Dr Henrik Zetterberg, PhDHenrik Zetterberg, MD, PhD
New study findings suggest that neurofilament light in the cerebrospinal fluid (cNfL) has multi-disease use as a biomarker for neuroaxonal damage, with the potential to help differentiate between frontotemporal dementia (FTD) and Alzheimer disease, and Parkinson disease from atypical parkinsonian syndromes.

Ultimately, the investigators noted that there may be a need for age- and sex-specific, as well as possibly disease-specific, reference values when using cNfL as a biomarker.

For some time already, the biomarker has been explored with interest in multiple sclerosis (MS) as a measurement of the effect of disease-modifying therapy. As an ultrasensitive assay has already been developed, and as such, serum NfL (sNfL) may replace cNfL. However, the investigators wrote that these findings can be readily translated to sNfL.

The systematic review and meta-analysis, including 10 years of data on cNfL, was conducted by the NFL Group, including Claire Bridel, MD, PhD, from the neurochemistry laboratory, department of clinical chemistry, VU University Medical Centre, and Henrik Zetterberg, MD, PhD, professor, neurochemistry, University of Gothenburg. They included 10,059 individuals and identified 35 separate diagnoses, including inflammatory diseases of the central nervous system (n = 2795), dementias and predementia stages (n = 4284), parkinsonian disorders (n = 984), and healthy controls (n = 1332).

“Although cNfL overlapped between most clinically similar conditions, its distribution did not overlap in FTD compared with other dementia subtypes or in [Parkinson] compared with atypical parkinsonian syndromes, indicating clinical potential in differentiating these conditions,” Bridel, Zetterberg, and others wrote.

The investigators compared cNfL in those who were not healthy controls and found it elevated for the majority of neurological conditions identified. Those who were HIV-positive with impaired cognition (iHIV), along with those with FTD, amyotrophic lateral sclerosis (ALS), and Huntington disease had the highest levels of cNfL. Although, there were no age-specific associations were observed with these diseases. Bridel et al. noted that this may suggest that some “pathological processes may cause plateau levels or mask age associations.”

There was an observed increase in the cNfL of the healthy controls with age—as well as most of the conditions identified, though the controls had the strongest association with age. About 66% of the diagnoses were observed to have a positive association with age and cNfL, which was hypothesized to reflect either a decrease in CSF clearance with age, the presence of a preclinical age-related neurological condition, or age-related neuronal loss. As such, age-specific reference values may be required, and cNfL’s diagnostic potential may decrease with age.

The group suggested that due to some conditions showing higher cNfL levels among men, most notably the controls, sex-specific reference values might be required. Ultimately, the clinical relevance of these findings, observed in a minority of the conditions, is still unknown. In 33.3% of diagnoses, including the healthy controls, MS, Alzheimer disease, and Parkinson disease, cNfL was higher in men than women.

All told, 46% of the variations in cNfL in the best-fitting model were explained by age, sex,  or cohort associations. This, Bridel et al. wrote, indicates that “many determinants of cNfL remain to be identified.” They noted that cNfL overlapped in conditions with the most clinically similar diagnoses, save for FTD and iHIV, which they observed to segregate from other dementias. Similar results were observed with Parkinson disease, which segregated from atypical parkinsonian syndromes.
 
They also suggested that disease duration and severity may be influencing factors in levels of cNfL, though the database for this metanalysis did not allow for this investigation. Studies aimed at specifically revealing any associations of these variables and others such as smoking, physical activity, and body size, are ongoing.

The group identified several limitations of the work, first being the use of clinical criteria for diagnoses, which they noted was mostly a concern among the dementias and parkinsonian syndromes. They also identified the inability to capture dementia of multifactorial origin, the limited data and age ranges for some conditions, and the inclusion of only studies which used a specific immunoassay, “in an attempt to reduce heterogeneity due to the analytical procedure,” they detailed.

REFERENCE
Bridel C, van Wieringen WN, Zetterberg H, Tijms BM, Teunissen CE. Diagnostic value of cerebrospinal fluid neurofilament light protein in neurology: A systematic review and meta-analysis. JAMA Neurol. Published online June 17, 2019. doi:10.1001/jamaneurol.2019.1534.