A phase I/II and pharmacokinetic study of irinotecan in combination with capecitabine as first-line therapy for advanced colorectal cancer

D. W. Rea1, J. W. R. Nortier2, W. W. Ten Bokkel Huinink3, S. Falk4, D. J. Richel5, T. Maughan6, G. Groenewegen7, J. M. Smit8, N. Steven1, J. M. Bakker10, D. Semiond11, D. J. Kerr9 and C. J. A. Punt12,*

1 CR UK Institute for Cancer Studies, University of Birmingham, Birmingham, UK; 2 Leiden University Medical Centre, Leiden; 3 Netherlands Cancer Institute, Amsterdam, The Netherlands; 4 Taunton and Somerset Hospital, Somerset, UK; 5 Academic Medical Centre, Amsterdam, The Netherlands; 6 Clinical Trials Unit Velindre Hospital NHS Trust, Whithchurch, UK; 7 University Medical Centre Utrecht, Utrecht; 8 Gelre Hospitals, Apeldoorn, The Netherlands; 9 University of Oxford, Oxford, UK; 10 Aventis Pharma B.V., The Netherlands; 11 Aventis Pharma, S.A., France; 12 Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands

* Correspondence to: Prof. Dr C. J. A. Punt, Department of Medical Oncology, Radboud University Nijmegen Medical Centre, PO Box 9101, 6500HB Nijmegen, The Netherlands. Tel: +31-24-3610353; Fax: +31-24-3540788; Email: c.punt{at}onco.umcn.nl


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Purpose:: The aim of this study was to determine in patients with previously untreated advanced colorectal cancer the maximum tolerated dose (MTD) and safety profile of irinotecan in combination with capecitabine, to identify a recommended dose and to determine the response rate and time to disease progression. In addition, we aimed to explore the pharmacokinetic parameters of irinotecan and capecitabine when used in different sequences of administration, with irinotecan infusion either prior to or after the first intake of capecitabine.

Patients and methods:: One hundred patients were included: 43 patients were recruited into an extended phase I trial of alternating escalation in dose of both drugs where irinotecan was administered intravenously (i.v) on day 1 after first intake of capecitabine taken from days 1–14 twice daily, with cycles repeated every 3 weeks. After the determination of recommended dose a further 57 patients were treated in a phase II evaluation with the reverse sequence of drugs on day 1. Pharmacokinetic analysis was performed in patients treated at the recommended dose in two cohorts of patients in which the sequence of the first administration of each drug was reversed.

Results:: The MTD of the combination was determined as irinotecan 300 mg/m2, with capecitabine 2000 mg/m2/day. Dose limiting toxicities were neutropenia and diarrhoea. The recommended dose is irinotecan intravenous (i.v.) 250 mg/m2 day 1 and capecitabine 2000 mg/m2/day days 1–14, every 3 weeks. Treatment was well tolerated, with diarrhoea the most common serious toxicity. Response rate in the phase II cohort was 42% [95% confidence interval (CI) 29% to 56%]. Median duration of response was 7.7 months (95% CI 7.5–8.9). Median time to progression was 8.3 months (95% CI 5.8–10). No significant effect on irinotecan pharmacokinetics was observed whatever the intake of capecitabine before or after irinotecan infusion. An effect of irinotecan on capecitabine and some capecitabine metabolites was observed, but irinotecan did not effect 5-fluorouracil (5-FU) pharmacokinetics.

Conclusions:: Irinotecan in combination with capecitabine is a well tolerated regimen with an activity comparable to, but more convenient than, irinotecan–5-FU i.v. combinations in patients with previously untreated advanced colorectal cancer. The pharmacokinetic data suggest that the sequence of administration does not impact significantly on the metabolism of the two drugs.

Key words: capecitabine, colorectal cancer, irinotecan, phase I, phase II pharmacokinetics


