Identifying Parkinson’s disease subtypes with motor and non-motor symptoms via model-based multi-partition clustering

Abstract

Identification of Parkinson’s disease subtypes may help understand underlying disease mechanisms and provide personalized management. Although clustering methods have been previously used for subtyping, they have reported generic subtypes of limited relevance in real life practice because patients do not always fit into a single category. The aim of this study was to identify new subtypes assuming that patients could be grouped differently according to certain sets of related symptoms. To this purpose, a novel model-based multi-partition clustering method was applied on data from an international, multi-center, cross-sectional study of 402 Parkinson’s disease patients. Both motor and non-motor symptoms were considered. As a result, eight sets of related symptoms were identified. Each of them provided a different way to group patients: impulse control issues, overall non-motor symptoms, presence of dyskinesias and pyschosis, fatigue, axial symptoms and motor fluctuations, autonomic dysfunction, depression, and excessive sweating. Each of these groups could be seen as a subtype of the disease. Significant differences between subtypes (P< 0.01) were found in sex, age, age of onset, disease duration, Hoehn & Yahr stage, and treatment. Independent confirmation of these results could have implications for the clinical management of Parkinson’s disease patients.

Data-Driven Subtyping of Parkinson’s Using Acoustic Analysis of Sustained Vowels and Cluster Analysis: Findings in the Parkinson’s Voice Initiative Study

“…Unsurprisingly, they found there are different ways to cluster PwP typically finding two or three clusters to group participants depending on the latent variables used. Collectively, it may not be straightforward to claim overall validity on the basis of the number of defined PD subtypes and agreement with the research literature, also in view of the findings in [15]. Nevertheless, we need to stress that our study has considerably more PwP than almost all existing studies in the PD subtype literature and hence as a result of greater statistical power this enables easier identification of emerging patterns and potentially exploring different ways to partition the data and explore larger numbers of resulting PwP groups.…”

“…recent work by Lee et al [57] and Brendel et al [14] who have reported the presence of three PD clusters. Rodriguez-Sanchez et al [15] approached the PD subtyping problem from a different perspective: instead of attempting to define PD subtypes directly using the variables available, they explored model-based multi-partition clustering to explore different approaches to define PD subtypes via the use of latent variables. Unsurprisingly, they found there are different ways to cluster PwP typically finding two or three clusters to group participants depending on the latent variables used.…”

“…For this reason, data-driven approaches to define PD subtypes have received considerable attention in the PD research community over the last few years using cluster analysis [10][11][12][13][14][15]. In the following, we aim to provide a broad overview of studies published in this area placing emphasis on recent work, however, we remark we did not attempt to pursue a systematic literature review.…”

Defining the unknowns for cell therapies in Parkinson's disease

“…Indeed, several studies have reported non-motor symptoms to have a greater impact on quality of life for people with PD than the motor symptoms ( Duncan et al, 2014 ; Hinnell et al, 2012 ). There is considerable variation in presentation ( Rodriguez-Sanchez et al, 2021 ), and treatments for non-motor symptoms remain a significant unmet need. While some cognitive and neuropsychiatric symptoms certainly arise from imbalances in other neurotransmitter systems, such as serotonergic and noradrenergic transmission, there is also increasing recognition of the role of dopamine in these cognitive and neuropsychiatric manifestations ( Fig.…”

Two distinct trajectories of clinical and neurodegeneration events identified in Parkinson's disease

“…Great efforts have been made to identify PD subtypes, from the initial classi cation based on a single motor symptom, to classi cation based on multidomain phenotypes including motor and nonmotor symptoms, neuroimage data, and biochemical markers [3][4][5][6][7][8] . On account of the inconsistent reliability and di culty in characterizing disease heterogeneity, the empirical classi cation methods fade away, and have been replaced by data-driven methods without a priori hypotheses [9][10][11][12][13][14] . These data-driven studies provide invaluable information toward the understanding of complex mechanisms but have limitations.…”