Monday, 20 March 2017

Predicting MS Activity

HÃ¥kansson I, Tisell A, Cassel P, Blennow K, Zetterberg H, Lundberg P, Dahle C, Vrethem M, Ernerudh J.Eur J Neurol. 2017 . doi: 10.1111/ene.13274. [Epub ahead of print]


Improved biomarkers are needed to facilitate clinical decision-making and as surrogate endpoints in clinical trials in multiple sclerosis (MS). We assessed whether neurodegenerative and neuroinflammatory markers in cerebrospinal fluid (CSF) at initial sampling could predict disease activity during 2 years of follow-up in patients with clinically isolated syndrome (CIS) and relapsing-remitting MS.


Using multiplex bead array and enzyme-linked immunosorbent assay, CXCL1, CXCL8, CXCL10, CXCL13, CCL20, CCL22, neurofilament light chain (NFL), neurofilament heavy chain, glial fibrillary acidic protein, chitinase-3-like-1, matrix metalloproteinase-9 and osteopontin were analysed in CSF from 41 patients with CIS or relapsing-remitting MS and 22 healthy controls. Disease activity (relapses, magnetic resonance imaging activity or disability worsening) in patients was recorded during 2 years of follow-up in this prospective longitudinal cohort study.


In a logistic regression analysis model, NFL in CSF at baseline emerged as the best predictive marker, correctly classifying 93% of patients who showed evidence of disease activity during 2 years of follow-up and 67% of patients who did not, with an overall proportion of 85% (33 of 39 patients) correctly classified. Combining NFL with either neurofilament heavy chain or osteopontin resulted in 87% overall correctly classified patients, whereas combining NFL with a chemokine did not improve results.


This study demonstrates the potential prognostic value of NFL in baseline CSF in CIS and relapsing-remitting MS and supports its use as a predictive biomarker of disease activity.

Neurofilaments in spinal fluid is a reasonable predictor of disease activity. We have said this many times


  1. Indeed. But is it a predictor of brain atrophy?

    1. Yes, data presented at ECTRIMS and submitted for publication by Jens Kuhle in Basel shows that it predicts brain atrophy.

    2. Then would you use LPs in your clinical monitoring of confirmed MSers?

    3. I can't find any references to BVL here:

    4. Anon: look for ARBA rates in the results.

      They are there but with a relatively weaker correlation of .36


Please note that all comments are moderated and any personal or marketing-related submissions will not be shown.