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
The results of systemic therapy for advanced colorectal cancer have improved significantly with the availability of cytotoxic drugs such as irinotecan and oxaliplatin, and monoclonal antibodies against growth factors and their receptors [1Go]. Irinotecan is a topoisomerase I inhibitor that has activity in 5-fluorouracil (5-FU)-pretreated advanced colorectal cancer and proven survival advantage when used as second-line monotherapy in this setting [2Go, 3Go]. Irinotecan administered in combination with 5-FU results in significant superior survival when used as first-line therapy compared with 5-FU monotherapy. Two separate trials have established this superiority when used in combination with either a bolus 5-FU regimen [4Go] or in combination with infusional 5-FU [5Go]. Recently, orally active fluoropyrimidine analogues have been introduced into routine clinical practice. Capecitabine is an oral precursor that is preferentially converted into the active compound 5-FU in malignant tissue [6Go] in three steps, and was shown to have improved tolerability and response rate compared with bolus 5-FU, with comparable time to progression and survival, in metastatic colorectal cancer [7Go, 8Go]. Capecitabine has gained widespread acceptance as an alternative to intravenous (i.v.) 5-FU [1Go]. A logical progression from these two developments is to combine irinotecan and capecitabine with the aim of maintaining the efficacy of a combination regimen and utilising the convenience of oral fluoropyrimidine therapy. We have conducted a phase I/II study of irinotecan and capecitabine in patients with advanced colorectal cancer using a 21-day cycle comprising irinotecan on day 1 with capecitabine administered twice-daily days 1–14. A clinically significant pharmacokinetic interaction between 5-FU–folinic acid (FA) and irinotecan has been reported, demonstrating a 40% reduction in the area under the curve (AUC) for the active irinotecan metabolite SN-38 when 5-FU was commenced following irinotecan compared with the reverse sequence where irinotecan was administered after completion of 5-FU infusion [9Go]. This interaction was associated with differences in the maximum tolerated dose (MTD) of irinotecan. Therefore, in the present study the two sequences of i.v. irinotecan, before or after capecitabine intake, were studied with a pharmacokinetic analysis of the metabolic interaction of these drugs to determine the importance of the sequence of administration of these drugs.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Patients
Between May 2000 and December 2003, patients were recruited from three centres to the phase I component and subsequently from nine centres to the phase II component. The main eligibility criteria were: histologically proven metastatic adenocarcinoma of the colon or rectum with measurable disease, no prior chemotherapy for metastatic disease, age ≥18 and ≤75 years, WHO performance status 0, 1 or 2, life expectancy >3 months, adequate haematological and biochemical parameters [haemaglobin ≥10 g/dl, neutrophils ≥2 x 109/l, platelets ≥100 x 109/l, bilirubin ≤1.25x institutional upper limit of normal range (ULN), aspartate aminotransferase and alanine aminotransferase ≤5 times ULN, serum creatinine ≤1.25 ULN or creatinine clearance >50 ml/min], and written informed consent. Main exclusion criteria were: pregnant or lactating women, inadequate contraception in women with childbearing potential, prior exposure to irinotecan and/or exposure to more than one thymidine synthetase inhibitor regimen, severe concomitant conditions, inflammatory bowel disease or malabsorption. The study was approved by appropriate ethics committees covering all sites.

Study design and treatment
This study comprised a phase I and a phase II component. The phase I study was conducted as an open label, non-randomised, dose-finding study using escalation of both study drugs. Five dose levels were tested with an irinotecan dose ranging from 250 to 300 mg/m2 in combination with capecitabine ranging from 1500 to 2500 mg/m2/day twice daily days 1–14. The primary objective in the phase I study was to determine the MTD of irinotecan and capecitabine and to determine a recommended dose for phase II study when irinotecan is administered i.v. on day 1 combined with a 14-day oral intake of capecitabine. A secondary objective was to define the safety profile of the combination at each dose level.

The primary objective of the phase II study was to determine the response rate after treatment at the recommended dose. Secondary objectives were to determine the duration of response and time to progression, and to further evaluate the safety profile of the recommended combination dosing and schedule.

For the phase I study capecitabine was administered orally twice daily on days 1–14 every 3 weeks. Dosing was ~12 h apart (9 a.m. and 9 p.m.) and taken within 30 min after a meal. Irinotecan was administered 2 h after the first dose of capecitabine as an i.v. infusion over 30 min on day 1 repeated every 21 days (sequence A). In the phase II study, the reverse sequence of administration was used (sequence B), where irinotecan was administered over 30 min followed 2 h after end of the irinotecan infusion by the first dose of oral capecitabine (Figure 1). 5-HT3 antagonists were administered within conventional institutional antiemetic protocols in routine use in participating centres. The administration of atropine sulphate 0.25 mg subcutaneously was widely adopted as prophylaxis against irinotecan-induced acute cholinergic-like syndrome.



View larger version (16K):
[in this window]
[in a new window]
 
Figure 1. Sequence and timing of drug administration on day 1 of cycle 1. n indicates numbers of patients contributing to pharmacokinetic evaluation in each schedule and dose. I.v., intravenous.

 
Dose escalation proceeded according to the dose schedule (see Table 2). Three patients were entered at each dose level and dose escalation was permitted if no dose-limiting toxicity (DLT) was encountered in the first cycle. Cohorts were expanded where DLT was experienced and patients replaced if unable to comply with cycle 1, unless because of DLT. MTD was defined as being reached if the first three patients experienced a DLT or if three or more patients in an expanded cohort experienced DLT.


View this table:
[in this window]
[in a new window]
 
Table 2. Dose-escalation schedule and dose-limiting toxicities in phase I cohort during first cycle

 
DLT was defined as any of the following experienced during the first cycle: grade 4 neutropenia >7 days, neutropenic fever, grade 4 thrombocytopenia, diarrhoea grade 3 or grade 2 >8 days, vomiting grade 4, hand–foot syndrome grade 3 for ≥2 days, any other grade ≥3 toxicity or inability to administer treatment on schedule. The recommended dose was the dose level below the MTD and a further 15 patients were to be treated at this dose level with pharmacokinetic evaluation. Within the phase II study a second cohort of 15 patients were treated with the reverse sequence (sequence B, Figure 1) with pharmacokinetic evaluation. This sequence was administered to all patients on the phase II trial.

