Enhancing the Efficacy of Radiomics-Based Prediction of Fuhrman Pathological Grading in Renal Clear Cell Carcinoma Using Multilayer Spiral CT Imaging.

Abstract

Background: Clear cell renal cell carcinoma (ccRCC) is the most prevalent subtype of renal cell carcinoma (RCC). Conventional pathological methods of Fuhrman pathological grading system have limitations. This study aims to investigate the efficacy of radiomics-based multilayer spiral computed tomography (CT) imaging of Fuhrman pathological grading in ccRCC.

Methods: A retrospective analysis was conducted on the clinical data of ccRCC patients admitted in our hospital from March 2023 to March 2024. The patients were classified as low-grade (Fuhrman pathological grades I and II) or high-grade (Fuhrman pathological grades III and IV). Statistical methods, including correlation analysis, receiver operating characteristic (ROC) curves and construction of a joint predictive model, were utilised to assess the predictive value of these imaging omics indicators for Fuhrman pathological grading in ccRCC. The primary outcome assessment parameter in this study was the predictive value of these imaging omics indicators for Fuhrman pathological grading in ccRCC.

Results: The clinical data from 101 ccRCC patients were examined, with 56 cases classified as low-grade and 45 cases as high-grade. The grey-level co-occurrence matrix (GLCM) features between low and high Fuhrman grading groups, including contrast (0.24 ± 0.08 vs. 0.33 ± 0.09), energy (0.73 ± 0.05 vs. 0.67 ± 0.06) and homogeneity (0.63 ± 0.05 vs. 0.57 ± 0.05), showed notable distinctions (p < 0.001). The CT imaging characteristics between low and high Fuhrman grading groups, including enhancement homogeneity (0.34 ± 0.08 vs. 0.26 ± 0.08) and washout half-time (28.57 ± 4.35 vs. 34.72 ± 5.62) demonstrated a substantial variation between the groups (p < 0.001). The enhancement homogeneity (r = 0.476), washout half-time (r = -0.519), contrast (r = 0.454), energy (r = -0.453) and homogeneity (r = -0.541) showed significant correlations with Fuhrman pathological grading. The predictive value of these features was evident, with a combined imaging genomics model exhibiting an area under the curve of 0.929.

Conclusions: This study demonstrated the potential of radiomics-based prediction using multilayer spiral CT imaging for accurately predicting Fuhrman pathological grading in ccRCC.