Understanding monogenic Parkinson’s disease at a global scale

Introduction

Although monogenic forms of Parkinson’s disease (PD) have been described worldwide, about three-quarters of all PD genetics studies were performed in individuals of European/White ancestry ¹, formally referred to as Caucasian, thereby severely limiting our current insight into genotype-phenotype relationships at a global, multi-ancestry scale and contributing further to healthcare and research disparities ². The first systematic approach to embrace monogenic PD worldwide, The Michael J. Fox Foundation Global Monogenic PD (MJFF GMPD) project, was built on the MDSGene Database (www.mdsgene.org) that compiles published genotype-phenotype relationships for monogenic PD ^{1, 3-5} and other movement disorders. Corresponding authors of PD-related articles included in MDSGene were contacted and asked to provide additional information on previously reported or newly identified individuals using an electronic case report form ^{1, 6}. Data were collected on almost 4,000 individuals from 89 centers in 41 countries, including affected and unaffected carriers of pathogenic variants in genes implicated in monogenic PD ¹. However, as the focus of this project was to identify established PD research centers with access to and funding for genetic testing, most of the provided information was on individuals of European/White ancestry. The Global Parkinson’s Genetics Program (GP2; https://gp2.org) complements this approach by including all centers with unsolved but suspected monogenic individuals, specifically addressing those that did not have access to genetic testing or that have not participated in PD research before ⁷. The aim is to extend the global network and include centers from formerly underrepresented countries and individuals with diverse ancestry ⁸. The Monogenic Network of GP2 (GP2 MN) focuses on identifying novel monogenic causes of PD but also characterizing genotype-phenotype relations of known monogenic PD at a global scale ⁹. Therefore, the present study builds on the efforts of The MJFF GMPD project, incorporating it into GP2’s MN, and specifically extending the diversity of the current global map of monogenic PD ⁸.

Methods

Recruitment of collaborating centers for The MJFF GMPD project was based on a systematic online approach to collect individual-level data on individuals with pathogenic variants in known PD genes (SNCA, LRRK2, VPS35, PRKN, PINK1, PARK7) and pathogenic or risk variants in GBA1 ¹. This included clinical and demographic data to analyze genotype–phenotype relationships ¹. In contrast, recruitment for the GP2 MN was more inclusive, and initially focused on individuals with an unsolved but suspected monogenic form of PD based on an early age at onset (AAO; ≤50 years of age) and/or a positive family history ⁹. Potential collaborators were additionally identified through personal contacts, participation in PD consortia, and advertising GP2 at congresses. Collaborators from both efforts were divided into three groups based on their involvement in both projects (Group 1), or only in The MJFF GMPD (Group 2) or GP2’s MN (Group 3).

Participants of The MJFF GMPD project received customized emails including relevant information about GP2 and an online survey gauging their interest in participating and their availability of DNA samples. Collaborators interested in joining GP2 were given standardized instructions for the onboarding process to GP2 MN ⁹. Further, all collaborators previously involved in The MJFF GMPD project (Groups 1 and 2) were asked for their permission to merge both efforts and transfer existing data and samples to GP2’s MN. Only centers with GP2 compliance approval until November 2023 are included in this manuscript. Compliance approval is based on an eligible consent form enabling international sample and data sharing and approval by the local Ethics Committee ⁹.

The assignment to underrepresented countries was based on the World Bank’s classification of income status (https://data.worldbank.org/country/XO), classifying countries with low and middle income as underrepresented.

Regardless of available genetic pre-screening, all samples from individuals with eligible clinical data that passed quality control (QC) underwent genotyping with the NeuroBooster Array (NBA), including 1.9 million markers from the Illumina Global Diversity Array and more than 95,000 neurological disease-oriented and population-specific variants, including several hundred known variants in PD-related genes ¹⁰. Genotyping did not include the analysis of copy number variants (CNV). Detected variants underwent a pathogenicity evaluation based on the criteria of the American College of Medical Genomics (ACMG) ¹¹ using Franklin (https://franklin.genoox.com/clinical-db/home), the human genomic variant search engine VarSome ¹², and MDSGene criteria (https://www.mdsgene.org/methods). Only (likely) pathogenic or known PD risk variants were included. Additionally, variants in LRRK2 were evaluated based on their impact on the LRRK2 kinase activity ¹³, while the GBA1-PD Browser (https://pdgenetics.shinyapps.io/gba1browser/) ¹⁴ was used to evaluate identified GBA1 variants. Efforts to perform short-read whole-genome sequencing including CNV analyses for this cohort are ongoing.

Percentages in the results section are given as valid percentages.

Data Sharing

GP2 partnered with the online cloud computing platform Accelerating Medicines Partnership - Parkinson’s Disease (AMP PD; https://amp-pd.org) to share data generated by GP2. Anonymized data can be shared upon request and qualified researchers are encouraged to apply for direct access to the data through AMP PD.

Results

To date, 100 centers from 46 countries (Figure 1) are included in the GP2 MN. All of the 89 centers previously identified through the MJFF GMPD project ¹ were invited to join the GP2 as part of the ‘merger’, and 38 (42.7 %) of these are now part of the GP2 MN. Reasons for non-participation were diverse, e.g., including no response to the invitation to participate, delay or failure of the onboarding process due to bureaucratic hurdles (e.g., consent forms not meeting the legal requirements of GP2), or inability to provide DNA samples in addition to clinical data.

• Download figure
• Open in new tab

Figure 1: World map of centers participating in the GP2 MN and the MJFF GMPD project

Colored are the countries from which one or more centers participated in the MJFF GMPD project and GP2 MN (blue) or only in GP2 MN (green). Colored in purple are new centers in underrepresented countries that were recruited based on the new approach of GP2 MN.