Treatment was repeated every 21 days. Treatment was discontinued for disease progression, unacceptable toxicity, delay of >36 days in instituting the next cycle of treatment, or at investigators discretion or patient request. Patients were followed up every 3 months after completion of treatment to evaluate resolution of treatment-related toxicity and time to progression.

Toxicity was evaluated weekly during treatment. Protocol specified dose reductions and delays based on previous cycle toxicity utilising both haematological nadirs and on day of next treatment haematological/biochemical parameters. In the phase I study, the dose was reduced to the dose level below. In the phase II study, irinotecan dose was reduced by 20% and capecitabine dose by 25% depending on the degree of toxicity. Grade 3 and 4 non-haematological toxicities and grade 2 hand–foot syndrome were required to resolve to grade ≤1 prior to redosing with a reduction in one or both drugs depending on the specific toxicity.

The primary end point of the phase II part was to assess the partial plus complete response rate associated with irinotecan plus capecitabine in patients with measurable advanced colorectal cancer. The study used a single-stage Fleming design, with a response rate of ≤20% selected as not worthy of further study and a response rate ≥35% sufficiently promising for further evaluation. With type I error set at 5%, type II error set at 20% and a power of 80%, a sample size of 50 evaluable patients was required. Allowing for a 10% rate of non-evaluable patients, 56 patients were included in the phase II trial. Response was evaluated by tumour assessments every three cycles during treatment including an end of treatment assessment. WHO response criteria were used for evaluation of response in the phase I trial and RECIST criteria were used to determine response in the phase II study.

Pharmacokinetics
The pharmacokinetic component of the study was intended to determine the pharmacokinetic parameters of irinotecan and the active metabolite SN-38 and of capecitabine and relevant capecitabine metabolites 5'-deoxy-5-fluorocytidine (5'-DFCR), 5'-deoxyfluorouridine (5'-DFUR), 5-FU and {alpha}-fluoro-ß-alanine (FBAL) during both sequences of administration during the extended phase I and phase II components of the study. Blood samples were drawn from an indwelling intravenous cannula and collected into heparin-coated tubes for irinotecan an into EDTA-containing tubes for capecitabine on day 1 of the first cycle as follows.

Sequence A (capecitabine preceding irinotecan)
Blood samples were collected immediately prior to capecitabine intake and at 0.5, 1, 2.5, 3, 4, 5, 7 and 10 h (prior to the evening dose) after the first oral morning capecitabine dose administered.

Samples were also taken immediately prior to irinotecan infusion and at 15 and 25 min after the start of infusion, and at 15, 30, 90 min, 2.5, 4.5, 7.5, 20.5, 23 and 25.5 h after the end of the infusion of irinotecan.

Sequence B (irinotecan preceding capecitabine)
Samples were taken immediately prior irinotecan infusion, at 15 and 25 min after the start of infusion and at 15, 30, 90 min and 2.5, 4, 6, 9, 23 and 28 h after the end of infusion of irinotecan. Samples were also taken immediately prior to intake of capecitabine, and at 0, 5, 1, 2, 3, 4, 5, 7 and 10 h after first oral morning dose administered. After sampling, all blood specimens were immediately put in an ice water bath (4 °C) until centrifugation at 2000 r.p.m. for 15 min at 4 °C. Then plasma samples were stored at –20 °C until analysis.

Plasma concentration of irinotecan and its metabolite SN-38 (total of lactone and carboxylate forms) were determined by validated high-performance liquid chromatography assay (HPLC) with fluorescence detection following solid phase extraction. The lower limit of quantitation was 10 µg/l for irinotecan and 2.5 µg/l SN-38 for a 50 µl sample volume.

The accuracy of the assay, defined as the present difference between the nominal and the mean measured concentrations of the quality controls, ranged from –1.4% to 4% for irinotecan and from –0.11% to 3.6% for SN-38 in plasma over the analysis period. The precision of the assay, established by the coefficient of variation of the quality controls, was lower than 5.5% for both compounds.

Concentrations of capecitabine and its metabolites 5'-DFCR, 5'-DFUR, 5-FU and FBAL were determined by HPLC with tandem mass-spectrometric detection. The lower limits of quantitation were 50 µg/l for 5'-DFUR, 11.3 µg/l for FBAL, 10 µg/l for capecitabine and 5'-DFCR, and 2 µg/l for 5-FU. The accuracy of the assay ranged from –8.3% to 4.4% for capecitabine, from –6% to 5.6% for 5'-DFCR, from –11% to 1.5% for 5'-DFUR, from –11% to 4.7% for 5-FU, and from –5.4% to 7.6% for FBAL over the analysis period. The precision of the assay was equal or lower than 4.6%, 7.9%, 8.6%, 12% and 22% for capecitabine, 5'-DFCR, 5'-DFUR, 5-FU and FBAL, respectively.

