|
|
Background Treatment with cetuximab, a monoclonal antibody directed against the epidermal growth factor receptor, improves overall and progression-free survival and preserves the quality of life in patients with colorectal cancer that has not responded to chemotherapy. The mutation status of the K-ras gene in the tumor may affect the response to cetuximab and have treatment-independent prognostic value.
Methods We analyzed tumor samples, obtained from 394 of 572 patients (68.9%) with colorectal cancer who were randomly assigned to receive cetuximab plus best supportive care or best supportive care alone, to look for activating mutations in exon 2 of the K-ras gene. We assessed whether the mutation status of the K-ras gene was associated with survival in the cetuximab and supportive-care groups.
Results Of the tumors evaluated for K-ras mutations, 42.3% had at least one mutation in exon 2 of the gene. The effectiveness of cetuximab was significantly associated with K-ras mutation status (P=0.01 and P<0.001 for the interaction of K-ras mutation status with overall survival and progression-free survival, respectively). In patients with wild-type K-ras tumors, treatment with cetuximab as compared with supportive care alone significantly improved overall survival (median, 9.5 vs. 4.8 months; hazard ratio for death, 0.55; 95% confidence interval [CI], 0.41 to 0.74; P<0.001) and progression-free survival (median, 3.7 months vs. 1.9 months; hazard ratio for progression or death, 0.40; 95% CI, 0.30 to 0.54; P<0.001). Among patients with mutated K-ras tumors, there was no significant difference between those who were treated with cetuximab and those who received supportive care alone with respect to overall survival (hazard ratio, 0.98; P=0.89) or progression-free survival (hazard ratio, 0.99; P=0.96). In the group of patients receiving best supportive care alone, the mutation status of the K-ras gene was not significantly associated with overall survival (hazard ratio for death, 1.01; P=0.97).
Conclusions Patients with a colorectal tumor bearing mutated K-ras did not benefit from cetuximab, whereas patients with a tumor bearing wild-type K-ras did benefit from cetuximab. The mutation status of the K-ras gene had no influence on survival among patients treated with best supportive care alone. (ClinicalTrials.gov number, NCT00079066
[ClinicalTrials.gov]
.)
K-ras, a small G-protein downstream of EGFR and an essential component of the EGFR signaling cascade, can acquire activating mutations in exon 2, thus isolating the pathway from the effect of EGFR2 and rendering EGFR inhibitors ineffective.3,4,5,6,7,8,9,10,11
We undertook correlative analyses to determine whether the mutation status of the K-ras gene modified the effect of cetuximab on overall survival and progression-free survival in the CO.17 trial. We also assessed the association of K-ras mutation status with overall survival and progression-free survival among patients receiving best supportive care alone.
Methods
A protocol committee that included members of the NCIC CTG and the AGITG designed the study. The NCIC CTG collected and analyzed the data and maintained full, unrestricted rights to submit the study data for publication. One of the academic authors reviewed all the data, another conducted the statistical analyses, and a third wrote the first draft of the manuscript; these authors vouch for the completeness and accuracy of the data presented. The initial draft of the manuscript was reviewed and commented on by all the authors as well as by employees of Bristol-Myers Squibb, who did not make substantive changes. The relevant institutional review boards approved the study protocol, including the use of archived tumor tissue. Written informed consent for tumor tissue research was obtained from the majority of patients. For use of tissue from those in whom a clinical decline prevented consent, approval was granted by the ethics board at the patient's institution on the basis of the patient's consent to the original research.
Patients and Trial Design
The trial design and eligibility criteria have been reported previously.1 The primary objective of the phase 3 CO.17 study was to examine the effect of cetuximab on survival among patients with advanced colorectal cancer in whom all chemotherapy for colorectal cancer had failed and for whom no other standard anticancer therapy was available. Eligible patients were enrolled in the trial between December 2003 and August 2005. None of the patients had received previous therapy directed against EGFR. After enrollment, patients were randomly assigned to receive cetuximab plus supportive care or supportive care alone. Cetuximab was given as an intravenous loading dose of 400 mg per square meter of body-surface area, administered over a period of 120 minutes, on day 1 of treatment, followed by an infusion of 250 mg per square meter, administered over a period of 60 minutes, once a week. Patients in both groups were evaluated for tumor response or progression every 8 weeks by means of radiologic imaging. Cetuximab therapy was continued until the disease progressed or until the patient could not tolerate the toxic effects.
