Content uploaded by Eric Van Cutsem
Author content
All content in this area was uploaded by Eric Van Cutsem
Content may be subject to copyright.
n engl j med
351;4
www.nejm.org july
22, 2004
The
new england journal
of
medicine
337
original article
Cetuximab Monotherapy and Cetuximab
plus Irinotecan in Irinotecan-Refractory
Metastatic Colorectal Cancer
David Cunningham, M.D., Yves Humblet, M.D., Ph.D., Salvatore Siena, M.D.,
David Khayat, M.D., Ph.D., Harry Bleiberg, M.D., Ph.D., Armando Santoro, M.D.,
Danny Bets, M.Sc., Matthias Mueser, M.D., Andreas Harstrick, M.D.,
Chris Verslype, M.D., Ph.D., Ian Chau, M.B., B.S.,
and Eric Van Cutsem, M.D., Ph.D.
From the Royal Marsden Hospital, London
and Surrey, United Kingdom (D.C., I.C.);
Saint-Luc University Hospital, Université
Catholique de Louvain, Brussels (Y.H.); Os-
pedale Niguarda Ca’ Granda, Milan (S.S.);
Hôpital Salpêtrière, Paris (D.K.); Institut
Jules Bordet, Brussels (H.B.); Istituto Clini-
co Humanitas, Rozzano-Milano, Italy (A.S.);
Merck, Amsterdam (D.B.); Merck, Darm-
stadt, Germany (M.M., A.H.); and Univer-
sity Hospital Gasthuisberg, Leuven, Bel-
gium (C.V., E.C.). Address reprint requests
to Dr. Cunningham at the Department of
Medicine, Royal Marsden Hospital, Downs
Rd., Sutton, Surrey SM2 5PT, United King-
dom, or at david.cunningham@icr.ac.uk.
N Engl J Med 2004;351:337-45.
Copyright © 2004 Massachusetts Medical Society.
background
The epidermal growth factor receptor (EGFR), which participates in signaling path-
ways that are deregulated in cancer cells, commonly appears on colorectal-cancer
cells. Cetuximab is a monoclonal antibody that specifically blocks the EGFR. We com-
pared the efficacy of cetuximab in combination with irinotecan with that of cetuximab
alone in metastatic colorectal cancer that was refractory to treatment with irinotecan.
methods
We randomly assigned 329 patients whose disease had progressed during or within
three months after treatment with an irinotecan-based regimen to receive either cetux-
imab and irinotecan (at the same dose and schedule as in a prestudy regimen [218 pa-
tients]) or cetuximab monotherapy (111 patients). In cases of disease progression, the
addition of irinotecan to cetuximab monotherapy was permitted. The patients were
evaluated radiologically for tumor response and were also evaluated for the time to tu-
mor progression, survival, and side effects of treatment.
results
The rate of response in the combination-therapy group was significantly higher than
that in the monotherapy group (22.9 percent [95 percent confidence interval, 17.5 to
29.1 percent] vs. 10.8 percent [95 percent confidence interval, 5.7 to 18.1 percent],
P=0.007). The median time to progression was significantly greater in the combina-
tion-therapy group (4.1 vs. 1.5 months, P<0.001 by the log-rank test). The median sur-
vival time was 8.6 months in the combination-therapy group and 6.9 months in the
monotherapy group (P=0.48). Toxic effects were more frequent in the combination-
therapy group, but their severity and incidence were similar to those that would be ex-
pected with irinotecan alone.
conclusions
Cetuximab has clinically significant activity when given alone or in combination with
irinotecan in patients with irinotecan-refractory colorectal cancer.
abstract
The New England Journal of Medicine
Downloaded from nejm.org on May 25, 2011. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.
n engl j med
351;4
www.nejm.org july
22
,
2004
The
new england journal
of
medicine
338
n the past decade, the median dura-
tion of survival among patients with ad-
vanced colorectal cancer has increased from
12 months to about 18 to 21 months, mainly owing
to the introduction of irinotecan and oxaliplatin.
Irinotecan (a semisynthetic camptothecin, which
inhibits topoisomerase I) and oxaliplatin (a third-
generation platinum compound) are widely used
in combination with fluorouracil and leucovorin as
first-line treatment for advanced colorectal can-
cer.
1-5
Irinotecan alone can also confer benefit in
patients with colorectal cancer that is refractory to
treatment with fluorouracil,
6,7
and oxaliplatin can
be effective in patients with disease that is refracto-
ry to irinotecan treatment.
8
Epidermal growth factor receptor (EGFR), a
member of the ErbB family of receptors, is relevant
in colorectal cancer because expression or up-reg-
ulation of the
EGFR
gene occurs in 60 to 80 percent
of cases.
9-11
Moreover, expression of the gene is as-
sociated with poor survival.
12,13
When inactive,
EGFR is a monomer, but when bound by epider-
mal growth factor or transforming growth factor
a
(TGF-
a
), it forms homodimers or heterodimers
with another member of the ErbB family of recep-
tors. Dimerization activates the intracellular tyro-
sine kinase region of EGFR, resulting in autophos-
phorylation and initiating a cascade of intracellular
events.
14
The EGFR signaling pathway regulates
cell differentiation, proliferation, migration, angio-
genesis, and apoptosis, all of which become de-
regulated in cancer cells.
15
Cetuximab (Erbitux,
Merck and Imclone Systems) is a chimeric IgG1
monoclonal antibody that binds to EGFR with high
specificity and with a higher affinity than either epi-
dermal growth factor or TGF-
a
, thus blocking lig-
and-induced phosphorylation of EGFR. In addition,
cetuximab enhances the effects of irinotecan
16
and
radiotherapy in experimental systems.
