FY2013 has seen significant progress toward accomplishing all of the Specific Aims.
For Aim 1, a major effort in the Translational Neuroradiology Unit (TNU) has been to define the relationship between MS lesions and the small veins around which they form. Based on images acquired at 3 and 7 tesla MRI, we have shown that central veins can be identified within most white matter MS lesions as well as lesions in an experimental autoimmune encephalomyelitis (EAE) model induced in the marmoset monkey. We have optimized and reported a technique for the detection and analysis of perivenular lesions in clinical care and research, which we call FLAIR*. We have used this technique to demonstrate that intralesional MS veins are smaller than uninvolved control veins, perhaps compressed by perivascular inflammatory cells or by fibrosis of the vascular wall. We have also found that extralesional MS veins appear larger than their counterparts in non-MS cases. By investigating developing MS lesions at high resolution using 7 tesla MRI, we have confirmed and extended earlier work in the lab demonstrating that opening of the blood-brain barrier in new MS lesions is a dynamic process that changes over time as lesions grow and begin to repair. Moreover, we have demonstrated that this opening of the blood-brain barrier is detectable on noncontrast 7 tesla imaging as a reduction of the T2* relaxation time constant and, more substantially, a shift in the phase of the MRI signal. In addition to perivascular abnormalities within the brain parenchyma itself, we have preliminarily reported the presence of blood-cerebrospinal fluid barrier opening within the subarachnoid space in up to 25% of MS cases, a new finding that is consistent with the presence of inflammation in this compartment. This may represent the first noninvasive demonstration of meningeal inflammation in vivo in MS, and studies to understand the characteristics of this finding, as well as its specificity and correlation with histopathological changes, are ongoing in the lab. Overall, our observations support a reinterpretation of blood-brain barrier opening in terms of a competition between tissue damage, which at least initially proceeds outward from the central vein, and tissue repair (or the prevention of damage), which is most intense at the periphery.
For Aim 2, in collaboration with the Advanced MRI Section (PI: Jeff Duyn) in the Laboratory of Functional and Molecular Imaging in NINDS, we are pursuing a related approach to myelin imaging. This approach, which uses gradient-echo imaging to assess the T2* time constants and frequency distribution of the MRI signal, offers several advantages over conventional approaches: It can be applied more readily on high-field MRI systems (3T and above) due to lower power deposition, amplifying both signal and contrast relative to background;data can be obtained much more rapidly;and sensitivity to myelin is increased because myelin itself, due both to its chemistry and to its highly ordered structure in white matter, induces susceptibility changes. Studies from this collaboration published during the current funding period have characterized the MRI signal that arises from a distinct pool of water protons with short relaxation times and substantial frequency shifts. The characteristics of this pool appear to be different in MS cases and within EAE lesions in the marmoset, and ongoing work is designed to more fully characterize these changes in both humans and animals. The results are accounted for by a quantitative model that fairly accurately describes the behavior of the myelin water signal at high field strength. We are also investigating the imaging correlates of axonal damage using a technique known as diffusion-weighted spectroscopy, which allows measurement of the diffusion properties of intracellular metabolites, particularly the intraneuronal metabolite N-acetylaspartate (NAA). So far, we have shown that that diffusion of NAA parallel to the axon is significantly lower in MS cases than in healthy volunteers and is moreover inversely correlated with the diffusion of water in the same direction. This result has been confirmed in a second cohort of MS cases and is consistent with biophysical models that suggest that axonal damage should reduce diffusivity and helps to resolve a paradox in the literature. Our results suggest that NAA diffusivity is a more specific marker of white matter integrity than water diffusivity.
For Aim 3, we have developed new and fully automated image segmentation techniques to identify the volumes of brain structures, including lesions. Two US patents are pending from this work were submitted during the current funding period. These methods allow us to investigate patterns of lesion growth and recovery and to learn how such patterns change over the course of the disease and in response to the initiation of different disease-modifying therapies. During the current funding period, we have also published a paper that develops a technique to integrate data derived from multiple scanners and scanning protocols in order to assess long-term changes in brain volume over time, a fundamental result of the MS disease process. The lack of a method to accomplish this is a substantial drawback in MS clinical research because scanning technology has been improving rapidly, and these improvements limit the ability to compare current scans with those obtained a decade or more earlier from the same individuals. The results demonstrate essentially linear decreases in gray matter volume over long periods of time with concomitant exponential increases in ventricular volume. Outside of technique development, in collaboration with the Myelin Repair Foundation we have developed a schema for proof-of-concept, short-term evaluation of therapies that promote tissue protection and repair in acute MS lesions. Such a trial design, which requires 6 months of testing in 15-20 people, improves dramatically on currently used methods that are based on 2 years of observation in 80-100 individuals. Detailed planning to test this trial design in relapsing-remitting MS is underway.

Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Zip Code
Absinta, Martina; Nair, Govind; Filippi, Massimo et al. (2014) Postmortem magnetic resonance imaging to guide the pathologic cut: individualized, 3-dimensionally printed cutting boxes for fixed brains. J Neuropathol Exp Neurol 73:780-8
Maggi, Pietro; Macri, Sheila M Cummings; Gaitan, Maria I et al. (2014) The formation of inflammatory demyelinated lesions in cerebral white matter. Ann Neurol :
Shiee, Navid; Bazin, Pierre-Louis; Cuzzocreo, Jennifer L et al. (2014) Reconstruction of the human cerebral cortex robust to white matter lesions: method and validation. Hum Brain Mapp 35:3385-401
Sweeney, Elizabeth M; Vogelstein, Joshua T; Cuzzocreo, Jennifer L et al. (2014) A comparison of supervised machine learning algorithms and feature vectors for MS lesion segmentation using multimodal structural MRI. PLoS One 9:e95753
Bryan, Frederick W; Xu, Zhoubing; Asman, Andrew J et al. (2014) Self-assessed performance improves statistical fusion of image labels. Med Phys 41:031903
Liu, Winston; Nair, Govind; Vuolo, Luisa et al. (2014) In vivo imaging of spinal cord atrophy in neuroinflammatory diseases. Ann Neurol 76:370-8
Sati, P; Thomasson, D M; Li, N et al. (2014) Rapid, high-resolution, whole-brain, susceptibility-based MRI of multiple sclerosis. Mult Scler 20:1464-70
Gao, Katherine C; Nair, Govind; Cortese, Irene C M et al. (2014) Sub-millimeter imaging of brain-free water for rapid volume assessment in atrophic brains. Neuroimage 100:370-8
Asman, Andrew J; Bryan, Frederick W; Smith, Seth A et al. (2014) Groupwise multi-atlas segmentation of the spinal cord's internal structure. Med Image Anal 18:460-71
Virtanen, J O; Wohler, J; Fenton, K et al. (2014) Oligoclonal bands in multiple sclerosis reactive against two herpesviruses and association with magnetic resonance imaging findings. Mult Scler 20:27-34

Showing the most recent 10 out of 26 publications