Tumor Collection and Processing
Formalin-fixed, paraffin-embedded samples of tumor tissue from archival specimens collected at the time of diagnosis were stored at a central tumor bank located at Queen's University in Kingston, Ontario, Canada. If tumor blocks were unavailable, unstained slides were retrieved. Assays of tissue samples for K-ras mutations were performed in a blinded fashion by members of the Department of Clinical Biomarkers–Oncology at Bristol-Myers Squibb, Hopewell, New Jersey. All available tissue samples were classified as having mutated or wild-type K-ras.
Nucleotide Sequence Analysis
Mutation analysis of K-ras was performed by extraction of genomic DNA from formalin-fixed, paraffin-embedded tissue slides or sections with the use of the QIAamp DNA Mini Kit (Qiagen). For K-ras analyses, the following nested primer sets for exon 2 were used: huKRAS2 ex2F, 5'GAATGGTCCTGCACCAGTAA3'; huKRAS2 ex2R, 5'GTGTGACATGTTCTAATATAGTCA3'; huKRAS ex2Fint, 5'GTCCTGCACCAGTAATATGC3'; and huKRAS2 ex2Rint, 5'ATGTTCTAATATAGTCACATTTTC3'. Each 25-µl polymerase-chain-reaction (PCR) cocktail contained at least 15 ng of genomic DNA, 2.5 mM magnesium chloride, 300 mM deoxynucleotide triphosphates, and 2.5 U of AmpliTaq Gold DNA Polymerase (Applied Biosystems). PCR conditions were as follows: 1 cycle at 94°C for 10 minutes; 45 cycles at 94°C for 30 seconds, 64°C for 30 seconds, and 72°C for 30 seconds; followed by 1 cycle at 72°C for 7 minutes. Primer extension sequencing was performed with the use of the BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems). Reactions were run on a 3730x1 DNA Analyzer (Applied Biosystems). Analyses of the DNA sequences were performed with the use of Mutation Surveyor v2.61 (SoftGenetics), along with visual inspection of each sample trace in both forward and reverse directions. Appropriate positive and negative controls were included for K-ras. The persons who performed the mutation analyses had no knowledge of the clinical outcome.
Statistical Analysis
All statistical analyses were performed by the NCIC CTG in accordance with a protocol for statistical analysis that was written before the assessment of K-ras mutation was performed. All randomly assigned patients for whom data on K-ras mutation status were available were included in the analysis. Overall survival, the primary end point, was defined as the time from randomization until death from any cause. The secondary end points were progression-free survival, defined as the time from randomization until the first objective evidence of disease progression or death from any cause; response rates, defined according to the Response Evaluation Criteria in Solid Tumors (RECIST); and quality of life, assessed with the use of the European Organisation for Research and Treatment of Cancer (EORTC) quality-of-life questionnaire (QLQ-C30). Wilcoxon tests were used to compare the two treatment groups with respect to the mean change from baseline in scores on the global quality-of-life scale. A difference of more than 10 points was considered to indicate clinical significance.12
The survival of patients in each K-ras–mutation group and treatment group was summarized with the use of Kaplan–Meier curves, and the difference between these groups was compared with the use of the log-rank test, with the hazard ratio and its 95% confidence intervals calculated from a Cox regression model with a single covariate. To assess whether K-ras was an independent prognostic factor for patients receiving supportive care, a multivariate Cox regression model was fitted to data for patients receiving supportive care alone; it included the following covariates specified by the protocol: Eastern Cooperative Oncology Group (ECOG) performance status (0 or 1 vs. 2, on a scale of 0 to 5, with lower scores representing a higher level of functioning), sex (male vs. female), age (
65 years vs. <65 years), baseline lactate dehydrogenase level (higher than the upper limit of the normal range vs. the upper limit or lower), baseline alkaline phosphatase level (higher than the upper limit of the normal range vs. the upper limit or lower), baseline hemoglobin (lower than the lower limit of the normal range vs. the lower limit or higher), number of disease sites (>2 vs. 
2), number of previous chemotherapy drug classes (>2 vs. 2), primary tumor site (rectum only vs. colon), and presence of liver metastases (yes vs. no). We used the Cox model, with treatment, K-ras–mutation status, and their interaction as covariates, to assess the interaction between treatment and K-ras–mutation status. All reported P values are two-sided and were not adjusted for multiple testing.
