The serum half-life of prostate-specific antigen (PSA) was calculated in 66 patients subsequent to radical prostatectomy. Comparing serum half-life to disease outcome in 37 patients after a minimum follow-up of two years, it was found that PSA serum half-life identifies patients with residual disease earlier and more reliably than the presence or absence of detectable PSA levels postoperatively. It is suggested that residual tumor affects the half-life by contributing to the serum level of PSA. When PSA serum half-life was calculated solely in potentially cured patients, we found a half-life of 1.6 days, which is considerably shorter than in previous reports based on patient populations regardless of the outcome of disease in the follow-up. To elucidate the route of PSA elimination, serial urine PSA levels were determined before and after radical prostatectomy, revealing strong evidence for the assumption that PSA is not eliminated by the kidneys in its unchanged form.