Comprehensive subtyping of Parkinson’s disease patients with similarity fusion: a case study with BioFIND data

Abstract

Parkinson’s disease (PD) is a complex neurodegenerative disorder with diverse clinical manifestations. To better understand this disease, research has been done to categorize, or subtype, patients, using an array of criteria derived from clinical assessments and biospecimen analyses. In this study, using data from the BioFIND cohort, we aimed at identifying subtypes of moderate-to-advanced PD via comprehensively considering motor and non-motor manifestations. A total of 103 patients were included for analysis. Through the use of a patient-wise similarity matrix fusion technique and hierarchical agglomerative clustering analysis, three unique subtypes emerged from the clustering results. Subtype I, comprised of 60 patients (~58.3%), was characterized by mild symptoms, both motor and non-motor. Subtype II, comprised of 20 (~19.4%) patients, was characterized by an intermediate severity, with a high tremor score and mild non-motor symptoms. Subtype III, comprised of 23 (~22.3%) patients, was characterized by more severe motor and non-motor symptoms. These subtypes show statistically significant differences when looking at motor (on and off medication) clinical features and non-motor clinical features, while there was no clear difference in demographics, biomarker levels, and genetic risk scores.

Social Determinants of Health and Suicidal Behaviors among Children: U.S. Longitudinal Adolescent Brain Cognitive Development (ABCD) Study

“…Specifically, we first split the entire cohort into 5 folds, then each time, we successively dropped 1 fold (20% children) and used the remaining 4 folds (80% children) to construct a subset. 48 We then re-identified SDoH patterns in each subset. In both sensitivity analyses, patterns of SDoH were re-identified using the hierarchical clustering algorithm, following the same criteria as in the primary analysis for determining the optimal number of clusters.…”

MNCD: A New Tool for Classifying Parkinson’s Disease in Daily Clinical Practice

“…However, combining both motor and NMS could be of great interest and usefulness [ 6 , 7 ]. Very recently, Brendel et al [ 51 ], using data from the BioFIND cohort with clustering analysis, identified three unique subtypes: subtype I, characterized by mild symptoms, both motor and non-motor; subtype II, characterized by an intermediate severity, with a high tremor score and mild non-motor symptoms; and, subtype III, with more severe motor and non-motor symptoms. Although different biological markers may be useful in the future when classifying PD for its management (i.e., genetic, neuroimaging, biochemical, etc.)…”

Deep phenotyping for precision medicine in Parkinson's disease

“…Another aim of precision medicine is to provide tailored treatment to individual patients, which can be achieved by classifying patients into finely grained disease subtypes that respond better to certain treatments. Unsupervised ML ( Box 1 ) methods like clustering ( Box 1 ) approaches and latent class ( Box 1 ) can reveal such subtypes ( Brendel et al, 2021 ). Precision medicine also aims to identify treatment tailored to a specific stage of a disorder.…”