Fluorouracil, Leucovorin, and Oxaliplatin With and Without Cetuximab in the First-Line Treatment of Metastatic Colorectal Cancer

  1. Carsten Bokemeyer,
  2. Igor Bondarenko,
  3. Anatoly Makhson,
  4. Joerg T. Hartmann,
  5. Jorge Aparicio,
  6. Filippo de Braud,
  7. Serban Donea,
  8. Heinz Ludwig,
  9. Gunter Schuch,
  10. Christopher Stroh,
  11. Anja H. Loos,
  12. Angela Zubel and
  13. Piotr Koralewski

+Author Affiliations

  1. From the University Hospital, Hamburg-Eppendorf; South West German Comprehensive Cancer, University Hospital Tübingen, Tübingen; Merck Serono, Darmstadt Germany; State Medical Academy, City Clinical Hospital #4, Dnepropetrovsk, Ukraine; Moscow City Oncology Hospital #62, Moscow Area, Russian Federation; Hospital Universitario La Fe, Valencia, Spain; Istituto Europeo di Oncologia, Milan, Italy; Institutul Oncologic “Prof. Dr. Al. Trestioreanu” Bucuresti, Bucharest, Romania; Wilhelminenspital der Stadt Wien, Vienna, Austria; and Wojewódzki Szpital Specjalistyczny Oddzia Chemioterapii, Krakow, Poland.
  1. Corresponding author: Carsten Bokemeyer, MD, Universitätsklinikum Hamburg-Eppendorf, Martinistrasse 52, 20246 Hamburg-Eppendorf, Germany; e-mail:c.bokemeyer@uke.uni-hamburg.de.
  1. Presented at the 43rd Annual Meeting of the American Society of Clinical Oncology, June 1-5, 2007, Chicago, IL; 14th European Cancer Conference, September 23-27, 2007, Barcelona, Spain; 44th Annual Meeting of the American Society of Clinical Oncology, May 30-June 3, 2008, Chicago, IL; and The ESMO International Symposium: 10th World Congress on Gastrointestinal Cancer, June 25-28, 2008, Barcelona, Spain.


Purpose This randomized study assessed whether the best overall response rate (ORR) of cetuximab combined with oxaliplatin, leucovorin, and fluorouracil (FOLFOX-4) was superior to that of FOLFOX-4 alone as first-line treatment for metastatic colorectal cancer. The influence of KRAS mutation status was investigated.

Patients and Methods Patients received cetuximab (400 mg/m2 initial dose followed by 250 mg/m2/wk thereafter) plus FOLFOX-4 (oxaliplatin 85 mg/m2 on day 1, plus leucovorin 200 mg/m2 and fluorouracil as a 400 mg/m2 bolus followed by a 600 mg/m2 infusion during 22 hours on days 1 and 2; n = 169) or FOLFOX-4 alone (n = 168). Treatment was continued until disease progression or unacceptable toxicity.KRAS mutation status was assessed in the subset of patients with assessable tumor samples (n = 233).

Results The confirmed ORR for cetuximab plus FOLFOX-4 was higher than with FOLFOX-4 alone (46% v 36%). A statistically significant increase in the odds for a response with the addition of cetuximab to FOLFOX-4 could not be established (odds ratio = 1.52; P = .064). In patients with KRAS wild-type tumors, the addition of cetuximab to FOLFOX-4 was associated with a clinically significant increased chance of response (ORR = 61% v 37%; odds ratio = 2.54; P = .011) and a lower risk of disease progression (hazard ratio = 0.57; P = .0163) compared with FOLFOX-4 alone. Cetuximab plus FOLFOX-4 was generally well tolerated.

Conclusion KRAS mutational status was shown to be a highly predictive selection criterion in relation to the treatment decision regarding the addition of cetuximab to FOLFOX-4 for previously untreated patients with metastatic colorectal cancer.


The epidermal growth factor receptor (EGFR) is a clinically validated anticancer molecular target that is expressed in the majority of colorectal tumors.1–5 Cetuximab (Erbitux, developed by Merck KGaA [Darmstadt, Germany], under license from Imclone New York, NY]) is an immunoglobulin G1 monoclonal antibody that specifically targets EGFR with high affinity, competitively inhibiting endogenous ligand binding and ligand-dependent downstream signaling.6,7 Binding to the tumor cell also initiates antibody-dependent cell-mediated cytotoxicity.8–10 Randomized phase II and III metastatic colorectal cancer (mCRC) studies have demonstrated the efficacy and tolerability of cetuximab, as monotherapy2,11 or in combination with irinotecan,2,12after the failure of previous chemotherapy regimens including irinotecan and/or oxaliplatin. Furthermore, in the first-line setting, building on promising phase II data,13the phase III Cetuximab Combined With Irinotecan in First-Line Therapy for Metastatic Colorectal Cancer (CRYSTAL) study has shown that the addition of cetuximab to infusional fluorouracil (FU)/leucovorin (LV)/irinotecan (FOLFIRI) significantly improves the progression-free survival (PFS) time, response rate, and R0 resection rate in mCRC patients, compared with FOLFIRI alone.14

Phase II studies examining the activity and tolerability of cetuximab combined with oxaliplatin and FU/LV (FOLFOX-4) have provided encouraging results.15,16 To further evaluate this combination, the Oxaliplatin and Cetuximab in First-Line Treatment of mCRC (OPUS) study compared the efficacy and safety of cetuximab plus FOLFOX-4 with FOLFOX-4 alone as first-line therapy for mCRC.

