Correction: Can we rely on magnetic resonance imaging for prostate cancer detection and surgical planning? Comprehensive analysis of a large cohort of patients undergoing transperineal mapped biopsies.

Abstract

Purpose: To evaluate MRI and histological concordance in prostate cancer (PCa) identification via mapped transperineal biopsies.

Methodology: Retrospective per-lesion analysis of patients undergoing MRI and transperineal biopsy at the Valencian Institute of Oncology (2016-2024) using CAPROSIVO PCa data. Patients underwent MRI, with or without regions of interest (ROI), followed by transperineal biopsies (3-5 cores/ROI, 20-30 systematic). Sensitivity (Se), specificity (Sp), negative predictive value (NPV), positive predictive value (PPV), and area under the curve (AUC) were calculated, considering PI-RADS 3 lesions as positive or negative. Gleason Grade Group (GG) > 1 defined clinically significant PCa (csPCa).

Results: 1817 lesions were analyzed from 1325 patients (median age 67, median PSA 6.3 ng/ml). 53% MRI were negative, GG > 1 prevalence was 29.1%. MRI-negative cases showed varying PCa rates: 57.4% negative, 34.3% GG 1, and 8.3% GG > 1. PI-RADS 3 lesions had mixed outcomes: 45.6% benign, 29.5% GG 1, and 24.9% GG > 1. 9.2% PI-RADS 4-5 lesions were negative, 23% GG 1, and 67.8% GG > 1. For PI-RADS 3 lesions considered positive, Se, Sp, NPV, PPV, and AUC were 84.9%, 68.4%, 91.7%, 52.6%, and 0.77 respectively. Considering PI-RADS 3 as negative yielded 70.6% Se, 86.2%, Sp, 87.7% NPV, 67.8% PPV, and 0.78 AUC.

Conclusion: MRI and mapped prostate biopsies exhibited moderate concordance. MRI could miss up to one in ten csPCa foci and misinterpret one in two ROIs. Careful MRI interpretation is crucial for optimizing patient care.