Sessions

Kathryn Fitzgerald, Sc.D., is an assistant professor of neurology at the Johns Hopkins University School of Medicine. Her research focuses on identifying diagnostic and prognostic factors related to multiple sclerosis (MS) and other neurologic diseases. She performs epidemiologic research integrating patient- and population-level data from people with MS. 

Dr. Fitzgerald earned her Doctor of Science in epidemiology at Harvard University, followed by a postdoctoral research fellowship in neuroimmunology and neurological infections at the Johns Hopkins University School of Medicine. Her research interests include how genetics, metabolic changes, hormones, diet, environmental factors and comorbid conditions affect the severity of MS symptoms.

Longitudinal epidemiology of MS in the Greater Hobart region, 1961 to 2019

BACKGROUND

The Greater Hobart region (42.3°S) of Tasmania has the highest frequencies of MS in Australia, this consistent across studies conducted over the last half century. However, the degree of excess compared to mainland sites like Newcastle (32.5°S) has declined over time.

OBJECTIVE

To assess the prevalence of MS in the Greater Hobart region in 2019, as well as the 2009-19 incidence and mortality rates, and to compare these estimates with prior measures in 1961, 1981, 2001, and 2009. 

METHOD

Cases were recruited from clinic-based samples and other multiple other data sources. Prevalence date was 1 June 2019. 2019 prevalence and 2009-19 incidence and mortality rates were estimated. Prevalence and incidence and mortality rates were age/sex-standardised to the 1961 and 1954 Hobart populations, respectively, to allow longitudinal comparisons. Female:male prevalence and incidence sex-ratios were also assessed. Differences between timepoints were assessed using Poisson regression. 

RESULTS

472 MS cases (female:71.4%) resident on prevalence day were identified, a crude prevalence=212.3/100,000 (155.8 age/sex-standardised), 57% increase vs 2009 and 388.0% vs 1961. The 2009-19 incidence rate=7.56/100,000 person-years (7.03 age/sex-standardised), 91% increase vs 2001-9 and 218.1% vs 1951-61. The 2009-19 mortality rate was 2.25/100,000 person-years (1.12 age/sex-standardised), comparable to 2001-9 (1.00). The age/sex-standardised prevalence-sex-ratio was 2.61, comparable to 2009 (2.65), but the incidence-sex-ratio was 2.68, 31% increase vs 2001-9 (2.05). 

CONCLUSION

Prevalence and incidence continue to be high in Hobart, although the differences compared to lower-latitude Newcastle have attenuated significantly. Changes in lifestyle may underlie the deterioration of the latitudinal gradient of MS in Australia. Further investigation is required to define the factors that drive these associations.

Cross-sectional analysis of stressful life events and relapse, disability, depression-risk, and fatigue in people with MS. 

BACKGROUND

Studies suggest stress increases the risk of relapses and reduces quality of life in people with MS but few have examined the effect of stressful life events (SLEs) on disease progression. 

OBJECTIVE

To investigate associations between the number of SLEs and individual SLEs and the frequencies of relapse, disability severity, depression-risk, and fatigue in people with MS. 

METHOD

Cross-sectional analysis of data from an online cohort of people with MS was performed (n=948). SLEs were assessed using a subset of 16 SLEs from the Holmes-Rahe Social Readjustment Rating Scale. SLE relationships with relapse, disability severity, depression-risk, and fatigue were assessed by log-binomial or log-multinomial regression, as appropriate, adjusted for age, sex, education, MS type, disability, fatigue, comorbidities, ongoing relapse symptoms, and antidepressant/anti-fatigue medication, as appropriate. 

RESULTS

Average SLE number was 1.8 (range 0-9). SLE number was associated with 10% (95%CI;1%-20%) and 8% (95%CI;3%-14%) more frequent depression-risk and fatigue, respectively, but no associations with relapse or disability were found. Depression-risk was more frequent in individuals with serious illness [30% (95%CI;3%-64%)], work/school/career crisis/serious disappointment [38% (95%CI;0%-88%)], and new family members [2.47-fold (95%CI;1.21-5.07)]. Fatigue was more frequent in individuals with serious illness [22% (95%CI;7%-39%)] and starting/resuming serious relationships [2.31-fold (95%CI;1.78-2.98)]. 

CONCLUSION

SLE number and individual SLEs were associated with greater depression-risk and fatigue. Defining the impact of SLEs on health outcomes may help inform behavioural and/or intervention strategies to improve health outcomes in people with MS.   

FUNDING

This research was generously funded by philanthropic gifts from Mr Wal Pisciotta, and anonymous donors.

Long Term Trajectories of Employment Status, Work hours and Disability Pension Status After an Initial Episode of CNS Demyelination. 

BACKGROUND

People with multiple sclerosis (PwMS) face significant work-related challenges. Long-term trajectories of employment related outcomes among PwMS have not previously been explored. 

OBJECTIVE

We aimed to explore distinctive trajectories of employment related outcomes or workforce participation in individuals with a first clinical diagnosis (FCD) of central nervous system (CNS) demyelination, and to examine individual’s sociodemographic and clinical factors associated with trajectory group membership. 

METHOD

We used data from the Ausimmune Longitudinal (AusLong) Study. Study participants diagnosed with FCD aged 18-59 years (n=279) were followed for more than 11 years. Those who had at least two observations during follow-up and had not been permanently out of workforce before the study, were included (n=207). Employment related outcomes were defined as employment status (full-time, part-time, and unemployed), average work hours per week and whether receiving a disability support pension (DP). We used Group Based Trajectory Modelling (GBTM) to identify distinctive groups of participants with common employment trajectory patterns, and factors associated with trajectory group membership were explored using log-binomial regression.

RESULTS

Distinct trajectories were identified for employment status, work hours per week and DP indicating substantial heterogeneity within the AusLong cohort. About one fourth of our sample had deteriorating employment status and there was noticeable variability in working hours of our participants during follow up. Notably, a minority group (18%) was identified with higher likelihood of receiving DP from early post FCD. Female sex was associated with membership in stable part-time (RR:4.2; 95%CI:2.0-8.6) and early deteriorated (RR:2.7; 95%CI:1.4-5.3) trajectories compared with stable full-time employment. Females also had consistently higher risk of being in deteriorated work hours trajectories than males. Moreover, every additional relapse during the 5 years post FCD was associated with 12% and 6% increased risk of being in early deteriorated and stable part-time trajectories compared to stable full-time trajectories, respectively. Additionally, progression to CDMS, higher number of comorbidities at baseline and higher number of relapses during five years post FCD were associated with greater risk of receiving a DP.  

CONCLUSION

The key findings of distinctive trajectories of employment status, work hours and DP among AusLong participants, suggest the considerable impact of MS that is initiated at the earliest stage of clinical CNS demyelination. The information provided in this research may be of higher importance for health professionals to closely monitor the employment outcomes in females and individuals with higher number of comorbidities and relapses. Therefore, necessary personalised interventions can be designed to suit individuals’ need and maintain them in the workforce.

Neurometabolic changes in RRMS: comparison between fingolimod and injectables therapies. 

BACKGROUND

Disease modifying therapies (DMTs) can reduce the frequency of relapses in multiple sclerosis (MS) patients and mitigate the long-term damage they cause. Treatment response studies using proton magnetic resonance spectroscopy (MRS) have demonstrated a significant increase in N-acetylaspartate/total creatine(NAA/tCr) in relapsing remitting MS (RRMS) patients correlated with maintaining axonal metabolic function. This is the first study to evaluate the neurometabolic effects of different DMTs on the MS brain, comparing volumetric and neurometabolic changes in RRMS patients on fingolimod or injectables to a matched healthy cohort. 

METHODS

Patients were recruited from a single center, on treatment for at least 6 months and stable with an EDSS<4. HCs were age and sex matched. All patients and HCs underwent EDSS as well as cognitive, fatigue and mental health assessments. MRS data was acquired from hippocampus, posterior cingulate gyrus(PCG) and pre-frontal cortex(PFC). Volumetric metrics and neurometabolic ratios were compared across cohorts using a univariate general linear model. MR metrics, clinical severity and neuropsychological scores were compared using parametric and non-parametric correlations. 

RESULTS

98 RRMS patients were recruited, comprising 52 on fingolimod, 27 on interferon and 19 on glatiramer acetate (GA). 51 healthy controls were age and sex matched. While clinical parameters showed no statistically significant differences between treatment cohorts, MR-metrics on individual and combined RRMS cohorts showed similar statistically significant volumetric differences: GM(-3%), WM(-5%) and CSF(+40%) compared to HCs. Compared to HCs, a reduction in NAA/tCr in hippocampus(-17%,p=0.001), PCG(-7%,p=0.001) and PFC(-9%,p=0.001) were detected in both cohorts. Volumetrics and neurometabolites in each location, moderately correlated with attention, cognition, fatigue, memory and depression.

CONCLUSION

This study is the first cross-sectional in-vivo investigation comparing the impact of fingolimod with interferon or GA on brain metabolism of RRMS patients. Although both treatment arms were not significantly different in volumetric and neurometabolic profiles, longitudinal studies may clarify brain metabolic changes over time and associations with treatment efficacy.

Developing a 18F radiotracer targeting the cell surface receptor MERTK to visualise reparative microglia in the demyelinated brain. 

BACKGROUND
Innate immune cells are activated within the brain in MS. These cells either worsen disease or promote repair, indicating need for techniques identifying their predominant response in individuals. Molecular imaging targeting MERTK, a receptor expressed selectively by reparative innate immune cells could fill this need.  

OBJECTIVE
Generation of a novel 18F radiotracer targeting MERTK for Positron Emission Tomography (PET) imaging. This will enable patients with MS and other neuroinflammatory CNS diseases to be more accurately phenotyped by determining the extent to which protective rather than pathogenic innate immune cells are present.  

METHOD
Mined data bases of kinase inhibitors of human MERTK, identifying those predicted to be CNS penetrant. Assessed biochemical properties of lead candidates. Identified suitable sites for installing 18F. Synthesised cold fluorinated analogues, confirmed high affinity for MERTK and undertook pharmacokinetic studies, including CNS penetrance. Generated histological sections from cuprizone challenged mice; performed in vitro autoradiography, testing for specific binding of the lead candidate 18F radiotracer targeting MERTK to demyelinated zones.  

