Magnetic resonance imaging-targeted biopsy and accuracy with radical prostatectomy specimens: a grading issue

To compare the accuracy of ISUP grade groups (GG) determined from MRI-targeted and systematic biopsies with radical prostatectomy (RP) specimens using the 2014 and 2019 ISUP recommendations, and to identify predictors of upgrading.

We analyzed 212 consecutive patients who underwent both MRI-targeted and systematic biopsies followed by RP at the Jules Bordet Institute from October 2019 to January 2025. GG were assigned per 2019 ISUP guidelines, using the highest Gleason score, either from the global score of MRI-targeted lesions (including perilesional cores) or from individual systematic biopsy cores. Sensitivity analyses explored an empirical update to the 2019. ISUP recommendations by excluding minor higher-grade patterns (< 5%). Accuracy was compared using McNemar's test and weighted kappa (κ) statistics. Predictors of upgrading were identified via multivariable logistic regression and CHAID analysis.

The 2019 ISUP recommendations improved concordance with RP specimens by 8% (CI: 0.02-0.14, p = 0.002), reduced downgrading rates by 11% (CI:0.16 - 0.06, p < 0.001) and shown upgrading in 16% of cases (n = 33/212), representing a 3% increase compared to 2014 ISUP criteria (CI: 0.0008-0.06, p = 0.03). Weighted κ-values improved between the 2014 (κ = 0.58), 2019 (κ = 0.72) and updated 2019 ISUP recommendations (κ = 0.77). PSA-level (odds ratio [OR]: 1.11, 95%CI: 1.04-1.18, p = 0.001) and number of cores with clinically significant (Cs-) PCa (OR: 0.67, 95%CI: 0.54-0.82, p < 0.001) were independent predictors of upgrading. CHAID analysis confirmed that patients with ≤ 4 positive cores had a higher risk of upgrading (20% vs. 8.2%).

The 2019 ISUP recommendations enhance PCa grading accuracy. The number of cores with csPCa is a key predictor of upgrading and should inform treatment decisions. These findings support the integration of the 2019 ISUP guidelines into clinical practice to improve risk stratification and minimize overtreatment.