The authors have declared that no competing interests exist.
Contributed some part of additional potentially eligible literature: FL. Conceived and designed the experiments: XW. Performed the experiments: ZJ CL FL. Analyzed the data: ZJ CL. Contributed reagents/materials/analysis tools: ZJ CL FL. Wrote the paper: ZJ CL.
The epidermal growth factor receptor (EGFR) gene copy number (GCN) has been previously demonstrated to correlate with the clinical outcome of colorectal cancer (CRC) treated with anti-EGFR monoclonal antibodies (mAbs), although it remains controversial. We conducted a systematic review and meta-analysis to assess EGFR GCN as a potential biomarker of survival for patients with advanced CRC receiving treatment with anti-EGFR mAbs.
We systematically identified articles investigating EGFR GCN by fluorescent or chromogenic
Among 13 identified studies, 10 (776 patients, 302 with increased GCN), 8 (893 patients, 282 with increased GCN) and 3 (149 patients, 66 with increased GCN) were eligible for the OS, PFS and TTP meta-analyses, respectively. Increased EGFR GCN was associated with increased OS (HR = 0.62; 95% CI 0.50–0.77;
Among the anti-EGFR-treated patients, increased EGFR GCN appears to be associated with improved survival outcomes. The effect on survival appears to be related to patients receiving the line of treatment.
The major prognostic determinant for patients with non-resectable metastic colorectal cancer (CRC) is the response to systemic therapy
Several clinical studies have shown that the presence of a KRAS mutation is a significant predictor of resistance to anti-EGFR mAbs
Moroni et al.
We performed a systematic computerized search of the MEDLINE (PubMed) database, EMbase and the Cochrane library (last search: August 10, 2012) to identify all published articles related to the identification of mutations in EGFR pertaining to CRC, using the algorithm: (epidermal growth factor receptor OR EGFR) AND (mutation OR polymorphism OR gene copy number OR GCN OR amplification OR gene status) AND (colorectal cancer OR CRC). Additional studies were identified by a hand search of references of original studies or review articles on this topic. Eligible studies included in the meta-analysis had to meet the following criteria: (a) a cohort colorectal cancer study; (b) hazard ratios (HRs) with corresponding 95% confidence intervals (CIs) comparing overall survival (OS), progression-free survival (PFS) or time-to-progression (TTP) stratified by EGFR gene copy number for patients receiving mono- or combination therapy with either cetuximab or panitumumab were reported or allowed the calculation; and (c) written in English.
Two investigators independently extracted data and reached a consensus on all of the items. The following information was extracted from each study: first author, years of publication, number of patients screened, ethnicity of study population, gender, proportion of increased GCN, the specific methods of gene copy number determination were recorded, as were the values for GCN cutoff, KRAS status, anti-EGFR mAbs, study design and also for data linking specific mutation to treatment outcome. Also, we categorized studies by line of treatment. When studies were conducted in mixed treatment settings, we operationally defined studies where at least 80% of patients had received previous chemotherapy as ‘second-line’ studies. Finally, we extracted HRs and their variance for the relevant survival outcomes comparing patients with increased and normal EGFR gene copy number receiving treatment with either cetuximab or panitumumab. The HR is the most appropriate metric for time-to-event outcomes
We used the HR and corresponding CI extracted from each study to assess between-study heterogeneity using the Q statistics
Subgroup analyses were performed to evaluate the effect of ethnicity (East Asian versus white), method of EGFR GCN determination (FISH versus chromogenic
Statistical analyses were conducted with Stata (version SE/10; StataCorp, College Station, TX).
