Background

Until recently, 5-fluorouracil (5-FU) in combination with folinic acid (FA) has been the mainstay of treatment for advanced colorectal cancer (CRC) [1, 2]. Of the numerous schedules and doses of 5-FU investigated, the Mayo Clinic bolus schedule, in which both 5-FU and FA are injected daily for five days every four weeks [3–5], emerged as one of the most widely used schedules worldwide for the first-line treatment of advanced CRC.

Irinotecan (CPT-11, Campto®) is an S-phase specific derivative of camptothecin, which interferes with DNA replication and cell division inhibiting topoisomerase I [6]. Irinotecan demonstrated antitumour activity against metastatic CRC when used alone either as first-line treatment [7–9] or as second-line treatment after the failure of 5-FU [9–12], with overall response rates ranging from 13% to 32%. In randomised phase III trials, irinotecan administered second-line prolonged survival significantly when compared with either best supportive care [13] or infusional 5-FU/FA [14]. The most frequently reported adverse events associated with irinotecan are neutropenia, delayed diarrhoea, acute cholinergic syndrome, alopecia, fatigue, nausea and vomiting [12].

Early studies in which irinotecan was combined or alternated with 5-FU/FA regimens showed promising first-line efficacy in advanced CRC [15–19]. Two studies had assessed the feasibility of irinotecan (administered every three weeks in the European study or weekly for four weeks followed by two weeks' rest in the US study) alternating with the Mayo Clinic schedule [16, 17]. Seventy patients [16] and 33 patients [17] were treated and the overall response rate was around 30%, with a median time to progression of seven months in both studies. The safety profile was satisfactory without overlapping toxicities. These data, coupled with the emerging data from the US and European phase III trials [20, 21], suggested that a study of irinotecan in combination with the Mayo Clinic schedule every four weeks was warranted.

The present phase I/II study was designed to determine the maximum tolerated dose (MTD), the recommended dose and the safety profile of irinotecan when combined with the 5-FU/FA Mayo Clinic schedule every four weeks in the treatment of advanced CRC. The antitumour activity of this combination was also evaluated.

Methods

Patient eligibility

Patients with histologically proven, measurable, advanced adenocarcinoma of the colon or rectum, with no potentially resectable metastases, were eligible for inclusion in the study. Other inclusion criteria were: age between 18 and 75 years; a World Health Organization (WHO) performance status (PS) 0–2; a life expectancy >3 months; adequate haematological function (haemoglobin ≥10 g/dL, neutrophils ≥2 × 109/L and platelets ≥150 × 109/L); satisfactory renal and hepatic functions (total bilirubin ≤1.25 × upper normal limit [UNL], creatinine ≤1.25 × UNL, aspartate aminotransferase [ASAT] and alanine aminotransferase [ALAT] ≤3 × UNL; or in the case of liver metastases: total bilirubin ≤1.5 × UNL, ASAT and ALAT ≤5 × UNL); no prior chemotherapy or only (neo) adjuvant chemotherapy completed more than six months prior to study entry, and no radiotherapy within the four weeks prior to study entry. Exclusion criteria included: evidence of brain metastases, current infection, unresolved bowel obstruction or sub-acute obstruction, uncontrolled Crohn's disease or ulcerative colitis, and a current history of chronic diarrhoea. This study was approved by the local Ethics Committees of each centre, and conducted in accordance with the Declaration of Helsinski. All patients provided written informed consent. Pretreatment evaluations included a complete medical history and physical examination, complete blood cell count and blood chemistry, electrocardiogram (ECG), complete tumour imaging (X-ray, computed tomography, ultrasound) performed within three weeks before first infusion, and an assessment of tumour markers. The first treatment administration was made within eight days of registration for the study.

Treatment

Irinotecan (Campto®) was supplied by Aventis Pharma (Antony, France) as a sterile solution of 20 mg/ml in 5 ml vials. 5-FU and FA were used in commercially available formulations. Irinotecan (starting at a dose of 250 mg/m2) was administered as a 30- to 90-minute intravenous [IV] infusion on day 1, immediately followed by FA 20 mg/m2/day IV bolus followed by 5-FU at a fixed dose of 425 mg/m2/day IV bolus, on days 1–5. All patients received a 15-minute bolus infusion of 5-FU/FA. The treatment was administered every four weeks up until disease progression, unacceptable toxicity or withdrawal of patient consent.

