This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Janmaat, M. L. Articles by Giaccone, G. Search for Related Content PubMed Articles by Janmaat, M. L. Articles by Giaccone, G. Social Bookmarking                            What's this?

In Reply

Vrije Universiteit Medical Center, Department of Medical Oncology, Amsterdam, the Netherlands

The letter by Dr Personeni provided some points that might help to interpret our data regarding the analysis of epidermal growth factor receptor (EGFR) gene copy number with chromogenic in situ hybridization (CISH) in advanced esophageal cancer in patients treated with gefitinib.¹

In our report, we chose to score the CISH analysis by taking into account the mean number of EGFR gene copies per cell. Dr Personeni proposes to take into account intratumoral genetic heterogeneity when these features are analyzed. Cross section analysis may indeed not be representative for the whole tumor, but because the relevance of intratumoral genetic heterogeneity is currently unknown, selecting specific foci is certainly not representative, and may be biologically irrelevant. Importantly, tumor heterogeneity might, in principle, also influence the scoring of the immunohistochemical analysis of EGFR protein. However, we show that by determining the mean protein expression level a significant correlation with clinical outcome can be made.

The aim of our study was to investigate the predictive value of EGFR gene copy number or protein expression in patients treated with gefitinib, not to determine the cause of EGFR overexpression. Thus, although EGFR gene copy number and protein expression were evaluated in serial sections, we did not compare these features on a cell-by-cell basis. However, after the comments of Dr Personeni regarding the correlation between immunohistochemistry (IHC) and CISH results, we re-evaluated the IHC and CISH stainings of two samples with 3+ EGFR expression and without gene amplification on a cell-by-cell basis. In these two samples, tumor cells that showed 3+ EGFR expression did not have EGFR gene amplification (Fig 1), indicating that EGFR overexpression in esophageal cancer is not necessarily the result of EGFR gene amplification. As discussed previously by Hanawa et al,² EGFR overexpression without gene amplification may be the result of transcriptional or post-transcriptional activation.

View larger version (94K): [in this window] [in a new window] [PowerPoint Slide for Teaching]   Fig 1. Epidermal growth factor receptor (EGFR) overexpression is not generally associated with EGFR gene amplification in esophageal cancer. (A) Tumor cells with 3+ EGFR immunostaining in esophageal adenocarinoma. (B) Chromogenic in situ hybridization analysis on an adjacent section shows that cells are negative for EGFR gene amplification.

First, the conclusions from our study require confirmation because of the small number of patients analyzed for both EGFR gene copy number and EGFR expression (n = 12). Unlike studies in non–small-cell lung cancer, in which large numbers of patients have been treated with gefitinib,³ the number of esophageal cancer patients treated with gefitinib is limited. Furthermore, only a fraction of patients treated with gefitinib in this study were evaluated for all biologic markers because of the limited amount of tumor material. In comparison, Hanawa et al² performed fluorescent in situ hybridization analysis on 83 tumors, which were selected on the basis of EGFR expression determined by IHC.

Second, before the introduction of a standardized system to score EGFR gene and protein status, comparisons between different studies remain difficult. Undoubtedly, differences in the used materials and methods will impact the results. For instance, the antibodies used for IHC are different in the two studies, which may affect the final scoring. Moreover, the methods (CISH v fluorescent in situ hybridization) and cut off points to determine EGFR gene amplification are also different. Using CISH we were obviously unable to normalize EGFR gene copy number for chromosome 7. Therefore, we were not able to detect polysomy in our study, but evaluated gene copy number per cell. In addition to differences in methods, differences in scoring definitions will lead to different outcomes. For example, we considered 3+ IHC staining as EGFR overexpression,¹ whereas Hanawa et al² judged 2+ and 3+ scores as EGFR overexpression. Thus, in order to make meaningful comparisons between different studies, we agree that validated, standardized methodology and definitions are warranted for EGFR gene and protein assessments.

In conclusion, we believe that our results, although conducted on a small number of patients, are valid and show that EGFR overexpression in esophageal cancer is not necessarily the result of EGFR gene amplification. In the current absence of a standardized system for scoring EGFR gene copy number and IHC staining, it is difficult to compare different studies. Finally, the impact of intratumoral genetic heterogeneity on tumor behavior and patient outcome remains to be determined.

Authors' Disclosures of Potential Conflicts of Interest

The authors indicated no potential conflicts of interest.

REFERENCES

1. Janmaat ML, Gallegos-Ruiz MI, Rodriguez JA, et al: Predictive factors for outcome in a phase II study of gefitinib in second-line treatment of advanced esophageal cancer patients. J Clin Oncol 24:1612-1619, 2006[Abstract/Free Full Text]

2. Hanawa M, Suzuki S, Dobashi Y, et al: EGFR protein overexpression and gene amplification in squamous cell carcinomas of the esophagus. Int J Cancer 118:1173-1180, 2006[CrossRef][Medline]

3. Cappuzzo F, Hirsch FR, Rossi E, et al: Epidermal growth factor receptor gene and protein and gefitinib sensitivity in non-small-cell lung cancer. J Natl Cancer Inst 97:643-655, 2005[Abstract/Free Full Text]

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Janmaat, M. L. Articles by Giaccone, G. Search for Related Content PubMed Articles by Janmaat, M. L. Articles by Giaccone, G. Social Bookmarking                            What's this?