Contributions
Abstract: EP1445
Type: ePoster
Abstract Category: Pathology and pathogenesis of MS - Imaging
Rationale: Detection of diffuse brain pathology in neuromyelitis optica spectrum disorder (NMOSD) using advanced magnetic resonance imaging (MRI) is inconsistent across studies and methods. Atlas-based assessment of myelin water fraction (fM) maps in multiple sclerosis (MS) has been shown to increase sensitivity to changes in myelin. We tested whether diffuse myelin changes could also be detected in NMOSD using this approach.
Methods: fM maps were obtained for 51 healthy controls (HC), 10 NMOSD, and 10 age, gender and EDSS-matched MS patients using a mcDESPOT protocol at 3T. A study-specific template was created using FSL tools to facilitate registration of atrophied brains. An fM atlas was derived from the HC data, including voxelwise mean and standard deviation, and z-score maps computed. A group-level normal-appearing white matter (NAWM) mask was created by subtracting each lesion mask from a white matter segmentation to preserve areas that were consistently non-lesional across subjects, and excluding voxels that reached an inter-subject coefficient of variation ≥ 20% in HC. Volume of abnormal fM (VAfM) was calculated as the number of voxels with a z-score ≤ -3 within supratentorial NAWM. Patient groups and a subset of 10 age and gender-matched HC were compared using non-parametric statistics (significant p-values and trends reported after Bonferroni correction).
Results: VAfM was higher in MS (mean±standard deviation: 378±542, p=0.01) and in NMOSD (126±205, p=0.1, non-significant) compared to HC (33±65). In MS, lesion volume was significantly correlated with VAfM (Spearman"s ρ=+0.9, p=0.02) and average NAWM z-score (ρ=-0.9, p=0.009); these relationships were absent in NMOSD (lesion volume with VAfM: ρ=0.3; with average NAWM z-score: ρ=0.04, both non-significant).
Conclusion: Diffuse myelin changes were observed in MS and linked to the extent of focal damage. VAfM was greater in NMOSD compared to HC, though not significantly; the lack of correlation with lesion load suggests that diffuse damage may be due to the presence of lesions not detected with conventional imaging, or secondary to Wallerian degeneration from lesions in areas that were not imaged (i.e. in spinal cord). In conditions so heterogeneous across subjects, further value can be extracted from fM z-score maps on a case-by-case basis, as they enable visualisation of abnormal areas at the single-subject level, and may be a useful method for individual disease description and tracking.
Disclosure:
A. Combes: has received partial PhD funding from Janssen.
P. Manogaran: has nothing to disclose.
I.M. Vavasour: has nothing to disclose.
K. McMullen: has nothing to disclose.
R. Carruthers: is the site Principal Investigator for studies funded by MedImmune, Seattle Genetics and Guthy Jackson; has received speaking fees for unbranded lectures from Biogen, Genzyme, Novartis, and Teva; has received consulting fees from EMD Serono, Genzyme.
D.K.B. Li: has acted as a consultant for Vertex Pharmaceuticals; has been on the scientific advisory board for Roche, Novartis, Nuron; has been on the data & safety advisory board for Opexa; is the director of the UBC MS/MRI Research Group which has been contracted to perform central analysis of MRI scans for therapeutic trials with Genzyme, Hoffmann-LaRoche, Merck-Serono, Nuron, Perceptives and Sanofi-Aventis.
G.J. Barker: has received a honorarium from GE Healthcare for teaching on a programming course, and acts as a consultant for IXICO.
A. Traboulsee: has been on the steering committee and acted as a consultant for Roche; has been a DSMB member, acted as a consultant and been on the speakers bureau for EMD Serono; has acted as a consultant for TevaNeuroscience, Biogen, Novartis; has been a speaker for TevaNeuroscience, Genzyme.
