0396

Detection of fluid/solid tissue behavior of neurotumors by magnetic resonance elastography

Kaspar Josche Streitberger¹, Ledia Lilaj¹, Felix Schrank¹, Jürgen Braun¹, Josef Käs², Karl Titus Hoffmann³, Martin Reiss-Zimmermann⁴, and Ingolf Sack⁵

¹Charité - Universitätsmedizin Berlin, Berlin, Germany, ²University Leipzig, Leipzig, Germany, ³University Hospital Leipzig, Leipzig, Germany, ⁴Neuroradiology Center, Erfurt, Germany, ⁵Radiology, Charité-Universitätsmedizin Berlin, Berlin, Germany

Synopsis

It is known that glioblastoma (GB) display high heterogeneity and porosity (higher water content) than meningioma (MEN) suggesting a higher fluidity of GB than MEN. However, we will demonstrate by MR elastography (MRE) in patients and phantoms that the concept of fluidity cannot be naively transferred to brain tumors. Instead, the macroscopic viscosity-fluidity behavior of GB and MEN can be understood MRE of materials that comprise various amounts of water including materials with fibrillary architecture such as tofu. Similar to tofu, viscosity and fluidity in GB and MEN seem to reduce with increasing water content, indicating an ‘anomalous’ viscosity-fluidity behavior.

Introduction

The biophysical properties of tumorous tissue are increasingly recognized as being important for tumor progression and malignant transformation. Elasticity and viscosity, i.e. respective solid and fluid behavior, are important classes of biophysical parameters that relate to the structural integrity of tissues and permissivity for cancer cells(1). MR elastography (MRE) applied to brain tumors demonstrated a wide variety of values ranging from fluid to solid tissue properties(2-6). A more fluid tissue characteristic favours metastatic cancer cell motility while solidity fosters cancer cell proliferation against a rigid microenvironment. A general pattern of viscoelastic changes of brain tissue in the course of tumor progression is still lacking. Therefore, we analyzed the degree of fluidity in two tumor entities, glioblastoma (GB) and meningioma (MEN), measured by in vivo MRE. It is known that GB display high heterogeneity and porosity (higher water content) than MEN suggesting a higher fluidity of GB than MEN. However, we will demonstrate that the concept of fluidity cannot be naively transferred to brain tumors since these entities gain their aggressiveness by displacing and invading healthy tissue unlike other solid tumors in which fluid cellular areas are associated with metastatic competent cells. Instead, the macroscopic viscosity-fluidity behavior of GB and MEN can be understood by MRE of phantom materials that comprise various amounts of water including materials with a fibrillary architecture such as tofu.

Methods

Eighteen patients with GB (N=9, 71±7yrs) or MEN (N=9, 57±14yrs) were investigated by standard MRI and multifrequency MRE. The MRE experimental setup and post processing routine is detailed in(3). In brief, seven stimulation frequencies from 30 to 60Hz were used to vibrate a head cradle connected to a piezoelectric driver. A stack of 15 slices of 2x2x2mm³ resolution was acquired by single-shot spin-echo EPI-MRE. Data analysis was based on multifrequency dual elasto-visco (MDEV) inversion, providing two independent parameters, the magnitude and the phase angle of the complex shear modulus (|G*| and φ) representing viscoelasticity and viscosity, respectively. Fluidity was defined as the area where fluid tissue properties dominate (φ>π/4) normalized by the total tumor area. To test the influence of extracellular water content on |G*| and φ, the same MRE protocol was applied to phantoms made of agarose and tofu blended with different amounts of water.