The definition of relevant molecular characteristics of an individual tumor (biomarker evaluation) will increasingly enable the selection of patients most likely to benefit from particular treatments.17 A recent subgroup analysis of a randomized phase III study of the immunoglobulin G2 EGFR-targeting monoclonal antibody panitumumab in patients with chemotherapy-refractory mCRC suggested a lack of clinical activity in patients whose tumors had activating mutations at codon 12 or 13 of the KRASgene.18 Small retrospective studies in chemotherapy-refractory patients have also suggested that clinical responses to cetuximab are confined to those mCRC patients (approximately 55% to 70%) whose tumors are wild type for KRAS.19–22 This assertion has recently been confirmed in subgroup analyses of the randomized phase III CRYSTAL and CO.17 studies.23,24 A subgroup analysis of the OPUS study therefore investigated whether tumor KRAS mutation status was predictive of a clinical response to cetuximab plus FOLFOX-4.


Patient Eligibility

Patients were eligible for inclusion if they were ≥ 18 years old; had a histologically confirmed, first-occurrence of a nonresectable, EGFR-expressing mCRC with at least one radiologically measurable lesion; a life expectancy of ≥ 12 weeks; an Eastern Cooperative Oncology Group Performance Status (ECOG PS) ≤ 2; and adequate hepatic, renal, and bone marrow function. Patients were ineligible if they were pregnant, had a history of previous exposure to EGFR-targeted therapy or previous chemotherapy (except adjuvant treatment) for mCRC, or had uncontrolled severe organ or metabolic dysfunction. The study was approved by relevant ethics committees and was conducted in accordance with the Declaration of Helsinki. All patients provided written, informed consent.

Study Design

This was an open-label, randomized, multicenter phase II study comparing the efficacy and safety of cetuximab combined with FOLFOX-4 versus FOLFOX-4 alone in the first-line treatment of EGFR-expressing mCRC. EGFR expression was determined using a DakoCytomation pharmDx immunohistochemistry kit (Dako, Glostrup, Denmark), with one tumor cell staining to any intensity being the minimum criterion required to confirm expression. Randomization (1:1) was carried out using a stratified permuted-block procedure, with ECOG PS (0 and 1 v 2) as a stratification factor.

On day 1 of a 14-day treatment cycle, patients received cetuximab (initial dose 400 mg/m2 infused during 2 hours, and 250 mg/m2 weekly during 1 hour thereafter) followed after 1 hour by FOLFOX-4 (oxaliplatin 85 mg/m2 on day 1, infused during 2 hours; LV 200 mg/m2, infused during 2 hours, followed by FU as a 400 mg/m2intravenous bolus then a 600 mg/m2 infusion during 22 hours on days 1 and 2) or FOLFOX-4 alone, until the occurrence of progressive disease (PD) or unacceptable toxicity. In the case of unacceptable toxicity or intolerance to FU/LV, oxaliplatin, or cetuximab, the agent responsible could be stopped and the patient could continue with the other study medication(s). Radiologic assessment of response was carried out every 8 weeks until PD or withdrawal for any reason.

The primary objective was to assess whether the best confirmed overall response rate of cetuximab plus FOLFOX-4 was superior to that of FOLFOX-4 alone. Tumor response including disease progression was assessed by an independent review committee according to modified WHO criteria. The independent review committee conducted a blinded review of images and clinical data using a common set of prespecified criteria. The data cutoff for the primary analysis of response was chosen to be 20 weeks after the last randomly assigned patient started study treatment (August 4, 2006) so that all patients could have at least two 8-week evaluations. Secondary objectives included treatment comparisons of the rate of curative metastatic surgery, duration of response, disease control rate, PFS time, OS time, and safety, all of which were based on data from a longer follow-up time (March 1, 2007). A retrospective subgroup analysis investigated associations between PFS and response rate and the KRAS mutation status of tumors.

Statistical Methods and Considerations

The primary efficacy analysis was performed on the intention-to-treat (ITT) population, defined as all randomly assigned patients who received at least one dose of study treatment. Assuming a difference in the rate of best confirmed response of at least 20% between the two treatments (ie, approximately 70% with cetuximab plus FOLFOX-4 and 50% with FOLFOX-4 alone for the stratum with ECOG PS 0 to 1, and 66% and 45%, respectively, for the ECOG PS 2 stratum), the Cochran-Mantel-Haenszel (CMH)25,26 odds ratio (OR) over the strata was expected to be 2.33. A sample size of approximately 146 patients per group was therefore calculated as necessary to detect an odds ratio of 2.33 at level α = .05, with a power of 90%. With the exception of the test of the null hypothesis of OR = 1 over the entire sample population, all further efficacy analyses were exploratory and P values were not adjusted for the multiplicity of statistical tests. Time-to-event data were analyzed using the Kaplan-Meier method.27 All analyses were repeated by KRAS mutation status in those ITT patients whose tumors were available and provided an assessable DNA sample (KRAS population).

