<scp>GP2</scp>: The Global Parkinson's Genetics Program

Abstract

Fundamentally, the identification of causes and contributors of disease represents the first step in an etiology-based understanding of disease. This, in turn, is a required step in the development of therapeutics targeting the underlying disease process. Parkinson's disease (PD) is believed to be a complex disorder, with disease liability driven in part by genetics. Current heritability estimates suggest that common genetic variability contributes ~22% of the disease liability in an average patient.1 Although this is believed to be an underestimate, it is also likely that there are nongenetic influences, such as environmental exposure or stochastic events. Genetics, however, has shown itself to be a robust, tractable, and reliable method to understand disease biology. Genetic understanding serves as a foundation for succeeding functional studies and as a central component of efforts to predict disease risk, onset, and progression and to understand disease mechanisms in individual patients. Without a reliable and complete foundation of genetic understanding, we limit our ability to develop and deploy treatments. A large number of risk loci and causative mutations for PD have been identified; however, it is clear that the majority of genetic risk remains to be found.1, 2 Although much can be done with existing knowledge, moving forward now to expand our genetic understanding will be the foundation that will support the development of a complete view of this network, providing an array of potential therapeutic opportunities. Increasing genetic information can only serve to improve our efforts to treat disease. Notably, our understanding of the genetic basis of PD has thus far largely been centered on research in individuals of northern European ancestry. Although some genetic discoveries have been made outside these populations, this work is the exception rather than the norm and generally focuses on the identification of rare mutations; little has been done in the identification of more common genes or genetic risk discovery.3, 4 Thus, we do not know if our current understanding is generalizable to the rest of the world and how the basis of disease varies across populations. Although it is tempting to argue that the genetic basis for PD will generally be the same across populations, we know that differences in genetic risk exist, and furthermore, there is evidence to suggest that genetic forms of disease can present differently across populations.3, 5-8 This fundamental limitation of current research creates an inequitable situation for patients. To facilitate the rapid expansion of our understanding of the genetic architecture of PD, both in terms of the depth and global context of this knowledge, we have created the Global Parkinson's Genetics Program (GP2; www.gp2.org). GP2 is the first supported resource project of the Aligning Science Across Parkinson's (ASAP) initiative, an audacious effort supporting PD research.9 GP2 is geared toward creating a worldwide collaborative effort that will first dramatically accelerate the identification of genetic contributors to disease and second establish a network of researchers who can best leverage this understanding to research, diagnose, and treat PD worldwide. Here we describe our mission, the path we have proposed to achieve this, and the core principles of data democratization, transparency, and diversity. The mission of GP2 is to drive transformational progress in our understanding of the genetic architecture of PD and to serve as a useful and actionable resource for the research and therapeutic development community. To fully realize this mission, GP2 will need to engage and mobilize a worldwide community of researchers and participants, generate and analyze genetic data on an extremely large scale, create an infrastructure that removes obstacles to data access, and make data and results accessible and useful to the broader community (Fig. 1). Broadly, there are 2 scientific arms to GP2, one centered in genetically complex, typical PD and the other in monogenic disease. Over the initial 5-year span of the GP2 program, our path will lead us to a dramatic increase in the number of known genes, disease-causing mutations, and risk loci for both rare monogenic and typical complex PD. Furthermore, this work will be extended, for the first time at scale, to underrepresented populations from around the world. We will generate dense genetic data in more than 150,000 participants, using a genotyping array specifically designed for this purpose. We will also generate whole-genome sequence data from more than 10,000 individuals to determine the genetic cause in as yet unsolved monogenic cases and to generate much needed reference data sets. Furthermore, we will use long-read DNA sequencing to support the analysis of structural and repeat variability that is relatively resistant to interrogation using traditional genome sequencing methods. For the most part, GP2 will use existing or ongoing patient and cohort collections and will work with established PD consortia; however, in some instances GP2 will support the collection of additional patients, particularly from underserved communities. To ensure a functional and efficient structure, we created a series of working groups and hubs that center on achieving specific aims and priorities within GP2. Although these groups have clear aims and deliverables, they function as a continuum with shared members (Fig. 2). The role of this group is to explore the genetic basis of typical, apparently sporadic PD. The foundation of this work will be the genotyping of 150,000 participants using