- Email: [email protected]
folinic acid or capecitabine plus celecoxib or placebo in the first-line treatment of patients with metastatic colorectal cancer. EORTC study 40015
original article
Annals of Oncology 19: 920–926, 2008 doi:10.1093/annonc/mdm544 Published online 6 December 2007
Irinotecan combined with infusional 5-fluorouracil/ folinic acid or capecitabine plus celecoxib or placebo in the first-line treatment of patients with metastatic colorectal cancer. EORTC study 40015 C.-H. Ko¨hne1*, J. De Greve2, J. T. Hartmann3, I. Lang4, P. Vergauwe5, K. Becker6, D. Braumann7, E. Joosens8, L. Mu¨ller9, J. Janssens10, C. Bokemeyer11, P. Reimer12, H. Link13, E. Spa¨th-Schwalbe14, H.-J. Wilke15, H. Bleiberg16, J. Van Den Brande17, M. Debois18, U. Bethe18 & E. Van Cutsem19
Received 29 July 2007; revised 28 October 2007; accepted 29 October 2007
Background: The study aimed to demonstrate the noninferiority of capecitabine to 5-fluorouracil (5-FU)/folinic acid (FA), in relation to progression-free survival (PFS) after first-line treatment of metastatic colorectal cancer and the benefit of adding celecoxib (C) to irinotecan/fluoropyrimidine regimens compared with placebo (P). Patients and methods: Patients were randomly assigned to receive FOLFIRI: irinotecan (180 mg/m2 i.v. on days 1, 15 and 22); FA (200 mg/m2 i.v. on days 1, 2, 15, 16, 29 and 30); 5-FU (400 mg/m2 i.v. bolus, then 22-h, 600 mg/m2 infusion) or CAPIRI: irinotecan (250 mg/m2 i.v. infusion on days 1 and 22); capecitabine p.o. (1000 mg/m2 b.i.d. on days 1–15 and 22–36). Patients were additionally randomly assigned to receive either placebo or celecoxib (800 mg: 2 · 200 mg b.i.d.). Results: The trial was closed following eight deaths unrelated to disease progression in the 85 enrolled (629 planned) patients. Response rates were 22% for CAPIRI + C, 48% for CAPIRI + P, 32% for FOLFIRI + C and 46% for FOLFIRI + P. Median PFS and overall survival (OS) times were shorter for CAPIRI versus FOLFIRI (PFS 5.9 versus 9.6 months and OS 14.8 versus 19.9 months) and celecoxib versus placebo (PFS 6.9 versus 7.8 months and OS 18.3 versus 19.9 months). Conclusion: Due to the small sample size following early termination, no definitive conclusions can be drawn in relation to the noninferiority of CAPIRI compared with FOLFIRI. Key words: capecitabine, celecoxib, colorectal, first line, 5-FU, irinotecan
introduction Colorectal cancer (CRC) remains one of the most common causes of cancer-related mortality [1]. In Europe alone, where it is the second most commonly diagnosed cancer, it is estimated *Correspondence to: Prof. C.-H. Ko¨hne, Department of Oncology and Hematology, Klinikum Oldenburg GmbH, Dr Eden Strasse 10, 26133 Oldenburg, Germany. Tel: +49-441-403-2611; Fax: +49-441-403-2654; E-mail: [email protected]
that CRC accounted for >207 000 deaths in 2006 [2]. Surgery is an effective curative treatment strategy for disease that is restricted to and resectable at the primary site. In addition, metastatic colorectal cancer (mCRC) may be curable, especially if hepatic and/or lung lesions are resectable, either at diagnosis or following effective first-line treatment [3]. The further improvement of first-line regimens is therefore of great importance, not only in relation to increasing the number of patients amenable to secondary surgery for
ª The Author 2007. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected]
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 26, 2015
Department of Oncology and Hematology, Klinikum Oldenburg, Oldenburg, Germany; 2Department of Oncology, Universitair Ziekenhuis Brussel, Vrije Universiteit, Brussels, Belgium; 3University Medical Center II, Eberhard-Karls-Universita¨t, Tu¨bingen, Germany; 4Department of Medical Oncology, National Institute of Oncology, Budapest, Hungary; 5Department of Internal Medicine, AZ Groeninghe, Campus Sint-Niklaas, Kortrijk, Belgium; 6Onkologischer Schwerpunkt Lerchenfeld, Hamburg; 7 II. Medizinische Abteilung, Allgemeines Krankenhaus Altona, Hamburg, Germany; 8Department of Hemato-Oncology, Ziekenhuis Netwerk Middelheim, Campus Middelheim, Antwerpen, Belgium; 9Onkologische Schwerpunktpraxis Leer, Leer, Germany; 10Department of Gastroenterology, Sint-Elisabeth Ziekenhuis, Turnhout, Belgium; 11Department of Oncology and Hematology, Universita¨tsklinikum Hamburg-Eppendorf, Hamburg; 12Medizinische Klinik III, Wu¨rzburg; 13Medizinische Klinik I, Westpfalz-Kinikum, Kaiserslautern; 14Medical Department, Hematology, Oncology and Gastroenterology, Vivantes Klinkum Spandau, Berlin; 15Department of Oncology, Kliniken Essen-Mitte, Essen, Germany; 16Department of Gastroenterology, Chemotherapy Unit, Institut Jules Bordet, Brussels; 17Department of Medical Oncology, Universitair Ziekenhuis Antwerpen, Edegem; 18European Organisation for Research and Treatment of Cancer (EORTC), EORTC Data Center, Brussels; 19 Digestive Oncology Unit, University Hospital Gasthuisberg, Leuven, Belgium
original article
1
original article
Annals of Oncology
Volume 19 | No. 5 | May 2008
were designed to show that the NSAID would similarly reduce the speed of recurrence of CRC in patients responding to first-line treatment.
patients and methods main eligibility criteria Patients, ‡18 years of age, with previously untreated (i.e. not treated with chemotherapy for metastatic disease), metastatic, histologically verified, adenocarcinoma of the colon or rectum were eligible. Other criteria for study entry were a World Health Organization performance status (PS) of two or less and measurable disease according to the Response Evaluation Criteria in Solid Tumors (RECIST) [17], located outside the field of any radiotherapy, which, if given, had been completed at least 4 weeks before randomization. Prior adjuvant therapy was allowed, provided that it had been completed 6 months before randomization, but no current or planned NSAID administration other than low-dose prophylactic aspirin was permitted and no other investigational drugs were to have been administered within 4 weeks before randomization. Patients were required to have adequate hematological, renal and hepatic function, no central nervous system metastases, second malignancies, severe cardiac disease, active Crohn’s disease or any other uncontrolled severe medical condition. Before patients were entered into the trial, they gave written informed consent in accordance with either local/national regulations or the principles defined in the Declaration of Helsinki, whichever provided the greater protection for the patient.
trial design This was a prospective, multicenter, two-by-two factorial, phase III, randomized study. As stated, the two main objectives were to demonstrate that capecitabine could replace 5-FU/FA as the fluoropyrimidine component of an irinotecan-including combination without compromising PFS and that COX-2 inhibition would improve PFS compared with placebo. The primary end point was therefore PFS. Secondary end points included safety (assessed in all patients receiving ‡1 dose of any study drug according to the National Cancer Institute Common Toxicity Criteria, version 2), response rate, time to treatment failure and overall survival (OS) assessed in the intention-to-treat (ITT) population. Treatment cycles lasted 6 weeks. Patients were randomly assigned to receive, in the first instance to, FOLFIRI or CAPIRI. The FOLFIRI arm received: irinotecan (180 mg/m2), as a 30- to 90-min i.v. infusion on days 1, 15 and 22; FA (200 mg/m2), as a 2-h infusion on days 1, 2, 15, 16, 29 and 30 (1 h after irinotecan on days 1, 15 and 29); 5-FU, as a 400-mg/m2 bolus given after FA, followed by 22-h continuous infusion (600 mg/m2) given after the bolus (days 1, 2, 15, 16, 29 and 30). The CAPIRI arm received: irinotecan (250 mg/m2), as a 30- to 90-min i.v. infusion on days 1 and 22 and capecitabine p.o. (1000 mg/m2, twice daily) on days 1–15 and 22–36. Within these arms, to complete the 2 · 2 design, patients were randomly assigned to receive either placebo or celecoxib (800 mg: as 2 · 200 mg, twice daily, before irinotecan, when administered). Treatment continued up to a planned total of six cycles or until disease progression, unacceptable toxicity or withdrawal of patient consent. Responding patients or those with stable disease were allowed to continue treatment beyond six cycles at the discretion of the investigator. Evaluation of disease status was carried out every 6 weeks during treatment and every 8 weeks subsequently until the documentation of disease progression. Response rate was assessed by the investigators through the radiological measurement of target lesions identified at baseline (maximum of 10 to be representative of all involved organs), according to RECIST criteria. The duration of response was defined as the time interval from when the measurement criteria for a complete or partial
doi:10.1093/annonc/mdm544 | 921
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 26, 2015
metastases but also in the context of disease management with regard to improving the quality of life and lengthening the survival of nonresectable patients. For many years, 5-fluorouracil (5-FU), generally combined with the biological response modifier folinic acid (FA) [4], has been the mainstay of treatment regimens for mCRC, with meta-analyses suggesting that infusional delivery is more effective [5] and associated with less grade 3/4 hematological toxicity than bolus administration [6]. The introduction of two further cytotoxic chemotherapy agents, irinotecan and oxaliplatin, have more recently allowed the development of multiline combination therapy for mCRC patients. The European Organisation for Research and Treatment of Cancer (EORTC) Gastrointestinal Group and others were able to demonstrate that median survival times of >20 months are now achievable [7, 8], with the use of three active drugs during the course of a patient’s treatment conferring maximum survival benefit [9]. It is anticipated that the use of targeted biological agents in this setting will further improve the performance of existing cytotoxic therapies [10]. Currently, infusional 5-FU/FA combined with irinotecan (FOLFIRI) or oxaliplatin (FOLFOX) are regarded as standard first-line chemotherapy regimens of equivalent efficacy. The devices used during the infusion process may be associated with infection or thrombosis [11]. One possible solution to this problem might be the oral delivery of fluorouracil. Capecitabine is an inactive oral prodrug which is preferentially enzymatically converted in tumors to fluorouracil [12]. Although there are no reported studies comparing the safety and efficacy of capecitabine with infusional 5-FU, the oral prodrug has been shown to be less toxic and at least as effective as the Mayo Clinic bolus 5-FU regimen [13, 14]. As combination therapy is now the first-line treatment of choice for most mCRC patients, opportunities to formally compare the efficacy and safety of infusional 5-FU/FA monotherapy versus capecitabine monotherapy are limited. Such information, however, may be derived indirectly by comparing regimens where the 5-FU/FA component of the comparator arm is replaced by capecitabine in the experimental arm. With this consideration in mind, the first main objective of this 2 · 2 study was to demonstrate that infusional 5-FU/FA component of FOLFIRI can be replaced by oral capecitabine (CAPIRI), without compromising efficacy, with the intention being to show noninferiority of CAPIRI to FOLFIRI in terms of progression-free survival (PFS). The second main objective of the study was to demonstrate that cyclooxygenase (COX)-2 inhibition by celecoxib improves PFS compared with placebo. COX-2 [15], the product of the PTGS2 gene, is a mitogen-inducible enzyme that, with the constitutively expressed COX-1, regulates a key step in the biosynthesis of prostaglandins. It is thought that COX-2 is involved in the process of inflammation [15]. Celecoxib is a nonsteroidal anti-inflammatory drug (NSAID) which shows high specificity for COX-2 over COX-1 inhibition. A double-blind placebo-controlled study demonstrated that the twice-daily administration of celecoxib was associated with a significant regression of colorectal adenomas in patients with the CRC-predisposing syndrome, familial adenomatous polyposis [16]. The celecoxib versus placebo arms of this trial
original article
Annals of Oncology
response were met until the first date that recurrent or progressive disease (PD) was documented. PFS was calculated as the time from randomization until the first report of PD or death. Patients with no evidence of PD at the time of their last visit were censored at that time.