Our initial search yielded 76 studies concerning EGFR-targeted treatment in CRC, which were assessed in full text. As indicated in the search flow diagram (
Auther (year) reference | Patients (N) | Ethnicity | Gender(M/F) | Increased EGFR gene copy number, n (%) | Detection method | GCNcutoff | KRAS status | anti-EGFR mAbs | Line of treatment (proportion of second-line orhigher) | Study design | Outcome reported |
Sartore-Bianchi (2007) |
58 | White | 33/25 | 19(33) | FISH | 2.5 | NR | panitumumab | ≥80% | retrospective | OS,PFS |
Cappuzzo (2008) |
85 | White | 54/31 | 43(51) | FISH | 2.92 | NR | cetuximab | ≥80% | retrospective | OS,TTP |
Gonçalves (2008) |
20 | White | 10/10 | 8(40) | FISH | 2 | mixed | cetuximab | ≥80% | retrospective | TTP |
Personeni (2008) |
87 | White | 49/38 | 33(38) | FISH | 2.83 | mixed | cetuximab | ≥80% | prospective | OS,PFS |
Scartozzi (2009) |
44 | White | 23/21 | 15(40) | FISH/CISH | 2.6/2.12 | wild-type | cetuximab | ≥80% | retrospective | OS,TTP |
Laurent-Puig (2009) |
96 | White | NR | 17(22) | FISH/CISH | 2/NR | wild-type | cetuximab | ≥80% | retrospective | OS,PFS |
Li (2010) |
74 |
East Asian | 43/31 | 53(77) | FISH | 2 | mixed |
cetuximab | ≥80% | retrospective | OS,PFS |
Tol (2010) |
277 | White | NR | 41(15) | FISH | # | mixed | cetuximab | <80% | retrospective | PFS |
Campanella (2010) |
101 | White | 62/39 | 56(56) | FISH | 2 | mixed | cetuximab | <80% | retrospective | PFS |
Bengala (2010) |
146 | White | 96/60 | 29(20) | FISH | 2.9 | mixed | NR | <80% | retrospective | OS,PFS |
Scartozzi (2011) |
90 | White | 59/31 | 43(48) | CISH | 2.12 | wild-type | cetuximab | ≥80% | retrospective | OS |
Lin (2011) |
42 | East Asian | 19/23 | 16(38) | qPCR | 1.5 | mixed | cetuximab | ≥80% | retrospective | OS |
Algars (2011) |
54 | White | 30/24 | 34(63) | SISH | 4 | mixed |
mixed | ≥80% | retrospective | OS,PFS |
of 74 patients, EGFR FISH analysis was successfully detected in 69 of the tumor samples.
also provided information for the outcome in wild-type populations.
an increased GCN was defined as an average of three or more locus copies per nucleus, or a locus to centromere ratio of two or more.
M, male; F, female; FISH, fluorescent
Eight of the studies employed FISH, one employed CISH and two employed both methods, two employed silver
Regarding OS, 10 studies involving 776 patients (302 with increased gene copy number, 39%) contributed data for the meta-analysis. There was no between-study heterogeneity (
Hazard ratios (HR) comparing patients with increased versus not increased EGFR gene copy number are presented. Each study is shown by the point estimate of the HR (square proportional to the weight of each study) and 95% confidence interval for the HR (extending lines); summary HR and their 95% confidence intervals by random-effects calculations are shown by diamonds. Value lower than one indicates that patients with increased EGFR gene copy number have improved survival compared to patients without increase in EGFR gene copy number.
The results of subgroup analysis are presented in
Comparison | Overall survival | Progression-free survival | ||
Number of studies, heterogenrity ( |
HR (95% CI); |
Number of studies, heterogenrity ( |
HR (95% CI); |
|
All studies | 10(0.886; 0) | 0.616(0.495–0.766); <0.001 | 8(0.004; 66) | 0.651(0.474–0.894); 0.008 |
Ethnicity | ||||
White | 8(0.885; 0) | 0.634(0.504–0.796); <0.001 | 7(0.008; 65.6) | 0.784(0.616–0.907); 0.003 |
East Asian | 2(0.405; 0) | 0.458(0.219–0.961); 0.039 | 1(NA) | 0.470(0.292–0.756); 0.002 |
KRAS status | ||||
wild-type | 5(0.504; 0) | 0.545(0.388–0.766); <0.001 | 3(0.002; 83.9) | 0.345(0.130–0.919); 0.033 |
mixed | 5(0.461; 0) | 0.575(0.428–0.771); <0.001 | 6(0.001; 75.2) | 0.688(0.564–0.840); <0.001 |
anti-EGFR mAbs | ||||
cetuximab | 7(0.945; 0) | 0.624(0.485–0.803); <0.001 | 5(0.074; 53.2) | 0.746(0.610–0.916); 0.004 |
others | 3(0.269; 23.8) | 0.593(0.384–0.915); 0.018 | 3(0.006; 80.5) | 0.551(0.373–0.814); 0.003 |
Detection method | ||||
FISH | 5(0.703; 0) | 0.660(0.488–0.894); 0.007 | 6(0.096; 46.5) | 0.747(0.614–0.910); 0.004 |
others | 5(0.780; 0) | 0.572(0.418–0.782); <0.001 | 2(0.003; 88.7) | 0.517(0.339–0.791); 0.002 |
Line of treatment | ||||
<80% second-line or higher | 1(NA) | 0.880(0.419–1.847); 0.735 | 3(0.121; 52.7) | 0.874(0.667–1.146); 0.330 |
≥80% second-line or higher | 9(0.908; 0) | 0.596(0.474–0.748); <0.001 | 5(0.019; 66.1) | 0.590(0.465–0.748); <0.001 |
Subgroup analyses was performed when at least two studies were in each subgroup.
