Dear Editor

We read with interest the paper by Borgquist et al (2008), published recently in Journal of Clinical Pathology¹. In their article the authors aimed to investigate the impact of ERbeta expression on breast cancer outcome using a cohort of 512 tumours represented in tissue microarray (TMA). Since the discovery of ERbeta over a decade ago, this has been the goal of many research groups. However progress in this area has been impeded by the lack of a consensus in terms of choice of primary antibody and cut-off value used to determine ERbets postivity^2,3. Additionally, we know that ERbeta exists as 5 isoforms (ERbeta1-5), each formed by alternative splicing of the last coding exon^4,5, which has further complicated interpretation of immunohistochemical studies. Comparative studies have been conducted to help determine the most suitable antibody for a number of applications and of these, 2 reliable antibodies have emerged for immunohistochemistry: 14C8 (AbCam, Cambridge, UK), located in the N-terminus of the ERbeta peptide and which detects most isoforms of ERbeta and PPG5/10 (Serotec, Oxford, UK) which is specific for ERbeta1^2,6,7. We routinely use both these antibodies in our immunohistochemical studies. Borgquist et al¹ used a different antibody, EMR02 (Novocastra, Newcastle-upon-Tyne), which is claimed through epitope mapping studies to show specific reactivity to a 17-amino acid sequence present in the C-terminus of full length wild type ERbeta and which is absent in ERbeta2/ERbetacx 8. While Borgquist et al¹ show immunohistochemical and Western blot analysis of ERbeta in a panel of breast cell lines we find it remarkable that they showed no staining of their breast TMAs especially as all the survival data presented in the paper was obtained from these. Using cell lines to validate antibodies is certainly a step in the right direction but it is insufficient as tissue characteristics and processing protocols will almost certainly differ substantially from those used in cell lines and these may alter antibody specificity and sensitivity.

In addition, the authors used a very low cut off value to define ERbeta positivity (1%) and with such a value we find it surprising that the percentage of ERbeta positive tumours was only 50% - substantially less than in many previous studies^6,9-12. We believe this could indicate an issue with antibody specificity and suggest this be confirmed through peptide absorption studies.

Finally, as part of or on-going programme to understand the significance of ERbeta in breast carcinogenesis, when EMR02 antibody became available, we compared its efficacy to the well-validated 14C8 and PPG5/10 antibodies which we use routinely. As illustrated in Figure 1, we were unable to achieve robust and consistent staining using EMR02, while TMA and full sections of breast carcinoma were clearly stained with 14C8 and PPG5/10.

Figure 1 - Serial sections of TMAs (top panel) and full sections (bottom panel) of breast carcinomas stained with EMR02 (a, d), 14C8 (b, e) and PPG5/10 (c, f). Consistent staining was only observed for 14C8 and PPG5/10 with EMR02 unable to detect the protein.

There still remains a great deal of controversy and indeed suspicion surrounding the potential importance of ERbeta in breast cancer. We believe to a large extent this is due to the use of poorly validated antibodies which have poisoned the field somewhat. We urge all scientists, clinicians and journals to be aware of the need for carefully validated studies to help eliminate these controversies which, more than a decade on, still continue to cloud ERbeta.

Valerie Speirs

References

1. Borgquist S, Holm C, Stendahl M, et al. Oestrogen receptors alpha and beta show different associations to clinicopathological parameters and their co-expression might predict a better response to endocrine treatment in breast cancer. J Clin Pathol 2008;61:197-3.

2. Carder PJ, Murphy CE, Dervan P, et al. A multi-centre investigation towards reaching a consensus on the immunohistochemical detection of ERbeta in archival formalin-fixed paraffin embedded human breast tissue. Breast Cancer Res Treat 2005;92:287-93.

3. Shaaban AM, Speirs V. Estrogen receptor beta - which one and where should we draw the line? Hum Pathol 2006;37:498.

4. Moore JT, McKee DD, Slentz-Kesler K, et al. Cloning and characterization of human estrogen receptor beta isoforms. Biochem Biophys Res Commun 1998;247:75-8.

5. Poola I, Abraham J, Baldwin K, et al. Estrogen receptors beta4 and beta5 are full length functionally distinct ERbeta isoforms: cloning from human ovary and functional characterization. Endocrine 2005;27:227-38.

6. Skliris GP, Parkes AT, Limer JL, et al. Evaluation of seven oestrogen receptor beta antibodies for immunohistochemistry, western blotting, and flow cytometry in human breast tissue. J Pathol 2002;197:155-62.

7. Weitsman GE, Skliris G, Ung K, et al. Assessment of multiple different estrogen receptor-beta antibodies for their ability to immunoprecipitate under chromatin immunoprecipitation conditions. Breast Cancer Res Treat 2006;100:23-31.

8. Rees ML, Marshall I, McIntosh GG, et al. Wild-type estrogen receptor beta expression in normal and neoplastic paraffin-embedded tissues. Hybrid Hybridomics 2004;23:11-8.

9. Saunders PT, Millar MR, Williams K, et al. Expression of oestrogen receptor beta (ERbeta1) protein in human breast cancer biopsies. Br J Cancer 2002;86:250-6.

10. Shaaban AM, O'Neill PA, Davies MP, et al. Declining estrogen receptor- beta expression defines malignant progression of human breast neoplasia. Am J Surg Pathol 2003;27:1502-12.

11. Fuqua SA, Schiff R, Parra I, et al. Estrogen receptor beta protein in human breast cancer: correlation with clinical tumour parameters. Cancer Res 2003;63:2343-39.

12. Nakopoulou L, Lazaris AC, Panayotopoulou, et al. The favourable prognostic value of oestrogen receptor beta immunohistochemical expression in breast cancer. J Clin Path 2004;57:523-28.