Bisphosphonates are used in oncology as a means of decreasing complications due to bone metastases, in association with anticancer treatment, especially in patients with breast cancer, prostate cancer and myeloma. Little is known about the effects of bisphosphonates on bone metastases from other tumors and in particular from tumors for which no effective treatment is available. We conducted a randomized, double-blind placebo-controlled trial of oral clodronate in patients with bone metastases from tumors poorly responsive to chemotherapy, with the aims of evaluating the effects of this drug on symptoms control and bone metastases evolution. Sixty-six patients with poorly responsive tumors such as non-small cell lung cancer (NSCLC), bladder cancer, gastrointestinal cancers, kidney cancer, melanoma and metastatic carcinoma of unknown origin entered the study. Patients were randomized to receive either clodronate 1,600 mg/day for one year or identical placebo-containing tablets. Various parameters such as Karnofsky performance status, pain score (measured by a visual-analogue scale) and analgesic requirement were recorded at monthly intervals. Of the 66 patients enrolled, 9 were observed for one month or less; 7 were followed for two months; only 50 patients were followed for more than 2 months and could be adequately evaluated. At 3 months both clodronate and placebo-treated patients had a decrease in Karnofsky performance status, with the decrease being more evident in the placebo group. Mean pain scores showed an increase of pain in patients receiving placebo and a decrease of pain in patients receiving clodronate, although the difference failed to be statistically significant. Analgesics requirement increased in both groups, but significantly more in patients receiving placebo (p = 0.042), in whom increase in opioid requirements was particularly evident. Toxicity was low, with occasional gastroenteric discomfort in both groups. The main problem of this study was the difficulty in recruiting an adequate number of patients and following them for a sufficient period of time: general conditions rapidly deteriorated in many patients, and approximately 25% of the 66 enrolled were not considered evaluable; few patients survived for the length of the study, one year. This might partly account for the lack of significance of some of the parameters under study. With these limits, oral clodronate demonstrated some efficacy in symptom control and in reducing the need for analgesics.