An ethynylestradiol (EE(2)) radioimmunoassay method (RIA) using a monoclonal antibody (MAB) reagent is described. 76 plasma and 67 buffer blanks spiked with known concentrations of EE(2) were used as positive samples. The steroid was extracted using dichloromethane method with an extraction efficiency of 78 +/- 2.6%. The plasma extracts were not purified. Separation of the free from bound tritiated EE(2) was by centrifugation with dextran-coated charcoal suspension in phosphate-gelatin buffer. Within 95% confidence limits, the sensitivity of the assay was 10 pg/ml in buffer samples, and 43 pg/ml in unpurified plasma samples. The accuracy and precision of the method were lower for the plasma samples than for the buffer solutions but within accepted limits (i.e., 80-120% for accuracy, and up to 20% for precision). The mean recovery accuracy (R) of the method for measuring 38 pg/ml EE(2) in buffer was 101 +/- 0.9% (n = 67), while for 2.39 pg/ml to 18.0 pg/ml EE(2) in plasma was 97 +/- 12% (n = 76). The buffer solutions and plasma samples containing 600 pg/ml EE(2) had a mean R of 102% (n = 18) and 90% (n = 16), respectively. The average interassay variation (%CV) between buffer samples was 14 +/- 1.2%, and between plasma samples was 19 +/- 8%. Plasma samples with 18 pg/ml and 2.39 pg/ml EE(2), together with the blank negative plasma samples showed high interassay variations (CV > 20%). Except for the above samples, however, the precision of the method in all sample groups was within accepted limits. The intra-assay variation of replicate assays of a single plasma sample containing 1.2 ng/ml EE(2) and 4 ng/ml levonorgestrel was 13% (n = 7). In the presence of very high concentration of closely related levonorgestrel, the R of the method was 106%. The increase in average variability (%CV) between results for the plasma samples (CV 19 +/- 8%) over that of the spiked buffer solutions (CV 14 +/- 1.2%) suggest some possible interference by co-extracted materials from plasma, for example lipids, which interfered in the ligand-antibody reactions. Further investigation may prove that a pre-assay purification of the sample, e.g., HPLC, would eliminate this interference. Nevertheless, the immunoassay described above completely eliminates the problem of cross-reactivity caused by the natural estrogens in plasma with the conventional polyclonal antibodies. It possesses an acceptable intra- and interassay variation with very high specificity to EE(2). It promises to be clinically useful in the assessment of extremely low levels of EE(2) expected in patients having a daily intake of 30 mu g of the estrogenic hormone contained in many oral contraceptives.