statistical methods and considerations
results patient demographics, treatment compliance and deaths The trial was activated on 21 May 2003. On 23 April 2004, after the enrollment of only 85 patients, recruitment was suspended as a consequence of seven deaths not due to disease progression. One more patient subsequently died following the suspension of recruitment. Six deaths occurred in patients receiving CAPIRI and two in those receiving FOLFIRI.
adverse events Three patients (4%) did not receive study drugs and are therefore not included in the safety analysis. At least one grade 3/4 adverse event (AE) was reported for 62% of patients. The combined frequency of grade 3/4 AEs was higher for patients in the capecitabine arms compared with those in the FOLFIRI arms (32/43, 74% versus 19/39, 49%). The most common grade 3/4 AE was diarrhea, followed by WBC toxicity. AEs occurring in two or more patients in any one arm are detailed in Table 3. Hand–foot syndrome (HFS), which is commonly associated with capecitabine administration, was
Table 1. Early deaths and relationship to study treatment Treatment
Total number of cycles
Time since last treatment (days)
Relatednessa
Agreed classificationa
CAPIRI + C CAPIRI + C CAPIRI + P CAPIRI + P CAPIRI + P CAPIRI + P FOLFIRI + C FOLFIRI + C
3 1 1 1 2 1 1 1
65 4 6 75 9 5 12 2
Nonrelated Exacerbated Exacerbated Related Related Related Related Exacerbated
Sudden cardiac arrest Thromboembolic event/pulmonary embolism Pulmonary embolism Diarrhea/neutropenia/septic shock Diarrhea/myocardial infarction Diarrhea/suspected pulmonary embolism Neutropenic sepsis Pneumonia
a
Classified by panel of experts. CAPIRI, capecitabine/irinotecan; C, celecoxib; P, placebo; FOLFIRI, 5-fluorouracil/folinic acid/irinotecan.
922 | Ko¨hne et al.
Volume 19 | No. 5 | May 2008
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 26, 2015
Patients were randomly assigned to receive centrally using a minimization technique [18]. Stratification factors were the institution, previous adjuvant therapy (yes/no) and risk groups, as defined by Ko¨hne et al. [19] [low risk: patients with a PS of 1 or 0 and only one tumor site; intermediate risk: patients with a good PS (<1), but more than one tumor site plus an alkaline phosphatase (ALP) level of <300 U/l, or those with a poor PS, a low white blood cell (WBC) count and only one tumor site; high risk: patients with good PS but more than one tumor site and a high ALP level, or a poor PS plus a high WBC count or a poor PS, a low WBC count and more than two tumor sites]. All efficacy analyses were carried out on the ITT population. Unacceptable inferiority of capecitabine over 5-FU/FA in relation to PFS, the primary end point, was defined by a hazard ratio (HR) ‡1.25. Given a one-sided alpha level of 2.5%, it was estimated that 632 events were needed to exclude a difference of this magnitude with 80% probability. This number of events would also allow the detection of a 2-month difference between the celecoxib and placebo arms with a power of 89% and a two-sided 5% significance level test. It was consequently estimated that 692 patients should be randomized (1 : 1) to generate this number of events. These sample size determinations were on the basis of the assumptions of an absence of interaction between the 5-FU versus capecitabine and the celecoxib versus placebo effects. Time to event variables were analyzed using the Kaplan–Meier technique [20] and differences summarized with HRs and confidence intervals (CIs). Quantitative variables were summarized using counts and percentages. The study was approved by the EORTC protocol review committee and by local authorities and ethic committees of the participating centers.
Celecoxib treatment allocation was unblinded at this stage. The individual hospital files were inspected and discussed with the physician to determine whether the deaths were related to or exacerbated by the study treatment (Table 1) and whether any risk factors for thromboembolic events or gastrointestinal toxicity were present at randomization. Five deaths in the CAPIRI arm and both of those in the FOLFIRI arm were deemed to be treatment related. Underlying risk factors could not be identified as a likely explanation for these fatal toxic effects. On the basis of the outcome of this review, the trial was officially closed on 12 January 2005. This report is consequently on the basis of the data available from each of the 85 eligible patients as of 16 December 2005. Median follow-up time was 14.6 months (95% CI 13.1–16.8). Patient characteristics, described in Table 2, were similar between treatment arms. Table 2 also details treatment exposure. Dose reductions, mainly as a consequence of gastrointestinal toxicity, were more common in the CAPIRI compared with the FOLFIRI arms, with 53% versus 33% of patients, respectively, experiencing at least one cycle with a reduction. Thirty-four patients (41.5%) experienced treatment delays, which were more common in the FOLFIRI compared with the CAPIRI arms, with 54% of patients versus 30%, respectively, experiencing at least one cycle with delay. The relative dose intensity for capecitabine and 5-FU did not differ markedly in their placebo arms (82.4% versus 84.8%, respectively) but was lower for capecitabine if celecoxib was also administered (66.4% versus 92.1% for 5-FU). Interestingly, very little difference in the irinotecan dose was observed across all study arms (Table 2; range 83.1%–88.4%).
original article
Annals of Oncology
Table 2. Patient characteristics and treatment exposure CAPIRI + P
FOLFIRI + C
FOLFIRI + P
Total
n = 23 12/11, (52/48)
n = 21 12/9, (57/43)
n = 19 15/4, (79/21)
n = 22 14/8, (64/36)
n = 85 53/32, (62/38)
12 (52) 8 (35) 3 (13)
8 (38) 8 (38) 5 (24)
10 (53) 8 (42) 1 (5)
14 (64) 6 (27) 2 (9)
44 (52) 30 (35) 11 (13)
63.0
65.0
(42–75) 15 (65) 8 (35)
(43–78) 11 (52) 10 (48)
12 (52) 6 (26) 5 (22)
10 (48) 6 (29) 5 (24)
14 (61) 8 (35) 1 (4) 12 (52) 11 (48) 3 (1–6)
Patients who started n = 23 chemotherapy Dose reductions in at least 13 (57) one cycle, n (%) Delays in at least one cycle, n (%) 8 (35) Median relative dose intensities, % (range) Capecitabine 66.4 (6.6–110.3) 5-FU Irinotecan 83.1 (32.6–103.3) Celecoxib/placebo 99.2 (66.7–100)
66.0 (54–77) 9 (47) 10 (53)
60.5
64.0
(45–75) 14 (64) 8 (36)
(42–78) 49 (58) 36 (42)
9 (47) 7 (37) 3 (16)
11 (50) 8 (36) 3 (14)
42 (49) 27 (32) 16 (19)
12 (57) 8 (38) 1 (5)
10 (53) 9 (47) 0 (0)
14 (64) 8 (36) 0 (0)
50 (59) 33 (39) 2 (2)
11 (52) 10 (48) 3 (0–6)
9 (47) 10 (53) 4 (0–6)
11 (50) 11 (50) 6 (0–6)
43 (51) 42 (49) 4 (0–6)
n = 20
n = 18
n = 21
n = 82
10 (50)
3 (17)
10 (48)
36 (44)
5 (25)
8 (44)
13 (62)
34 (41)
82.4 (47.5–119.6) 83.6 (47.5–101.7) 98.3 (59.5–101.2)
92.1 (21.2–107.4) 88.4 (20.9–98.6) 96.2 (37.6–100.0)
83.8 (51.7–103.1) 84.0 (51.6–103.3) 94.4 (81.6–100.0)
73.7 86.8 84.1 96.9
(6.6–119.6) (21.2–107.4) (20.9–103.3) (37.6–101.2)
CAPIRI, capecitabine/irinotecan; C, celecoxib; P, placebo; FOLFIRI, 5-fluorouracil/folinic acid/irinotecan; WHO, World Health Organization; PS, performance status; 5-FU, 5-fluorouracil.