KRAS Mutation Detection

Tumor DNA was extracted from formaldehyde-fixed paraffin-embedded tissues, and screened for the presence of KRAS codon 12 and 13 mutations using a polymerase chain reaction (PCR) clamping and melting curve technique adapted from Chen et al28 (full methods information is listed in the Appendix, online only). Briefly, a one-step Lightcycler PCR reaction (LightMix, k-ras Gly12; TIB MOLBIOL, Berlin, Germany) incorporated a codon 12 to 14 spanning locked nucleic acid oligomer to suppress amplification of the wild-type sequence.29 Post-PCR hybridization and melting curve analysis using fluorescently tagged oligonucleotides incorporated in the original PCR reaction permitted the identification and discrimination of distinct KRAS codon 12 and 13 missense mutations.

Safety Analysis

The safety population comprised all patients who received at least one dose of study treatment. All adverse events (AEs) reported with onset on or after the first day of study medication and captured up to 6 weeks after the end of the last administration of study treatment were coded using the Medical Dictionary for Regulatory Activities version 8.1 and summarized by worst severity per patient according to the National Cancer Institute Common Toxicity Criteria version 2.0.


Patient Demographics

Three hundred forty-four patients were randomly assigned (three of them in error). Of the eligible patients, 337 received study treatment at 79 centers. Study accrual lasted from July 2005 until March 2006, with 169 patients receiving cetuximab plus FOLFOX-4 and 168 patients receiving FOLFOX-4 alone (Fig 1). Baseline demographic and disease characteristics were generally well balanced between the treatment groups (Table 1). DNA suitable for KRAS mutation analysis was extracted from the tumor samples of 233 patients, 113 and 120 of whom had received cetuximab plus FOLOFOX-4 and FOLFOX-4 alone, respectively. These patients comprised the KRAS population. In general, the population with tissue available forKRAS analysis (n = 233) was considered to be representative of the overall ITT population (n = 337) in relation to demographic parameters (Table 1).

Fig 1.

Fig 1.

Disposition of subjects at the time of the data cutoff March 1, 2007. EGFR, epidermal growth factor receptor; FOLFOX-4, oxaliplatin, leucovorin, and fluorouracil; ITT, intention to treat.

Treatment Compliance

The median duration of cetuximab treatment was 24 weeks, with 84% of patients having a relative dose-intensity (RDI) of ≥ 80%. Similar numbers of patients in both arms had RDIs of ≥ 80% for oxaliplatin (75% and 80% of patients receiving cetuximab plus FOLFOX-4 and FOLFOX-4, respectively) and FU (67% and 70% of patients, respectively). Reductions and delays in cetuximab dosing were primarily because of skin reactions, and delays in chemotherapy dosing were because of hematologic, GI, or neurologic reactions. In the KRAS population, comparable levels of exposure were reached; 86% of patients who received cetuximab had a RDI ≥ 80%. RDIs of ≥ 80% were also reported for 73% and 81% of patients receiving oxaliplatin and for 65% and 72% of patients receiving FU in the cetuximab plus FOLFOX-4 and FOLFOX-4 arms, respectively.


Efficacy data are summarized in Table 2. The best confirmed overall response rate in the cetuximab plus FOLFOX-4 group was 46%, and in the FOLFOX-4–alone group, this value was 36%. The comparison between the two treatment groups resulted in a common ECOG PS–adjusted OR of 1.52 (95% CI, 0.98 to 2.36), indicating an increase in the chance of a response in the cetuximab plus FOLFOX-4 arm by 52%. This increase was not statistically significant at the 5% level (ECOG PS–adjusted CMH test, P = .064). A slightly greater treatment effect was observed in the group of participants (n = 153 v 152) with ECOG PS 0 to 1 (OR = 1.65; 95% CI, 1.04 to 2.60; CMH test, P = .032). Regarding the secondary efficacy end points, although a marked benefit in relation to PFS time was not detected (Table 2; Fig 2A), the addition of cetuximab to FOLFOX-4 was associated with an approximate doubling of the R0 resection rate from 2.4% (four of 168 patients) for FOLFOX-4 alone to 4.7% (eight of 169 patients) for cetuximab plus FOLFOX-4. The median duration of response was also longer in the cetuximab group (9.0 v 5.7 months).

Efficacy Data in the ITT andKRAS Populations

Fig 2.

Fig 2.

Kaplan and Meier plots for progression-free survival time for patients in each study arm in (A) the intention to treat (ITT) population and (B) the KRAS population. Cet., cetuximab; FOLFOX-4, oxaliplatin, leucovorin, and fluorouracil; mo., months.

