Multiple Sclerosis Biomarkers – moving towards clinical applications
Publication Date
December 13, 2019
Creator
Abstract
Background: Previously proposed multiple sclerosis (MS) biomarkers, including the central vein sign (CVS), which can help differentiate between MS and non-MS, and brain atrophy, which has been shown to predict disability, hold the promise of better diagnosis and monitoring of MS patients. However, findings from research MRI studies need further validation on conventional hospital MRI, before they can be incorporated into clinical practice.
Methods: Using 3T MRI, I tested the diagnostic usefulness of the CVS in three studies: (1) a large, cross-sectional, multi-centre study of patients with MS and mimicking disorders, (2) a prospective single-centre study of patients with atypical first presentation of a clinically isolated syndrome (CIS) and (3) a prospective single-centre study of patients with a typical CIS and MS mimicking disorders. I also explored the utility of routine, non-standardised MRI scans in atrophy estimation and proposed a novel “Frankenstein” approach for atrophy estimation in scans with missing slices.
Results: The CVS achieved high sensitivity (up to 100%) and specificity (up to 90%) in diagnosing MS against its imaging mimics and in predicting a final diagnosis of MS in patients presenting with both typical and atypical CIS. Moreover, the detection of the CVS was largely independent of the MR sequence or scanner used. Brain atrophy estimations from suboptimal clinical MRI scans improved with the “Frankenstein” approach, increasing successful processing rate from 0% to 59%.
Conclusions: The CVS appears to be a highly sensitive and specific marker of MS lesions at 3T and its diagnostic value, as a stand-alone test and in combination with the current diagnostic criteria, should be assessed in large, prospective, multi-centre studies. Assessment of brain atrophy remains challenging in clinical practice. The “Frankenstein” approach appears promising in reducing atrophy estimation failures but future studies using a larger cohort of patients are required to further validate its performance.
Item Type
ethesis
Thesis Type
PhD
Subjects (LC)
Associated Schools / Departments
School of Psychology (UK)
eprints ID
59301
UoN Repository URI
Except where otherwise noted, this item's license is described as
File(s)![Thumbnail Image]()
Name
Thesis4259916final.pdf
Type
Full-text
Description
Examined
Size
2.63 MB
Format
Adobe PDF
Checksum (MD5)
82daa592978b0c1ffc0bc5d2c93da326