original articles Annals of Oncology

Annals of Oncology 25: 17561761, 2014

doi:10.1093/annonc/mdu230
Published online 18 June 2014

Clinical activity of FOLFIRI plus cetuximab according

to extended gene mutation status by next-generation
sequencing: ndings from the CAPRI-GOIM trial
F. Ciardiello1,*, N. Normanno2, , E. Maiello3, E. Martinelli1, T. Troiani1, S. Pisconti4, F. Giuliani5,
C. Barone6, G. Carten7, A. M. Rachiglio2, V. Montesarchio8, G. Tonini9, D. Rizzi5, S. Cinieri10,
R. Bordonaro11, A. Febbraro12, F. De Vita1, M. Orditura1, F. Fenizia2, M. Lambiase2, A. Rinaldi13,
F. Tatangelo2, G. Botti14 & G. Colucci5
1
Department of Clinical and Experimental Medicine F. Magrassi, Medical Oncology, Second University of Naples, Naples; 2Cell Biology and Biotherapy Unit, National
Cancer Institute Fondazione Giovanni Pascale, Naples; 3Medical Oncology, Hospital Casa Sollievo Della SofferenzaSan Giovanni Rotondo (Foggia), San Giovanni
Rotondo; 4Department of Medical Oncology, Hospital SS. Annunziata, Taranto; 5Department of Medical Oncology, National Cancer Institute Giovanni Paolo II, Bari;
6
Department of Medical Oncology, University Hospital A. Gemelli, Rome; 7Department of Medical Oncology, Hospital A. Cardarelli, Naples; 8Department of Medical
Oncology, Hospital Monaldi- Azienda Ospedaliera dei Colli, Napoles; 9Department of Medical Oncology, Univeristy Hospital Campus Bio-Medico di Rome, Rome;
10
Department of Medical Oncology, Hospital A. Perrino, Brindisi; 11Department of Medical Oncology, Hospital Garibaldi, Nesima, Catania; 12Department of Medical
Oncology, Hospital Sacro Cuore di Ges, Fatebenefratelli, Benevento; 13Department of Medical Oncology, Hospital Polo Occidentale, Castellaneta, Bari; 14Department
of Pathology, National Cancer Institute Fondazione Giovanni Pascale, Naples, Italy

Received 14 March 2014; revised 26 May 2014; accepted 10 June 2014

Background: Treatment with antiepidermal growth factor receptor (anti-EGFR) monoclonal antibodies has been
restricted to metastatic colorectal cancer (mCRC) patients with RAS wild-type tumors. Next-generation sequencing
(NGS) allows the assessment in a single analysis of a large number of gene alterations and might provide important pre-
dictive and prognostic information.
Patients and methods: In the CAPRI-GOIM trial, 340 KRAS exon 2 wild-type mCRC patients received rst-line
FOLFIRI plus cetuximab. Tumor samples (182/340, 53.5%) were assessed by NGS to search for mutations in 22 genes
involved in colon cancer.
Results: Objective responses in the NGS cohort were observed in 104/182 patients [overall response rate (ORR) 57.1%;
95% condence interval (95% CI) 52% to 66.4%] with a median progression-free survival (mPFS) of 9.8 (95% CI 8.711.5)
months. NGS analysis was successfully completed in all 182 samples. One or more gene mutations (up to ve) were detected
in 124/182 (68.1%) tumors within 14/22 genes for a total of 206 mutations. KRAS exon 2 mutations were identied in 29/182
(15.9%) samples, dened as wild type by local laboratory assessment. Frequently mutated genes were: TP53 (39.6%),
KRAS exons 3/4 (8.8%), NRAS exons 2/3 (7.1%), PIK3CA exons 9/20 (13.2%), BRAF (8.2%). FOLFIRI plus cetuximab treat-
ment determined ORR of 62.0% (95% CI 55.5% to 74.6%) with mPFS of 11.1 (95% CI 9.212.8) months in patients with
KRAS and NRAS wild-type tumors. Conversely, ORR was 46.6% (95% CI 39.957.5%) with mPFS of 8.9 (95% CI 7.49.6)
months in patients with KRAS or NRAS mutations. Similarly, the subgroup of patients carrying KRAS, NRAS, BRAF, or
PIK3CA mutations showed a worse outcome, although this might be due to a prognostic effect.
Conclusions: This study demonstrates that NGS analysis in mCRC is feasible, reveals high level of intra and intertumor het-
erogeneity, and identies patients that might benet of FOLFIRI plus cetuximab treatment.
Key words: colorectal cancer, next-generation sequencing, cetuximab, KSA/NRAS, BRAF, PIK3CA

