Neurofilaments: from a two towards a three-layered cake!


We are happy to announce that our new paper on how to use cerebrospinal fluid neurofilament light chain (CSF NfL) in daily clinical practice has appeared online (open access) in the journal of Neurology Neuroimmunology and Neuroinflammation

In this paper, we have documented how the BartsMS team is already using for more than four years NfL values obtained with a lumbar puncture to advise pwMS on which disease-modifying treatment strategy would be best for them. This means that we only accept that pwMS are free of disease activity when the following three domains are evaluated: 

1. Clinical – Presence of relapses 

2. Radiological – New or enhancing lesions on brain or spinal cord MRI

3. Cerebrospinal fluid – Neurofilament light chains above age-related references values

This three-layered approach is novel, as the majority of current MS neurologists base their treatment strategies solely on the first/clinical and second/radiological domains. However, we demonstrate that MS disease activity is most frequently represented by elevated CSF NfL (85 out of 203 pwMS) compared to clinical (81 out of 203 pwMS) and MRI activity (65 out of 203 pwMS).

In addition, it was also apparent that 22 out of 203 pwMS (approximately 10 %) had increased CSF NfL as the only sign of disease activity, and this was especially true for people with the progressive disease stage. 

The implications of these findings are tantalising, as it adds a whole new layer on top of the traditional disease activity work-up. It also means that those with progressive disease are no longer side-lined as inactive in so far as their disease is concerned. The technique we have used to measure the NfL values is CE-certified and available for commercial use in the European Union. This means that European MS neurologist can pick up the phone as we speak and discuss with their clinical laboratories how to implement CSF NfL measurements in their clinical practice. 

Our findings also question the current classifications of MS, moving towards a more biological signature of disease activity than a clinical one. The current MS classification system that Lublin et al. introduced in 2013 still makes the distinction between relapsing and progressive pwMS. However, as our data show that progressive pwMS can still manifest with very high CSF NfL levels reflecting ongoing and undoubtfully harmful inflammation this classification system feels outdated. Shouldn’t we be focusing on active versus inactive disease instead? . Labelling pwMS as active versus inactive based on our three-layered disease-activity cake would therefore reflect much better the treatment dilemmas neurologists face today. What do you think?


CSF neurofilament light chain testing as an aid to determine treatment strategies in MS

Saúl Reyes*,  Ide Smets*, David Holden, Karina Carrillo-Loza, Tatiana Christmas, Lucia Bianchi, Francesca Ammoscato, Benjamin Turner,  Monica Marta,  Klaus Schmierer, Gavin Giovannoni, Sharmilee Gnanapavan

Objective To evaluate the use of CSF neurofilament light chain (NfL) measurements in clinical practice as well as their effect on treatment strategies and outcomes in patients with MS.

Methods This was an observational cohort study of patients with MS who had a CSF NfL measurement between December 2015 and July 2018 as part of their routine clinical care. Treatment strategies were classified as “No Treatment/No Escalation” (no treatment or no escalation of treatment) or “Treatment/Escalation” (first-line injectable/oral disease-modifying therapies (DMTs), highly active DMTs, or treatment escalation). Change in Expanded Disability Status Scale (EDSS) scores was evaluated after 1-year follow-up.

Results Of 203 patients with MS, 117 (58%) had relapsing-remitting MS. Disease activity was most frequently indicated by elevated CSF NfL (n = 85), followed by clinical (n = 81) and MRI activity (n = 65). CSF NfL measurements were independently associated with clinical (p = 0.02) and MRI activity (p < 0.001). Of those with elevated CSF NfL as the only evidence of disease activity (n = 22), 77% had progressive MS (PMS). In patients with PMS, 17 (20%) had elevated CSF NfL as the sole indicator of disease activity. Elevated CSF NfL resulted more frequently in Treatment/Escalation than normal CSF NfL (p < 0.001). Median EDSS change at follow-up was similar between patients receiving No Treatment/No Escalation and Treatment/Escalation decisions (p = 0.81).