Pharmacokinetic analysis was performed by non compartmental analysis using WinNonlin software, version 3.3 (Pharsight Corp., Montain View, CA, USA). The following parameters were determined for irinotecan, capecitabine and their metabolites: peak concentration (Cmax), time to Cmax (Tmax), area under the curve from 0 to infinity (AUC) or from 0 to time (t) the last quantifiable concentration (AUC0–t), half-life of the terminal phase (t1/2), the metabolic ratio of AUC0–ts of SN-38 to irinotecan and, finally, plasma clearance (CL) and volume of distribution at steady state (Vss) for irinotecan only.

Statistical analysis was performed using SAS software (SAS version 8.02; SAS Institute Inc, Cary, NC, USA) with a significance level of {alpha} =0.05.

The effect of capecitabine on irinotecan pharmacokinetics was carried out on dose-normalised Cmax, AUC0–t or AUC of irinotecan and SN-38 and on metabolic ratio, after log-transformation using the Proc-Mixed procedure with group taken as fixed effect and with reverse sequence as reference. The 90% confidence intervals (CIs) were estimated.

The effect on irinotecan on the pharmacokinetics of capecitabine was carried out on Cmax, AUC0–t and AUC of capecitabine, 5'-DFCR, 5'-DFUR, 5-FU and FBAL, and on metabolite ratios after log-transformation using the Proc-Mixed procedure with group taken as fixed effect and with initial sequence as reference. The 90% CIs were estimated.


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
A total of 100 patients were included. Of the 43 patients entered into the phase I study, 19 patients were treated at the recommended dose with irinotecan after capecitabine. Fifty-seven patients were treated within the phase II study. All patients were eligible and evaluable for toxicity.

Phase I study
Patient characteristics for the 43 patients evaluated in the phase I cohort are summarised in Table 1. DLTs were experienced at all dose levels and are summarised in Table 2. Dose escalation proceeded to dose level 5 (irinotecan 300 mg/m2 and capecitabine 2500 mg/m2/day). At this dose the level grade 3 and 4 toxicity encountered in later cycles was regarded as unacceptable and cohort expansion at dose level 4 was considered appropriate. MTD was also encountered at dose level 4 (irinotecan 300 mg/m2 and capecitabine 2000 mg/m2/day), at which level five DLTs were experienced in the nine patients treated: neutropenia grade 4 (three patients) and diarrhoea grade 3 (one patient) and grade 4 (one patient). The recommended dose was therefore established at irinotecan 250 mg/m2 and capecitabine 2000 mg/m2/day. As specified per protocol, this dose level was further expanded to a total of 19 patients. This dose level was well tolerated with none of the 19 patients experiencing a haematological DLT. DLT at the recommended dose was experienced in four of 19 patients comprising one patient with grade 3–4 diarrhoea lasting 11 days resulting in grade 3 dehydration. In addition, two further patients developed grade 3 dehydration, and one patient developed grade 3 nausea and vomiting. One patient treated at dose level 1 died from complications arising from treatment-related infection without neutropenia. All treatment-related grade 3 and 4 toxicity experienced at each dose level is summarised in Tables 3 and 4 for haematological and non-haematological toxicities, respectively. At the recommended dose the most significant toxicity was diarrhoea, experienced as grade 3 or 4 in seven out of 19 patients, which was associated with grade 3 dehydration in three patients.


View this table:
[in this window]
[in a new window]
 
Table 1. Patient characteristics

 

View this table:
[in this window]
[in a new window]
 
Table 3. Grade 3 and 4 haematological toxicity (worst grade per patient) in phase I cohort

 

View this table:
[in this window]
[in a new window]
 
Table 4. Phase I grade 3 and 4 non haematological toxicity worst grade by patient and dose level

 
Efficacy in phase I cohort
Thirty-nine patients were evaluable for response. Responses were seen at levels 2, 3 and 4. Two complete responses and 11 partial responses were observed in 43 patients. Overall response rate was 30% (95% CI 17% to 46%). Overall response rate of the recommended dose level 3 was 32% (95% CI 13% to 57%). Median duration of response was 5.2 months (95% CI 4.4–17.5). Median time to progression was 7 months (95% CI 5.3–17.5).

Phase II study
Patient characteristics of the 57 patients treated within the phase II protocol are summarised in Table 1. All patients were eligible for safety and were treated with at least one cycle of chemotherapy at the recommended dose using administration sequence B (irinotecan preceding capecitabine). After evaluation of the first three cycles of the first 15 patients an acceptable safety profile was observed. Accrual was continued for additional 42 patients. A total of 314 cycles were administered (median six per patient; range 1–15). The administration of the combination was delayed in 37 cycles (12%) and the doses were reduced in five cycles (2%) for haematological toxicity and in 14 cycles (14%) for non-haematological toxicity. Treatment was generally well tolerated with a low incidence of grade 3 and 4 toxicity and no treatment-related deaths. Grade 3 neutropenia was reported as worst haematological toxicity in 12 patients and grade 4 neutropenia in five patients, occuring in 8% and 2% of cycles, respectively. There was only one episode of febrile neutropenia. Grade 4 diarrhoea was experienced by seven patients (12%), and grade 3 diarrhoea by four patients (7%), occurring in 2% and 1% of cycles, respectively. There was one episode of grade 3 dehydration. Four patients experienced grade 3 hand–foot syndrome and three patients experienced grade 3 vomiting. Thromboembolic complications were encountered in three patients. The most common all-grade toxicities per patient are summarised in Figure 2.



