The value of pretreatment clinical and biochemical parameters in patients with newly diagnosed untreated prostate carcinoma and no indications for bone metastases on the bone scintigram.

Between 10% and 25% of patients with newly diagnosed prostate cancer without bone metastases at the time of diagnosis will develop metastases during follow-up. To determine the value of clinical and biochemical parameters for assessment of prognosis at the time of diagnosis, a retrospective study was performed in 124 consecutive patients with newly diagnosed prostate cancer without bone metastases. The mean follow-up was 41 months, during which time 36 patients died and 15 patients developed metastases. Bone scans were classified from 0 (=normal) through 2 (=abnormal, but not typical for metastases) and were correlated with age, alkaline phosphatase (AP), prostate-specific antigen (PSA), tumour grade, T-stage and N-stage. In patients with a class 2 scan, additional roentgenograms and follow-up were used to exclude metastases at initial stage. All parameters, including therapy, were finally correlated with the development of metastases and survival. For survival 38 patients with proven metastases were used as controls. For all parameters tested, no statistically significant differences were found between the three bone scan classifications. The interval between diagnosis and the development of metastases ranged from 12 to 72 months. For the risk of development of metastases only PSA was found to be a significant correlate (P=0.0075). However, when tumour stages were clustered in limited disease (T0-2) and extensive disease (T3-4), the incidence of metastases was significantly higher in patients with extensive disease than in those with limited disease (P=0.0021). Finally, age, PSA and Anderson classification were found to be significant correlates of survival, but in stepwise analysis PSA was selected as the most prognostic variable (P<0.0001). In contrast with a typical pattern of metastases on bone scintigraphy, an abnormal scan (class 1 and 2) at the time of diagnosis is not a poor prognostic parameter of the risk of death. In conclusion, in patients with prostate cancer without bone metastases at the time of diagnosis, pretreatment PSA and tumour stage can be used for the assessment of risk of development of metastases during follow-up and survival. For this purpose, tumour stage should be clustered in limited and extensive disease.