Abstract:Background: Parkinson's Disease (PD) has been associated with a distinct odour, which emanates from the skin and is strongest in sebum-rich areas. Thermal Desorption Gas Chromatography Mass Spectrometry (TDGCMS) has revealed volatile signatures from sebum that distinguish individuals with Parkinson's Disease (PD) from healthy controls. Here, we hypothesized that the sebum volatilome of individuals will alter as the disease progresses, since we have shown that endogenous metabolites do alter in abundance at different stages of the disease. We sought to determine features that could be used as progression markers. Methods: Sebum was sampled from participants using cotton gauze and the headspace from these swabs was analysed directly with TDGCMS using established methods. Supervised and unsupervised multivariate analysis was used to compare the relative abundances of significant volatile metabolites found for each phenotype. Our primary model was built on data from subjects with clinically established PD (n=46) and healthy controls (n=28) sampled at two sites. The volatilome profiles obtained for these PD and control cohorts were compared with the profile of participants with idiopathic REM sleep behaviour disorder (iRBD) to investigate metabolite changes in probable prodromal PD. A subset were also subjected to olfactory analysis. We also compared a subset of PD participants sampled at yearly intervals for a total of three years and built a subsequent model from this data. For all models the significant features were putatively annotated based on accurate mass and fragmentation patterns against spectral libraries. Findings: Classification between PD and control validated our previously published work. Significant features were investigated, and the majority that were putatively annotated were alkanes and fatty acid methyl esters (FAMEs), with other metabolites identified as an aldehyde, purine, oleamide and tropinone. Data from iRBD samples were then included in the analysis and we found that samples from individuals with PD and with iRBD could be classified completely from each other and from control samples. We found 27 significant features whose abundance in samples from individuals with iRBD was intermediate between that found for PD and control samples. In olfactory analysis of a subset all PD samples were correctly identified, and 3/9 of the iRDB samples were classified PD. On clinical follow up 2 of these showed PD symptoms. Further, when analysing the volatilome from longitudinal PD sampling, almost a two-thirds of the significant features showed differential regulation over the three visits. Interpretation: There are differential volatile metabolite signatures in sebum from PD and control participants even with low (n < 50) numbers of samples which is useful when comparing to rare and prodromal phenotypes. In this pilot investigation, we find that iRBD volatile metabolite signature can be separated from PD and control subjects using multivariate analysis, and almost two-thirds of the significant features can be seen to alter at disease classification stage, indicating possible early markers of disease and its progression. In longitudinal samples up or down regulation over 3 years also provides evidence for the presence of progression markers in sebum. When considering PD vs. control participants, more than 85% of the identifiable metabolites were putatively annotated as hydrocarbons and fatty acid methyl esters (FAMEs). Our prior work indicates that these components arise from larger lipid molecules that decompose during ionisation in TDGCMS. Validation of the change in the volatilome of individuals was provided by odour analysis.

A Fast and Non-Invasive Artificial Intelligence Olfactory-Like System That Aids Diagnosis of Parkinson's Disease

Artificial Intelligent Olfactory System for the Diagnosis of Parkinson’s Disease

An Automated Hybrid Approach Via Deep Learning and Radiomics Focused on the Midbrain and Substantia Nigra to Detect Early-Stage Parkinson’s Disease

Non-invasive diagnostic tool for Parkinson's disease by sebum RNA profile with machine learning

Diagnosis of Parkinson's Disease via the Metabolic Fingerprint in Saliva by Deep Learning

Sensitivity of Sniffer Dogs for a Diagnosis of Parkinson's Disease: A Diagnostic Accuracy Study.

Molecular Features of Parkinson's Disease in Patient‐Derived Midbrain Dopaminergic Neurons

Detection of Skin Α-Synuclein Using RT-QuIC As a Diagnostic Biomarker for Parkinson’s Disease in the Chinese Population

Paper Spray Ionization Ion Mobility Mass Spectrometry of Sebum Classifies Biomarker Classes for the Diagnosis of Parkinson’s Disease

Exploring the application and challenges of fNIRS technology in early detection of Parkinson's disease

Development of a Simplified Smell Test to Identify Patients with Typical Parkinson's as Informed by Multiple Cohorts, Machine Learning and External Validation

Classification of Parkinson's Disease and Delineating Progression Markers from the Sebum Volatilome

Application of Odor Identification Test in Parkinson's Disease in China: A Matched Case-Control Study

Odor Identification Test in Idiopathic REM-Behavior Disorder and Parkinson's Disease in China

Automatic Classification Of Early Parkinson'S Disease With Multi-Modal Mr Imaging

Evaluating the Diagnostic Potential of Combined Salivary and Skin Biomarkers in Parkinson's Disease

Evaluation of Parkinsons disease with early diagnosis using single-channel EEG features and auditory cognitive assessment

Diagnostic Performance of Artificial Intelligence-Assisted PET Imaging for Parkinson’s Disease: a Systematic Review and Meta-Analysis

A Markerless 2D Video, Facial Feature Recognition–Based, Artificial Intelligence Model to Assist With Screening for Parkinson Disease: Development and Usability Study

Diagnostic performance of artificial intelligence-assisted PET imaging for Parkinson's disease: a systematic review and meta-analysis

Shoupa: An AI System for Early Diagnosis of Parkinson's Disease