Results
Characteristics of the Patients
Of 572 patients who underwent randomization, 287 were assigned to receive cetuximab plus supportive care, and 285 were assigned to receive supportive care alone. A total of 394 tumor specimens (198 from the cetuximab group and 196 from the supportive-care group) were examined for K-ras–mutation status (accounting for 68.9% of the total study population). Tumor specimens from the remaining patients could not be retrieved despite our best efforts; the most common reasons were lack of patient consent, insufficient tissue, and refusal or inability of the laboratory of origin to release the tissue for research. A K-ras mutation was detected in 40.9% of tumor specimens in the cetuximab group and in 42.3% of tumor specimens in the supportive-care group. Table 1 shows the K-ras mutations we identified and the distribution according to the treatment group. Table 2 summarizes the baseline demographic and disease characteristics of the total study population and of the patients who were evaluated for the K-ras mutation. The group of patients with K-ras mutations and the group with wild-type K-ras were similar with respect to baseline characteristics, including ECOG performance status and the other variables associated with survival in the multivariate analysis. The distribution of these characteristics among the patients who were evaluated for K-ras mutations was similar to that in the total study population. In the supportive-care group, 13 patients had protocol violations and crossed over to cetuximab treatment. Four of them received cetuximab before progression of the disease, and nine received cetuximab after progression. All of the patients who had been assigned to the cetuximab group and who were included in the K-ras–mutation analysis did receive cetuximab.
|
|
Among patients with mutated K-ras tumors, the median overall survival was 4.5 months in the group receiving cetuximab and 4.6 months in the group receiving supportive care only, with 1-year overall survival rates of 13.2% and 19.6%, respectively (hazard ratio for death in the cetuximab group as compared with the supportive-care group, 0.98; 95% confidence interval [CI], 0.70 to 1.37; P=0.89) (Figure 1A). Among patients with wild-type K-ras tumors, the median overall survival was 9.5 months in the cetuximab group as compared with 4.8 months in the supportive-care group, with 1-year overall survival rates of 28.3% and 20.1%, respectively (hazard ratio, 0.55; 95% CI, 0.41 to 0.74; P<0.001). In a multivariate Cox regression model, this difference remained significant after adjustments for potential prognostic factors specified in the protocol (hazard ratio, 0.62; 95% CI, 0.44 to 0.87; P=0.006) (Figure 1B). The effect of cetuximab on overall survival was significantly greater among the patients with wild-type K-ras tumors than among those with mutated K-ras tumors (P=0.01 for the interaction between K-ras–mutation status and the assigned treatment).
|
Among patients with mutated K-ras tumors, the median progression-free survival was 1.8 months in both the cetuximab and supportive-care groups (hazard ratio for progression or death in the cetuximab group as compared with the supportive-care group, 0.99; 95% CI, 0.73 to 1.35; P=0.96) (Figure 2A). The precipitous drop in progression-free survival at 8 weeks was consistent with the tumor-imaging schedule mandated by the protocol. For patients with wild-type K-ras tumors, the median progression-free survival was 3.7 months in the cetuximab group and 1.9 months in the supportive-care group (hazard ratio for progression or death in the cetuximab group as compared with the supportive-care group, 0.40; 95% CI, 0.30 to 0.54; P<0.001) (Figure 2B). This difference remained significant after adjustment for potential prognostic factors specified in the protocol (adjusted hazard ratio, 0.42; 95% CI, 0.30 to 0.58; P<0.001). The effect of cetuximab on progression-free survival was significantly greater among the patients with wild-type K-ras tumors than among those with mutated K-ras tumors (P<0.001 for the interaction between K-ras–mutation status and the assigned treatment).
|
None of the patients in the supportive-care group had an objective tumor response. In the cetuximab group, the response rate among patients with wild-type K-ras tumors was 12.8%, whereas only one patient with a mutated K-ras tumor (1.2%) had a response. Significantly more patients with wild-type K-ras tumors than patients with mutated K-ras tumors had a rash during cetuximab treatment or within 30 days after completion of the treatment (94.9% vs. 84.0%, P=0.02).
Quality of Life
Among patients with wild-type K-ras tumors, those in the cetuximab group had an improvement in global health status at 8 weeks, whereas those in the supportive-care group had a deterioration in global health status (mean change in score, 3.2 vs. –7.7 points; difference, 10.9; 95% CI, 4.2 to 17.6; P=0.002). The patients in the cetuximab group also had less deterioration at 16 weeks than did those in the supportive-care group (mean change in score, –0.2 vs. –18.1 points; difference, 17.9; 95% CI, 7.6 to 28.2; P<0.001). Among patients with mutated K-ras tumors, there was no significant difference in global health status between the cetuximab group and the supportive-care group at 8 weeks (mean change in score, –4.7 and –9.6 points, respectively; difference, 4.9; 95% CI, –4.2 to 14.0; P=0.53) or at 16 weeks (mean change in score, –9.5 and –13.9 points, respectively; difference, 4.4; 95% CI, –9.2 to 17.9; P=0.62).
