Introduction. The study aimed to evaluate the performance of multiparametric magnetic resonance imaging (mpMRI) and the Martini model to predict extraprostatic extension (EPE) and biochemical recurrence (BCR) of prostate cancer (PCa). Materials and methods. 61 patients underwent a radical laparoscopic prostatectomy. The preoperative risk of EPE was determined using mpMRI and the Martini model. Results. MpMRI predicts the presence of EPE of PCa with a sensitivity and specificity of 47.4% and 85.7%, respectively (AUC 0.66, 95% CI: 0.51-0.82, p = 0.046). The Martini model's sensitivity was higher, but the specificity was lower than that of mpMRI and was 84.2% and 66.7%, respectively (AUC 0.78, 95% CI: 0.66-0.89, p < 0.001). Univariate and multivariate Cox analysis indicated that EPE in mpMRI (HR 6.6, 95% CI: 1.8-24.1), and the presence of positive surgical margins (PSM) (HR 7.1, 95% CI: 1.9-26.7) are independent factors increasing the probability of BCR. Conclusions. MpMRI and Martini model are valuable tools in local staging of PCa, managing and predicting the oncological treatment outcomes of patients with PCa.