Deep phenotyping for precision medicine in Parkinson's disease

Abstract

A major challenge in medical genomics is to understand why individuals with the same disorder have different clinical symptoms and why those who carry the same mutation may be affected by different disorders. In every complex disorder, identifying the contribution of different genetic and non-genetic risk factors is a key obstacle to understanding disease mechanisms. Genetic studies rely on precise phenotypes and are unable to uncover the genetic contributions to a disorder when phenotypes are imprecise. To address this challenge, deeply phenotyped cohorts have been developed for which detailed, fine-grained data have been collected. These cohorts help us to investigate the underlying biological pathways and risk factors to identify treatment targets, and thus to advance precision medicine. The neurodegenerative disorder Parkinson's disease has a diverse phenotypical presentation and modest heritability, and its underlying disease mechanisms are still being debated. As such, considerable efforts have been made to develop deeply phenotyped cohorts for this disorder. Here, we focus on Parkinson's disease and explore how deep phenotyping can help address the challenges raised by genetic and phenotypic heterogeneity. We also discuss recent methods for data collection and computation, as well as methodological challenges that have to be overcome.

Keywords: Genetics; Phenotyping; Precision medicine.

Fig. 1.

Parkinson's disease (PD) is characterised by a high degree of heterogeneity. At diagnosis, >50% of dopaminergic neurons are already lost, and patients can show any combination of motor, neuropsychiatric and autonomic symptoms of differing severity. The blue area indicates the variability in the loss of dopaminergic neurons over time. RBD, rapid eye movement sleep behaviour disorder. See Glossary (Box 1) for descriptions of the symptoms.

Fig. 2.

GBA mutations: genetic heterogeneity in human versus disease models. GBA mutations are the most common genetic risk factor for Parkinson's disease (PD). However, GBA mutations are also found in different human disorders, such as Gaucher's disease (GD), dementia with Lewy bodies (DLB) and rapid eye movement sleep behaviour disorders (RBDs), and in healthy individuals. Different models of GBA mutations, including mouse and human induced pluripotent stem cell (iPSC)-derived neuron models, develop the same observable phenotypes. Together, this suggests that other genetic or non-genetic factors contribute to GBA-mutant PD in the human population.

Fig. 3.

The merit of deep phenotyping for precision medicine. In traditional clinical practice, the same treatment strategy is applied to anyone diagnosed with a disorder. This means that a diagnosis is made and a treatment given based on a predefined set of signs and symptoms. The outcome, including treatment effectiveness, can thus be varied due to heterogeneity within that disorder. Precision medicine uses fine-grained information gained through deep phenotyping and genetics to match the best treatment to an individual patient. In addition, it can aid in the monitoring and identification of at-risk individuals and enable preventative interventions.

Fig. 4.

Towards precision medicine by integrating multi-modal biomedical data. A number of the research objectives can be explored with genetic data and deep phenotyping. Deep phenotyping provides data on many different scales, such as environmental factors, lifestyle, multi-omics, diverse biological samples, imaging, behaviour, etc. Such complex data benefit from the advent of machine learning, such that large-scale, heterogeneous, multi-modal phenotypic and genetic data can be translated into meaningful information about risk, diagnosis, prognosis and stratification.

Fig. 5.

Clinical phenotype-driven versus biomarker-driven research. (A) Biomarkers are useful for identifying differences between clinical phenotypes and clinical subgroups, and in providing a differential diagnosis. (B) Biomarkers can also differentiate disease subtypes, which can then be associated with clinical phenotypes and behaviour. For example, in patients with pure synucleinopathy, we expect to only see PD-specific biomarkers (red), whereas in those with AD co-pathologies, we expect abnormalities in AD-specific biomarkers as well (blue). The distinction between DLB and PDD is defined by the '1-year rule': if the onset of dementia symptoms is within 1 year of parkinsonism, the disorder is called DLB; if parkinsonism is present for more than 1 year before the onset of dementia, the disorder is called PDD. AD, Alzheimer's disease; DLB, dementia with Lewy bodies; MCI, mild cognitive impairment; PD, Parkinson's disease; PD-D, Parkinson's disease dementia.

Fig. 6.

Overview of challenges and resulting opportunities. Deep phenotyping produces large amounts of data, which present various challenges in three domains: data storage, analysis and application. However, these challenges give us the opportunity to set global standards and, once the infrastructure is in place, to gain valuable novel insights into disorders that can guide us to precision medicine.

Fig. 7.

A model for precision medicine in diagnosing and treating PD. To evaluate how pharmacological interventions might reverse the early (pre)clinical symptoms of PD, the features of in vitro disease models, such as lysosomal dysfunction in fibroblasts with GBA mutations, need to be linked to the phenotypes of RBD/PD patients, such as their sleep and biomarker profiles. The UK Biobank population could also be profiled to identify the earliest features of RBD and thus help more at-risk patients. Cognitron is an artificial intelligence tool to evaluate mental skills of an individual (Hampshire et al., 2021; https://www.cognitron.co.uk/). GBA, glucocerebrosidase gene; MRI, magnetic resonance imaging; PD, Parkinson's disease; PSG, polysomnography; RBD, rapid eye movement sleep behaviour disorder; WES, whole-exome sequencing.