an array designed by us specifically for this purpose. The Neuro Booster Array is centered on the backbone of the global diversity array (GDA; 1.8+ million variants; https://www.illumina.com/products/by-type/microarray-kits/infinium-global-diversity.html) but also includes more than 95,000 custom content variants that include neurological disease-oriented content and population-specific boosters (article in preparation; https://github.com/GP2code/Neuro_Booster_Array). Broadly, we expect to generate data on ~100,000 northern European ancestry individuals and more than 50,000 subjects from underrepresented populations from around the world. We have established collaborations to collect and assess cases of Black American, East Asian, African, Indian, Caribbean and Central/South American provenance. Our estimates of sample numbers are driven by a combination of resource availability and pragmatism. However, we hope to significantly increase the collection and genetic analysis of patients from underrepresented groups well beyond these numbers, so 50,000 can be considered a lower bound. This will depend on the availability of additional subjects and how these cohorts are able to align with the GP2 general principles around data access and use. It is also worth noting that the majority of GP2 genotyping will be centered on persons with PD because abundant data on controls already exist; however, some of our genetic resources will be centered on generating genetic data in controls, primarily for populations in which no such data exist or for whom there are valuable PD related data in these particular control samples (eg, biomarker data). Data from underrepresented populations will be generated from a variety of sources. GP2 has already initiated partnerships with academic research centers in the United States to improve representation of Black Americans within the project. Samples and data are being collected from East Asia by the International Parkinson's Disease Genomics Consortium (IPDGC) East Asia group, with efforts ongoing in Taiwan, Japan, South Korea, and China. Likewise, IPDGC Africa has initiated collaborations across Africa beginning with patients from Nigeria, Egypt, Ethiopia, Ghana, Mali, Tanzania, Senegal, South Africa, Sudan, and Zambia. The Genetic Epidemiology of Parkinson's disease (GEoPD) Consortium has developed collaborations across underrepresented populations from North and sub-Saharan Africa, Australia, and Asia. The Luxembourgish-German Indian Alliance on Neurodegenerative Diseases and Therapeutics (Lux-GIANT) has formed a collaborative group to investigate PD patients across India.10 Last, the Latin American Research Consortium on the Genetics of Parkinson's Disease (LARGE-PD) group is a fully active collaborative group collecting and investigating patients from Argentina, Brazil, Chile, Costa Rica, Colombia, Ecuador, Honduras, Mexico, Peru, Puerto Rico, Uruguay, and the West Indies.11 As GP2 continues, there will be room to expand to other countries and populations underserved in our current research. These data, collectively, will afford the opportunity to rapidly detect novel genetic risk for PD. Critically, the availability of similar data across ancestral groups will allow an assessment of the varied genetic contribution in different ancestries, including the identification of population-specific loci, an understanding of the differences in the heritable component of disease between groups, and the generation of population-specific genetic risk profiles. Notably, these data collectively provide the opportunity to refine association signals, with transethnic fine-mapping. A crucial step will be integration of clinical phenotype data. We know that there are diverse outcomes of PD including the rate of motor and cognitive deterioration and medication side effects such as levodopa-induced dyskinesias.12 We believe that these will be in part genetically determined13, 14 and that understanding the associated genes and pathways will lead to new biological insights and importantly personalized treatments. A barrier to this is the harmonization of data, which is a major goal of the cohort integration group. Although modern genetic methods provide tools for the rapid discovery of rare causal or high-risk mutations, several barriers exist that limit the efficiency of identifying novel causes of disease. First, multiplex families are overall rare and dispersed. Second, the generation and analysis of genetic data are specialized and expensive. Third, existing genetic data are not harmonized and often not available for sharing. And, fourth, penetrance is reduced in dominantly inherited forms and, although high in recessive forms, age dependent. The latter results in frequent absence of the most prominent red flag for the occurrence of a monogenic form of PD, that is, positive family history, so that a significant proportion of patients miss out on genetic testing and research because they are not deemed good candidates. Collectively, this means that finding segregating mutations, or mutations in the same putative novel gene, is difficult; this, in turn, has resulted in the publication of a growing number of potentially disease-associated mutations that are preliminary and can be quite misleading or confusing to the field. Importantly, and unlike findings from complex genetic studies of PD, these putative new monogenic causes are often readily implemented in PD gene panels for diagnostic testing by genetic testing companies, posing an additional challenge to patients, unaffected carriers, genetic counselors, and physicians in terms of interpretation of the ensuing test results. The monogenic disease arm of GP2 aims to