Background: The longitudinal dynamics of neurofilament light chain (NfL) in multiple system atrophy (MSA) were incompletely illuminated. This study aimed to explore whether the plasma NfL (pNfL) could serve as a potential biomarker of clinical diagnosis and disease progression for MSA.
Methods: We quantified pNfL concentrations in both a large cross-sectional cohort with 214 MSA individuals, 65 PD individuals, and 211 healthy controls (HC), and a longitudinal cohort of 84 MSA patients. Propensity score matching (PSM) was used to balance the age between the three groups. The pNfL levels between groups were compared using Kruskal-Wallis test. Linear mixed models were performed to explore the disease progression-associated factors in longitudinal MSA cohort. Random forest model as a complement to linear models was employed to quantify the importance of predictors.
Results: Before and after matching the age by PSM, the pNfL levels could reliably differentiate MSA from HC and PD groups, but only had mild potential to distinguish PD from HC. By combining linear and nonlinear models, we demonstrated that pNfL levels at baseline, rather than the change rate of pNfL, displayed potential prognostic value for progression of MSA. The combination of baseline pNfL levels and other modifiers, such as subtypes, Hoehn-Yahr stage at baseline, was first shown to improve the diagnosis accuracy.
Conclusions: Our study contributed to a better understanding of longitudinal dynamics of pNfL in MSA, and validated the values of pNfL as a non-invasive sensitive biomarker for the diagnosis and progression. The combination of pNfL and other factors is recommended for better monitoring and prediction of MSA progression.
Keywords: Biomarker; Diagnosis; Multiple system atrophy; Parkinson’s disease; Plasma neurofilament light chain; Prediction.
© 2023. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany.