Mengge He1,2, Yang Song3, Haiming Li4,5, Jing Lu1, Yongai Li1, Shaofeng Duan6, Jinwei Qiang1. 1. Department of Radiology, Jinshan Hospital, Fudan University, Shanghai, China. 2. The Shanghai Institution of Medical Imaging, Fudan University, Shanghai, China. 3. Shanghai Key Laboratory of Magnetic Resonance, East China Normal University, Shanghai, China. 4. Department of Radiology, Fudan University Shanghai Cancer Center, Shanghai, China. 5. Department of Oncology, Shanghai Medical College, Fudan University, Shanghai, China. 6. GE Healthcare, Shanghai, China.
Abstract
BACKGROUND: The accurate preoperative differentiation between borderline and malignant epithelial ovarian tumors (BEOTs vs. MEOTs) is crucial for determining the proper surgical strategy and improving the patient's postoperative quality of life. Several diffusion and perfusion MRI technologies are valuable for the differentiation; however, which is the best remains unclear. PURPOSE: To compare the whole solid-tumor volume histogram analysis of diffusion-weighted imaging (DWI), diffusion kurtosis imaging (DKI), intravoxel incoherent motion (IVIM), and dynamic contrast-enhanced MRI (DCE-MRI) in the differentiation of BEOTs vs. MEOTs and to identify the correlations between the perfusion parameters from IVIM and DCE-MRI. STUDY TYPE: Retrospective. POPULATION: Twenty patients with BEOTs and 42 patients with MEOTs. FIELD STRENGTH/SEQUENCE: 1.5T/DWI, DKI, and IVIM models fitting from 13 different b factors and 40 phases DCE-MRI. ASSESSMENT: Histogram metrics were derived from the apparent diffusion coefficient (ADC), diffusion kurtosis (K), diffusion coefficient (Dk), pure diffusion coefficient (D), pseudodiffusion coefficient (D*), perfusion fraction (f), volume transfer constant (Ktrans ), rate constant (kep ), and extravascular extracellular volume fraction (ve ). STATISTICAL TESTS: The Mann-Whitney U-test and receiver operating characteristic curve were used to determine the best histogram metrics and parameters. Multivariate logistic regression analysis was used to determine the best combined model for each two from the four technologies. Spearman's rank correlation was used to analyze the correlations between the IVIM and DCE-MRI parameters. RESULTS: ADC, D, Dk, and D* were significantly higher in BEOTs than in MEOTs (P < 0.05). K, Ktrans , kep , and ve were significantly lower in BEOTs than in MEOTs (P < 0.05). The 10th percentile of Dk was the most reliable single metric, with an area under the curve (AUC) of 0.921. Dk combined with Ktrans yielded the highest AUC of 0.950. A weak inverse correlation was found between D and Ktrans (r = -0.320, P = 0.025) and between D and kep (r = -0.267, P = 0.037). DATA CONCLUSION: The 10th percentile of Dk was the most valuable metric and Dk combined with Ktrans had the best performance for differentiating BEOTs from MEOTs. There was no evident link between perfusion-related parameters derived from IVIM and DCE-MRI. LEVEL OF EVIDENCE: 4 Technical Efficacy Stage: 2 J. Magn. Reson. Imaging 2020;52:257-268.
BACKGROUND: The accurate preoperative differentiation between borderline and malignant epithelial ovarian tumors (BEOTs vs. MEOTs) is crucial for determining the proper surgical strategy and improving the patient's postoperative quality of life. Several diffusion and perfusion MRI technologies are valuable for the differentiation; however, which is the best remains unclear. PURPOSE: To compare the whole solid-tumor volume histogram analysis of diffusion-weighted imaging (DWI), diffusion kurtosis imaging (DKI), intravoxel incoherent motion (IVIM), and dynamic contrast-enhanced MRI (DCE-MRI) in the differentiation of BEOTs vs. MEOTs and to identify the correlations between the perfusion parameters from IVIM and DCE-MRI. STUDY TYPE: Retrospective. POPULATION: Twenty patients with BEOTs and 42 patients with MEOTs. FIELD STRENGTH/SEQUENCE: 1.5T/DWI, DKI, and IVIM models fitting from 13 different b factors and 40 phases DCE-MRI. ASSESSMENT: Histogram metrics were derived from the apparent diffusion coefficient (ADC), diffusion kurtosis (K), diffusion coefficient (Dk), pure diffusion coefficient (D), pseudodiffusion coefficient (D*), perfusion fraction (f), volume transfer constant (Ktrans ), rate constant (kep ), and extravascular extracellular volume fraction (ve ). STATISTICAL TESTS: The Mann-Whitney U-test and receiver operating characteristic curve were used to determine the best histogram metrics and parameters. Multivariate logistic regression analysis was used to determine the best combined model for each two from the four technologies. Spearman's rank correlation was used to analyze the correlations between the IVIM and DCE-MRI parameters. RESULTS: ADC, D, Dk, and D* were significantly higher in BEOTs than in MEOTs (P < 0.05). K, Ktrans , kep , and ve were significantly lower in BEOTs than in MEOTs (P < 0.05). The 10th percentile of Dk was the most reliable single metric, with an area under the curve (AUC) of 0.921. Dk combined with Ktrans yielded the highest AUC of 0.950. A weak inverse correlation was found between D and Ktrans (r = -0.320, P = 0.025) and between D and kep (r = -0.267, P = 0.037). DATA CONCLUSION: The 10th percentile of Dk was the most valuable metric and Dk combined with Ktrans had the best performance for differentiating BEOTs from MEOTs. There was no evident link between perfusion-related parameters derived from IVIM and DCE-MRI. LEVEL OF EVIDENCE: 4 Technical Efficacy Stage: 2 J. Magn. Reson. Imaging 2020;52:257-268.
Authors: Dirk Timmerman; François Planchamp; Tom Bourne; Chiara Landolfo; Andreas du Bois; Luis Chiva; David Cibula; Nicole Concin; Daniela Fischerova; Wouter Froyman; Guillermo Gallardo Madueño; Birthe Lemley; Annika Loft; Liliana Mereu; Philippe Morice; Denis Querleu; Antonia Carla Testa; Ignace Vergote; Vincent Vandecaveye; Giovanni Scambia; Christina Fotopoulou Journal: Int J Gynecol Cancer Date: 2021-06-10 Impact factor: 3.437