Subscribe to eTOC

New Data to Support Neurofilament Levels as A Biomarker in Multiple Sclerosis


Investigators reported increased CSF levels of neurofilament heavy chain protein in subtypes of multiple sclerosis (MS) compared with healthy controls, providing supports for its potential use as a biomarker for disease progression.

In the last few years, investigators have examined CSF levels of neurofilament (Nf) as a potential biomarker of neuronal damage in Alzheimer disease, amyotrophic lateral sclerosis, and neuromyelitis optica. Neurofilaments, which are composed of light, medium, and heavy chain units, are the structural proteins involved in neuronal homeostasis and axonal transport.

Now, a team of Swiss investigators led by Jens Kuhle, MD, a neuroimmunologist in the departments of biomedicine and neurology the University Hospital Basel in Switzerland, report increased CSF levels of neurofilament heavy chain protein (NfH) in subtypes of multiple sclerosis (MS) compared with healthy controls. In addition, they found NfH levels are higher in patients experiencing clinical relapse compared with subjects with stable disease. The findings were published Feb. 23 in the online edition of Neurology.


The investigators measured CSF levels of NfH in 63 patients with clinically isolated syndrome (CIS) and 87 with definite MS, compared with 73 healthy controls. They also measured NfH levels among patients with various MS subtypes: relapsing-remitting MS (RRMS), secondary progressive MS (SPMS), and primary progressive MS (PPMS). Samples were obtained between 1998 and 2007.

The NfH levels significantly correlated with disability as measured by Expanded Disability Status Scale (EDSS) scores in patients with CIS and RRMS (r=0.33, p=0.001) and in patients during clinical relapse (r=.35, p=0.01). The correlation was most prominent in patients with RRMS during relapse (r=0.54, p=0.01).

To take into account that NfH levels, like most biomarkers, are age-dependent, the investigators stratified patients by age. The CSF levels of NfH were increased with age in controls (r=0.50, p<0.0001) and CIS (r=0.50, p<0.0001), but were less evident in RRMS (r=0.35, p=0.027) and absent in patients with SPMS and PPMS.

“Our electrochemilunimescence (ECL)-based assay is 60- to 80-fold more sensitive than the conventional methods,” Dr. Kuhle said. Along with higher sensitivity to conventional assays, the ECL-based assay provides a broader dynamic range and requires less sample volume, all of which make it possible to measure very low levels and therefore to analyze small differences, he added.

“We think the age correlation has only been borne out by being able to measure very low CSF NfH concentrations,” he said.

“It seems most likely that the predominant cause of the increase of NfH with age is ongoing neurodegeneration that accelerates with age,” Dr. Kuhle said.

This finding suggests that age-related NfH increases will need to be considered to distinguish these changes from disease-related NfH increases, he added.


Commenting on the study, Shiv Saidha, MD, a clinical neuroimmunologist at the Johns Hopkins University School of Medicine, pointed out that whereas the increases in NfH in patients with CIS and RRMS appeared to correlate with age, this was not the case in patients with SPMS or PPMS. “This suggests that acceleration in neurodegeneration in this group of patients is not exclusively age-driven,” he said.

Dr. Saidha, who was not involved with the study, said the paper offers an important contribution to the understanding of NfH levels and degree of neurodegeneration by MS subtype and disease activity. But, he added, more work is needed for the biomarker to be a useful clinical tool.

A more practical and immediate use, he said, would be to use the NfH levels as “a putative outcomes measure in clinical trials of new MS drugs and other neuroprotective agents for other neurological conditions.” He cited a 2011 study in the Annals of Neurology that showed the efficacy of natalizumab therapy in normalizing light chain Nf levels in relapsing MS patients that mirrored reductions in EDSS scores, relapses, and CSF inflammatory parameters in patients with stable or inactive disease.

“Similar studies of NfH are warranted and will help validate NfH as a truly useful biomarker,” he said.

In an editorial accompanying the study, Gavin Giovannoni, MBBCh, PhD, professor of neurology at the Centre for Neuroscience and Trauma at the Barts and the London School of Medicine and Dentistry, and Avendra Nath, MD, professor of neurology at Johns Hopkins University, expressed a similar sentiment. “Elevated CSF Nf is a simple indicator of axonal damage, and is predictive of severity and poor recovery from acute attacks and the development of long-term disability in patients with MS,” they wrote. “We would encourage the MS community to take these observations on board and to insist on the inclusion of this valuable biomaker in all future clinical trials.”

Richard M. Ransohoff, MD, director of the Neuroinflammation Research Center at the Lerner Research Institute and staff neurologist at the Mellen Center for MS Treatment and Research of the Cleveland Clinic, was less sanguine about the study findings, however. He said that although it is conceptually important to find a biomarker of axonal injury in early MS, he doesn't think the current study provides meaningful clinical correlations on which to draw conclusions about the potential of this biomarker in MS.

In particular, he said that the use of EDSS to measure disability is imprecise and a “terrible overall measure of MS.”

“If you have the same patient examined on the same day by two different neurologists, one-half hour apart, they will come up with a different EDSS by easily one-half point,” Dr. Ransohoff said. “So the clinical correlations in this paper are not meaningful enough for general application.”

Despite these reservations, Dr. Ransohoff emphasized the need for further research to find a meaningful predictor of disability that may complement current reliance on imaging.

“One of the things we agonize about is that we don't have very powerful predictors of prognosis early in the disease and therefore we struggle to determine which patients should be treated aggressively and monitored closely and which can be treated less aggressively and monitored less closely based on real data,” he said. The main struggle is finding a way to measure the neuroprotective effect of a drug should there be a drug that could exert this effect in MS.


Kuhle J, Leppert D, Lindberg RIP, et al. Neurofilament heavy chain in CSF correlations with relapses and disability in multiple sclerosis. Neurology 2011; E-pub 2011 Feb. 23.
    Giovannoni G, Nath A. Editorial: After the storm: Neurofilament levels as a surrogate endpoint of neuroaxonal damage. Neurology 2011;E-pub 2011 Feb. 23.
      Gunnarsson M, Malmestrom C, Axelsson M, et al. Axonal damage in relapsing multiple sclerosis is markedly reduced by natalizumab. Ann Neurol 2011;69:83–89.