introduction option [1]. A subset of mCRC is dependent on epidermal

Molecular targeted agents with chemotherapy in treatment of
metastatic colorectal cancer (mCRC) is an effective therapeutic
*Correspondence to: Dr Fortunato Ciardiello, Via Pansini 5, 80131 Naples. Tel: +39-081-
5666745; Fax: +39-081-5666732; E-mail: fortunato.ciardiello@unina2.it
growth factor receptor (EGFR) activation and treatment with
anti-EGFR monoclonal antibodies (moAbs), such as cetuximab
or panitumumab, in combination with standard chemotherapy
(FOLFIRI or FOLFOX) is an effective therapeutic approach [2, 3].
Activating mutations in exon 2 of KRAS gene predict mCRC re-
sistance to anti-EGFR therapies, bringing to the rst molecular

These authors equally contributed to this paper. marker for clinical use in this disease [4]. Mutations in exon 2

The Author 2014. Published by Oxford University Press on behalf of the European Society for Medical Oncology.
All rights reserved. For permissions, please email: journals.permissions@oup.com.
Annals of Oncology
(codon 12 or 13) of the KRAS gene account for 90% of KRAS
mutations in mCRC [4]. Less-frequent KRAS mutations in exons
3 and 4, or NRAS mutations are markers of resistance to anti-
EGFR moAbs in mCRC [58]. The European Medicine Agency
restricted the use of anti-EGFR moAbs to mCRC patients whose
tumors are wild type for both KRAS and NRAS genes.
In colorectal cancer, other genes encoding for key intracellu-
lar molecular transducers of EGFR activation, such as BRAF,
PIK3CA, and PTEN, could be potentially associated with resist-
ance to moAbs [810]. However, no rm conclusions can be
drawn since studies of EGFR moAbs as rst-line treatment of
mCRC suggest prognostic rather than predictive effect of BRAF
mutations, insufcient data are available for PIK3CA mutations
[2, 3].
Since molecular alterations in different signal transduction
genes play a role in resistance to anti-EGFR agents, optimization
of these therapies in mCRC could require more extended molecu-
lar classication. Next-generation sequencing (NGS) techniques
are powerful diagnostic tools that could allow the assessment in
a single analysis of a large number of gene alterations for better
patients selection treatment. NGS can provide information on
inter- and intratumor heterogeneity that might allow a better
stratication of patients eligible for personalized therapy.
However, the feasibility of NGS in clinical diagnostics needs to
be demonstrated.
The Cetuximab After Progression in KRAS wIld-type colo-
rectal cancer patients (CAPRI)Gruppo Oncologico dellItalia
Meridionale (GOIM) study explores the role of EGFR inhibition in
second-line treatment of KRAS exon 2 wild-type mCRC patients
after progression from rst-line treatment with cetuximab; mCRC
patients were treated with FOLFIRI plus cetuximab in rst line
and at progression were randomized to receive FOLFOX alone
or FOLFOX plus cetuximab. Three hundred forty patients were
enrolled in rst line. The second-line part of the trial is currently
ongoing.
Here, we report the results of NGS mutational analysis of a
subgroup of patients enrolled in CAPRI trial. Correlation between
mutational prole of these patients and efcacy of rst-line
FOLFIRI plus cetuximab therapy is also described.
patients and methods
study design and patient population
CAPRI is a nonprot academic, open-label, multicenter study carried out by
the GOIM cooperative group (Eudract number: 2009-014041-81; see supple-
mentary Materials, available at Annals of Oncology online). Twenty-ve
centers participated to the trial (see supplementary Materials, available at
Annals of Oncology online). Here, we present a retrospective, descriptive ana-
lysis of mutational prole carried out by NGS of a subset of patients enrolled
in CAPRI trial. We also provide in a descriptive manner the correlation
