2963
Towards clinical translation of single-subject characterization of T1 changes to capture the extent of focal tissue damage in multiple sclerosis1Institute of Diagnostic and Interventional Neuroradiology, University Hospital Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany, 2Advanced Clinical Imaging Technology, Siemens Healthineers International AG, Lausanne, Switzerland, 3CIBM Center for Biomedical Imaging, Geneva, Switzerland, 4Department of Radiology, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland, 5LTS5, École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland, 6Center of Clinical Neuroscience, Department of Neurology, University Hospital Carl Gustav Carus,Technische Universität Dresden, Dresden, Germany
Synopsis
Keywords: Multiple Sclerosis, Quantitative Imaging, White matter abnormalities
Motivation: Multiple sclerosis involves both focal and diffuse tissue damage, necessitating rapid MRI methods that can be easily incorporated into clinical routine and can capture disease-specific microstructural changes.
Goal(s):
The purpose of this study was to capture the magnitude of qT1 changes within lesions.
Approach:
Using MP2RAGE, qT1 changes within lesions could be measured as T1 z-scores.
Results: A total of n=3511 individual lesions were examined, of which 49.5% had a mean z-score greater than 2, reflecting deficient tissue integrity. The deficient volume fraction, reflecting the damage within the whole white matter, can be used to characterize the tissue destruction burden of MS patients.
Impact: T1 mapping using a rapid MP2RAGE sequence can be incorporated into clinical practice to determine the severity of damage in multiple sclerosis lesions, particularly important for detecting disease severity of progressive forms of multiple sclerosis.
Introduction
Conventional MRI provides important information for diagnosis and prognosis of patients with multiple sclerosis (MS) and for monitoring efficacy of treatments.1 White matter lesions (WML) show pathologically heterogeneous patterns of demyelination2 and a wide range of altered T1 values3. Quantifying damage to CNS tissue in MS patients could improve understanding of how the disease develops and progresses. Quantitative MRI (qMRI) could provide more sensitive measures of pathology to monitor therapeutic neuroprotective interventions. QMRI techniques have not yet entered clinical routine due to limited availability of sufficiently rapid acquisition techniques. One of the methods addressing this fact is the compressed sensing magnetization prepared 2 rapid acquisition gradient echoes (MP2RAGE) sequence4,5, providing rapid 3D T1 mapping. Alterations of qT1 reflect tissue demyelination, axonal loss, edema, and chronic inflammation that are hallmarks of disease-related pathology.The goal of this study was to introduce a clinically interpretable marker of tissue destruction: the proposed marker aims at capturing the different degrees of qT1 change within lesions and normal-appearing white matter (NAWM).
Methods
Demographics and MR protocoln=47 MS patients (Table 1) were scanned at 3T (Siemens MAGNETOM Prisma). Conventional T1-weighted MP-RAGE and 3D T2-weighted FLAIR images and a compressed sensing MP2RAGE research application sequence, which allows whole-brain 1mm isotropic T1 mapping in 3:20 minutes, were acquired (Table 2).
Image processing and statistics
The research application Morphobox6 was used to derive segmentation masks of brain tissues from MPRAGE images, and MS lesions were segmented on FLAIR images using a deep neural network trained on weak labels7 and fine-tuned on manual annotations. A mask for NAWM was generated by removing segmented lesions from the WM mask.
Atlases of reference T1 values for healthy brain tissues were established following 8,9. Voxel-based assessment of T1 alterations were estimated by means of z-scores, i.e., number of standard deviations away from age-/sex-matched reference mean values. Average absolute T1-z-scores were computed in NAWM and lesions. Voxels with a z-score of 2 or higher were considered deficient. The ratio of deficient voxel in the investigated tissue (i.e., NAWM or lesions) was defined as deficient volume fraction (DVF)10
Finally, Spearman correlation coefficients were calculated between T1-z-scores (NAWM or lesions) and EDSS, using a significance level of 0.05.
Results
Different lesion morphological patterns using T1-z-scores are displayed in Figure 1. T1-z-scores allow to detect very different degrees of tissue change as opposed to signal intensity in the appearance of the conventional FLAIR lesions. They allow evaluation of diffuse tissue changes in the NAWM. Figure 1 shows that individual lesion z-scores can be highly variable between lesions. The z-score distribution within lesions was either homogeneous (Figure 1A) or heterogeneous with higher z-score values in the center (Figure 1B). Figure 2 shows the distribution of individual lesion T1-z scores.. The mean z-score in WML was 2.5 (SD 2.4). Interestingly, 34% of WMLs had a z-score less than 1, suggesting that for this proportion of lesions the T1 is within the range of the normative healthy controls. The distribution of z-scores within lesions allowed comparison of variability within and between patients.For the entire cohort, mean DVF in the NAWM was 5.6% (SD 5.9%), ranging from 0.4% to 24.7% and mean DVF in WML was 14.4% (SD 16.8%), ranging from 0.5% to 67.0%. The correlation between DVF in the NAWM and WML tissue compartments with EDSS was (r = 0.24, p = 0.11) vs. (r = 0.32, p = 0.027), as shown in Figure 3.
