Intervention after PSA failure: examination of intervention time and subsequent outcomes from a prospective patient database.

PURPOSE: To investigate timing of intervention for solitary biochemical relapse following external beam radiation therapy (EBRT) for localized prostate cancer, and to assess the impact on survival. METHODS AND MATERIALS: Of 1499 men treated with EBRT from 1994 to 2000, 544 men had prostate-specific antigen (PSA) relapse by the Vancouver Criteria. Patients with near-simultaneous clinical relapse (n = 79) were excluded, leaving 465 for analysis. Pretreatment prognostic factors, the time that biochemical relapse was realized (trigger PSA time), and postrelapse doubling time (PSAdt) were examined for effect on the timing of intervention and cause-specific survival (CSS).
RESULTS: Median actuarial time from trigger-PSA time to intervention was 30 months. The only factor associated with early intervention on multivariate analysis (MVA) was faster PSAdt (p < 0.0001). Not significant were initial PSA, Gleason score, T stage, or adjuvant androgen ablation use or duration. Seventy-five percent of interventions occurred with PSAdt faster than 12 months. The 5-year CSS was 89%. Faster PSAdt (p = 0.0007), higher Gleason score (p = 0.018), and earlier intervention (p = 0.0006) were significantly associated with increased prostate cancer death on MVA. The 5-year CSS is 19% where the PSAdt was <3 months, 84% where 3-6 months, 93% where 6-12 months, and 98% where >12 months (p < 0.0001).
CONCLUSIONS: Men with a biochemical relapse post-EBRT with PSAdt >1 year had excellent 5-year CSS (98%). Rapid PSAdt is associated with increased mortality on MVA. The association of earlier intervention with increased mortality may be due to the selection of those with occult metastasis for early intervention.