A paper in the December 3 JAMA Neurology adds yet more evidence that neurofilament light (NfL) protein is a general biomarker of neurodegeneration. Scientists led by Bob Olsson and Erik Portelius, University of Gothenburg, Sweden, found above-control levels of NfL in the cerebrospinal fluid of people with a wide variety of neurodegenerative disorders—some diagnosed by autopsy. The more severe a person’s disease, the higher his or her CSF NfL, supporting the idea that the protein could track disease progress. “It’s a nice replication study,” said Douglas Galasko, University of California, San Diego. “NfL could be used to extend the diagnostic panel of CSF biomarkers.”

  • NfL is elevated across the board in neurodegenerative diseases.
  • Levels are highest when disease is most severe.
  • NfL could help track disease progression over time.

NfL is part of the axonal cytoskeleton. When neurons die, they release it into the cerebrospinal fluid. Olsson and colleagues had previously reported higher CSF NfL among 3,356 people in Sweden who had various kinds of dementia, including frontotemporal (FTD), vascular, early and late-onset Alzheimer’s disease, AD with mixed pathology, dementia with Lewy bodies (DLB), and Parkinson’s disease with dementia (PDD), all diagnosed clinically (Skillbäck et al., 2014). In the current study, Nicholas Cullen at the University of Pennsylvania, Philadelphia, and others examined 838 patients who had been treated at UPenn’s six different neurodegenerative disease centers beginning in 1996; in 120 of them, diagnosis was made at autopsy.

Baseline CSF was sampled from 114 people with mild cognitive impairment (MCI), 397 with AD, 96 with FTD, 68 with amyotrophic lateral sclerosis (ALS), 41 with PD, 19 with PD/MCI, 29 with PDD, 33 with DLB, 21 with corticobasal syndrome (CBS), 20 with progressive supranuclear palsy (PSP), and 75 who were aging normally. All had undergone follow-up cognitive testing for one to 18 years.

“To our knowledge, this is the first time all these diseases are compared with each other in a single cohort in the same study,” wrote the authors.

“This is an impressive cohort,” said Anne Fagan, Washington University in St. Louis. “These are relatively large numbers for the different subcategories of neurodegenerative disorders.”

In all patients, baseline levels exceeded those of controls, who averaged 536 pg/mL. ALS and FTD patients had the most—4,185 and 2,094 pg/mL, respectively. In parkinsonian disorders, concentrations ranged from 619 pg/mL in PD with normal cognition to 981 in PDD, 1,281 pg/mL in CBS, and 1,578 in PSP. With an average of 951 pg/mL, AD patients had slightly more than the MCI average of 831, though there was considerable overlap between those groups.

NfL correlated negatively with baseline score on the mini-mental state examination (MMSE), meaning the more NfL a person had in his or her CSF, the worse the score on this cognition screen. Higher baseline NfL also meant that the MMSE score declined faster on subsequent cognitive tests. Among 85 PD patients whose cognition was staged not on the MMSE but either the Montreal Cognitive Assessment (MoCA) or Dementia Rating Scale (DRS), the correlation with baseline NfL held up, as well.

Adjusted for age and sex, NfL correlated with other neuronal markers of disease, including total CSF tau in controls and people with MCI, AD, bvFTD, ALS, DLB, PD/MCI, and the semantic form of PPA. Not so for PD, PDD, CBS, logopenic PPA, non-fluent agrammatic PPA, or PSP. While tau aggregates are found in CBS and PSP, they predominantly occur in astrocytes, which may explain no correlation between CSF tau and neurofilament light in these two disorders (Nov 2018 conference news). In 60 autopsied brains in which TDP-43 immunostaining was undertaken, CSF NfL concentrations correlated with TDP-43 load in the majority of regions sampled.

The study demonstrates that NfL is a reliable biomarker for neurodegeneration, said co-author Henrik Zetterberg, also from the University of Gothenburg. Zetterberg and colleagues found that NfL in plasma mirrors levels in CSF, therefore scientists can likely get the same result from a simple blood test, he told Alzforum (Mar 2017 news). That would simplify enrollment, particularly for longitudinal studies that track changes in NfL levels over time.

NfL could improve clinical trials, hopefully allowing researchers to monitor whether a drug slows neurodegeneration, Zetterberg said. In multiple sclerosis, CSF NfL falls in patients treated with the therapeutic antibody natalizumab (Gunnarsson et al., 2011). Zetterberg also suggested retroactively measuring NfL in samples collected in previous clinical trials to see if they show any signal of reduced neurodegeneration.

Fagan found it interesting that NfL and tau did not correlate in some clinical groups, saying this deserves follow up. She thinks that NfL will help characterize patients in clinical trials, especially for predicting when a person’s cognition will decline. Galasko believes NfL will likely help distinguish atypical parkinsonian disorders from idiopathic PD. Both these commentators, and others, wondered if NfL could be the “N” in the proposed A/T/N classification for Alzheimer’s diagnosis, which stands for amyloid, tau, and neurodegeneration, respectively (Nov 2018 news). 

Not all researchers consider NfL quite ready for prime time. “In general, we should be cautious in interpreting NfL as a marker of neurodegeneration, as there is no orthogonal marker of neurodegeneration for corroboration,” wrote William Hu, Emory University, Atlanta. “Future autopsy work examining whether CSF NfL levels correlate with the degree of neuronal loss within each disorder will greatly improve our understanding of what these changes truly represent,” Hu wrote. Zetterberg noted that longitudinal magnetic resonance imaging studies demonstrate a robust association between egional brain atrophy, a marker of neurodegeneration, and CSF NfL (Zetterberg et al., 2016).—Gwyneth Dickey Zakaib


No Available Comments

Make a Comment

To make a comment you must login or register.


News Citations

  1. International Symposium Puts PSP/CBD on the Map
  2. Blood Neurofilament Light a Promising Biomarker for Alzheimer’s?
  3. Is NfL the New 'N' in A/T/N Classification of Alzheimer’s?

Paper Citations

  1. . CSF neurofilament light differs in neurodegenerative diseases and predicts severity and survival. Neurology. 2014 Nov 18;83(21):1945-53. Epub 2014 Oct 22 PubMed.
  2. . Axonal damage in relapsing multiple sclerosis is markedly reduced by natalizumab. Ann Neurol. 2011 Jan;69(1):83-9. PubMed.
  3. . Association of Cerebrospinal Fluid Neurofilament Light Concentration With Alzheimer Disease Progression. JAMA Neurol. 2016 Jan 1;73(1):60-7. PubMed.

Further Reading


  1. . Neurofilaments as biomarkers in neurological disorders. Nat Rev Neurol. 2018 Oct;14(10):577-589. PubMed.

Primary Papers

  1. . Association of Cerebrospinal Fluid Neurofilament Light Protein Levels With Cognition in Patients With Dementia, Motor Neuron Disease, and Movement Disorders. JAMA Neurol. 2018 Dec 3; PubMed.