Discussion and Conclusion
The rapid acquisition of CS-MP2RAGE permits T1 mapping to be incorporated into clinical imaging and allows the detection of NAWM damage and severity of focal lesion damage. We identified patterns regarding extent and heterogeneity of damage within and around the lesion. 49.5% of the lesions had a mean T1-z-score greater than 2, reflecting detectable tissue damage. In contrast, tissue integrity was preserved in 34% of lesions with no or little T1 change. Capturing individual lesion T1-z-scores provides insights into the intra- and inter-patient variability of tissue destruction. T1 mapping allows the definition of a marker of tissue destruction burden (i.e., DVF). The inter-individual DVF range reflects highly variable WM involvement among patients. The weak association of DVF in WML with the EDSS score may be explained by the small sample size and unbalanced number of patients with high EDSS score in our cohort. Cutoff z-scores defining evident tissue destruction are not yet established but are critical for clinical application11. DVF could be an easily interpretable and comparable score to quantify tissue damage across patients.Acknowledgements
No acknowledgement found.References
1. Rovira À, Wattjes MP, Tintoré M, et al. MAGNIMS consensus guidelines on the use of MRI in multiple sclerosis—clinical implementation in the diagnostic process. Nat Rev Neurol. 2015;11(8):471-482. doi:10.1038/nrneurol.2015.106
2. Metz I, Weigand SD, Popescu BFG, et al. Pathologic heterogeneity persists in early active multiple sclerosis lesions. Annals of Neurology. 2014;75(5):728-738. doi:10.1002/ana.24163
3. Vrenken H, Geurts JJG, Knol DL, et al. Whole-Brain T1 Mapping in Multiple Sclerosis: Global Changes of Normal-appearing Gray and White Matter. Radiology. 2006;240(3):811-820. doi:10.1148/radiol.2403050569
4. Marques JP, Kober T, Krueger G, van der Zwaag W, Van de Moortele PF, Gruetter R. MP2RAGE, a self bias-field corrected sequence for improved segmentation and T1-mapping at high field. NeuroImage. 2010;49(2):1271-1281. doi:10.1016/j.neuroimage.2009.10.002
5. Mussard E, Hilbert T, Forman C, Meuli R, Thiran J, Kober T. Accelerated MP2RAGE imaging using Cartesian phyllotaxis readout and compressed sensing reconstruction. Magnetic Resonance in Med. 2020;84(4):1881-1894. doi:10.1002/mrm.28244
6. Schmitter D, Roche A, Maréchal B, et al. An evaluation of volume-based morphometry for prediction of mild cognitive impairment and Alzheimer’s disease. NeuroImage: Clinical. 2015;7:7-17. doi:10.1016/j.nicl.2014.11.001
7. Fartaria MJ, Bonnier G, Roche A, et al. Automated detection of white matter and cortical lesions in early stages of multiple sclerosis. Magnetic Resonance Imaging. 2016;43(6):1445-1454. doi:10.1002/jmri.25095
8. Piredda GF, Hilbert T, Granziera C, et al. Quantitative brain relaxation atlases for personalized detection and characterization of brain pathology. Magnetic Resonance in Med. 2020;83(1):337-351. doi:10.1002/mrm.27927
9. Caneschi S. A normative cortical T1 atlas for single-subject pathology detection at 7T. Oral presented at: Advances in Quantitative Neuroimaging; June 5, 2023; Toronto, Canada. https://cds.ismrm.org/protected/23MPresentations/abstracts/0274.html
10. Kitzler HH, Su J, Zeineh M, et al. Deficient MWF mapping in multiple sclerosis using 3D whole-brain multi-component relaxation MRI. NeuroImage. 2012;59(3):2670-2677. doi:10.1016/j.neuroimage.2011.08.052
11. Granziera C, Wuerfel J, Barkhof F, et al. Quantitative magnetic resonance imaging towards clinical application in multiple sclerosis. Brain. 2021;144(5):1296-1311. doi:10.1093/brain/awab029
Figures
Figure 2: Distribution of individual lesion z-scores among patients. The median of the whole cohort is indicated by a blue line with dashed lines marking the quartiles. Patients are sorted based on the median z-score of their lesions.
Figure 3: Correlation-plots of A) DVF in WML with EDSS and B) DVF in NAWM with EDSS. The MS courses are color-coded.
Table 1: Patients‘ demographics.
Table 2: MRI sequence parameters.