restricted to 14 patients who received the drug (seven in each arm), reaching grade 3 in only two (one in each arm). Five patients in the FOLFIRI arms experienced grade 1 HFS reactions.
efficacy The best overall response to treatment is detailed in Table 3. The number of patients in each arm was too small to allow reliable conclusions to be drawn. Response rates, however, were higher in both placebo arms compared with the corresponding celecoxib arms: 22% for CAPIRI plus celecoxib versus 48% for CAPIRI plus placebo and 32% for FOLFIRI plus celecoxib versus 46% for FOLFIRI plus placebo. The effect of capecitabine versus 5-FU/FA and celecoxib versus placebo on PFS and OS were analyzed with stratification for the nonexamined variable. Survival data are detailed in Table 4 and Figure 1. Median PFS times were shorter for CAPIRI versus
Volume 19 | No. 5 | May 2008
FOLFIRI (5.85 versus 9.6 months) and celecoxib versus placebo (6.9 versus 7.75 months). Median OS times were shorter for CAPIRI versus FOLFIRI (14.75 versus 19.9 months) and similar for celecoxib versus placebo (18.3 versus 19.9 months).
discussion This study aimed to demonstrate in patients with mCRC that capecitabine was not inferior to 5-FU/FA in terms of PFS and that the addition of celecoxib to CAPIRI and/or FOLFIRI improved PFS compared with placebo. Seven deaths unrelated to disease progression, however, occurred early on in the study (five in the CAPIRI arms), with a further death (CAPIRI arm) occurring after the suspension of recruitment. The deaths were primarily linked to gastrointestinal or thromboembolic events. Sudden deaths linked to such causes have previously been noted for regimens combining irinotecan
doi:10.1093/annonc/mdm544 | 923
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 26, 2015
Variable Sex (M/F), n (%) Primary site, n (%) Colon Rectum Colorectal Age, years Median Range £65 years, n (%) >65 years Risk group, n (%) [19] Good Intermediate Poor WHO PS, n (%) 0 1 2 Metastatic sites, n (%) 1 >1 Median number of cycles of chemotherapy, n (range)
CAPIRI + C
original article
Annals of Oncology
Table 3. Best overall response to treatment and grade 3/4 adverse events reported for two or more of the patients, in any one arm, who started treatment CAPIRI + P
FOLFIRI + C
FOLFIRI + P
Total
Best overall response n (%) CR PR SD Progressive disease Early death Not assessable Response rate (CR + PR) Disease control rate (CR + PR + SD)
n = 23 0 5 (22) 12 (52) 2 (9) 2 (9) 2 (9) 5 (22) 17 (74)
n = 21 1 (5) 9 (43) 5 (24) 2 (10) 3 (14) 1 (5) 10 (48) 15 (71)
n = 19 1 (5) 5 (26) 7 (37) 2 (11) 3 (16) 1 (5) 6 (32) 13 (68)
n = 22 0 10 (45) 9 (41) 3 (14) 0 0 10 (45) 19 (86)
n = 85 2 (2) 29 (34) 33 (39) 9 (11) 8 (9) 4 (5) 31 (36) 64 (75)
Adverse event, n (%) Diarrhea Vomiting Nausea Gastrointestinal, other Cardiovascular Febrile neutropenia Hepatic toxicity White blood cells Hemoglobin Renal toxicity All grade 3/4 events
n = 23 9 (39) 2 (9) 1 (4) 2 (9)b 3 (13)b 0 2 (9) 3 (13)c 2 (9) 2 (9)b 19 (83)
n = 20 7 (35) 1 (5)b 1 (5) 0 1 (5) 2 (10)b 1 (5) 3 (15)c 0 0 13 (65)
n = 18 3 (17) 1 (6) 1 (6) 0 1 (6) 2 (11)b 1 (6) 2 (11)b 0 0 8 (44)
n = 21 2 (10)b 1 (5) 2 (10) 1 (5) 0 0 1 (5) 4 (19)b 0 0 11 (52)
n = 82 21 (26) 5 (6) 5 (6) 3 (4) 5 (6) 4 (5) 5 (6) 12 (15) 2 (2) 2 (2) 51 (62)
a
Confirmation was documented in only 74% of responders, although intended in each case. One patient at grade 4. c Two patients at grade 4. CAPIRI, capecitabine/irinotecan; C, celecoxib; P, placebo; FOLFIRI, 5-fluorouracil/folinic acid/irinotecan; CR, complete response; PR, partial response; SD, stable disease. b
and bolus 5-FU/FA [21]. Following the consideration of all available information, a decision was taken to close the trial after the enrollment of only 85 of the planned 692 patients had been achieved. Patients who had tolerated the treatment were allowed to continue with their assigned regimen. Given the small number of patients involved, no formal definitive conclusions can be drawn from the study data. Although caution is required in relation to the small numbers of analyzable patients, it should be noted that due primarily to a higher level of treatment-associated toxicity, patients receiving CAPIRI tended to receive fewer cycles of planned treatment than those receiving FOLFIRI and more patients required a dose reduction (53% versus 33, respectively). The efficacy data from this study are nevertheless consistent with those reported for the randomized, 3 · 2 factorial BICC-C trial, which assessed whether celecoxib added to FOLFIRI, CAPIRI or a modified irinotecan, bolus 5-fluorouracil and folinic acid (m-IFL) regimen improved efficacy and/or reduced toxicity. Median time to progression and OS times in this trial were longer in the patients who received FOLFIRI compared with those who received CAPIRI or m-IFL [22]. The most common grade 3/4 AE observed in the current study was diarrhea, which occurred significantly more frequently in the patients receiving CAPIRI than FOLFIRI (37% versus 13%). The dose levels of capecitabine and irinotecan initially selected were the same as those recommended, and found to be well tolerated by 76 patients
924 | Ko¨hne et al.
Table 4. Progression-free and overall survivals for CAPIRI versus FOLFIRI and celecoxib versus placebo Parameter PFS Median, months (95% CI) 1 year, % (95% CI) Hazard ratio, (95% CI) Overall survival Median, months (95% CI) 1 year, % (95% CI) Hazard ratio, (95% CI)
PFS Median, months (95% CI) 1 year, % (95% CI) Hazard ratio, (95% CI) Overall survival Median, months (95% CI) 1 year, % (95% CI) Hazard ratio, (95% CI)
CAPIRI, n = 44
FOLFIRI, n = 41
5.9 (4.4–8.9) 22.6 (11.4–36.2) 1.00
9.6 (6.9–10.9) 29.3 (16.4–43.4) 0.76 (0.48–1.21)
14.75 (10.7–18.3) 53.5 (36.0–68.2) 1.00
19.9 (18.9, NR) 84.9 (69.4–92.9) 0.31 (0.14–0.71)
Celecoxib, n = 42
Placebo, n = 43
6.9 (5.5–10.4) 20.1 (9.5–33.4) 1.00
7.8 (6.0–12.0) 31.4 (18.2–45.5) 0.90 (0.56–1.43)
18.3 (10.2, NR) 65.2 (48.1–77.8) 1.00
19.9 (16.7, NR) 74.6 (57.6–85.5) 0.80 (0.39–1.63)
CAPIRI, capecitabine/irinotecan; FOLFIRI, 5-fluorouracil/folinic acid/ irinotecan; PFS, progression-free survival; CI, confidence interval; NR, not reached.
Volume 19 | No. 5 | May 2008
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 26, 2015
CAPIRI + C a
Annals of Oncology
original article
in a recent phase I/II trial [23]. Similarly, in a large phase III study of combination chemotherapy with capecitabine, irinotecan and oxaliplatin in 820 advanced CRC patients, CAPIRI was again found to be generally well tolerated [24]. In the current study, considering the observed toxicity, CAPIRI at this dose and schedule, however, was not considered feasible. Toxicity was also generally elevated in the CAPIRI arm of the BICC-C trial compared with the FOLFIRI or m-IFL arm, with the regimen associated with the highest rates of nausea, vomiting, diarrhea, dehydration and HFS. It is perhaps noteworthy that the current and BICC-C studies, both of which report high levels of toxicity with CAPIRI, similarly included the use of celecoxib. These analyses raise the question of whether a lower capecitabine dose may have been more effective. In this context, it is interesting to note that in the TREE-2 study, capecitabine was reduced to 850 mg/m2 twice daily in combination with oxaliplatin (CAPOX) and bevacizumab after high levels of toxicity were reported for the corresponding CAPOX arm in the initial TREE-1 investigation, during which capecitabine had been administered at 1000 mg/m2 twice daily [25]. Similarly, in the phase III X-ACT study of adjuvant capecitabine versus 5-FU/FA, the 1250-mg/m2 twice-daily dose of capecitabine required modification (delay, reduction and interruption) in 57% of patients [26]. One factor that should be noted is that the tolerability of fluoropyrimidines may differ according to the geographical region from which patients are drawn [27]. Further studies to determine the most appropriate dose of capecitabine in CAPIRI and other combination regimens for particular geographic and/or ethnic patient groups may therefore be warranted.