Effect of Mutation Status in the Supportive-Care Group
In the supportive-care group, there was no significant difference in overall survival between patients with wild-type K-ras tumors and those with mutated K-ras tumors. As seen above, the median overall survival among the patients with mutated K-ras tumors was 4.6 months, as compared with 4.8 months among those with wild-type K-ras tumors, with 1-year overall survival rates of 19.6% and 20.1%, respectively (hazard ratio for death among patients with K-ras mutations, 1.01; 95% CI, 0.74 to 1.37; P=0.97) (Figure 3). The median progression-free survival was 1.8 months and 1.9 months for patients with mutated and wild-type K-ras tumors, respectively (hazard ratio for progression, 1.12; 95% CI, 0.84 to 1.49; P=0.46). The differences remained nonsignificant after adjustment for other factors specified by the protocol (adjusted hazard ratio for death, 1.33; 95% CI, 0.95 to 1.86; P=0.10; adjusted hazard ratio for progression, 1.24; 95% CI, 0.90 to 1.70; P=0.19).
|
Our findings show that the mutation status of the K-ras gene is associated with overall survival among patients with advanced colorectal cancer who are being treated with cetuximab after previous chemotherapy has failed. Treatment with cetuximab as compared with supportive care alone was associated with almost a doubling of the median overall and progression-free survival among patients with wild-type K-ras tumors. There was no significant survival benefit from cetuximab, however, among patients with tumors that had K-ras mutations. Several small, retrospective studies have shown an association between K-ras–mutation status and responsiveness of a colorectal tumor to cetuximab.5,6,7,13 Treatment with another monoclonal antibody directed against EGFR, panitumumab, has been compared with supportive care in a phase 3 trial.14 A K-ras analysis showed, as with our findings, that the benefit associated with panitumumab was limited to patients with wild-type K-ras tumors.3 The group with wild-type K-ras tumors had a progression-free survival benefit from treatment with panitumumab, although an overall survival benefit was not shown.
We also examined the treatment-independent effect of K-ras–mutation status. The clinical significance of K-ras mutations has varied in reported studies, most of which have been compromised by a small sample size. In the largest reported series, K-ras mutations, as compared with wild-type K-ras, were associated with a risk of death that was increased by 26%. In particular, a mutation in K-ras codon 12 with substitution of valine for glycine (G12V), which was found in 8.6% of all patients, had a significant effect on failure-free and overall survival.8,15 Mutations in K-ras, B-type Raf kinase (BRAF), or phosphatidylinositol 3-kinase (PI3K) genes were also significantly associated with shorter survival in another study, which involved 586 patients.16 Less favorable clinical outcomes, particularly shorter survival, have been associated with the presence of K-ras mutations at codon 13.11 No association with the outcome was found in a study of stage 3 colon cancer, but only 55 cases were examined.17 In the randomized trial of panitumumab therapy as compared with supportive care, multivariate analysis confirmed that wild-type K-ras was significantly associated with overall survival, but since most patients in the supportive-care group crossed over and received the monoclonal antibody when their disease progressed, the significance of K-ras–mutation status was ultimately uncertain.3
We examined the association of K-ras–mutation status with survival among patients in the supportive-care group, in whom no effect of cetuximab would be expected. Such an analysis represents the best method for evaluating the significance of a variable, since its potential predictive effect is not relevant. We did not observe a significant difference in survival according to K-ras–mutation status in the supportive-care group.
Although K-ras mutations may be a common genetic aberration involved in carcinogenesis, mutations of other genes, such as phosphatase and tensin homologue (PTEN), BRAF, or PI3K, that can lead to unrestricted growth of cancer cells may also be useful biomarkers. Loss of PTEN activity, for example, has been associated with lack of efficacy of cetuximab in 11 patients with colorectal cancer.7 Further upstream in the signaling pathway, high expression of EGFR ligands, particularly amphiregulin and epiregulin, has been observed in tumors that respond to cetuximab, and these ligands may be targets for new treatments.18 Immunohistochemical studies suggest that EGFR, the principal target of cetuximab, is not a useful predictive factor19,20,21,22; responses to cetuximab have been reported in patients with advanced colorectal cancers that do not have immunohistochemical evidence of EGFR expression.23,24 The problems with immunohistochemical assessment include inaccuracy of and variations in the procedure, heterogeneity of EGFR expression in the tumor, and variable affinity of EGFR for cetuximab.25,26 Assessment of the EGFR copy number may be more reliable, but the findings have been inconsistent.27,28,29 These discrepant results highlight the need for standardization of the measurement of EGFR expression and activity.
