Influence of estrogen receptor variants on the determination of ER status in human breast cancer.

Determination of estrogen receptor alpha (ER) status in breast cancer is an important predictive factor for clinical response to endocrine therapy. We have recently shown that discrepancies in ER status determined by immunohistochemical assay (ER-IHA) can occur between amino-terminal (1D5) and carboxyl-terminal (AER-311) targeted ER antibodies and that those tumors which demonstrate discordance are associated with increased expression of truncated ER variant mRNAs. In this study, we have explored this observation to examine if ER variant expression can exert a direct effect on ER-IHA or whether this association is attributable to the characteristics of the antibodies. ER negative cos-1 cells were transfected with expression vectors containing wild type ER (wt-ER) and/or a frequently expressed truncated variant, ER-clone-4 variant. We found that ER-IHA performed with the same N- and C-terminal targeting ER antibodies on cos-1 cells expressing wt-ER alone demonstrated no difference in signals by western blot (P > 0.1). However, co-expression of wt-ER and the truncated ER-clone-4 variant, resulted in discordant IHA results with relatively higher ER-IHA H-scores from N-terminal antibodies (P < 0.03). Furthermore, re-examination of a subset of breast tumors previously studied by ER-IHA showed persistent concordance in 4/5 cases and persistent differences in 3/5 cases with a different pair of ER antibodies. We conclude that the presence of truncated ER variant proteins can interfere with the interpretation of ER status determined by IHA and that this may account for some of the inconsistencies between ER status and response to endocrine therapy.