Salvage radiation therapy (SRT) is standard treatment for patients after radical prostatectomy (RP). However, the optimal timing of SRT remains to be elucidated. We retrospectively reviewed 133 prostate cancer (PCa) patients who underwent SRT for biochemical recurrenceafter RP. Disease progression was defined as repeated prostate-specific antigen (PSA) level more than 0.2ng/mL, greater than the post-SRT nadir or radiographic progression. A receiver operating characteristic curve analysis was used to identify the optimal pre-SRT PSA level for predicting progression after SRT. Cox regression analyses were performed to elucidate the association between clinicopathologic characteristics and disease progression. Fifty-one PCa patients (38.4%) experienced disease progression after SRT. The optimal cutoff value of the pre-SRT PSA for predicting disease progression was 0.44ng/mL. In multivariable analysis, pre-SRT PSA >0.44ng/mL was a significant independent predictor of post-SRT disease progression [hazard ratio (HR): 2.02, P=0.02]. Although the pre-SRT PSA >0.44ng/mL did not maintain its independent association with disease progression in the multivariable analysis of patients with adverse pathology (HR: 1.63, P=0.22), PSA within 4weeks after RP as a continuous variable was significantly associated with disease progression (HR: 1.19, P=0.04). Our results highlight thatin PCa patients who undergo RP, SRT should be performed before their PSA reaches 0.44ng/mL. In patients with adverse pathology disease, a high PSA level within the 4weeks after RP might identify those who are likely to have disease progression, andthese patients might require systemic therapy.