KRAS mutations in codons 12/13 (hereinafter called KRAS mutations) were detected in the tumor tissue of 99 of 233 (42%) patients, 52 of whom received cetuximab plus FOLFOX-4 and 47 of whom received FOLFOX-4 alone (Table 2). No mutations were detected in the tumors of 134 patients (58%; hereinafter called KRAS wild-type tumors), 61 of whom received cetuximab plus FOLFOX-4 and 73 of whom received FOLFOX-4 alone. In patients whose tumors were scored as KRAS wild type, the addition of cetuximab to FOLFOX-4 was associated with a 2.54-fold increased chance for response compared with FOLFOX-4 alone (61% v 37%; OR = 2.54; 95% CI, 1.24 to 5.23; ECOG PS–adjusted CMH test, P = .011). Conversely, for patients with tumorKRAS mutations, the response rate for cetuximab plus FOLFOX-4 was lower, compared with FOLFOX-4 alone (33% v 49%; OR = 0.51; 95% CI, 0.22 to 1.15; ECOG PS–adjusted CMH test, P = .106). The R0 resection rate was more than doubled in patients with KRAS wild-type tumors who received cetuximab plus FOLFOX-4 (six of 61 patients; 9.8%) compared with those who received FOLFOX-4 alone (three of 73 patients; 4.1%). In contrast, in patients whose tumors carried KRASmutations, R0 resection rates were similar in both treatment arms (one in 52 and one in 47 patients, respectively; 1.9% and 2.1%). However, due to the small patient numbers, these data have to be interpreted with caution.

The risk of disease progression (Figs 2A and 2B) was similar for both the ITT (cetuximab plus FOLFOX-4 v FOLFOX-4 alone; hazard ratio [HR] = 0.93; log-rank test, P = .6170) and the KRAS population (HR = 0.93; log-rank test, P = .6609), adding additional evidence consistent with the representability of the KRASpopulation. A marked risk reduction for progression was evident between the study arms when comparing patients according to whether KRAS mutations were detected in their tumor DNA. Specifically, as summarized in Table 2 and Figure 3, the addition of cetuximab to FOLFOX-4 was associated with a 43% reduction in the risk of progression in patients with KRAS wild-type tumors (Fig 3C; HR = 0.57; log-rank test,P = .0163). In those patients receiving cetuximab plus FOLFOX-4, the risk of disease progression was lower for those whose tumors were wild type for KRAS, compared with those whose tumors carried mutations (Fig 3A; HR = 0.45; log-rank test, P = .0009). However, in patients whose tumors carried mutations of KRAS, the risk of disease progression was found to be lower for those receiving FOLFOX-4 alone compared with those receiving cetuximab plus FOLFOX-4 (Fig 3D; HR = 1.83; log-rank test, P = .0192). For patients receiving FOLFOX-4 alone, a trend for reduced risk of disease progression associated with tumor KRAS mutation was also apparent (Fig 3B; HR = 1.40; log-rank test, P = .1655).

Fig 3.

Fig 3.

Kaplan and Meier plots for progression-free survival time according to tumorKRAS mutation status in patients (pts) who received (A) cetuximab (Cet.) plus oxaliplatin, leucovorin, and fluorouracil (FOLFOX-4) or (B) FOLFOX-4 alone and according to treatment arm for pts whose tumors were (C) wild type or (D) mutant for KRAS. Progr., progression; Cens., censored; HR, hazard ratio.


The most common AEs at any grade were skin and subcutaneous tissue disorders (90%) and GI disorders (78%) in the cetuximab plus FOLFOX-4 group, and GI disorders (73%) and myelotoxicity (71%) in the FOLFOX-4–alone group. The most common grade 3/4 AEs were neutropenia (30% v 34%), rash (11% v 0.6%), and diarrhea (8% v 7%) respectively, for the cetuximab plus FOLFOX-4 versus the FOLFOX-4– alone group (Table 3).

Relevant Grade 3 or 4 Adverse Events Occurring in ≥ 3% of Patients in Either Group

In the cetuximab plus FOLFOX-4 group, cetuximab was discontinued in 39 patients (23%), chemotherapy was discontinued in 51 patients (30%), and both were discontinued in 16 patients (9%) due to AEs. Chemotherapy was discontinued in 42 patients (25%) in the FOLFOX-4–alone group due to AEs. The most common reasons for cetuximab discontinuation were hypersensitivity reactions (4%) and rash (4%), in accordance with guidelines prespecified in the protocol, and for chemotherapy discontinuation in the cetuximab plus FOLFOX-4 group, neutropenia (4%) and neuropathy (3%). In the FOLFOX-4–alone group, peripheral sensory neuropathy (5%), neuropathy and paresthesia (4%), and neutropenia (2%) were the most common reasons for treatment discontinuation.

At the time of analysis, 111 patients had died; 60 patients (35%) in the cetuximab plus FOLFOX-4 group and 51 patients (30%) in the FOLFOX-4–alone group, mainly due to PD (94 patients). No deaths were related to cetuximab.