View larger version (23K):
[in this window]
[in a new window]
 
Figure 2. Toxicity in phase II cohort.

 
The relative dose-intensity for irinotecan was 99.96% (range 83% to 104%) and for capecitabine was 96.77% (17% to 111%).

Efficacy in phase II cohort
Of the 57 patients entered, 46 patients were evaluable for response. Eleven patients were not evaluable for response because they discontinued treatment before the first scheduled evaluation without definite proof of progressive disease. The reasons for withdrawal from the study as assessed by the local investigator were bowel obstruction (four patients), deterioration of performance status (three), nausea/vomiting (one), coronary vein spasms (one), cholinergic syndrome (one) and patient refusal (one).

Two complete responses and 22 partial responses were observed. Stable disease was recorded as best response in 15 patients. The overall response rate (intention-to-treat) was 42% (95% CI 29% to 56%). The median duration of response was 7.7.months (95% CI 7.5–8.9). The median time to progression was 8.3 months (95% CI 5.8–10).

Pharmacokinetics
Pharmacokinetic evaluation was performed in 17 out of 19 patients recruited to the pharmacokinetic protocol in sequence A (capecitabine preceding irinotecan) and in 15 out of 16 patients recruited in sequence B (irinotecan preceding capecitabine). The doses administered in these patients were 250 mg/m2 irinotecan and 2000 mg/m2/day capecitabine in both sequences, with the exception of four patients who received 300 mg/m2 irinotecan and 2000 mg/m2/day capecitabine (sequence A) (Figure 1).

The mean (± SD) irinotecan and SN-38 pharmacokinetic parameters are summarised in Table 5A and B, and the mean (± SD) capecitabine and its metabolites pharmacokinetic parameters are reported in Table 5C. Irinotecan was rapidly metabolised, with SN-38 peak plasma concentrations achieved within 0.83–1 h (median time after the start of infusion), in both sequences.


View this table:
[in this window]
[in a new window]
 
Table 5A. Pharmacokinetic results. Pharmacokinetic parameters of irinotecan on day 1 at cycle 1 [expressed as mean (SD)]

 

View this table:
[in this window]
[in a new window]
 
Table 5B. Pharmacokinetic parameters of SN-38 on day 1 at cycle 1 [expressed as mean (SD)]

 

View this table:
[in this window]
[in a new window]
 
Table 5C. Phamacokinetic parameters of capecitabine and its metabolites on day 1 at cycle 1 [expressed as mean (SD)] in patients receiving 1000 mg/m2 twice daily capecitabine with 250 or 300 mg/m2 irinotecan

 
Interpatient variability of irinotecan pharmacokinetic parameters was generally higher in the initial sequence (A) than the reverse sequence (B) (e.g. 40% to 54% versus 27% for AUC), with the exception of Cmax, for which it was in the same range (10% to 22%). For SN-38, the interpatient variability was high and similar in both sequences (48% to 124%).

The CL and Vss were 14 l/h/m2 and 80 l/m2 when irinotecan was administered as a 30-min infusion at 250 mg/m2 prior to capecitabine (sequence B), and 12.6 l/h/m2 and 83 l/m2, respectively, when capecitabine was administered 2 h before irinotecan infusion (sequence A). Cmax and AUC0–t of SN-38 were 62 µg/l and 396 µg·h/l, respectively, when irinotecan was administered prior capecitabine, and 48.2 µg/l and 269 µg·h/l, respectively, when capecitabine was administered 2 h before irinotecan infusion at 250 mg/m2.

Statistical analysis showed no statistically significant differences in pharmacokinetic parameters for both irinotecan and SN-38 between the two groups, but mean pharmacokinetic parameters of SN-38 were lower in patients taking capecitabine first (sequence A), with large confidence intervals (Table 5D).


View this table:
[in this window]
[in a new window]
 
Table 5D. Statistical analysis of phamacokinetic parameters

 
Capecitabine pharmacokinetics were characterised by a rapid absorption after oral dosing, with maximum plasma concentrations reached at 0.5 h when capecitabine was given 2 h prior to irinotecan (sequence A) or 2 h after the infusion of irinotecan (sequence B).

Capecitabine was thereafter rapidly metabolised into 5'-DFCR, 5'-DFUR, 5-FU and FBAL, with peak levels occurring between 0.5 and 1 h post dosing for 5'-DFUR, 5'-DFCR and 5-FU, and at ~2 h for FBAL.

