Purpose Adjuvant hormonal therapy may be beneficial in patients who are treated with RRP and found to have adverse pathological findings. We assessed the natural history of detectable PSA in such patients with particular emphasis on the prognostic usefulness of PSADT. Materials and Methods We identified 903 patients treated with RRP and adjuvant hormonal therapy (started less than 90 days postoperatively) for prostate cancer at our institution between 1990 and 1999. PSADT was calculated by log linear regression in men with 2 or more PSA measurements available at least 90 days apart. CSS and sRFS were estimated by the Kaplan-Meier method and analyzed using Cox proportional hazard models. Results At a median followup of 9.1 years PSA had become detectable in 369 of 771 patients (47.9%) who achieved an undetectable nadir. PSADT evaluable in 463 patients was less than 12 months in 68 (14.6%) and more than 10 years in 283 (61.1%). N stage and Gleason score were significantly associated with sRFS and CSS. PSADT was a significant predictor of sRFS and CSS in N+ and N0 cases with a cancer death HR of 0.55 (95% CI 0.43 to 0.71) and 0.50 (95% CI 0.31 to 0.79), respectively. The association between PSADT and survival persisted after multivariate adjustment for preoperative PSA, specimen Gleason score and seminal vesicle invasion. Conclusions This study demonstrates that many patients have slow progression despite increasing PSA following RRP and adjuvant hormonal therapy. Nodal status, cancer grade and PSADT are predictive of sRFS and CSS, and may be a useful means of selecting patients for future adjuvant therapy trials.