients using the weekly regimen (n = 225). This was higher than our experience in the present study (30% of patients with grade 3/4 neutropenia). Similarly, Van Cutsem et al. [16], using the alternating schedule, reported 64% of patients with grade 3/4 neutropenia (n = 33), which was higher than that reported in the present trial (22% of grade 3/4 neutropenia). However, the frequency of severe neutropenia was higher (46% of grade 3/4 toxicity) in the European randomized phase III study than in the present study (8% grade 3/4 toxicity) [20]. The incidence of grade 3/4 delayed diarrhea was approximately 20% in arms A and C (18% for the weekly schedule and 22% for the alternating regimen),

7 1112 Figure 1. Kaplan Meier estimate of time to progression (intention-to treat population). Figure 2. Kaplan Meier estimate of overall survival (intention-to-treat population). which was comparable to the previously reported data [16, 19]. Delayed diarrhea also occurred in the biweekly schedule combining CPT-11 with bolus and, with 15% of patients exhibiting grade 3/4 toxicity. This incidence of severe diarrhea was similar to that observed in the European randomized phase III study, and was well managed by following the antidiarrheal guidelines with intensive and early use of loperamide. However, this regimen was distinguishable from the other two regimens by virtue of the very low incidence of grade 3/4 neutropenia (8% of patients and 2% of cycles).

8 1113 Table 4. Hematological toxicity (numbers of patients) related to CPT-11 and 5-FU/folinic acid treatments (worst grade by patient) Neutropenia Arm A, CPT-11 + bolus 5-FU/FA (n = 51) Overall 35 (69%) 25 (49%) 30 (60%) Grade 3 8 (16%) 2 (4%) 7 (14%) Grade 4 7 (14%) 2 (4%) 4 (8%) Febrile neutropenia a 3 (6%) 0 0 Leukopenia Overall 36 (71%) 26 (51%) 23 (46%) Grade 3 6 (12%) 0 6 (12%) Grade 4 1 (2%) 0 2 (4%) Thrombocytopenia Overall 4 (8%) 1 (2%) 2 (4%) Grade 4 1 (2%) 0 1 (2%) Anemia Overall 32 (63%) 39 (76%) 31 (62%) Grade 3 3 (6%) 1 (2%) 2 (4%) Grade 4 1 (2%) 0 0 CPT-11 and 5-FU/FA (n = 50) a Defined as fever 38 C concomitant with a grade 3/4 neutropenia in the absence of documented infection. Table 5. Number of patients (and percentage) with grade 3/4 non-hematological toxicity related to CPT-11 and 5-FU/FA treatments (n = 154) Toxicity Arm A, CPT-11 + bolus 5-FU/FA (n = 51) Alopecia grade 2 15 (30%) 21 (40%) 27 (54%) Diarrhea 9 (18%) 8 (15%) 11 (22%) Vomiting 4 (8%) 4 (8%) 4 (8%) Nausea 3 (6%) 3 (6%) 4 (8%) Stomatitis 3 (6%) 0 3 (6%) Infection 3 (6%) 1 (2%) 0 Cholinergic syndrome 1 (2%) 0 1 (2%) CPT- 11 and 5-FU/FA (n = 50) The other severe non-hematological toxicities occurred at a low incidence in all treatment arms and included mainly alopecia, and nausea/vomiting. No case of severe stomatitis was noted in the biweekly regimen, while 6% of patients experienced grade 3/4 toxicity in both the weekly regimen and the alternating regimen. The occurrence of cholinergic syndrome was low in all three arms. In conclusion, the three different therapeutic regimens combining CPT-11 with concomitant (weekly or biweekly) or alternated bolus and/or were highly active in patients with advanced colorectal cancer, and were associated with an easily managed safety profile, which ensured good schedule compliance. The low incidence of grade 3/4 neutropenia observed with the biweekly combination of CPT-11 and 5-FU/FA (arm B), means that this regimen has the best tolerability profile of the three regimens tested. Acknowledgements We are indebted to N. Cussac for diligent writing support and meticulous data review, as well as to B. Raoult and M. Berlion. This work was supported by Aventis Pharma, Antony, France and was presented in preliminary forms at the 20th Annual Meeting of

