Predictive Value of Quantitative MRI Parameters for Clinically Significant Prostate Cancer: A Comparison Between PI-RADS v2.1 Category 4 and 5 Lesions on Biparametric MRI
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Purpose Evaluating the role of quantitative MRI parameters in differentiating PI-RADS v2.1 category 4 and 5 lesions for improved detection of clinically significant prostate cancer(csPCa). Methods In a retrospective analysis of 153 biopsy-proven patients with suspected csPCa undergoing PI-RADS v2.1 MRI, two radiologists independently measured lesion dimensions (maximum diameter, mean diameter, area, volume) on tri-planar T2-weighted images(T2WI). Metric differences were assessed between csPCa and benign prostatic hyperplasia (BPH), and across prostatic zones. For PI-RADS ≥ 4 lesions, the diagnostic performance of axial plane measurements was compared using Receiver operating characteristic (ROC)analysis. Results A total of 165 nodules from 153 patients were analyzed, comprising 108 csPCa (69 in the peripheral zone (PZ), 39 in the transition zone (TZ)) and 57 non-csPCa (20 PZ, 37 TZ). csPCa nodules exhibited significantly larger maximum and mean diameters than BPH nodules across all imaging planes (axial, coronal, sagittal; all p < 0.05), with no significant differences in these diameters among the planes themselves. Similarly, csPCa nodules demonstrated greater area and volume than BPH nodules in both the TZ and PZ. For diagnosing csPCa, the respective Area Under the Curves (AUCs)for maximum diameter, mean diameter, area, and volume were approximately 0.701, 0.723, 0.722, and 0.811 in the TZ, and 0.733, 0.738, 0.722, and 0.819 in the PZ. Conclusion In PI-RADS v2.1 category ≥ 4 nodules, which exhibited no dominant growth orientation, volume outperformed diameter and area metrics for diagnosing csPCa, with proposed thresholds of 1.07 mm³ and 1.23 mm³ for differentiating categories 4 and 5, respectively.