To see whether progestins prevent estrogen action in breast cancer cells, we have studied in vitro the effect of R5020 on the cell growth and the synthesis of secreted proteins in T47D and R27 human breast cancer cells. While R5020 had no effect on cell growth when tested alone, it significantly inhibited the growth of both cell lines in the presence of estradiol (1 nM). The effect was most clear-cut after 10-12 days of treatment and was dose dependent, a half-maximal inhibition occurred with 1 nM R5020. R27, a cloned MCF7 variant resistant to Tamoxifen, remained responsive to R5020, which prevented the effect of 17 beta-estradiol (E2) and inhibited cell growth in the presence of Tamoxifen. This suggests that the two antiestrogens are acting through different mechanisms. Dihydrotestosterone and dexamethasone did not reproduce or inhibit the effect of R5020 on cell growth. R5020 was ineffective in a rat tumor cell line containing androgen and glucocorticoid receptors but lacking progesterone receptors and estrogen receptors. These results suggest that R5020 is probably acting via progesterone receptors rather than via the androgen or glucocorticoid receptors. Using sodium dodecyl sulfate-polyacrylamide gel electrophoresis, we have shown that R5020 specifically decreases the production of the 52K protein, a major protein released by R27 cells after E2 stimulation. We conclude that R5020 has an antiestrogenic activity on breast cancer cells in culture, since it prevents the stimulation of cell growth and protein synthesis by E2.