RESULTS
Our lead candidate, [18F]-MIPS15692, is selective for MERTK over other receptor tyrosine kinases. Pharmacokinetic analysis demonstrates key characteristics for a radiotracer, including rapid plasma clearance, high brain to plasma ratio and unbound brain to plasma ratio (Kp,uu) >1, demonstrating rapid cerebral uptake. In vitro autoradiography shows high affinity binding (KD 12.6nM) in regions of increased MERTK expression (Bmax=896 fmol/mg protein) within demyelinated corpus callosum and hippocampus, blocked by competition studies.  

CONCLUSION
These results provide proof-of-concept evidence for utility of [18F]MIPS15692 in imaging MERTK expression in demyelinating disease. Future studies will investigate [18F]MIPS15692 as a PET radiotracer in vivo and its ability to stratify the nature of  innate immune activity between patients, to facilitate targeted therapeutic intervention.  

FUNDING
NHMRC project grant APP1142814, NHMRC Investigator grant APP1175775, MSRA project grant 20-119, Perpetual Foundation

NgR-Fc protein delivered by hematopoietic cells enhances neurorepair in a multiple sclerosis model .

ABSTRACT

Nogo receptor 1 (NgR1) is the high affinity receptor for the potent myelin-associated inhibitory factors (MAIFs) that make up part of the inflammatory extracellular milieu during experimental autoimmune encephalomyelitis (EAE). Signaling through the NgR1 complex has been shown to be associated with axonal degeneration in an animal model of MS and neuronal loss of this receptor homologue. In a disease, the specific manner is associated with preserving axons even in the context of neuroinflammation. The local delivery of NgR(310)ecto-Fc, a therapeutic fusion protein has been successfully applied as a treatment in animal models of spinal cord injury and glaucoma. As multiple sclerosis (MS) and EAE exhibit large numbers of inflammatory cell infiltrates within central nervous system (CNS) lesions, in this study, we utilized transplantable HSCs as a cellular delivery method of the NgR(310)ecto-Fc fusion protein. There were no immunomodulatory effects of the transplanted, modified HSCs with LV-NgR(310)ecto-Fc on immune cell lineages in EAE-induced recipient female mice. Moreover, diffusion tensor imaging (DTI)-MRI is expected to evaluate the potential relationship between grey matter lesions and white matter lesions microstructurally. We identified CNS infiltrating macrophages as the immune-positive cell type expressing for the myc-tagged NgR(310)ecto-Fc. These differentiated phagocytes predominated the extensive immune cell infiltrated regions at the peak of clinical disease with extensive demyelination and associated engulfment of NgR-Fc protein-MAIF complexes. Notably, three animals transplanted with LV-NgR(310)ecto-Fc-overexpressing HSCs, recovered from the peak of neurological symptoms associated with EAE, exhibiting axonal regeneration and eventual remyelination in the white matter tracts. These results suggest that HSCs can be utilized as carriers of the therapeutic NgR-Fc protein for specific delivery into multifocal EAE lesions and can potentiate neurological recovery.

ACKNOWLEDGEMENT
This project is supported by Trish MS Research Foundation.

Dietary plasmalogen supplementation stimulates remyelination and restores neuromotor function in a mouse model of demyelination.

BACKGROUND

Over 70% of the myelin membrane is comprised of lipids with phospholipid plasmalogens being one of the most abundant lipids. In Multiple Sclerosis (MS), progressive loss of myelin is characterised by a net loss of lipids, particularly in myelin-plasmalogens. 

OBJECTIVES

To investigate whether dietary-induced plasma-plasmalogens supports central myelin homeostasis and neuromotor function in the cuprizone mouse model of demyelination. 

METHODS

16-week-old male C57BL/6J mice were randomised to receive either control chow, cuprizone-enriched diet (0.2%), alkylglycerol-supplemented (plasmalogen precursor; 1%), or a combination of cuprizone and alkylglycerol. After 4 weeks of intervention, CNS myelin homeostasis was determined by in vivo Magnetic Resonance Imaging (MRI; Magnetisation Transfer Imaging) and neuromotor coordination, balance and function was assessed by the Rotarod performance test. 

RESULTS

At 4 weeks of intervention, dietary plasmalogen intervention ameliorated cuprizone-induced demyelination in the corpus callosum. Brain myelin abundance in cuprizone plus plasmalogen treated mice was comparable to control mice on normal chow. In alignment with these findings, dietary-plasmalogen and cuprizone treated mice had significantly improved neuromotor coordination compared to cuprizone-only treated mice with average performance (latency) time of 68.4 s and 33.9 s (control group 60.2 s), respectively.  

CONCLUSION

Dietary-induced increase in plasma plasmalogens were associated with markedly enhanced myelin homeostasis and alleviation of neuromotor deficits, in a murine-model of chemically-induced demyelination. The results of this study strongly support the potential of dietary modulation of myelin-lipid homeostasis and presents as an easily translatable therapeutic strategy to stimulate remyelination in MS.

Changes to immune cell gene expression following vitamin D supplementation in people with a first demyelinating event: a PrevANZ substudy. 

BACKGROUND

Vitamin D deficiency is a risk factor for multiple sclerosis (MS). Predominantly in vitro studies have demonstrated immunoregulatory effects by vitamin D. How vitamin D supplementation alters immune cell activity remains unclear. We sought to investigate this in people with a first demyelinating event (FDE). 

OBJECTIVE

We aimed to assess effects of vitamin D supplementation at various doses on gene expression of peripheral immune cells in patients with a FDE, and to identify pathways modulated by vitamin D. 

METHOD

As a substudy of PrevANZ, participants were randomised to either 10,000, 5000 or 1000 IU daily of vitamin D3, or placebo. Peripheral blood was collected at baseline and 3 months in PAXgene tubes. The transcriptome was assessed by RNAseq. EdgeR was used to identify differentially expressed genes (DEGs) at 3 months compared to baseline, and ClusterProfiler/MSigDB used to determine enriched pathways. 

RESULTS

55 participants had baseline and 3-month samples. We identified evidence for differential expression of 2355 genes after supplementation with 5000 and 10,000 IU (P<0.05). RNA processing and translation pathways were upregulated after supplementation, while the oxidative phosphorylation pathway was downregulated. The KEGG antigen processing and presentation pathway was enriched by upregulated MHC II genes. TNF-alpha signalling was downregulated. No significant pathways/gene sets were up/downregulated following supplementation with 1000 IU/placebo (FDR<0.05). 

CONCLUSION

Three months of high dose vitamin D supplementation altered peripheral immune cell gene expression and specific pathways. Our results support a threshold/dose-dependent effect for vitamin D. We provide new insights into vitamin D’s in vivo immunomodulatory effects to inform future supplementation studies.

Vitamin D metabolism genes, 25(OH)D, and 1,25(OH)2D; interplay and association with a first clinical demyelination event.

BACKGROUND

Multiple sclerosis is a disorder with genetic and environmental contributions. Factors of relevance include the CYP27B1 gene and low serum 25(OH)D levels. The interplay with 1,25(OH)2D, the active form of vitamin D, is not well understood.

OBJECTIVE

We aim to evaluate (i) the association between 25(OH)D & 1,25(OH)2D, (ii) factors predicting 1,25(OH)2D after 25(OHD) levels are taken into account, (iii) the interplay between vitamin D metabolism genes and vitamin D metabolites for FCD risk. 

METHOD

This investigation utilised data from the Early Environment and Type 1 Diabetes Prevention Project; (EET1DPP; 241 cases, 322 controls), the Southern Tasmanian MS Longitudinal study (MSL; 198 prevalent MS cases), and the Ausimmune Study (279 FCD cases, 577 controls). Serum vitamin D metabolites, SNP data and other factors were examined. A genetic function score (GFS) was constructed using key vitamin D metabolism genes. A variety of statistical approaches were used. 

RESULTS

Correlation between 25(OH)D and 1,25(OH)2D was not high (r=0.11-0.58 across samples). In some samples, accounting for Vitamin D-related genes and their interaction with 25(OH)D levels improved the prediction of 1,25(OH)2D from 25(OH)D alone. Vitamin D-related SNPs were associated with FCD risk, either directly or exhibiting interaction with 1,25(OH)2D levels. The CYP27B1 and GC gene variants were associated with FCD, with an interaction between the GC variant and 1,25(OH)2D levels. 

CONCLUSION

Incorporating genetic information may improve 1,25(OH)2D levels prediction beyond 25(OH)D levels alone in some settings. The demonstrated interplay between these vitamin D metabolites and related genes requires replication

FUNDING

Multiple funding sources contributed to these three studies including: MS Research Australia, National Health and Medical Research Council of Australia, and the National Multiple Sclerosis Society of the United States of America.

Professor Gabrielle Belz trained in veterinary medicine and surgery at the University of Queensland and has made major contributions to the field of immunology for which she received a Doctor of Veterinary Science. Her prime research interests are in the areas of understanding how the fine tuning of the mucosal immune system of the gut and lung impact immune homeostasis and how the dysregulation of this system results in both local and systemic disease though networks such as the gut-brain axis. Her lab aims to unravel how innate and adaptive immune cells are wired to generate tolerance and memory – core processes protecting the local and distant organs of the body. She has published over 200 peer-reviewed original papers in leading scientific journals and has received a number of awards including the Gottschalk Medal from the Australian Academy of Science and Howard Hughes Medical Institute Investigator Awards. She is the Chair of Immunology, University of Queensland Diamantina Institute and continues to be excited by the unexpected and new discoveries linking innate cells in dynamic physiological regulatory circuits in immune protection.

NET inhibition as a novel treatment for autoimmunity. 

BACKGROUND

Neutrophil extracellular traps (NETs) are weblike structures released by neutrophils upon activation. Recent evidence has implicated NETs in RRMS. Thus, inhibiting histones, the major component of NETs, should reduce the severity of symptoms in our experimental autoimmune encephalomyelitis (EAE) MS model. 

OBJECTIVE

We here aim to optimise our lead histone inhibitor mCBS for future use in MS by seeking orally deliverable derivatives. We investigated the effects of these modified inhibitors on immune cell populations in our mouse model as well as their efficacy to inhibit histone meditated T cell activation in vitro. 

