While hormone receptor values are extremely valuable in breast cancer management, many variables leading to falsely negative receptor results should be considered before clinical decision making in the setting of metastatic disease. At the Papanicolaou Comprehensive Cancer Center in Miami, Florida 271 patients with metastatic breast cancer received tamoxifen over a 4-year period. Only 40 of the 204 patients (19.6%) with available estrogen receptor assay information had pretreatment receptor values classically considered receptor-poor. While four patients had inevaluable tamoxifen trials and 12 patients had "compassionate" use in end-stage situations, 24 patients received tamoxifen therapy in the face of receptor-poor values because of clinical or histopathological correlates suggesting the possibility of false negative assay results. Six of 36 evaluable patients (16.6%), including those treated on "compassionate grounds," responded to tamoxifen, while 6 of 24 more highly selected patients (25%) responded. Response durations in these six patients were 11, 12, 28, 28+, 49, and 51 months, respectively; two of these six also had significant, objective tamoxifen withdrawal responses of 9 and 14 months. Based on these results, it is urged that receptor values are studied and integrated with classical clinical and histopathologic variables before making a clinical decision so that patients with receptor-poor assay results are not prematurely labeled hormonally unresponsive.