Conclusions CSF NfL measurements informed treatment strategies, alongside clinical and MRI measures. CSF NfL levels were the only indicator of disease activity in a subset of patients, which was more pronounced in patients with PMS. Elevated CSF NfL was associated with more Treatment/Escalation strategies, which had an impact on EDSS outcomes at 1 year.

CoI: multiple

About the author

Ide Smets


  • This is why I read your excellent blog! Clearly explained with food porn your revolutionary strategies. Hope the laggards actually read and not take a let them eat cake attitude to starving our so called progressive MS comrades,

  • There is little doubt that NFL monitoring is also changing the way we think about active MS and in particular active advanced or progressive MS. This is one of the reasons why we are trying to modify the course of advanced MS.

    • But what worries me is inactive MS (no clinical or radiological signs) where the patient is still progressing / accruing disability. Does nfl monitoring pick this up and if so what are the treatment options?

      • Well, read the post above – 20% were designated “active ” based on CSF NfL alone. However, from pathology data we know that ~4/5 pwMS have chronic/active lesions, so the assumption should rather be that MS is active unless proven otherwise. However, it will take time to change perceptions, and I hope ChariotMS will help since disease activity “as measured by MRI or neurofilament status” will not be among eligibility criteria. Reference values are also an issue – we need more granularity to increase sensitivity for elevated levels.

        • Yup, it’s going to need some further work with predictive modelling and discrimination is this. Hopefully, a nice new research and clinical outcome measure in the end.

    • Great research and clinical application. In Sweden Uppsala they also routinely check CSF NfL levels. Less invasive, however, is to check NFl levels in blood. Do you think blood testing is as accurate as CSF testing or that in time blood testing can replace CSF testing? 

      • If depends who you read….according tothe NFL mafia it is ready to but our data did not replicate their perfect fit

      • It will depend on how you use the NfL, but in our paper we use single-point measurements to draw conclusions upon the treatment strategy of individual pwMS. Personally, I don’t think the levels of blood NfL are sufficiently different from healthy controls to use blood NfL in the same way as we did here. However, blood NfL has very interesting applications when it comes to long-term disease monitoring as they can be measured serially.

  • But then again, none of those 3 criterias correlate with BVL! How do I know? I am a case-in-point.

    When will that be added to your decision tree? What happen to that Dutch volume measurement software Dr. K was involved with years ago?

    • Thanks for kick up the backside Tony, we’re working on that as well. It’s a Belgian company, and in spite of the eternity it’s taken to this point, we’re still hopeful we’ll be able to start using semi-automated MRI “reading” in the near future…

      • I had lunch with a young radiologist with a muscular and bone sub-speciality yesterday who says that tumour size change is now being monitored by a software. I did not ask for details as I was worried that would disclose the fact that I know slightly more than the general population (keep my MS for myself for now).
        Now if a software can measure a tumour today reliably, why not the whole brain I thought?
        He went on to say that NHS Wales bought him two 8Ks monitors for £15,000 each and he is now working from home and that no-non serious MRI assessment will be done in 2020 because of backlog and lack of resources. So long for MS monitoring I thought, but then again, I did not ask specifically about that.

        • I wish such complications were not needed
          (worried about what a statement or question might reveal!)
          But it can’t be helped I suppose, such is life

  • Just received a letter from my neurologist. After an mdt meeting the team decided my symptoms are due to one or two of the following conditions- radiation myelopathy, past b12 deficiency or MS. Could this test help to diagnose me more accurately. The only treatment I am on is the b12 injections and baclofen. Would ideally like a dmd if it is MS.

  • Good work, thank you

    Isn’t it a problem that collecting the CSF needs a lumbar puncture?

    I have never had a NFL level test or a even lumbar puncture for oligoclonal bands

    The doctors did not feel it’s necessary, and I have never asked for one or even asked about one (because of some nightmare stories about days of severe headaches )

By Ide Smets



Recent Posts

Recent Comments