METHOD

EAE was induced in C57BL/6J mice, these mice were scored daily for disease parameters, and blood immune cell phenotypes were monitored throughout disease using flow cytometry. Histone inhibitor (mCBS) was administered either as a single dose before disease onset or continual dosing throughout the disease course.  In addition, naïve CD4+ T cells were isolated and polarised into different subsets in the presence of histones, with or without mCBS. 

RESULTS

The administration of histone inhibitor, mCBS, over the course of disease showed a significant decrease in the clinical score of EAE mice. We have also seen a marked increase in the differentiation of naïve T cells to specific subsets, notably Th17 cells, in the presence of low-level histones. This effect was prevented by mCBS. Current studies are examining the ability of mCBS derivatives with oral availability to inhibit EAE. 

CONCLUSION

Histone inhibition appears to have a marked effect on the course of EAE in mice, making this a potential new therapy to reduce symptoms for people living with MS. It has wider implications for other autoimmune diseases where immune responses involve NETs and, thus, the release of histones.

Long-term Immune Reconstitution after Autologous Haematopoietic Stem Cell Transplantation in Multiple Sclerosis Patients. 

BACKGROUND

Multiple Sclerosis (MS) is an autoinflammatory disease of central nervous system (CNS) characterised by demyelination, axonal degeneration, and dysregulated immunological self-tolerance. Studies indicate that autologous haematopoietic stem cell transplantation (AHSCT) improves clinical outcomes in MS, but evidence of how AHSCT induces clinical remission is limited. 

OBJECTIVE

Our group has previously reported early immunological reconstitution (IR) in MS patients up to 1yr post-AHSCT. However, long-term immunological effects of AHSCT have yet to be elucidated. The present study aims to investigate the maintenance of IR in MS patients at 2yrs and 3yrs post-AHSCT. 

METHOD

High-dimensional immunophenotyping of peripheral blood mononuclear cells from MS patients (n=23) at pre-AHSCT, 2yr and 3yr post-AHSCT timepoints were performed using two custom-designed 18-colour flow cytometry panels. Frequencies of key immune subsets involved in inflammation, immunoregulation, and CNS-trafficking were examined. Statistical analysis included linear mixed-effects model and Holm-Sidak test (p<0.05), with logarithmic transformations if required after residual analysis, using GraphPad Prism 8.  

RESULTS

Significant decreases were observed in pro-inflammatory immune populations, Th17 at 2yr (p<0.01) and 3yr (p<0.0001), putative CNS-homing (CD4+CD38-CD45RA-CD25med/-CD127-/+CD49d+β7-CCR6+CD73+) at 2yr (p<0.001) and 3yr (p<0.0001), circulating Tfh at 3yr (p<0.01), and MAIT cells at 2yr (p<0.0001) and 3yr (p<0.0001) post-AHSCT. We observed significantly increased CD56hi NK cells at 2yr (p<0.05), terminally differentiated CD8+ T-cells at 2yr (p<0.0001) and 3yr (p<0.001), whereas CD39+ T-regulatory cells remained unchanged at 2yr and 3yr post-AHSCT. 

CONCLUSION

Long-term suppression of the pro-inflammatory immune profile following AHSCT is indicative of sustained immune tolerance, which is further accompanied by long-term elevation of anti-inflammatory populations. These findings demonstrate that AHSCT induces substantial recalibration of the immune system in MS patients persisting out to 3yrs post-transplant.

Liposomal drug delivery: targeting therapies for neuroinflammation. 

BACKGROUND

Liposomes are drug delivery nanoparticles which may improve treatment targeting in neuroinflammation. Liposomes accrue at sites of disrupted vasculature, such as the inflamed central nervous system, where they are internalised by phagocytic immune cells, which may offer direct targeting to key pathogenic immune populations. 

OBJECTIVE

We aim to determine the impact of liposomal encapsulation on drug efficacy in the context of neuroinflammation, using liposomal mitoxantrone as proof-of-concept. Further, we aim to delineate liposome uptake by different immune cell populations and assess the mechanistic importance of these interactions. 

METHOD

Mice were induced with experimental autoimmune encephalomyelitis (EAE) and treated weekly with 0.5mg/kg liposomal mitoxantrone (LMTX) or freely delivered mitoxantrone (MTX). To assess treatment efficacy, mice were scored daily for clinical disease, and pathogenic immune cell populations in blood and CNS tissue were monitored using flow cytometry. To examine liposome delivery characteristics, EAE-induced mice were dosed with fluorescence-tagged liposomes and interactions with immune cells detected using flow cytometry. 

RESULTS

Weekly delivery of 0.5mg/kg liposomal mitoxantrone prevents clinical signs of EAE, whereas freely delivered MTX offers minimal protection. Examination of treatment impact on immune populations has identified an enhanced impact on pathogenic immune populations including B cells and IL-17-producing T cells. We have further identified a specific myeloid cell population depleted by LMTX treatment, which further seems to interact preferentially with liposomes and is correlated with disease severity. 

CONCLUSION

We have identified that liposomal encapsulation improves drug efficacy in EAE, highlighting the potential application of these drug vehicles in treatment of neuroinflammation. Further, we have initial insights regarding liposome-interacting immune cell populations, and are currently dissecting the importance of these interactions to drug mode-of-action. 

FUNDING

MS Research Australia Project Grant, Australia Government Research Training Program, Gordon Ada Scholarship.

Going beyond tokenistic: the experience of being a consumer involved in research about multiple sclerosis. 

BACKGROUND

The ANU Our Health in Our Hands (OHIOH) personalised medicine program has incorporated people living with MS in its Health Experience Team and the broader research program, with the aim of enhancing the quality, relevance, credibility and legitimacy of research. 

OBJECTIVE

To describe people living with MS’ experiences of being involved in research about MS. 

METHOD

Using story telling techniques, four Health Experience Team members living with MS were asked to create self-filmed vignettes reflecting on their involvement in the OHIOH program. Each team member reflected on one of four projects they had contributed to. 

RESULTS

Health Experience Team members living with MS felt supported and that their contributions were valued. Examples described included contributions to research question development, qualitative data analysis, preparation of peer reviewed papers, creation of a toolkit for collaboration between people living with MS and researchers, and production of conference presentations. As co-researchers team members were named as investigators on grant applications and as co-authors on peer-reviewed papers. 

CONCLUSION

For PwMS, being involved in research can be empowering and provide a meaningful way to contribute to both the research community and the broader MS community. Further evaluation of this model of consumer involvement may elucidate detailed impacts on the research and those involved. 

ACKNOWLEDGEMENT

This research was funded by and has been delivered in partnership with Our Health in Our Hands (OHIOH), a strategic initiative of the Australian National University, which aims to transform health care by developing new personalized health technologies and solutions in collaboration with patients, clinicians and health-care providers.

PEGylation of Gas6 for therapeutic use in demyelinating disease.

BACKGROUND

Gas6 is a TAM (Tyro3, Axl and Mertk) receptor ligand and is an attractive candidate in potentiating remyelination. However, it has a molecular weight of 75kD  limiting its access to the CNS and therefore its potential as a therapeutic agent in multiple sclerosis.

Objective: To modify Gas6 to increase its half-life by adopting an NHS-PEGylation strategy and to assess the bioactivity of native Gas6 and the synthesised derivatives via a viability assay using the rat oligodendroglial cell line, CG4. The efficacy of PEGylated-Cas6 will then be assessed in-vivo. 

METHODS

PEGylation was carried out by reacting 5 kD NHS-PEG (Creative pegworks, USA) with Gas6. The identity of the final products was confirmed by barium iodide staining followed by western blotting. To check for bioactivity, CG4 cells were cultured on Poly-D-Lysine coated wells at a density of ~150 cells/well in different culture conditions. Live/dead cells were counted after 48 hours using Ethidium homodimer/Calcein AM under an inverted florescent microscope, blinded. 

RESULTS

Successful NHS-PEGylation of Gas6 was confirmed by western blotting. In the viability assay, in the absence of insulin (control) there was compromised viability (34.6% viable cells ± 1.4 SE), whereas viability was potentiated by un-modified/native Gas6 (46.7% viable cells ± 1.4 SE). 5kD-PEGylated versions of Gas6 had similar efficacy in promoting the survival of CG4 cells (46.0% viable cells ± 1.4 SE). This effect was abolished in the presence of Axl-Fc which binds soluble Gas6 but remained unchanged in the presence of the irrelevant receptor, IgG- Fc. 

CONCLUSION

NHS-Pegylated versions of the Gas6 protein have been produced, and a robust viability assay established, which is useful in determining the bioactivity of Gas6 derivatives. This is a starting point to compare the efficacy of modified and un-modified Gas6 in promoting remyelination in animal models of MS. 

ACKNOWLEDGEMENT

Negar Asadian is supported by the Australian Government Research Training Program.

Laquinimod as a novel regulator of Kynurenine Pathway in Neural Stem Cells and Monocytes. 

BACKGROUND

Our research uncovered that Kynurenine pathway (KP), the canonical route of amino acid tryptophan metabolism, is dysregulated in MS, producing increasing levels of metabolites corresponding to disease severity. KP enzymes including rate-limiting IDO-1 are activated by interferon-gamma. Nanomolar Quinolinic acid (QUIN) metabolite potently kills oligodendrocytes/neurons. 

OBJECTIVE

Laquinimod is a drug that for relapsing-remitting MS showed clinical efficacy in reducing brain atrophy/relapse risk, suggesting a neuroprotective action. It’s effects on the KP are currently unknown. We aimed to investigate if Laquinimod reduces KP dysregulation and elevated QUIN levels in two relevant celltypes. 

METHOD

Laquinimod was tested on both cultured human monocytes and cultured mouse subventricular zone mouse neural stem cells. Monocytes were treated with/without interferon-gamma. SVZ neurospheres were dissociated and plated for experiment on poly-L-ornithine/laminin-coated dishes. Apoptosis and cell health parameters were analysed on a Muse cytometer. A p value <0.5 was regarded as statistically significant. 

