Paired cerebrospinal fluid metabolomic changes across repeated anodal transcranial direct current stimulation in Parkinson’s disease: an exploratory within-subject study

Abstract

Transcranial direct current stimulation (tDCS) has been investigated as an adjunctive intervention for Parkinson’s disease (PD), but its biochemical correlates in the human central nervous system remain poorly characterized. We examined participant-level cerebrospinal fluid (CSF) metabolomic changes across a five-day course of repeated anodal tDCS. Fourteen patients with PD received five consecutive daily sessions of anodal tDCS over the primary motor cortex. Clinical outcomes and CSF metabolomic profiles were assessed before and after the intervention. After analytical-quality filtering, feature-level paired analyses were conducted in 12 complete polar pairs and 13 complete non-polar pairs, with Benjamini-Hochberg false discovery rate (FDR) correction. Complementary sensitivity analyses assessed the robustness of the polar and non-polar findings. UPDRS-III and NMSS-K showed significant within-participant decreases ( p = 0.035 and p < 0.001, respectively), while MoCA-K showed a significant within-participant increase ( p = 0.011). Of 201 polar annotations, 156 passed analytical-quality filters; 16 features met the combined criteria of analytical quality, FDR support in both complete-pair and all-profile analyses, concordant direction, and at least 1.2-fold change. The most coherent endogenous pattern included increased hypoxanthine, xanthine, inosine, L-tyrosine, 3-O-methyl-L-DOPA, and p-hydroxyphenyllactic acid. Of 189 non-polar annotations, 169 passed analytical-quality filters; 17 features were FDR-supported and 14 met the predefined primary-candidate criteria. Twelve of the 14 primary candidates also met the PQN candidate criteria, whereas the remaining two retained PQN FDR support but fell below the predefined fold-change threshold. Representative non-polar changes included increased CE 16:1 and several other cholesteryl esters and decreased sphingomyelin, phosphatidylcholine, and ether-phosphatidylcholine species. Repeated anodal tDCS was accompanied by short-term clinical and CSF metabolomic changes in this small within-subject cohort. The most consistently supported within-study molecular patterns involved purine-related and tyrosine/levodopa-related metabolites and scale-corrected differences among selected non-polar lipids. Because the study lacked a sham control, the findings cannot establish treatment efficacy or tDCS-specific causality. They provide feasibility data, candidate molecular domains, and preliminary variance and effect-size information for future randomized sham-controlled studies.

The authors' abstract, as published at the source. Journal of NeuroEngineering and Rehabilitation, 2026 · DOI ↗