Differentiation and quantification of inflammation, demyelination and axon injury or loss in multiple sclerosis

Brain. 2015 May;138(Pt 5):1223-38. doi: 10.1093/brain/awv046. Epub 2015 Feb 26.

Abstract

Axon injury/loss, demyelination and inflammation are the primary pathologies in multiple sclerosis lesions. Despite the prevailing notion that axon/neuron loss is the substrate of clinical progression of multiple sclerosis, the roles that these individual pathological processes play in multiple sclerosis progression remain to be defined. An imaging modality capable to effectively detect, differentiate and individually quantify axon injury/loss, demyelination and inflammation, would not only facilitate the understanding of the pathophysiology underlying multiple sclerosis progression, but also the assessment of treatments at the clinical trial and individual patient levels. In this report, the newly developed diffusion basis spectrum imaging was used to discriminate and quantify the underlying pathological components in multiple sclerosis white matter. Through the multiple-tensor modelling of diffusion weighted magnetic resonance imaging signals, diffusion basis spectrum imaging resolves inflammation-associated cellularity and vasogenic oedema in addition to accounting for partial volume effects resulting from cerebrospinal fluid contamination, and crossing fibres. Quantitative histological analysis of autopsied multiple sclerosis spinal cord specimens supported that diffusion basis spectrum imaging-determined cellularity, axon and myelin injury metrics closely correlated with those pathologies identified and quantified by conventional histological staining. We demonstrated in healthy control subjects that diffusion basis spectrum imaging rectified inaccurate assessments of diffusion properties of white matter tracts by diffusion tensor imaging in the presence of cerebrospinal fluid contamination and/or crossing fibres. In multiple sclerosis patients, we report that diffusion basis spectrum imaging quantitatively characterized the distinct pathologies underlying gadolinium-enhanced lesions, persistent black holes, non-enhanced lesions and non-black hole lesions, a task yet to be demonstrated by other neuroimaging approaches. Diffusion basis spectrum imaging-derived radial diffusivity (myelin integrity marker) and non-restricted isotropic diffusion fraction (oedema marker) correlated with magnetization transfer ratio, supporting previous reports that magnetization transfer ratio is sensitive not only to myelin integrity, but also to inflammation-associated oedema. Our results suggested that diffusion basis spectrum imaging-derived quantitative biomarkers are highly consistent with histology findings and hold promise to accurately characterize the heterogeneous white matter pathology in multiple sclerosis patients. Thus, diffusion basis spectrum imaging can potentially serve as a non-invasive outcome measure to assess treatment effects on the specific components of underlying pathology targeted by new multiple sclerosis therapies.

Keywords: diffusion basis spectrum imaging; diffusion tensor imaging; inflammation; multiple sclerosis; white matter injury.

Publication types

  • Research Support, N.I.H., Extramural
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • Axons / pathology*
  • Demyelinating Diseases / pathology*
  • Diffusion Tensor Imaging / methods
  • Humans
  • Inflammation / pathology
  • Magnetic Resonance Imaging / methods
  • Multiple Sclerosis / pathology*
  • Myelin Sheath / pathology
  • Nerve Fibers, Myelinated / pathology*