between mutational prole of these patients and efcacy of rst-line
FOLFIRI plus cetuximab therapy. The protocol was approved in each center
by local independent Ethics Committee.
multiple gene mutation analysis
by next-generation sequencing
Tumor samples were analyzed with the Ion AmpliSeq Colon and Lung
Cancer Panel (Life Technologies) using Ion Torrent semiconductor
Volume 25 | No. 9 | September 2014
original articles
sequencing. The panel contains primer pairs to analyze over 500 known
mutations and eventually novel mutations in 87 hotspot regions of the fol-
lowing 22 genes: ALK, EGFR, ERBB2, ERBB4, FGFR1, FGFR2, FGFR3,
MET, DDR2, KRAS, PIK3CA, BRAF, AKT1, PTEN, NRAS, MAP2K1,
STK11, NOTCH1, CTNNB1, SMAD4, FBXW7, TP53 [11] (supplementary
Figure S1, available at Annals of Oncology online). A detailed protocol is pro-
vided in supplementary Materials, available at Annals of Oncology online.
results
patients characteristics
From 15 July 2009 to 1 June 2013, 340 mCRC patients [intention-
to-treat (ITT) population], with KRAS exon 2 wild-type tumors,
determined by local laboratories, were enrolled in 25 Italian
centers in CAPRI-GOIM trial (see supplementary Table S1, avail-
able at Annals of Oncology online). Patients received at least one
infusion of FOLFIRI plus cetuximab with a median treatment
duration of 12 cycles (6 months; range, 229 months) (see sup-
plementary Table S2, available at Annals of Oncology online for
toxicities).
multiple gene mutation analysis
by next-generation sequencing
For 182 of 340 (53.5%) patients, FFPE tumor tissues, available
in participating centers, were centrally collected and analyzed
with Ion AmpliSeq Colon and Lung Cancer Panel. A 2% sen-
sitivity threshold has been set for this panel following the results
of a validation study [11]. High-quality DNA was extracted from
all 182 samples thus multiple gene mutation assessment was
possible in all these cases. As shown in Table 1A, no mutations
in 22 tested genes were found in 58 of 182 (31.9%) cases;
whereas one or more genes were mutated in 124 of 182 (68.1%)
samples. Among the 22 genes, mutations were only found in 14
genes. The most frequently mutated gene was TP53 (72/182,
39.5%) (Table 1B). Two different co-existing mutations in TP53
gene were observed in seven cases. Forty-six KRAS gene muta-
tions were detected, with one tumor sample having two different
KRAS mutations. Therefore, KRAS gene was mutated in 45 of
182 (24.7%) cases. Surprisingly, KRAS exon 2 (i.e. codon 12 or
13) mutations were identied in 29 of 182 (15.9%) samples, al-
though they were previously classied as having a KRAS exon 2
wild-type tumor by local laboratory measurement and, for this
reason, enrolled in the current study. All KRAS exon 2 muta-
tions were conrmed by using Sanger sequencing or the
Therascreen KRAS RGQ kit. Mutations in exons 3 or 4 of KRAS
gene were detected in 16 of 182 (8.8%) cases, while NRAS exons
2 or 3 mutations were found in 13 of 182 (7.1%) samples. All
KRAS and NRAS mutations were point mutations in previously
reported hot spot regions (data not shown). PIK3CA gene muta-
tions occurred in 24 of 182 (13.2%) cases. Two co-existing add-
itional PIK3CA mutations were detected in 2 of these 24 cases
for a total of 26 mutations. Sixteen PIK3CA mutations were
found in exon 9; whereas 10 mutations were located in exon 20.
BRAF mutations were identied in 15 of 182 (8.2%) tumors, with
10 cases showing a V600E mutation, while other 5 samples were
mutated in other regions of the BRAF gene. Less-frequent muta-
tions were found in other nine genes, including MET (7/182,
doi:10.1093/annonc/mdu230 |
original articles Annals of Oncology

Table 1. (A) Twenty-two multiple gene mutation analysis in mCRC treated with FOLFIRI + cetuximab; (B) Number of cases (>2%) with most
frequently gene mutations; (C) Number of cases (2%) with less-frequent gene mutations

