Bin Bo1, Tianyang Sheng2, Weijun Tang3, Yibo Zhao4,5, Yudu Li4,6, Wen Jin4,5, Rong Guo4,7, Xiangjun Chen2, Zhi-Pei Liang4,5, and Yao Li1
1School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China, 2Department of Neurology, Huashan Hospital, Fudan University, Shanghai, China, 3Department of Radiology, Huashan Hospital, Fudan University, Shanghai, China, 4Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana-Champaign, Urbana, IL, United States, 5Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, IL, United States, 6National Center for Supercomputing Applications, University of Illinois at Urbana-Champaign, Urbana, IL, United States, 7Siemens Medical Solutions USA, Inc., Urbana, IL, United States
Synopsis
Keywords: Multiple Sclerosis, Multiple Sclerosis
Motivation: Multiple sclerosis (MS) is characterized by diverse metabolic alterations. 1H-MRSI provides a unique capability for non-invasive mapping of neurometabolites but is often limited in resolution, scanning time, and brain coverage.
Goal(s): Our goal was to demonstrate the feasibility of high-resolution whole-brain 1H-MRSI for characterizing metabolic alterations in MS.
Approach: 3D 1H-MRSI scanning using SPICE technology (scan time: 10 minutes, resolution: 2×3×3 mm3, FOV: 240×240×120 mm3) was performed on 44 MS patients.
Results: N-acetylaspartate (NAA), myo-inositol (mI), creatine, and choline levels altered among different lesion types and peri-plaque regions. NAA and mI/NAA differentiated RRMS and PMS patients in association with clinical scores.
Impact: High-resolution whole-brain 1H-MRSI
provides a promising tool for non-invasive metabolic imaging to characterize MS
pathophysiology.
Introduction
Multiple
sclerosis (MS) is a chronic inflammatory disease
characterized by inflammation, gliosis, demyelination, remyelination, and
axonal loss, all contributing to clinical disability and disease progression 1,2. Lesion appearance on T1- and
T2-weighted MRI has limited capabilities in characterizing MS lesions and
identifying diffuse damage in areas of normal-appearing white matter (NAWM) due
to their low sensitivity and specificity 3. Proton MR spectroscopic
imaging (1H-MRSI) provides a unique capability for non-invasive
mapping of neurometabolites associated with MS pathology, including
N-acetylaspartate (NAA) as a marker for neuronal/axonal integrity, myo-inositol (mI) as a marker for astroglial activation
and gliosis, creatine (Cr) as a marker for glial
cell density, and choline (Cho) as a marker for cell
membrane turnover 4–6. However, the low spatial resolution,
long data acquisition time, and limited brain coverage have significantly hindered
the clinical translation of MRSI. In this study, we demonstrated the
feasibility and potential of fast, high-resolution (2×3×3 mm3),
whole-brain metabolic imaging of MS using a 1H-MRSI technology known
as SPICE
(SPectroscopic Imaging by exploiting spatiospectral CorrElation) at 3T 7–11.
We evaluated the metabolic alterations and investigated the underlying
pathological substrates associated with various lesion types and diffuse damage
in NAWM. Furthermore, we determined if the detected metabolic alterations were
associated with clinical outcomes.Methods
Forty-four MS patients (34 females;
mean age, 34.3±8.1 years), including 32 relapsing-remitting MS
(RRMS) and 12 progressive MS (PMS) patients, and 13 healthy controls (HCs, 8
females; mean age, 26.1±6.1 years) were enrolled in our study. Patients
underwent neurological examination with expanded disability status scale (EDSS,
range: 0-6.5) and symbol digit modalities test (SDMT, range:5-74). The study
was approved by the IRB of Huashan Hospital, Shanghai, China. All subjects
underwent MRI on
a 3T system (MAGNETOM Prisma,
Siemens Healthcare, Erlangen, Germany). The imaging
protocol included 3D-MRSI using SPICE (2.0×3.0×3.0
mm3; FOV = 240×240×120
mm3; TR/TE = 160/1.6 ms; scan time = 10:03 minutes), T1w-MP2RAGE (1.0×1.0×1.0
mm3, TR/TE/TI = 4000/2.22/1270 ms; FOV = 216×256×192
mm3), and T2w-FLAIR (0.5×0.5×1.0
mm3, TR/TE = 5000/394 ms; FOV = 256×204×160
mm3). The spatiospectral functions of the neurometabolites were reconstructed
from the 3D-MRSI data using a union-of-subspaces model, incorporating
pre-learned spectral basis functions 7-11. The spectral
quantification was performed using an improved LCModel-based algorithm 11.
White
matter lesions (WML) were segmented on co-registered FLAIR and T1w images following
previous literatures 12,13. The lesions were
classified into: (1) T1-dark: T1w black holes; (2) T1-hypointense: FLAIR
hyperintense and T1w hypointense; (3) T1-isointense: FLAIR hyperintense and T1w
isointense 14. The WML mask was dilated isotropically
by 4 mm to generate peri-plaque (PP) and NAWM regions 15. Corpus
callosum (CC), frequently affected in MS, was segmented based on the JHU-ICBM-labels
atlas intersected with the NAWM mask.
