Background: Heterozygous carriers of Parkin mutations are suggested to be at risk of developing Parkinson's disease, while biallelic variants are associated with typical autosomal recessive early-onset PD. Investigating unaffected heterozygous mutation carriers holds the potential of a deeper understanding of monogenic PD and has implications for PD in general, in particular regarding the prodromal phase.
Objectives: To discriminate healthy Parkin mutation carriers from healthy non-mutation carriers using a multimodal approach.
Methods: Twenty-seven healthy heterozygous Parkin mutation carriers (13 female. age: 48 ± 13 years) and 24 healthy non-mutation carriers (14 female. age: 48 ± 15 years) from the CHRIS study (Cooperative Health Research in South Tyrol) were recalled based on their genetic profile and underwent a blinded assessment of motor and non-motor PD symptoms, transcranial sonography and sensor-based posturography and gait analyses under different conditions with increasing difficulty. For the latter, gradient-boosted trees were used to discriminate between carriers and non-carriers. The classification accuracy and the area under the curve of the receiver-operator characteristics curve were calculated.
Results: We observed no differences concerning motor or non-motor symptoms and substantia nigra hyperechogenicity. The best gradient-boosted trees-based model on posturography measurements (tandem feet, eyes closed, firm surface), however, showed a classification accuracy of up to 86%. The best-performing gradient-boosted trees-based model for gait analyses showed a balanced accuracy of up to 87% (dual-tasking).
Conclusions: Sensor-based quantification of movements allows to discriminate unaffected heterozygous mutation carriers from mutation-free controls. Thereby, it is crucial to challenge the motor system with more difficult tasks to unmask subtle motor alterations.
Keywords: Gait analysis; Genetic PD; Parkin; Parkinson's disease (PD); Wearables.
Copyright © 2021 Elsevier Ltd. All rights reserved.