9 1114 Table 6. Death (numbers of patients) related or unrelated to study treatment within 60 days after the first infusion Death Arm A, CPT-11 + bolus 5-FU/FA (n = 51) Treatment-induced 2 (4%) 0 0 Treatment-unrelated 2 (4%) 2 (4%) 0 CPT- 11 and 5-FU/FA (n = 50) the American Society of Clinical Oncology, May 2000, New Orlean, and at the 21st Annual Meeting of the American Society of Clinical Oncology, May 2001, San Francisco. Additional principal participating investigators of the Irinotecan Study Group were: P. P. Bapsy (India), S. Ben Ahmed (Tunisia), T. Salek (Slovakia), S. Donea (Romania), D. A. Gerges (Lebanon), D. C. Doval (India), E. Buyukunal (Turkey), I. N. Chernozemsky (Bulgaria), V. I. Tzekova (Bulgaria), H. A. El-Azim (Egypt), H. Errihani (Morocco), R. Samlali (Morocco), J. S. Godz (Poland), N. Ghilezan (Romania), B. A. Berdov (Russia), G. N. Tchaikovsky (Russia), M. Wagnerova (Slovakia), N. F. Aykan (Turkey), J. Zidan (Israel) and G. Chachine (Lebanon). References 1. Machover D. A comprehensive review of 5-fluorouracil and leucovorin in patients with metastatic colorectal carcinoma. Cancer 1997; 80: Advanced Colorectal Cancer Meta-Analysis Project. Modulation of fluorouracil by leucovorin in patients with advanced colorectal cancer: evidence in terms of response rate. J Clin Oncol 1992; 10: Advanced Colorectal Cancer Meta-Analysis Project. Meta-analysis of randomized trials testing the biochemical modulation of fluorouracil by methotrexate in metastatic colorectal cancer. J Clin Oncol 1994; 12: Creemers GJ, Lund B, Verweij J. Topoisomerase I inhibitors: topotecan and irinotecan. Cancer Treat Rev 1994; 20: Rothenberg ML. Topoisomerase I inhibitors: review and update. Ann Oncol 1997; 8: Conti J, Kemeny N, Saltz L et al. Irinotecan is an active agent in untreated patients with metastatic colorectal cancer. J Clin Oncol 1996; 14: Pitot HC, Wender DB, O Connell M et al. Phase II trial irinotecan in patients with metastatic colorectal carcinoma. J Clin Oncol 1997; 15: Rougier P, Bugat R, Douillard J et al. Phase II study of irinotecan in the treatment of advanced colorectal cancer in chemotherapy-naive patients and pretreated patients with fluorouracil-based chemotherapy. J Clin Oncol 1997; 15: Vanhoefer U, Harstrick A, Achterrath W et al. Irinotecan in the treatment of colorectal cancer: clinical overview. J Clin Oncol 2001; 19: Rothenberg M, Eckardt J, Kuhn J et al. Phase II trial of irinotecan in patients with progressive or rapidly recurrent colorectal cancer. J Clin Oncol 1996; 14: Van Cutsem E, Cunningham D, Ten Bokkel Huinink WW et al. Clinical activity and benefit of irinotecan (CPT-11) in patients with colorectal cancer truly resistant to 5-fluorouracil (5-FU). Eur J Cancer 1999; 35: Cunningham D, Pyrhönen S, James RD et al. Randomised trial of irinotecan plus supportive care versus supportive care alone after fluorouracil failure for patients with metastatic colorectal cancer. Lancet 1998; 352: Rougier P, Van Custem E, Bajetta E et al. Randomised trial of irinotecan versus fluorouracil by continuous infusion after failure in patients with metastatic colorectal cancer. Lancet 1998; 352: Saltz LB, Kanowitz J, Kemeny NE et al. Phase I clinical and pharmacokinetic study of irinotecan, fluorouracil, and leucovorin in patients with advanced solid tumors. J Clin Oncol 1996; 14: Rothenberg ML, Pazdur R, Rowinsky EK et al. A phase II multicenter trial of alternating cycles of irinotecan and 5-FU/LV in patients with previously untreated metastatic colorectal cancer. Proc Am Soc Clin Oncol 1997; 16: 266a (Abstr 944). 16. Van Cutsem E, Pozzo C, Starkhammar H et al. A phase II study of irinotecan alternated with five days bolus of 5-fluorouracil and leucovorin in first line chemotherapy of metastatic colorectal cancer. Ann Oncol 1998; 9: Vanhoefer U, Harstrick A, Köhne CH et al. Phase I study of a weekly schedule of irinotecan, high dose leucovorin, and infusional fluorouracil as first-line chemotherapy in patients with advanced colorectal cancer. J Clin Oncol 1999; 17: Ducreux M, Ychou M, Seitz JF et al. Irinotecan combined with bolus fluorouracil, continuous infusion fluorouracil, and high-dose leucovorin every two weeks (LV5FU2 regimen): a dose finding and pharmacokinetic study in patients with pretreated metastatic colorectal cancer. J Clin Oncol 1999; 17: Saltz LB, Cox JV, Blanke C et al. Irinotecan plus fluorouracil and leucovorin for metastatic colorectal cancer. N Engl J Med 2000; 343: Douillard JY, Cunningham D, Roth AD et al. Irinotecan combined with fluorouracil compared with fluorouracil alone as first-line treatment for metastatic colorectal cancer: a multicentre randomised trial. Lancet 2000; 355: