The Role of Simultaneous Standardized Uptake Value and MRI Diffusion Weighted Heterogeneity Index from Hybrid PET/MR in the Evaluation of Brain Metastases

Abstract

\n BackgroundPET/MRI is a hybrid imaging modality and uses for evaluating oncology patients with its benefits of combination of soft tissue contrast and glucose metabolism. AimsTo evaluate the role of simultaneous derived apparent diffusion coefficient (ADC) heterogeneity index and 2-[18F]-fluoro-2-deoxy-D-glucose (18F-FDG) standardized uptake value (SUV) from hybrid PET/MRI to differentiate brain metastasis (BM) from normal appearing brain parenchyma (NABP) and to find out the efficiency of combination of ADCCV and SUV against conventional ADC parameter. Study DesignWhole-body PET/MRI was performed to evaluate proven 49 BM of 26 oncology patients (15 females, 11 males; mean age 63±16 years), sourced from different primary cancer. MethodsBrain sequences, which were dixon and diffusion weighted imaging (DWI) protocols with simultaneous PET were used to calculate coefficient of variance of the ADC (ADCCV) and SUVmax. All images were assessed by three radiologists and the same size of VOI was placed on BM and NABP. Inter-rater reliability was tested by inter-class correlation (ICC). The correlation of ADCCV and SUVmax and the differences in ADC values and SUVmax between BM and NABP were investigated. ResultsThe excellent consistency was found between raters at ADCmean (0.972) and ADCCV (0.995). There was a moderate correlation between ADCCV and SUVmax (r=0.585) and a negligible inverse correlation between ADCmean and SUVmax (r=-0.154). A statistically significant difference between BM and NABP was determined for ADCCV (p<0.001) and SUVmax (p=0.007). An area under the curve (AUC) of 0.663, 0.966, 0.571, 0.696 and 0.971 were obtained with ROC analysis of SUVmax, ADCCV, ADCmean, SUVmax+ADCmean and SUVmax+ADCCV, respectively.ConclusionADCCV may be considered as a new parameter in multi-parametric MRI that quantitatively discriminates BM from NABP with excellent interrater reliability.