False negatives in oestrogen-receptor assay

Abstract

Stefan Aebi and colleagues (May 27, p 1869)1Abei S Castiglione-Gertsch M Gelber RD et al.Is chemotherapy alone adequate for young women with oestrogen-receptor-positve breast cancer?.Lancet. 2000; 355: 1869-1874Summary Full Text Full Text PDF PubMed Scopus (246) Google Scholar did not note the limitation of the dextran-coated charcoal (DCC) assay method used to measure and define the oestrogen-receptor status in the over-viewed International (Ludwing) Breast Cancer Study Group. The DCC method uses tritiated oestradiol as a ligand to measure oestrogen receptor in tissue homogenates. It has several limitations, including false-negative results if the sample is not snap frozen immediately, through contamination by non-malignant cells and disturbance by endogenous and exogenous oestrogen administration. However, a more important limitation has come to light. When the assay was in common use, only one oestrogen receptor was known, since renamed ERα. A further oestrogen receptor has been identified, named ERβ, which is expressed in breast cancer.2Kupier GGJM Enmark E Pelto-Hukko M Nilsson S Gustafsson J-A Cloning of a novel estrogen receptor expressed in rat prostate and ovary.Proc Natl Acad Sci USA. 1996; 93: 5925-5930Crossref PubMed Scopus (4182) Google Scholar, 3Jarvinen TAH Pelto-Hukko M Hollo K Isola J Estrogen receptor ß is coexpressed with ERα and PR and associated with nodal status, grade and proliferation rate in breast cancer.Am J Pathol. 2000; 156: 29-35Summary Full Text Full Text PDF PubMed Scopus (258) Google Scholar The two receptors represent separate gene products with distinct biological roles and ligand-binding specificity. Since the two wild-type receptors can bind oestradiol, interpretation of DCC data must take into account that any value represents the sum of the binding capacity of these receptors, which Aebi and colleagues did not note in their article. The issue is further complicated by the presence of ERβ variants, which differ in their interaction with oestradiol and ERα, compared with a wild-type receptor. A C-terminal splice variant of ERβ, known as βcx, exists.4Ogawa S Inoue S Wantanbe T et al.Molecular cloning and characterization of human estrogen receptor (ßcx: a potential inhibitor of estrogen action in human.Nucleic Acids Res. 1998; 26: 3505-3512Crossref PubMed Scopus (406) Google Scholar This variant has no oestrogen-binding ability and preferentially forms heterodimers with ERα rather than ERβ. In addition on binding ERα, it acts as a dominant negative receptor, inhibiting its ability to regulate gene transcription. A further variant with an 18 aminoacid insertion, known as β2,5Hanstein B Liu H Yancisin MC et al.Functional analysis of a novel estrogen receptor isoform.Mol Endocrinol. 1999; 13: 129-137Crossref PubMed Scopus (129) Google Scholar has an eightfold lower affinity for oestrogen than wild type ERβ. The presence of homodimers or heterodimers containing these variants could affect the DCC assay result negatively, potentially leading to some women being given the wrong oestrogen-receptor status. We have confirmed the presence of ERβ in 19 breast cancers investigated, as well as the βcx variant in eight investigated samples. The presence and relative concentrations of ERα and ERβ receptors and ERβ variants might, therefore, result in tumours with different proliferative responses to oestradiol and could explain the different outcomes seen for the oestrogen receptor positive and negative younger women. In addition, the effect on DCC assay data has to be borne in mind when such data are used in the future. False negatives in oestrogen-receptor assayAuthors' reply Full-Text PDF