The long-term clinical impact of biochemical recurrence of prostate cancer 5 or more years after radical prostatectomy

PURPOSE: Information regarding the clinical impact of delayed (5 years or greater) biochemical failure (BF) after radical prostatectomy (RP) is lacking. We undertook an investigation to differentiate the innocuous recurrence of serum prostate specific antigen (PSA) from that which heralds an eventual clinical failure (CF), and to determine if there is a period following RP when a patient is cured of clinical disease. MATERIALS AND METHODS: Men with clinically localized prostate cancer (PCA) undergoing RP (1987 to 1995) were identified from our longitudinal PCA registry. Outcome measurements were based on the detection of post-RP serum PSA 0.4 ng/ml or greater, clinical identification of cancer recurrence and disease related death. RESULTS: Following RP in 3,903 eligible men, 33% had a detectable PSA (median followup 8.8 years). Of these BFs 27% occurred after 5 or more disease-free years. Currently, 29% of all men with BF have clinical evidence of PCA, with 8% dying of PCA (median actuarial survival time from CF to death 9.8 years). Progression from BF to CF was not significantly altered by the disease-free interval (p = 0.544). A PSA doubling time less than 12 months significantly increased the risk of CF regardless of the interval from surgery. Risk factors for BF were significant throughout the duration of followup. CONCLUSIONS: Patients are at prolonged risk for BF and CF following RP. Regardless of the timing of the initial PSA recurrence the PSA doubling time is the most powerful predictor of progression, stratifying patients with BF into high and low risk groups for CF.