Contributions
Abstract: 171
Type: Oral
Abstract Category: Pathology and pathogenesis of MS - Imaging
Background: Progressive multifocal leukoencephalopathy (PML) is a demyelinating condition of the central nervous system. Recent studies using susceptibility weighted imaging (SWI) have described a hypointense band near the gray-white junction (GWJ); it has been suggested that this is due to iron deposition. The time evolution of the band has not yet been reported, and conclusive evidence establishing whether iron is responsible for the finding, and where it accumulates, is lacking.
Methods: Using a rapid 0.65mm isotropic susceptibility-based MRI technique on a Siemens 3-tesla (3T) scanner, we investigated the features and temporal evolution of PML lesions on T2*-weighted (T2*w) images acquired for 10 PML patients (4 natalizumab-treated MS, 3 lymphoma, 1 idiopathic low CD4, 2 HIV infection) using a 32-channel head coil. Scans were repeated after a mean(SD) of 6(4) months after baseline.
5 slabs of postmortem brain tissue from 4 PML patients (2 HIV infection and 2 natalizumab-treated MS) were imaged at 7T with 0.4mm isotropic gradient-echo imaging. Four of these slabs also underwent X-ray fluorescence (XRF) imaging for iron. 18 blocks of tissue from these slabs were then studied by staining for Luxol fast blue, 3,3"-diaminobenzidine (DAB) Turnbull (for iron), and immunohistochemistry (IHC) for myelin proteolipid protein (PLP). Infection with JC virus was confirmed using IHC for T and VP1 antigens. IHC for CD68, GFAP, and T and VP1 antigens was also performed after DAB Turnbull staining.
Results: PML lesions were visible as asymmetric WM hyperintensities on T2*w images. A hypointense band was noted at the GWJ of the PML lesions, but not at the white matter border, in all cases. The volume of the band was 4.9(3.5) ml, increasing by 1.2(1.7) ml at follow-up (n=6), generally concordant with expansion of the PML lesion. XRF confirmed the hypointense band on postmortem scans corresponded to regions with high iron content. The band and XRF findings corresponded to the accumulation of iron-rich CD68+ macrophages. Iron was not present in astrocytes or JC-virus-infected cells.
Conclusions: A T2*-hypointense band at the GWJ is characteristic of PML lesions on MRI. The band may enlarge over time and corresponds histologically to the accumulation of iron within macrophages in the lower layers of the cortex. Further work will focus on establishing the specificity of the band for PML and on understanding its mechanistic basis and preferential spatiotemporal localization.
Disclosure:
V V Sethi: nothing to disclose
G Nair: nothing to disclose
S K Ha: nothing to disclose
B Popescu: Dr Popescu served as a speaker for Teva Innovation Canada, received an honorarium for publishing in Continuum: Lifelong Learning in Neurology and presenting at the 67th American Academy of Neurology Annual Meeting, Washington, and receives research support from the Canada Research Chairs program, University of Saskatchewan, Government of Saskatchewan and Saskatchewan Health Research Foundation.
M K Schindler: nothing to disclose
B R Smith: nothing to disclose
M Absinta: nothing to disclose
P Sati: nothing to disclose
C Wüthrich: nothing to disclose
S Webb: nothing to disclose
Kumar S: nothing to disclose
J Ohayon: nothing to disclose
C Lucchinetti: receives funding support from Biogen and Novartis but neither is directly related to this work.
I J Koralnik: is funded by NIH grants R01 NS047029 and NS 074995 and a research grant from Biogen
E O Major: nothing to disclose
A Nath: nothing to disclose
I Cortese: nothing to disclose
D S Reich: nothing to disclose
Abstract: 171
Type: Oral
Abstract Category: Pathology and pathogenesis of MS - Imaging
Background: Progressive multifocal leukoencephalopathy (PML) is a demyelinating condition of the central nervous system. Recent studies using susceptibility weighted imaging (SWI) have described a hypointense band near the gray-white junction (GWJ); it has been suggested that this is due to iron deposition. The time evolution of the band has not yet been reported, and conclusive evidence establishing whether iron is responsible for the finding, and where it accumulates, is lacking.
Methods: Using a rapid 0.65mm isotropic susceptibility-based MRI technique on a Siemens 3-tesla (3T) scanner, we investigated the features and temporal evolution of PML lesions on T2*-weighted (T2*w) images acquired for 10 PML patients (4 natalizumab-treated MS, 3 lymphoma, 1 idiopathic low CD4, 2 HIV infection) using a 32-channel head coil. Scans were repeated after a mean(SD) of 6(4) months after baseline.
5 slabs of postmortem brain tissue from 4 PML patients (2 HIV infection and 2 natalizumab-treated MS) were imaged at 7T with 0.4mm isotropic gradient-echo imaging. Four of these slabs also underwent X-ray fluorescence (XRF) imaging for iron. 18 blocks of tissue from these slabs were then studied by staining for Luxol fast blue, 3,3"-diaminobenzidine (DAB) Turnbull (for iron), and immunohistochemistry (IHC) for myelin proteolipid protein (PLP). Infection with JC virus was confirmed using IHC for T and VP1 antigens. IHC for CD68, GFAP, and T and VP1 antigens was also performed after DAB Turnbull staining.
Results: PML lesions were visible as asymmetric WM hyperintensities on T2*w images. A hypointense band was noted at the GWJ of the PML lesions, but not at the white matter border, in all cases. The volume of the band was 4.9(3.5) ml, increasing by 1.2(1.7) ml at follow-up (n=6), generally concordant with expansion of the PML lesion. XRF confirmed the hypointense band on postmortem scans corresponded to regions with high iron content. The band and XRF findings corresponded to the accumulation of iron-rich CD68+ macrophages. Iron was not present in astrocytes or JC-virus-infected cells.
Conclusions: A T2*-hypointense band at the GWJ is characteristic of PML lesions on MRI. The band may enlarge over time and corresponds histologically to the accumulation of iron within macrophages in the lower layers of the cortex. Further work will focus on establishing the specificity of the band for PML and on understanding its mechanistic basis and preferential spatiotemporal localization.
Disclosure:
V V Sethi: nothing to disclose
G Nair: nothing to disclose
S K Ha: nothing to disclose
B Popescu: Dr Popescu served as a speaker for Teva Innovation Canada, received an honorarium for publishing in Continuum: Lifelong Learning in Neurology and presenting at the 67th American Academy of Neurology Annual Meeting, Washington, and receives research support from the Canada Research Chairs program, University of Saskatchewan, Government of Saskatchewan and Saskatchewan Health Research Foundation.
M K Schindler: nothing to disclose
B R Smith: nothing to disclose
M Absinta: nothing to disclose
P Sati: nothing to disclose
C Wüthrich: nothing to disclose
S Webb: nothing to disclose
Kumar S: nothing to disclose
J Ohayon: nothing to disclose
C Lucchinetti: receives funding support from Biogen and Novartis but neither is directly related to this work.
I J Koralnik: is funded by NIH grants R01 NS047029 and NS 074995 and a research grant from Biogen
E O Major: nothing to disclose
A Nath: nothing to disclose
I Cortese: nothing to disclose
D S Reich: nothing to disclose