address these obstacles and thereby create an efficient infrastructure to accelerate the identification of novel genetic causes of apparently monogenic PD. Leveraging the above-described global network of researchers contributing patient samples to Neuro Booster Array genotyping and including already-existing resources from the monogenic field, such as IPDGC,15 the GEoPD, and the Michael J. Fox Foundation Global Genetic Parkinson's Disease Study Group,16 the monogenic arm will collect >5000 patients and families in whom a monogenic cause may be suspected. Particular emphasis will be placed on families from underrepresented populations. All currently known PD genes have been found in various populations around the globe, however, some occur at highly variable and population-specific frequencies, the most striking examples being the p.G2019S mutation in the LRRK2 gene17 and GBA mutations including p.N370S18 and p.K198E.19 In addition, it is conceivable that population-specific hereditary forms of PD may exist, as exemplified by X-linked dystonia-parkinsonism, a condition exclusively present in patients of Filipino ancestry,20 for which the underlying genetic cause has been identified as well as genetic age-at-onset modifiers.21-23 This condition has served as an important model for basal ganglia disease.24, 25 The monogenic arm will collect families, singleton cases, and patient-parent trios and prioritize these for whole genome sequency or long-read DNA sequencing based on a number of different criteria: family history and availability of samples from several affected (and unaffected) family members, age at onset, ethnicity (with a focus on underrepresented populations), and level of available genetic prescreening. Importantly, all patients enrolled in the monogenic arm of the project will also undergo Neuro Booster Array genotyping, which, based on its PD-related custom content, will result in the identification of a sizable number of patients with mutations in known PD genes. Assuming that an average of ~10%–15% of all PD patients carry a mutation in a known PD gene (3%–5%) or a high-risk variant in GBA (8%–10%),26 we estimate a total of ~15,000 monogenic or high-risk variant carriers to be detected in the total GP2 sample set. It is at this important interface, where the complex genetic and monogenic arms will interact most closely: although the complex genetics arm will identify carriers of known PD-causing mutations that can then be enrolled in various additional investigations, such as genetic studies of age at onset, the monogenic will all of patients to the monogenic to also patient and control cohorts the complex genetics This will create not only for the discovery of novel genetic causes of PD but also for a understanding of the known genetic forms of PD. Over the the genetics has made toward data available to the research community. However, barriers Data are highly across there is often for data access data and data use can be In addition, the to results or analyze data can be and the of data analysis can be A of GP2 is the generation of data and that can be readily and by the research community. To achieve this, we are several to data in one and provide all analysis with the context in a We will as much information as and in the an Although the of data be a a large number of only and of these can be shared data will be however, we will only a to access these the data access process. step in the for cohorts is the and by the and This group with and to for data sharing. Our ability to with these our of data and are if we to forward with the cohort GP2. We will a model in which researchers analyze data in rather than data to This several It means that we can by control and and these as a common It means we can create collaborative and and resources by working across a shared In GP2, we can be both and in our which can be shared with the research and to This the data, and individuals with the but the resources to access and analyze these data. The model of to the data, rather than is one being or by major and global genetics and such as the International Disease Alliance at and the To achieve these our current model is centered around with other is a developed and by the specifically for research. access and analysis of data, the need for data allow researchers to whole such as sequence data sample and across populations and, with an and Thus, by and and This researchers to use and and It is not to make data available to the PD research community. A part of GP2 is to develop that will and researchers around the world so that they may using GP2 data. We that a significant barrier to to work with genetic data is These to establish in and a of new Our of has been focus by the and we have our efforts to This has the of these more readily accessible to from around the world. We also to support clinical in that most from Our hope is that these resources will increase the efficiency of and in GP2, particularly for where such is to For individuals who drive and we will support a of individual in the form of as at centers of and We are to the of and researchers who have been underrepresented so far in research, and we will research and partnerships a network of GP2 We believe that a component of research to underserved and underrepresented populations is of studies by an active and role in the global and and other infrastructure support for researchers to on As we to support the development of and all data are to contributing on of harmonization and a shared the individual can access the and control that have been to generate these data. can also use the from to collaborative analysis in the barrier of providing GP2 also aims to