The main circulating compounds were capecitabine primary metabolite 5'-DFCR and FBAL (5-FU metabolite), in most of the patients in both groups.

A moderate to high interpatient variability was observed in capecitabine and its metabolites for Cmax, and to a lesser extent for AUC, the coefficient of variation varying from 33% to 63% for Cmax and from 13% to 45% for AUC of capecitabine, 5'-DFCR, 5'-DFUR and 5-FU. The interpatient variability of the exposures (Cmax and AUC) of FBAL was lower (18% to 27%).

The statistical analysis showed a statistically significant difference between the two sequences for Cmax and AUC of capecitabine, with an increased Cmax of 64% and an increased AUC of 72% in patients receiving irinotecan first (sequence B). The AUC of 5'-DFCR was found to be significantly 17.2% lower in sequence B while Cmax was also 18.5% lower but in a non-statistically significant manner.

No significant difference in pharmacokinetic parameters of 5'-DFUR, 5-FU and FBAL was detected between the two sequences (Table 5D).

The proportion of patients experiencing any grade 3 or 4 clinical toxicity over the first cycle of treatment in patients treated at the recommended dose was similar, with 21% with sequence A and 20% in sequence B.


    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
This study has established the safety profile of combining irinotecan and capecitabine as first-line treatment for advanced colorectal cancer when irinotecan is administered i.v. on day 1 in combination with capecitabine administered as a twice daily dose on days 1–14 every 3 weeks. The phase I component of this study has established the MTD of irinotecan as 300 mg/m2 and capecitabine as 2000 mg/m2/day, with diarrhoea as the most common DLT. The recommended dose for the phase II study of irinotecan 250 mg/m2 i.v. followed by capecitabine 2000 mg/m2/day was well tolerated, with diarrhoea the most frequently observed toxicity. Grade 3–4 diarrhoea occurred in 19% of patients. A concern with irinotecan used in single-agent studies at higher doses or in combination with bolus doses of 5-FU is the combination of severe diarrhoea and simultaneous myelosuppression. The diarrhoea, indicative of damage to the colonic mucosa, providing an entry portal for gut acquired systemic infection, can result in the potential lethal combination of uncontrolled sepsis and dehydration. In our study we encountered only moderate myelosuppression. Compared with larger studies with infusional 5-FU–leucovorin (LV) regimens in combination with irinotecan, the incidence of grade 3–4 neutropenia of 30% in our study is similar to the study of Tournigand et al. [10Go] (24%) and to the weekly schedule in the study by Douillard et al. [5Go] (29%), but somewhat less than the 2-weekly schedule (46%). Compared with the incidence of 19% grade 3–4 diarrhoea in our study these figures are 13% [10Go], 44% and 13% [5Go], respectively. In our study, cases of severe diarrhoea were accompanied by neutropenic sepsis in only one patient. This observation, and the absence of any treatment-related mortality in the phase II study, is reassuring, and is parallel to the experience of irinotecan in combination with infusional 5-FU–LV regimens [5Go, 10Go]. Patient education is essential in limiting the impact of toxicity, and clear instructions should be provided to patients on the management of diarrhoea and the importance of seeking specialist advice if marked diarrhoea occurs. The incidence of any grade of hand–foot syndrome was 41%, including 7% of grade 3, which is less than experienced with single-agent capecitabine at 2500 mg/m2/day [7Go]. The combination of irinotecan and capecitabine is highly active, with a response rate in our phase II series of 42% (95% CI 29% to 56%). The response rate and median time to progression (8.3 months) are comparable to the results for combinations of 5-FU and irinotecan in randomised studies: Saltz et al. [4Go] (bolus 5-FU) reported 39% and 7 months, Douillard et al. [5Go] (infusional 5-FU) 41% and 6.7 months, and Tournigand et al. [10Go] (infusional 5-FU) 56% and 8.5 months, respectively. Obviously, a true comparison can only be made in a prospective randomised study.

The pharmacokinetics of irinotecan were similar whatever the sequence of administration used, and are consistent with published data obtained from phase I trials of irinotecan administered as a single agent [11Go, 12Go]. The administration of capecitabine 2 h before irinotecan did not modify the pharmacokinetics of irinotecan and its metabolite SN-38, suggesting the absence of a pharmacokinetic interaction of capecitabine on irinotecan with this schedule of administration.

However, irinotecan administration 2 h before administration of capecitabine increased the exposure of capecitabine by 72% and decreased the exposure of 5'-DFCR by 17%. The exposure of 5'-DFUR, 5-FU and FBAL appeared to be not affected by the administration of irinotecan.

Of note, no major differences were observed in 5-FU AUC, whatever the administration order used, and the systemic exposures of 5-FU were consistent with published data when capecitabine is given alone [13Go]. The evaluation of pharmacokinetic interaction was of particular interest because of the complexity of the metabolism of both irinotecan and capecitabine. The principle issue of interest is that the carboxylesterases metabolise irinotecan into its active metabolite SN-38, and also capecitabine into 5'-DFCR. Consequently, a pharmacokinetic drug–drug interaction is theoretically possible.

