Abstract
BACKGROUND AND OBJECTIVES: Deep brain stimulation (DBS) is an established treatment for Parkinson disease (PD). In patients carrying GBA1 variants (GBA-PD), concerns persist that DBS may accelerate cognitive decline. This study investigated the potential additive effects of GBA1 genotype and DBS on long-term motor and nonmotor outcomes. METHODS: This multicenter retrospective, controlled, Italian study included 3 groups: DBS-treated PD patients either carrying or noncarrying GBA1 variants (DBS-nonGBA-PD and DBS-GBA-PD) and GBA-PD patients who fulfilled DBS criteria but eventually were not operated. As secondary aims, we assessed the clinical outcomes of DBS-GBA-PD stratified by GBA1 variant classes and by different DBS targets. Cognitive, motor, and other nonmotor features were collected at baseline and after 1, 3 and, when available, 5 years. Between-group comparisons used χ(2) and Kruskal-Wallis tests with Bonferroni correction. Longitudinal changes were analyzed with linear mixed-effects models. Subgroup analyses were performed by GBA1 variant class and DBS target. RESULTS: A total of 615 participants were included: 430 DBS-nonGBA-PD (age 57.4 ± 7.7 years, 32% female), 109 DBS-GBA-PD (age 53.5 ± 8.4 years, 38% female), and 76 nonDBS-GBA-PD (age 57.7 ± 8.1 years, 37% female). At baseline, groups were largely matched for clinical features. Longitudinally, both DBS groups showed marked motor improvement (dyskinesias, on-off phenomenon, and wearing-off, all p vs T0 < 0.001), a benefit which was absent in nonDBS-GBA-PD. At 5 years, dementia occurred more frequently in DBS-GBA-PD and nonDBS-GBA-PD compared with DBS-nonGBA-PD (25.5% vs 36.8% vs 10.8%, p < 0.001). Hallucinations and urinary problems increased in both GBA-PD groups than nonGBA-PD (p-between <0.001 and 0.02, respectively), regardless of DBS. No relevant differences emerged on stratification for variant classes or DBS targets, up to 3 years postsurgery. DISCUSSION: Despite its retrospective design, this study supports DBS as a valid therapeutic option for GBA-PD, providing prolonged benefits on motor symptoms and quality of life. The accelerated cognitive decline observed in GBA-PD, compared with non-mutated participants, was similarly present in both operated and non-operated groups, suggesting it is driven by the genotype rather than DBS itself. CLASSIFICATION OF EVIDENCE: This study provides Class III evidence that DBS does not worsen cognitive function in patients with GBA1-associated PD.