To determine the potential role of p53 inactivation in prostate cancer, we studied a well characterized cohort of 86 patients treated with radical prostatectomy. We analyzed patterns of p53, mdm2, and p21/WAF1 expression by immunohistochemistry. Results were then correlated with clinicopathological parameters of poor outcome, including time to PSA relapse. In addition, data were also correlated with proliferative index, as assessed by Ki67 antigen detection. p53-positive phenotype, defined as identification of nuclear immunoreactivity in > 20% of tumor cells, was observed in 6 of 86 cases (7%). An association was observed between p53-positive phenotype and decreased time to PSA relapse (P < 0.01). mdm2-positive phenotype, defined as > or = 20% of tumor cells displaying nuclear immunoreactivity, was observed in 28 of 86 cases (32.5%). mdm-2-positive phenotype was found to be associated with advanced stage (P = 0.009). p21-positive phenotype, defined as > 5% of tumor cells with nuclear immunoreactivity, was observed in 28 of 86 cases (32.5%). An association was observed between p21-positive phenotype and high Ki67 proliferative index (P = 0.002). Patients with p21-positive phenotype had a significant association with decreased time to PSA relapse (P = 0.0165). In addition, a significant association was found between p21-positive phenotype and coexpression of mdm2 (P < 0.01). Forty-three of 86 cases (50%) were found to have one or more alterations, and patients with any alteration were found to have a higher rate of PSA relapse (P < 0.01). It is our hypothesis that a pathway of prostate cancer progression involves p53 inactivation caused by mdm2 overexpression and that p21 transactivation in this setting is due to an alternative signaling system, rather than through a p53-dependent mechanism.