A phase 2 study of weekly cetuximab and irino-
tecan in 121 patients who had colorectal cancer
that was refractory to fluorouracil and irinotecan
found a 17 percent response rate.
17
In a study of
57 patients with EGFR-positive colorectal cancer
that was refractory to both fluorouracil and irinote-
can, 8.8 percent of the patients had a partial re-
sponse to cetuximab monotherapy, and 36.8 per-
cent had stable disease.
18
In the present trial, we compared the combina-
tion of cetuximab and irinotecan with cetuximab
monotherapy in patients with EGFR-expressing
irinotecan-refractory colorectal cancer.
patients
We considered patients eligible if they were more
than 18 years of age and had stage IV, histologi-
cally confirmed colorectal adenocarcinoma. Other
criteria for eligibility were a Karnofsky performance-
status score of 60 or more, adequate hematologic
function (hemoglobin, at least 9 g per deciliter
[5.6 mmol per liter]; neutrophil count, at least 1500
per cubic millimeter; and platelet count, at least
100,000 per cubic millimeter), renal function (se-
rum creatinine, less than 1.5 times the upper limit
of normal), and liver function (bilirubin, not more
than 1.5 times the upper limit of normal; aspartate
aminotransferase and alanine aminotransferase,
not more than 5 times the upper limit of normal).
To be eligible, patients must also have received
one of several qualifying, prestudy irinotecan regi-
mens for at least six weeks and must have had doc-
umented progression of disease during receipt of
this regimen or within three months thereafter.
These regimens were irinotecan at a dose of 125 mg
per square meter of body-surface area given weekly
for four consecutive weeks, followed by two weeks’
rest, as a single agent or in combination with fluo-
rouracil and leucovorin; irinotecan at a dose of
180 mg per square meter given every two weeks in
combination with fluorouracil and leucovorin; and
irinotecan at a dose of 350 mg per square meter
given every three weeks as a single agent. Disease
progression was documented by computed tomog-
raphy (CT) or magnetic resonance imaging (MRI);
new lung lesions could be documented by chest ra-
diography. At least one unidimensionally measur-
able lesion was required, as was immunohisto-
chemical evidence of EGFR expression, either in the
primary tumor or in at least one metastatic lesion.
All the patients signed a consent form. The study
was approved by the institutional ethics commit-
tees of all the participating centers.
study design and treatment
This open-label, randomized trial was conducted
in 56 centers in 11 European countries (the partic-
ipating investigators are listed in the Appendix). Be-
fore randomization, the EGFR status of the tumor
was determined at a central location in Germany
by immunohistochemical analysis of a paraffin-
embedded tumor specimen with the use of an EGFR
diagnostic kit (Dako Cytomation). Random assign-
ment of the eligible patients to either cetuximab in
i
methods
The New England Journal of Medicine
Downloaded from nejm.org on May 25, 2011. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.
n engl j med
351;4
www.nejm.org july
22, 2004
cetuximab in irinotecan-refractory colorectal cancer
339
combination with irinotecan or cetuximab alone in
a ratio of 2:1 was carried out centrally by an inde-
pendent randomization service. Randomization was
performed by a minimization technique, with strat-
ification according to Karnofsky performance sta-
tus, previous treatment with or without prior use
of oxaliplatin, and treatment center.
Cetuximab was given at an initial dose of 400 mg
per square meter, followed by weekly infusions of
250 mg per square meter. A histamine-receptor an-
tagonist was given as premedication before at least
the first infusion. Patients assigned to the combi-
nation-therapy group also received irinotecan at
the same dose as that given during their most re-
cent prestudy therapy. All the patients were to be
treated until disease progression or unacceptable
toxic effects occurred. In the case of disease progres-
sion, patients assigned to the monotherapy group
could continue to receive cetuximab, and irinote-
can could be added at the same dose as that given
during their most recent prestudy therapy.
Tumor response was evaluated every 6 weeks
for the first 24 weeks and thereafter every 3 months
with the use of consistent imaging techniques
(CT or MRI). Assessment was performed by the
investigators, who used the Response Evaluation
Criteria in Solid Tumors (RECIST)
19
and by an in-
dependent review committee consisting of three
radiologists and one oncologist who used modi-
fied World Health Organization (WHO) criteria.
20
In brief, on the basis of the WHO criteria, a com-
plete response was defined as the complete disap-
pearance of all measurable lesions, without the ap-
pearance of any new lesions. A partial response was
defined as a reduction in bidimensionally measur-
able lesions by at least 50 percent of the sum of the
products of their largest perpendicular diameters
and an absence of progression in other lesions,
without the appearance of any new lesions. Stable
disease was defined as a reduction in tumor volume
of less than 50 percent or an increase in the volume
of one or more measurable lesions of less than 25
percent, without the appearance of any new lesions.
Progressive disease was defined as an increase in
the size of at least one bidimensionally measurable
lesion by at least 25 percent and the appearance of
new lesions. The change from use of the RECIST
criteria to the WHO criteria followed a Food and
Drug Administration advisory that indicated that
use of the WHO criteria would facilitate an inde-
pendent review of three major studies to be submit-
ted for approval of cetuximab in the United States
and Europe.
The independent review committee, which was
blinded to the treatment assignment, assessed dis-
ease progression during the irinotecan regimen
given before enrollment in the study and the re-
sponse and time to progression during the study.
To do so, the committee used the WHO criteria and
compared CT scans or MRI images obtained be-
fore enrollment and during the study. Toxic effects
were assessed according to the National Cancer In-
stitute Common Toxicity Criteria, version 2. Modi-
fications of the dose of cetuximab were made only
in cases of toxic effects to the skin, and modifica-
tions in the dose of irinotecan were made in cases
of hematologic or nonhematologic toxic effects.
The investigators and representatives from
Merck designed the study. The data were collected
Figure 1. Design of the Study.
