The Hospital for Sick Children
Neurology

Author Of 2 Presentations

Pediatric MS Oral Presentation

PS07.04 - Fibre-specific white matter differences in children with pediatric acquired demyelinating syndromes compared to healthy children

Speakers
Presentation Number
PS07.04
Presentation Topic
Pediatric MS
Lecture Time
13:27 - 13:39

Abstract

Background

White matter (WM) microstructural changes occur in youth with multiple sclerosis (MS) and myelin oligodendrocyte glyoprotein (MOG)-associated disorders. While diffusion tensor imaging has been extensively used to characterize white matter, this method lacks microstructural and pathological specificity. ‘Fixel Based Analysis’ (FBA) statistically estimates changes in diffusion MRI connectivity that is specific to micro and macro-structure. WM damage that leads to less densely packed axons in a fiber bundle causes a decrease in fibre density (FD). If the number of axons is not reduced but occupies less area, then fibre cross-section (FC) will decrease. Last, if the density of axons within a fibre bundle and the area the bundle occupies are reduced, then fibre density and cross-section (FDC) will decrease.

Objectives

To use whole-brain FBA to measure differences in FD, FC, FDC in youth with demyelinating syndromes compared to healthy controls.

Methods

We evaluated group differences in the FBA metrics between 28 typically developing children (17F; age 15.0±2.6y), 19 children with MS (13F; 16.9±1.1y; disease duration (DD)=0.1-11.7y; expanded disability status scale(EDSS):median=1.5,range=0-4.5), and 11 children with MOG (8F;12.1±2.8y; DD=0.5-6.4y;EDSS:m=1.0,r=0-3). Multi-shell diffusion-weighted imaging of the brain was acquired with echo planar imaging on a 3T MRI scanner and was pre-processed to correct for distortions and movement. Whole-brain group FBA was performed on FD, FC and FDC to test differences between groups adjusting for age, sex, total intracranial volume, EDSS and DD (p<0.05, family-wise error (FWE) corrected).

Results

Participants with MS and MOG showed reduced FD, FC and FDC relative to typically developing children (FWE corrected p<0.05). Differences in FD were found within splenium, superior longitudinal fasciculus and optic radiations. MS patients had reduced FDC within the corticospinal tract and cerebellar peduncle compared to MOG patients. In participants with MS and MOG, decreased FD within the brain stem, cerebellar peduncles and corona radiata was associated with increased DD and EDSS.

Conclusions

Our preliminary findings showed that patients with demyelinating disorders display decreased axonal density and fibre bundle size in multiple WM tracts relative to typically developing children, which were related to clinical outcomes (EDSS, DD). These changes were more pronounced in MS compared to MOG participants in selected WM tracts.

Collapse
Machine Learning/Network Science Oral Presentation

PS16.03 - Use of machine learning classifiers based on structural and functional visual metrics to predict diagnosis in children with acquired demyelination.

Speakers
Presentation Number
PS16.03
Presentation Topic
Machine Learning/Network Science
Lecture Time
13:15 - 13:27

Abstract

Background

Predicting diagnosis in youth at the first episode of demyelination is feasible in some but not all cases. Machine learning classifiers (MLC) can be trained to identify relationships between numerous multimodal input features and disease classifications to provide highly accurate predictions.

Objectives

To assess performance of machine learning classifiers for early disease diagnosis based on visual metrics in youth with demyelination.

Methods

Standardized clinical and visual data was prospectively collected at disease onset from 141 pediatric subjects with acquired demyelinating syndromes (ADS) and 75 healthy controls (HC). Participants were recruited through The Hospital for Sick Children (Toronto, Ontario (2010-2020)) and University of Calgary (2010-2017). Patients were classified using consensus definitions of demyelinating disorders and serum antibody testing for myelin oligodendrocyte glycoprotein (MOG) and aquaporin 4 (AQP4). Twenty-two auto-segmented Optical Coherence Tomography (OCT) features, 4 functional visual and 4 clinical features were used in a stratified manner alone or in combination to identify which combination of features provided the highest predictive accuracy. These input features were analyzed using 9 supervised MLC (Random Forest (RF), AdaBoost, XGBoost, Decision Tree (DT), Logistic Regression, Support Vector Machines (SVM), k-Nearest Neighbors, Stochastic Gradient Descent, Multinomial Naive Bayes). Data was split 80/20 between training and test sets. Backward feature selection was performed to re-run the analysis with best scoring predictor features in the MLC with highest predictive accuracy.

Results

AdaBoost, SVM, and DT were the best performing MLC with a test set accuracy between 82-88% in distinguishing between ADS and HC eyes. Multiple sclerosis (MS) was distinguished from HC with 92% accuracy. In descending order, fovea thickness, inferotemporal ganglion cell layer (GCL) thickness, low contrast visual acuity, outer inferior macular thickness, temporal peripapillary retinal nerve fiber layer and superior GCL thicknesses were the most important contributors to disease classification.

