Immunohistochemically Determined Estrogen and Progesterone Receptor Levels: A Comparison of Three Antibodies with the Ligand-Binding Assay

Abstract: Traditionally, estrogen and progesterone receptor levels have been determined by biochemical ligandbinding assays, but more recently immunohistochemical techniques have become available. They have gained popularity due to their low cost, smaller sample size requirements, and direct visualization capability of reaction location. Several antibody clones are commercially available and antibodies directed against the estrogen receptor (ER) are supplied by Ventana Medical Systems (Tucson, AZ), Abbott Laboratories (Abbott Park, IL), and lmmunotech Westbrook, ME). Antibodies directed against the progesterone receptor (PgR) are supplied by Ventana Medical Systems (Tucson, AZ), lmmunotech (Westbrook, ME), and Becton-DickinsonKelI Analysis Systems (San Jose, CA). Computer-assisted image analysis using the CAS ZOOTM (Becton-DickinsonKIS, San Jose, CA) allows quantitation of immunohistochemically determined receptor levels. Correlation of quantitated immunohistochemical ER levels with values determined by ligand-binding assay revealed the Ventana antibody to most closely predict the ligand-binding results (wk = .667). The Ventana anti-progesterone antibody quantitation most closely correlated with the ligand-binding results (wk = 435) for determination of PgR. Progesterone receptor level as determined by any of the tested methods did not stratify patients into favorable and unfavorable prognostic groups. Estrogen receptor level as determined by the Ventana antibody was the most predictive of patient outcome but this relationship did not reach statistical significance (p = .09). Most discrepancies between the ligand-binding assay and the immunohistochemical assays were associated with one of three factors: (a) low volume of neoplastic cells present due either to small sample size or high stromal content, (b) premenopausal status with circulating endogenous estrogens potentially occupying receptor sites, (c) presence of benign breast epithelium resulting in a false-positive ligand-binding assay.