RESULTS

Laquinimod reduced monocyte basal IDO-1 expression by 18.44+6.1% versus control. In interferon-gamma-treated monocytes, Laquinimod increased levels of neuroprotective Kynurenic acid and significantly increased it in 24-hour pre-treated monocytes prior to 72 hours interferon-gamma. Laquinimod significantly reduced neurotoxic QUIN in interferon-gamma-treated monocytes. Apoptosis was reduced in mouse SVZ cells treated with Laquinimod and QUIN, restoring to control. Cell confluence declined with Laquinimod but cell cycle parameters unaltered, suggesting alteration in cell-size. 

CONCLUSION

Monocytes highly upregulate IDO-1 with IFN-gamma, secrete QUIN, cross the blood-brain barrier and localise to lesions. QUIN may further compromise oligodendrocyte survival and neuronal atrophy. Laquinimod was protective on neural stem cells. Enhancing stem cell survival and modulating monocytes underpins possible novel MS therapeutic strategies. 

FUNDING

We acknowledge funding of the Peter Duncan Neuroscience Research Unit at St. Vincent’s Centre for Applied Medical Research.

Whole-blood methylation as a prognostic biomarker for multiple sclerosis.

BACKGROUND

The lack of prognostic biomarkers in multiple sclerosis makes it difficult for clinicians to optimise care. Variation in disease severity is incompletely explained by genetics, suggesting genetic and environmental interactions are involved. DNA methylation is one epigenetic mechanism by which gene-environment interactions can be assessed. 

OBJECTIVES

To date, no epigenetic predictors of severity have been identified. We aimed to identify whole blood and immune cell specific DNA methylation patterns associated with mild and severe relapse-onset multiple sclerosis, and to further test the utility of DNA methylation as a prognostic biomarker. 

METHODS

We compared whole-blood methylation patterns between 119 and 116 relapse-onset females with mild and severe disease, respectively. Severity was determined with longitudinal Age-Related Multiple Sclerosis Severity (ARMSS) scores using MSBase Registry data. Methylation was measured with the Illumina EPIC array, and pre-processed and analysed using the ChAMP package. Cell-type specific analysis was performed with the EpiDISH package. A predictive model was built with glmnet. FDR<0.05 was used to assess significance.

RESULTS

We identified associations between whole-blood methylation and disease severity. 548 differentially methylation positions (DMPs) were identified; 230 DMPs were hypermethylated, and 318 were hypomethylated in the severe group. Sensitivity analyses demonstrated no major effects of covariates on methylation. No cell-specific DMPs were identified. We showed that whole-blood methylation could accurately predict categorical disease severity using methylation levels at 890 CpG sites (AUC=0.91).

CONCLUSION

Whole-blood DNA methylation patterns can be used to accurately predict disease severity in females with relapse-onset multiple sclerosis. DNA methylation has potential as a prognostic biomarker for multiple sclerosis

FUNDING SOURCES

MSRA Project Grant (18-0424) 
RMH Home Lottery Grant MH2013-055
MSBase Foundation Project Grant 
Charity Works for MS Project Grant 
Pennycook Foundation Grant 2018 
Monash University Project Grant

Disease reactivation after cessation of disease-modifying therapy in relapsing-remitting multiple sclerosis. 

BACKGROUND

Multiple sclerosis treatment interruptions are common and leave patients vulnerable to breakthrough disease activity.  The safe duration of the ‘switchover time’ for different therapies is not known. 

OBJECTIVES

To evaluate the rate of return of disease activity after cessation of multiple sclerosis (MS) disease-modifying therapy.  

METHOD

This was a retrospective cohort study from two large observational MS registries: MSBase and OFSEP.  Patients with relapsing-remitting MS who ceased a disease-modifying therapy and were followed up for the subsequent 12-months were included.  The primary study outcome was annualised relapse rate in the 12 months after disease-modifying therapy discontinuation stratified by patients who did, and did not, commence a subsequent therapy.  Predictors of relapse and disability accumulation after treatment discontinuation were a secondary outcome. 

RESULTS

18 029 eligible treatment discontinuation epochs were identified.  Rates of relapse started to increase 2-months after natalizumab cessation.  Commencement of a subsequent therapy within 2-4 months reduced the magnitude of disease reactivation.  After discontinuation of fingolimod, rates of relapse increased overall, and stabilised faster in patients who started a new therapy within 1-2 months.  Predictors of relapse were higher relapse rate in the year before cessation, female sex, younger age and higher EDSS.  

CONCLUSION

Understanding the rate of disease reactivation after discontinuing different MS immunotherapies will help guide optimal wash-out times for therapeutic agents during treatment sequencing. 

United they stand, divided they fall: Machine Learning accurately predicts multiple sclerosis severity where GWAS fails.

BACKGROUND

One of the fundamental, unanswered questions with respect to Multiple Sclerosis (MS), is the source of the underlying disease heterogeneity observed clinically. To-date genetic studies of MS severity have been largely unsuccessful in identifying genetic variants that may explain this heterogeneity in clinical course.

OBJECTIVE

To replicate previously published putative genetic markers of severity. To identify novel single nucleotide variants (SNVs) associated with MS severity. Finally, to determine whether machine learning approaches may be more powerful in finding genetic associations with severity than traditional genome-wide association study (GWAS) approaches. 

METHODS

We genotyped individuals of European ancestry with relapse-onset MS, enrolled in the MSBase Registry, using the Illumina MegaEx array; and imputed against the Haplotype Reference Consortium panel. Fixed-effects meta-analyses tested associations with three MS severity phenotypes: longitudinal age-related MS severity scale (ARMSS), longitudinal MS severity scale, and age at onset. Gradient boosted tree machine learning was used to look for non-linear associations between genetic variants and phenotypes of interest.    

RESULTS

1,813 individuals passed quality control procedures. We replicated one putative non-HLA severity variant rs868824, intronic to IMMP2L, associated with age at onset (b=-1.0935; p=4.31x10-4). No novel SNVs were associated with our longitudinal severity phenotypes (p>5x10-8). Using 62k variants with p<0.01, gradient boosted trees predicted severity based on our longitudinal ARMSS phenotype with 85% accuracy in a testing set that the training algorithm had never seen (70-30% split).

CONCLUSION

Using machine learning we were able to accurately classify individuals as having mild or severe MS where GWAS failed. Our results demonstrate that MS severity is not under strong genetic control. That like MS risk, many variants collectively contribute miniscule amounts to disease severity.

MS Specialist Nurse Care: Patterns of access and impact on health outcomes.

BACKGROUND

MS specialist nurses are an integral part of the multidisciplinary health care team of people with MS. However, changes in traditional funding models have led to a reduction in the number of MS nurses, and concern about inequity of access.

OBJECTIVE

To describe the patterns of access to MS nurses, to examine whether people with MS without access had worse disease outcomes, and to describe the overall value of MS nurses and the perceived benefits of MS nurse care.

METHOD

1,417 participants of the Australian MS Longitudinal Study completed the 2020 MS Nurses Survey and 2019 Disease Course Survey. Participants were asked about MS nurse care access, utilisation (frequency, duration, satisfaction, types of support, consultation type), perceived effects and overall value. Regression models were used to assess associations with the severity of 13 MS symptoms, disability, self-reported progression in the past 12 months, and health-related quality of life. 

RESULTS

We found that 33% of people with MS did not have access to MS nurse care, and that people without access had consistently worse health outcomes compared to those with access, even after adjustment for education, age, MS duration, onset type and remoteness. MS nurses reduced the need for other health professionals and prevented hospital admissions. There were high levels of satisfaction with the frequency and types of services provided.

CONCLUSION

MS nurses seem to offer a service that benefits people with MS, and people who are not able to access this service were adversely affected in terms of health outcomes. It calls for policies and practices to improve the access to MS nurse services.

FUNDING

The Australian MS Longitudinal Study is supported by MS Research Australia.
The AMSLS MS Nurses Study was supported by funding from MS Australia.

BACKGROUND

Difficulties with sleep are common in people with MS (pwMS), and self-reported sleep quality is reduced in pwMS compared to healthy individuals. 

OBJECTIVE

To examine the relationship between sleep-related measures and 13 common MS symptoms, and to determine the independent effect of sleep-related measures on health-related quality of life in a cross-sectional analysis of data 1717 pwMS from the Australian MS Longitudinal Study. 

METHODS

Sleep quality was assessed in the MS Nurses and Sleep questionnaire (2020) using the Pittsburgh Sleep Quality Index (PSQI), daytime sleepiness the Epworth Sleepiness Scale, severity of restless legs the International Restless Legs Syndrome Study Group Rating Scale, insomnia using the Menzies MS Insomnia Index. Quality of life was assessed by the Assessment of Quality of Life questionnaire (AQoL-8D). We analysed data using hierarchical cluster analysis and multivariable linear regression. 

RESULTS

67% of pwMS had poor sleep (PSQI>5). Poor sleep was associated with obesity, greater disability, worse MS symptom scores. MS symptoms clustered into 7 clusters, with sleep-related measures clustering separately. Three clusters were independently associated with PSQI score (‘pain and sensory symptoms’, ‘anxiety and depression’, ‘fatigue and cognitive symptoms’). All measures of sleep quality were associated with AQoL-8D utility scores independent of symptoms and increasing variance explained (Rsq=58.6% to 65.1%). 

CONCLUSION

Sleep quality and dysfunction cluster independently of common MS symptoms despite being correlated. Poor sleep has substantial deleterious effects on health-related quality of life independent of other MS symptoms. 

FUNDING

The Australian MS Longitudinal Study is funded by MS Research Australia.

Novel Plasma and Brain Proteins implicated in Multiple Sclerosis Susceptibility as Potential Targets for Drug Repurposing. 

BACKGROUND
Understanding how variations in the plasma and brain proteome contribute to multiple sclerosis (MS) susceptibility can provide important insights to guide drug repurposing and therapeutic development for MS.  

OBJECTIVE
This proteome-wide analysis aims to identify plasma and brain proteins associated with MS risk, followed by assessing differential gene expression patterns in relevant immune and brain cell types for the associated proteins.  

