Genetic, magnetic resonance imaging and body fluid biomarker associations with severity of multiple sclerosis

Multiple sclerosis is a chronic and progressive neuroinflammatory disease that leads to demyelination and neurodegeneration in the central nervous system (CNS). Previous research has identified a wide range of environmental, lifestyle and genetic factors which increase MS susceptibility. However, the pathomechanisms that influence the severity of MS are largely unknown, and adequate biomarkers of disease severity are consequently lacking. Therefore, the aim of my thesis was to; 1) assess associations between the nerve injury biomarker neurofilament light (NfL) and brain atrophy and lesion volumes; 2) assess which brain/lesion volume measures show the strongest longitudinal association with clinical MS disability measures and to what degree these associations were affected by age; and to 3) identify genetic variants associated with brain atrophy, lesion volumes and plasma NfL (pNfL) levels in persons with MS.

In Study I, we assessed how cerebrospinal fluid (CSF) and pNfL levels were associated with T1- and T2-lesion volumes as well as whole-brain, cortical and subcortical grey matter, white matter and thalamic volume fractions of total intracranial volume based on magnetic resonance imaging (MRI). High baseline CSF and pNfL levels were associated with lower whole-brain, subcortical grey matter, thalamic, white matter and corpus callosal volume fractions over time. A further analysis showed that there was an association between baseline pNfL and baseline cortical grey matter fractions also in absence of radiological signs of inflammatory disease activity. A topographic analysis of cortical thickness showed that loss of cortical volume preferentially involved frontotemporal cortical regions. These findings indicate that NfL levels contribute information about MS severity not provided by traditional MRI lesion metrics.

In Study II, we showed that associations between baseline MRI variables, and baseline physical disability and self-reported impact of MS rapidly increased in strength in individuals beyond approximately 40-50 years of age. In separate longitudinal analyses using linear mixed-effects models, we showed that among the recorded brain volume measures, cortical and subcortical grey matter and thalamic volume fractions at baseline were the strongest predictors of future worsening in clinical disability over a median of approximately ten years’ follow-up time. They were also stronger predictors than T1- and T2-lesion volumes.

In Study III, we assessed if a weighted risk score comprising 12 known MS risk human leukocyte antigen (HLA) alleles was associated with baseline and longitudinal MRI measures as described in Studies I and II. While this risk score was not significantly associated with baseline MRI measures, we found that a high score was associated with lower cortical grey matter fractions longitudinally. A further analysis showed that this effect was primarily driven by the HLA-DRB1*15:01 allele. These results suggest that MS HLA risk variants not only affect inflammatory, but also neurodegenerative aspects of the disease.

In Studies IV and V, we performed genome-wide association studies of pNfL levels and whole-brain volume fractions, respectively, in persons with MS (and controls in Study IV). While no genome-wide significant associations were found in Study IV, gene set analyses highlighted a neural crest and odontogenesis development pathway in the regulation of pNfL levels, and a weighted MS susceptibility polygenic risk score was associated with higher pNfL levels in MS with statistical significance. These findings suggest that there is some degree of genetic regulation of pNfL levels, which partially overlap with MS risk.

In Study V, we identified a genome-wide significant locus upstream of the glycerol kinase 2 (GK2) gene, previously implicated in the propensity for tobacco smoking, which is a known MS risk and severity factor. Gene set analyses in Study V also implicated Hypoxia Inducible Factor-1 (HIF1) in the regulation of whole-brain volume fractions, indicating that iron metabolism and response to hypoxia play a role in the neurodegenerative processes in MS.

List of scientific papers

I. Benjamin V. Ineichen, Thomas Moridi, Ewoud Ewing, Russell Ouellette, Ali Manouchehrinia, Leszek Stawiarz, Daniel Ferreira, Sebastian J. Muehlboeck, Jens Kuhle, Eric Westman, David Leppert, Jan Hillert, Tomas Olsson, Ingrid Kockum, Fredrik Piehl, Tobias Granberg. Neurofilament light chain as a marker for cortical atrophy in multiple sclerosis without radiological signs of disease activity. J Intern Med. 2021 Aug;290(2):473-476.
https://doi.org/10.1111/joim.13286

II. Thomas Moridi, Leszek Stawiarz, Kyla A McKay, Benjamin V Ineichen, Russell Ouellette, Daniel Ferreira, J-Sebastian Muehlboeck, Eric Westman, Ingrid Kockum, Tomas Olsson, Fredrik Piehl, Jan Hillert, Ali Manouchehrinia, Tobias Granberg. Association between brain volume and disability over time in multiple sclerosis. Mult Scler J Exp Transl Clin. 2022 Dec;8(4):20552173221144230.
https://doi.org/10.1177/20552173221144230

III. Thomas Moridi, Pernilla Stridh, Ali Manouchehrinia, Leszek Stawiarz, Daniel Ferreira, Russell Ouellette, J-Sebastian Muehlboeck, Eric Westman, Tomas Olsson, Fredrik Piehl, Jan Hillert, Ingrid Kockum, Tobias Granberg. HLA genetic burden is associated with longitudinal cortical atrophy in multiple sclerosis. [Submitted]

IV. Thomas Moridi, Pernilla Stridh, Adil Harroud, Klementy Shchetynsky, Ali Manouchehrinia, Jens Kuhle, Jan Hillert, Fredrik Piehl, Tomas Olsson, Ingrid Kockum. A genome-wide association study of plasma neurofilament light chain levels in multiple sclerosis. [Manuscript]

V. Thomas Moridi, Pernilla Strid, Adil Harroud, Klementy Shchetynsky, Leszek Stawiarz, Daniel Ferreira, J-Sebastian Muehlboeck, Eric Westman, Jan Hillert, Fredrik Piehl, Tomas Olsson, Tobias Granberg, Ingrid Kockum. A genome-wide association study of whole-brain volume fractions in multiple sclerosis. [Manuscript]