Neurofilament-light chain (NfL) has emerged as one of the more promising biomarkers in multiple sclerosis. Neurofilaments are proteins expressed by axons and released into the cerebrospinal fluid, a process that is accelerated with neuroaxonal damage. It has not been determined if elevated concentrations of NfL are due to increased expression and/or release of NfL, impaired clearance of NfL by microglia, or other factors. While the current understanding of NfL is incomplete, an elevated concentration in CSF, plasma or serum, although not specific to MS, is a potentially useful biomarker of axonal damage.
In MS, elevated NfL levels are associated with inflammatory disease activity, including relapses and lesion load (Kuhle et al. Mult Scler 2013;19:1597-1603. Disanto et al. Ann Neurol 2017;81:857-870). Accordingly, some MS clinics are now routinely collecting NfL as an aid to evaluating prognosis, disease severity and treatment response. To assist this process, a new review has summarized some of the key issues in the collection, storage and interpretation of NfL in clinical practice (Arslan et al. Clin Chem Lab Med 2023;61:1140-1149. Free full text at www.degruyter.com/document/doi/10.1515/cclm-2023-0036/html).
Although NfL levels are about 50-fold higher in CSF than in blood, blood sampling is generally preferred to avoid lumbar puncture. Single-molecule array (Simoa), which enables detection of individual NfL molecules, has greater analytical sensitivity than ELISA or electrochemiluminescence (ECL) methods (Kuhle et al. Clin Chem Lab Med 2016;54:1655-1661). The Quanterix NF-light assay, which calibrates using recombinant human NfL, may be more reproducible than ‘home brew’ assays that use NfL obtained from bovine brain. The two calibrators are correlated but the Quanterix results must be multiplied by five to make them comparable to bovine calibrator results (Hendricks et al. Bioanalysis 2019;11:1405-1418).
Serum and plasma levels of NfL are correlated but serum levels are about 10% higher than those analysed from EDTA plasma (Hviid, et al. Scand J Clin Lab Invest 2020;80:291-295). Test results are more variable if centrifugation is delayed 6-24 hours after sample collection (van Lierop et al. Clin Chem Lab Med 2022;60:842-850). Centrifugation temperature does not appear to be a factor. NfL levels in serum and plasma samples are generally stable following prolonged exposure to room temperature and multiple freeze-thaw cycles.
Several factors will influence test results. Of these, the most important is age and several groups have reported age-based reference values for serum and plasma NfL (Valentino et al. Mult Scler Relat Disord 2021;54:103090. Simrén et al. Brain Commun 2022;4:fcac174; free full text at https://academic.oup.com/braincomms/article/4/4/fcac174/6628649). As a rough guide, the normal sNfL value is <10 pg/mL in healthy individuals aged <40 years, increasing 2-4% per year (Valentino 2021). NfL levels are inversely correlated with body-mass index and blood volume. NfL concentrations are higher in patients with kidney disease.
NfL is subject to diurnal variation. Preliminary results indicate that sNfL levels are >10% higher in the morning compared to the evening, so morning sampling is preferred (Benedict et al. Neurology 2020;94:e1181-1189).
It should be noted that normal serum or plasma NfL levels do not exclude neurodegenerative or other conditions associated with low-grade axonal injury. In such cases, cNfL may be more appropriate since it is more sensitive in detecting axonal damage.