Biofluid biomarkers across neurological disorders: a systematic meta‐analysis of shared and unique molecular signatures

TL;DR

This study identifies shared and unique protein biomarkers in cerebrospinal fluid and plasma across neurological disorders, revealing insights into their diagnostic potential and biological heterogeneity.

Contribution

The study introduces a meta-analysis framework to compare dementia-specific biomarkers across biofluids and disorders, revealing matrix-specific marker behavior.

Findings

Effect sizes for seven proteins show opposite directions in CSF and plasma, suggesting CSF-specific utility.

THOP1 is elevated in prodromal Parkinson's disease, indicating potential for early diagnosis.

DDC and CRH are uniquely dysregulated in Parkinson's-plus syndromes, highlighting shared biological mechanisms.

Abstract

A comprehensive analysis of proximity‐extension assay (PEA) proteomic studies have led to the development and validation of novel cerebrospinal fluid (CSF) biomarker panels for Alzheimer's disease (AD; Del Campo et al., 2022 Nat Aging) and dementia with Lewy bodies (DLB; Del Campo et al., 2023 Nat Comm.). However, to what extent the markers are dementia‐ and matrix‐ (CSF or plasma) specific is largely unknown. To establish the clinical context for these novel markers, data from PEA proteomic studies in CSF and plasma will be collected and meta‐analyzed across neurological disorders.

A systematic review following PRISMA‐guidelines is being conducted to collect relevant CSF and plasma PEA studies. For six CSF and plasma studies with readily available data in dementia and movement disorders, Cohen's d effect sizes were calculated for seven proteins identified as AD‐specific (ABL1, ITGB2, ENO2, SMOC2, and THOP1) or DLB‐specific (DDC and CRH) – measuring the mean difference between disease and control groups. A random‐effects meta‐analysis model was applied within the same disorders and plotted, Figure 1.

Analysis showed that for all seven proteins, when comparing effect sizes between CSF and plasma, the direction of change is opposite. This suggests that these markers may only be useful in CSF, and in plasma may even associate with other disorders; as THOP1 seems to be most increased in prodromal PD, Figure 1. Within‐disease analysis revealed variability mainly in preclinical AD, for ABL1, ITGB2, and THOP1, and prodromal PD for DDC, supporting the biological heterogeneity of these groups. Additionally, DDC and CRH in CSF are distinctly dysregulated for Parkinson's‐plus syndromes (PD, DLB, PSP, CBS) highlighting the overlapping biological mechanisms.

Our meta‐analysis across multiple cohorts is expanding to include around 150 studies in dementias, movement disorders and in psychiatric conditions. The framework of analysis established in this study provides a foundation for future expansion of the database in various neurological disorders, such as psychiatric disorders, and to other proteomic platforms. These findings may contribute to the development of more accurate diagnostic tools and offer insights into the complex interplay of protein markers in neurodegenerative diseases.

Figures