Volume 19 | No. 5 | May 2008
Celecoxib has been shown to reduce the incidence of colorectal polyps and induce their regression in the Min mouse model of the human CRC-predisposing syndrome, adenomatous polyposis coli (APC) [28]. Similarly, in humans, the daily administration of celecoxib to APC patients has been shown to lead to a significant reduction in the number of colorectal polyps [16]. More recently, in a blinded review of a large randomized trial, the twice-daily use of this agent in polypectomy patients, however, was associated with a dose-related increase in the risk of serious cardiovascular events [29]. While such associations argue strongly against the use of celecoxib at this dose and schedule in a chemopreventive setting, the additional toxicity risks may be acceptable in the treatment of advanced disease, provided that they are balanced by a significant benefit in survival. The data from this study, however, indicate no such efficacy benefit. Indeed, in view of the fact that response rates were lower in both celecoxib arms compared with the corresponding placebo arms for both regimens, it is possible that celecoxib may actually reduce the response to chemotherapy, a hypothesis that warrants further preclinical exploration. Celecoxib did not appear to modulate the toxicity of the chemotherapy; thus it is very unlikely that the toxicity observed with CAPIRI was due to celecoxib. A recent randomized phase II trial involving 81 mCRC patients has also failed to show a benefit from adding celecoxib (400 mg b.i.d.) to FOLFIRI [30]. The response rates in the FOLFIRI versus FOLFIRI plus celecoxib arm were 41% versus 35% with median time to progression and OS times of 8 versus 9 months and 16 versus 19 months, respectively. Toxicity was generally mild in both arms, with
doi:10.1093/annonc/mdm544 | 925
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 26, 2015
Figure 1. Kaplan–Meier plots of progression-free survival (panels A and B) and overall survival (panels C and D) plots for CAPIRI versus FOLFIRI (A and C) and celecoxib versus placebo (B and D). CAPIRI, capecitabine/irinotecan; FOLFIRI, 5-fluorouracil/folinic acid/irinotecan.
original article gastrointestinal disorders being most common. Although this study confirms that FOLFIRI is an effective and welltolerated regimen, data from larger randomized studies will be needed before formal definitive conclusions about the benefits of adding celecoxib to this combination can be drawn. It should be noted in this context that in relation to time to progression and OS, celecoxib failed to significantly improve the efficacy of first-line irinotecan/fluoropyrimidine combinations compared with placebo in the 430 patient BICC-C study [22]. In summary, CAPIRI at this dose and schedule was not considered feasible in this setting. Small sample size and confounding safety issues did not allow valid conclusions to be drawn concerning the relative efficacy of CAPIRI versus FOLFIRI. Consistent with other studies, no benefit was seen from adding celecoxib to irinotecan/fluoropyrimidine regimens.
Roche; Pharmacia (currently Pfizer); Aventis (currently sanofiaventis Group).
acknowledgements We would like to thank the patients who took part in this study and the medical and support staff at the collaborating centers. The contents of the present publication as well as the data collection and data analysis are solely the responsibility of the authors.
references 1. Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, 2002. CA Cancer J Clin 2005; 55: 74–108. 2. Ferlay J, Autier P, Boniol M et al. Estimates of the cancer incidence and mortality in Europe in 2006. Ann Oncol 2007; 18: 581–592. 3. Folprecht G, Grothey A, Alberts S et al. Neoadjuvant treatment of unresectable colorectal liver metastases: correlation between tumour response and resection rates. Ann Oncol 2005; 16: 1311–1319. 4. Thirion P, Michiels S, Pignon JP et al. Modulation of fluorouracil by leucovorin in patients with advanced colorectal cancer: an updated meta-analysis. J Clin Oncol 2004; 22: 3766–3775. 5. Meta-analysis Group In Cancer. Efficacy of intravenous continuous infusion of fluorouracil compared with bolus administration in advanced colorectal cancer. J Clin Oncol 1998; 16: 301–308. 6. Toxicity of fluorouracil in patients with advanced colorectal cancer: effect of administration schedule and prognostic factors. Meta-Analysis Group In Cancer. J Clin Oncol 1998; 16: 3537–3541. 7. Ko¨hne CH, van Cutsem E, Wils J et al. Phase III study of weekly high-dose infusional fluorouracil plus folinic acid with or without irinotecan in patients with metastatic colorectal cancer: European Organisation for Research and Treatment of Cancer Gastrointestinal Group Study 40986. J Clin Oncol 2005; 23: 4856–4865. 8. Tournigand C, Andre T, Achille E et al. FOLFIRI followed by FOLFOX6 or the reverse sequence in advanced colorectal cancer: a randomised GERCOR study. J Clin Oncol 2004; 22: 229–237. 9. Grothey A, Sargent D. Overall survival of patients with advanced colorectal cancer correlates with availability of fluorouracil, irinotecan, and oxaliplatin regardless of whether doublet or single-agent therapy is used first line. J Clin Oncol 2005; 23: 9441–9442.
926 | Ko¨hne et al.
10. Mocellin S, Lise M, Nitti D. Targeted therapy for colorectal cancer: mapping the way. Trends Mol Med 2005; 11: 327–335. 11. Mueller BU, Skelton J, Callender DP et al. A prospective randomized trial comparing the infectious and noninfectious complications of an externalized catheter versus a subcutaneously implanted device in cancer patients. J Clin Oncol 1992; 10: 1943–1948. 12. Meropol NJ. Oral fluoropyrimidines in the treatment of colorectal cancer. Eur J Cancer 1998; 34: 1509–1513. 13. Cassidy J, Twelves C, Van Cutsem E et al. First-line oral capecitabine therapy in metastatic colorectal cancer: a favorable safety profile compared with intravenous 5-fluorouracil/leucovorin. Ann Oncol 2002; 13: 566–575. 14. Van Cutsem E, Hoff PM, Harper P et al. Oral capecitabine vs intravenous 5fluorouracil and leucovorin: integrated efficacy data and novel analyses from two large, randomised, phase III trials. Br J Cancer 2004; 90: 1190–1197. 15. Brown JR, DuBois RN. COX-2: a molecular target for colorectal cancer prevention. J Clin Oncol 2005; 23: 2840–2855. 16. Steinbach G, Lynch PM, Phillips RK et al. The effect of celecoxib, a cyclooxygenase-2 inhibitor, in familial adenomatous polyposis. N Engl J Med 2000; 342: 1946–1952. 17. Therasse P, Arbuck SG, Eisenhauer EA et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 2000; 92: 205–216. 18. Pocock SJ, Simon R. Sequential treatment assignment with balancing for prognostic factors in the controlled clinical trial. Biometrics 1975; 31: 103–115. 19. Ko¨hne CH, Cunningham D, Di CF et al. Clinical determinants of survival in patients with 5-fluorouracil-based treatment for metastatic colorectal cancer: results of a multivariate analysis of 3825 patients. Ann Oncol 2002; 13: 308–317. 20. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457–481. 21. Rothenberg ML, Meropol NJ, Poplin EA et al. Mortality associated with irinotecan plus bolus fluorouracil/leucovorin: summary findings of an independent panel. J Clin Oncol 2001; 19: 3801–3807. 22. Fuchs C, Marshall J, Mitchell E et al. A randomized trial of first-line irinotecan/ fluoropymidine combinations with or without celecoxib in metastatic colorectal cancer (BICC-C). ASCO Annual Meeting (Post Meeting Proceedings). 2006; 18S: (Abstr 3506). 23. Rea DW, Nortier JW, Ten Bokkel Huinink WW et al. A phase I/II and pharmacokinetic study of irinotecan in combination with capecitabine as first-line therapy for advanced colorectal cancer. Ann Oncol 2005; 16: 1123–1132. 24. Koopman M, Antonini NF, Douma J et al. Sequential versus combination chemotherapy with capecitabine, irinotecan, and oxaliplatin in advanced colorectal cancer (CAIRO): a phase III randomised controlled trial. Lancet 2007; 370: 135–142. 25. Hochster H, Welles L, Hart L et al. Safety and efficacy of bevacizumab (Bev) when added to oxaliplatin/fluoropyrimidine (O/F) regimens as first-line treatment of metastatic colorectal cancer (mCRC): TREE 1 & 2. ASCO Annual Meeting (Post Meeting Proceedings). 2005; 23: (Abstr 3515). 26. Twelves C, Wong A, Nowacki MP et al. Capecitabine as adjuvant treatment for stage III colon cancer. N Engl J Med 2005; 352: 2696–2704. 27. Haller DG, Cassidy J, Clarke S et al. Tolerability of fluoropyrimidines appears to differ by region. ASCO Annual Meeting (Post Meeting Proceedings). 2006; 24: (Abstr 3514). 28. Jacoby RF, Seibert K, Cole CE et al. The cyclooxygenase-2 inhibitor celecoxib is a potent preventive and therapeutic agent in the min mouse model of adenomatous polyposis. Cancer Res 2000; 60: 5040–5044. 29. Solomon SD, McMurray JJ, Pfeffer MA et al. Cardiovascular risk associated with celecoxib in a clinical trial for colorectal adenoma prevention. N Engl J Med 2005; 352: 1071–1080. 30. Maiello E, Giuliani F, Gebbia V et al. FOLFIRI with or without celecoxib in advanced colorectal cancer: a randomized phase II study of the Gruppo Oncologico dell’Italia Meridionale (GOIM). Ann Oncol 2006; 17 (Suppl 7): vii55–vii59.
