17 Nov 2017

It works! Neurofilament light chain (NfL) has emerged as the most promising fluid marker for tracking the progression of neurodegenerative disease. Unlike many other biomarkers, blood levels correlate closely with those in cerebrospinal fluid, raising the possibility of monitoring brain disease with a simple blood draw. Two recent Neurology papers broadened the potential applications for the marker, suggesting that serum NfL may detect both degeneration at early stages and response to treatment. In the October 25 Neurology, researchers led by Nick Fox at University College London reported that serum NfL starts to climb before symptoms emerge in people who carry a familial Alzheimer’s mutation. Meanwhile, researchers led by Lenka Nováková at the University of Gothenburg, Sweden, found that serum NfL levels dropped in response to disease remission or treatment in people with multiple sclerosis. That work appeared in the October 27 Neurology. “This could be a very accessible, non-disease-specific neurodegenerative marker,” Fox told Alzforum.

Other researchers agreed that the case for clinical use of this biomarker is growing. “I strongly believe that blood NfL concentration would be a valuable tool for clinical neurologists in a variety of scenarios to help gauge the presence of disease, some measure of severity, and potentially response to disease-modifying therapy,” Adam Boxer at the University of California, San Francisco, wrote to Alzforum. Christopher Shaw at King’s College London noted that in the field of neurodegenerative disease, “NfL is the best biomarker for disease activity that we have.” 

Numerous studies have correlated high NfL in CSF and in blood with the severity of neurodegenerative diseases such as AD (Nov 2015 news; Jun 2016 news; Mar 2017 news). NfL levels are elevated in people with mild cognitive impairment as well, but it was unclear how early in disease the biomarker rises.

To answer this, Fox and colleagues turned to families with a history of early onset AD. Because family members who inherit the pathological mutation are certain to get the disease, researchers can investigate very early preclinical changes in this group. First author Philip Weston recruited 48 young to middle-aged people from families with an APP or PS1 mutation. Eighteen of them already had symptoms of AD, while the other 30 did not. Genetic testing subsequently showed that 19 of those 30 carried an AD gene. On average, these carriers were 10 years away from their parent’s age of onset. Participants took several cognitive tests and donated blood at baseline. NfL levels were measured with ultra-sensitive single molecule array (Simoa) assays (Apr 2016 conference news). A subset of participants underwent structural MRI scans, with 33 of those returning about one year later for a follow-up scan.

Asymptomatic carriers had a slight but statistically significant increase in serum NfL compared with noncarriers, 16.7 versus 12.7 pg/mL. In people with symptomatic disease, NfL surged to 46 pg/mL. This latter figure is similar to the plasma NfL Simoa measurements reported in other recent studies of MCI and AD patients (Mattsson et al., 2017; Zhou et al., 2017). Despite the average differences between the diagnostic groups in this study, individual values overlapped greatly (see image). Also seen in other studies, this overlap means that NfL by itself cannot be used as a diagnostic marker of disease stage, Fox noted.

However, NfL does seem to reflect disease severity within a given person. In asymptomatic carriers, its level correlated inversely with the estimated years to disease onset. Across all mutation carriers, higher NfL associated with worse cognitive scores on the MMSE and CDR sum of boxes. To estimate cognitive decline in mutation carriers, the researchers compared participants’ scores on the Wechsler Abbreviated Scale of Intelligence (WASI), which measures current IQ, with their scores on the National Adult Reading Test (NART), which provides a measure of their previous, healthy IQ. If the WASI was lower than expected in comparison with the NART, it suggested cognitive loss. This estimated decline also correlated with NfL.

NfL flagged structural brain changes, as well. High NfL correlated with low baseline brain volume and hippocampal volume, as well as a faster decline in whole-brain volume. Intriguingly, people with high NfL tended to have greater ventricular volume and more expansion of this space over time. Why this occurs is unclear, although one possibility is that it reflects neuroinflammation, Fox noted.

Commenters said these data strengthen the case for NfL tracking progression. They questioned how early the marker might detect disease, given that the increase in asymptomatic carriers was small. “It remains to be seen whether this marker will translate into the preclinical population,” said Sid O’Bryant at the University of North Texas Health Science Center, Fort Worth. Doug Galasko at the University of California, San Diego, agreed. He noted that to truly gauge serum NfL’s usefulness as a progression marker, researchers need to study how its levels change over time in individuals (see comment below).

Fox and colleagues are doing those longitudinal studies now in their familial cohort. “For some biomarkers, the rate of change is more relevant than the baseline value,” Fox noted. He will examine whether the rate of change better predicts progression than the absolute value does, and whether people with higher NfL progress sooner to symptomatic disease. If so, NfL levels could be used to select participants for clinical trials, he suggested.

“If NfL could be used to predict progression, it would revolutionize therapeutic trials,” O’Bryant said. However, he noted that it is possible that high NfL, instead of identifying those people who are furthest along in the disease process, distinguishes a subgroup of patients who have more aggressive disease or carry more pathological comorbidities, for example α-synuclein or TDP-43 deposits in addition to Aβ. That possibility should be investigated, O’Bryant believes.

Other hurdles remain before serum NfL advances to clinical use. In an accompanying Neurology editorial, Michelle Mielke at the Mayo Clinic in Rochester, Minnesota, noted that researchers need to characterize serum NfL across the general population and find out how it varies with age, race, sex, and different medical conditions, as well as how it compares with other markers of neurodegeneration such as total tau. “Understanding these aspects will help fast-track serum NfL and other potential blood-based markers into useful clinical practice,” Mielke wrote. Researchers also stressed the need to standardize measurement methods for NfL, as is being done for CSF biomarkers (Aug 2015 conference news; Apr 2017 conference news). 