The pharmacokinetic evaluation performed in this study has not identified a significant effect of capecitabine on irinotecan pharmacokinetics, which contrasts with the observations with infusional 5-FU reported by Falcone et al. [9Go]. These authors reported a reduced clearance of irinotecan when administered in combination with 5-FU and an associated clinical impact with increased toxicity when 5-FU was administered first. Capecitabine and its immediate metabolites appear not to produce the same interaction, and the systemic levels of 5-FU encountered with clinically effective doses of capecitabine appear to have no significant effect on irinotecan metabolism. Unless very dramatic or long lasting, effects of irinotecan on capecitabine pharmacokinetics would not be expected to give rise to clinically observable effects, since capecitabine administration covers a 14-day period. The effect of irinotecan on capecitabine and capecitabine metabolites is, however, complex; capecitabine levels were higher when preceded by irinotecan but this was not reflected in lower concentrations of downstream metabolites, in particular blood 5'-DFUR and 5-FU and FBAL levels were not affected. The lower conversion of capecitabine into 5'-DFCR could be explained by a competition for irinotecan conversion into SN-38 by carboxylesterases.

An interaction of either drug on intracellular drug or metabolite concentrations of the other cannot be excluded, but any such effect is unlikely to be clinically relevant since we were unable to detect any major differences in toxicity profile in relation to the order of drug administration. We can conclude from this study that the timing and order of drug administration on day 1 of this regimen is not as critical as has been demonstrated for irinotecan in combination with 5-FU.

Several studies on the combination of irinotecan and capecitabine have been reported (Table 6). Differences in schedule concern the administration of irinotecan (weekly, weekly x2 followed by 1 week rest, or 3-weekly) and the starting day of capecitabine (day 1 or 2). Our results are similar to the study by Park et al. [14Go], who administered irinotecan at 240 mg/m2 and capecitabine at 2000 mg/m2. This study reported a response rate of 44% with a time to progression of 6.7 months. The toxicity profile was similar to that seen in our study. Bajetta et al. [15Go] initially used in their 3-weekly irinotecan schedule a dose of irinotecan 300 mg/m2 and capecitabine 2500 mg/m2/day, but since this was associated with unacceptable toxicity doses were reduced to irinotecan 240 mg/m2 and capecitabine 2000 mg/m2/day. This arm of their randomised study reported a response rate of 47% and progression-free survival of 8.3 months. Toxicity was more manageable when irinotecan was administered only at day 1 compared with day 1 and 8 of a 3-weekly schedule. Goel et al. [16Go] explored day 1 irinotecan dosing in combination with capecitabine days 2–15 on a 21-day cycle in a mixed population of patients with advanced gastrointestinal tumours. In their study using fixed dosing of capecitabine, the recommended dose was irinotecan 275 mg/m2 with capecitabine 2300 mg/m2/day. Diarrhoea and fatigue were the predominant DLTs. Tewes et al. [17Go] administered irinotecan as a weekly administration for 6 weeks. Diarrhoea, neutropenia and asthenia were the DLTs, and the response rate was 38%. In a recent randomised phase II study by Borner et al. [18Go], a 3-weekly schedule of irinotecan had a comparable response rate compared to a weekly schedule in combination with capecitabine (35% and 34%, respectively), but the 3-weekly schedule seemed advantageous in terms of toxicity (except for alopecia) and survival, although the study was not designed for these parameters.


View this table:
[in this window]
[in a new window]
 
Table 6. Studies with irinotecan plus capecitabine in advanced colorectal cancer

 
Taking these results together, no outright preference for any schedule in terms of efficacy or toxicity is obvious. However, our 3-weekly schedule with irinotecan on day 1 and capecitabine on days 1–14 does appear more convenient compared with other irinotecan/capecitabine schedules, and certainly when compared with 5-FU/FA infusional schedules, which involve ambulatory pumps, permanent vascular access devices and more frequent patient visits. Small but potentially clinically significant differences in efficacy may exist, and phase III studies are currently ongoing to more accurately compare the relative efficacy and toxicity of this regimen in comparison with irinotecan in combination with infusional 5-FU–FA. Our recommended dose of capecitabine and irinotecan is currently being tested in the CAIRO study of the Dutch Colorectal Cancer Group (DCCG), which investigates the sequential versus concomitant use of capecitabine, irinotecan and oxaliplatin. Accrual of 820 patients was completed in December 2004. This regimen is also currently under investigation in the adjuvant context in the recently opened UK Quasar II study. These and other studies will provide relevant data for the use of capecitabine–irinotecan as part of the standard treatment of colorectal cancer.


    Acknowledgements
 
This study was supported by Aventis Pharma, and was presented in part at the 38th annual meeting of the American Society of Clinical Oncology (ASCO) 2002, and the 12th meeting of the European Cancer Conference (ECCO) 2003.

