INTRODUCTION
Prostate cancer (PCa) is the second most common cancer in men1. Today, the gold standard treatment for localized prostate cancer is radical prostatectomy. Biochemical recurrence (BCR) can be seen in the early or late period after RP. BCR is an important condition in patients who have undergone radical prostatectomy as it indicates that clinical recurrence may also occur. Characteristics of patients such as prostate biopsy Gleason score (GS), clinical stage, preoperative total prostate specific antigen (PSA) value may differ. Various models have been developed using these pre-operative characteristics to predict BCR 2.
In clinical practice, the D’Amico risk classification and the Cancer of prostate risk assessment (CAPRA) are frequently used models to predict BCR. In D’Amico risk classification, pre-operative total PSA, clinical stage and prostate biopsy GS score are used 3. In addition to these parameters, in CAPRA scoring, patient age and positive core ratio are also used 4. On the other hand, in addition to frequently used models, models that are relatively less used in clinical practice have been defined. Yoshida et al. defined The prostate cancer risk index (PRIX) 5. In addition, the external validation of the PRIX scoring has been performed6. Soga et al. defined The GP Score, a Simplified Formula (Bioptic Gleason Score Times Prostate Specific Antigen) and demonstrated that it could predict BCR after RP 7.
The models used to predict BCR after RP have been compared. D’Amico risk classification, CAPRA scoring and Stephenson nomogram, which are frequently used in clinical practice, were compared in the same patient group 8. Although there is studies that indicate predictive values of GP and PRIX score in BCR, to best our knowledge there is no head to head comparison of GP score and PRIX score with other well-known models in predicting BCR. These models are relatively new and use of GP score is easier than other prediction models. It was known that predictions of BCR can be varied between pre-operative models. So in our study we evaluated accuracy of four pre-operative models (GP score, PRIX, D’Amico risk classification, CAPRA) in predicting BCR after RP in Turkish patients.