A
n/N (%)

Analyzed for 22 gene mutations 182/340 (53.5%)

Wild type in all 22 gene analyzed 58/182 (31.9%)
Mutated at 1 of 22 genes analyzed 124/182 (68.1%)
Total mutations 206
Mutated genes 14/22
B
Gene Number of cases (>2%) with mutations, n (%) (N = 182 analyzed)
TP53 72 (39.5%)
KRAS 45 (24.7) 30 at codon 12 or 13 (16.5%); 16 at other (8.8%)
PIK3CA 24 (13.2%) 16 at exon 9 (8.8%); 10 at exon 20 (5.5%)
BRAF 15 (8.2%) 10 at codon 600 (5.5%); 5 at other (2.7%)
NRAS 13 (7.1%)
MET 7 (3.8%)
FBXW7 9 (4.9%)
C
Gene Number of cases (2%) with mutations, n (%) (N = 182 analyzed)
EGFR 2 (1.1%)
CTNNB1 2 (1.1%)
FGFR3 2 (1.1%)
SMAD4 2 (1.1%)
ERBB2 1 (0.55%)
FGFR2 1 (0.55%)
PTEN 1 (0.55%)

3.8%) (Table 1C). Mutations in the EGFR and in the ERBB2
genes occurred very rarely (in two and in one cases, respectively).
Multiple gene analysis by NGS identied several tumor samples
with co-existing mutations in different genes (Table 2 and supple-
mentary Table S3, available at Annals of Oncology online). In this
respect, 206 mutations were detected in 124 samples. Thirty of
45 cases with mutated KRAS gene had concomitant mutations
in other genes, including TP53, PIKCA exons 9 and 20, BRAF,
MET, PTEN, EGFR, or ERBB2. However, the presence of KRAS
and NRAS mutations was mutually exclusive. The analysis by
NGS revealed that, of 15 tumor samples with BRAF mutations,
BRAF was the only mutated gene in three cases. In the other 12
cases, BRAF mutations co-existed with other mutations, includ-
ing TP53, KRAS, and PIK3CA exon 9 and exon 20. A single
PI3KCA mutation was found in only 5 of 24 samples with
mutated PIK3CA gene. More frequently, PIK3CA mutations were
detected together with other gene mutations, including TP53,
KRAS, NRAS, and/or BRAF. In one case, a total of ve different
mutations were detected (PIK3CA exon 9 and exon 20, KRAS,
ERBB2, TP53).
clinical activity of FOLFIRI plus cetuximab
according to multiple gene mutation analysis
by next-generation sequencing
We evaluated if NGS cohort of 182 patients was representative
of entire ITT population of 340 patients. Clinical activity of
FOLFIRI plus cetuximab was similar in this subgroup of patients
when compared with ITT group (Table 3A). An overall response
rate (ORR) of 57.1% and 56.4% were, respectively, reported in
NGS and ITT populations. The median progression-free survival
(mPFS) in NGS cohort of patients was of 9.8 months, whereas a
median PFS of 9.9 months was observed in ITT population. We
evaluated if the presence of gene mutations that could correlate
with anti-EGFR moAbs efcacy in mCRC, such KRAS, NRAS,
BRAF, and PIK3CA, affect the clinical activity of FOLFIRI plus
cetuximab treatment. As illustrated in Table 3B, among the
patients whose tumors were analyzed by NGS, 124 of 182 (68.1%)
tumor specimen resulted wild type for both KRAS and NRAS
genes. Treatment determined in the KRAS/NRAS wild-type
cohort an ORR of 62.0%. Stable disease was reported in 35 of 124
(28.2%) patients, and 12 of 124 (9.7%) patients experienced pro-
gressive disease. The median PFS in the KRAS/NRAS wild-type
group was of 11.1 months. On the contrary, mCRC patients with
a tumor harboring KRAS or NRAS mutations achieved an ORR
of 46.6% with a median PFS of 8.9 months. We evaluated the clin-
ical activity of FOLFIRI plus cetuximab in patients whose tumors
did not carry mutations in KRAS, NRAS, BRAF, and PIK3CA
genes (Table 3C). FOLFIRI plus cetuximab treatment determined
an ORR of 64.4% with mPFS of 11.3, whereas the ORR was
47.4% with mPFS of 7.7 months in patients with a mutation in
any of these four genes. However, we must acknowledge that this
difference could be due to a prognostic rather than predictive
effect of BRAF and PI3CA mutations. No difference was found
between the presence or the absence of TP53 gene mutations and
clinical activity of FOLFIRI plus cetuximab (data not shown).