Volume 19 | No. 5 | May 2008
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 26, 2015
funding
Annals of Oncology
Annals of Oncology 19: 920–926, 2008 doi:10.1093/annonc/mdm544 Published online 6 December 2007
Irinotecan combined with infusional 5-fluorouracil/ folinic acid or capecitabine plus celecoxib or placebo in the first-line treatment of patients with metastatic colorectal cancer. EORTC study 40015 C.-H. Ko¨hne1*, J. De Greve2, J. T. Hartmann3, I. Lang4, P. Vergauwe5, K. Becker6, D. Braumann7, E. Joosens8, L. Mu¨ller9, J. Janssens10, C. Bokemeyer11, P. Reimer12, H. Link13, E. Spa¨th-Schwalbe14, H.-J. Wilke15, H. Bleiberg16, J. Van Den Brande17, M. Debois18, U. Bethe18 & E. Van Cutsem19
Received 29 July 2007; revised 28 October 2007; accepted 29 October 2007
Background: The study aimed to demonstrate the noninferiority of capecitabine to 5-fluorouracil (5-FU)/folinic acid (FA), in relation to progression-free survival (PFS) after first-line treatment of metastatic colorectal cancer and the benefit of adding celecoxib (C) to irinotecan/fluoropyrimidine regimens compared with placebo (P). Patients and methods: Patients were randomly assigned to receive FOLFIRI: irinotecan (180 mg/m2 i.v. on days 1, 15 and 22); FA (200 mg/m2 i.v. on days 1, 2, 15, 16, 29 and 30); 5-FU (400 mg/m2 i.v. bolus, then 22-h, 600 mg/m2 infusion) or CAPIRI: irinotecan (250 mg/m2 i.v. infusion on days 1 and 22); capecitabine p.o. (1000 mg/m2 b.i.d. on days 1–15 and 22–36). Patients were additionally randomly assigned to receive either placebo or celecoxib (800 mg: 2 · 200 mg b.i.d.). Results: The trial was closed following eight deaths unrelated to disease progression in the 85 enrolled (629 planned) patients. Response rates were 22% for CAPIRI + C, 48% for CAPIRI + P, 32% for FOLFIRI + C and 46% for FOLFIRI + P. Median PFS and overall survival (OS) times were shorter for CAPIRI versus FOLFIRI (PFS 5.9 versus 9.6 months and OS 14.8 versus 19.9 months) and celecoxib versus placebo (PFS 6.9 versus 7.8 months and OS 18.3 versus 19.9 months). Conclusion: Due to the small sample size following early termination, no definitive conclusions can be drawn in relation to the noninferiority of CAPIRI compared with FOLFIRI. Key words: capecitabine, celecoxib, colorectal, first line, 5-FU, irinotecan
introduction Colorectal cancer (CRC) remains one of the most common causes of cancer-related mortality [1]. In Europe alone, where it is the second most commonly diagnosed cancer, it is estimated *Correspondence to: Prof. C.-H. Ko¨hne, Department of Oncology and Hematology, Klinikum Oldenburg GmbH, Dr Eden Strasse 10, 26133 Oldenburg, Germany. Tel: +49-441-403-2611; Fax: +49-441-403-2654; E-mail: [email protected]
that CRC accounted for >207 000 deaths in 2006 [2]. Surgery is an effective curative treatment strategy for disease that is restricted to and resectable at the primary site. In addition, metastatic colorectal cancer (mCRC) may be curable, especially if hepatic and/or lung lesions are resectable, either at diagnosis or following effective first-line treatment [3]. The further improvement of first-line regimens is therefore of great importance, not only in relation to increasing the number of patients amenable to secondary surgery for
ª The Author 2007. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected]
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 26, 2015
Department of Oncology and Hematology, Klinikum Oldenburg, Oldenburg, Germany; 2Department of Oncology, Universitair Ziekenhuis Brussel, Vrije Universiteit, Brussels, Belgium; 3University Medical Center II, Eberhard-Karls-Universita¨t, Tu¨bingen, Germany; 4Department of Medical Oncology, National Institute of Oncology, Budapest, Hungary; 5Department of Internal Medicine, AZ Groeninghe, Campus Sint-Niklaas, Kortrijk, Belgium; 6Onkologischer Schwerpunkt Lerchenfeld, Hamburg; 7 II. Medizinische Abteilung, Allgemeines Krankenhaus Altona, Hamburg, Germany; 8Department of Hemato-Oncology, Ziekenhuis Netwerk Middelheim, Campus Middelheim, Antwerpen, Belgium; 9Onkologische Schwerpunktpraxis Leer, Leer, Germany; 10Department of Gastroenterology, Sint-Elisabeth Ziekenhuis, Turnhout, Belgium; 11Department of Oncology and Hematology, Universita¨tsklinikum Hamburg-Eppendorf, Hamburg; 12Medizinische Klinik III, Wu¨rzburg; 13Medizinische Klinik I, Westpfalz-Kinikum, Kaiserslautern; 14Medical Department, Hematology, Oncology and Gastroenterology, Vivantes Klinkum Spandau, Berlin; 15Department of Oncology, Kliniken Essen-Mitte, Essen, Germany; 16Department of Gastroenterology, Chemotherapy Unit, Institut Jules Bordet, Brussels; 17Department of Medical Oncology, Universitair Ziekenhuis Antwerpen, Edegem; 18European Organisation for Research and Treatment of Cancer (EORTC), EORTC Data Center, Brussels; 19 Digestive Oncology Unit, University Hospital Gasthuisberg, Leuven, Belgium
original article
1
original article
Annals of Oncology
Volume 19 | No. 5 | May 2008
were designed to show that the NSAID would similarly reduce the speed of recurrence of CRC in patients responding to first-line treatment.
patients and methods main eligibility criteria Patients, ‡18 years of age, with previously untreated (i.e. not treated with chemotherapy for metastatic disease), metastatic, histologically verified, adenocarcinoma of the colon or rectum were eligible. Other criteria for study entry were a World Health Organization performance status (PS) of two or less and measurable disease according to the Response Evaluation Criteria in Solid Tumors (RECIST) [17], located outside the field of any radiotherapy, which, if given, had been completed at least 4 weeks before randomization. Prior adjuvant therapy was allowed, provided that it had been completed 6 months before randomization, but no current or planned NSAID administration other than low-dose prophylactic aspirin was permitted and no other investigational drugs were to have been administered within 4 weeks before randomization. Patients were required to have adequate hematological, renal and hepatic function, no central nervous system metastases, second malignancies, severe cardiac disease, active Crohn’s disease or any other uncontrolled severe medical condition. Before patients were entered into the trial, they gave written informed consent in accordance with either local/national regulations or the principles defined in the Declaration of Helsinki, whichever provided the greater protection for the patient.
trial design This was a prospective, multicenter, two-by-two factorial, phase III, randomized study. As stated, the two main objectives were to demonstrate that capecitabine could replace 5-FU/FA as the fluoropyrimidine component of an irinotecan-including combination without compromising PFS and that COX-2 inhibition would improve PFS compared with placebo. The primary end point was therefore PFS. Secondary end points included safety (assessed in all patients receiving ‡1 dose of any study drug according to the National Cancer Institute Common Toxicity Criteria, version 2), response rate, time to treatment failure and overall survival (OS) assessed in the intention-to-treat (ITT) population. Treatment cycles lasted 6 weeks. Patients were randomly assigned to receive, in the first instance to, FOLFIRI or CAPIRI. The FOLFIRI arm received: irinotecan (180 mg/m2), as a 30- to 90-min i.v. infusion on days 1, 15 and 22; FA (200 mg/m2), as a 2-h infusion on days 1, 2, 15, 16, 29 and 30 (1 h after irinotecan on days 1, 15 and 29); 5-FU, as a 400-mg/m2 bolus given after FA, followed by 22-h continuous infusion (600 mg/m2) given after the bolus (days 1, 2, 15, 16, 29 and 30). The CAPIRI arm received: irinotecan (250 mg/m2), as a 30- to 90-min i.v. infusion on days 1 and 22 and capecitabine p.o. (1000 mg/m2, twice daily) on days 1–15 and 22–36. Within these arms, to complete the 2 · 2 design, patients were randomly assigned to receive either placebo or celecoxib (800 mg: as 2 · 200 mg, twice daily, before irinotecan, when administered). Treatment continued up to a planned total of six cycles or until disease progression, unacceptable toxicity or withdrawal of patient consent. Responding patients or those with stable disease were allowed to continue treatment beyond six cycles at the discretion of the investigator. Evaluation of disease status was carried out every 6 weeks during treatment and every 8 weeks subsequently until the documentation of disease progression. Response rate was assessed by the investigators through the radiological measurement of target lesions identified at baseline (maximum of 10 to be representative of all involved organs), according to RECIST criteria. The duration of response was defined as the time interval from when the measurement criteria for a complete or partial
doi:10.1093/annonc/mdm544 | 921
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 26, 2015
metastases but also in the context of disease management with regard to improving the quality of life and lengthening the survival of nonresectable patients. For many years, 5-fluorouracil (5-FU), generally combined with the biological response modifier folinic acid (FA) [4], has been the mainstay of treatment regimens for mCRC, with meta-analyses suggesting that infusional delivery is more effective [5] and associated with less grade 3/4 hematological toxicity than bolus administration [6]. The introduction of two further cytotoxic chemotherapy agents, irinotecan and oxaliplatin, have more recently allowed the development of multiline combination therapy for mCRC patients. The European Organisation for Research and Treatment of Cancer (EORTC) Gastrointestinal Group and others were able to demonstrate that median survival times of >20 months are now achievable [7, 8], with the use of three active drugs during the course of a patient’s treatment conferring maximum survival benefit [9]. It is anticipated that the use of targeted biological agents in this setting will further improve the performance of existing cytotoxic therapies [10]. Currently, infusional 5-FU/FA combined with irinotecan (FOLFIRI) or oxaliplatin (FOLFOX) are regarded as standard first-line chemotherapy regimens of equivalent efficacy. The devices used during the infusion process may be associated with infection or thrombosis [11]. One possible solution to this problem might be the oral delivery of fluorouracil. Capecitabine is an inactive oral prodrug which is preferentially enzymatically converted in tumors to fluorouracil [12]. Although there are no reported studies comparing the safety and efficacy of capecitabine with infusional 5-FU, the oral prodrug has been shown to be less toxic and at least as effective as the Mayo Clinic bolus 5-FU regimen [13, 14]. As combination therapy is now the first-line treatment of choice for most mCRC patients, opportunities to formally compare the efficacy and safety of infusional 5-FU/FA monotherapy versus capecitabine monotherapy are limited. Such information, however, may be derived indirectly by comparing regimens where the 5-FU/FA component of the comparator arm is replaced by capecitabine in the experimental arm. With this consideration in mind, the first main objective of this 2 · 2 study was to demonstrate that infusional 5-FU/FA component of FOLFIRI can be replaced by oral capecitabine (CAPIRI), without compromising efficacy, with the intention being to show noninferiority of CAPIRI to FOLFIRI in terms of progression-free survival (PFS). The second main objective of the study was to demonstrate that cyclooxygenase (COX)-2 inhibition by celecoxib improves PFS compared with placebo. COX-2 [15], the product of the PTGS2 gene, is a mitogen-inducible enzyme that, with the constitutively expressed COX-1, regulates a key step in the biosynthesis of prostaglandins. It is thought that COX-2 is involved in the process of inflammation [15]. Celecoxib is a nonsteroidal anti-inflammatory drug (NSAID) which shows high specificity for COX-2 over COX-1 inhibition. A double-blind placebo-controlled study demonstrated that the twice-daily administration of celecoxib was associated with a significant regression of colorectal adenomas in patients with the CRC-predisposing syndrome, familial adenomatous polyposis [16]. The celecoxib versus placebo arms of this trial
original article
Annals of Oncology
response were met until the first date that recurrent or progressive disease (PD) was documented. PFS was calculated as the time from randomization until the first report of PD or death. Patients with no evidence of PD at the time of their last visit were censored at that time.
