Clinically significant cut-off values to discriminate between receptor-positive and -negative, and the prognostic value of estrogen receptors (ER) and progesterone receptors (PgR) measured by enzyme immunoassay (EIA) have not yet been established. We have therefore measured ER and PgR by EIA in cytosols from 205 primary breast cancer biopsies. Clinically significant cut-off values (30 fmol/mg protein for ER; 27 fmol/mg protein for PgR), as related to tumor recurrence (median follow-up, 47 months), have been established by isotonic regression analysis. These data were compared to those obtained by simultaneously performed dextran-coated charcoal (DCC) assays (cut-off values: 18 fmol/mg protein for ER, and 26 fmol/mg protein for PgR) on the same cytosols, and to DCC assays performed previously (up to 10 years ago) on cytosols prepared from other parts of the tissue biopsies (cut-off values: 18 fmol/mg protein for ER, and 23 fmol/mg protein for PgR). Using the cut-off values for the EIA and the DCC assays performed on the same cytosols, the discrepancies between receptor status appeared less than 10% both for ER and for PgR. Furthermore, the concentrations of ER or PgR detected with the EIA or DCC assay were highly and significantly correlated (Spearman rank correlations: for ER, Rs = 0.94; for PgR, Rs = 0.88; P less than 0.0001). After classification in different phenotypes with respect to ER/PgR status (+/+, +/-, -/+, and -/-), analysis for relapse-free survival and overall survival showed equal prognostic power in the comparable groups in the order, from favorable to unfavorable, of +/+ greater than +/-(-/+) greater than -/- (chi2: P less than 0.0001), irrespective of the assay which has been used for quantification of the receptor. It is concluded that both the conventionally used DCC and the newly available EIA methods are equally useful for assessing ER and PgR status.