One unanswered question in AD is whether NfL levels will drop with treatment. Data from mouse models hint at this, but so far there are no human data on treatment effects in AD (Bacioglu et al., 2016). However, the findings from MS patients suggest that NfL levels can respond rapidly to changes in disease intensity.

In the Swedish study, Nováková and colleagues followed 138 people with multiple sclerosis who changed their treatment regimen over the course of a year, and compared them with 10 untreated MS patients and 42 healthy controls. The untreated participants maintained stable NfL levels of about 26 pg/mL, compared with 10 pg/mL in healthy controls. However, in 50 patients who began a disease-modifying treatment, median NfL dropped to about 20 pg/mL. Another 68 switched from a first-line therapy, such as glatiramer acetate or methotrexate, to a more effective second-line therapy such as cyclophosphamide or rituximab, and their NfL dropped on average from 17.7 to 12.4 pg/mL. By contrast, the 20 patients who switched between treatments of similar efficacy stayed stable, with a median NfL of just over 15 pg/mL.

Adding to the evidence that NfL reflects an active disease process, among the 98 participants with the relapse-remitting form of MS, NfL rose during relapses and dropped during remission. Among all MS patients, serum NfL correlated with the number of lesions seen on brain scans.

“Serum NfL could be used in clinical practice as a tool to monitor treatment efficacy in MS,” Nováková wrote to Alzforum. She believes the findings may be relevant for neurodegenerative disease as well, given that NfL levels reflect brain damage in numerous conditions.

O’Bryant agreed. “This is really exciting data, because it suggests that an effective disease-modifying therapy may change NfL levels,” he said. If the finding holds up in AD, serum NfL could serve as a surrogate outcome marker, he suggested.—Madolyn Bowman Rogers

Comments

1. • Douglas Galasko
     University of California, San Diego
   • Posted: 17 Nov 2017

   The study of NfL in autosomal-dominant AD (ADAD) provides exciting preliminary evidence supporting this as a progression marker. The effect size was much more impressive in symptomatic patients than in preclinical carriers, and it will be interesting to see how within-subject longitudinal serum or plasma NfL changes over time. In principle, this may provide one way to help monitor a therapeutic intervention that slows down neurodegeneration.

   The study in MS takes monitoring of progression and of therapeutic intervention one step further, because of the availability of longitudinal plasma samples. The data overall support the use of NfL in the early stages of relapsing-remitting MS and in progressive MS, across groups of patients, and showed a beneficial effect of induction of disease-modifying therapy. It will be interesting to see how well NfL may behave as a way to monitor individual patients with MS, particularly in comparison to serial MRI.

   View all comments by Douglas Galasko
2. • Takaomi Saido
     RIKEN Center for Brain Science
   • Posted: 27 Nov 2017

   This is very interesting, but what is the underlying mechanism?

   Why NfL? Why not other neurofilament components? There must be answers.

   View all comments by Takaomi Saido

Make a Comment

To make a comment you must login or register.

References

News Citations

1. Neurofilament Light Chain: A Useful Marker for AD Progression? 6 Nov 2015
2. Blood NfL Looks Good as Progression and Outcome Marker 9 Jun 2016
3. Blood Neurofilament Light a Promising Biomarker for Alzheimer’s? 31 Mar 2017
4. WANTED: Biomarkers for Drug Trials in Frontotemporal Dementia 25 Apr 2016
5. CSF Aβ Assays Remain Fickle: Robots to the Rescue? 19 Aug 2015
6. Are CSF Assays Finally Ready for Prime Time? 15 Apr 2017

Paper Citations

1. Mattsson N, Andreasson U, Zetterberg H, Blennow K, Alzheimer’s Disease Neuroimaging Initiative. Association of Plasma Neurofilament Light With Neurodegeneration in Patients With Alzheimer Disease. JAMA Neurol. 2017 May 1;74(5):557-566. PubMed.
2. Zhou W, Zhang J, Ye F, Xu G, Su H, Su Y, Zhang X, Alzheimer’s Disease Neuroimaging Initiative. Plasma neurofilament light chain levels in Alzheimer's disease. Neurosci Lett. 2017 May 22;650:60-64. Epub 2017 Apr 18 PubMed.
3. Bacioglu M, Maia LF, Preische O, Schelle J, Apel A, Kaeser SA, Schweighauser M, Eninger T, Lambert M, Pilotto A, Shimshek DR, Neumann U, Kahle PJ, Staufenbiel M, Neumann M, Maetzler W, Kuhle J, Jucker M. Neurofilament Light Chain in Blood and CSF as Marker of Disease Progression in Mouse Models and in Neurodegenerative Diseases. Neuron. 2016 Jul 6;91(1):56-66. Epub 2016 Jun 9 PubMed.

Further Reading

News

1. Neurofilament Light Chain as Prognostic Biomarker in ALS 10 Mar 2017
2. Touchdown for NfL: Blood Test Tells Parkinson's from Related Disorders 10 Feb 2017
3. Fluid NfL Shines, Tau PET Dims, in the Hunt for FTD Biomarkers 29 Sep 2016
4. Biomarkers Differentiate Parkinsonian Diseases and Forecast Decline 21 Aug 2015
5. Blood Marker May Predict ALS Progression 4 Jun 2015
6. For Hockey Players, Brain Damage Can Be Measured in Blood 20 Mar 2014