Neurometabolic
levels were compared among different regions, different lesion types, and
different subject groups using paired t-tests, mixed model analysis of
covariance (ANCOVA) correcting for lesion volume, and one-way ANCOVA correcting
for age/gender, respectively. Associations between neurometabolic levels and
clinical scores were evaluated using partial Spearman correlation controlling for
age, gender, and disease duration. Bonferroni correction was used for multiple
comparisons.Results
Figure 1 demonstrates representative
3D neurometabolite maps obtained from a healthy subject, an RRMS patient and a
PMS patient, respectively. As shown in Figure 2, NAA increased from WML to PP and
from PP to NAWM, mI, and Cho decreased from PP to NAWM, and Cr increased from
WML to PP (all P < 0.001). Relative metabolic ratios including
mI/NAA, mI/Cr, NAA/Cr, and Cho/Cr differed among three regions (P <
0.001). As shown in Figure 3, all neurometabolites in T1-dark and
T1-hypointense lesions showed lower levels than T1-isointense lesions (P
< 0.001). Lower NAA/Cr level was found in T1-dark lesion than T1-hypointense
lesion (P = 0.008). Among the three subject groups, significant
differences were found in mI/NAA of NAWM from both whole-brain and CC (P
≤ 0.001) (Fig. 4). PMS patients exhibited higher mI level in CC than RRMS
patients (P < 0.001). The patients’ mI/NAA in whole-brain NAWM was
associated with EDSS (r = 0.459, P = 0.003) and SDMT scores (r =
- 0.589, P = 0.004). Both mI and mI/NAA in CC NAWM were correlated with
patient EDSS (r = 0.417, P = 0.008; r = 0.426, P = 0.007)
and SDMT scores (r = - 0.626, P = 0.002; r = - 0.603, P = 0.004) (Fig.
5).Conclusion
We demonstrated
the feasibility of high-resolution whole-brain 3D 1H-MRSI
for characterizing metabolic alterations in MS patients. The work may lay a
foundation for a more comprehensive clinical study on using non-invasive
metabolic imaging to characterize MS pathophysiology. Acknowledgements
This work was supported by Shanghai
Pilot Program for Basic Research—Shanghai Jiao Tong University (21TQ1400203), the
Program for Professor of Special Appointment (Eastern Scholar) at Shanghai
Institutions of Higher Learning, Key Program of Multidisciplinary Cross
Research Foundation of Shanghai Jiao Tong University (YG2021ZD28, YG2023ZD22), and
New Faculty Start-up Foundation of Shanghai Jiao Tong University
(23X010501992).References
- Trapp BD, Peterson J, Ransohoff RM, et al. Axonal transection in the lesions of multiple sclerosis. New England Journal of Medicine 1998;338(5):278–285.
- Confavreux C, Vukusic S, Adeleine P. Early clinical predictors and progression of irreversible disability in multiple sclerosis: An amnesic process. Brain 2003;126(4).
- Enzinger C, Barkhof F, Ciccarelli O, et al. Nonconventional MRI and microstructural cerebral changes in multiple sclerosis. Nat Rev Neurol 2015;11(12).
- De Stefano N, Filippi M, Miller D, et al. Guidelines for using proton MR spectroscopy in multicenter clinical MS studies. Neurology 2007;69(20).
- Kirov II, Tal A, Babb JS, et al. Serial proton MR spectroscopy of gray and white matter in relapsing-remitting MS. Neurology 2013;80(1).
- Llufriu S, Kornak J, Ratiney H, et al. Magnetic resonance spectroscopy markers of disease progression in multiple sclerosis. JAMA Neurol 2014;71(7).
- Liang Z.-P. Spatiotemporal imaging with partially separable functions. In: Proc. 4th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, Arlington, VA, USA, 2007; pp. 988–91.
- Lam F, Liang Z.-P. A subspace approach to high-resolution spectroscopic imaging. Magn Reson Med 2014;71(4).
- Lam F, Ma C, Clifford B, Johnson CL, Liang Z.-P. High-resolution 1H-MRSI of the brain using SPICE: Data acquisition and image reconstruction. Magn Reson Med. 2016;76(4):1059-1070.
- Ma C, Lam F, Johnson CL, Liang Z.-P. Removal of nuisance signals from limited and sparse 1H MRSI data using a union-of-subspaces model. Magn Reson Med. 2016;75(2):488-497.
- Li Y, Lam F, Clifford B, Liang Z.-P. A subspace approach to spectral quantification for MR spectroscopic imaging. IEEE Trans Biomed Eng. 2017;64(10):2486-2489.
- Valverde S, Oliver A, Roura E, et al. Automated tissue segmentation of MR brain images in the presence of white matter lesions. Med Image Anal 2017;35.
- Meier DS, Guttmann CRG, Tummala S, et al. Dual-sensitivity multiple sclerosis lesion and CSF segmentation for multichannel 3T brain MRI. Journal of Neuroimaging 2018;28(1).
- Schiavi S, Petracca M, Sun P, et al. Non-invasive quantification of inflammation, axonal and myelin injury in multiple sclerosis. Brain 2021;144(1).
- Rahmanzadeh R, Lu PJ, Barakovic M, et al. Myelin and axon pathology in multiple sclerosis assessed by myelin water and multi-shell diffusion imaging. Brain 2021;144(6).