Steven Kwok Keung Chow1, Angela Walls1, Andrew Dwyer1, Kieran O’Brien2, Stephanie Withey2, Nicole Dmochowska3, Aidan Cousins4, and Benjamin Thierry3
1Clinical Research and Imaging Centre, South Australian Health and Medical Research Institute, Adelaide, Australia, 2Siemens Healthcare Pty Ltd., Adelaide, Australia, 3Future Industries Institute, University of South Australia, Adelaide, Australia, 4Ferronova Pty Ltd., Adelaide, Australia
Synopsis
Keywords: Preclinical Image Analysis, Quantitative Susceptibility mapping, QSM , SPION, Iron concentration
Motivation: Progressing the clinical translation of QSM is important for imaging of biometal dysregulation in neurodegenerative disease but available methods for calibrating QSM sequences between systems are limited.
Goal(s): To develop and validate a superparamagnetic iron oxide nanoparticle (SPION) phantom replicating human brain iron concentration.
Approach: Dilutions of Ferrotrace SPION (0.1-25 $$$\mu$$$g/mL) were scanned with research application QSM gradient echo sequence using two different 3T systems and multiple analysis methods.
Results: A good linear fit was demonstrated between QSM values and SPION concentration across a clinically relevant interval and different QSM analysis methods. Values differed between two scanners but there was high within-scanner concordance.
Impact: A SPION based phantom replicating in vivo iron of healthy and diseased brain could be an invaluable calibration and QA tool for QSM clinical translation, normative datasets and emerging SPION-based theranostics for brain cancer.
Introduction
Quantitative Susceptibility
Mapping (QSM) is increasing being translated into clinical practice1 across
a range of vendors, field and gradient strengths. Dysregulation of biometals is
manifested in a variety of pathologic processes and increasingly implicated in
neurodegenerative disease where brain iron content can increase up 5 times compared
to healthy controls2. QSM requires attention to several methodologic
issues but limited resources are available for clinical sites to reference
human brain iron across different systems3. Superparamagnetic iron
oxide nanoparticles (SPIONs) are stable and dilutable indicating the potential to
be established as an iron brain phantom4. Several SPION-based
tracers are also emerging candidates for brain cancer theranostics and reliable
in vivo quantification will be an important step in new biomarkers for
tumour imaging. This study investigates the quantification of SPIONs mimicking iron
concentration in the human brain for QSM MRI sequence validation in two
scanners. Material and Methods
A QSM phantom was made using SPION (FerroTrace;
Ferronova Pty Ltd, Adelaide, SA, Australia). The FerroTrace was diluted with phosphate-buffered
saline (PBS) solution into 0.1, 0.25, 0.5, 0.75, 1.0, 2.5, 5, 10, 25
$$$\mu$$$g/mL concentrations
and a PBS solution as a control. The phantom was scanned on a MAGNETOM Skyra
3T (software version VE11C; Siemens Healthcare; Erlangen, Germany) and MAGNETOM
Cima.X 3T (software version XA61A; Siemens Healthcare; Erlangen, Germany) using
a 64-channel head/neck coil. Repeatability test-retest scans were also acquired
one week after the original scan. Research application QSM gradient echo
sequence parameters were TEs = 6.32, 10.63, 5.42, 20.21, 25 ms; TR = 29 ms;
bandwidth = 310 Hz/Px; flip angle = 12º; matrix size = 224 × 180; number of
slices = 120; voxel size = 1×1×1 mm; and scan time = 290s. The SEPIA5
pipeline control tool, MATLAB (R2021a, The MathWorks, USA), was used to
post-process the QSM data. QSM phase images used Projection onto Dipole Field
(PDF)6 as background field removal and Laplacian (MEDI)7 phase unwrapping, Morphology
Enabled Dipole Inversion (MEDI)8 and Thresholded k-space Division
(TKD)9. QSM maps were quantified by manually drawing the ROIs using
ImageJ (1.46r software, NIH) and the data were averaged across three adjacent
slices. Statistical analysis was performed using a non-parametric Mann-Whitney
