Relationship between radioligand binding assay, immunoenzyme assay and immunohistochemical assay for estrogen receptors in human breast cancer and association with tumor differentiation.

We have studied the merit of a new enzyme immunoassay (EIA) in relation to the results obtained with a conventional dextran-coated charcoal assay (DCC) of estrogen receptors (ER) in cytosols and nuclear extracts of human breast cancer tissue. The results of the two assays were related to cytosolic progesterone receptor content (PgR), semiquantified ER content in formalin-fixed paraffin embedded tissue specimens and tumor differentiation. The EIA was found stable at low cytosol protein concentrations (0.5 mg/ml). The EIA and DCC assays were highly correlated both in cytosols (r = 0.92, n = 57) and nuclear extracts (r = 0.82, n = 25), but the EIA slightly overestimated the ER values in both ER fractions. A significant correlation between ER in nuclear (ER(N] and cytosolic (ER(C] fractions was established with both assays (DCC: r = 0.90, n = 56; EIA: r = 0.83, n = 24). A qualitative relationship was established between PgR and ER fractions as determined with both assays, the best quantitative association was between PgR and ER(N(DCC] (r = 0.58, n = 34, P less than 0.001). A significant qualitative and quantitative relationship was found between semiquantified ER content in formalin-fixed, paraffin-embedded tissue and ER(C(DCC] (r = 0.88), ER(N(DCC] (r = 0.86], ER(C(EIA] (r = 0.60), ER(N(CIA] (r = 0.64) and PgR (r = 0.65). Finally, we found tumor differentiation to be significantly associated with ER content as determined with all assays except for ER(N(EIA]. We recommend the use of the DCC assay for routine analysis of ER until the clinical correlation of EIA results has been established.