Purpose: We investigated the usefulness of bone alkaline phosphatase isoenzyme and prostate specific antigen (PSA) determined by radioimmunoassay to predict bone scan evidence of metastasis in newly diagnosed untreated and treated prostate cancer. Materials and Methods: We analyzed bone alkaline phosphatase enzyme concentrations in 350 men, including 150 controls, 100 with benign prostatic hyperplasia and 100 with prostate cancer (52 with stages T1 to 4, MO and 48 with stages T1 to 4, M1 to 4). We also analyzed bone alkaline phosphatase enzyme concentrations in 61 stages T1 to 4, MO prostate cancer cases during followup after radical prostatectomy or hormonal therapy, and 17 had clinical progression (9 with local, 5 with lymph node and 3 with bone metastases). Simultaneously, we analyzed PSA concentrations. Results: Average bone alkaline phophatase enzyme levels were 12, 11.1 and 10.0 ng./ml. in the control, benign prostatic hyperplasia and stage MO prostate cancer groups, respectively (p not significant), and 83.2 ng./ml. in patients with stage M1 to 4 disease (p < 0.001). Considering that to diagnose bone metastasis the cutoff for bone alkaline phosphatase enzyme and PSA is 30 ng./ml. and 100 ng./ml., respectively, clinical effectiveness was 93.7% and 81.8%, respectively. Finally, measurement of both substances at:the same time increased clinical effectiveness to 97.9%. During followup a bone alkaline phosphatase enzyme level that becomes greater than 30 ng./ml. (0% in the local and lymphatic progression groups, and 100% in the bone metastasis group) indicates the need to perform a bone scan. Conclusions: We recommend the clinical use of bone alkaline phosphatase enzyme determined by radioimmunoassay and PSA measurement for the diagnosis of bone metastases and progression of prostate cancer because of the good sensitivity and specificity.