Related Experiment Video
Updated: Jun 11, 2025
A Cognitive Fusion-guided Prostate Biopsy Using Multiparametric Magnetic Resonance Imaging and Transrectal Ultrasound
Published on: March 21, 2025
Preoperative multiparametric magnetic resonance imaging based risk stratification system for predicting biochemical
Cagri Akpinar1, Digdem Kuru Oz2, Alkan Oktar3
1Department of Urology, Ankara Etlik City Hospital, Ankara, Turkey.
Background:
Multiparametric magnetic resonance imaging (mpMRI) is used as a current marker in preoperative staging and surgical decision-making, but current evidence on predicting post-surgical oncological outcomes based on preoperative mpMRI findings is limited. In this study We aimed to develop a risk classification based on mpMRI and mpMRI-derived biopsy findings to predict early biochemical recurrence (BCR) after radical prostatectomy.
Methods:
Between January 2017 and January 2023, the data of 289 patients who underwent mpMRI, transrectal ultrasound-guided cognitive and fusion targeted biopsies, and subsequent radical prostatectomy (RP) with or without pelvic lymph node dissection in a single center were retrospectively re-evaluated. BCR was defined as a prostate specific-antigen (PSA) ≥ 0.2 ng/mL at least twice after RP. Multivariate logistic regression models tested the predictors of BCR. The regression tree analysis stratified patients into risk groups based on preoperative mpMRI characteristics. Receiver operating characteristic (ROC)-derived area under the curve (AUC) estimates were used to test the accuracy of the regression tree-derived risk stratification tool.
Results:
BCR was detected in 47 patients (16.2 %) at a median follow-up of 24 months. In mpMRI based multivariate analyses, the maximum diameter of the index lesion (HR 1.081, 95%Cl 1.015-1.151, p = 0.015) the presence of PI-RADS 5 lesions (HR 2.604, 95%Cl 1.043-6.493, p = 0.04), ≥iT3a stage (HR 2.403, 95%Cl 1.013-5.714, p = 0.046) and ISUP grade ≥4 on biopsy (HR 2.440, 95%Cl 1.123-5.301, p = 0.024) were independent predictors of BCR. In regression tree analysis, patients were stratified into three risk groups: maximum diameter of index lesion, biopsy ISUP grade, and clinical stage on mpMRI. The regression tree-derived risk stratification model had moderate-good accuracy in predicting early BCR (AUC 77 %) CONCLUSION: Straightforward mpMRI and mpMRI-derived biopsy-based risk stratification for BCR prediction provide an additional clinical predictive model to the currently available pathological risk tools.
