GCH1 genetic variation as a prognostic factor in Parkinson disease across populations

Background: Pathogenic variants in GCH1 have been associated with Parkinson's disease (PD), but the clinical phenotype and longitudinal disease course of GCH1-associated PD remain incompletely characterized. Objectives: To characterize the genetic spectrum, clinical phenotype, and longitudinal progression of GCH1-associated PD across multiple populations. Methods: Whole-genome sequencing (WGS) and clinical exome sequencing (CES) data from the Global Parkinson Genetics Program (GP2) were analyzed together with unpublished and published GCH1-associated PD patients. Variant pathogenicity was classified according to ACMG criteria. Demographic, clinical, and longitudinal features were compared between GCH1 P/LP variant carriers and non-carrier PD patients; individuals with known pathogenic variants in PD-associated genes were excluded from both groups. Results: In the GP2 cohort (PD, n=22,825; controls, n=4,453), 16 pathogenic or likely pathogenic (P/LP) GCH1 variants were identified in 58 individuals, including 54 PD patients, one control, and three individuals with other neurodegenerative phenotypes (two with progressive supranuclear palsy and one with dementia with Lewy bodies). In the pooled-ancestry WGS analysis, GCH1 P/LP variants were enriched in PD patients versus controls (0.267% vs 0.023%; OR=11.854; 95% CI=1.620-86.699; p=0.0006). Variant frequencies in PD patients ranged from 0.121% to 0.714% across ancestries. In the CES cohort, P/LP variants were identified in 0.201% of PD patients. After integrating GP2 with additional unpublished and published datasets, 119 GCH1-associated PD patients were analyzed. Compared with noncarriers, carriers of P/LP GCH1 variants had an earlier age at onset (mean [SD], 53.7 [14.8] vs 59.2 [11.7] years; P < .001), lower levodopa equivalent daily dose requirements (mean [SD , 467.5 [331.7] vs 680.3 [466.4] mg/d; P < .001), and a higher frequency of a family history of Parkinson disease (47.6% vs 19.9%; P < .001). Adjusted Cox models showed significant delayed progression to motor fluctuations (HR=0.32, 95% CI=0.16-0.62) and levodopa-induced dyskinesias (HR=0.51, 95% CI=0.30-0.87). Conclusion: GCH1 pathogenic variants were associated with a clinically distinct phenotype characterized by earlier disease onset and slower progression of motor complications. These findings suggest that GCH1 genetic variants may serve as genetic biomarkers for patient stratification and prognosis in PD.