| Ciardiello et al. Volume 25 | No. 9 | September 2014

Annals of Oncology original articles
discussion

TP53
NGS is a fascinating novel technique that allows obtaining infor-

18
3
9
8
3
1
1
1
1
1

5
mation on mutational status of a large number of genes in a
single analysis. In fact, this study is the rst to demonstrate that
FBXW7

mutational analysis of mCRC with NGS correlates with patients

5

5
outcome.
Multiple gene mutation analysis of patients enrolled in the
FGFR2

CAPRI trial was carried out with the Ion AmpliSeq Colon
1

1
and Lung Cancer Panel [11]. Analysis with this platform was
successfully conducted in all 182 samples retrospectively
ERBB2

obtained from several different surgical pathologies (>20).

1

1
1

1
Therefore, our ndings suggest that NGS analysis might be feas-
ible in a scenario resembling the clinical diagnostic setting.
PTEN

Nevertheless, NGS is a complex technique that should be

1

carried out only in highly specialized centers with expert per-

FGFR3

sonnel. Dedicated external quality assessment (EQA) schemes

for NGS should also be developed to ensure adequate quality of
1

1
analysis.
CTNNB1

The results of the CAPRI-GOIM trial conrmed the efcacy

of the rst-line combination of FOLFIRI plus cetuximab in
KRAS exon 2 wild-type mCRC patients as previously reported
1

[1, 12]. More recently, other less-frequent RAS gene mutations

SMAD4

(KRAS exons 3 and 4 and NRAS exons 2, 3, and 4) are predict-

ive markers of resistance to anti-EGFR therapies in mCRC
1

patients [68, 1315]. The analysis with the Ion AmpliSeq

EGFR

Colon and Lung Cancer Panel identies the subgroup of mutant

1

RAS patients that do not benet of anti-EGFR moAbs treat-

ment. Differential clinical activity of FOLFIRI plus cetuximab
MET

was observed in mCRC patients whose tumors were wild type

1
1

for both KRAS and NRAS genes compared with patients whose
PIK3CA ex20

tumors were carrying KRAS or NRAS mutations with ORR of

62.0% and mPFS of 11.1 months versus ORR of 46.6% and
mPFS of 8.9 months, respectively. Similarly, patients carrying
either KRAS, NRAS, BRAF, or PIK3CA mutations showed a
5

3
2

2
3

worse outcome when compared with quadruple wild-type

PIK3CA ex9
Table 2. Coexisting mutations in different KRAS, NRAS, BRAF, and PIK3CA

patients. However, this could be due to prognostic effects of

BRAF and PIK3CA mutations in mCRC rather than to the true
predictive value of these mutations for FOLFIRI plus cetuximab
9
1
1

2
1
1

treatment. The results of CRYSTAL, PRIME, and FIRE-3 trials

BRAF

do suggest that BRAF mutations are strong negative prognostic

biomarkers for mCRC patients treated with chemotherapy alone
4

1
3

1
2
9

(FOLFIRI or FOLFOX), chemotherapy plus anti-EGFR moAbs

NRAS

(cetuximab and panitumumab), and chemotherapy (FOLFIRI)

