Objective: Define lipid signatures in sporadic, LRRK2, and GBA1 PD and pre-manifest carriers and assess medication impacts.

Background: Lysosomal dysfunction contributes to PD, particularly in LRRK2 and GBA1 carriers. Many lipids and metabolites are processed in lysosomes, and CNS impairment may alter CSF profiles. We analyzed CSF samples from the PPMI cohort, including unaffected carriers and PD patients, to characterize lysosomal dysfunction and differentiate disease- and treatment-related effects.

Method: CSF samples were analyzed from controls (N=120), LRRK2 carriers (N=120), GBA carriers (N=100), sPD (N=120), LRRK2 PD (N=120), and GBA PD (N=60). Among PD patients, 21% (sporadic), 12% (LRRK2), and 18% (GBA) were untreated. A total of 507 metabolites were analyzed, with 135 classified as lipids. Levels were estimated relative to internal standards and batch adjusted. The model was run separately for PD, LRRK2, or GBA classification using GLMM-LASSO with ANCOVA-adjusted residuals to remove covariate effects. Medication was a random effect. LASSO-selected key features were subsequently ranked using RFE with 5-fold cross-validation to identify the most predictive subset.

Results: Dopamine and polyamines were the strongest PD discriminators but were significantly influenced by medication. Lysosomal lipid accumulation was more pronounced in untreated patients, suggesting a link to both disease pathology and treatment status. In genetic subgroups, GBA carriers exhibited sphingolipid alterations, particularly in hexosylceramides, attenuated by MAO-B inhibitors. LRRK2 carriers showed DHA lipid enrichment independent of medication, suggesting intrinsic lysosomal dysfunction. SVM-RFE achieved AUCs of 0.7 for PD, 0.6 for LRRK2, and 0.6 for GBA

Conclusion: Although not top features in the ML model, lysosomal lipids distinguished PD from controls primarily for untreated PD patients, limiting their utility for overall classification but underscoring its relevance to disease and treatment status. DHA lipid enrichment in LRRK2 carriers was independent of disease and treatment status, suggesting an intrinsic lysosomal dysfunction signature. ML models classified PD and genetic subtypes with varying success, reinforcing the influence of lysosomal lipid metabolism. Findings highlight medication effects on metabolic biomarkers and the importance of lysosomal lipid metabolism in PD subtype differentiation.

« Back to 2025 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/lysosomal-lipid-dysregulation-in-parkinsons-disease-genetic-and-medication-effects-in-csf-metabolomics/