577 Patients screened
474 Patients had tumors positive
for EGFR
329 Patients underwent
randomization
218 Patients randomly assigned to
receive cetuximab and irinotecan
111 Patients randomly assigned to
receive cetuximab monotherapy
56 Patients received irinotecan
because of disease progression
145 Patients did not undergo
randomization
Poor physical condition
or death (n=31)
No disease progression
before trial recruitment
finished (n=57)
Eligibility criteria not met
(n=41)
Refusal to enter the study
(n=9)
No information (n=7)
The New England Journal of Medicine
Downloaded from nejm.org on May 25, 2011. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.
n engl j med
351;4
www.nejm.org july
22
,
2004
The
new england journal
of
medicine
340
and analyzed by medical and statistical representa-
tives from Merck, who worked in conjunction with
the investigators. All the investigators had access
to the primary data and participated in writing the
manuscript. All the participating institutions re-
ceived grant support from Merck for conducting
the study.
statistical analysis
The primary end point was the rate of confirmed
radiologic tumor response, as assessed by the in-
dependent review committee, in the intention-to-
treat population. Differences in response rates be-
tween the two groups were evaluated by means of a
two-sided Fisher’s exact test. A P value of less than
0.05 was considered to indicate statistical signifi-
cance. Secondary end points included the time to
progression, the duration of response, overall sur-
vival time, and the incidence of adverse effects. The
time to progression was calculated as the period
from the date of randomization to the first observa-
tion of disease progression or to death from any
cause within 60 days after randomization or the
most recent tumor assessment. The overall survival
time was calculated as the period from the date of
randomization until death from any cause or until
the date of the last follow-up, at which point data
were censored. Both the time to progression and
overall survival time were estimated by the Kaplan–
Meier method
21
and compared between the two
groups with use of the log-rank test.
22
Stratified
permutation tests were carried out to explore the
association between tumor response and rash and
between tumor response and EGFR expression.
The planned sample size for the study was based
on power calculations related to the estimation of
the confidence interval expected for the combina-
tion-therapy group. Anticipating a response rate of
19 percent in this group, we determined that 150
patients would be sufficient to exceed a clinically
relevant nominal response rate of 10 percent, with
85 percent power. A response rate of 5 percent in
the cetuximab-monotherapy group was expected.
Given this expectation, a 2:1 randomization ratio
was considered appropriate and ethically justified.
Therefore the study was originally set up to assign
225 patients randomly to one of the two treatment
groups. This sample size would also allow detec-
tion of a statistically significant difference in terms
of response rates between the two groups, with a
power of about 80 percent at a significance level of
0.05 by means of a two-sided Fisher’s exact test.
However, after the study had commenced and the
majority of patients had been enrolled, the Swed-
ish Medical Products Agency advised that patients
whose disease had progressed at the end of irino-
tecan treatment or progressed within one month
thereafter would be considered to have disease that
was strictly refractory to irinotecan. Therefore, the
sample size was increased to a total of 300 patients
to ensure that the study objectives could be met for
this subgroup of patients.
Predefined subgroup analyses included analy-
ses of the patients whose disease progressed dur-
ing or within one month after the end of the pre-
study irinotecan treatment (as described above) and
of patients who had received previous oxaliplatin
treatment.
Table 1. Baseline Characteristics of the Patients.
Characteristic
Cetuximab
plus
Irinotecan
(N=218)
Cetuximab
Monotherapy
(N=111)
Total
(N=329)
Age — yr
Median
Range
59
26–82
58
39–84
59
26–84
Sex — no. (%)
Male
Female
143 (65.6)
75 (34.4)
63 (56.8)
48 (43.2)
206 (62.6)
123 (37.4)
Race — no. (%)
White
Black
Asian
214 (98.2)
2 (0.9)
2 (0.9)
109 (98.2)
0
2 (1.8)
323 (98.2)
2 (0.6)
4 (1.2)
Karnofsky performance-status score
— no. (%)
<80
80–100
25 (11.5)
193 (88.5)
15 (13.5)
96 (86.5)
40 (12.2)
289 (87.8)
Previous adjuvant therapy — no. (%) 59 (27.1) 37 (33.3) 96 (29.2)
No. of previous cancer treatments
— no. (%)
One
Two
More than two
41 (18.8)
79 (36.2)
98 (45.0)
27 (24.3)
41 (36.9)
43 (38.7)
68 (20.7)
120 (36.5)
141 (42.9)
Prestudy irinotecan treatment
— no. (%)
125 mg/m
2
/wk for 4 wk followed
by 2-wk rest
180 mg/m
2
every 2 wk
350 mg/m
2
every 3 wk
Other regimens
33 (15.1)
124 (56.9)
57 (26.1)
4 (1.8)
20 (18.0)
54 (48.6)
31 (27.9)
6 (5.4)
53 (16.1)
178 (54.1)
88 (26.7)
10 (3.0)
Prior oxaliplatin treatment — no. (%) 135 (61.9) 71 (64.0) 206 (62.6)
Percentage of cells positive for
epidermal growth factor
receptor — no. (%)
0 to <10
10 to <20
20 to <40
≥40
94 (43.1)
23 (10.6)
39 (17.9)
62 (28.4)
41 (36.9)
22 (19.8)
16 (14.4)
32 (28.8)
135 (41.0)
45 (13.7)
55 (16.7)
94 (28.6)
The New England Journal of Medicine
Downloaded from nejm.org on May 25, 2011. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.
n engl j med
351;4
www.nejm.org july
22, 2004
cetuximab in irinotecan-refractory colorectal cancer
341
The cutoff point for survival data was January
2003; for safety data, it was November 2002. All
analyses were performed using SAS software (ver-
sion 8.2).