METHOD
Integrating plasma proteomics (n=3,301) and brain proteomics (n=376 discovery; n=152 replication) into MS genome-wide association studies (n=14,802 cases and 26,703 controls), we employed summary-based methods to identify candidate proteins involved in MS susceptibility. We further examined differential gene expression across five immune cell types (n=73 cases and 97 controls) for MS-associated plasma proteins, and across twenty brain cell clusters (17,799 nuclei from n=4 cases and 5 controls) for MS-associated brain proteins.  

RESULTS
We identified 39 novel proteins associated with MS risk. Based on five identified plasma proteins, we found consistent patterns of differential TAPBPL upregulation and MAPK3 downregulation in immune cells from MS patients. Among the 34 candidate brain proteins, 18 were replicated in a smaller cohort and ten showed robust differential gene expression in functional subclusters of brain cells. Further, we highlighted existing approved therapeutics that target the associated proteins.  

CONCLUSION 
The associated proteins present a robust set of candidate biomarkers and therapeutic targets for MS. Findings from this study highlight the complexity of disease-associated proteins with cell-specific transcriptional profiles, providing useful insights that could inform future studies of drug repurposing and novel therapeutic development. 

ACKNOWLEDGEMENTS 
YZ was supported by a MS Research Australia postdoctoral fellowship as well as an NHMRC investigator grant L1 (GNT1173155). We also thank the investigators who have made their datasets freely available to undertake research of this type.

GluK4-containing kainite receptors impair action potential propagation, synaptic communication and motor function in mice.

Multiple Sclerosis (MS) is an autoimmune and neurodegenerative disease with no clear cause and no known cure.  It is estimated that, depending on the population studied, genetics account for 15–72% of MS risk.  Genome-wide association studies have identified >200 non-MHC loci that are each associated with a very small increase in MS risk (odds ratios ≤1.3) and account for <20% of MS heritability.  We instead study families with an unusually high incidence of MS in which MS-associated genetic signals are amplified.  In one specific family four of seven siblings developed MS and we recently identified a variant in the GRIK4 gene carried by the affected but not the unaffected family members and that is extremely rare in the population.  GRIK4 encodes the Gluk4 high-affinity subunit of ionotropic kainite (glutamate) receptors.  To determine how dysregulated kainite receptor signalling might contribute to MS pathophysiology, we examined the cognitive and motor function of Grik4 knockout (Grik4-/-) mice across adulthood.  Grik4-/- mice were hyperactive and displayed anxiolytic and anti-depressant behaviours, but also developed a progressive motor impairment relative to wildtype mice.  Neuron, oligodendrocyte progenitor cell and oligodendrocyte number were normal.  However, by P180 electron micrographs revealed that the size of the periaxonal space was increased and the callosal axons showed other signs of degeneration.  Whole cell patch clamp recordings from pyramidal neurons in the motor cortex of acute brain slices also revealed a ~50% reduction in the number of excitatory synaptic inputs received, which could readily account for the impaired motor function.  Additionally, the amplitude of the myelinated callosal compound action potential was reduced by ~30% in Grik4-/- mice relative to wildtype mice.  Additional pharmacological studies revealed that this was the result of impaired potassium channel function in Grik4-/- mice. These data suggest that Gluk4-containing kainite receptors are critical for action potential maintenance and synaptic health in the motor cortex indicating that dysfunctional glutamate signalling can directly impair axon health.

Acceptability study of ELEVIDA, an online intervention for fatigue in multiple sclerosis.

BACKGROUND

Fatigue is one of the most common and disabling symptoms reported by people with multiple sclerosis (MS), yet effective management is limited. A German version of ELEVIDA, based on the principles of cognitive behaviour therapy (CBT), has shown to be effective in reducing MS-related fatigue. 

OBJECTIVE

To pilot test the acceptability, feasibility and suitability of use in Australia of an English version of ELEVIDA.

METHOD

We recruited volunteers with MS aged 18 years and over who self-reported at least mild fatigue (scoring ≥43 on the Fatigue Scale for Motor and Cognitive Functions scale), mild or no self-reported cognitive difficulties (scoring ≤32 on the Multiple Sclerosis Neuropsychological Screening Questionnaire) and some mobility (EDSS<8). Participants were asked to ‘think aloud’ and provide feedback when completing the nine online conversations. 

RESULTS

Fifteen people living with MS-related fatigue completed the ELEVIDA program (mean age=58.87 years, SD=10.52, 10 females, 5 males, 9 with relapsing MS, 6 with progressive MS). Over 90% of the participants rated the ELEVIDA program as acceptable, good or very good; 80% of participants said they would recommend ELEVIDA to others living with MS-related fatigue, and almost 70% of participants found the ELEVIDA program helpful for treating their fatigue. 

CONCLUSION

The English version of ELEVIDA is acceptable, feasible and helpful for people living with MS-related fatigue. Participant ratings and feedback from this study will contribute to the optimisation of the English version of ELEVIDA for larger studies aimed at improving MS-related fatigue in Australia.

Epstein-Barr virus infection dysregulates the MS risk miRNA expressions in B cells. 

BACKGROUND 

Although over 200 risk genetic loci have been identified for Multiple Sclerosis (MS), the causes of MS largely remain unknown. However, Epstein-Barr virus (EBV) infection has identified as necessary but not sufficient to cause MS. 

OBJECTIVE 

To interrogate the host response to EBV infection that contributes to MS, we aimed to identify the potential contribution of EBV-encoded and host cellular miRNAs to MS pathogenesis. 

METHODS

We identified differentially expressed host miRNAs in EBV infected B cells (LCLs) and putative host-EBV miRNA interactions with MS risk loci. We estimated the genotype effect of MS risk loci on the identified putative miRNA:mRNA interactions in silico. Lastly, we tested the genotype effect of the MS risk SNP rs10271373 on expression of ZC3HAV1, an MS risk gene with antiviral functions.

RESULTS  

We found that hsa-let-7b-5p may interact with ZC3HAV1 differently in LCLs compared to B cells. In vitro assays indicated that the protective allele of rs10271373 increases ZC3HAV1 expression in LCLs, but not in B cells, which is consistent with increased interferon response, promoting viral mRNA degradation, and repressing viral replication process via ZC3HAV1 and so decreased immune evasion by EBV. 

CONCLUSION 

Taken together, this provides evidence that EBV infection dysregulates the B cell miRNA machinery, including MS risk miRNAs, which may contribute to MS pathogenesis via interaction with MS risk genes either directly or indirectly.

Effect of multiple sclerosis disease modifying therapies, natalizumab, ocrelizumab, fingolimod and cladribine, on peripheral immune cell counts. 

BACKGROUND

Varying effects on peripheral immune cell counts after DMT have been shown. Alterations in cell counts are generally on target effects in keeping with the DMT mechanism of action. Variations in other cells counts are possible off target effects that may be of significance. 

OBJECTIVE

Monitor peripheral immune cell counts and immune cell ratios in multiple sclerosis (MS) patients on ocrelizumab, natalizumab, fingolimod and cladribine to decipher the effect of each disease modifying treatments on counts over time. 

METHOD

Our study included MS patients from Royal Melbourne Hospital and Box Hill Hospital. Peripheral immune cell counts done as part of routine clinical care was retrospectively collected, and the data was entered into iMed (MSBASE) patient register. Immune cell counts of patients treated with natalizumab (n=268 patients), ocrelizumab (n=185), fingolimod (n=265) and cladribine (n=48) were extracted. Data was subsequently analysed using piecewise segmented regression models for pre- and post-treatment. 

RESULTS

After 2 years, peripheral lymphocyte counts were shown to significantly increase with natalizumab (p=8×10−32) and decrease with fingolimod (p=2×10−30) or ocrelizumab therapy (p=0.02). Natalizumab and fingolimod also significantly reduced neutrophil (p=0.006; p=0.00002) and eosinophil counts (p=2×10−9; p=0.00004). Cladribine was shown to cause a reduction in peripheral monocyte count (p=0.00009). Effect on neutrophil-to-lymphocyte ratio (NLR) and the monocyte-to-lymphocyte ratio (MLR) biomarkers of each treatment was shown.

CONCLUSION

ocrelizumab, fingolimod, natalizumab and cladribine affect different peripheral immune cell types in varying degrees, with possible implications on efficacy and safety monitoring of each drug.

Smoking, infectious mononucleosis, early-life exposures and risk of progressive-onset Multiple Sclerosis: a case-control study. 

BACKGROUND

Risk factors in Multiple Sclerosis (MS) have been identified in samples that contain predominantly cases with relapse-onset MS (ROMS). It is unknown whether the risk factors differ in people with a progressive-onset MS (POMS). 

OBJECTIVE

Examine the association between smoking, infectious mononucleosis, early life exposures and POMS and compare the magnitude of association with POMS. 

METHOD

The case-control study included 147 POMS cases, 264 ROMS cases and 558 community controls. Questionnaires were used to collect information on smoking, infections, breastfeeding and vitamin use in early life. Logistic regression was used to examine associations. Odds ratios (ORs) were adjusted for age, sex and latitude band. 

RESULTS

Compared with controls, POMS cases were more likely to have smoked ≥20 pack-years (AOR 2.83, 1.53-5.24) and ROMS cases were more likely to have ever been a smoker (AOR 1.48, 1.09-2.01). POMS cases (AOR 1.75, 1.05-2.93) and ROMS cases (AOR 1.74, 1.21-2.49) were both more likely to have had infectious mononucleosis. In childhood, POMS cases were less likely to use vitamin supplements (AOR 0.52, 0.30-0.91) or being breastfed (AOR 0.49, 0.31-0.79). 

CONCLUSION

POMS and ROMS share some commonly known risk factors. Some protective factors were identified in POMS that were not observed in ROMS.

Ethan G. Hughes is an Assistant Professor in the Cell and Developmental Biology Department at the University of Colorado School of Medicine. He leads a laboratory focused on understanding how neuron-glial interactions modulate brain function and contribute to pathology in neurodegenerative disease. Dr. Hughes obtained his B.S. degree in Biology with Special Honors from George Washington University (2003) and his Ph.D. in the laboratory of Rita Balice-Gordon at the University of Pennsylvania (2009) studying the mechanisms of synapse formation and maintenance. As a postdoctoral fellow under the mentorship of Dwight Bergles at Johns Hopkins University, Dr. Hughes examined cellular dynamics of oligodendrocyte lineage cells and myelination in the adult brain using longitudinal in vivo two-photon microscopy. In 2016, Dr. Hughes joined the faculty at the University of Colorado School of Medicine. He is an awardee of the Whitehall Foundation Fellowship, Marilyn Hilton Award for Innovation in MS Research, Webb-Waring Biomedical Research Award, and a ACTRIMS Young Investigator Award. He serves on study sections for the National Multiple Sclerosis Society and Congressionally Directed Medical Research Programs, Multiple Sclerosis Research Program for the Department of Defense.