The rationale for testing cetuximab in combination with FOLFOX-4 rested on preclinical data suggesting that cetuximab improved the antitumor activity of oxaliplatin and also on promising clinical phase II study data.16,30 The confirmed best overall response rate for the ITT population, as assessed by blinded independent review, was increased by 10 percentage points by the addition of cetuximab to FOLFOX-4, compared with FOLFOX-4 alone (46% v 36%). Although this difference showed a clinically relevant and strong trend, a statistically significant increase in the odds for a response with the addition of cetuximab to FOLFOX-4 could not be established (P = .064).

Although there was no apparent difference in PFS time between the ITT population treatment arms, when retrospective efficacy analyses were carried out in the KRASpopulation according to KRAS mutation status, striking differences were apparent in relation to tumor response and the risk of disease progression. Patients whose tumors were wild type for KRAS had a clinically relevant increase in the chance of response (61% v 37%; OR = 2.54; P = .011) and decrease in the risk of disease progression (HR = 0.57; P = .016) in the cetuximab plus FOLFOX-4 arm compared with those in the FOLFOX-4–alone arm. Somewhat unexpected was the observation that patients whose tumors carried a mutation in KRAS were more likely to derive a clinical benefit if treated with FOLFOX-4 alone rather than cetuximab plus FOLFOX-4 (Fig 3D). Similarly, considering patients receiving FOLFOX-4 alone, there was a trend for improved progression-free survival time in patients whose tumors carried a KRASmutation (Fig 3B).

It remains to be seen whether the apparently better outcome of patients whose tumors carried KRAS mutations who were treated with FOLFOX-4 alone compared with those treated with cetuximab plus FOLFOX-4 was a chance finding, or whether it might have a biologic explanation. Given that a comparable effect was not observed in equivalent patient groups in the CRYSTAL study when cetuximab was added to FOLFIRI,23 the type of chemotherapy (platinum based compared with nonplatinum based) may be a factor in any possible interaction. It is interesting to note in this context that the subgroup of genotyped patients in the randomized phase III CAIRO2 study whose tumors had a mutation in the KRAS gene (n = 196) who received capecitabine/oxaliplatin plus bevacizumab (n = 103) had significantly better PFS than those who received the same regimen plus cetuximab (n = 93).31

The OPUS and CAIRO2 studies therefore raise the possibility that in relation to tumors with mutations in the KRAS gene, the addition of EGFR-targeted agents to oxaliplatin-based chemotherapy may impair the efficacy of a cytotoxic component of the combined regimen. In addition, a specific interaction between bevacizumab and cetuximab in relation to the treatment of tumors with KRAS mutations cannot be excluded. Experimental studies are therefore warranted to explore such hypothetical treatment effects in KRAS mutant mCRCs with a view to devising novel clinical strategies to allow extension of the benefit of anti-EGFR therapy to patients whose tumors have KRAS mutations.

In line with recent findings from a phase I dose-escalation study of cetuximab with FOLFIRI,32 and the first-line data from the phase III CRYSTAL study,23 the results of the current analysis confirm published data from previously treated mCRC, regarding restriction of the activity of cetuximab to patients whose tumors are wild type forKRAS.19–22,24 The KRAS data also emphasize the fact that the higher clinical benefit associated with cetuximab essentially may be masked, or at the least underestimated, when considering overall populations of unselected patients in randomized studies, as illustrated by the nonsignificant results in response rate and the comparable risk of disease progression in the ITT population.

The combination of cetuximab and FOLFOX-4 was generally well tolerated, and there was no evidence to suggest that cetuximab increased the frequency or severity of the known toxicities of oxaliplatin and FU or vice versa. In addition, the incidence and severity of skin reactions, infusion-related reactions, and mucositis were consistent with the known safety profile of cetuximab.

In conclusion, this study demonstrated that to a clinically meaningful extent, the addition of cetuximab to FOLFOX-4 as first-line treatment for mCRC increased the overall response rate and reduced the risk of disease progression in patients whose tumors were wild type for KRAS, compared with FOLFOX-4 alone. These data confirm that KRAS mutation status is a powerful predictive factor in relation to the efficacy of cetuximab treatment, and are a clear demonstration of the potential and increasing importance of tumor biomarker analysis as a component of treatment selection in the clinic.


Although all authors completed the disclosure declaration, the following author(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a “U” are those for which no compensation was received; those relationships marked with a “C” were compensated. For a detailed description of the disclosure categories, or for more information about ASCO’s conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.