Cases with multiple mutations/total mutated cases.
1

plus bevacizumab [1214]. Furthermore, PIK3CA mutation has

no effect on the efcacy of FOLFIRI plus cetuximab in the FIRE-
KRAS

3 trial [14]. Therefore, no evidence sustains the use of BRAF and

18
4
9
5
1
1
1

1
1
1

PIK3CA mutations for the selection of mCRC patients sensitive

Mutated cases (N = 182 analyzed)

to anti-EGFR moAbs, although further studies should be

carried out to explore this hypothesis.
A limit of the Ion AmpliSeq Colon and Lung Cancer Panel
used in this study is that it does not cover exon 4 NRAS muta-
PIK3CA ex9 (14/16)a
PIK3CA ex20 (7/10)a

tions. However, in this study, KRAS exon 3 or 4 and NRAS exon

CTNNB1 (2/2)a

2 or 3 mutations were detected in 15.9% of cases, a frequency

KRAS (30/45)a

BRAF (12/15)a

SMAD4 (2/2)a

FBXW7 (9/9)a
TP53 (36/72)a
NRAS (5/13)a

FGFR3 (2/2)a

ERBB2 (1/1)a
FGFR2 (1/1)a
PTEN (1/1)a
EGFR (1/2)a

similar to other studies, such as PRIME, PEAK, and FIRE-3

MET (4/7)a

trials, in which no NRAS exon 4 mutations were as well iden-

tied [6, 13, 14].
Centralized, highly sensitive multiple gene analysis by NGS
a

also enabled the identication of 29 additional patients with

Volume 25 | No. 9 | September 2014 doi:10.1093/annonc/mdu230 |

original articles Annals of Oncology

Table 3. First-line efcacy data according to gene mutations

A B C
Clinical activity of Total patients Patients with 22 KRAS/NRAS wt KRAS/NRAS mt KRAS/RAS/ KRAS/NRAS/
FOLFIRI + cetuximab (N = 340) multiple gene (n = 124) (n = 58) BRAF/PIK3CA BRAF/PIK3CA
mutation analysis wt (n = 104) mt (n = 78)
(n = 182)

Complete response, % 26/340 (7.6%) 12/182 (6.6%) 8/124 (6.4%) 4/58 (6.9%) 8/104 (7.7%) 4/78 (5.1%)
Partial response, % 166/340 (48.8%) 92/182 (50.5%) 69/124 (55.6%) 23/58 (39.7%) 59/104 (56.7%) 33/78 (42.3%)
Stable disease, % 115/340 (33.8%) 61/182 (33.5%) 35/124 (28.2%) 26/58 (44.8%) 28/104 (26.9%) 33/78 (42.3%)
Progressive disease, % 33/340 (9.7%) 17/182 (9.3%) 12/124 (9.7%) 5/58 (8.6%) 9/104 (8.6%) 8/78 (10.3%)
ORR, % (95% CI) 192/340 (56.4%) 104/182 (57.1%) 77/124 (62.0%) 27/58 (46.6%) 67/104 (64.4%) 37/78 (47.4%)
(52.663.2%) (52.066.4%) (55.574.6%) (39.9 57.5%) (58.276.6%) (39.061.2%)
Median PFS, months 9.9 (8.811.3) 9.8 (8.711.5) 11.1 (9.212.8) 8.9 (7.49.6) 11.3 (9.413.2) 7.7 (5.49.4)
(95% CI)

(A) Clinical activity of FOLFIRI plus cetuximab of the ITT population (340) and NGS cohort (182). (B) Clinical activity of FOLFIRI plus cetuximab
in KRAS/NRAS wild type cohort and KRAS/NRAS mutant. (C) Clinical activity of FOLFIRI plus cetuximab in KRAS/NRAS/BARF and PIK3CA
wild-type cohort and KRAS/NRAS/BARF/PIK3CA mutant.
wt, wild type; mt, mutant.