Concomitant medication included subcutaneous atropine 0.25 mg as curative treatment for severe cholinergic symptoms, including early diarrhoea. For the treatment of delayed diarrhoea, high-dose loperamide (2 mg every two hours for at least 12 hours) was given as soon as the first liquid stool occurred and continued for up to 12 hours after the last liquid stool, without exceeding a total treatment duration of 48 hours. If the diarrhoea persisted for more than 48 hours despite loperamide treatment, or in the case of severe diarrhoea or diarrhoea associated with vomiting, fever or severe neutropenia, patients received a seven-day course of prophylactic broad spectrum oral antibiotics (fluoroquinolone or cotrimoxazole) and were hospitalised for rehydration. Patients with febrile neutropenia had to be hospitalised to receive IV antibiotic therapy.

Study design

Phase I study

The starting dose of irinotecan was 250 mg/m2 (dose level I) and the subsequent planned dose levels were 300 mg/m2 (dose level II) and 350 mg/m2 (dose level III). At least three consecutive patients were to be entered at each dose level with a one-week interval between the entry of the first patient and the next two patients. If one out of three patients at one dose level experienced a dose-limiting toxicity (DLT) in the first cycle, at least three more patients were entered at the same dose level. If three out of six patients experienced the same DLT, this dose level was considered to be the MTD of irinotecan in this combination. In the absence of DLT, the next dose level was explored. Once the MTD, as defined above, was reached, the dose immediately below the MTD was explored for determination of the recommended dose. At least six patients were to be included at the MTD and at the recommended dose. Intra-patient dose escalation was not allowed. DLT during the first cycle was defined as the occurrence of any grade 3 or 4 non-haematological toxicity (except alopecia), platelets <25 × 109/L, symptomatic thrombocytopenia (haemorrhage), febrile neutropenia (fever ≥38.0°C with concomitant grade 3–4 neutropenia in the absence of documented infection), grade 3–4 infection, grade 4 neutropenia ≥7 days, grade 3–4 neutropenia associated with severe infection. Second and further cycles of treatment were administered when the neutrophil count was ≥1.5 × 109/L, platelets ≥100 × 109/L and non-haematological toxicity (except alopecia) ≤grade 2, otherwise treatment had to be delayed but not for more than two weeks. If the delay was greater than two weeks, the patient was withdrawn from the study.

Phase II study

According to the optimal two-stage Simon design [22] used to define the total number of patients required to detect the activity of the treatment, up to 43 evaluable patients had to be included in the study. Dose reductions of 20% for irinotecan and 5-FU were planned in case of grade 3–4 haematological and/or non-haematological toxicities. There was no dose reduction schedule for FA. As for the phase I part of the study, if the treatment delay was greater than two weeks, the patient was withdrawn from the study.

Relative dose intensities (RDI, the ratio between the actual dose delivered and the planned dose) for the different components of the treatment were regimen were calculated for both the phase I and phase II parts of the study.

Toxicity and response evaluation

Toxicity, graded according to National Cancer Institute Common Toxicity Criteria (NCI-CTC) version 1, was assessed by clinical examination every cycle, weekly for haematology and every two cycles for biochemistry. Tumour responses were assessed every eight weeks (or two cycles of treatment) using the same method of assessment used at baseline, and according to WHO criteria. Responses were defined as follows: complete response (CR), partial response (PR), minor response (MR), stable disease (SD), and progressive disease (PD). The overall response rate was defined as the percentage of patients with a CR or PR. The duration of response, in responding patients, was calculated from the date of first infusion to the date of documented progression. The time to disease progression (TTP) was calculated from the time of the first infusion to the date of documented progression or death due to malignant disease. Deaths for other reasons than malignant disease were censored.

Statistical analysis

Analyses were performed using SAS® Software version 8.0 within Windows. The primary end-point of the phase I part of the study was to analyse the DLT observed during the first cycle and to determine the MTD of irinotecan in combination with the 5-FU/FA Mayo Clinic schedule. The secondary objectives were to determine the recommended dose of irinotecan for the phase II part of the study and the safety profile of the combination regimen. In the phase II part of the study, the response rate was reported within 95% confidence intervals (CI). Efficacy analyses were performed for both the intent-to-treat (ITT) and evaluable (eligible patients with measurable lesion, who had received a minimum of two cycles of treatment, i.e. eight weeks on study) populations.