Results

Fig.1 shows histological slides for two representative cases of GB and MEN with demarcation of structural features indicating higher extracellular water content in GB than MEN. Consistently, T2*-weighted contrast from the magnitude MRE signal (|S*|) showed higher intensities in GB than MEN (Fig.2). MEN shows smooth boundaries while GB has an elongated less regular shape. MRE demonstrates that both entities can have very low |G*|-values comparable to normal brain but differ in their viscous properties. MEN has high φ-values, while GB is clearly less viscous, consistent to Fig.3 in which data of all cases enrolled in this study are shown. GB and MEN are well separated based on φ and fluidity but overlap in |G*|. Interestingly, fluidity is higher in MEN than GB despite the lower extracellular water content, indicating an anomalous macroscopic viscoelastic (VE) behavior. Fig.4 summarizes all cases by plotting |G*| versus φ. Note that, unlike(3,4), we here show absolute values of both parameters. Figure 4 also displays the VE-behavior in our phantoms. Adding water to agarose reduces stiffness and increases viscosity (blue). Tofu clearly shows an inverse effect, i.e. viscosity and fluidity reduce with increasing water content, motivating the term ‘anomalous’ VE behavior (red).

Discussion

The differences in the mechanical properties of GB and MEN determine the spreading behaviour of these two different tumor entities. GB have a low stiffness and viscosity, i.e. a low resistance in agreement with the high water content. MEN have a high viscosity and thus behaves more cohesive in accordance with reports by surgeons. As a matter of fact GB grow infiltrative with irregular shape, high heterogeneity and porosity, while MEN grow slowly, with more regular, encapsulated boundaries and homogenous, compact consistency. We show that the mechanical behavior of neurotumors can be understood by observations in biological materials with varying water contents such as tofu and agarose. Similar to the differences encountered in our groups of tumors, tofu changes with increasing amounts of water from a highly viscous, dissipative fluid state into a soft more solid state due to inhibition of hydrophobic protein interactions. Our study demonstrates that fluidity and viscosity are a potentially sensitive biomarker for the integrity and malignancy of tumor masses.

Acknowledgements

Support of the German Research Foundation (GRK2260, BIOQIC) is gratefully acknowledged.

References

1. Hirsch, S., Braun, J. & Sack, I. Magnetic Resonance Elastography: Physical Background And Medical Applications, (Wiley-VCH, 2017).

2. Murphy, M.C., et al. Preoperative assessment of meningioma stiffness using magnetic resonance elastography. J Neurosurg 118, 643-648 (2013).

3. Streitberger, K.-J., et al. High-resolution mechanical imaging of glioblastoma by multifrequency magnetic resonance elastography. PLoS One 9, e110588 (2014).

4. Reiss-Zimmermann, M., et al. High Resolution Imaging of Viscoelastic Properties of Intracranial Tumours by Multi-Frequency Magnetic Resonance Elastography. Clin Neuroradiol 25, 371-378 (2015).

5. Yin, Z., et al. Slip Interface Imaging Predicts Tumor-Brain Adhesion in Vestibular Schwannomas. Radiology 277, 507-517 (2015).

6. Murphy, M.C., Huston, J., 3rd & Ehman, R.L. MR elastography of the brain and its application in neurological diseases. Neuroimage (2017).

Figures

Figure 1: Histological analysis of representative cases of glioblastoma (GB) and meningioma (MEN) demonstrating hemorrhages and necrotic tissue areas in GB versus solid fibrillary structures in MEN.

Figure 2: Juxtaposition of meningioma and glioblastoma. |S*|: magnitude of the MRE scan, |G*|: stiffness, |φ|: viscosity. Red dashed lines demarcate tumor regions.

Figure 3: Group means of stiffness, viscosity and fluidity for meningioma, glioblastoma, and reference tissues (Ref). There is a trend of GB to be softer than MEN; however, the two entities are fully separated by viscosity, indicating that both GB and MEN are similarly soft (very soft compared to surrounding soft brain matter) but significantly differ in their liquid properties. The apparently high fluidity of MEN is further corroborated by the proposed fluidity marker on the right-hand side, suggesting anomalous VE-fluidity in MEN (statistics based on the Wilcoxon rank-sum test).

Figure 4: Stiffness (|G*|) versus viscosity (|φ|) plot from all cases including tumorous and normal-appearing brain tissue. Additionally, phantom data are shown for agarose (blue) and tofu (red), both blended with different amounts of water, demonstrating inverse tendencies of increasing and decreasing viscosity when water is added.

Proc. Intl. Soc. Mag. Reson. Med. 27 (2019)

0396