Employment or Leadership Position: Christopher Stroh, Merck Serono (C); Anja H. Loos, Merck Serono (C); Angela Zubel, Merck Serono (C)

Consultant or Advisory Role: Carsten Bokemeyer, Merck Serono (C)

Stock Ownership: None Honoraria: Carsten Bokemeyer, Merck Serono, Sanofi-aventis; Joerg T. Hartmann, Merck Serono Research Funding: Carsten Bokemeyer, Merck Serono; Gunter Schuch, Merck Serono; Christopher Stroh, Merck SeronoExpert Testimony: None Other Remuneration: Serban Donea, Merck Serono, Sanofi-aventis


Conception and design: Carsten Bokemeyer, Angela Zubel

Administrative support: Joerg T. Hartmann, Anja H. Loos

Provision of study materials or patients: Carsten Bokemeyer, Igor Bondarenko, Anatoly Makhson, Joerg T. Hartmann, Jorge Aparicio, Filippo de Braud, Serban Donea, Heinz Ludwig, Gunter Schuch, Piotr Koralewski

Collection and assembly of data: Igor Bondarenko, Anatoly Makhson, Joerg T. Hartmann, Jorge Aparicio, Christopher Stroh, Piotr Koralewski

Data analysis and interpretation: Carsten Bokemeyer, Anatoly Makhson, Joerg T. Hartmann, Filippo de Braud, Christopher Stroh, Anja H. Loos

Manuscript writing: Carsten Bokemeyer, Joerg T. Hartmann, Christopher Stroh, Anja H. Loos

Final approval of manuscript: Carsten Bokemeyer, Igor Bondarenko, Anatoly Makhson, Joerg T. Hartmann, Jorge Aparicio, Filippo de Braud, Serban Donea, Heinz Ludwig, Gunter Schuch, Christopher Stroh, Anja H. Loos, Angela Zubel, Piotr Koralewski


We thank the patients enrolled, the medical staff who contributed to their care, and the study team at Merck KGaA, particularly Esther Prang, Barbara Kraus, and Ilhan Celik. We also like thank the investigators at the 87 centers who recruited patients for EGFR expression screening (detailed in the Appendix).

The Appendix is included in the full-text version of this article, available online atwww.jco.org. It is not included in the PDF version (via Adobe® Reader®).

