This study investigated the prognostic value of inflammation-based scores in patients with high-risk localized prostate cancer who underwent radical prostatectomy with or without neoadjuvant androgen deprivation therapy (ADT). Inflammation-based scores included the neutrophil-to-lymphocyte ratio (NLR), derived NLR (dNLR), platelet-to-lymphocyte ratio (PLR), prognostic nutritional index (PNI), and plasma fibrinogen. A total of 440 patients (380 patients treated without neoadjuvant ADT and 60 patients treated with neoadjuvant ADT) were retrospectively evaluated in our medical center. Receiver operating characteristic (ROC) curves and Kaplan-Meier analyses were performed to compare the prognostic value of these scores. Univariate and multivariate Cox regression analyses were also performed. For all patients, dNLR and PNI were predictive of biochemical recurrence (.=0.041 and <0.001, respectively). Subgroup analysis of neoadjuvant strategies was also performed. For patients treated with neoadjuvant ADT, no selected inflammation-based scores were significantly correlated with biochemical recurrence (.>0.05). In contrast, for patients treated without neoadjuvant ADT, NLR (area under the ROC curve [AUC] =0.576, P=0.033), dNLR (.=0.585 and 0.017), PLR (AUC =0.582, P=0.024), and PNI (AUC =0.622, P<0.001) were predictive of biochemical recurrence. Kaplan-Meier analyses showed that dNLR (.=0.044), PLR (.=0.028), and PNI (.=0.004) were significantly associated with biochemical recurrence. Based on multivariable models, PNI was an independent predictor of biochemical recurrence (hazard ratio: 0.56, 95% confidence interval: 0.35-0.90, P=0.016). High dNLR, high PLR, and low PNI were associated with poor biochemical recurrence-free survival in patients undergoing radical prostatectomy for high-risk localized prostate cancer not treated with neoadjuvant ADT. In particular, PNI was an independent prognostic factor for biochemical recurrence.