Conclusions

MLC can be used to combine visual metrics and clinical parameters to distinguish ADS from HC, and to predict MS. In addition to commonly used clinical metrics, we identified other structural and functional metrics that contribute importantly to classification. Among the machine learning algorithms tested, AdaBoost, SVM and DT performed best for this model.

Collapse

Author Of 4 Presentations

Imaging Poster Presentation

P0546 - Axonal and myelin volume fractions and imaging g-ratio in pediatric MS and MOG-associated disorders. (ID 1520)

Abstract

Background

Previous studies have described extensive microstructural brain tissue abnormalities in pediatric MS patients. However, available techniques do not distinguish the extent to which such abnormalities are due to axonal loss or demyelination. Further, little is known about microstructural brain tissue changes in MOG-associated disorders (MOGad).

Objectives

To apply a combined analysis of magnetization transfer saturation (MTsat) and multi-shell diffusion-weighted imaging (DWI) with computation of myelin and axonal volume fractions (MVF and AVF) and imaging g-ratio (the ratio between inner and outer diameter of the myelin sheath); to investigate the specific relationship between these metrics in the corpus callosum (CC) and within brain white matter lesions (WML) of pediatric MS and MOGad.

Methods

We acquired standardized 3T brain MRI in 26 healthy controls (HC) (58% females (F), mean age [years (y) (range)] 15y (9-19)); 16 MS (69% F, 17y (14-18), disease duration (DD) 3y (1-7), time from last relapse (TLR) 2y (0-6)); and 11 MOGad (72% F, 12y (8-18), DD 3y (0-6), TLR 1y (0-3), 8/11 relapsing). WML and CC were segmented according to establishes procedures. DWI processing was performed with FSL and DMIPy; MTsat, MVF, AVF, and g-ratio were computed using the Jargon data management system. We used general linear models to model average MVF, AVF, and g-ratio in the CC and WML of each group, including the factors age, DD, and the interaction term group*DD. Models including sex were tested, and all exhibited lower AIC.

Results

Relative to HC, MS showed decreased CC MVF (-0.018/y, p=0.0304) and AVF (-0.0069/y; p=0.053) and corresponding increased CC g-ratio (0.0084/y, p=0.059) with increased DD. Relative to HC, MOGad showed decreased CC MVF (-0.017/y, p=0.0304) and AVF (-0.0081/y, p=0.014) with increased DD, without significant CC g-ratio changes. Both MS and MOGad showed decreased average WML MVF compared to HC WM (-0.19, p<10-8; and -0.2, p<10-8). MOGad also showed decreased average WML AVF (-0.067, p=0.0048) compared to HC. Average WML g-ratio was higher in MS than MOGad (0.17, p=0.0102), but not significantly different from HC in either group. WML MVF, AVF, and g-ratio did not change significantly with DD in MS or MOGad compared to HC.

Conclusions

Both pediatric MS and MOGad exhibited MRI correlates of axonal loss and demyelination in the CC and WML. Our measures of axonal loss in MOGad reinforces recent work warning of potentially long-term impacts on the brain from non-MS demyelinating pathologies.

Collapse
Imaging Poster Presentation

P0607 - MRI Characterization of Damage in and Around Lesions in Pediatric MS and MOG-Associated Disorders (ID 1847)

Abstract

Background

Multiple sclerosis (MS) and MOG-associated disorders (MOGad) are characterized by hyperintense white matter (WM) lesions on T2/FLAIR MRI. Conventional imaging is sensitive but does not inform on the specific pathological substrate. Magnetization transfer saturation provides a good myelin measure, and multishell diffusion is sensitive to the axon + myelin assembly. Together, these can be modelled to estimate myelin volume fraction (MVF), axonal volume fraction (AVF) and imaging g-ratio.

Objectives

To quantify gradients of damage to axons and myelin in lesions and surrouding normal appearing white matter, in pediatric MS and MOGad.

Methods

15 MS [67% females (F), mean (range) age [years (y)]: 17y (14-18), disease duration (DD) 3y (0-6), time from last relapse (TLR) 2y (0-6)] and 7 MOGad [86% F, 13y (8-18), DD 3y (0-6), TLR 1y (0-3), 6/7 relapsing] participants received 3T brain MRI. MVF, AVF and g-ratio were computed according to established procedures. T2 lesions were segmented according to standardized pipelines and WM masks by multi-atlas segmentation. Euclidean distance transforms labelled voxels in normal-appearing WM with the distance to the nearest lesion voxel, and voxels inside lesions with the distance to the nearest non-lesional WM voxel. Mean MVF, AVF and g-ratio were computed on each isodistant surface. Data were modeled using linear mixed models with distance, diagnosis, and their interaction. Knots were used at 0 and 2mm distance.