statistical methods and considerations
results patient demographics, treatment compliance and deaths The trial was activated on 21 May 2003. On 23 April 2004, after the enrollment of only 85 patients, recruitment was suspended as a consequence of seven deaths not due to disease progression. One more patient subsequently died following the suspension of recruitment. Six deaths occurred in patients receiving CAPIRI and two in those receiving FOLFIRI.
adverse events Three patients (4%) did not receive study drugs and are therefore not included in the safety analysis. At least one grade 3/4 adverse event (AE) was reported for 62% of patients. The combined frequency of grade 3/4 AEs was higher for patients in the capecitabine arms compared with those in the FOLFIRI arms (32/43, 74% versus 19/39, 49%). The most common grade 3/4 AE was diarrhea, followed by WBC toxicity. AEs occurring in two or more patients in any one arm are detailed in Table 3. Hand–foot syndrome (HFS), which is commonly associated with capecitabine administration, was
Table 1. Early deaths and relationship to study treatment Treatment
Total number of cycles
Time since last treatment (days)
Relatednessa
Agreed classificationa
CAPIRI + C CAPIRI + C CAPIRI + P CAPIRI + P CAPIRI + P CAPIRI + P FOLFIRI + C FOLFIRI + C
3 1 1 1 2 1 1 1
65 4 6 75 9 5 12 2
Nonrelated Exacerbated Exacerbated Related Related Related Related Exacerbated
Sudden cardiac arrest Thromboembolic event/pulmonary embolism Pulmonary embolism Diarrhea/neutropenia/septic shock Diarrhea/myocardial infarction Diarrhea/suspected pulmonary embolism Neutropenic sepsis Pneumonia
a
Classified by panel of experts. CAPIRI, capecitabine/irinotecan; C, celecoxib; P, placebo; FOLFIRI, 5-fluorouracil/folinic acid/irinotecan.
922 | Ko¨hne et al.
Volume 19 | No. 5 | May 2008
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 26, 2015
Patients were randomly assigned to receive centrally using a minimization technique [18]. Stratification factors were the institution, previous adjuvant therapy (yes/no) and risk groups, as defined by Ko¨hne et al. [19] [low risk: patients with a PS of 1 or 0 and only one tumor site; intermediate risk: patients with a good PS (<1), but more than one tumor site plus an alkaline phosphatase (ALP) level of <300 U/l, or those with a poor PS, a low white blood cell (WBC) count and only one tumor site; high risk: patients with good PS but more than one tumor site and a high ALP level, or a poor PS plus a high WBC count or a poor PS, a low WBC count and more than two tumor sites]. All efficacy analyses were carried out on the ITT population. Unacceptable inferiority of capecitabine over 5-FU/FA in relation to PFS, the primary end point, was defined by a hazard ratio (HR) ‡1.25. Given a one-sided alpha level of 2.5%, it was estimated that 632 events were needed to exclude a difference of this magnitude with 80% probability. This number of events would also allow the detection of a 2-month difference between the celecoxib and placebo arms with a power of 89% and a two-sided 5% significance level test. It was consequently estimated that 692 patients should be randomized (1 : 1) to generate this number of events. These sample size determinations were on the basis of the assumptions of an absence of interaction between the 5-FU versus capecitabine and the celecoxib versus placebo effects. Time to event variables were analyzed using the Kaplan–Meier technique [20] and differences summarized with HRs and confidence intervals (CIs). Quantitative variables were summarized using counts and percentages. The study was approved by the EORTC protocol review committee and by local authorities and ethic committees of the participating centers.
Celecoxib treatment allocation was unblinded at this stage. The individual hospital files were inspected and discussed with the physician to determine whether the deaths were related to or exacerbated by the study treatment (Table 1) and whether any risk factors for thromboembolic events or gastrointestinal toxicity were present at randomization. Five deaths in the CAPIRI arm and both of those in the FOLFIRI arm were deemed to be treatment related. Underlying risk factors could not be identified as a likely explanation for these fatal toxic effects. On the basis of the outcome of this review, the trial was officially closed on 12 January 2005. This report is consequently on the basis of the data available from each of the 85 eligible patients as of 16 December 2005. Median follow-up time was 14.6 months (95% CI 13.1–16.8). Patient characteristics, described in Table 2, were similar between treatment arms. Table 2 also details treatment exposure. Dose reductions, mainly as a consequence of gastrointestinal toxicity, were more common in the CAPIRI compared with the FOLFIRI arms, with 53% versus 33% of patients, respectively, experiencing at least one cycle with a reduction. Thirty-four patients (41.5%) experienced treatment delays, which were more common in the FOLFIRI compared with the CAPIRI arms, with 54% of patients versus 30%, respectively, experiencing at least one cycle with delay. The relative dose intensity for capecitabine and 5-FU did not differ markedly in their placebo arms (82.4% versus 84.8%, respectively) but was lower for capecitabine if celecoxib was also administered (66.4% versus 92.1% for 5-FU). Interestingly, very little difference in the irinotecan dose was observed across all study arms (Table 2; range 83.1%–88.4%).
original article
Annals of Oncology
Table 2. Patient characteristics and treatment exposure CAPIRI + P
FOLFIRI + C
FOLFIRI + P
Total
n = 23 12/11, (52/48)
n = 21 12/9, (57/43)
n = 19 15/4, (79/21)
n = 22 14/8, (64/36)
n = 85 53/32, (62/38)
12 (52) 8 (35) 3 (13)
8 (38) 8 (38) 5 (24)
10 (53) 8 (42) 1 (5)
14 (64) 6 (27) 2 (9)
44 (52) 30 (35) 11 (13)
63.0
65.0
(42–75) 15 (65) 8 (35)
(43–78) 11 (52) 10 (48)
12 (52) 6 (26) 5 (22)
10 (48) 6 (29) 5 (24)
14 (61) 8 (35) 1 (4) 12 (52) 11 (48) 3 (1–6)
Patients who started n = 23 chemotherapy Dose reductions in at least 13 (57) one cycle, n (%) Delays in at least one cycle, n (%) 8 (35) Median relative dose intensities, % (range) Capecitabine 66.4 (6.6–110.3) 5-FU Irinotecan 83.1 (32.6–103.3) Celecoxib/placebo 99.2 (66.7–100)
66.0 (54–77) 9 (47) 10 (53)
60.5
64.0
(45–75) 14 (64) 8 (36)
(42–78) 49 (58) 36 (42)
9 (47) 7 (37) 3 (16)
11 (50) 8 (36) 3 (14)
42 (49) 27 (32) 16 (19)
12 (57) 8 (38) 1 (5)
10 (53) 9 (47) 0 (0)
14 (64) 8 (36) 0 (0)
50 (59) 33 (39) 2 (2)
11 (52) 10 (48) 3 (0–6)
9 (47) 10 (53) 4 (0–6)
11 (50) 11 (50) 6 (0–6)
43 (51) 42 (49) 4 (0–6)
n = 20
n = 18
n = 21
n = 82
10 (50)
3 (17)
10 (48)
36 (44)
5 (25)
8 (44)
13 (62)
34 (41)
82.4 (47.5–119.6) 83.6 (47.5–101.7) 98.3 (59.5–101.2)
92.1 (21.2–107.4) 88.4 (20.9–98.6) 96.2 (37.6–100.0)
83.8 (51.7–103.1) 84.0 (51.6–103.3) 94.4 (81.6–100.0)
73.7 86.8 84.1 96.9
(6.6–119.6) (21.2–107.4) (20.9–103.3) (37.6–101.2)
CAPIRI, capecitabine/irinotecan; C, celecoxib; P, placebo; FOLFIRI, 5-fluorouracil/folinic acid/irinotecan; WHO, World Health Organization; PS, performance status; 5-FU, 5-fluorouracil.
