Academy of Diagnostics & Laboratory Medicine - Scientific Short
Scientific Shorts are brought to you by the
Neurofilament proteins are cylindrical proteins that are expressed in large caliber myelinated axons. They provide a structural backbone to neuronal cytoskeleton and determine axonal diameter; hence, they are markers of neuroaxonal injury and insult to brain in multiple conditions, including but not limited to multiple sclerosis (MS), Alzheimer’s Disease, and Amyotrophic Lateral Sclerosis etc1. Neurofilament proteins consist of subunits known as neurofilament light-chain (NfL), medium-chain (NfM), heavy-chain (NfH), α-internexin, and peripherin. Specifically in MS, Elevated NfL levels in blood or cerebrospinal fluid (CSF) correlate with gadolinium-enhancing lesions, T2 lesion volume, and progression of disability based clinical scales including but not limited to Expanded Disability Status Scale, and response to Disease Modifying Therapy1.
Neurofilament levels can be measured in CSF and blood (serum or plasma). Although CSF levels are approximately 40 times higher than blood, the invasive nature of CSF collection makes it less practical to use for monitoring response of therapy or relapse. In past few years, the NfL protein has gained attention as commercial vendors worked on regulatory approval of this assay for MS as an aid in the detection of disease activity. Roche Elecsys NfL assay received US FDA approval and breakthrough device designation in 2023 for the population between 18-55 years old to aid in disease activity for relapsing remitting and secondary progressive MS. Siemens Healthineers NfL Atellica assay received Health Canada approval in 2024 to aid in the detection of disease activity for patients 18-55 years old with relapsing MS. It is important to keep in mind that evidence of the clinical utility of NfL came primarily from Quanterix Simoa assay used in clinical trials. Recent studies comparing commercial assays with the Simoa assay showed variable degrees of bias, specifically, the Elecsys assay demonstrated an 84% negative bias, while the Atellica assay showed a 13% negative bias relative to Simoa. . The degree of proportional bias observed among these commercial assays is due to a lack of standardized reference material, detection methods, and the use of different antibodies2. Both Simoa and Atellica methods use Uman antibodies, suggesting the possible reason of the lesser variability in slope, while no antibody information is available for the Elecsys method2.
Interpretation of NfL is affected by confounders including age, body weight, sex and race, as well as other comorbidities such as diabetes and other neurological conditions5. Previous studies has shown that serum NfL levels rise steadily by 2.2% every year and the slope of increase is steeper between the age of 50-60 years3. While Neurofilament light (NfL) levels in cerebrospinal fluid (CSF) remain unaffected by body mass index (BMI)3, blood-based concentrations (serum/plasma) are significantly influenced by BMI , and must be accounted for when interpreting serum or plasma results. When baseline sNFL are not available, transformation of absolute value to z-scores or percentiles can begin to reliably correct for confounding factors. Previously, an NfL reference database has been designed by a Swiss group4 to adjust BMI and age adjusted sNFL percentile value and z-score in both adult5 and pediatric populations6. Scores and percentiles (considered interchangeable) are drawn from the control population and have been validated in controlled trials as a way to express deviation from normal, adjusting for age and BMI. A previously published study suggested if z-score>1.25, odds ratio (OR) of worsening of disease is 2.28 while with a z-score>1.75, OR is 3.85 in MS7. Any z-score >1.5 and ≤2 (97.7th percentile) suggests NfL is elevated while z-score >2 indicated NfL concentration is strongly elevated7. It is important to keep in mind the original reference database was based on the Simoa assay. Importantly, the previous version of the Simoa assay (Simoa® NF-lightTM V1) has been used to derive the reference database. This has been replaced by a new version (Simoa® NF-lightTM V2) and may require a conversion factor for future studies.
Another clinical interpretation can be provided by calculation of the reference change value, a significant change of NfL concentrations between patient’s two samples using biological variation in accordance with EFLM biological variation database and assay’s analytical performance7. For instance, assuming an 8.6% within-subject variation (from the EFLM database) and an analytical variation of 6-7%, the change needed between consecutive measurements to exceed both biological and analytical variation and thus be clinically relevant is approximately 23-31%. This value should be calculated by an individual laboratory with consideration to their own instrument’s typical analytical variation.
In summary, until the assays are standardized, assay-specific reference intervals provide valuable information to interpret the results, but additional information on age and BMI-adjusted and z-score reference values for vendor-specific assays will help in the additional clinical interpretation of the assay.