Results

MVF decreased towards the center of lesions (MOGad: -0.03/mm; MS: -0.05/mm; p values (ps)<0.002; difference n.s.) as did AVF (MOGad: -0.03/mm; MS: -0.01/mm; ps<0.0002; difference p=0.02); this graded damage extended to 2mm outside lesions. Beyond this, AVF continued to increase (MOGad: 0.001/mm; MS: 0.0003/mm; ps<10-6; difference p<10-6). Inside lesions, g-ratio increased towards the center in MS (0.03/mm, p<10-6) and decreased in MOGad (p=0.15; MOGad-MS difference p<10-4). G-ratio rose with distance outside lesions (MOGad: 0.001/mm; MS: 0.0004/mm; ps<10-4; difference p<10-5). AVF and g-ratio were similar between groups (within 2%) at 20mm from lesions; MVF was higher in MS (14%, p=0.08).

Conclusions

MS and MOGad showed myelin and axonal loss of decreasing severity with distance from lesion center, and this damage extended outside visible lesions. However, MOGad exhibited more severe axonal loss within and near lesions. The corresponding decreasing g-ratio relative to MS may indicate preferential loss of small axons in MS, or relatively better remyelination in MOGad.

Collapse
Pediatric MS Poster Presentation

P1077 - Evaluating visual evoked cortical magnetic responses in youth with demyelination (ID 1537)

Speakers
Presentation Number
P1077
Presentation Topic
Pediatric MS

Abstract

Background

Traditional fundoscopy and structural studies using optical coherence tomography (OCT) suggest that youth with MS have an injury to the temporal quadrant of the retinal nerve fibre layer (RNFL). However, full-field evoked responses may not adequately capture focal injury in different quadrants of the retina, hence missing important information about distinctive disease processes associated with different forms of acquired demyelinating syndromes (ADS). The utility of hemifield evoked responses in the evaluation of youth with ADS/MS is unknown.

Objectives

To evaluate amplitude and latency differences in evoked responses between youth with ADS, MS and healthy control (HC) using hemifield assessments.

Methods

This cross-sectional study included youth with ADS (n=22, MS=11, and Myelin Oligodendrocyte Glycoprotein antibody-related disorders (MOG)=11) and 21 age/sex-matched HC. Evaluations were performed at a minimum of 6 months after a known episode of optic neuritis (ON). Binocular evoked responses to a reversing checkerboard stimulus in each hemifield (nasal and temporal) were recorded in a whole-head 151 channel CTF magnetoencephalography system (VSM MedTech, Vancouver, CA) following a standardized protocol. 1,2,3 Amplitude and latency were then identified using source reconstructed visual evoked responses (MATLAB 8.0, The MathWorks and BrainWave v 3.2)4,5,6. Binomial regression was performed with hemifield as an interaction term to predict disease classification. GEE was used to account for multiple observations in each patient. The analysis was completed using R 4.0.

Results

When assessed separately by hemifield (nasal and temporal), latency did not distinguish HC from ADS or HC from MS. Decreases in amplitude (0.99±1.00nAm) measures overall corresponded to increased probability of MS. This relationship was stronger when measured in the temporal hemifield (X2=0.98). Effect size was small, but statistically significant (p=0.023).

Conclusions

In the chronic state, using visual evoked fields (VEF), the amplitude is more likely than latency to distinguish youth with MS from HC. This effect is accentuated in the temporal hemifield. This reinforces the knowledge that focal axonal injury may occur early on in youth with MS. While significant, the effect size of this relationship was small, suggesting that VEF temporal hemifield amplitude alone may not be adequate to distinguish MS from HC. Future studies should concentrate on the contribution of other morphological features of the VEF waveform to distinguishing MS from other populations.

Collapse
Pediatric MS Poster Presentation

P1079 - Optic nerve head volume decreases in pediatric multiple sclerosis but not in pediatric MOG-related disorders (ID 1357)

Speakers
Presentation Number
P1079
Presentation Topic
Pediatric MS

Abstract

Background

Children with multiple sclerosis (MS) and myelin oligodendrocyte glycoprotein (MOG)-associated syndromes exhibit changes in the retinal nerve fiber layer thickness (RNFL) and ganglion cell inner plexiform layer thickness (GCIPL), but knowledge about abnormalities in the optic nerve head (ONH) morphology in these groups is lacking. Previous studies have characterized deformations of the ONH structures associated with glaucoma and increased raised intracranial pressure and others demonstrated the utility of the assessment of these 3D parameters in the management of adults with aquaporin-4-related NMOSD and acute optic neuritis. No information regarding the utility of these measures in childhood demyelinating syndromes is available.

