Skip to main content
Evidence-Based Supplement Research
Evidence-Based Supplement Research

Cerebrospinal fluid biomarkers for diagnosis of Parkinson's disease: a systematic review and network meta-analysis.

  • 2025-12
  • Journal of neurology 272(12)
    • Siming Li
    • Chen Yang
    • Jiayi Wu
    • Yuanchu Zheng
    • Zhenwei Yu
    • Genliang Liu
    • Yaqin Yang
    • Tao Feng

Study Design

Type
Systematic Review
Population
4925 PD patients, 698 multiple system atrophy (MSA), 177 progressive supranuclear palsy (PSP), 78 dementia with Lewy body (DLB) and 3072 healthy controls (HCs) or non-neurological controls (NNCs)
Methods
Comprehensive research on PubMed, Web of Science, Embase and Cochrane library from inception until December 31st, 2024; random effects models for sensitivity, specificity, positive likelihood ratio (PLR) and negative likelihood ratio (NLR), diagnostic odds ratio (DOR), and 95% CIs; summary receiver operating characteristic (SROC) curves; network meta-analysis based on the ANOVA model and surface under the cumulative ranking curve (SUCRA) scores

Objectives

There are multiple cerebrospinal fluid (CSF) biomarkers with potential for distinguishing Parkinson's disease (PD) from atypical parkinsonian syndromes (APSs) and controls; however, consensus on their diagnostic performance remains limited. This study aims to systematically evaluate and rank the diagnostic accuracy of CSF biomarkers in differentiating PD from APS and controls through network meta-analysis, determining the clinical diagnostic yield of these biomarkers and whether they should be considered as first-line diagnostic and differentiating tools.

Methods

A comprehensive research was performed on PubMed, Web of Science, Embase and Cochrane library from inception until December 31st, 2024. Random effects models for sensitivity, specificity, positive likelihood ratio (PLR) and negative likelihood ratio (NLR), diagnostic odds ratio (DOR), and 95% CIs were used to calculate test accuracy. In addition, summary receiver operating characteristic (SROC) curves were used to summarize the overall diagnostic performance. The network meta-analysis based on the ANOVA model and surface under the cumulative ranking curve (SUCRA) scores were selected to rank the diagnostic performance of the included biomarkers by calculating the relative sensitivity, specificity, and diagnostic odds ratio (DOR).

Results

Seventy eligible studies containing 4925 PD patients, 698 multiple system atrophy (MSA), 177 progressive supranuclear palsy (PSP), 78 dementia with Lewy body (DLB) and 3072 healthy controls (HCs) or non-neurological controls (NNCs) were included in our meta-analysis. CSF Alpha-synuclein seed amplification assays (α-syn SAAs) showed high diagnostic accuracy with pooled sensitivity of 0.91 (95% CI 0.89-0.92) and specificity of 0.95 (95% CI 0.94-0.96) in distinguishing PD from HC or NNCs. The Area Under the Curve (AUC) of the SROC curve of CSF Neurofilament Light Chain (NfL) was 0.91, which effectively distinguished between PD and MSA. Both α-syn SAAs and NfL demonstrated similarly high diagnostic accuracy for differentiating PD from PSP. Only CSF Aβ1-42 has been evaluated for the differential diagnosis between PD and DLB, which demonstrated low pooled sensitivity (0.59, 95%CI 0.49-0.69) and specificity (0.68, 95%CI 0.56-0.78). Network meta-analysis using ANOVA model showed that among 13 CSF biomarkers, α-syn SAAs had the highest relative DOR (285.01, 95% CI 156.72-475.89), relative sensitivity (1.29, 95% CI 1.19-1.41) and specificity (1.54, 95% CI 1.39-1.72), while the SUCRA curves also reach the same conclusion, indicating the highest diagnostic accuracy in distinguishing PD from HCs or NNCs. Notably, these studies have general heterogeneities.

Conclusions

CSF α-syn SAAs could serve as promising biomarkers for distinguishing PD from PSP and HCs or NNCs. Meanwhile, NfL is suitable for differentiating PD from MSA and PSP. In addition, the confirmation of these CSF biomarkers, as well as the discovery of new biomarkers, requires extensive study in large, independent cohorts. The main limitations of this analysis are the reliance on evidence derived from case-control studies, which are susceptible to bias, and the high level of heterogeneity observed for some biomarkers.

Research Insights

    Back to top