S. Kolind: has received speaking fees/been on the scientific advisory board for Roche, Genzyme, EMD Serono.
Abstract: EP1445
Type: ePoster
Abstract Category: Pathology and pathogenesis of MS - Imaging
Rationale: Detection of diffuse brain pathology in neuromyelitis optica spectrum disorder (NMOSD) using advanced magnetic resonance imaging (MRI) is inconsistent across studies and methods. Atlas-based assessment of myelin water fraction (fM) maps in multiple sclerosis (MS) has been shown to increase sensitivity to changes in myelin. We tested whether diffuse myelin changes could also be detected in NMOSD using this approach.
Methods: fM maps were obtained for 51 healthy controls (HC), 10 NMOSD, and 10 age, gender and EDSS-matched MS patients using a mcDESPOT protocol at 3T. A study-specific template was created using FSL tools to facilitate registration of atrophied brains. An fM atlas was derived from the HC data, including voxelwise mean and standard deviation, and z-score maps computed. A group-level normal-appearing white matter (NAWM) mask was created by subtracting each lesion mask from a white matter segmentation to preserve areas that were consistently non-lesional across subjects, and excluding voxels that reached an inter-subject coefficient of variation ≥ 20% in HC. Volume of abnormal fM (VAfM) was calculated as the number of voxels with a z-score ≤ -3 within supratentorial NAWM. Patient groups and a subset of 10 age and gender-matched HC were compared using non-parametric statistics (significant p-values and trends reported after Bonferroni correction).
Results: VAfM was higher in MS (mean±standard deviation: 378±542, p=0.01) and in NMOSD (126±205, p=0.1, non-significant) compared to HC (33±65). In MS, lesion volume was significantly correlated with VAfM (Spearman"s ρ=+0.9, p=0.02) and average NAWM z-score (ρ=-0.9, p=0.009); these relationships were absent in NMOSD (lesion volume with VAfM: ρ=0.3; with average NAWM z-score: ρ=0.04, both non-significant).
Conclusion: Diffuse myelin changes were observed in MS and linked to the extent of focal damage. VAfM was greater in NMOSD compared to HC, though not significantly; the lack of correlation with lesion load suggests that diffuse damage may be due to the presence of lesions not detected with conventional imaging, or secondary to Wallerian degeneration from lesions in areas that were not imaged (i.e. in spinal cord). In conditions so heterogeneous across subjects, further value can be extracted from fM z-score maps on a case-by-case basis, as they enable visualisation of abnormal areas at the single-subject level, and may be a useful method for individual disease description and tracking.
Disclosure:
A. Combes: has received partial PhD funding from Janssen.
P. Manogaran: has nothing to disclose.
I.M. Vavasour: has nothing to disclose.
K. McMullen: has nothing to disclose.
R. Carruthers: is the site Principal Investigator for studies funded by MedImmune, Seattle Genetics and Guthy Jackson; has received speaking fees for unbranded lectures from Biogen, Genzyme, Novartis, and Teva; has received consulting fees from EMD Serono, Genzyme.
D.K.B. Li: has acted as a consultant for Vertex Pharmaceuticals; has been on the scientific advisory board for Roche, Novartis, Nuron; has been on the data & safety advisory board for Opexa; is the director of the UBC MS/MRI Research Group which has been contracted to perform central analysis of MRI scans for therapeutic trials with Genzyme, Hoffmann-LaRoche, Merck-Serono, Nuron, Perceptives and Sanofi-Aventis.
G.J. Barker: has received a honorarium from GE Healthcare for teaching on a programming course, and acts as a consultant for IXICO.
A. Traboulsee: has been on the steering committee and acted as a consultant for Roche; has been a DSMB member, acted as a consultant and been on the speakers bureau for EMD Serono; has acted as a consultant for TevaNeuroscience, Biogen, Novartis; has been a speaker for TevaNeuroscience, Genzyme.
S. Kolind: has received speaking fees/been on the scientific advisory board for Roche, Genzyme, EMD Serono.