restricted to 14 patients who received the drug (seven in each arm), reaching grade 3 in only two (one in each arm). Five patients in the FOLFIRI arms experienced grade 1 HFS reactions.
efficacy The best overall response to treatment is detailed in Table 3. The number of patients in each arm was too small to allow reliable conclusions to be drawn. Response rates, however, were higher in both placebo arms compared with the corresponding celecoxib arms: 22% for CAPIRI plus celecoxib versus 48% for CAPIRI plus placebo and 32% for FOLFIRI plus celecoxib versus 46% for FOLFIRI plus placebo. The effect of capecitabine versus 5-FU/FA and celecoxib versus placebo on PFS and OS were analyzed with stratification for the nonexamined variable. Survival data are detailed in Table 4 and Figure 1. Median PFS times were shorter for CAPIRI versus
Volume 19 | No. 5 | May 2008
FOLFIRI (5.85 versus 9.6 months) and celecoxib versus placebo (6.9 versus 7.75 months). Median OS times were shorter for CAPIRI versus FOLFIRI (14.75 versus 19.9 months) and similar for celecoxib versus placebo (18.3 versus 19.9 months).
discussion This study aimed to demonstrate in patients with mCRC that capecitabine was not inferior to 5-FU/FA in terms of PFS and that the addition of celecoxib to CAPIRI and/or FOLFIRI improved PFS compared with placebo. Seven deaths unrelated to disease progression, however, occurred early on in the study (five in the CAPIRI arms), with a further death (CAPIRI arm) occurring after the suspension of recruitment. The deaths were primarily linked to gastrointestinal or thromboembolic events. Sudden deaths linked to such causes have previously been noted for regimens combining irinotecan
doi:10.1093/annonc/mdm544 | 923
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 26, 2015
Variable Sex (M/F), n (%) Primary site, n (%) Colon Rectum Colorectal Age, years Median Range £65 years, n (%) >65 years Risk group, n (%) [19] Good Intermediate Poor WHO PS, n (%) 0 1 2 Metastatic sites, n (%) 1 >1 Median number of cycles of chemotherapy, n (range)
CAPIRI + C
original article
Annals of Oncology
Table 3. Best overall response to treatment and grade 3/4 adverse events reported for two or more of the patients, in any one arm, who started treatment CAPIRI + P
FOLFIRI + C
FOLFIRI + P
Total
Best overall response n (%) CR PR SD Progressive disease Early death Not assessable Response rate (CR + PR) Disease control rate (CR + PR + SD)
n = 23 0 5 (22) 12 (52) 2 (9) 2 (9) 2 (9) 5 (22) 17 (74)
n = 21 1 (5) 9 (43) 5 (24) 2 (10) 3 (14) 1 (5) 10 (48) 15 (71)
n = 19 1 (5) 5 (26) 7 (37) 2 (11) 3 (16) 1 (5) 6 (32) 13 (68)
n = 22 0 10 (45) 9 (41) 3 (14) 0 0 10 (45) 19 (86)
n = 85 2 (2) 29 (34) 33 (39) 9 (11) 8 (9) 4 (5) 31 (36) 64 (75)
Adverse event, n (%) Diarrhea Vomiting Nausea Gastrointestinal, other Cardiovascular Febrile neutropenia Hepatic toxicity White blood cells Hemoglobin Renal toxicity All grade 3/4 events
n = 23 9 (39) 2 (9) 1 (4) 2 (9)b 3 (13)b 0 2 (9) 3 (13)c 2 (9) 2 (9)b 19 (83)
n = 20 7 (35) 1 (5)b 1 (5) 0 1 (5) 2 (10)b 1 (5) 3 (15)c 0 0 13 (65)
n = 18 3 (17) 1 (6) 1 (6) 0 1 (6) 2 (11)b 1 (6) 2 (11)b 0 0 8 (44)
n = 21 2 (10)b 1 (5) 2 (10) 1 (5) 0 0 1 (5) 4 (19)b 0 0 11 (52)
n = 82 21 (26) 5 (6) 5 (6) 3 (4) 5 (6) 4 (5) 5 (6) 12 (15) 2 (2) 2 (2) 51 (62)
a
Confirmation was documented in only 74% of responders, although intended in each case. One patient at grade 4. c Two patients at grade 4. CAPIRI, capecitabine/irinotecan; C, celecoxib; P, placebo; FOLFIRI, 5-fluorouracil/folinic acid/irinotecan; CR, complete response; PR, partial response; SD, stable disease. b
and bolus 5-FU/FA [21]. Following the consideration of all available information, a decision was taken to close the trial after the enrollment of only 85 of the planned 692 patients had been achieved. Patients who had tolerated the treatment were allowed to continue with their assigned regimen. Given the small number of patients involved, no formal definitive conclusions can be drawn from the study data. Although caution is required in relation to the small numbers of analyzable patients, it should be noted that due primarily to a higher level of treatment-associated toxicity, patients receiving CAPIRI tended to receive fewer cycles of planned treatment than those receiving FOLFIRI and more patients required a dose reduction (53% versus 33, respectively). The efficacy data from this study are nevertheless consistent with those reported for the randomized, 3 · 2 factorial BICC-C trial, which assessed whether celecoxib added to FOLFIRI, CAPIRI or a modified irinotecan, bolus 5-fluorouracil and folinic acid (m-IFL) regimen improved efficacy and/or reduced toxicity. Median time to progression and OS times in this trial were longer in the patients who received FOLFIRI compared with those who received CAPIRI or m-IFL [22]. The most common grade 3/4 AE observed in the current study was diarrhea, which occurred significantly more frequently in the patients receiving CAPIRI than FOLFIRI (37% versus 13%). The dose levels of capecitabine and irinotecan initially selected were the same as those recommended, and found to be well tolerated by 76 patients
924 | Ko¨hne et al.
Table 4. Progression-free and overall survivals for CAPIRI versus FOLFIRI and celecoxib versus placebo Parameter PFS Median, months (95% CI) 1 year, % (95% CI) Hazard ratio, (95% CI) Overall survival Median, months (95% CI) 1 year, % (95% CI) Hazard ratio, (95% CI)
PFS Median, months (95% CI) 1 year, % (95% CI) Hazard ratio, (95% CI) Overall survival Median, months (95% CI) 1 year, % (95% CI) Hazard ratio, (95% CI)
CAPIRI, n = 44
FOLFIRI, n = 41
5.9 (4.4–8.9) 22.6 (11.4–36.2) 1.00
9.6 (6.9–10.9) 29.3 (16.4–43.4) 0.76 (0.48–1.21)
14.75 (10.7–18.3) 53.5 (36.0–68.2) 1.00
19.9 (18.9, NR) 84.9 (69.4–92.9) 0.31 (0.14–0.71)
Celecoxib, n = 42
Placebo, n = 43
6.9 (5.5–10.4) 20.1 (9.5–33.4) 1.00
7.8 (6.0–12.0) 31.4 (18.2–45.5) 0.90 (0.56–1.43)
18.3 (10.2, NR) 65.2 (48.1–77.8) 1.00
19.9 (16.7, NR) 74.6 (57.6–85.5) 0.80 (0.39–1.63)
CAPIRI, capecitabine/irinotecan; FOLFIRI, 5-fluorouracil/folinic acid/ irinotecan; PFS, progression-free survival; CI, confidence interval; NR, not reached.
Volume 19 | No. 5 | May 2008
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 26, 2015
CAPIRI + C a
Annals of Oncology
original article
in a recent phase I/II trial [23]. Similarly, in a large phase III study of combination chemotherapy with capecitabine, irinotecan and oxaliplatin in 820 advanced CRC patients, CAPIRI was again found to be generally well tolerated [24]. In the current study, considering the observed toxicity, CAPIRI at this dose and schedule, however, was not considered feasible. Toxicity was also generally elevated in the CAPIRI arm of the BICC-C trial compared with the FOLFIRI or m-IFL arm, with the regimen associated with the highest rates of nausea, vomiting, diarrhea, dehydration and HFS. It is perhaps noteworthy that the current and BICC-C studies, both of which report high levels of toxicity with CAPIRI, similarly included the use of celecoxib. These analyses raise the question of whether a lower capecitabine dose may have been more effective. In this context, it is interesting to note that in the TREE-2 study, capecitabine was reduced to 850 mg/m2 twice daily in combination with oxaliplatin (CAPOX) and bevacizumab after high levels of toxicity were reported for the corresponding CAPOX arm in the initial TREE-1 investigation, during which capecitabine had been administered at 1000 mg/m2 twice daily [25]. Similarly, in the phase III X-ACT study of adjuvant capecitabine versus 5-FU/FA, the 1250-mg/m2 twice-daily dose of capecitabine required modification (delay, reduction and interruption) in 57% of patients [26]. One factor that should be noted is that the tolerability of fluoropyrimidines may differ according to the geographical region from which patients are drawn [27]. Further studies to determine the most appropriate dose of capecitabine in CAPIRI and other combination regimens for particular geographic and/or ethnic patient groups may therefore be warranted.
