Cerebrospinal fluid and blood neurofilament light chain in Parkinson's disease and atypical parkinsonian syndromes: a systematic review and Bayesian network meta-analysis.

Abstract

Background and objective: The value of neurofilament light chain (NfL) levels as a biomarker for the diagnosis and differential diagnosis in patients with Parkinson's disease (PD) and atypical parkinsonian syndromes (APS) remains controversial. Furthermore, few studies have directly compared NfL levels among specific APS categories. This study aimed to compare cerebrospinal fluid (CSF) and blood NfL levels among PD, APS, other PD-related disorders, and controls, as well as rank NfL levels across these groups.

Methods: PubMed, Embase, Web of Science, and the Cochrane Library were searched from the inception up to November 1st, 2024, to identify eligible studies reporting CSF or blood NfL concentrations in PD, PD dementia (PDD), multiple system atrophy (MSA), progressive supranuclear palsy (PSP), dementia with Lewy bodies (DLB), corticobasal syndrome (CBS), vascular parkinsonism (VP), essential tremor (ET), idiopathic rapid eye movement sleep behavior disorder (iRBD), and controls. The Bayesian approach was utilized to estimate the standardized mean difference (SMD) and the associated 95% credible intervals (CrIs) of NfL levels. The surface under the cumulative ranking curve (SUCRA) was employed to evaluate the ranking probabilities of NfL levels. Subgroup analysis and meta-regression were conducted to explore the sources of heterogeneity.

Results: The present network meta-analysis (NMA) included 78 studies with 13,120 participants (4050 controls, 5021 PD, 191 PDD, 1173 MSA, 887 PSP, 1254 DLB, 319 CBS, 160 ET, 65 iRBD, and 0 VP). Of these, the NMA of CSF NfL included 34 studies with 6,013 participants, while the NMA of blood NfL included 49 studies with 7,787 participants. Both CSF and blood NfL levels were significantly elevated in patients with PD and APS compared to controls. Compared to PD patients, CSF NfL levels were significantly elevated in MSA (SMD 1.85; 95% CrI 1.55-2.15), CBS (1.42; 1.08-1.75), PSP (1.35; 1.06-1.64), and DLB 0.52; 0.20-0.85) patients. Similarly, blood NfL levels were significantly higher in patients with MSA (1.36; 1.02-1.71), PDD (1.19; 0.65-1.72), PSP (1.15; 0.77-1.54), CBS (0.92; 0.11-1.72), and DLB (0.63; 0.14-1.12) compared to PD. Among APS, CSF NfL levels in MSA patients were significantly higher than those in PSP, DLB, and CBS patients, while blood NfL levels in MSA patients were significantly higher only compared to DLB. In both CSF and blood NfL, MSA patients exhibited the highest probability of ranking first for NfL level elevations (CSF: SUCRA = 0.998; blood: SUCRA = 0.925). Age significantly influenced the SMD of the comparison between MSA and PD in CSF NfL (β = -0.15; p = 0.016).

Conclusions: CSF and blood NfL levels in PD and APS are higher than those in controls, and all APS categories show higher levels than PD, suggesting that NfL levels may serve as a potential biomarker for the differential diagnosis between PD and APS. However, caution is warranted when using NfL as a diagnostic biomarker for PD. Significant differences in NfL levels are also observed between certain APS categories. Patients with MSA exhibit the highest NfL levels among PD and related disorders.