Blue: Argentina, Australia, Austria, Belgium, Canada, Denmark, France, Germany, Greece, Hungary, Israel, Italy, Japan, Luxembourg, Malaysia, Netherlands, Norway, Poland, Portugal, Saudi Arabia, Singapore, Slovakia, South Africa, South Korea, Spain, Sweden, Taiwan, Thailand, Turkey, UK, USA

Green: Chile, Czechia, New Zealand, Switzerland

Purple: Bangladesh, Colombia, Indonesia, Iran, Mongolia, Nepal, Nigeria, Pakistan, Philippines, Sri Lanka, Tunisia

Nineteen (19%) centers from 15 countries included in the GP2 MN were from underrepresented countries (Figure 1), eleven countries of which were not previously part of The MJFF GMPD project.

Thirty-four centers and the PDGene consortium (https://www.parkinson.org/advancing-research/our-research/pdgeneration) already submitted samples and clinical data to the GP2 MN for further analysis. For 66 centers, the onboarding process is still ongoing. To date, 5,374 DNA or blood samples have been sent to the coordinating GP2 site in Luebeck, Germany, including 468 samples with pending clinical data. The vast majority (n=4,631) of these 4,906 ready-to-analyze samples are from affected individuals with PD/parkinsonism (Table 1), and only a subset (n=275) is from unaffected relatives or unaffected mutation carriers. The median AAO was 49 (Interquartile Range [IQR]: 41-61) years, and 55.7% had an AAO ≤ 50 years (17.5% missing data).

Genotyping of the GP2 MN cohort with the NBA identified (likely) pathogenic single nucleotide variants (SNVs) in known PD genes (including GBA1) in 131 (4.9%) and PD risk variants in GBA1 and LRRK2 (LRRK2 risk variants previously associated with PD in Asian populations) in 272 (10.3%) out of 2,648 individuals tested thus far; the genetic details are listed in Table 2. Heterozygous variants were identified in GBA1 (n=84), LRRK2 (n=12), and SNCA (n=14), and one carrier of a pathogenic variant in VPS35 was found. Sixteen carriers of homozygous variants in PINK1 and 4 PRKN homozygous variant carriers were identified but no carriers of biallelic variants in PARK7 (DJ1). Further, genotyping identified variants of uncertain significance (VUS) in known PD genes in 33 individuals (Table 2).

Discussion

To date, 100 centers from 46 countries are included in the GP2 MN, 38% of which were previously recruited through The MJFF GMPD project ¹. An important reason for MJFF GMPD centers not to participate in GP MN was that the principal investigator was no longer active, demonstrating the importance of dedicated researchers to drive projects like GP2’s MN. Both projects followed different approaches when recruiting collaborators. The ‘merger’ of both efforts with complementary approaches enables the comprehensive identification of centers working in the PD genetics field, including not only centers identifiable through publications on PD genetics (MJFF GMPD) but also less visible centers (GP2 MN). GP2’s focus on underrepresented regions brought in groups that were not detectable through MJFF GMPD’s publication-based approach, but merging both efforts facilitates the build-up of an inclusive study population with a diverse ancestral background.

Sixty-two centers not previously participating in The MJFF GMPD project were recruited for GP2’s MN, including 16 centers from underrepresented countries. However, the research infrastructure in these countries often posed a challenge to the onboarding process and required intensified efforts and support. Our goal is to overcome these bureaucratic and logistical hurdles to enable global representation of monogenic PD. One way of doing this will be using and supporting established PD consortia in these areas, e.g., LARGE-PD (https://large-pd.org/).

So far, ∼5,400 samples have been included in the GP2 MN and ∼3,900 individuals in The MJFF GMPD project. In The MJFF GMPD project, 3,185 affected and 703 unaffected mutation carriers were included ¹. Genetic testing of the GP2 MN cohort with the NBA has thus far identified (likely) pathogenic variants in 121 (including GBA1 variant carriers), PD risk variants in LRRK2 and GBA1 in 252, and VUS in 33 out of 2,648 tested individuals. The low frequency of (likely) pathogenic variants compared to previous studies (Westenberger et al., in press) is likely due to the initial selection criteria of GP2’s MN, which prioritized individuals with a negative genetic pre-screening since a major aim of the MN is to discover new PD-related genes ⁹. While GP2’s MN initially focused on unsolved cases, The MJFF GMPD project collected data and samples from individuals with known monogenic causes. Integrating The MJFF GMPD project into GP2 has widened the scope of GP2’s MN. Merging both projects will not only allow creating a sufficient infrastructure to identify the clinical spectrum of monogenic PD across diverse ancestries but also enable investigating factors modifying monogenic PD by identifying a large number of mutation carriers. Together, the combined effort now allows for a more comprehensive, cohesive, and collaborative approach to understanding monogenic PD. Identifying patients with monogenic PD provides the basis for recruiting individuals with genetic PD/parkinsonism for future gene-specific clinical trials which is a dedicated aim of one of the GP2’s special interest groups.

Our efforts of combining two large genetic PD research initiatives underline that the future of research should be based on team science approaches to combine data into even larger, standardized data sets to generate meaningful and globally relevant results ⁶. A collaborative mindset is indispensable to sharing expertise internationally and facilitating new research opportunities which will eventually enable the development of personalized therapies.

Funding Sources

This research was supported in part by the Intramural Research Program of the NIH, National Institute on Aging (NIA), National Institutes of Health, Department of Health and Human Services; project number ZO1 AG000534, as well as the National Institute of Neurological Disorders and Stroke.

GP2 is funded by the Aligning Science Across Parkinson’s (ASAP) initiative and implemented by The Michael J. Fox Foundation for Parkinson’s Research (https://gp2.org).