Objectives

To compare 3D ONH characteristics of pediatric multiple sclerosis and MOG-associated syndromes to healthy controls (HC).

Methods

Standardized OCT data (Cirrus HD-OCT (Carl Zeiss Meditec, Dublin, CA)) were collected on HC [n=27(16 female); median age at OCT of 15.0 years (13.7-16.3)], pediatric MS patients [n=17(11 female), median age at diagnosis of 16.9 years(15.4-18.4); median disease duration of 2.6 years(0.2-5.4)] and children with MOG-related disorders [ n=13( 11 female); median age at diagnosis of 11.5 years (7.3-15.7), median disease duration of 1.7 years(0.4-3)]. ONH parameters (total ONH volume (TV), Bruch’s membrane opening (BMO), region volume (BMO-Vol), and BMO minimal rim width (BMO-MRW)) were computed by triangulated 3D surface reconstruction. Mean peripapillary RNFL and GCIPL were derived from manufacturer’s fully-automated segmentation software, and results were reviewed with manual correction where necessary. Multivariable mixed effects models were used to model the 5 OCT parameters between groups, accounting for age at OCT, eye-specific number of optic neuritis episodes, and a subject-specific random intercept. Results were Bonferroni adjusted for multiple comparisons (p = 0.01).

Results

MS patients showed lower ONH TV (-.241, SE.073; p=.002), BMO-Vol (-.217, SE.083; p=.012), and BMO-MRW (-0.42 dMRW, SE.014; p=.004) compared to HC. ONH parameters were not different in MOG patients compared to controls. Neither RNFL nor GCIPL were significantly different between HC and MS. GCIPL was lower in MOG than HC (-7.9, SE 3.045; p.012).

Conclusions

Our analysis showed ONH volume loss in pediatric MS despite no significant differences in RNFL or GCIPL compared to HC. These changes were not observed in MOG patients. ONH analysis may therefore be superior to RNFL and GCIPL in detecting anterior visual pathway injury in children with MS early in the disease course. Larger studies are needed to confirm these associations.

Collapse

Presenter Of 1 Presentation

Imaging Poster Presentation

P0546 - Axonal and myelin volume fractions and imaging g-ratio in pediatric MS and MOG-associated disorders. (ID 1520)

Abstract

Background

Previous studies have described extensive microstructural brain tissue abnormalities in pediatric MS patients. However, available techniques do not distinguish the extent to which such abnormalities are due to axonal loss or demyelination. Further, little is known about microstructural brain tissue changes in MOG-associated disorders (MOGad).

Objectives

To apply a combined analysis of magnetization transfer saturation (MTsat) and multi-shell diffusion-weighted imaging (DWI) with computation of myelin and axonal volume fractions (MVF and AVF) and imaging g-ratio (the ratio between inner and outer diameter of the myelin sheath); to investigate the specific relationship between these metrics in the corpus callosum (CC) and within brain white matter lesions (WML) of pediatric MS and MOGad.

Methods

We acquired standardized 3T brain MRI in 26 healthy controls (HC) (58% females (F), mean age [years (y) (range)] 15y (9-19)); 16 MS (69% F, 17y (14-18), disease duration (DD) 3y (1-7), time from last relapse (TLR) 2y (0-6)); and 11 MOGad (72% F, 12y (8-18), DD 3y (0-6), TLR 1y (0-3), 8/11 relapsing). WML and CC were segmented according to establishes procedures. DWI processing was performed with FSL and DMIPy; MTsat, MVF, AVF, and g-ratio were computed using the Jargon data management system. We used general linear models to model average MVF, AVF, and g-ratio in the CC and WML of each group, including the factors age, DD, and the interaction term group*DD. Models including sex were tested, and all exhibited lower AIC.

Results

Relative to HC, MS showed decreased CC MVF (-0.018/y, p=0.0304) and AVF (-0.0069/y; p=0.053) and corresponding increased CC g-ratio (0.0084/y, p=0.059) with increased DD. Relative to HC, MOGad showed decreased CC MVF (-0.017/y, p=0.0304) and AVF (-0.0081/y, p=0.014) with increased DD, without significant CC g-ratio changes. Both MS and MOGad showed decreased average WML MVF compared to HC WM (-0.19, p<10-8; and -0.2, p<10-8). MOGad also showed decreased average WML AVF (-0.067, p=0.0048) compared to HC. Average WML g-ratio was higher in MS than MOGad (0.17, p=0.0102), but not significantly different from HC in either group. WML MVF, AVF, and g-ratio did not change significantly with DD in MS or MOGad compared to HC.

Conclusions

Both pediatric MS and MOGad exhibited MRI correlates of axonal loss and demyelination in the CC and WML. Our measures of axonal loss in MOGad reinforces recent work warning of potentially long-term impacts on the brain from non-MS demyelinating pathologies.

Collapse