Background

There is a clear need for biomarker development in progressive multiple sclerosis (PMS). Serum neurofilament light (sNFL), and to a lesser extent neurofilament heavy (sNFH), are leading candidates. Whilst sNFL in particular has been examined in relapsing remitting MS, data from PMS are limited and in contrast to immunosuppressive treatments, candidate neuroprotective treatments show a variable effect on these biomarkers [1]. We examine data from the original MS-STAT trial [2].

Objectives

To analyses serum neurofilament data from the MS-STAT trial, assessing cross-sectional and longitudinal sNFL and sNFH and their relationship with MRI and clinical variables.

Methods

Serum samples were acquired at months 0, 6, 12 and 24. sNFL and sNFH were quantified using Single Molecule Array (Simoa). Linear mixed models were used to assess for treatment effects between simvastatin and placebo, and regression, linear mixed and bivariate models to assess association with MRI and clinical variables.

Results

Median baseline sNFL was 14.6pg/ml (IQR 10.8-20.2) and sNFH 64.2pg/ml (23.9-119), rising to 16.0pg/ml (11.9-22.2) and 70.9pg/ml (26.5-123) by 24 months. Higher baseline sNFL was associated with greater subsequent brain atrophy. Similarly, higher sNFL was associated with higher T2 lesion volume (T2LV), and increases in sNFL were associated with increases in T2LV.

High baseline sNFL was associated with worse baseline EDSS (95% CI of coefficient: 0.120 to 0.633), 9-hole peg test (-0.009 to -0.002) and 25 foot walk (-0.515 to -0.019), and a greater deterioration in 25 foot walk from baseline to 2 years (-0.448 to -0.152). There was no evidence of an association between sNFH and these variables.

There was no evidence of a simvastatin treatment effect on either sNFL (95% CI -0.07 to 0.10, p=0.716) or sNFH (95% CI -0.17 to 0.22, p=0.827).

Conclusions

The mode of action of simvastatin in reducing brain atrophy remains to be elucidated, with actions upon vascular comorbidity and intermediate metabolites from cholesterol synthesis pathways as possible candidates [3]. Simvastatin is not immunosuppressive, and in common with other purportedly neuroprotective treatments, there is no evidence from this study that sNFL or sNFH can act as biomarkers of simvastatin treatment. The relationship between sNFL and T2LV supports the hypothesis that in MS, sNFL may act predominantly as a marker of neuroinflammation, even in this typical SPMS cohort.

1. Williams T et al (2020) Neurofilaments in progressive multiple sclerosis: a systematic review. J Neurol.

2. Chataway J et al (2014) Effect of high-dose simvastatin on brain atrophy and disability in secondary progressive multiple sclerosis (MS-STAT): a randomised, placebo-controlled, phase 2 trial. Lancet

3. Palladino R et al (2020) Evaluating the Risk of Macrovascular Events and Mortality among People with Multiple Sclerosis in England. JAMA Neurol

Background

Cognitive dysfunction is common in multiple sclerosis (MS) and is particularly prevalent in patients with progressive forms of the disease [1]. Exploring neurofilament associations has clear value in trying to develop treatments for cognition.

Objectives

To perform a post-hoc analysis of the MS-STAT trial [2], assessing serum neurofilament light (sNFL) and heavy (sNFH) as predictive biomarkers of future cognitive performance.

Methods

Serum samples were analysed for sNFL and sNFH using Single Molecule Array (Simoa). Cognition was assessed at months 0, 12 and 24 with a detailed neuropsychometric battery including the Wechsler Abbreviated Scale of Intelligence (WASI) and Brain Injury Rehabilitation Trust Memory and Information Processing Battery (BMIPB), as previously described [3]. Multivariate regression models were used for cross-sectional analysis, and linear mixed models were used to assess the predictive value of dichotomised baseline sNFL and sNFH on changes in cognition. All analyses controlled for the established MRI variables of whole brain volume (WBV) and T2 lesion volume (T2LV) in order to assess the extent to which sNFL and sNFH provided additional prognostic value.

Results

Cross-sectional analyses:

After adjusting for demographics, T2LV and WBV, neither sNFL nor sNFH demonstrated cross-sectional associations with current cognitive performance.

Longitudinal analyses:

Patients with high baseline sNFL experienced significantly greater cognitive decline from 0 to 24 months in WASI full-scale IQ (95% CI of coefficient -0.238 to -0.024), WASI Verbal IQ (-0.281 to -0.011), and in both immediate and delayed BMIPB figure recall (-0.489 to -0.046 and -0.399 to -0.025, respectively). sNFH was not associated with changes in cognitive performance.

Conclusions

Our results demonstrate the prognostic value of sNFL on future changes in cognitive performance in SPMS, assessed through a detailed neuropsychometric battery. sNFL remained significantly associated with future cognitive decline whilst controlling for established MRI biomarkers, suggesting that it may provide additional utility in identifying those who may benefit most from interventions aimed at preventing cognitive decline.

1. Sumowski JF et al (2018) Cognition in multiple sclerosis: State of the field and priorities for the future. Neurology

2. Chataway J et al (2014) Effect of high-dose simvastatin on brain atrophy and disability in secondary progressive multiple sclerosis (MS-STAT): a randomised, placebo-controlled, phase 2 trial. Lancet

3. Chan D et al (2017) Effect of high-dose simvastatin on cognitive, neuropsychiatric, and health-related quality-of-life measures in secondary progressive multiple sclerosis: secondary analyses from the MS-STAT randomised, placebo-controlled trial. Lancet Neurol