Between July 2001 and May 2002, 577 patients were
screened; 474 of them (82.1 percent) had EGFR-
positive tumors. Of these 474 patients, 329 were
randomly assigned either to combination treatment
with cetuximab and irinotecan (218 patients) or to
cetuximab monotherapy (111 patients). Figure 1
shows the design of the trial. Major protocol devi-
ations were reported in three patients. One patient
in each group had a tumor with no evidence of
EGFR expression, and one patient in the monother-
apy group had no evidence of metastatic colorectal
cancer at the time of enrollment. There were no
major imbalances between the two groups in terms
of baseline characteristics (Table 1). Nearly 80 per-
cent of the patients who underwent randomization
had received two or more previous regimens of
treatment for cancer. All the patients had received
irinotecan, and 206 (62.6 percent) had received ox-
aliplatin.
The overall response rate (the rate of complete
response plus the rate of partial response) in the in-
tention-to-treat population was 22.9 percent (95
percent confidence interval, 17.5 to 29.1 percent) in
the combination-therapy group and 10.8 percent
(95 percent confidence interval, 5.7 to 18.1 percent)
in the monotherapy group (P=0.007) (Table 2). The
median duration of response was 5.7 months in
the combination-therapy group and 4.2 months in
the monotherapy group. Disease control (complete
response plus partial response plus stable disease)
was achieved in 55.5 percent of patients receiving
combination treatment and in 32.4 percent of pa-
tients treated with cetuximab alone (P<0.001).
When the more stringent definition for irinotecan
resistance (progression during or within one month
after irinotecan therapy) was applied, the response
rates were 25.2 percent (95 percent confidence in-
terval, 18.1 to 33.4 percent) and 14.1 percent (95
percent confidence interval, 7.0 to 24.4 percent) in
the combination-therapy and monotherapy groups,
respectively (P=0.07), although the number of pa-
tients in this comparison (206) was small.
Of special interest is the finding that cetuximab-
based therapy was similarly effective in patients who
had previously received oxaliplatin in addition to
irinotecan before entering the study. In these mul-
tiply pretreated patients, the response rate was 22.2
percent in the combination-therapy group and 8.5
percent in the monotherapy group (P=0.01). Sta-
tistical regression modeling revealed that the dif-
ferences in response rates between the two treat-
ment groups were maintained after adjustment for
age, sex, performance status, and number of prior
treatment regimens (data not shown).
The degree of EGFR expression, either as the
percentage of EGFR-positive tumor cells or as the
maximal staining intensity per cell, did not correlate
significantly with the clinical response rate (P=0.87
and P=0.64, respectively) (Table 3). However, the
response rates in patients with skin reactions af-
ter cetuximab treatment were higher than those in
patients without skin reactions (25.8 percent vs.
6.3 percent in the combination-therapy group
[P=0.005] and 13.0 percent vs. 0 percent in the
monotherapy group).
results
* Values in brackets are 95 percent confidence intervals.
† The overall response rate is the sum of the rate of complete response and the
rate of partial response.
‡ The rate of disease control is the sum of the rates of complete response, par-
tial response, and stable disease.
Table 2. Rates of Radiologic Response.*
Subgroup and Variable
Cetuximab
plus
Irinotecan Cetuximab P Value
Intention-to-treat population
No. of patients
218 111
Response — no. (%)
Complete response 0 0
Partial response 50 (22.9) 12 (10.8)
Stable disease 71 (32.6) 24 (21.6)
Progressive disease 68 (31.2) 59 (53.2)
Could not be evaluated 29 (13.3) 16 (14.4)
Overall response† 50 (22.9
[17.5–29.1])
12 (10.8
[5.7–18.1])
0.007
Disease control‡ 121 (55.5
[48.6–62.2])
36 (32.4
[23.9–42.0])
<0.001
Subgroup with progression during or within
4 wk after prestudy irinotecan
No. of patients 135 71
Response — no. (%) 34 (25.2
[18.1–33.4])
10 (14.1
[7.0–24.4])
0.07
Subgroup with prior oxaliplatin therapy
No. of patients 135 71
Response — no. (%) 30 (22.2
[15.5–30.2])
6 (8.5
[3.2–17.5])
0.01
The New England Journal of Medicine
Downloaded from nejm.org on May 25, 2011. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.
n engl j med
351;4
www.nejm.org july
22
,
2004
The
new england journal
of
medicine
342
In the intention-to-treat analysis, the median
time to progression of disease was 4.1 months in
the combination-therapy group and 1.5 months in
the monotherapy group. The hazard ratio for dis-
ease progression in the combination-therapy group
was 0.54 (95 percent confidence interval, 0.42 to
0.71), indicating a 46 percent reduction in the risk
of progression with combination therapy as com-
pared with monotherapy (P<0.001 by the log-rank
test) (Fig. 2). In the combination-therapy group, the
percentages of patients remaining free of progres-
sion at three and six months were 54 percent (95
percent confidence interval, 47 to 61 percent) and
30 percent (95 percent confidence interval, 23 to 37
percent), respectively. In the monotherapy group,
the percentages of patients remaining free of pro-
gression at three and six months were 28 percent
(95 percent confidence interval, 19 to 37 percent)
and 8 percent (95 percent confidence interval, 1 to
14 percent), respectively. The times to progression
were similar among patients whose disease pro-
gressed during or within one month after prestudy
irinotecan therapy and among those who had pre-
viously received oxaliplatin.
