Real-world analytical and clinical evaluation of the Siemens Atellica® immunoassay for plasma neurofilament light chain in multiple sclerosis

Background: Plasma neurofilament light chain (pNfL) is an established biomarker of neuroaxonal injury in multiple sclerosis (MS), but analytical variability across platforms may limit clinical comparability. Objective: We compared pNfL values obtained using Lumipulse® and Atellica® assays, and evaluated the clinical correlates (associations with disability status, disease activity, MRI outcomes, and treatments) of Atellica® pNfL in a real-world MS cohort. Methods: Baseline plasma samples from people with MS were analysed using both immunoassays. Agreement was assessed using Passing-Bablok regression, Lin's concordance correlation coefficient, and Bland-Altman analysis. Associations between Atellica® pNfL and demographic and clinical variables were evaluated using multivariable regression. Results: We included 394 people with MS (mean age 41.0 ± 13.0 years; 61.5% female; median EDSS 2.5). Lumipulse® and Atellica® pNfL values showed strong association, with Passing-Bablok regression indicating proportional bias (slope 0.78; 95% CI 0.70-0.86; intercept 3.24; 95% CI 2.46-3.90) and moderate concordance (ρc = 0.49). Higher Atellica® pNfL levels were independently associated with older age and greater disability, but not with short-term clinical or radiological activity. Conclusions: Atellica® and Lumipulse® showed strong association but only moderate concordance in pNfL measurements. These findings indicate that the two platforms cannot be considered interchangeable. Atellica® pNfL reflects disability status and represents a clinically meaningful biomarker when interpreted within assay-specific frameworks in real-world MS care.