Abstract Purpose To evaluate the potential predictive value of the preoperative serum albumin to globulin ratio (AGR) for oncological outcomes in patients treated with radical prostatectomy (RP) for clinically non-metastatic prostate cancer (PCa). Methods Pre-operative AGR was assessed in a multi-institutional cohort of 6041 patients treated with RP. Logistic regression analyses were performed to assess the association of the AGR with advanced disease. We performed Cox regression analyses to determine the relationship between AGR and biochemical recurrence (BCR). Results The optimal cut-off value was determined to be 1.31 according to receiver operating curve analysis. Compared to patients with a higher AGR, those with a lower preoperative AGR had worse BCR-free survival (P < 0.01) in the Kaplan–Meier analysis. Pre- and post-operative multivariable models that adjusted for the effects of established clinicopathologic features, confirmed its independent association with BCR [hazard ratio (HR) 1.52, 95% confidence interval (CI) 1.31–1.75, P < 0.01, HR 1.55, 95% CI 1.34–1.79, P < 0.01, respectively]. However, the addition of AGR to established prognostic models did not improve their discrimination. Conclusion While AGR is significantly associated with BCR, in the present study, the clinical impact of AGR was not large enough to affect patient management. Longer follow-up is necessary to observe the true effect of AGR.