Bruce joined the University of Colorado and the Section of Developmental Biology in 2008 after having begun his faculty career at Vanderbilt University in 1998. He got his training as a Drosphila developmental biologist during his Ph.D. work at the University of Utah and as a zebrafish developmental neurobiologist as a postdoctoral fellow at the University of Oregon. His lab at the University of Colorado investigates the genetic, molecular and cellular basis of brain development and myelination using zebrafish as a model system. Particularly areas of interest in the lab currently include mechanisms that guide specification of neural progenitors for oligodendrocyte fate, mechanisms that regulate formation of specific amounts of myelin on select axons during development, and the role of microglia in shaping developmental myelination.

Short-term myelin loss is sufficient to alter neuronal synapses: insight into disease progression in MS.

BACKGROUND

For information processing and cognition, connections between neurons in brain circuits adapt in response to experience (synaptic plasticity). People with MS can experience cognitive and sensorimotor processing symptoms, which has historically been attributed to myelin loss and slower action potential conduction. It is also possible that myelin loss impacts synaptic communication, however, this has not been directly explored. 

OBJECTIVE

Investigate how short-term myelin loss affects synapses in excitatory neurons, using a transgenic mouse model of “primary progressive” demyelination. 

METHODS

Myelin loss was initiated in adult mice by conditionally deleting myelin regulatory factor (Myrf) from oligodendrocytes (Myrf-deleted).  In separate cohorts, we performed immunohistochemistry to detect oligodendrocytes; assessed performance in motor tasks; and measured compound action potentials (CAPs) in the corpus callosum of Myrf-deleted and control mice.  We crossed Myrf-deleted with Thy1-YFPH mice to image a subset of pyramidal neurons and quantify post-synaptic structures (dendritic spines). We assessed synaptic functional changes using whole cell patch-clamp electrophysiology, measuring mini post-synaptic currents (mPSCs) in control and Myrf-deleted motor cortical neurons. 

RESULTS

Myrf-deleted mice developed a progressive motor phenotype which was associated with significant reduced number of ASPA+ myelinating oligodendrocytes and reduced myelinated axon CAP amplitude.  Pyramidal neurons in the motor cortex of Myrf-deleted mice received similar number of inhibitory and excitatory signals (mIPSC and mEPSC frequency), and similar density of dendritic spines, suggesting the number of synaptic connections is unchanged. However, in Myrf-deleted mice a higher proportion of dendritic spines were large (mushroom) spines, suggesting that they respond more strongly to glutamate. 

ONCLUSION

Short term myelin loss results in the inappropriate strengthening of synapses between neurons, enhancing excitability. This early change may make neurons vulnerable to irreversible excitotoxic damage, revealing a potential mechanism linking myelin and neuron loss leading to disease progression in MS. 

FUNDING

NHMRC (1030939, 1045240 and 1077792); ARC DP180101494; MS Research Australia (11-014, 15-054, 16-105, 19-0696; 17-0223).

How do microglia influence myelination?: Investigating the role of microglial Mertk on oligodendrogenesis.

BACKGROUND

Microglia have recently been identified as regulators of myelination. Two proposed mechanisms by which they do this are by refining myelin and releasing growth factors to support oligodendrocyte survival and differentiation. Mertk is a receptor expressed on microglia that mediates immunoregulatory functions including myelin phagocytosis.

OBJECTIVE

Given the involvement of microglial in supporting oligodendrocyte lineage cells, and the role of Mertk in microglial function, this project aimed to investigate the influence of microglial Mertk on oligodendrocyte development and subsequent myelination.

METHOD

We generated Mertk conditional knockout (cKO) mice, in which Mertk is deleted from Cx3Cr1-expressing cells, including microglia. To mark newly-generated cells during development, we administered the thymidine-analogue EdU over the course of 3 days, before collecting brain tissue at postnatal day(P) 8 and P28. We performed immunostaining to determine numbers of newly-generated microglia, oligodendrocyte precursor cells (OPCs) and mature oligodendrocytes. We also analysed the ultrastructure of cKO myelin throughout aging.

RESULTS

At P8, Mertk-deficient animals showed a reduction in newly-generated cells, however this was not due to any changes in microglial proliferation. At P28, cKO mice also had reduced proportions of newly-differentiated mature oligodendrocytes compared to WT, suggesting impaired OPC proliferation and/or maturation. At 8-10 weeks, cKO myelin appeared decompacted, suggesting an enduring effect of Mertk-deletion on the oligodendrocyte population. Despite abnormal myelination, microglia, OPC and mature oligodendrocyte numbers were unchanged.

CONCLUSION

These data highlight the importance of microglia in myelination and identify microglial Mertk as an important regulator of oligodendrogenesis. Further work is ongoing to determine the exact mechanism(s) by which this occurs, and the consequences of this developmental phenotype on the outcomes of demyelinating disease.

FUNDING

Linda Nguyen is supported by the Australian Commonwealth Government Research Training Program Scholarship.

Demyelination modifies communication between neurons and oligodendrocyte progenitor cell.  

BACKGROUND

Multiple Sclerosis lesion repair involves the generation of oligodendrocytes from oligodendrocyte progenitor cells (OPCs).  OPCs receive synaptic inputs from neurons, allowing them to monitor and respond to neuronal activity.  The number of synaptic connections can change with demyelination and may regulate myelin repair.  

OBJECTIVE

To characterise neuron to OPC synaptic connectivity following demyelination and during remyelination. 

METHOD

A subset of OPCs was fluorescently labelled by injecting a monosynaptic variant of the rabies virus (SADΔG-GFP-EnvA) into the corpus callosum of cuprizone demyelinated adult mice. Specifically, Pdgfrα-CreERT2::Rosa26-gp4-TVA mice received 0.2% (w/w) cuprizone for 5-weeks and were injected with virus 24h after cuprizone withdrawal.  Once injected, the virus moved across the synapse to label all connected neurons.  The number and distribution of labelled cells was quantified in immunolabelled tissue sections. 

RESULTS

In control mice, ~1-20 callosal OPCs and ~400-1900 cortical and subcortical neurons were labelled, with ~84% of inputs received from neurons in the ipsilateral hemisphere.  A large proportion of the labelled neurons (~59%) were CA1 hippocampal neurons.  Following demyelination, the number of GFP+ neurons decreased by ~80%. OPCs received ~8% fewer inputs from neurons in the contralateral cortex and ~45% fewer inputs from hippocampal neurons. 

CONCLUSION

These data suggest that immediately following demyelination, callosal OPCs are less likely to form synaptic connections with axons.  As the cuprizone model of demyelination rapidly demyelinates the axons of cortical neurons (more than hippocampal neurons), OPCs may prioritise synaptic connections with demyelinated neurons. 

FUNDING

Dementia Australia, MS Research Australia, NHMRC, ARC, MRFF.

Innate immunity impacts social-cognitive functioning in people with multiple sclerosis and healthy individuals: implications for IL-1ra and urinary immune markers. 

BACKGROUND

Social-cognitive difficulties are being increasingly recognised as a possible outcome of multiple sclerosis (MS), a chronic and progressive neuroinflammatory and neurodegenerative disease.  

OBJECTIVE

In this pilot investigation, we examined the relationship between various serum- and urine-derived pro- and anti-inflammatory markers (i.e., cytokines, chemokines, and growth and colony-stimulating factors) and social-cognitive functioning in people with MS and demographically-matched healthy individuals.  

METHOD

Participants comprised 20 pwMS (17 relapsing-remitting, 1 secondary-progressive, 2 primary-progressive) and 20 healthy control individuals without MS, demographically matched for sex, age, and education.  Social cognitive functioning was objectively assessed using The Awareness of Social Inference Test – Short (TASIT-S) and subjectively assessed using self-reports of abilities in emotion recognition, emotional empathy, and cognitive theory of mind.  

RESULTS

In MS, higher serum concentration of pro-inflammatory markers IL-1β, IL-17, TNFα, IP-10, and MIP-1α, and higher urine concentration of GM-CSF, IP-10, and MIP-1β were associated with poorer social-cognitive functioning.  Higher serum concentration of the anti-inflammatory marker IL-1ra was associated with better social-cognitive functioning (i.e., self-reported emotional empathy and TASIT-S sarcasm detection performance).  The findings were mixed for the anti-inflammatory serum markers IL-4 and IL-1 0, i.e., objective versus subjective social-cognitive ability. 

CONCLUSION

The findings suggest that biologically-derived disease-related inflammatory processes, distinct from inflammatory dysregulation that occur during an MS relapse, may, in part, underlie difficulties in social-cognitive abilities in people with MS.

FUNDING

The authors acknowledge the importance of funding from a Multiple Sclerosis Research Australia (MSRA) Incubator Grant (2016-17) awarded to Cynthia A. Honan, Chai K. Lim, and Skye McDonald.

Exploring the utility of MOG antibody epitopes in predicting a relapsing disease course.  

BACKGROUND

Autoantibodies against myelin oligodendrocyte glycoprotein (MOG) have been used to define a novel class of CNS demyelinating disorders. In adults, these disorders mainly affect the optic nerve and spinal cord, with approximately 50% of patients relapsing. Currently, there is no method of predicting relapse.

OBJECTIVE

We aimed to identify the prognostic utility of two key MOG antibody epitopes, Proline42 (P42) and Histidine103/Serine104 (H103/S104), in predicting a relapsing disease course.

METHOD

A flow cytometry live cell-based assay was utilised to determine the level of antibody binding to wild-type MOG as well as two mutants of MOG, P42S and H103A/S104E, in 306 MOG antibody-positive adults. Clinical information including lesion localisation and disease course were collected from 177 of these patients.

