The levels of prolactin receptors (PRL-R), based on a new micro-method, estrogen receptors (ER) and progesterone receptors (PG-R), were determined in 159 breast cancer specimens. Sixty-seven of 159 tumors (42%) had PRL-R levels of 20 fmol/mg protein or higher, and were regarded as PRL-R positive. In 60 of the 159 samples (38%), no PRL-R could be detected, and the remaining 32 samples (20%) were considered borderline. While a positive correlation was found between the presence of ER and PGR, no correlation was detected between PRL-R and any steroid receptors (r = -0.024 for ER vs PRL-R, 0.052 for PGR vs PRL-R and 0.002 for ER + PGR vs PRL-R). Furthermore, PRL-R in levels of 20 fmol/mg protein or higher were found in 35% of samples in which no steroid receptors were detected as well as in 38% and in 27% of samples which exhibited positive ER or PGR respectively (greater than or equal to 20 fmol/mg protein). On the other hand, in 47% of the samples possessing both ER and PGR, the PRL receptors could not be found. These results clearly demonstrate that in the human breast cancer, the presence of PRL-R is independent of the status of either ER or PGR. It is suggested that the measurement of PRL-R could serve as a chemical marker to guide a possible therapeutic use of PRL-suppressing drugs in women with breast cancer.