Comparison of sex steroid receptor determinations in human breast cancer by enzyme immunoassay and radioligand binding

Abstract

Most investigators comparing ligand-binding procedures for quantifying estrogen and progestin receptors in human breast cancer with procedures employing monoclonal antibody-based methods have utilized an inappropriate variety of reaction conditions, including the elimination of sodium molybdate in the steroid-binding assays. We studied 197 biopsies of human breast cancer, comparing the results of simultaneous measurements of both estrogen and progestin receptors in identical cytosols by enzyme immunoassay and by radioligand binding using the commercially available kits developed by Abbott Laboratories and by DuPont/NEN Products, respectively. Regression analyses comparing the results from the two procedures indicated a linear relationship, with correlation coefficients ranging from 0.79 to 0.93 for both types of receptors over a wide range of data. Using the widely established cutoff value of 10 fmol/mg of cytosol protein for the ligand-binding method, and calculating sensitivity and specificity limits according to McNeil et al. (1975), an equivalent cutoff value of 15 fmol/mg of cytosol protein was determined for the enzyme immunoassay of these receptors. Endocrine status of the patient did not appear to alter the cutoff values of either estrogen or progestin receptors when determined by enzyme immunoassay. We recommend that these cutoff values be considered until the results of clinical correlations are completed.