The Kaplan-Meier method was used to analyse TTP and overall survival. Safety analyses were performed on all patients receiving at least one administration of irinotecan.

Results

Patient characteristics

Between June 19, 1999 and March 26, 2001, 63 patients from three countries (Austria, South Africa, and United Kingdom) were entered: 14 patients in the phase I part of the study and 49 patients in the subsequent phase II part of the study (Table 1). Four patients were not eligible in the phase II part of the study due to the absence of adequate biochemical/haematological function (one of these patients also had no measurable lesion and one had received adjuvant chemotherapy less than six months prior to study entry). Demographic and baseline disease characteristics are listed in Table 1. Seven men and 7 women with a median age of 59 years entered the phase I part of the study. Thirty-three men and 16 women with a median age of 63 years entered the phase II part of the study. In both parts of the study, the majority of patients had a WHO PS of 1 and two organs involved (median 2, range 1–4), with the liver being the most common site of metastatic disease. Most patients underwent prior surgery (86% in both parts of the study). Prior radiotherapy was given to 21% of phase I patients and to 29% of phase II patients. Prior adjuvant chemotherapy had been administered to 29% and 24% of phase I and phase II patients, respectively.

Table 1 Patient characteristics
Full size table

Maximum tolerated dose, recommended dose, phase I part of the study

Six patients received irinotecan at dose level I (250 mg/m2). Eight patients received irinotecan at dose level II (300 mg/m2). DLTs in the first cycle of treatment are shown in Table 2. The MTD was determined to be irinotecan 300 mg/m2 combined with 5-FU 425 mg/m2 and FA 20 mg/m2. At this dose level, a total of four patients (out of eight) experienced a grade 3 or 4 non-haematological toxicity, comprising diarrhoea (three patients), fatigue or pain (one patient each). The haematological DLT was grade 3–4 neutropenia with severe infection (one patient). The recommended dose level for irinotecan was defined as 250 mg/m2 combined with 5-FU 425 mg/m2 and FA 20 mg/m2. Of the six patients treated at this dose level, four experienced a non-haematological grade 3–4 toxicity, comprising stomatitis, nausea, fatigue, diarrhoea or pulmonary toxicity. One patient also experienced febrile neutropenia.

Table 2 Dose-limiting toxicity during first cycle of irinotecan combined with 5-FU/FA Mayo Clinic schedule according to dose levels in phase I study (n = 14)
Full size table

Treatment compliance

Phase I

In total, 60 cycles of irinotecan combined with 5-FU/FA were administered to 14 patients (median number of cycles four, range, 1–8). Six cycles (10%) were delayed (22% at dose level I and 5% at dose level II), and in two cycles (3%), the dose was reduced due to non-haematological toxicity (one cycle) and prescription error (one cycle).

Phase II

In total, 218 cycles were administered to 49 patients (median four cycles, range, 1–12). The median RDI was 90% (range, 52–110%) for irinotecan, 88% (range, 19–108%) for 5-FU and 98% (range, 20–135%) for FA. Atotal of 22 cycles (10%) had to be delayed due to haematological (seven cycles) and non-haematological toxicity (six cycles), non-study-related adverse events (two cycles) and other reasons (seven cycles: patient convenience, five cycles; surgery, two cycles). The dose of both irinotecan and 5-FU was reduced in 34 cycles (16%), due to haematological toxicity (15 cycles), non-haematological adverse events (17 cycles), haematological and non-haematological toxicity (one cycle) and weight loss (one cycle).

Haematological toxicity

Phase I

Neutropenia was the main adverse event. At dose level I, grade 3–4 neutropenia occurred in 83% (5/6) of patients (Table 3). The median time to neutrophil nadir was 13 days. The median durations of grade 3 and grade 4 neutropenia were one day and two days, respectively. At dose level II, grade 3–4 neutropenia occurred in 63% (5/8) of patients (Table 3). The median time to neutrophil nadir was 20.5 days. At dose level I, grade 3–4 thrombocytopenia occurred in 17% (1/6) of patients. At dose level II, no grade 3–4 thrombocytopenia was observed. Febrile neutropenia was seen in one patient at dose level I, whilst one patient had severe anaemia at dose level II (Table 3).