Volume 19 | No. 5 | May 2008
Celecoxib has been shown to reduce the incidence of colorectal polyps and induce their regression in the Min mouse model of the human CRC-predisposing syndrome, adenomatous polyposis coli (APC) [28]. Similarly, in humans, the daily administration of celecoxib to APC patients has been shown to lead to a significant reduction in the number of colorectal polyps [16]. More recently, in a blinded review of a large randomized trial, the twice-daily use of this agent in polypectomy patients, however, was associated with a dose-related increase in the risk of serious cardiovascular events [29]. While such associations argue strongly against the use of celecoxib at this dose and schedule in a chemopreventive setting, the additional toxicity risks may be acceptable in the treatment of advanced disease, provided that they are balanced by a significant benefit in survival. The data from this study, however, indicate no such efficacy benefit. Indeed, in view of the fact that response rates were lower in both celecoxib arms compared with the corresponding placebo arms for both regimens, it is possible that celecoxib may actually reduce the response to chemotherapy, a hypothesis that warrants further preclinical exploration. Celecoxib did not appear to modulate the toxicity of the chemotherapy; thus it is very unlikely that the toxicity observed with CAPIRI was due to celecoxib. A recent randomized phase II trial involving 81 mCRC patients has also failed to show a benefit from adding celecoxib (400 mg b.i.d.) to FOLFIRI [30]. The response rates in the FOLFIRI versus FOLFIRI plus celecoxib arm were 41% versus 35% with median time to progression and OS times of 8 versus 9 months and 16 versus 19 months, respectively. Toxicity was generally mild in both arms, with
doi:10.1093/annonc/mdm544 | 925
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 26, 2015
Figure 1. Kaplan–Meier plots of progression-free survival (panels A and B) and overall survival (panels C and D) plots for CAPIRI versus FOLFIRI (A and C) and celecoxib versus placebo (B and D). CAPIRI, capecitabine/irinotecan; FOLFIRI, 5-fluorouracil/folinic acid/irinotecan.
original article gastrointestinal disorders being most common. Although this study confirms that FOLFIRI is an effective and welltolerated regimen, data from larger randomized studies will be needed before formal definitive conclusions about the benefits of adding celecoxib to this combination can be drawn. It should be noted in this context that in relation to time to progression and OS, celecoxib failed to significantly improve the efficacy of first-line irinotecan/fluoropyrimidine combinations compared with placebo in the 430 patient BICC-C study [22]. In summary, CAPIRI at this dose and schedule was not considered feasible in this setting. Small sample size and confounding safety issues did not allow valid conclusions to be drawn concerning the relative efficacy of CAPIRI versus FOLFIRI. Consistent with other studies, no benefit was seen from adding celecoxib to irinotecan/fluoropyrimidine regimens.
Roche; Pharmacia (currently Pfizer); Aventis (currently sanofiaventis Group).
acknowledgements We would like to thank the patients who took part in this study and the medical and support staff at the collaborating centers. The contents of the present publication as well as the data collection and data analysis are solely the responsibility of the authors.
references 1. Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, 2002. CA Cancer J Clin 2005; 55: 74–108. 2. Ferlay J, Autier P, Boniol M et al. Estimates of the cancer incidence and mortality in Europe in 2006. Ann Oncol 2007; 18: 581–592. 3. Folprecht G, Grothey A, Alberts S et al. Neoadjuvant treatment of unresectable colorectal liver metastases: correlation between tumour response and resection rates. Ann Oncol 2005; 16: 1311–1319. 4. Thirion P, Michiels S, Pignon JP et al. Modulation of fluorouracil by leucovorin in patients with advanced colorectal cancer: an updated meta-analysis. J Clin Oncol 2004; 22: 3766–3775. 5. Meta-analysis Group In Cancer. Efficacy of intravenous continuous infusion of fluorouracil compared with bolus administration in advanced colorectal cancer. J Clin Oncol 1998; 16: 301–308. 6. Toxicity of fluorouracil in patients with advanced colorectal cancer: effect of administration schedule and prognostic factors. Meta-Analysis Group In Cancer. J Clin Oncol 1998; 16: 3537–3541. 7. Ko¨hne CH, van Cutsem E, Wils J et al. Phase III study of weekly high-dose infusional fluorouracil plus folinic acid with or without irinotecan in patients with metastatic colorectal cancer: European Organisation for Research and Treatment of Cancer Gastrointestinal Group Study 40986. J Clin Oncol 2005; 23: 4856–4865. 8. Tournigand C, Andre T, Achille E et al. FOLFIRI followed by FOLFOX6 or the reverse sequence in advanced colorectal cancer: a randomised GERCOR study. J Clin Oncol 2004; 22: 229–237. 9. Grothey A, Sargent D. Overall survival of patients with advanced colorectal cancer correlates with availability of fluorouracil, irinotecan, and oxaliplatin regardless of whether doublet or single-agent therapy is used first line. J Clin Oncol 2005; 23: 9441–9442.
926 | Ko¨hne et al.
10. Mocellin S, Lise M, Nitti D. Targeted therapy for colorectal cancer: mapping the way. Trends Mol Med 2005; 11: 327–335. 11. Mueller BU, Skelton J, Callender DP et al. A prospective randomized trial comparing the infectious and noninfectious complications of an externalized catheter versus a subcutaneously implanted device in cancer patients. J Clin Oncol 1992; 10: 1943–1948. 12. Meropol NJ. Oral fluoropyrimidines in the treatment of colorectal cancer. Eur J Cancer 1998; 34: 1509–1513. 13. Cassidy J, Twelves C, Van Cutsem E et al. First-line oral capecitabine therapy in metastatic colorectal cancer: a favorable safety profile compared with intravenous 5-fluorouracil/leucovorin. Ann Oncol 2002; 13: 566–575. 14. Van Cutsem E, Hoff PM, Harper P et al. Oral capecitabine vs intravenous 5fluorouracil and leucovorin: integrated efficacy data and novel analyses from two large, randomised, phase III trials. Br J Cancer 2004; 90: 1190–1197. 15. Brown JR, DuBois RN. COX-2: a molecular target for colorectal cancer prevention. J Clin Oncol 2005; 23: 2840–2855. 16. Steinbach G, Lynch PM, Phillips RK et al. The effect of celecoxib, a cyclooxygenase-2 inhibitor, in familial adenomatous polyposis. N Engl J Med 2000; 342: 1946–1952. 17. Therasse P, Arbuck SG, Eisenhauer EA et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 2000; 92: 205–216. 18. Pocock SJ, Simon R. Sequential treatment assignment with balancing for prognostic factors in the controlled clinical trial. Biometrics 1975; 31: 103–115. 19. Ko¨hne CH, Cunningham D, Di CF et al. Clinical determinants of survival in patients with 5-fluorouracil-based treatment for metastatic colorectal cancer: results of a multivariate analysis of 3825 patients. Ann Oncol 2002; 13: 308–317. 20. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457–481. 21. Rothenberg ML, Meropol NJ, Poplin EA et al. Mortality associated with irinotecan plus bolus fluorouracil/leucovorin: summary findings of an independent panel. J Clin Oncol 2001; 19: 3801–3807. 22. Fuchs C, Marshall J, Mitchell E et al. A randomized trial of first-line irinotecan/ fluoropymidine combinations with or without celecoxib in metastatic colorectal cancer (BICC-C). ASCO Annual Meeting (Post Meeting Proceedings). 2006; 18S: (Abstr 3506). 23. Rea DW, Nortier JW, Ten Bokkel Huinink WW et al. A phase I/II and pharmacokinetic study of irinotecan in combination with capecitabine as first-line therapy for advanced colorectal cancer. Ann Oncol 2005; 16: 1123–1132. 24. Koopman M, Antonini NF, Douma J et al. Sequential versus combination chemotherapy with capecitabine, irinotecan, and oxaliplatin in advanced colorectal cancer (CAIRO): a phase III randomised controlled trial. Lancet 2007; 370: 135–142. 25. Hochster H, Welles L, Hart L et al. Safety and efficacy of bevacizumab (Bev) when added to oxaliplatin/fluoropyrimidine (O/F) regimens as first-line treatment of metastatic colorectal cancer (mCRC): TREE 1 & 2. ASCO Annual Meeting (Post Meeting Proceedings). 2005; 23: (Abstr 3515). 26. Twelves C, Wong A, Nowacki MP et al. Capecitabine as adjuvant treatment for stage III colon cancer. N Engl J Med 2005; 352: 2696–2704. 27. Haller DG, Cassidy J, Clarke S et al. Tolerability of fluoropyrimidines appears to differ by region. ASCO Annual Meeting (Post Meeting Proceedings). 2006; 24: (Abstr 3514). 28. Jacoby RF, Seibert K, Cole CE et al. The cyclooxygenase-2 inhibitor celecoxib is a potent preventive and therapeutic agent in the min mouse model of adenomatous polyposis. Cancer Res 2000; 60: 5040–5044. 29. Solomon SD, McMurray JJ, Pfeffer MA et al. Cardiovascular risk associated with celecoxib in a clinical trial for colorectal adenoma prevention. N Engl J Med 2005; 352: 1071–1080. 30. Maiello E, Giuliani F, Gebbia V et al. FOLFIRI with or without celecoxib in advanced colorectal cancer: a randomized phase II study of the Gruppo Oncologico dell’Italia Meridionale (GOIM). Ann Oncol 2006; 17 (Suppl 7): vii55–vii59.
Volume 19 | No. 5 | May 2008
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 26, 2015
funding
Annals of Oncology