Association of pre-clinical depressive symptoms and its plasma metabolomic signature with Parkinson’s disease: a community-based longitudinal study
Accumulating evidence has suggested that depression is associated with an increased risk of Parkinson’s disease (PD), yet whether pre-clinical depressive symptoms in PD development and the potential mediating effects of metabolomic profiles remain unclear. We aimed to examine the association between depressive symptoms and PD risk, evaluate effect modification by genetic susceptibility, and assess metabolomic mediation. This prospective study included 451,922 PD-free participants from the UK Biobank. Depressive symptoms were collected by a 2-item Patient Health Questionnaire (range, 0–6; ≥3 indicates possible depressive disorder) at baseline. Identification of PD was based on medical records. PD-related polygenic risk score (PRS PD ), which incorporates established genes for PD based on external GWAS meta-analyses, was tertiled as low, moderate, and high. Baseline plasma metabolites were quantified via NMR spectroscopy from blood samples. Data were analyzed using Cox regression. Over a median follow-up of 14.6 years, 3108 (0.7%) participants developed PD. Higher depressive symptom severity showed linear association with PD risk (hazard ratio [HR]: 1.11, 95% confidence interval: 1.07–1.15). Participants with possible depressive disorder had 36% higher PD risk (HR: 1.36, 1.15–1.61). Those with both possible depressive disorder and high genetic risk had the highest PD risk (HR 2.15, 1.74–2.65; P for interaction = 0.028). Metabolomic signature, incorporating fatty acids and lipoproteins, accounted for 15% (model-based proportion) of the depressive symptoms-PD association. Pre-clinical depressive symptoms are associated with a moderately increased risk of PD, particularly among people with a high genetic susceptibility to PD. Metabolomic profiles account for a significant portion of this relationship.