KRAS exon 2-mutated tumors among the 182 cases that had
been previously classied as KRAS wild-type exon 2 by local
pathology laboratories. To our knowledge, this is the rst report
of a centralized and repeated analysis of KRAS exon 2 mutations
within a clinical trial in mCRC patients treated with anti-EGFR
drugs based on local pathology laboratory assessment. There are
several potential explanations for the difference between original
local laboratory assessment and centrally carried out NGS evalu-
ation. The NGS technique that has been used in the present
study has a sensitivity of 2%, which is much higher when com-
pared with standard Sanger sequencing. In this regard, recent
studies have suggested that low levels of KRAS mutations might
lead to resistance to anti-EGFR moAbs [15]. NGS can also
detect mutations that are not recognized by other methods used
for KRAS testing in clinical practice. However, in the present
study, only 6 cases had a percentage of KRAS exon 2 mutant
alleles <10%, 8 samples had a percentage of mutant alleles
between 10% and 20%, whereas 15 tumors carried >20% KRAS
exon 2 mutant alleles (data not shown). Therefore, most of these
mutations could be identied by Sanger sequencing. In addition,
only one of the identied mutations is not included in the list of
variants detected by commercially available kits (i.e. G13 V).
Finally, we might have carried out NGS analysis on a different
histology section from the one originally used by the local path-
ology laboratory. Nevertheless, a heterogeneous pattern of
KRAS mutations is unlikely to occur in tumors with high fre-
quency of mutant alleles, which were the majority of the cases.
Taken together, these data suggest that false-negative results in
KRAS exon 2 testing could occur in routine clinical practice, as
also suggested by the results of European and Italian EQA
schemes for KRAS mutation assessment [16, 17].
An important nding of the present study is that by using
NGS it is possible to detect multiple gene mutations within the
same tumor sample. Among the 22 genes that were studied, 14
were mutated with a total of 206 gene mutations being found in
124 of 182 samples, with an average of 1.67 mutations per
sample. Genes whose mutations are considered mutually exclu-
sive, such as BRAF and KRAS, were found mutated in the same
tumor sample in some cases, as also recently reported [15]. In
contrast, no sample was found to carry NRAS and BRAF muta-
tions. Although the number of NRAS mutant cases in our study
was low, this might suggest a different pattern of tumorigenesis
for KRAS and NRAS. More importantly, identication of rare
mutations as well as increased knowledge on the comprehensive
mutational prole of each individual tumor might improve de-
velopment of personalized medicine in mCRC. A number of
novel agents targeting molecular alterations in genes involved in
tumor growth and progression are in clinical development, NGS
screening of a wide array of genes might facilitate the identica-
tion of mCRC patients that might be enrolled in trials with new
drugs. Several initiatives are ongoing in including a European
database collecting the results of the different NGS-based
screening programs would help to accelerate the progress of per-
sonalized medicine in mCRC.
In most cases, we found that coexisting mutations occurred at
different frequencies in the involved genes. Because the fre-
quency of the mutation is directly related to the number of
tumor cells that carry the mutant gene, these ndings suggest
that tumor clones with different mutational prole might be
present in the same sample. These results suggest that, in agree-
ment with recent ndings in other tumors [18], CRC is charac-
terized by intertumor heterogeneity, since we could recognize
subgroups of tumors carrying specic molecular alterations, as
well as by intratumor heterogeneity due to the presence of cancer
cells with different molecular alterations in the same sample.
Further studies will be needed to investigate these hypotheses and
how tumor heterogeneity could affect response to therapy with
anti-EGFR drugs.
In summary, the results of present study conrm the efcacy
of cetuximab in combination with FOLFIRI in rst-line treat-
ment of mCRC patients with KRAS and NRAS wild type
tumors. To our knowledge, this is the rst trial in which

