Neurofilament Light Chain Leads Candidates For Blood-Based Biomarker Of MS Disease Activity


It’s taken almost 20 years, but the search for a widely usable, blood-based biomarker of multiple sclerosis (MS) disease activity is finally yielding results. New technology now allows ultralow levels of neurologically-derived substances to be detected in the plasma or serum. Previously, these substances were detectable only in the cerebrospinal fluid (CSF).

Neurofilament light chain (NfL) is the candidate biomarker currently causing the most stir among the MS research community as it’s the biomarker most likely to make it from research labs into routine clinical practice in the near future. It’s a protein associated exclusively with neurons such that, when detected in the CSF or blood, it can give an indication of neurologic damage. Indeed, baseline levels of NfL in both CSF and peripheral blood have been shown to be higher in individuals with MS than in healthy controls, to increase over time with increasing neurologic damage, and to decrease in response to disease-modifying therapies (DMTs) (Neuropsychiatr Dis Treat. 2018;14:2241-54).

“I’ve been working on neurofilament levels now for decades, so to get to the point where we are now, shows you how long it takes for a biomarker to get into clinical practice and for an innovation to be adopted,” Gavin Giovannoni, MBBCh, PhD, said in an interview.

Dr. Giovannoni, who is professor and chair of neurology at the Blizard Institute of Cell and Molecular Science, Barts and The London School of Medicine and Dentistry, Queen Mary University of London, explained that neurofilament proteins are the major structural proteins of axons. They are composed of light-, medium-, and heavy-chain subunits that, until recently, were reliably detectable only in the CSF, which limited the potential use of neurofilaments as a routine biomarker of MS disease activity (Brain. 2018;141[8]:2235-7).

“When you damage the nervous system, you release these proteins, and the level that is released is proportional to the amount of damage,” he said.

Dr. Gavin Giovannoni
The breakthrough for using NfL as a possible blood-based MS biomarker has come thanks to the availability of “a new, very sensitive assay that allows you to detect the neurofilament levels in the peripheral blood,” Dr. Giovannoni said. This assay, called SIMOA, which stands for single molecule array, is basically an ultrasensitive enzyme-linked immunosorbent assay (ELISA). In addition to detecting the presence of NfL, it can also detect tau and other neurologic proteins associated with brain pathology.

“We are in the process of validating the assay, ticking all the boxes to make sure that it is reliable and reproducible enough to be used in clinical practice,” Dr. Giovannoni said. “At the moment, the peripheral blood levels are really only a research tool and are not used in routine practice, as yet,” Dr. Giovannoni stressed.

Measuring NfL in CSF is possible although still not universally employed in clinical practice. “The assay that we use in CSF has been validated, and it’s got a quality (CE) mark from the European Union; so we are increasingly doing lumbar punctures on people to measure NfL in the spinal fluid.”

Staining of human cerebellum with an antibody to the phosphorylated form of the heavy or high molecular weight neurofilament subunit NF-H. This protein is found in axons of basket cells which are seen surrounding the large Purkinje cells.

Credit: Antibody and image generated by EnCor Biotechnology Inc. GerryShaw/Wikimedia Commons/CC BY-SA 3.0

NfL levels in the peripheral blood are lower but correlate with levels in the CSF, observed Charlotte Teunissen, PhD, who is a professor of neurochemistry at VU University Medical Center, Amsterdam. Being able to measure levels in blood rather than CSF, “offers a lot of opportunities,” she said in an interview. “With blood, you can do repeated collection and repeated analysis, so you can really use neurofilament light for monitoring.”

“MS treatment is becoming more complicated every day as new treatments arrive on the market, so it’s difficult to decide which treatment to use in an individual patient,” Dr. Teunissen explained. Currently, according to the U.S. National MS Society website there are 15 DMTs approved by the Food and Drug Administration. “As a doctor, you want to know, at the earliest possible stage, whether a treatment is effective – or if you need to stop one and start another type of treatment. For that, neurofilament light could be a very good biomarker.”

Dr. Charlotte Teunissen

Neurofilaments are released with axonal injury. Early assays were only able to detect neurofilaments at the higher levels found in cerebrospinal fluid. New assays are now able to detect neurofilaments at the lower levels found in blood samples.

NfL is not specific to MS, therefore its role is limited to monitoring MS disease activity rather than to making an MS diagnosis. “Neurofilament light is elevated in almost any neurological disease that you can imagine. For diagnosis, one usually wants to use something that is really specific for that disease,” Dr. Teunissen explained. “There are a few diseases such as ALS [amyotrophic lateral sclerosis], frontotemporal dementia, and HIV dementia where levels can be extremely elevated and for those diseases, you can use neurofilament light levels to support a diagnosis.”

“People like to think that it may be a diagnostic marker, but at the end of the day, it is not specific enough” for MS, Dr. Giovannoni observed. To sort out MS from other conditions, “you will need other tools.” Blood testing for NfL, he said, “could complement a diagnosis, but in itself is not a diagnostic marker.”

“I think all of the studies now show that NfL is a very predictive biomarker … baseline levels predict outcome in the future, so people with raised levels do worse. (NfL levels are) also very predictive of disease activity. In other words, if the levels are raised it means the MS is active. It can be used to assess whether the treatments are failing and also can be used in clinical trials to assess whether treatments are working.”

