One hundred thirty-three consecutive, previously untreated patients who had metastatic breast cancer were treated with a combination of 5-fluorouracil, doxorubicin (Adriamycin), and cyclophosphamide (FAC). They were randomly assigned to receive nonspecific immunotherapy with a heptavalent pseudomonas vaccine. Sixty-five patients were treated with pseudomonas vaccine, whereas 68 did not receive immunotherapy. In addition, all patients with estrogen receptor-positive tumors or tumors with an estrogen receptor status were also treated with tamoxifen. To allow clinical assessment of hormone sensitivity in vivo, tamoxifen was started 6 weeks before chemotherapy except in patients who had life-threatening disease. After the initial 6 weeks of tamoxifen, 3% of patients had achieved a complete remission, 9% a partial remission, while 16% achieved a minor response. The maximum response after tamoxifen and chemotherapy included complete remissions in 20% of patients and partial remissions in 61% of patients for an overall remission rate of 81%. The median response duration was 15 months, and the median survival time, 27 months. There were no differences in remission rate, remission duration, or survival time between the groups treated with or without pseudomonas vaccine. Eleven patients with limited metastatic disease received radiotherapy consolidation to initially involved sites. In these patients the median time from radiotherapy to progression of disease was 33 months, and the median survival time was 46 months. We conclude that nonspecific immunotherapy with pseudomonas vaccine failed to increase remission rate or survival time. Furthermore, the addition of tamoxifen to FAC chemotherapy did not improve the remission rate or duration compared to a recent, historical control group of patients treated with only FAC chemotherapy.