Differential effects of severe vs mild GBA mutations on Parkinson disease

Abstract

Objective: To better define the genotype-phenotype correlations between the type of GBA (glucosidase, beta, acid) mutation, severe or mild, and the risk and age at onset (AAO), and potential mechanism of Parkinson disease (PD).

Methods: We analyzed 1,000 patients of Ashkenazi-Jewish descent with PD for 7 founder GBA mutations, and conducted a meta-analysis of risk and AAO according to GBA genotype (severe or mild mutation). The meta-analysis included 11,453 patients with PD and 14,565 controls from worldwide populations. The statistical analysis was done with and without continuity correction (constant or empirical), considering biases that could potentially affect the results.

Results: Among Ashkenazi-Jewish patients with PD, the odds ratios for PD were 2.2 and 10.3 for mild and severe GBA mutation carriers, respectively. The observed frequency of severe GBA mutation carriers among patients with PD was more than 4-fold than expected (4.4% vs 0.9%, respectively, p < 0.0001, Fisher exact test). In the different models of the meta-analysis, the odds ratios for PD ranged between 2.84 and 4.94 for mild GBA mutation carriers and 9.92 and 21.29 for severe GBA mutation carriers (p < 1 × 10(-6) for all analyses). Pooled analysis demonstrated AAO of 53.1 (±11.2) and 58.1 (±10.6) years for severe and mild GBA mutation carriers, respectively (p = 4.3 × 10(-5)).

Conclusions: These data demonstrate that mild and severe heterozygous GBA mutations differentially affect the risk and the AAO of PD. Our results have important implications for genetic counseling and clinical follow-up.

Figure 1. Meta-analysis of severe and mild GBA mutations and Parkinson disease risk

(A) Forest plot of studies with data on mild GBA mutations, using an empirical continuity correction (see methods) for studies with zero cases with mild GBA mutations. The analysis included data from 31 studies with a total of 11,453 cases and 14,565 controls. The p value for heterogeneity was 0.62. (B) Forest plot of studies with data on severe GBA mutations, using an empirical continuity correction (see methods) for studies with zero cases with severe GBA mutations. The analysis included data from 31 studies with a total of 11,453 cases and 14,565 controls. The p value for heterogeneity was 0.94. FE = fixed effect; Ref = reference number.