Plasma LC-HRMS metabolomics in parkinsonian syndromes: comparative evaluation of sample preparation protocols for exploratory metabolic signatures

Untargeted plasma metabolomics by LC-HRMS is increasingly used to explore candidate biomarkers in neurodegenerative disease, yet the analytical output and downstream biological interpretability remain highly dependent on end-to-end workflow choices. This limits reproducibility and cross-study comparability in parkinsonian syndromes where objective fluid biomarkers for differential diagnosis are still lacking. Here, we implemented a QC-anchored, untargeted LC-HRMS plasma platform to compare sample preparation strategies and to explore disease-associated metabolic patterns across Parkinson’s disease (PD), multiple system atrophy–parkinsonian subtype (MSA-P), and progressive supranuclear palsy–parkinsonism (PSP-P) in a prospective single-center cohort (n = 102; 57 PD, 19 MSA-P, 26 PSP-P). Three protein precipitation-based workflows were compared within a harmonized analytical framework: tube precipitation, 96-well precipitation-filtration, and 96-well phospholipid removal. Integrated preparation QCs, system QCs, and blanks supported performance monitoring, while MS2 annotation enabled interpretation of disease-associated signals. Phospholipid removal provided the most favorable trade-off for discovery and quantitation, delivering reduced phospholipid-related chromatographic background, improved quantitative behavior, and a higher density of MS2-supported annotations despite a modest reduction in raw feature counts. Differential abundance was assessed using empirical Bayes moderated linear models implemented in limma. Pairwise contrasts were defined for PD vs. MSA-P, PD vs. PSP-P, and MSA-P vs. PSP-P, and P values were adjusted using the Benjamini–Hochberg false-discovery rate (FDR). Across workflows, the most robust exploratory disease-associated pattern was observed for PD versus PSP-P, whereas MSA-P comparisons yielded limited FDR-significant features. Because of the absence of MSI Level 1 annotation, the hypothesis-generating pathway-level was based on exploratory metabolomics interpretation, requiring targeted confirmation.