Participating Centers

Investigators at 87 centers in 13 countries enrolled patients onto the initial epidermal growth factor receptor screening phase of the study. Austria: Richard Greil (St Johanns Spital-Landeskrankenhaus, Salzburg), Heinz Ludwig, (Wilhelminenspital der Stadt Wien, Wien), Dieter Lutz (A. ö. KH der Elisabethinen Linz, Linz), Hellmut Samonigg (Medizinische Universitätsklinik, Graz), Johannes Schüller (KH der Stadt Wien Rudolfstiftung, Wien), Ewald Wöll (Krankenhaus St Vinzenz Zams, Zams); Belgium: Filip Baert (Heilig Hartziekenhuis, Roeselare), Alain Bols (AZ Sint-Jan, Brugge), Geert D’Haens (Imeldaziekenhuis, Bonheiden), Jos Janssens (Sint-Elisabethziekenhuis, Turnhout), Eric Joosens (AZ Middelheim, Antwerpen), Joseph Kerger (UCL Mont Godinne, Yvoir), Jeroen Mebis (Virga Jesse Ziekenhuis, Hasselt), Marc Peeters (UZ Gent, Gent), Eric Van Cutsem (UZ Gasthuisberg, Leuven); France: Dominique Béchade (Hôpital d’Instruction des Armées du Val de Grâce, Paris), Christophe Borg (Hôpital Jean Minjos, Besancon Cedex), Marian Gil Delgado (Groupe Hospitalier Pitié-Salpétrière, Paris Cedex 13), Jean Louis Misset (Hôpital Saint Louis, Paris), Eric Raymond (Hôpital Beaujon, Clichy), Jean-Philippe Wagner (Clinique Sainte Anne, Strasbourg), Marc Ychou (C.R.L.C. Val d’Aurelle-Paul Lamarque, Montpellier Cedex 5); Germany: Carsten Bokemeyer (Universitätsklinikum Hamburg-Eppendorf, Hamburg-Eppendorf), Wolfgang Fischbach (Klinikum Aschaffenburg Medizinische Klinik, Aschaffenburg), Gunnar Folprecht (Universitätsklinikum Dresden, Dresden), Frank Gieseler (Universitätsklinikum Schleswig-Holstein, Kiel), Joerg T. Hartmann (Universitätsklinikum Tübingen, Tübingen), Andreas Hochhaus (Universitätsklinikum Mannheim, Mannheim), Hendrik Kröning (Gemeinschaftspraxis, Magdeburg), Rainer Lipp (Innovation Onkologie Research & Consulting GmbH, Hamburg), Tanja Trarbach (Universitätsklinikum der GHS-Essen, Essen), Karin Weigang-Köhler (Klinikum Nürnberg Nord, Nürnberg); Greece: George Fountzilas (General Hospital Papageorgiou of Thessaloniki, Thessaloniki); Israel: Dan Aderka (Oncology Institute, The Chaim Sheba M.C, Tel-Hashomer), Alexander Beny (Rambam Medical Center, Haifa), Baruch Brenner (Rabin Medical Center, Petah Tiqva), Arie Figer (Tel-Aviv Sourasky Medical Center, Tel-Aviv), Baruch Klein (Oncology Institute, Meir Hospital, Sapir M.C., Kfar-Saba); Italy: Enrico Cortesi (Università degli Studi di Roma La Sapienza, Roma), Maria Giulia Zampino (Istituto Europeo di Oncologia, Milano), Luigi Dogliotti (SCDU Oncologia Medica-Azienda Sanitaria Ospedaliera San Luigi di Orbassano, Orbassano), Margherita Girino (IRCCS Policlinico S. Matteo, Pavia), Alberto Zaniboni (Fondazione Poliambulanza-Istituto Ospedaliero, Brescia); Poland: Kazimierz Drosik (Clinical Hospital, Opole), Malgorzata Foszczynska-Kloda (Regionalny Szpital Onkologiczny, Szczecin), Piotr Koralewski (Wojewódzki Szpital Specjalistyczny Oddział Chemioterapii, Krakow), Marek Nowacki (Klinika Nowotworów Przewodu Pokarmowego Centrum Onkologii Instytut im. Marii Sklodowskiej Curie, Warszawa), Elzbieta Staroslawska (Centrum Onkologii Ziemii Lubelskiej Oddział Chemioterapii, Lublin), Cezary Szczylik (Klinika Onkologii CSK WAM, Warszawa), Marek Wojtukiewicz (Białostockie Centrum Onkologii Oddział Onkologii Klinicznej, Bialystok), Jerzy Zaluski (Wielkopolskie Centrum Onkologii, Oddzial Chemioterapii, Poznan); Portugal: Paulo Cortes (Hospital de Santa Maria, Lisboa), Francisco Pimentel (Hospital de São Sebastião, Santa Maria da Feira); Romania: Cristina Cebotaru (Institutul Oncologic “Ion Chiricuta,” Cluj-Napoca), Mihai Ciochinaru (Spitalul Clinic de Urgenta Militar Central “Dr. Carol Davila,” Bucharest), Serban Donea (Institut Oncologic Al. Trestioreanu Bucuresti, Bucharest), Iuliu Ionas (Spitalul Municipal Onesti, Onesti), Lucia Milian (Spitalul Clinic Judetean Oradea, Oradea), Anca Niculescu (Cabinetul de Oncologie Medicala “Oncomed,” Timisoara), Ozana Radu (Spitalul Judetean Alba, Alba Iulia), Constantin Volovat (Centrul de Oncologie Medicala, Iasi); Russian Federation: Lilya Akhmadulina (KLPU of City Oncology Dispensary, Kazan), Boris Berdov (Medical Radiological Research Center, Obninsk), Vasily Borisov (Moscow City Oncology Dispensary, Moscow), Rustem Khasanov (Clinical Oncology Dispensary, Kazan), Mikhail Kopp (Samara Oncology Center, Samara), Anatoly Makhson (Moscow City Oncology Hospital # 62, Moscow Area), Irina Mitashok (Krasnodar Regional Oncology Dispensary, Krasnodar), Yury Shelygin (State Scientific Center of Coloproctology, Moscow), Escender Topuzov (St. Petersburg Mechnikov State Medical Academy. Surgical Disease Clinic #1, St Petersburg); Spain: Rocío García-Carbonero (Hospital Severo Ochoa, Madrid), Carlos García Girón (Hospital General Yagüe, Burgos), Ignacio Juez Martel (Hospital Josep Trueta (ICO Girona), Gerona), Guillermo López-Vivanco (Hospital de Cruces, Bilbao), Mercedes Salgado Fernández (Complejo Hospitalario Ourensano (Hospital Santa María Nai), Ourense), Antonio Sánchez (Hospital Universitario Puerto de Hierro, Madrid), Isabel Sevilla (Hospital Virgen de la Victoria, Málaga), Jorge Aparicio Urtasún (Hospital Universitario La Fé, Valencia), Vincent Valenti (Hospital Universitari de Sant Joan, Tarragona); Ukraine: Borys Bolyukh (Vinnitsa Regional Oncology Center, Vinnitsa State Medical University, Vinnitsa), Igor Bondarenko (Dnipropetrovsk State Medical Academy, City Clinical Hospital #4, Dnepropetrovsk), Tetyana Danylova (Crimean Republican Oncology Center, Simferopol), Oleksandr Dudnichenko (Kharkov Regional Oncology Center, Kharkov Academy of Postgraduate Education, Kharkov), Olena Kolesnik (Institute of Oncology, Academy of Medical Science of Ukraine, Kyiv), Oksana Tarasova (Institute of Medical Radiology AMS Ukraine, Kharkov), Igor Shchepotin (Kyiv City Oncological Center, Kyiv), Yaroslav Shparyk (Lviv Oncological Medical and Diagnostic Center, Lviv).

Methods: KRAS Mutation Analysis


Formaldehyde-fixed, paraffin-embedded (FFPE) tumor biopsy samples, as paraffin blocks or as 3- to 5-μm paraffin sections mounted on glass slides, were sent to a central facility (Merck Serono, Darmstadt, Germany) for KRAS mutation analysis. Serial sections were cut from each block for pathologic analysis (3 μm) and DNA extraction (10 μm, stored at −80°C pending isolation). A pathologist confirmed the presence of invasive tumor cells in the tissue through examination of hematoxylin and eosin–stained sections. If invasive tumor cells could not be confirmed in a particular sample, the biopsy was excluded from further analysis.