test.Results
Within the range of 0.1
to 1.0 $$$\mu$$$g/mL of FerroTrace there was
a good linear correlation between QSM signal and particle concentration. This interval
encompasses QSM values seen in health and early neurodegenerative disease. Higher
particle concentrations tested up to 25 $$$\mu$$$
g/mL showed a mild
saturation effect causing a non-linear component. QSM values up to 1.0 $$$\mu$$$
g/mL from each scanner
are shown in Figure 1. The MEDI method (Figure 1a) and TKD method (Figure 1b)
show good concordance. There was no significant within-scanner difference on
repeated measurement (p < 0.005). Full range data using the MEDI method is
shown in Figure 2.Discussion
The concordance of QSM result across analysis
methods and repeated measurement with a good linear relationship to SPION concentration
within a clinically relevant interval indicates a SPION based phantom mimicking
iron concentration of the healthy and diseased brain is feasible. QSM is
impacted by many imaging factors including field strength, gradients, geometry,
homogeneity and sequencing evident in difference obtained with the two scanners
tested. A phantom-based validation
curve may calibrate QSM results between scanners, particularly where data is
referenced to normative ranges, or where research is conducted across multisite
studies or longitudinal timepoints. The relationship between QSM and true
concentration of iron species in the human brain is complex and susceptibility of
SPIONs is higher than iron states in vivo1,10. As a result, dilute
SPIONs concentrations correlate to QSM values likely to be encountered in
practice. This study focuses on commonly used phase and QSM calculation methods.
Further investigation is necessary to evaluate different vendors, field
strengths and long-term reproducibility metrics before suitable to guide human
brain studies. Conclusion
A SPION based phantom demonstrated good linear
fit in a clinically relevant interval with concordant results across analysis
method and within-scanner repeated measurement. Differences between scanners
can be calibrated. By mimicking the susceptibility of human brain iron, a SPION
phantom could function as an invaluable quality assurance tool, improving the
accuracy and reliability of QSM for neurodegenerative applications and emerging
SPION-based theranostics.Acknowledgements
No acknowledgement found.References
- Wang
Y, Spincemaille P, Liu Z, et al. Clinical quantitative susceptibility mapping
(QSM): Biometal imaging and its emerging roles in patient care. J Magn Reson
Imaging. 2017;46(4):951-971.
- Kumar,
P , Bulk M, Webb A, et al. A novel approach to quantify different iron forms in
ex-vivo human brain tissue. Sci. Rep. 2016;6(1):38916.
- De
Barros A, Arribarat G, Combis J, et al. Matching ex vivo MRI With Iron
Histology: Pearls and Pitfalls. Front Neuroanat. 2019;13:68.
- Deh K, Zaman M, Vedvyas Y, et al. Validation of MRI
quantitative susceptibility mapping of superparamagnetic iron oxide
nanoparticles for hyperthermia applications in live subjects. Sci Rep.
2020;10(1):1171.
- Chan
K, Marques J. SEPIA - Susceptibility mapping pipeline tool for phase images. NeuroImage.
2021;227(February 2021):117611.
- Liu T, Khalidov I, de Rochefort L, et al. A novel background field
removal method for MRI using projection onto dipole fields (PDF). NMR Biomed.2011;24(9):1129-1136.
- Li
W, Wu B, and Liu C, Quantitative susceptibility mapping of human brain reflects
spatial variation in tissue composition, NeuroImage. 2011;55(4):1645-1656.
- Liu J, Liu T, Rochefort L, et al. Morphology enabled
dipole inversion for quantitative susceptibility mapping using structural
consistency between the magnitude image and the susceptibility map. NeuroImage.
2012;59(3):2560-2568.
- Wharton
S, Schäfer A, and Bowtell R. Susceptibility mapping in the human brain using
threshold-based k-space division. Magn. Reson. Med. 2010;63(5):1292-1304.
- Lide
D R. Handbook of Chemistry and Physics. 81st edition, 2004. Boca
Raton, FL: CRC Press; 2004.