RESULTS

Of the patients who bound to a non-P42 epitope, 70% exhibited a relapsing disease course. This was enriched in patients with unilateral optic neuritis (UON), where 94% of non-P42 patients relapsed. This is notable as UON patients account for a third of patients with a relapsing course in our cohort. In contrast, preliminary evidence suggests that for patients with transverse myelitis, a non-H103/S104 epitope may help predict a relapsing course.

CONCLUSION

Our research supports the prognostic utility of MOG antibody epitopes in predicting a relapsing course, which will enable clinicians to modify treatment in susceptible patients.

CLADIN: CLADribine and INnate Immune responses. 

BACKGROUND

Cladribine (Mavenclad®) is a purine analog, and an oral treatment for relapsing remitting MS (RRMS). The mechanism of action of Cladribine on peripheral innate immune cells, in particular monocytes, is unclear, and forms the basis of this study. 

OBJECTIVE

The primary objective is to examine the effect of Cladribine on innate immune responses. The secondary objective is to assess patient outcomes (EDSS, relapse, new lesions on MRI), and evaluation blood biomarkers of disease activity over time. 

METHOD

This is Phase IV, multi-centre, 3 year trial. Patients planned for commencement of Cladribine are recruited into the study. Peripheral innate immune phenotypes (monocyte numbers and activation states) were measured at various times prior and after commencement of therapy. In addition, 14 cytokines and neurofilament light is planned to be measured over the course of treatment. The findings are linked to clinical outcomes (EDSS, relapse, MRI activity). 

RESULTS

At the time of submission, 35 patients have been recruited into the study. There was a significant reduction in monocyte count at week 1 compared to pre-treatment levels (p<0.005) which recovered by 2 months post medication administration. While monocytes expressing HLA-DR and CD14 remained relatively stable, those expressing CD16 (i.e. the ‘proinflammatory’ or ‘non-classical’) monocytes decreased at week 1 (p<0.005).  

CONCLUSION

This study demonstrates a novel mechanism of action for Cladribine, highlighting that it exerts its effects acutely on peripheral monocytes. The laboratory data will be linked to clinical data to decipher what innate immune parameters translate to better patient outcome.

Professor Godfrey is an NHMRC Investigator Fellow, a Fellow of the Australian Academy for Health and Medical Sciences, Past President of the Australasian Society for Immunology and founder and Past President of the Melbourne Immunotherapy Network. His area of interest is the field of unconventional T cells including their development, function and therapeutic potential. He has also led a research program to develop a SARS-CoV-2 vaccine which is about to enter into a phase I clinical trial. Godfrey has published over 250 scientific papers that have collectively been cited over 36000 times and he is a Clarivate Highly Cited Researcher.

Targeting Epstein-Barr virus-encoded EBNA2 with novel antisense oligonucleotides as a therapeutic strategy for MS.

BACKGROUND

Multiple Sclerosis (MS) involves over 200 genetic risk variants and environment factors. Epstein-Barr virus is considered necessary for MS and the EBV protein EBNA2 is implicated in MS. Yet, no EBV-targeting therapies exist and the benefits of targeting EBV in MS are unclear. 

OBJECTIVE

The objective of this study was to develop antisense oligonucleotides (ASOs) to specifically target EBNA2 in EBV-infected B cells, to provide a proof-of-concept for the targeting of EBV associated diseases including MS. 

METHOD

ASOs were designed to target mature EBNA2 mRNA. The pipeline for choosing the EBNA2 target sequence and ASOs was based on RNA accessibility and the predicted specificity of the target sequence. The ASOs were tested in EBV-infected B cells, lymphoblastoid cell lines (LCLs), for reduction of EBNA2 mRNA and EBNA2-regulated genes. 

RESULTS

The most effective ASO (ASO2) reduced EBNA2 mRNA expression and reduced LCL proliferation and EBV DNA copy number, indicating disruption to the EBNA2-driven latency III growth phase. The EBV oncogene LMP1, and LMP2A, were reduced by ASO2, and the MS risk gene TRAF3 was significantly altered.  ASO disruption of the EBV latency III growth programme could be beneficial in MS as it is implicated in MS risk. 

CONCLUSION

This work indicates that EBNA2 targeting by small nucleic acid inhibitors could provide a therapeutic strategy to target EBV generally or more specifically could be used in EBNA2-related autoimmune disorders including MS.

Identifying genetic and epigenetic signatures for predicting conversion to clinically definite multiple sclerosis in the AusLong study cohort.

BACKGROUND

Being able to accurately predict the conversion from clinically isolated syndrome (CIS) to clinically definite MS (CDMS) at the time of first neurological event is vital for initiating the optimal course of treatment at the early stages of disease. 

OBJECTIVE

To use an ensemble machine-learning approach to identify genetic and epigenetic signatures that can predict conversion to CDMS in a longitudinal study of an Australian cohort (AusLong Study). 

METHOD

The study included 241 participants at first neurological event (baseline). After 10 years, 32 participants remained CIS whilst 190 had converted to CDMS. Genome-wide SNP genotype and CpG methylation data were obtained in participants at baseline using GSA and EPIC arrays, respectively. To classify groups, linear and non-linear regression models were trained using GLMNet and XGBoost algorithms, respectively. The predictive accuracy of models was tested on unseen data using AUCs. 

RESULTS

The best model for GLMNet selected 15 features (all SNPs) and yielded an AUC of 0.72. XGBoost selected 89 features including 19 SNPs and 64 CpGs and yielded an AUC of 0.76. Regressing the conjoint effects of the GLMNet and XGBoost scores on outcome resulted in a final AUC of 0.79 (95% CI: 0.65 – 0.94, P=0.003). The signature showed gene enrichment for neurogenesis processes and the G-protein signalling pathway (P<0.005). 

CONCLUSION

Using machine-learning algorithms this study revealed a genetic-epigenetic signature associated with conversion to CDMS in the Auslong cohort. Future work should include all available baseline data (clinical and environmental) to maximise predictive value. Fully optimised models will require validation in an independent cohort.

Impact of pregnancy history on whole-blood methylation in women with relapse-onset multiple sclerosis.

BACKGROUND

There is evidence that pregnancy is beneficial to women with multiple sclerosis (MS) by reducing disease progression and associated disability. The mechanism that drives this effect is unknown, but converging evidence suggests a role for epigenetic mechanisms altering immune and central nervous system function. 

OBJECTIVE

The effect of pregnancy on the epigenome is understudied in MS. Identifying the molecular mechanisms responsible could identify novel therapeutic targets. We aimed to identify whole blood and immune cell specific DNA methylation patterns associated with parity in relapse-onset multiple sclerosis. 

METHOD

We compared whole-blood methylation between 96 pairs of nulliparous and parous females, matched by age and Age-Related Multiple Sclerosis Severity (ARMSS) scores (n=192). Methylation was measured with the Illumina EPIC array, and pre-processed and analysed using the ChAMP package. Cell-type specific analysis was performed using the EpiDISH package. Methylation age was calculated with the methyAge package. Methylation age acceleration (MAA) was calculated by regressing methylation age on chronological age. 

RESULTS

We identified associations between whole-blood methylation and parity. Parous females showed hypomethylation at genes enriched in protein-protein interactions at synapses and axon guidance pathways. Cell-type specific analysis showed different monocyte proportions between groups, but no cell-specific differential methylation. Parity reduced median MAA by 1.7 to 2.1 years. The median time from last pregnancy to blood collection was 16.7 years, demonstrating long-term effects of pregnancy on methylation age. 

CONCLUSION

Whole-blood methylation patterns are associated with parity in females with relapse-onset multiple sclerosis. The reduced MAA in parous females suggests a potential molecular mechanism underpinning the long-term effect of pregnancy in women with multiple sclerosis. 

FUNDING

MSRA Project Grant (18-0424)
RMH Home Lottery Grant MH2013-055
MSBase Foundation Project Grant
Charity Works for MS Project Grant
Pennycook Foundation Grant 2018
Monash University Project Grant

Serine and kynurenine metabolites in genetic susceptibility of multiple sclerosis. 

BACKGROUND

Metabolites are small molecules reflecting a spectrum of biological processes. Several metabolites including kynurenine have been studied in multiple sclerosis (MS), though their causal relationship with MS remain unclear. Understanding the causality between specific metabolites and MS risk can help identify diagnostic and prognostic biomarkers. 

OBJECTIVE

We systematically screened 174 metabolites of seven groups (amino acids, acylcarnitines, phosphatidylcholines, hexoses, biogenic amines, lysophosphatidylcholines, sphingomyelins). Among 29 metabolites with sufficient number of genetic instruments, we aimed to assess their genetic causal relationships with MS risk using six different Mendelian randomisation (MR) methods. 

METHOD

Summary data from a cross-platform metabolomics study (n=8,569-86,507) and MS genome-wide association studies (n=14,802 MS cases, 26,703 controls) were first analysed using generalised summary-data-based Mendelian randomisation, followed by sensitivity analyses using five MR methods (Mendelian randomisation-Egger, inverse-variance weighting, weighted median, weighted mode, and causal analysis using summary effect estimates). We first examined the effects of metabolites on MS susceptibility, and then assessed whether MS affected the metabolite levels. 

RESULTS

We found strong genetic evidence of serine level positively influencing MS risk (odds ratio=1.67, p=1.73x10-20), and a positive association between kynurenine level with MS susceptibility (odds ratio=1.30, p=1.80x10-5). Upon sensitivity analysis, we observed consistent association signals with MS risk for serine metabolite level but not kynurenine. Additionally, MS had a nominal positive effect on kynurenine level but not serine, which supported a pleiotropic association between kynurenine and MS risk. 

CONCLUSION

Serine metabolite was found to positively influence MS risk robust across different MR models, while the results also revealed a complex relationship between kynurenine and MS. The identified metabolites are candidate biomarkers that warrant further investigations to confirm their roles in MS aetiology.

Yasmine Probst is an Associate Professor with the School of Medicine at the University of Wollongong and Research Fellow at the Illawarra Health and Medical Research Institute. She holds dual Masters degrees in Dietetics and Health Informatics and is recognised as an Advanced Accredited Practising Dietitian with the Dietitians Association of Australia and a Fellow of the Australasian College of Health Informatics. Yasmine teaches a key research theory subject within the Nutrition and Dietetic programs at the University of Wollongong. As a person living with MS, her research and teaching focuses on nutrition for people living with MS.

