Multi-omics study identifies diagnostic metabolic signatures of early Parkinson's disease associated with dysregulated glutathione and TCA cycle metabolism.

BACKGROUND

The development of non-invasive diagnostic tools for Parkinson's disease (PD) remains an unmet clinical need. Given the contribution by peripheral metabolism and environmental factors toward PD pathogenesis, comprehensive profiling of the plasma phenome could enable early disease detection. Accordingly, this study aimed to develop a non-invasive diagnostic panel for PD, to comprehensively characterize potential environmental triggers of PD pathogenesis, and to delineate the relationship between peripheral metabolism and disease progression.

METHODS

We performed quantitative plasma phenome profiling - including metabolomics, lipidomics and bile acid analysis - in two independent cohorts of early PD (n = 226, median disease duration <5 years) comprising predominantly treated and treatment-naive patients, respectively, and healthy controls (HCs; n = 122) to identify PD-associated metabolites. Longitudinal analyses were conducted in a patient subset over a two-year follow-up. Longitudinal analyses (2 years) and trans-omics integration were applied.

RESULTS

Trans-omics integration showed glutathione metabolism dysregulation, with reduced plasma γ-glutamyl peptides and elevated sphingomyelin linked to motor dysfunction; higher baseline γ-glutamyl peptides predicted worse motor prognosis in PD. Reduced sphingosine-1-phosphates (S1Ps) and lysophosphatidic acids (LPAs) were correlated with abated tricarboxylic acid (TCA) cycle and motor deterioration. A three-metabolite panel (pyroglutamic acid, 2-methylbutyrylglycine, α-ketoisovaleric acid), selected via LASSO regression, robustly discriminated early PD from HCs, consistently outperforming clinical indices and lipid/bile acid panels (Discovery AUROC = 0.97 [0.96-0.99]; Validation AUROC = 0.83 [0.75-0.91]).

CONCLUSION

Our findings demonstrate the clinical utility of plasma metabolites as robust, medication-independent biomarkers for early PD diagnosis, and delineate disease-associated metabolic deficits as potential targets for nutritional/metabolic intervention.