The inactivation of the effect of cetuximab by K-ras mutation is biologically plausible and is supported by previous studies. A lack of activity of therapy directed against EGFR, particular tyrosine kinase inhibitors, has also been shown in patients who have non–small-cell lung cancer with K-ras mutations.30,31,32 Our analysis was focused on K-ras and was based on a statistical plan that was specified before the assessment of K-ras status was performed. Evaluation of multiple correlative biomarkers was not performed.
Treatment with cetuximab, which is relatively expensive, would be most cost-effective if it were given to patients with the highest chance of benefiting from it. Our analysis identified a biomarker that would effectively exclude a clinically significant proportion of patients with colorectal cancer — those with tumors bearing K-ras mutations (42%) — from receipt of a therapy offering little prospect of a benefit. Nevertheless, there were also patients with wild-type K-ras tumors who did not have a response to cetuximab and in whom the tumor rapidly progressed. Additional reliable and easily measured biomarkers are clearly needed to improve the identification of patients who will benefit from treatment with cetuximab.
Supported by the National Cancer Institute of Canada, ImClone Systems, and Bristol-Myers Squibb.
Presented in part in abstract form at the World Congress on Gastrointestinal Cancer, Barcelona, June 25–28, 2008.
Drs. Karapetis and Shapiro report receiving consulting fees from Merck Serono; Drs. Jonker and Au, receiving consulting fees from Bristol-Myers Squibb; Drs. O'Callaghan and Tu, being employees of the National Cancer Institute of Canada Clinical Trials Group, which has received grant support from Bristol-Myers Squibb and Amgen Canada; Drs. Langer and Khambata-Ford, owning equity in and being employees of Bristol-Myers Squibb; Dr. Zalcberg, receiving lecture and consulting fees from Amgen and ImClone; and Drs. Simes and Zalcberg, receiving research grants from Amgen, Merck Serono, Bristol-Myers Squibb, and Alphapharm (Australia). No other potential conflict of interest relevant to this article was reported.
Source Information
From Flinders Medical Centre and Flinders University, Adelaide, Australia (C.S.K.); Bristol-Myers Squibb Research and Development, Princeton, NJ (S.K.-F.); Ottawa Hospital Research Institute, University of Ottawa, Ottawa (D.J.J.); National Cancer Institute of Canada Clinical Trials Group, Kingston, ON (C.J.O., D.T., S.R., L.S.); Austin Health, Melbourne, Australia (N.C.T.); National Health and Medical Research Council Clinical Trials Centre, University of Sydney, Sydney (R.J.S.); Allan Blair Cancer Centre, Regina, SK, Canada (H.C.); Cabrini Hospital and Alfred Hospital, Melbourne, Australia (J.D.S.); Queen Elizabeth Hospital and University of Adelaide, Adelaide, Australia (T.J.P.); Cross Cancer Institute, Edmonton, AB, Canada (H.-J.A.); Bristol-Myers Squibb, Wallingford, CT (C.L.); Princess Margaret Hospital, Toronto (M.J.M.); and Peter MacCallum Cancer Centre and University of Melbourne, Melbourne, Australia (J.R.Z.).
Address reprint requests to Dr. Karapetis at the Department of Medical Oncology, Flinders Medical Centre, Flinders Dr., Bedford Park, SA 5042, Australia, or at c.karapetis{at}flinders.edu.au.
References
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||
Related Letters:
K-ras Mutations and Cetuximab in Colorectal Cancer
Marchetti A., Gasparini G., Albitar M., Yeh C., Ma W., De Roock W., Lambrechts D., Tejpar S., Winder T., Scheithauer W., Lang A., Gattenlohner S., Germer C., Muller-Hermelink H.-K., Codacci-Pisanelli G., Spinelli G., Tomao S., Karapetis C., O'Callaghan C., Khambata-Ford S.
Extract |
Full Text |
PDF
N Engl J Med 2009;
360:833-836, Feb 19, 2009.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
