Objectives: To report our findings regarding to the natural history of prostate cancer (PCa) that shows recurrence after radical prostatectomy (RP), in terms of time to development of metastatic disease and death from PCa. To identify independent predictors of PSA recurrence., Material and Methods: Retrospective analysis of 227 patients with clinically localized PCa who underwent RP. The event PSA recurrence was defined as the presence of a postoperative PSA level of 0,2 ng/ml or higher at least 3 months after surgery. Hence, cases with shorter follow-up time were excluded from analysis. No adjuvant therapy (radiotherapy or hormonal therapy) was performed in the included population. Recurrence free survival was calculated during the follow-up period (Kaplan-Meier analysis). Uni and multivariate study was performed in order to assess the ability of factors as preoperative PSA level, Gleason score in surgical specimen, capsular penetration, positive surgical margins (excluding urethral), extracapsular extension, positive pelvic lymph nodes, and seminal vesicle invasion, to predict PSA recurrence. Finally, we selected the group of patients with PSA recurrence and calculated the probability of being free from distant metastatic disease during the follow-up period. Also, function of disease-specific survival was calculated., Results: A total of 208 records were finally included in the study. Median age was 61 years. A total of 47 (22.6%) presented with extracapsular extension. Median follow-up time was 35.8 months, and 49 (23.6%) developed PSA recurrence. Recurrence free survival was 79.9% and 67.4% at 2 and 5 years, respectively. Only three factors were identified with the aid of multivariate analysis as independent predictors of recurrence: preoperative PSA >= 10 ng/ml (hazard ratio--HR--3.03), Gleason score in surgical specimen 8 or higher (HR 3.42), and the finding of capsular penetration (HR 2.17). When only patients with PSA recurrence were considered, 16.3% developed distant metastasis. Probabilities of being free from distant disease after PSA recurrence were 97.7% and 86.9% at 2 and 5 years respectively (actuarial median time 110.8 months). Only 2 patients died from PCa, therefore disease-specific mortality analysis was not performed., Conclusions: Although an important proportion of patients present with PSA recurrence after RP in our setting, the prognosis in term of development of metastatic disease is acceptable in the short-medium term. Anyway, further analysis will be needed to ascertain the evolution of these patients in the long term.