Table 3 Haematological toxicity of irinotecan combined with 5-FU 425 mg/m2 and FA 20 mg/m2 Mayo Clinic schedule (worst grade by patient)
Full size table

Phase II

Neutropenia was also the main adverse event for this phase of the study, with grade 3–4 toxicity reported for 77% (38/49) of patients. Three patients experienced febrile neutropenia. The median time to neutrophil nadir was 15 days. The median duration of grade 3 and grade 4 neutropenia was one day. Grade 3–4 thrombocytopenia and anaemia were infrequent and occurred in one and two patients, respectively.

Non-haematological toxicities

Phase I

The most frequent grade 3–4 non-haematological adverse events possibly or probably related to the treatment were delayed diarrhoea, fatigue, pain, stomatitis, alopecia, nausea, anorexia, vomiting, fever and infection. Grade 3–4 adverse events possibly or probably related to treatment included delayed diarrhoea (57% of patients), fatigue (36% of patients), stomatitis (36% of patients), and pain (29% of patients) (Table 4). Hepatic tolerance was very good (no grade 3/4 increase for transaminases).

Table 4 Most common grade 3–4 non-haematological toxicity possibly or probably related to irinotecan combined with 5-FU/FA Mayo Clinic schedule according to dose levels (worst grade by patient)
Full size table

Phase II

The safety profile of the recommended dose combination was close to that observed for the phase I patients. Grade 3–4 adverse events related to study drugs included diarrhoea in 33% of patients, stomatitis in 26% of patients, fatigue in 22% of patients, infection in 16% of patients, nausea in 16% of patients, vomiting in 14% of patients, and pain in 18% of patients (Table 4). Hepatic tolerance was good: one patient had a grade 3 increase in alkaline phosphatase and one had a grade 3 increase in bilirubin. However, three (out of 49) patients died of neutropenic septic shock during the first cycle due to probable or possible treatment-related toxicity (these deaths occurred within the first fifteen days of treatment). One of these patients had a WHO PS of 1 and two had a WHO PS of 2.

Efficacy results

Phase I

In the phase I part of the study, three eligible patients were not evaluable for efficacy because two full cycles of treatment were not completed (n = 1) and/or the method of tumour assessment following treatment differed from that used at baseline (n = 3). PRs were seen in 18% (2/11) of evaluable patients (0/4 at dose level I and 2/7 at dose level II). In addition, 45% of evaluable patients had at least stabilisation of their disease (four MR and one SD).

Phase II

In the phase II part of the study, nine of the eligible patients were not evaluable for efficacy because two full cycles of treatment were not completed (n = 8) and/or the method of tumour assessment following treatment differed from that used at baseline (n = 9). Among evaluable patients there were three CR and 10 PR, giving an overall response rate of 36% (13/36) (29% in the ITT population) (Table 5). Seven patients had a MR and nine patients had SD, leading to 44% SD. The median duration of stabilisation was 5.8 months (95% CI, 4.0–8.3 months). The median duration of response was 10.4 months (95% CI, 5.9–11.1 months). The median TTP was 7.0 months (95% CI, 4.5–10.1 months) in the evaluable population and 5.9 months (95% CI, 4.0–8.1 months) in the ITT population (Figure 1). The median survival time was 12.0 months (95% CI, 8.0–15.2 months) (Figure 2).

Table 5 Efficacy of irinotecan 250 mg/m2 combined with 5-FU 425 mg/m2 and FA 20 mg/m2 Mayo Clinic schedule in patients in phase II study
Full size table
Figure 1
figure 1

Kaplan-Meier plot of time to progression in ITT population Three patients were censored (1 at 0.4 months, 2 at 0.5 months) due to deaths from neutropenic sepsis related to study treatment. The patient censored at 2.7 months died of bowel obstruction not related to study treatment.

Full size image
Figure 2
figure 2

Kaplan-Meier plot of overall survival in ITT population

Full size image

Discussion

The main objective of this multicentre, open phase I/II study was to evaluate the efficacy of the combination of irinotecan administered as a 30- to 90-minute IV infusion combined with the commonly used 5-FU/FA Mayo Clinic schedule (bolus infusion daily for five days, every four weeks) in patients with advanced CRC after determination of the recommended dose in the phase I part of study. Delayed diarrhoea was dose limiting at an irinotecan dose level of 300 mg/m2 (dose level II) combined with 5-FU 425 mg/m2 and FA 20 mg/m2. The recommended dose for the phase II part of the study was, therefore, irinotecan 250 mg/m2 combined with 5-FU 425 mg/m2 (as a 15-minute IV bolus) and FA 20 mg/m2. At this dose level, 18 cycles of irinotecan were administered to six patients in the phase I part of the study and 218 cycles to 49 patients in the phase II part of the study. In the phase II study, the median RDI was superior or equal to 87% for each drug.

