Identifying Genetic Markers Associated with the Progression of Cognitive Decline in Parkinson's Disease: A Call Out for Replication.

Abstract

Among patients with Parkinson's disease (PD), PD dementia (PDD) is among the greatest detriments to quality of life, and it is a hallmark of the progression of PD, making patient follow-up essential over time. To date, there have been three large, longitudinal, genome-wide studies for genetic determinants of dementia: Iwaki et al.1 (n = 4093, 12 cohorts), Tan et al.2 (n = 3364, 3 cohorts), and most recently, Liu et al.3 These authors have described longitudinal genome-wide survival studies in the search for genetic variants that influence the progression of PD to dementia endpoints in 3821 patients of European ancestry from 15 different cohorts in a span of 31 years. The most relevant outcome of this work is the discovery of a potential novel genome-wide signal within the RIMS2 locus, putatively associated with PDD. The RIMS2 variant was highlighted as a stronger predictor of PDD than GBA and APOE, which, surprisingly, did not reach genome-wide significance. The role of RIMS2 on PDD warrants further study because of the fact that: (1) its rare allele frequency could limit imputation accuracy4 and statistical power, and (2) it is not reported to be in linkage disequilibrium with any coding variant or act as a quantitative trait locus of expression or splicing of any gene. Because heterogeneity values across the 15 cohorts under study were not specified and authors report some genomic inflation (used to correct the test statistics downward), analyzing the contribution of RIMS2 to cognitive progression for each cohort is essential to ascertain that inflation is due to polygenicity (multigenic inheritance) and not population substructure (differences in allele frequencies between cases and controls as a result of systematic differences in ancestry) or other confounders. Two other suggestive association signals were reported in the TMEM108 and WWOX loci. Previous genome-wide association studies aimed at exploring cognitive progression in PD did not nominate any of these three loci.1, 2 In this study,3 the reported effect estimate for RIMS2 is massive, with a relatively small sample size for low minor allele frequency of the rare variant of interest. Previous studies should have detected this association if the effect estimate generalized across datasets. Moreover, cognitive polygenic hazard scores revealed that 18% of investigated cases carried at least one of the three nominated prognosis locus variants. However, the best model to predict whether a patient will experience development of dementia within 10 years from disease onset also included GBA and APOE loci. The authors also performed a polygenic risk score including 90 PD susceptibility loci associated with dementia prognosis, but no significant association was detected, indicating that PD progression may be controlled by separate genetic mechanisms and influenced by specific not yet discovered loci.5 This study3 represents a potential yet debatable step forward in our understanding of genetic markers associated with PD progression to dementia. Further robust replication in additional cohorts of diverse ancestry populations is imperative to confidently link this locus to PDD. This work was carried out with the support and guidance of the ‘GP2 Trainee Network’ which is part of the Global Parkinson's Genetics Program and funded by the Aligning Science Across Parkinson's (ASAP) initiative. Both authors contributed equally to the initial manuscript preparation, manuscript editing, and commentary. Data sharing is not applicable to this article as no new data were created or analyzed in this study.