Comparison of 7-tesla and conventional 3-tesla magnetic resonance imaging for prostate cancer staging.
[BACKGROUND] Conventional 3T magnetic resonance imaging (MRI) shows limited sensitivity for predicting extraprostatic extension (EPE) and frequently underestimates prostate tumor size versus pathology
APA
Kim TM, Song SH, et al. (2025). Comparison of 7-tesla and conventional 3-tesla magnetic resonance imaging for prostate cancer staging.. Prostate international, 13(4), 227-233. https://doi.org/10.1016/j.prnil.2025.06.002
MLA
Kim TM, et al.. "Comparison of 7-tesla and conventional 3-tesla magnetic resonance imaging for prostate cancer staging.." Prostate international, vol. 13, no. 4, 2025, pp. 227-233.
PMID
41472924
Abstract
[BACKGROUND] Conventional 3T magnetic resonance imaging (MRI) shows limited sensitivity for predicting extraprostatic extension (EPE) and frequently underestimates prostate tumor size versus pathology. Ultrahigh-field (7T) MRI-performed on prostatectomy specimens after surgery-may overcome these limitations through superior spatial resolution and tissue contrast.
[MATERIALS AND METHODS] In this prospective study, 20 patients with biopsy-proven prostate cancer underwent preoperative 3T MRI followed by 7T MRI of radical prostatectomy specimens. Two experienced uroradiologists independently assigned EPE scores (a 5-point scale modified from International Society of Urological Pathology (ISUP) guidelines) and measured tumor dimensions on both modalities; histopathology served as the reference standard. A clarity score quantified the deviation of each EPE score from an equivocal midpoint. Statistical analyses included logistic regression for EPE prediction, receiver operating characteristic curve analysis for diagnostic performance, Wilcoxon signed-rank testing for paired comparisons of clarity scores and size discrepancies, weighted kappa for inter-reader agreement, and intraclass correlation coefficients (ICCs) for MRI-pathology size concordance.
[RESULTS] The 7T MRI EPE score was the only significant predictor of pathologic EPE (odds ratio: 4.40, = 0.02) and achieved an area under the receiver operating characteristic curve of 0.86 versus 0.61 for 3T ( = 0.06). Clarity scores improved on 7T ( < 0.01), with inter-reader agreement rising from κ = 0.32 (3T) to κ = 0.53 (7T). Both modalities underestimated tumor size, but the median discrepancy was significantly lower with 7T (2.0 mm vs. 6.0 mm; < 0.01), and size concordance improved from ICC = 0.48 (3T) to ICC = 0.70 (7T).
[CONCLUSIONS] The 7T MRI enhances the diagnostic accuracy, clarity, and reproducibility of EPE assessment and reduces tumor size underestimation compared to 3T MRI, supporting its potential role in optimizing preoperative prostate cancer staging.
[MATERIALS AND METHODS] In this prospective study, 20 patients with biopsy-proven prostate cancer underwent preoperative 3T MRI followed by 7T MRI of radical prostatectomy specimens. Two experienced uroradiologists independently assigned EPE scores (a 5-point scale modified from International Society of Urological Pathology (ISUP) guidelines) and measured tumor dimensions on both modalities; histopathology served as the reference standard. A clarity score quantified the deviation of each EPE score from an equivocal midpoint. Statistical analyses included logistic regression for EPE prediction, receiver operating characteristic curve analysis for diagnostic performance, Wilcoxon signed-rank testing for paired comparisons of clarity scores and size discrepancies, weighted kappa for inter-reader agreement, and intraclass correlation coefficients (ICCs) for MRI-pathology size concordance.
[RESULTS] The 7T MRI EPE score was the only significant predictor of pathologic EPE (odds ratio: 4.40, = 0.02) and achieved an area under the receiver operating characteristic curve of 0.86 versus 0.61 for 3T ( = 0.06). Clarity scores improved on 7T ( < 0.01), with inter-reader agreement rising from κ = 0.32 (3T) to κ = 0.53 (7T). Both modalities underestimated tumor size, but the median discrepancy was significantly lower with 7T (2.0 mm vs. 6.0 mm; < 0.01), and size concordance improved from ICC = 0.48 (3T) to ICC = 0.70 (7T).
[CONCLUSIONS] The 7T MRI enhances the diagnostic accuracy, clarity, and reproducibility of EPE assessment and reduces tumor size underestimation compared to 3T MRI, supporting its potential role in optimizing preoperative prostate cancer staging.