By January 2003, 215 of the 329 patients in the
intention-to-treat analysis (65.3 percent) had died
(140 in the combination-therapy group and 75 in
the monotherapy group). The median overall sur-
vival time was 8.6 months in the combination-ther-
apy group and 6.9 months in the monotherapy
group. The hazard ratio for death with combina-
tion therapy, as compared with monotherapy, was
0.91 (95 percent confidence interval, 0.68 to 1.21),
and there was no statistically significant differ-
ence between the two groups (P=0.48 by the log-
rank test) (Fig. 3). In the combination-therapy
group, the overall survival rates at 6 and 12 months
were 66 percent (95 percent confidence interval,
60 to 72 percent) and 29 percent (95 percent con-
fidence interval, 22 to 37 percent), respectively. In
the monotherapy group, the overall survival rates at
6 and 12 months were 58 percent (95 percent con-
fidence interval, 49 to 67 percent) and 32 percent
(95 percent confidence interval, 23 to 41 percent),
respectively. The median survival times among
patients with skin reactions and those without skin
reactions were 9.1 and 3.0 months, respectively, in
the combination-therapy group and 8.1 and 2.5
months, respectively, in the monotherapy group.
Fifty-six patients in the monotherapy group re-
ceived additional irinotecan after disease progres-
sion was recognized. Two of these 56 patients (3.6
percent) had a partial response, and an additional
20 (35.7 percent) had stable disease after crossing
over. The median time to progression among the
patients who crossed over to receive additional
irinotecan was 1.4 months. These results may in
part explain the absence of a significant difference
in overall survival between the two groups.
All the patients received at least one infusion of
cetuximab. The median number of cetuximab infu-
sions was 18 in the combination-therapy group
and 7 in the monotherapy group. Table 4 summa-
rizes the most frequently observed grade 3 or 4 ad-
verse events related to treatment. Four patients who
were randomly assigned to the combination-thera-
py group did not receive irinotecan and therefore
were evaluated for safety as part of the monother-
apy group, and two patients who were randomly
assigned to the combination-therapy group did
* EGFR denotes epidermal growth factor receptor.
Table 3. Influence of Epidermal Growth Factor Receptor Expression and Rash
on Rates of Response.*
Variable
Cetuximab
plus
Irinotecan
(N=218)
Cetuximab
Monotherapy
(N=111)
P Value
for
Trend
no./total no. (%)
Percentage of EGFR-expressing cells
0.87
≤10
>10 to ≤20
>20 to ≤35
>35
25/109 (22.9)
4/20 (20.0)
6/27 (22.2)
15/62 (24.2)
4/56 (7.1)
5/16 (31.3)
0/7 (0)
3/32 (9.4)
EGFR staining intensity 0.64
Faint
Weak or moderate
Strong
11/53 (20.8)
22/89 (24.7)
17/75 (22.7)
1/21 (4.8)
7/55 (12.7)
4/34 (11.8)
Acne-like rash
None vs. any
None
Any
8/48 (16.7)
42/170 (24.7)
2/28 (7.1)
10/83 (12.0)
0.18
None vs. grade 1 or 2 vs. grade 3 or 4
None
Grade 1 or 2
Grade 3 or 4
8/48 (16.7)
29/148 (19.6)
13/22 (59.1)
2/28 (7.1)
9/79 (11.4)
1/4 (25.0)
0.001
Skin reaction
None vs. any
None
Any
2/32 (6.3)
48/186 (25.8)
0/19 (0)
12/92 (13.0)
0.005
None vs. grade 1 or 2 vs. grade 3 or 4
None
Grade 1 or 2
Grade 3 or 4
2/32 (6.3)
32/157 (20.4)
16/29 (55.2)
0/19 (0)
10/86 (11.6)
2/6 (33.3)
<0.001
The New England Journal of Medicine
Downloaded from nejm.org on May 25, 2011. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.
n engl j med
351;4
www.nejm.org july
22, 2004
cetuximab in irinotecan-refractory colorectal cancer
343
not receive any study medication. In 4 of the 329
enrolled patients (1.2 percent), severe anaphylactic
reactions to cetuximab developed, and the treat-
ment was discontinued. There were no treatment-
related deaths. Acne-like rash occurred in about 80
percent of the patients in each group, but grade 3 or
4 toxic effects on the skin were observed in only
9.4 percent and 5.2 percent of the combination-
therapy and monotherapy groups, respectively. In
225 of the 253 patients in whom any form of acne-
like rash developed (88.9 percent), the rash ap-
peared within the first three weeks after the start of
cetuximab treatment. Diarrhea and neutropenia
were more frequent among the patients receiving
irinotecan in combination with cetuximab than
they were among those receiving cetuximab mono-
therapy, but the frequency of each was in the range
that would be expected for irinotecan alone.
In this randomized trial, cetuximab alone or in com-
bination with irinotecan had clinical activity in irino-
tecan-refractory colorectal cancer, confirming the
results of phase 2 studies.
17,18
The combination-
therapy group had a significantly higher response
rate and a significantly longer time to progression
than the monotherapy group, suggesting that the
combination of irinotecan and cetuximab should
be preferred for patients with irinotecan-refractory
cancer. Moreover, the number of previous treatment
regimens and previous use or nonuse of oxalipla-
tin did not affect the efficacy of the cetuximab-and-
irinotecan combination. Cetuximab monotherapy
also had activity and only mild toxic effects and thus
may be an option for patients who are not consid-
ered candidates for further treatment with irinote-
can-based chemotherapy or who choose not to re-
ceive such treatment.
The effectiveness of the combination of irino-
tecan and cetuximab in patients with irinotecan-
refractory tumors suggests that cetuximab may
circumvent irinotecan resistance. Cells acquire irino-
tecan resistance by several mechanisms.
23
EGFR
inhibition by cetuximab may overcome this resis-
tance by abrogating drug efflux,
23-28
restoring ap-
optosis,
29
or impairing DNA-repair activity.