Subgroup analysis was not performed for TTP as only three studies provided information for this outcome.
FISH, fluorescent
There was extreme heterogeneity among the 8 studies including PFS (
Sensitivity analysis was performed both by sequential remove of individual studies and cumulative statistics for all comparisons of all subjects and subgroups. It was shown that with the passage of time and increasing the sample size, the results of OS and PFS become more stable. The pooled HRs were not influenced by any individual study.
Funnel plots and Egger’s test were performed to assess publication bias. The data suggested that there was no evidence of publication bias for the study’s primary outcome, OS (Begg’s test
In the present study, we collected all available studies and carried out a meta analysis to examine the association of variations of EGFR GCN with prognosis of advanced CRC patients. Ten studies involving 776 patients on OS, eight studies involving 893 patients on PFS and three studies on TTP were critically reviewed. We subgrouped the articles into five groups (ethnicity, KRAS status, anti-EGFR mAbs, detection method and line of treatment). Meta-analysis showed increased EGFR GCN was significantly associated with improved OS and PFS but not TTP. The median OS of patients harboring increased GCN showed 1.61-fold increase, the median PFS showed 1.54-fold increase. A meta-analysis of these studies confirms that increased EGFR GCN is indeed associated with a moderate OS and PFS benefit, from anti-EGFR treatment for metastatic CRC patients. Similarly, EGFR gene copy number has also been evaluated as a potential predictor of response of tyrosine kinase inhibitors (TKIs) in non-small-cell lung cancer patients, and a meta analysis has demonstrated an association between increased EGFR copy number, and improved survival outcomes
The anti–EGFR monoclonal antibody is effective in prolonging survival in patients with metastatic colorectal cancer after failure of conventional chemotherapy
EGFR is a transmembrane tyrosine kinase receptor that, on ligand binding, triggers two main signaling pathways, the RAS-RAF-MAPK axis, which is mainly involved in cell proliferation, and the PI3K-PTEN-AKT pathway, which is mainly involved in cell survival and motility
Current obstacles for a future clinical application of EGFR GCN are mainly concentrated on the following two aspects: detection methods and difficult technical reproducibility. FISH technique has been used in most previous studies, but the FISH results are challenging to interpret and the lack of standardization of analytical methods and scoring systems may partly explain why the EGFR GCN evaluation has not been incorporated into clinical practice yet
There are several limitations kept in consideration in this meta-analysis. First, most of the studies were not conclusive because they evaluated limited patient series that were nonhomogeneously treated. Second, relatively small sample sizes included in East Asians may also influence the results, and further studies are necessary to detect the potential role of GCN. Third, primarily the unavailability of individual patient data that would allow correction for potential confounding factors such as age, gender, or additional genetic aberrations. Finally, different detection methods used in the studies included in the analysis may have different quality control issues.
In conclusion, our meta-analysis provides evidence that EGFR gene copy number is a prognostic marker for survival among patients receiving anti-EGFR mAbs for advanced colorectal cancer. Furthermore, according to our results, the prognostic ability of EGFR gene copy number appears to be significantly stronger among those populations that received second-line or higher treatment.