| Ciardiello et al. Volume 25 | No. 9 | September 2014

Annals of Oncology
molecular characterization carried out with multiple gene NGS-
based technique has been correlated with clinical outcome in
mCRC patients treated with an anti-EGFR moAbs.
acknowledgements
FC and NN thank the Associazione Italiana per la Ricerca sul
Cancro (AIRC).
funding
This work has been supported by Associazione Italiana per la
Ricerca sul Cancro (AIRC) (no grant number).
disclosure
The authors have declared no conicts of interest.
references
1. Ciardiello F, Tortora G. EGFR antagonists in cancer treatment. N Engl J Med 2008;
358: 11601174.
2. Van Cutsem E, Khne CE, Lang I et al. Cetuximab plus irinotecan, uorouracil, and
leucovorin as rst-line treatment for metastatic colorectal cancer: updated analysis
of overall survival according to tumor KRAS and BRAF mutation status. J Clin
Oncol 2011; 29: 20112019.
3. Douillard JY, Siena S, Cassidy J et al. Randomized, phase III trial of panitumumab
with infusional uorouracil, leucovorin, and oxaliplatin (FOLFOX4) versus FOLFOX4
alone as rst-line treatment in patients with previously untreated metastatic
colorectal cancer: the PRIME study. J Clin Oncol 2010; 28: 46974705.
4. Normanno N, Tejpar S, Morgillo F et al. Implications for KRAS status and EGFR-
targeted therapies in metastatic CRC. Nat Rev Clin Oncol 2009; 6: 519527.
5. De Roock W, Claes B, Bernasconi D et al. Effects of KRAS, BRAF, NRAS, and
PIK3CA mutations on the efcacy of cetuximab plus chemotherapy in
chemotherapy-refractory metastatic colorectal cancer: a retrospective consortium
analysis. Lancet 2010; 11: 753762.
6. Schwartzberg LS, Rivera F, Karthaus M et al. Analysis of KRAS/NRAS mutations in
PEAK: a randomized phase II study of FOLFOX6 plus panitumumab ( pmab) or
Volume 25 | No. 9 | September 2014
original articles
bevacizumab (bev) as rst-line treatment (tx) for wild-type (WT) KRAS (exon 2)
metastatic colorectal cancer (mCRC). J Clin Oncol 2013; 13(suppl 3): abstr 3631.
7. Stintzing S, Jung A, Rosssius L et al. Analisi of KRAS/NRAS and BRAF mutations
in FIRE-3: a randomized phase III study of Folri plus cetuximab or bevacizumab as
rst-line treatment for wild-type (WT) KRAS (exon 2) metastatic colorectal cancer
(mCRC) patients. Eur J Cancer 2013; 49(suppl 3): LBA17.
8. Sartore-Bianchi A, Di Nicolantonio F, Nichelatti M et al. Multi-determinants
analysis of molecular alterations for predicting clinical benet to EGFR-targeted
monoclonal antibodies in colorectal cancer. PLoS One 2009; 4: e7287.
9. Tabernero J, Cervantes A, Rivera F et al. Pharmacogenomic and pharmacoproteomic
studies of cetuximab in metastatic colorectal cancer: biomarker analysis of a phase I
dose-escalation study. J Clin Oncol 2010; 28: 11811189.
10. De Roock W, De Vriendt V, Nomanno N et al. KRAS, BRAF, PIK3CA, and PTEN
mutations: implications for targeted therapies in metastatic colorectal cancer.
Lancet Oncol 2011; 12: 594603.
11. Normanno N, Petraroli R, Rico A et al. The OncoNetwork Consortium: A European
Collaborative Research study on the development of an Ion AmpliSeq gene panel
targeting hot spots in colon and lung cancers. AACR 2013; abstract 36.
12. Van Cutsem E, Khne CH, Hitre E et al. Cetuximab and chemotherapy as
initial treatment for metastatic colorectal cancer. N Engl J Med 2009; 360:
14081417.
13. Douillard JY, Oliner KS, Siena S et al. Panitumumab-FOLFOX4 treatment and RAS
mutations in colorectal cancer. N Engl J Med 2013; 369: 10231034.
14. Stintzing S, Jung A, Rossius L et al. Mutations within the EGFR signaling pathway:
inuence on efcacy in FIRE-3. A randomized phase III study of FOLFIRI plus
cetuximab or bevacizumab as rst-line treatment for wild-type (WT) KRAS (exon 2)
metastatic colorectal cancer (mCRC) patients. J Clin Oncol 2014; 32(suppl 3):
abstr 445.
15. Molinari F, Felicioni L, Buscarino M et al. Increased detection sensitivity for KRAS
mutations enhances the prediction of anti-EGFR monoclonal antibody resistance in
metastatic colorectal cancer. Clin Cancer Res 2011; 17: 49014914.
16. Normanno N, Pinto C, Castiglione F et al. KRAS mutations testing in colorectal
carcinoma patients in Italy: from guidelines to external quality assessment. PLoS
One 2011; 6: e29146.
17. Bellon E, Ligtenberg MJ, Tejpar S et al. External quality assessment for KRAS
testing is needed: setup of a European program and report of the rst joined
regional quality assessment rounds. Oncologist 2011; 16: 467478.
18. Gerlinger M, Rowan AJ, Horswell S . Intratumor heterogeneity and branched
evolution revealed by multiregion sequencing. N Engl J Med 2012; 366:
883892.
doi:10.1093/annonc/mdu230 |