Abstract
BACKGROUND: Multiparametric Magnetic Resonance Imaging (mpMRI), including diffusion-weighted imaging (DWI), plays a key role in detecting and stratifying prostate cancer (PCa). The Apparent Diffusion Coefficient (ADC) aids in tissue characterization and may assist in distinguishing between Gleason score 7 subgroups (GS 3 +4 and GS 4 +3). This study aimed to evaluate mean ADC and ADC(ratio) (ratio of tumor-to-normal ADC) in differentiating these subgroups and to assess the influence of magnetic field strength (1.5 T vs. 3.0 T) on diagnostic performance. METHODS: This retrospective study included 48 patients with histologically confirmed GS 7 PCa who underwent mpMRI at 1.5 T or 3.0 T. Two experienced radiologists independently and blindly delineated tumor and reference ROIs. Inter-observer agreement was evaluated using the intraclass correlation coefficient (ICC). Correlation, t-tests, and receiver operating characteristic (ROC) analyses were used to assess relationships, group differences, and optimal cut-off values. RESULTS: Excellent inter-observer agreement was found for both mean ADC and ADC(ratio) (ICC > 0.9). At 3.0 T, mean ADC (10.9 ×10⁻⁴ mm²/s vs 8.55 ×10⁻⁴ mm²/s, p = 0.011) and ADC(ratio) (0.68 vs 0.54, p = 0.003) significantly distinguished GS 3 + 4 from GS 4 + 3, while no differences were observed at 1.5 T. Spearman's ρ confirmed stronger correlations at 3.0 T (ADCratio ρ=-0.510; p = 0.004). For the combined (3.0 T + 1.5 T) dataset, ADC(ratio) achieved an AUC of 0.748 and mean ADC an AUC of 0.718 (p > 0.05). CONCLUSION: ADC(ratio) demonstrated high reproducibility and stable diagnostic performance across scanners, supporting its potential as a reliable imaging biomarker for prostate cancer grading.