A phase II randomised controlled trial of an early tailored cognitive behavioural therapy based intervention for depression in those newly diagnosed with multiple sclerosis (action-ms): preliminary results. 

BACKGROUND

Depression is high in those newly diagnosed with multiple sclerosis. There is a lack of research examining the effectiveness of tailored interventions for depression in individuals newly diagnosed with multiple sclerosis. 

OBJECTIVE

We examined the effectiveness and cost-effectiveness of an 8-week tailored cognitive behavioural therapy (CBT) based psychological intervention for depression and related factors such as anxiety, fatigue, pain and sleep difficulties in those newly diagnosed with multiple sclerosis. 

AIMS OF THE STUDY

To assess efficacy of the CBT intervention, compared to supported listening (SL), in producing significant and clinically meaningful reductions in depressive sand anxiety symptoms, and levels of fatigue, sleep disturbance, and pain impact; and increases in physical and mental health related quality of life, illness acceptance, social support, coping and resilience. The study will also assess cost-effectiveness of providing the CBT intervention.

METHOD

This study is a two-arm parallel group randomised controlled trial with the  primary site being the Royal Melbourne Hospital. To date, 38 participants have been randomly allocated to the tailored CBT (n=20) and SL (n=18) interventions. Both interventions consist of 1-hour sessions and are delivered over 8 weeks by clinical psychologists. Clinical interview and self -report data are collected at pre, post and 3-month follow up time points. 

PRIMARY OUTCOME

Clinically significant change of 10 points or more on the BDI-II at post-intervention.

SECONDARY  OUTCOME

Level of depression at 3-months follow up and level of anxiety at post-intervention.

TERTIARY OUTCOME

Level of anxiety at 3-months follow up,  pain and fatigue impact, MS illness acceptance, sleep quality, coping, social support, resilience at post-intervention and 3-months follow up. 

METHOD

Presently, we have recruited 38 participants from MS hospital clinics around Melbourne into the randomised controlled trial who were within 5 years of having received a diagnosis of MS and scored within the mild to moderate categories of the BDI-II. 

RESULTS

We found that the CBT-based intervention significantly reduced level of depressive symptoms compared to the supportive listening intervention at post-intervention and that this was maintained at a 3-month follow-up intervention. 

CONCLUSION

The CBT-based tailored intervention is an effective intervention in the treatment of depression, anxiety, fatigue, pain and sleep disturbance and should be offered to individuals who have been newly diagnosed with multiple sclerosis.

Are the eyes the window to the MS brain? Delineating the influence of visual processing impairment on cognitive function. 

BACKGROUND

Visual impairments are highly prevalent in MS, reflecting deficits across 3 visual processing stages: (1) afferent transmission of visual information, (2) cognitive processing and (3) efferent generation of an eye movement. However, it is unclear how deficits in each of these stages impacts cognitive function. 

OBJECTIVE

This study sought to determine whether different visual processing phenotypes exist in MS, based on the presence of deficits in the 3 stages of visual processing, and how cognitive function is differentially affected in each visual processing phenotype. 

METHOD

150 MS patients underwent comprehensive assessment of afferent visual processing (optical coherence tomography, visual evoked potentials and visual acuity), cognitive visual processing (cognitive ocular-motor assessment) and efferent visual processing (eye-movement irregularities ­- INO, nystagmus, diplopia). Clinical assessment included the EDSS, MRI, SDMT and BDI. To understand the relationship between visual processing and cognition, visual processing phenotypes were identified using latent-class modelling and inter-phenotype differences in clinical outcomes were examined. 

RESULTS

Preliminary results indicate that afferent visual processing impairment (thinning of retinal nerve fibre) more greatly affects SDMT performance, compared with cognitive visual processing impairment (cognitive ocular-motor performance). Given this finding, we predict that distinct visual processing phenotypes exist in MS that are differentially associated with cognition. Furthermore, we anticipate that these visual processing phenotypes will significantly differ in lesion volume, lesion location, atrophy in visual networks, disability, fatigue and depression. 

CONCLUSION

The identification of distinct MS visual processing phenotypes may help to determine whether patients experience predominantly vision-based or cognition-based difficulties in their everyday functioning. Furthermore, characterisation of specific cognitive
phenotypes may assist in the selection of tailored interventions for patients, such as targeted cognitive rehabilitation.

FUNDING

Sanofi Genzyme

Worsening longitudinal reaction time trajectories using the MSReactor computerised battery predicts confirmed EDSS progression. 

BACKGROUND

Longitudinal cognitive trajectories in MS are heterogenous and difficult to measure. Computerised tests designed to screen broad cognitive domains may be a reliable method to detect trajectories of cognitive performance. 

OBJECTIVE

To validate a latent class model to identify trajectories of reaction time change in people with relapsing remitting MS (pwRRMS).

METHOD

The MSReactor battery computerised cognitive battery assesses psychomotor function (SiRT), visual attention (ChRT) and working memory (OBK). We used Latent class analysis (LCA) to model longitudinal reaction times and classify participants into discrete cognitive trajectories. We determined the minimum number of tests required to accurately classify an individual into a longitudinal trajectory. Survival analyses was performed for inter-class differences in time to 6-month confirmed disability progression (CDP) and time to sustained reaction time change.    

RESULTS

pwRRMS (n=460) with at least 3 MSReactor tests over minimum of 180 days follow up were included. Three latent trajectories were identified for each task. Just 3, 4 and 5 tests per participant could classify participants into trajectories with 81%, 61% and 64% accuracy for the tasks, respectively. All tasks detected a group of pwRRMS with increased probability of sustained reaction time worsening. The ChRT and OBK tasks identified a group of pwRRMS who had an increased probability of 6-month CDP (HR 3.7 and HR2.6, respectively). 

CONCLUSION

Latent class modelling of longitudinal reaction times can detect discrete trajectories of cognitive performance. All models identified a group of RRMS more likely to sustain significant reaction time changes. The ChRT and OBK tasks identified a group of RRMS more likely to experience disability progression

Associations of DMT therapies with COVID-19 severity in multiple sclerosis: an international cohort study. 

BACKGROUND

People with multiple sclerosis (MS) are a vulnerable group for severe COVID-19, particularly those taking immunosuppressive disease-modifying therapies (DMTs). 

OBJECTIVE

To assess characteristics of COVID-19 severity in an international sample of people with MS, including hospitalisation, ICU admission, requiring artificial ventilation, and death. 

METHOD

Data from 12 data-sources in 28 countries were aggregated. Demographic and clinical covariates were queried, alongside COVID-19 clinical severity outcomes, hospitalisation, admission to ICU, requiring artificial ventilation, and death. Characteristics of outcomes were assessed in patients with suspected/confirmed COVID-19 using multilevel mixed-effects logistic regression. 

RESULTS

657 (28.1%) with suspected and 1,683 (61.9%) with confirmed COVID-19. Older age, progressive MS-phenotype, and higher disability associated with worse COVID-19 outcomes. Ocrelizumab and rituximab associated with hospitalisation (aOR=1.75,95%CI=1.29-2.38; aOR=2.76,95%CI=1.87-4.07) and ICU admission (aOR=2.55,95%CI=1.49-4.36; aOR=4.32,95%CI=2.27-8.23) vs pooled-other-DMTs but only rituximab with artificial ventilation (aOR=6.15,95%CI=3.09-12.27). Similar associations seen compared to dimethyl fumarate and to natalizumab. No associations observed between DMTs and death. 

CONCLUSION

Using the largest cohort of people with MS and COVID-19 available, we demonstrated consistent associations of rituximab and ocrelizumab with worse COVID-19, suggesting their use may be a risk factor for more severe COVID-19.

Novel placement of wearable sensors detects subclinical gait instability in non-disabled people with multiple sclerosis in clinical setting using calculated local divergent exponent. 

BACKGROUND

The local divergence exponent (LDE) measures gait stability, the ability to deal with step-to-step perturbations. We recently showed lab-based LDE is sensitive to subclinical gait deterioration in people with MS (PwMS). However, studies using the LDE in PwMS have not been conducted in clinical settings. 

OBJECTIVE

We aim to examine LDE sensitivity utilizing wearable sensors during overground walking in a clinical setting to provide known-groups validation, compare output from novel sensor placement at the sternum with traditional placement at the sacrum, and explore covariates to further develop construct validity. 

METHOD

51 PwMS (relapsing-remitting, EDSS 0-4) and 23 age-matched healthy controls (HC) walked for 6 minutes along a 20m walkway while wearing APDM sensors (Oregon, USA). LDE was calculated using 3D acceleration data (sacrum and sternum sensors) using Matlab (Natwick, USA). One-way ANOVA was used to determine group differences. Post hoc tests were conducted to examine differences between EDSS-categorized subgroups. Correlations between significant outcome measures and clinical outcomes were also performed. 

RESULTS

LDEsacrum and LDEtrunk were significantly different between PwMS and HC (F(1,74)=4.29, p=0.042 and F(3,33=9.45, p<0.001, respectively). LDEtrunk was significantly different between HC and all groups, including MS-EDSS0-1 (mean difference 0.06, p=0.03). Using LDE calculated over each axis, LDE-ML (mediolateral) correlated with EDSStotal (LDEsacrum-ML: rs=0.40,p<0.01; LDEtrunk-ML: rs=0.47,p<0.01) particularly Functional Sensory scores (Sacrum rs=0.32, p=0.02; Trunk rs=0.38, p=0.01). Pyramidal scores correlated highest with LDEsacrum-AP‎ (rs=-0.44, p=0.001) and Cerebellar with LDEtrunk-AP (rs=-0.35, p=0.01). 

CONCLUSION

Single-sensor-derived LDE distinguished PwMS from HC in a clinical setting. Novel sensor placement at the sternum was sensitive to subclinical gait instability (EDSS 0-1), demonstrating potential capacity as a digital biomarker for early detection of disease, including future potential to monitor disease via smart phone. 

FUNDING

Multiple Sclerosis Research Australia