30,31
An acne-like or maculopapular rash, a charac-
teristic side effect of EGFR blockade, is due to the
role of EGFR in maintaining the integrity of the
skin. Response and survival have been shown to be
correlated with the severity of the rash in phase 2
studies,
32
and the correlation is confirmed by our
study, although it is based on subgroup analyses.
Since no dose-limiting toxicity was observed in
phase 1 studies of cetuximab given according to
the currently recommended dosage regimen,
33
in-
dividualized dose titration based on the occurrence
and severity of rash may improve the effectiveness
of cetuximab treatment. The rates of incidence of
other hematologic and nonhematologic toxic effects
in the combination-therapy group were similar to
those reported in randomized studies of irinotecan
as second-line therapy.
6,7
In contrast to the blockade of HER2 by trastu-
zumab in breast cancer,
34
there was no apparent re-
lationship between the efficacy of cetuximab and
the level of EGFR in the tumor. A similar observa-
tion has been made with gefitinib, another EGFR
tyrosine kinase inhibitor.
35
Only patients with im-
discussion
Figure 2. Time to Disease Progression in the Two Study Groups.
The hazard ratio for disease progression in the combination-therapy group as
compared with the monotherapy group was 0.54 (95 percent confidence in-
terval, 0.42 to 0.71) (P<0.001 by the log-rank test). The points on the curves
represent the dates on which a patient’s data were censored.
Cetuximab plus irinotecan
Cetuximab
Patients Free
of Progression (%)
75
50
25
0
0 2 4 6 8 10
12
Time to Progression (months)
100
Figure 3. Overall Survival in the Two Study Groups.
The hazard ratio for death in the combination-therapy group as compared
with the monotherapy group was 0.91 (95 percent confidence interval, 0.68
to 1.21) (P=0.48 by the log-rank test). The points on the curves represent the
dates on which a patient’s data were censored.
75
50
25
0
0.0 2.5 5.0 7.5 12.510.0 15.0
Patients Surviving (%)
17.5
Overall Survival Time (months)
100
Cetuximab plus irinotecan
Cetuximab
The New England Journal of Medicine
Downloaded from nejm.org on May 25, 2011. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.
n engl j med
351;4
www.nejm.org july
22
,
2004
The
new england journal
of
medicine
344
munohistochemical evidence of EGFR expression
were included in our study, and therefore whether
patients without EGFR expression would benefit
from cetuximab is unknown. EGFR expression has
not been shown to be a predictor of responsiveness
to gefitinib in non–small-cell lung cancer.
36,37
It is
possible that the level of activated, phosphorylated
EGFR is more important than the total EGFR level
35
or, in lung cancer, the presence of mutations in the
ATP-binding site of EGFR,
38
in the prediction of
efficacy.
Patients with advanced colorectal cancer who
receive fluoropyrimidine, irinotecan, and oxalipla-
tin in combination or sequentially may survive 18
to 21 months. However, once these three standard
drugs have failed, there are no accepted treatment
options. Therefore, the observed tumor-response
rate of 22.9 percent among patients with irinotecan-
refractory colorectal cancer is clinically important.
Cetuximab compares favorably with oxaliplatin-
based therapy in patients with irinotecan-refrac-
tory disease. The 22.9 percent rate of response to
cetuximab in combination with irinotecan was high-
er than the response rate reported for oxaliplatin
in combination with infused fluorouracil and leu-
covorin (9.6 percent),
8
but the median time to pro-
gression and the survival time were similar with the
two regimens. Furthermore, previous oxaliplatin
treatment did not negate any benefit of cetuximab.
The one-year survival rates in this group of heavily
pretreated patients — 29 percent in the combina-
tion-therapy group and 32 percent in the monother-
apy group — are encouraging. Moreover, about 48
percent of the patients obtained clinically mean-
ingful disease control.
Supported by grants to all the participating institutions from
Merck (Darmstadt, Germany).
Dr. Cunningham reports having received consulting and lecture
fees from Merck, Aventis, and Sanofi-Synthelabo; Dr. Humblet lec-
ture fees from Aventis, Novartis, Janssen, and Amgen; Dr. Siena con-
sulting fees from Amgen and Roche and lecture fees from Merck,
Amgen, and Roche; Dr. Khayat consulting fees from Merck, Sanofi-
Synthelabo, and Aventis and lecture fees from Merck and Aventis;
and Dr. Van Cutsem lecture fees from Merck.
We are indebted to all the patients who took part in the study
and to their families.
appendix
The following investigators participated in the study: the Netherlands — G.S. Liem, H. Goey, C.J. van Groeningen, and R.J.M. van Leendert;
United Kingdom — D. Cunningham, L. Samuel, T. Hickish, M. Hill, T. Iveson, J. Ledermann, P. Mainwaring, and J. Cassidy; Germany — C.
Bokemeyer, M. Möhler, N. Niederle, H.-J. Schmoll, C. Peschel, and E. Kettner; France — M. Ychou, A. de Gramont, M. Ducreux, M.C. Ka-
minsky, J.-F. Seitz, D. Khayat, and Y. Becouarn; Austria — W. Scheithauer, H. Ludwig, J. Schüller, H. Zwierzina, H. Samonigg, and H. Haus-
maninger; Belgium — E. Van Cutsem, J. van Laethem, H. Bleiberg, and Y. Humblet; Sweden — H. Mellstedt and M. Albertsson; Spain — A.
Cervantes Ruipérez, J. Tabernero Caturla, E. Diaz-Rubio, P. Gascón Vilaplana, M. Navarro Garcia, and A. Antón Torres; Italy — E. Bajetta,
P. Conte, F. de Braud, A. Santoro, S. Siena, S. Artale, F. Grossi, A.R. Bianco, and A. Sobrero; Switzerland — R. Herrmann and C. Sessa; and
Norway — S. Dueland.
references
1.
de Gramont A, Figer A, Seymour M, et
al. Leucovorin and fluorouracil with or with-
out oxaliplatin as first-line treatment in ad-
vanced colorectal cancer. J Clin Oncol 2000;
18:2938-47.
