Quantification of Gleason Pattern 4 at MRI-Guided Biopsy to Predict Adverse Pathology at Radical Prostatectomy in Intermediate-Risk Prostate Cancer Patients

SIMPLE SUMMARY: The amount of Gleason pattern 4 (GP4) in biopsy material may be used for individual risk stratification in prostate cancer patients. The utility of this parameter is potentially most significant in patients that fall into the intermediate-risk category, in whom various alternative strategies are under consideration. We aim to assess the performance of multiple methods of GP4 quantification in predicting the features of advanced disease at final surgery. Our retrospective analysis of data from 123 patients who underwent magnetic-resonance imaging-guided biopsy and radical prostatectomy revealed that the commonly used method of risk assessment with GP4 amount relative to cancer length may be a poor predictor of high-risk disease, as compared to other quantification methods, including our newly developed concept of GP4 volume. The results of this study may serve as the basis for further research aimed at refining the risk-assessment strategies in prostate cancer. ABSTRACT: Background: Data on Gleason pattern 4 (GP4) amount in biopsy tissue is important for prostate cancer (PC) risk assessment. We aim to investigate which GP4 quantification method predicts adverse pathology (AP) at radical prostatectomy (RP) the best in men diagnosed with intermediate-risk (IR) PC at magnetic resonance imaging (MRI)-guided biopsy. Methods: We retrospectively included 123 patients diagnosed with IR PC (prostate-specific antigen <20 ng/mL, grade group (GG) 2 or 3, no iT3 on MRI) at MRI-guided biopsy, who underwent RP. Twelve GP4 amount-related parameters were developed, based on GP4 quantification method (absolute, relative to core, or cancer length) and site (overall, targeted, systematic biopsy, or worst specimen). Additionally, we calculated PV×GP4 (prostate volume × GP4 relative to core length in overall biopsy), aiming to represent the total GP4 volume in the prostate. The associations of GP4 with AP (GG ≥ 4, ≥pT3a, or pN1) were investigated. Results: AP was reported in 39 (31.7%) of patients. GP4 relative to cancer length was not associated with AP. Of the 12 parameters, the highest ROC AUC value was seen for GP4 relative to core length in overall biopsy (0.65). an even higher AUC value was noted for PV × GP4 (0.67), with a negative predictive value of 82.8% at the optimal threshold. Conclusions: The lack of an association of GP4 relative to cancer length with AP, contrasted with the better performance of other parameters, indicates directions for future research on PC risk stratification to accurately identify patients who may not require immediate treatment. Incorporating formulas aimed at GP4 volume assessment may lead to obtaining models with the best discrimination ability.