The combination of irinotecan and 5-FU/FA demonstrated clinical activity, as evidenced by an overall response rate of 29% among the ITT population and of 36% among the evaluable population. This is similar to the response rate reported recently in a phase II study of an alternating regimen of irinotecan and the Mayo Clinic schedule (30%) [23]. However, the response rate is lower than that reported among the ITT populations of a phase II study investigating the Nordic bi-weekly schedule of irinotecan/bolus FU/FA (39%) [24]. While our study was being conducted, results were published from the two pivotal phase III randomised trials confirming the benefits of irinotecan combined with either the bi-weekly de Gramont or the weekly German A.I.O. (Arbeitsgemeinschaft Internische Onkologie) infusional regimens (European trial) [21] or the weekly bolus Saltz regimen (North American trial) [20] compared with 5-FU/FA alone. The response rates in the ITT populations of the irinotecan-containing arms of both trials (35% and 50%) were higher than that seen in our current study. The median TTP was similar in our study, the phase II study using the alternating irinotecan/Mayo Clinic schedule and the phase III trials (approximately 7.0 months). However, the median survival time in the ITT populations were higher in both the phase II alternating irinotecan/Mayo Clinic study (17 months) [23] and the phase III trials (14.8 and 17.4 months) [20, 21] than in our study (12.0 months). One possible explanation for the shorter median survival time observed in our study could be the relatively lower dose of irinotecan used (250 mg/m2 every four weeks), compared with the doses used in the European trial (80 mg/m2/week or 180 mg/m2/2 weeks) [21] and the American trial (125 mg/m2/week) [20]. However, the recently reported trial of alternating irinotecan/Mayo Clinic schedule used an irinotecan dose of 350 mg/m2 every six weeks [23]. It is possible therefore that the schedule of administration may be an influencing factor in outcome.

In the present study, neutropenia was the main toxicity. The incidence of grade 3–4 neutropenia in the phase II part of the study (77% of patients) was comparable to that reported by Buroker et al. with the 5-FU/FA Mayo Clinic schedule alone [4]. However, it was higher than that reported with an alternating schedule of irinotecan/Mayo Clinic schedule [23] and that reported by Saltz et al. in their phase III study comparing irinotecan/bolus 5-FU/FA (54%) with 5-FU/FA Mayo Clinic schedule alone (66%) [20]. Although the incidence of grade 3–4 delayed diarrhoea in the present study (33% of patients in the phase II study) was comparable to that obtained with irinotecan as a single agent [20, 25, 26], it was higher than that reported with the alternating irinotecan/Mayo Clinic schedule (22%) [23] and the Saltz irinotecan/bolus 5-FU/FA (23%) regimen [20]. The incidences of grade 3–4 infection (16% of patients) and grade 3–4 stomatitis (26% of patients) were also significant in the phase II part of our study.

Three treatment-related deaths occurred within the 60 days after the first infusion and were due to complications following haematological toxicity. Since the initiation of this trial, data have been published regarding the association between some regimens of bolus 5-FU/FA, in combination with irinotecan or oxaliplatin, and early deaths of patients [27–29]. Recommendations of an independent panel for the use of irinotecan plus bolus 5-FU/FA include the weekly assessment of patients, particularly up until week 3 or 4 of treatment, and the prompt management of diarrhoea and/or neutropenia and/or fever [28].

Conclusions

Although infusional regimens of 5-FU/FA in combination with irinotecan seem to have better safety profiles than the use of bolus 5-FU/FA, they are not always a practical therapeutic option, and bolus-based regimens still have a role to play in the treatment of metastatic CRC. The results of the present phase I/II study show that the combination of irinotecan (250 mg/m2) and the Mayo Clinic schedule of bolus 5-FU/FA administered once every four weeks demonstrated activity in the first-line treatment of metastatic CRC. However, this regimen was associated with a relatively high level of neutropenia. Three treatment-related deaths amongst 49 patients were reported. Therefore, this regimen requires a close follow-up of the patients. Infusional 5-FU/FA in combination with irinotecan should be the regimen of choice instead of bolus 5-FU/FA.