provide several of support beyond We have a data analysis core that not only serves to but also to support members in the form of in and by providing a data The work for GP2 will provide data and fundamental insights for PD research. We believe that this work of the path from biological understanding to therapeutic the work from GP2 will lead to a significant increase in the proportion of the known heritable component of identifying novel risk mutations, and This will improve the genetic component of risk and will provide us with to determine genetic of the common disease are on the expansion of efforts underrepresented populations will also lead to the identification of new loci that are or of in the northern European These data will allow us to determine heritability across ancestral groups and to risk profiles. It will provide the genetic basis of differences in disease and as well as which groups may be particularly to a therapeutic these data allow transethnic to be to the in which functional risk and to identify the functional risk, GP2 the opportunity to the genetic basis of variability in disease. at onset, and the of and are to genetic more this work the for the of individual risk for and of disease. complex genetic with data in monogenic cases also the identification of genetic of disease Furthermore, these data suggest that such work may effects of the same risk loci in primarily monogenic and complex a potential functional basis for differences or at different between patients. This is of to our in who are for potential of the disease process. are a variety of by the GP2 that are not part of its most and there is significant potential for additional work made the generation of GP2 data. Our with the the use of genetics in biomarker development genetics in disease an of the of ancestry on and the use of genetics to improve biomarker by for genetic to disease. studies have collected additional data that can be with genetics. The availability of and data an opportunity to understand the genetic basis of in these and may insights on disease data from largely methods such as and with genetics the opportunity to loci in both and an that has been to understand and improve biomarker The of GP2 in collecting samples from underrepresented populations also the opportunity to studies in these the most level this may include of patients from countries or ancestral groups in which little has been It can also include work on and on disease and the use of genetics. We believe these and in the of and beyond can be for research. Our in GP2 is not to create an that of but rather to create a around which such studies can be Our hope is that individual within the GP2 network will new of and form and collaborations with other to explore these GP2 and our data and can serve as an for such The identification of novel gene mutations and loci will the underlying basis and of disease. these results with other data can provide evidence on the context of genetic risk, a step in the of Likewise, the more genetic information we the more complete is the we can of the of disease. has supported a series of within the Research with a major focus on work to determine the of genetics. GP2 will be a resource for these In the relatively time that we have been able to identify genetic risk at scale, the to understanding has been However, particularly in the of to the gene, and biological the context and on can be in traditional and to disease pathways and of the path a fundamental understanding of the and depth of genetic in PD will our ability to understand the disease process. work can afford us insights the of of there are in the disease and the of these mechanisms across a typical PD Critically, if we or mechanisms in of patients, it will be important to patients with personalized This is, in an of the of testing LRRK2 in LRRK2 GP2 will improve of disease risk and This information will serve to facilitate clinical supporting in patients and outcomes based on of As there are a large number of potential and associated with the GP2 data and collaborative We are at the beginning of our but it is to the to on priorities and may as we The results of our current work the The efforts of the monogenic group are already providing a understanding of is to be we identify to monogenic are we to identifying the majority of these to determine which mutation carriers will and Likewise, data from the complex genetics group will address similar We will be able to understand we have biological in common risk loci the at which identified loci are to new disease an for association We will also have a good of our sequencing efforts in complex disease do we significant additional discoveries or using this method the is and it is now that it be particularly important to our work in underrepresented We know our current efforts in underrepresented groups a significant and a major step forward for the field. However, they are a It will be important to on this and to in research within and by underserved so that we can realize the potential of this Last, we also the genetic In genetics has a We have a clear path forward and on the to This is also for the to from genetic to and there may be some to GP2 some of these particularly around the of resources to make this However, of how we to these we believe that GP2 will serve as the on this and a foundation for the global PD research community. We are the progress we have made so far and to be this with our This research supported by the Aligning Science Across Parkinson's the Research on of of and project and the Michael J. Fox Foundation for Parkinson's of GP2 and made to this Global Parkinson's Genetics Program

Topics & Concepts