Head-to-head comparison of four plasma neurofilament light chain (NfL) immunoassays

Background: Neurofilament Light Chain (NfL) is an emerging blood biomarker of neuro-axonal injury and neurodegeneration with the potential to be used in the clinical management of various neurological conditions. Various NfL immunoassays are in development on high-throughput automated systems, but little information is available related to the comparability between assays. In this study, we performed a head-to-head comparison of four NfL immunoassays using plasma samples from individuals with various neurological conditions. Methods: EDTA plasma samples in which NfL was ordered clinically were stratified according to diagnosis. NfL concentrations (pg/mL) in plasma were obtained using the Quanterix Simoa (R), the Roche Elecsys, the Siemens Healthineers Atellica (R) IM, and the Fujirebio Lumipulse (R) NfL assays. Passing-Bablok regression analyses were performed to assess the correlation and bias between methods. Additionally, the distribution of NfL concentrations for each assay was assessed in three disease groups: amyotrophic lateral sclerosis (ALS) upon initial diagnosis, ALS treated, and multiple sclerosis (MS). Results: The R-2 between assays were all >= 0.95, however, significant proportional bias was observed between some assays. In particular, the Roche Elecsys assay NfL concentrations were significantly lower (similar to 85 %) when compared against the other three assays. The four assays were comparable with regards to the percentage of patients that were identified as having an elevated NfL result in the various clinical groups: ALS initial diagnoses (83-94 %), ALS untreated (93-100 %), and MS (8-18 %). Conclusions: This is the first study describing a head-to-head comparison of four automated NfL immunoassays. We demonstrate that there is a strong correlation between assays but a lack of standardization which is evident by the bias observed between some of the evaluated methods. These analytical differences will be important to consider when using NfL as a biomarker of neurodegeneration.