An obvious problem for the surgeon or oncologist treating breast cancer has been to identify the patients likely to respond to endocrine manipulation. Until recently, clinical factors such as previous response to hormone therapy, disease-free interval, age and menopausal status, and location of the dominant metastatic lesion were the principal criteria for selecting therapeutic regimens for these women. Recently, the measurement of steroid hormone receptors has become an important laboratory test. Progress during the last decade has shown that: the most reliable methods of determining estrogen receptors (ER) and progestin receptors (PR) are multipoint titration analysis using dextran-coated charcoal, and sucrose density gradient centrifugation; 55% to 65% of primary breast tumors contain more than 10 femtomole/mg cytosol protein of ER; 45% to 55% of metastatic breast tumors contain more than 10 fmol/mg cytosol protein of ER; ER are present more often in tumors of postmenopausal women compared with those of premenopausal women; benign breast lesions such as fibrocystic disease and fibroadenomas usually contain less than 10 fmol/mg cytosol protein of ER; 90% of male breast carcinomas contain ER; approximately 55% of women with breast tumors containing ER respond objectively to endocrine therapy, either additive or ablative; less than 3% of women with breast tumors lacking ER respond objectively to hormone therapy. In addition, it has been suggested that the absence of ER in a breast tumor correlates well with an increased response to cytotoxic chemotherapy; 45% to 60% of primary or metastatic breast tumors contain PR. Also, the presence of both ER and PR in a breast tumor indicates a 75% to 80% likelihood that the patient will respond to endocrine manipulation, either additive or ablative; it has been suggested that the presence of the 8 Svedberg form of ER in a breast tumor (as detected by sucrose gradient centrifugation) improves the accuracy of selecting the patient likely to respond to endocrine therapy; and there appears to be a relationship between the quantity of ER in a breast tumor and a patient's response to endocrine therapy. The incidence of response to hormone therapy increases with increasing ER levels.