AUTHOR=Wang Shuang , Ji Tuo , Yu Dan , Dai Yimeng , Zhang Butian , Liu Lin 

TITLE=Grading of clear cell renal cell carcinoma using diffusion MRI with a multimodal apparent diffusion model

JOURNAL=Frontiers in Oncology

VOLUME=Volume 15 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2025.1507263

DOI=10.3389/fonc.2025.1507263

ISSN=2234-943X

ABSTRACT=ObjectiveTo assess the feasibility of utilizing parameters derived from a multimodal apparent diffusion (MAD) model to distinguish between low- and high-grade clear cell renal cell carcinoma (ccRCC).MethodDiffusion-weighted imaging (DWI) scans with 12 b-values (0 - 3000 s/mm²) were conducted on 54 patients diagnosed with ccRCC (30 low-grade and 24 high-grade). The MAD model parameters, including diffusion coefficients (Dr, Dh, Dui, Df) representing restricted diffusion, hindered diffusion, unimpeded diffusion, and flow, respectively, were computed. Proportions corresponding to these diffusion types (fr, fh, fui, ff) and the heterogeneous nature of hindered diffusion (αh) were also obtained. Parameters were compared between low- and high-grade groups. Receiver operating characteristic (ROC) curves were used to evaluate the diagnostic performance of these parameters, compared with the apparent diffusion coefficient (ADC) from a mono-exponential model.ResultSignificant differences between low- and high-grade ccRCC were observed in Dh (low-grade: 1.360 ± 0.11 μm2/ms; high-grade group, 1.254 ± 0.13 μm2/ms; P = 0.0327), fr (low-grade: 0.06 ± 0.005; high-grade: 0.08 ± 0.009; P = 0.0233), and αh (low-grade: 0.872 ± 0.22; high-grade: 0.896 ± 0.39; P = 0.0294). Additionally, the ADC values (low-grade: 0.924 ± 0.08 μm2/ms; high-grade group, 0.854 ± 0.04 μm2/ms; P = 0.0323) showed statistical significance. The combination of Dh, fr, and αh provided the highest diagnostic accuracy of 0.667, with a sensitivity of 0.750, specificity of 0.734, and area under the curve of 0.796, outperforming individual parameters and ADC.ConclusionThe MAD diffusion model shows promise as a non-invasive imaging tool for distinguishing between low- and high-grade ccRCC, which may aid in preoperative planning and personalized treatment strategies.