0586
High-frequency electrical conductivity imaging of rat brain tumor1Centre for Preclinical Imaging, University of Liverpool, Liverpool, United Kingdom, 2Department of Biomedical Engineering, Kyung Hee University, Seoul, Korea, Republic of, 3School of Engineering, Macquarie University, Sydney, Australia
Synopsis
Magnetic Resonance Electrical Property Tomography (MREPT) allowed the evaluation of the intra- and extracellular ionic changes in a rat model of brain tumor. A multi-echo T2-weighted pulse sequence was used with 10 echos to map the B1 field and reconstruct the tissue conductivity at 9.4T. Higher conductivity values were measured in the tumors compared to the healthy brain tissue, suggesting an increased ionic content and mobility in the tumor.
Introduction
Magnetic Resonance Electrical Properties Tomography (MREPT) has recently been described as an MRI technique to measure tissue conductivity and permittivity, which is primarily derived from the magnitude phase and phase magnitude information of B1 maps1,2. Tissue conductivity maps provide novel contrast mechanism and reflect changes in ionic content. Although MREPT at low fields (3T) has been reported in breast tumors3, its application at high fields for assessing preclinical tumor models has not been reported. The purpose of this study was to compare the utility of ionic conductivity measured with MREPT with mean diffusivity values in the characterization of an intracranial model of rat gliomas.Methods
Animal model: Brain tumors were induced by transcranial injection of F98 cells in the right cortex. Six F344 female (100-120 g) rats were injected with 50,000 F98 cells suspended in 5 µL PBS solution.
MRI: Longitudinal MRI, including DTI and MREPT was performed on day 8, 11 and 14 after tumor cell implantations. In vivo conductivity data were acquired using an MSME sequence using: TR = 4341 ms, echo-spacing = 8 ms, 10 echos, 2 averages, FOV = 40x20 mm, matrix = 128x64, slice thickness = 0.3 mm, resolution = 0.313x0.313x0.3 mm, scan duration = 9 min 15 s. In vivo DTI datasets were acquired using an EPI-DTI sequence with the parameters: TE/TR = 23/2500 ms, 5 averages, b-values = 0,1000,2000 s/mm2, 15 directions, respiratory triggered scan duration ≈ 55 min.
Image processing: Mean diffusivity (MD) maps were calculated from DTI data using DKE (Medical University of South Carolina, USA). Conductivity (σH) maps were reconstructed from the B1 phase maps acquired using the MSME sequence4. Mean diffusivity maps and conductivity maps were co-registered with T2-weighted images and median values were extracted from the tumor center, tumor rim, peritumoral edema, contralateral cortex and ventricles for CSF analysis.
Results
Figure 1 shows T2-weighted images (top row) covering the tumor from a representative rat, with their corresponding conductivity (middle) and mean diffusivity (bottom) maps. The conductivity was higher in the ventricles and the tumor compared to the contralateral healthy brain (Figure 2). The central necrotic core of the tumor demonstrated particularly higher conductivity compared to the viable tumoral rim.
The conductivity measurements were compared to the mean diffusivity in the same volumes-of-interest longitudinally 8, 11, and 14 days after tumor cell injection (Figure 2). A signed-rank Wilcoxon test indicated a significant increase in conductivity in the tumor (rim, center and edema) compared to the contralateral cortex. However, no significant difference was observed between the tumor and the edema, when the data was analyzed longitudinally. For comparison, MD was significantly higher in the tumor compared to the contralateral cortex, and also and significantly higher in the edematous region compared to the tumor rim at all time-points.
Data from all time points was combined to see if the conductivity and MD values differentiated the various tissue types regardless of tumor size (Figure 3). The signed-rank Wilcoxon test demonstrated significantly higher conductivity in the tumoral rim (mean = 0.81±0.18 S/m) compared to the contralateral cortex (mean = 0.43±0.05 S/m), furthermore, a significantly lower conductivity was also observed in the edema (mean = 0.66±0.14 S/m) compared to the tumor rim.
Discussion
Higher ionic conductivity in the ventricles and edema is probably due to the higher fluid content which was confirmed by the high mean diffusivity values. The higher conductivity in the tumor rim and center suggests increased ionic conductivity concentration and mobility in the tumor compared to the normal brain, which is also reported in Na-MRI of brain tumors indicating an imbalance in the Na+/K+ pumps5,6. Higher MD values in the tumor probably reflect increased water molecule mobility in the increased extracellular space.
When all data was combined, tumor edema had a significantly higher conductivity than the contralateral cortex but was lower than the tumor, which may provide a way to differentiate the tumor from edema, especially in non-enhancing tumors, where it is difficult to estimate the tumor margins.
Conclusion
Measurement of ionic conductivity from a relatively fast and simple multi-echo T2-weighted sequence provides a novel contrast, which can be implemented on any scanner with acceptable data acquisition times. The information may be complementary to other contrast mechanisms including mean diffusivity and may aid in better characterization of brain tumors.Acknowledgements
This work was partially funded by the MRC-KHIDI UK-Korea Partnering award: MC_PC_17108 (Poptani). Imaging data were obtained at the Centre for Preclinical Imaging (CPI) of the University of Liverpool.
References
1. Katscher U, Berg CAT. Electric properties tomography: Biochemical, physical and technical background, evaluation and clinical applications. NMR in Biomedicine 2017;30(8):e3729.
2. Minhas AS, Kim YT, Kim HJ, Woo EJ, Kim M, Kim D, Seo JK. Feasibility of dual-frequency conductivity imaging using MREIT and MREPT. 2011 13-16 May 2011. p 68-71.
3. Kim S-Y, Shin J, Kim D-H, Kim MJ, Kim E-K, Moon HJ, Yoon JH. Correlation between conductivity and prognostic factors in invasive breast cancer using magnetic resonance electric properties tomography (MREPT). European Radiology 2016;26(7):2317-2326.
4. Gurler N, Ider YZ. Gradient-based electrical conductivity imaging using MR phase. Magnetic Resonance in Medicine 2016;77(1):137-150.
5. Thulborn KR, Davis D, Adams H, Gindin T, Zhou J. Quantitative tissue sodium concentration mapping of the growth of focal cerebral tumors with sodium magnetic resonance imaging. Magnetic Resonance in Medicine 1999;41(2):351-359.
6. Nielles-Vallespin S, Weber M-A, Bock M, Bongers A, Speier P, Combs SE, Wöhrle J, Lehmann-Horn F, Essig M, Schad LR. 3D radial projection technique with ultrashort echo times for sodium MRI: Clinical applications in human brain and skeletal muscle. Magnetic Resonance in Medicine 2006;57(1):74-81.
Figures
