An estimated 20-40% of men experience a biochemical recurrence within 10 years of definitive prostate cancer treatment. No single prostate-specific antigen (PSA) value is invariably associated with clinical metastasis or cancer-specific survival; PSA kinetics might prove to be a more important predictor of eventual progression-free survival and cancer-specific survival than absolute PSA level alone. With only one-third of patients progressing from biochemical recurrence to clinical disease, therapeutic morbidity should not outpace risk of disease progression. Salvage radiation therapy following radical prostatectomy has widely variable long-term biochemical control rates (from 18 to 64% depending on the follow-up period). Early hormonal therapy delivered as castration or complete androgen blockade might delay clinical metastasis in patients with high-risk pathologic disease; however, the adverse effects and morbidity of long-term therapy must not be underestimated. Non-steroidal antiandrogens as monotherapy for early biochemical recurrence, particularly for younger men who wish to preserve their libido and sexual potency, have received considerable attention, but there are conflicting data on long-term outcomes. Because of their favorable adverse-effect profiles, non-traditional therapies that exert localized hormonal or cellular effects are receiving considerable attention for treatment of early, PSA-only recurrence. Data from animal models provide a rationale for the use of these therapies, but there is a lack of evidence to support prolongation of progression-free survival or cancer-specific survival.