Received for publication February 26, 2005. Accepted for publication March 2, 2005.


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
1. Punt CJA. New options and old dilemmas in the treatment of patients with advanced colorectal cancer. Ann Oncol 2004; 15: 1453–1459.[Abstract/Free Full Text]

2. 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: 1413–1418.[CrossRef][ISI][Medline]

3. Rougier P, Van Cutsem E, Bajetta E et al. Randomised trial of irinotecan versus fluorouracil by continuous infusion after fluorouracil failure in patients with metastatic colorectal cancer. Lancet 1998; 352: 1407–1412.[CrossRef][ISI][Medline]

4. Saltz LB, Cox JV, Blanke C et al. Irinotecan plus fluorouracil and leucovorin for metastatic colorectal cancer. Irinotecan Study Group. N Engl J Med 2000; 343: 905–914.[Abstract/Free Full Text]

5. 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: 1041–1047.[CrossRef][ISI][Medline]

6. Miwa M, Nishida UM, Ishikawa ST et al. Design of a novel oral fluoropyrimidine carbamate, capecitabine which generates 5-fluorouro-uracil selectively in tumours by enzymes concentrated in human liver and cancer tissue. Eur J Cancer 1998; 34: 1274–1281.[CrossRef][ISI][Medline]

7. Van Cutsem E, Twelves C, Cassidy J et al. Oral capecitabine compared with intravenous fluorouracil plus leucovorin in patients with metastatic colorectal cancer: results of a large phase III study. J Clin Oncol 2001; 19: 4097–4106.[Abstract/Free Full Text]

8. Hoff PM, Ansari R, Batist G et al. Comparison of oral capecitabine versus intravenous fluorouracil plus leucovorin as first-line treatment in 605 patients with metastatic colorectal cancer: results of a randomized phase III study. J Clin Oncol 2001; 19: 2282–2292.[Abstract/Free Full Text]

9. Falcone A, Di Paolo A, Masi G et al. Sequence effect of irinotecan and fluorouracil treatment on pharmacokinetics and toxicity in chemotherapy-naive metastatic colorectal cancer patients. J Clin Oncol 2001; 19: 3456–3462.[Abstract/Free Full Text]

10. Tournigand C, André T, Achille E et al. FOLFIRI followed by FOLFOX6 or the reversed sequence in advanced colorectal cancer: a randomized GERCOR study. J Clin Oncol 2004; 22: 229–237.[Abstract/Free Full Text]

11. Chabot GG, Abigerges D, Catimel G et al. Population pharmacokinetics and pharmacodynamics of irinotecan (CPT-11) and active metabolites SN-38 during phase I trials. Ann Oncol 1995; 6: 141–151.[Abstract]

12. Abigerges D, Chabot GG, Armand JP et al. Phase I and pharmacologic studes of the camptothecin analog irinotecan administered every 3 weeks in cancer patients. J Clin Oncol 1995; 13: 210–221.[Abstract/Free Full Text]

13. Pronk LC, Vasey P, Sparreboom A et al. A phase I and pharmacokinetic study of the combination of capecitabine and docetaxel in patients with advanced solid tumours. Br J Cancer 2000; 83: 22–29.[CrossRef][Medline]

14. Park SH, Bang SM, Cho EK et al. First-line chemotherapy with irinotecan plus capecitabine for advanced colorectal cancer. Oncology 2004; 66: 353–357.[CrossRef][ISI][Medline]

15. Bajetta E, Di Bartolomeo M, Mariani L et al. Randomized multicenter Phase II trial of two different schedules of irinotecan combined with capecitabine as first-line treatment in metastatic colorectal carcinoma. Cancer 2004; 100: 279–287.[CrossRef][ISI][Medline]

16. Goel S, Jhawer M, Rajdev L et al. Phase I clinical trial of irinotecan with oral capecitabine in patients with gastrointestinal and other solid malignancies. Am J Clin Oncol 2002; 25: 528–534.[CrossRef][ISI][Medline]

17. Tewes M, Schleucher N, Achterrath W et al. Capecitabine and irinotecan as first line chemotherapy in patients with metastatic colorectal cancer: results of an extended phase I study. Ann Oncol 2003; 9: 1442–1448.[CrossRef]

18. Borner MM, Bernhard J, Dietrich D et al. A randomized phase II trial of capecitabine and two different schedules of irinotecan in first-line treatment of metastatic colorectal cancer: eficacy, quality-of-life and toxicity. Ann Oncol 2005; 16: 282–288.[Abstract/Free Full Text]





This Article
Abstract
Full Text (PDF)
All Versions of this Article:
16/7/1123    most recent
mdi227v1
E-letters: Submit a response
Alert me when this article is cited
Alert me when E-letters are posted
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Disclaimer
Request Permissions
Google Scholar
Articles by Rea, D. W.
Articles by Punt, C. J. A.
PubMed
PubMed Citation
Articles by Rea, D. W.
Articles by Punt, C. J. A.