DNA extraction.

For FFPE tissue blocks, DNA was extracted from three to six 10-μm sections (depending on biopsy size). If tumor samples were available only as sections mounted on glass slides, the complete sample was scraped off and transferred into a 1.5-mL tube using a fresh scalpel for each slide, to avoid cross contamination.

Paraffin was dissolved by adding 800 μL xylene to each sample followed by gentle mixing. Subsequently, 400 μL ethanol (100% v/v) was added to each tube. The mixture was centrifuged for 2 minutes at 13,000 rpm in a microcentrifuge and the supernatant was carefully removed and discarded. If any remaining paraffin was visible, the xylene and ethanol extraction was repeated. The pellet was washed with 1,000 μL of ethanol (100% v/v) and the tubes were centrifuged for 2 minutes at 13,000 rpm. The supernatant was again carefully removed and the pellet containing the DNA dried for 10 minutes at 55°C. Genomic DNA was isolated from the dried pellet using the DNeasy Blood & Tissue Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. The concentration of each extracted DNA sample was determined by measuring the absorption at 260 nm using a Nanodrop UV/VIS spectrometer (Thermo Fisher Scientific, Wilmington, DE). Purity was assessed by determination of the 260/280 nm ratio. The DNA quality was further assessed in aKRAS-independent quantitative polymerase chain reaction (qPCR) amplification. Genomic DNA from cell lines was also isolated using the DNeasy Blood & Tissue Kit, according to the manufacturer’s instructions and quantified and quality controlled as for the FFPE material.

Methods: KRAS Mutation Detection by Locked Nucleic Acid–Mediated qPCR Clamping and Melting Curve Analysis

Locked nucleic acid–mediated qPCR clamping.

PCR reactions were performed on a LightCycler 2.0 system, using a KRAS mutation detection–specific program with additional melting curve analysis, in a one-step qPCR amplification and detection run using the LightMix k-ras Gly12 assay (TIB MOLBIOL, Berlin, Germany). The methodology was adapted from Chen et al.28 PCR reactions incorporated: a KRAS-specific primer pair (KRAS forward, KRAS reverse), which amplified a 164 base pair genomic fragment from KRAS exon 2; two hybridization sensor probes covering the mutation hotspot region around codons 12 and 13; and an anchor probe (Anchor FL), which was 3′-end labeled with fluorescein. The sensor probes were 5′-end labeled with LC Red dye and 3′-end phosphorylated to prevent extension. Amplification of KRAS wild-type DNA in the PCR reaction was inhibited by temperature stable binding of a KRAS wild-type specific locked nucleic acid sequence (2′-4′ bridged ribose backbone).

The mutation analysis was performed in duplicate for each tumor sample. The PCR reaction was performed in a total volume of 20 μL containing: 7.4 μL PCR-grade water, 1.6 μL 25 mM MgCl2, 4.0 μL of reagent/primer mix (parameter specific reagents containing both PCR primers, one anchor, two sensor-probes, and the LNA), 2.0 μL of qPCR Mastermix (Roche Diagnostics, Penzberg, Germany), and 5 μL of the genomic DNA solution (50 ng). PCR reactions incorporating a 1:10 mixture of DNA (1 ng) derived from cell lines with defined KRAS mutations and DNA (10 ng) from cell lines with wild-type KRAS were included as positive controls in each run. PCR reactions with 10 ng of DNA derived from cells with wild-type KRAS were performed as negative controls in each run. The qPCR conditions comprised a precycling incubation of 600 seconds at 95°C followed by 45 cycles of 95°C for 10 seconds, 58°C for 5 seconds, and 72°C for 12 seconds.

Melting curve analysis.

After the qPCR run was complete, a melting curve analysis was performed by measuring the fluorescence of the samples with increasing temperature. Initially, the PCR products were denatured at 95°C. Subsequent cooling of the reaction mix allowed the anchor probe and the mutation-specific sensor probes to anneal to the amplified genomic KRAS sequence, resulting in increased fluorescence due to fluorescence resonance energy transfer between annealed anchor and sensor probe dyes. The melting curve was assessed by heating the sample from 40°C to 85°C, with continuous monitoring of fluorescence in the 640/530-nm channel. At a certain temperature, depending on the identity of the KRAS mutation, binding of the sensor probe to the DNA strand was disrupted. This resulted in spatial separation of the fluorescence resonance energy transfer partners, and as a consequence, in a decrease of fluorescence. The change of fluorescence intensity was converted to a melting peak (Tm) by plotting the negative derivative of fluorescent signal corresponding to the temperature (−dF/dT). The melting curve analysis program therefore comprised a preincubation of 20 seconds at 95°C, followed by 20 seconds at 58°C, followed by 10 seconds at 40°C, followed by heating to 85°C at 0.3°C/second. Different individual KRAS mutations displayed different characteristicTm values. Each could be identified by comparing the Tm of the experimental sample with the Tm values of a panel of positive controls.