|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 23, No 31 (November 1), 2005: pp. 8118-8119 © 2005 American Society of Clinical Oncology. DOI: 10.1200/JCO.2005.03.0130
Epidermal Growth Factor Receptor Expression in Breast CancerLaboratory of Cellular and Molecular Biology, Center for Cancer Research, Bethesda, MD
Cancer Therapeutics Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD
Laboratory of Cellular and Molecular Biology, Center for Cancer Research, Bethesda, MD To the Editor: We read with interest the article describing the relationship between epidermal growth factor receptor (EGFR) expression and ErbB-2 expression and prognosis in breast cancer.1 In the work by DiGiovanna et al only 15% of the breast cancers expressed EGFR, but there was a striking correlation between overexpression of ErbB-2 and EGFR expression and between overexpression of activated ErbB-2 and EGFR expression. The coexpression of EGFR with either ErbB-2 or activated ErbB-2 carried a poor prognosis. While these data are provocative and suggest that EGFR contributes to the activation of ErbB-2 in breast cancer, we have reservations about the conclusions based on our experience with the antibody to EGFR that was used (anti-EGFR 1005, Santa Cruz Biotechnology, Santa Cruz, CA). In our experience with this antibody, it cross reacts with ErbB-2 when ErbB-2 is expressed at high levels, as is frequently seen in breast cancers with ErbB-2 amplification. We first noted this when evaluating human breast cancer cell lines for EGFR expression (Fig 1). In a cell line that expresses EGFR (ie, MCF10A, Fig 1A, panel 1, lane 1), the anti-EGFR (1005) antibody detected the EGFR at the expected molecular weight. In the SKBr3 cell line, we observed an abundant protein that migrated at a molecular weight that was slightly higher than the EGFR detected in MCF10A (Fig 1A, panel 1, compare lanes 1 and 3). Extended exposure of this immunoblot detected a faint band in SKBr3 cells that migrates at the same size as the EGFR seen in MCF10A (data not shown). No band was detected in the ZR75-1 cell line by this antibody. When we probed these samples with another anti-EGFR antibody (anti-EGFR 2232 from Cell Signaling, Beverly, MA), we detected a strong signal for EGFR only in the MCF10A cell line (Fig 1A, panel 2, lanes 1 to 3). Extended exposure of the immunoblot with the anti-EGFR (2232) antibody detected a faint band in SKBr3 that migrates at the same size as the EGFR seen in MCF10A (data not shown). SKBr3 is derived from a breast cancer patient with amplified ErbB-2 and the cell line maintains both ErbB-2 gene amplification and protein overexpression.2 This suggested that the higher molecular weight protein detected with the anti-EGFR (1005) antibody in the SKBr3 cell line was the highly overexpressed ErbB-2. To confirm this we stripped and reprobed the blot originally probed with the anti-EGFR (1005) antibody with an antibody specific for ErbB-2 (Neomarkers Ab-1, Freemont, CA). This antibody detected a band in the SKBr3 sample that comigrated with the band detected by the anti-EGFR (1005) antibody used in the paper (Fig 1A, panel 3, lane 3). Both MCF10A and ZR75-1 expressed ErbB-2, but at significantly lower levels than SKBr3 (Fig 1A, panel 3, lanes 1 to 3). These data suggested that the anti-EGFR (1005) antibody cross reacts with ErbB-2 when ErbB-2 is expressed at high levels. To confirm this, we transfected HEK293T cells with empty vector, EGFR, or increasing amounts of ErbB-2 plasmid (Fig 1A, lanes 4 to 8). As in the breast cancer cell lines, the anti-EGFR (1005) antibody used by the authors detected EGFR in cells transfected with EGFR plasmid (Fig 1A, panel 1, lane 5) as well as a protein that migrates at a slightly higher molecular weight than EGFR in the cells transfected with high amounts of ErbB-2 plasmid (Fig 1A, panel 1, lanes 6 to 8). Reprobing of this blot with the anti-EGFR (2232) antibody or with the ErbB-2 specific antibody confirmed that the transfected cells expressed the appropriate proteins (Fig 1A, panels 2 and 3, lanes 4 to 8). As suggested by the data with the breast cancer cell lines, the cross reactivity of the anti-EGFR (1005) antibody with ErbB-2 was only seen at high levels of transfected ErbB-2 expression comparable to the expression seen in the SKBr3 cell line (eg, Fig 1A, compare lanes 3 and 7 in panels 1 and 3). Our immunoblotting experiments with the anti-EGFR (1005) antibody were performed using 0.2 µg/mL of antibody, the same as in the paper by DiGiovanna et al. Additional experiments, using a 10-fold lower dilution of the antibody showed similar cross reactivity (data not shown). The authors show an immunoblot that demonstrates no cross reactivity of the anti-EGFR (1005) with ErbB-2. Our data show that the cross reactivity is only apparent at high levels of ErbB-2. It is possible that the amount of ErbB-2 precipitated in the experiment shown in the paper is not sufficient to elicit cross reactivity. However, we observe cross reactivity in a physiologically relevant context since the SKBr3 cell line is derived from a patient with amplified and overexpressed ErbB-2.
We next tested whether the anti-EGFR (1005) antibody cross reacted with ErbB-2 when used for immunohistochemistry. The anti-EGFR (1005) antibody stained both MCF10A and SKBr3 cells (Fig 1B, top panels). In contrast, a second anti-EGFR antibody (31G7 from Zymed, Carlsbad, CA) stained MCF10A cells but not SKBr3 cells (Fig 1B, middle panels). Similarly, an ErbB-2 specific antibody (CB11 from Biogenex, San Ramon, CA) stained SkBr3 cells, but not MCF10A cells (Fig 1B, lower panels). These data are consistent with the cross reactivity seen on the immunoblot for the anti-EGFR (1005) antibody. Together, our immunoblotting and immunohistochemistry data suggest that the anti-EGFR (1005) antibody cross reacts with high levels of ErbB-2, levels frequently seen in patients with breast cancer. In this light, the EGFR expression data from the paper by DiGiovanna et al could be interpreted as detecting ErbB-2 expression in those patients with the highest ErbB-2 expression rather than EGFR coexpression. Previous work has demonstrated that ErbB-2 activation and prognosis correlate with ErbB-2 expression levels and thus the prognostic information gained from EGFR expression in this article could be an indication of ErbB-2 levels rather than EGFR coexpression.3 While our observations about the ErbB-2 cross reactivity of the anti-EGFR (1005) antibody do not rule out a role for EGFR in breast cancer, we feel that the data in the paper by DiGiovanna et al need to be interpreted with caution and that their observations should be confirmed with a more specific anti-EGFR antibody. Authors' Disclosures of Potential Conflicts of Interest The authors indicated no potential conflicts of interest.
REFERENCES
1. DiGiovanna MP, Stern DF, Edgerton SM, et al: Relationship of epidermal growth factor receptor expression to ErbB-2 signaling activity and prognosis in breast cancer patients. J Clin Oncol 23:1152-1160, 2005 2. Kraus MH, Popescu NC, Amsbaugh SC, et al: Overexpression of the EGF receptor-related proto-oncogene erbB-2 in human mammary tumor cell lines by different molecular mechanisms. Embo J 6:605-610, 1987[Medline]
3. Thor AD, Liu S, Edgerton S, et al: Activation (tyrosine phosphorylation) of ErbB-2 (HER-2/neu): A study of incidence and correlation with outcome in breast cancer. J Clin Oncol 18:3230-3239, 2000
Related Reply
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2005 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
