Effects of GBA1 Variants in Patients With Parkinson's Disease and Levodopa-Carbidopa Intestinal Gel: A Nation-Wide, Multicenter, Longitudinal, "Real-World" Study. The EPIC Study

Abstract

Background: The outcome of levodopa/carbidopa intestinal gel (LCIG) in Parkinson's disease carriers of GBA1 mutations (GBA-PD) remains uncertain.

Objective: To evaluate the safety and efficacy of LCIG in a large PD cohort, focusing on GBA1 variants.

Methods: This multicenter, retrospective, longitudinal "real-world" study included consecutive patients with advanced PD treated with LCIG at 31 Italian centers; data were collected at baseline, 1-, 5-year, and last-available follow-up.

Results: Data from 512 PD patients (59% male, mean age and disease duration at LCIG initiation 67.0 ± 8.0 and 12.9 ± 5.0 years, respectively) were analyzed. GBA1 genotyping was available for 306 patients (60%), of whom 40 (13%) had GBA1 mutations or risk variants. Mean follow-up on LCIG was 3.9 ± 2.9 years; 5-year follow-up data were available for 159 subjects. At baseline, GBA-PD had a younger age, shorter PD duration, worse cognition, and more hallucinations than noncarriers. At 1- and 5-year follow-up, LCIG improved motor and non-motor symptoms, OFF-time, and dyskinesias in the entire population. In GBA-PD, MDS-UPDRS parts I, II, and III scores did not change, while part IV score improved significantly less than in noncarriers; cognition and orthostatic hypotension symptoms worsened more rapidly. Multivariate analysis of predictors for adverse events and LCIG discontinuation found no significant contribution from GBA1 mutation status.

Conclusions: GBA1 status does not increase the risk of adverse events or LCIG discontinuation. LCIG is a safe option for advanced GBA-PD, even in patients with cognitive impairment at baseline. However, GBA-PD experiences lower efficacy on motor disability and complications and faster cognitive decline than noncarriers.

Keywords: GBA1; LCIG; Parkinson's disease; levodopa–carbidopa intestinal gel; real world.

FIGURE 1

Study flowchart.

FIGURE 2

Adjusted cumulative incidence for LCIG discontinuation. We used Kaplan–Meier method, reporting GBA‐PD vs. PD noncarriers (panel A) and GBA‐PD according to mutations status vs. PD noncarriers (panel B).

FIGURE 3

Unadjusted values of selected motor and non‐motor symptoms in GBA‐PD and PD noncarriers. Prevalence of cognitive impairment (panel A); mean and standard deviation of MDS‐UPDRS parts I and II scores (panel B) and part IV subscores related to dyskinesias and OFF‐time (panel C) and prevalence of FOG and recurrent falls (panel D).