2.
Douillard JY, Cunningham D, Roth AD,
et al. Irinotecan combined with fluorouracil
compared with fluorouracil alone as first-
line treatment for metastatic colorectal can-
cer: a multicentre randomised trial. Lancet
2000;355:1041-7. [Erratum, Lancet 2000;
355:1372.]
3.
Giacchetti S, Perpoint B, Zidani R, et al.
Phase III multicenter randomized trial of ox-
aliplatin added to chronomodulated fluoro-
uracil-leucovorin as first-line treatment of
metastatic colorectal cancer. J Clin Oncol
2000;18:136-47.
4.
Goldberg RM, Sargent DJ, Morton RF,
et al. A randomized controlled trial of fluoro-
uracil plus leucovorin, irinotecan, and oxali-
* Four patients who were randomly assigned to the combination-therapy group
did not receive irinotecan and were thus evaluated for safety in the monother-
apy group. Two patients who were randomly assigned to the combination-
therapy group did not receive any study medication.
Table 4. Grade 3 or 4 Adverse Events Related to Treatment.*
Adverse Event
Cetuximab
plus Irinotecan
(N=212)
Cetuximab
Monotherapy
(N=115) P Value
no. of patients (%)
Any
138 (65.1) 50 (43.5) <0.001
Anemia 10 (4.7) 3 (2.6) 0.55
Neutropenia 20 (9.4) 0 <0.001
Thrombocytopenia 1 (0.5) 1 (0.9) 1.00
Diarrhea 45 (21.2) 2 (1.7) <0.001
Asthenia 29 (13.7) 12 (10.4) 0.49
Acne-like rash 20 (9.4) 6 (5.2) 0.20
Nausea and vomiting 15 (7.1) 5 (4.3) 0.47
Abdominal pain 7 (3.3) 6 (5.2) 0.39
Stomatitis 5 (2.4) 1 (0.9) 0.67
Dyspnea 3 (1.4) 15 (13.0) <0.001
Fever 5 (2.4) 0 0.17
Hypersensitivity reaction 0 4 (3.5) 0.01
Death 0 0 1.00
The New England Journal of Medicine
Downloaded from nejm.org on May 25, 2011. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.
n engl j med
351;4
www.nejm.org july
22, 2004
cetuximab in irinotecan-refractory colorectal cancer
345
platin combinations in patients with previ-
ously untreated metastatic colorectal cancer.
J Clin Oncol 2004;22:23-30.
5.
Saltz LB, Cox JV, Blanke C, et al. Irinote-
can plus fluorouracil and leucovorin for met-
astatic colorectal cancer. N Engl J Med 2000;
343:905-14.
6.
Cunningham D, Pyrhonen 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;
352:1413-8.
7.
Rougier P, Van Cutsem E, Bajetta E, et al.
Randomised trial of irinotecan versus fluo-
rouracil by continuous infusion after fluo-
rouracil failure in patients with metastatic
colorectal cancer. Lancet 1998;352:1407-
12. [Erratum, Lancet 1998;352:1634.]
8.
Rothenberg ML, Oza AM, Bigelow RH,
et al. Superiority of oxaliplatin and fluoro-
uracil-leucovorin compared with either
therapy alone in patients with progressive
colorectal cancer after irinotecan and fluo-
rouracil-leucovorin: interim results of a
phase III trial. J Clin Oncol 2003;21:2059-69.
9.
Messa C, Russo F, Caruso MG, Di Leo A.
EGF, TGF-alpha, and EGF-R in human colo-
rectal adenocarcinoma. Acta Oncol 1998;
37:285-9.
10.
Porebska I, Harlozinska A, Bojarowski
T. Expression of the tyrosine kinase activity
growth factor receptors (EGFR, ERB B2,
ERB B3) in colorectal adenocarcinomas and
adenomas. Tumour Biol 2000;21:105-15.
11.
Salomon DS, Brandt R, Ciardiello F,
Normanno N. Epidermal growth factor-rela-
ted peptides and their receptors in human
malignancies. Crit Rev Oncol Hematol 1995;
19:183-232.
12.
Goldstein NS, Armin M. Epidermal
growth factor receptor immunohistochem-
ical reactivity in patients with American Joint
Committee on Cancer Stage IV colon adeno-
carcinoma: implications for a standardized
scoring system. Cancer 2001;92:1331-46.
13.
Mayer A, Takimoto M, Fritz E, Schelland-
er G, Kofler K, Ludwig H. The prognostic
significance of proliferating cell nuclear an-
tigen, epidermal growth factor receptor, and
mdr gene expression in colorectal cancer.
Cancer 1993;71:2454-60.
14.
Klapper LN, Kirschbaum MH, Sela M,
Yarden Y. Biochemical and clinical implica-
tions of the ErbB/HER signaling network of
growth factor receptors. Adv Cancer Res
2000;77:25-79.
15.
Ciardiello F, Tortora G. A novel approach
in the treatment of cancer: targeting the epi-
dermal growth factor receptor. Clin Cancer
Res 2001;7:2958-70.
16.
Prewett MC, Hooper AT, Bassi R, Ellis
LM, Waksal HW, Hicklin DJ. Enhanced anti-
tumor activity of anti-epidermal growth fac-
tor receptor monoclonal antibody IMC-C225
in combination with irinotecan (CPT-11)
against human colorectal tumor xenografts.
Clin Cancer Res 2002;8:994-1003.
17.
Saltz L, Rubin MS, Hochster HS, et al.