“I think all of the studies now show that NfL is a very predictive biomarker … baseline levels predict outcome in the future, so people with raised levels do worse.”

Blood testing for NfL is unlikely to eliminate the need for MRI scans, but might be able to reduce their frequency. MRIs and NfL measures each provide different information and they are essentially complementary to each other. Blood testing for NfL “is easier to do, and it will almost certainly be cheaper and easier to interpret” than an MRI, Dr. Giovannoni said. Just as muscle diseases are monitored by measuring creatinine kinase levels and adjusting treatments accordingly, perhaps MS could be monitored with NfL levels, he said. “I see us moving into that space, we’ll be doing blood levels and if the levels aren’t coming down, we will switch treatments. The therapeutic aim will be to normalize the levels and keep them in the normal range.”

NfL may become a treatment target in itself, Dr. Giovannoni further suggested. “At the moment, we try to treat people to the target of NEDA (no evidence of disease activity) … we want them to be relapse free and MRI activity free. But going forward, we’re going to go beyond NEDA-3. We want to target neurofilament levels; we want to normalize neurofilament levels.”

Color enhanced transmission electron micrograph of neurofilaments in an astrocyte.
Image credit: Dennis Kunkel Microscopy/Science Source

Dr. Teunissen said the future roles of MRI and blood NfL measurement in MS management will need to be defined. “I can imagine that, when the NfL result is negative on occasions where there are no clinical signs of change and you would normally ask for an MRI, that you could say ‘OK, in this case, I think we don’t need an MRI.’ ”

Both Dr. Teunissen and Dr. Giovannoni are optimistic that testing for NfL in the blood could become routine within the next 1-2 years; the assay and equipment are available, and it’s now down to validating the method and getting the right cutoff levels and standardization across labs.

“I can imagine that, when the NfL result is negative on occasions where there are no clinical signs of change and you would normally ask for an MRI, that you could say ‘OK, in this case, I think we don’t need an MRI.’ ”

We are also developing other blood tests, but that’s at a very-early stage,” Dr. Teunissen said. “One of those is for glial fibrillary acidic protein (GFAP). So that’s another blood biomarker we’ll probably hear more of in the future.” A recent article (Mult Scler. 2018 Dec 20. doi: 10.1177/1352458518819380) described how serum GFAP also correlates with MS disease severity, with associations seen between higher levels of this protein and MS disability, longer disease duration, a progressive disease course, and MRI pathology.

Most of the other potential biomarkers have currently been detected only in the CSF, which is of course closer to the disease process and its value for monitoring MS disease severity and possibly for diagnosing MS shouldn’t be forgotten, Dr. Teunissen said. “The CSF is closer to the brain tissues and therefore closer to the brain pathology. That’s why I think we still need CSF for the discovery of and (validation) of novel biomarkers.”

Some of the interesting CSF-based biomarkers under investigation include chitinase and chitinase-3-like proteins. These are “a family of glycoproteins that bind and hydrolyze chitin,” which may be “promising for discriminating MS, optic neuritis, and clinically isolated syndrome (CIS),” based on a recent review paper (Degener Neurol Neuromuscul Dis. 2017;7:19-29).

Circulating microRNAs also are under investigation. These often are packaged in exosomes and can be detected in the CSF or peripheral blood using various methods such as quantitative polymerase chain reaction or microRNA array analysis. While promising, the latter produces just a number, Dr. Teunissen said. “It’s also interesting and it is about to be an important technology, but with microRNA, it also is a number. It doesn’t give you a feeling of biological mechanism, and [the measure] can also relate to different biological mechanisms so that makes it hard to interpret the meaning of an elevation or a decrease.”

Then there are the immunoglobulins, IgG and IgM, which can be seen as oligoclonal bands in the CSF, and two brain-specific soluble cell adhesion proteins, contactin-1 and contactin-2, that are expressed on the axonal membranes of neurons. These could be early markers of axonal dysfunction that could be detected in the CSF (Mult Scler J Exp Transl Clin. 2018;4[4]:1-10).

Given all of this, what does the future hold? Will there be a single biomarker that is tested, such as NfL or GFAP in the blood? Perhaps a panel of biomarkers in the blood or CSF? “I think that is the future. Not the immediate future, but it will go in that direction,” Dr. Teunissen said.

Dr. Teunissen and Dr. Giovannoni had no relevant financial disclosures in relation to their comments reported in this article.


Cai L, et al. Neurofilament light chain as a biological marker for multiple sclerosis: a meta-analysis study. Neuropsychiatr Dis Treat. 2018; 14: 2241-54.

Giovannoni G. Peripheral Blood Neurofilament Light Chain Levels: The Neurologist’s C-reactive Protein? Brain. 2018;141(8):2235-37.

Högel H et al. Serum glial fibrillary acidic protein correlates with multiple sclerosis disease severity. Mult Scler. 2018 Dec 20 [Epub ahead of print]

Harris VK, et al. Biomarkers of multiple sclerosis: current findings (2017 review). Degener Neurol Neuromuscul Dis. 2017;7:19-29.

Chatterjee M, et al. Contactin-1 and contactin-2 in cerebrospinal fluid as potential biomarkers for axonal domain dysfunction in multiple sclerosis. Mult Scler J Exp Transl Clin. 2018;4(4):1-10.