OBJECTIVE To present the current state of systemic radiopharmaceutical therapy for the palliation of pain in individuals with metastatic cancer and to evaluate the palliative effect and degree of hemotoxicity of strontium chloride 89 (89Sr) in patients with painful osteoblastic metastases primarily from prostate and breast cancer. METHODS A MEDLINE search through December 1994 was performed to identify English-language studies that met the following criteria. All eligible studies reported treatment of patients with painful osteoblastic bony metastases primarily from prostate or breast cancer treated with intravenous 89Sr. For study eligibility, evaluation of clinical response as assessed by the Karnofsky index, need for pain medication, or changes in mobility or sleep patterns was required. Hemotoxicity data were a requirement. A minimum of 10 prostate cancer cases was necessary for study inclusion. Only those studies assessing clinical response following one injection of 89Sr were included. Preliminary reports of cooperative studies were not included. Doses of 89Sr ranged from 0.6 MBq/kg (16 microCi/kg) to 400 MBq (10.8 mCi) per patient. Evaluation of patients for at least 3 months following 89Sr treatment was required. In addition, two studies examining issues of cost with regard to 89Sr treatment were identified. METHODS Baseline pain assessment and periodic pain estimates as measured by the Karnofsky index, medication diaries, changes in mobility, sleep patterns, and/or ability to work were the basis for assessment of response. Baseline and periodic complete blood cell counts were the basis for hemotoxicity evaluation. RESULTS Palliation and hemotoxicity data were analyzed separately for each study. Some improvement occurred in as many as approximately 80% of patients. Several studies demonstrated complete relief of pain in at least 10% of patients The nadir of platelet and white blood cell counts appears at approximately 4 to 8 weeks following injection, with a partial return to baseline by 12 weeks. As many as 10 injections spaced 3 months apart have been given to some patients with repeated palliative effect and without serious hemotoxicity. Reinjection may be limited by a platelet count below 60 x 10(9)/L, a white blood cell count below 2.4 x 10(9)/L, or the absence of osteoblastic skeletal metastasis as seen on bone scan. Studies examining treatment costs suggest that 89Sr may decrease costs associated with palliation of pain due to metastatic disease. CONCLUSIONS As many as 80% of selected patients with painful osteoblastic bony metastases from prostate or breast cancer may experience some pain relief following 89Sr administration. In addition, as many as 10% or more may become pain free. Duration of clinical response may average 3 to 6 months in some cases. Hemotoxicity is mild. A decrease in treatment costs with administration of 89Sr to patients with painful osteoblastic bony metastases from prostate cancer may occur. These observations reflect the preliminary nature of knowledge in this field and point to the need for larger clinical trials of the use of 89Sr palliation.