Cetuximab (IMC-C225) plus irinotecan
(CPT-11) is active in CPT-11 refractory co-
lorectal cancer that expresses epidermal
growth factor receptor. Prog Proc Am Soc
Clin Oncol 2001;20:3a. abstract.
18.
Saltz LB, Meropol NJ, Loehrer PJ Sr, Nee-
dle MN, Kopit J, Mayer RJ. Phase II trial of
cetuximab in patients with refractory colo-
rectal cancer that expresses the epidermal
growth factor receptor. J Clin Oncol 2004;
22:1201-8.
19.
Therasse P, Arbuck SG, Eisenhauer EA,
et al. New guidelines to evaluate the re-
sponse to treatment in solid tumors: Euro-
pean Organization for Research and Treat-
ment of Cancer, National Cancer Institute of
the United States, National Cancer Institute of
Canada. J Natl Cancer Inst 2000;92:205-16.
20.
Miller AB, Hoogstraten B, Staquet M,
Winkler A. Reporting results of cancer treat-
ment. Cancer 1981;47:207-14.
21.
Kaplan EL, Meier P. Nonparametric esti-
mation from incomplete observations. J Am
Stat Assoc 1958;53:457-81.
22.
Peto R, Peto J. Asymptotically efficient
invariant test procedures. J R Stat Soc [A]
1972;135:185-206.
23.
Xu Y, Villalona-Calero MA. Irinotecan:
mechanisms of tumor resistance and novel
strategies for modulating its activity. Ann
Oncol 2002;13:1841-51.
24.
Chen ZS, Furukawa T, Sumizawa T, et al.
ATP-dependent efflux of CPT-11 and SN-38
by the multidrug resistance protein (MRP)
and its inhibition by PAK-104P. Mol Phar-
macol 1999;55:921-8.
25.
Chu XY, Suzuki H, Ueda K, Kato Y, Ak-
iyama S, Sugiyama Y. Active efflux of CPT-11
and its metabolites in human KB-derived cell
lines. J Pharmacol Exp Ther 1999;288:735-
41.
26.
Meyers MB, Merluzzi VJ, Spengler BA,
Biedler JL. Epidermal growth factor receptor
is increased in multidrug-resistant Chinese
hamster and mouse tumor cells. Proc Natl
Acad Sci U S A 1986;83:5521-5.
27.
Meyers MB, Yu P, Mendelsohn J. Cross-
talk between epidermal growth factor re-
ceptor and P-glycoprotein in actinomycin
D-resistant Chinese hamster lung cells. Bio-
chem Pharmacol 1993;46:1841-8.
28.
Naruse I, Ohmori T, Ao Y, et al. Antitu-
mor activity of the selective epidermal growth
factor receptor-tyrosine kinase inhibitor
(EGFR-TKI) Iressa (ZD1839) in an EGFR-
expressing multidrug-resistant cell line in
vitro and in vivo. Int J Cancer 2002;98:
310-5.
29.
Sclabas GM, Fujioka S, Schmidt C, Fan
Z, Evans DB, Chiao PJ. Restoring apoptosis
in pancreatic cancer cells by targeting the
nuclear factor-kappaB signaling pathway
with the anti-epidermal growth factor anti-
body IMC-C225. J Gastrointest Surg 2003;7:
37-43.
30.
Bandyopadhyay D, Mandal M, Adam L,
Mendelsohn J, Kumar R. Physical interac-
tion between epidermal growth factor recep-
tor and DNA-dependent protein kinase in
mammalian cells. J Biol Chem 1998;273:
1568-73.
31.
Huang SM, Harari PM. Modulation of
radiation response after epidermal growth
factor receptor blockade in squamous cell
carcinomas: inhibition of damage repair, cell
cycle kinetics, and tumor angiogenesis. Clin
Cancer Res 2000;6:2166-74.
32.
Saltz L, Kies MS, Abbruzzese J, Azarnia
N, Needle MN. The presence and intensity
of the cetuximab-induced acne-like rash
predicts increased survival in studies across
multiple malignancies. Prog Proc Am Soc
Clin Oncol 2003;22:204. abstract.
33.
Baselga J, Pfister D, Cooper MR, et al.
Phase I studies of anti-epidermal growth fac-
tor receptor chimeric antibody C225 alone
and in combination with cisplatin. J Clin
Oncol 2000;18:904-14.
34.
Slamon DJ, Leyland-Jones B, Shak S, et
al. Use of chemotherapy plus a monoclonal
antibody against HER2 for metastatic breast
cancer that overexpresses HER2. N Engl J
Med 2001;344:783-92.
35.
Ciardiello F, Tortora G. Epidermal
growth factor receptor (EGFR) as a target in
cancer therapy: understanding the role of
receptor expression and other molecular
determinants that could influence the re-
sponse to anti-EGFR drugs. Eur J Cancer
2003;39:1348-54.
36.
Kris MG, Natale RB, Herbst RS, et al. Ef-
ficacy of gefitinib, an inhibitor of the epider-
mal growth factor receptor tyrosine kinase,
in symptomatic patients with non-small cell
lung cancer: a randomized trial. JAMA 2003;
290:2149-58.
37.
Fukuoka M, Yano S, Giaccone G, et al.
Multi-institutional randomized phase II tri-
al of gefitinib for previously treated patients
with advanced non-small-cell lung cancer.
J Clin Oncol 2003;21:2237-46.
38.
Lynch TJ, Bell DW, Sordella R, et al. Acti-
vating mutations in the epidermal growth
factor receptor underlying responsiveness
of non–small-cell lung cancer to gefitinib.
N Engl J Med 2004;350:2129-39.
Copyright © 2004 Massachusetts Medical Society.
The New England Journal of Medicine
Downloaded from nejm.org on May 25, 2011. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.
