Odors from earwax may help distinguish individuals with Parkinson’s disease (PD) from those without the condition, a new study suggests.
Researchers found that four volatile organic compounds (VOCs) in ear canal secretions significantly differed between participants with and without PD.
The compounds — ethylbenzene, 4-ethyltoluene, pentanal, and 2-pentadecyl-1,3-dioxolane — may represent potential biomarkers. An artificial intelligence olfactory (AIO)-based screening model used in the study identified those with PD with 94% accuracy.
“The accuracy of the model really surprised us,” study investigator Hao Dong, Research Center for Frontier Fundamental Studies, Zhejiang Lab, Hangzhou, China, MD, told Medscape Medical News.
However, the study was a “small-scale, single-center experiment,” he noted in a press release.
“The next step is to conduct further research at different stages of the disease, in multiple research centers, and among multiple ethnic groups in order to determine whether this method has greater practical application value,” Dong said.
The findings were published online recently in Analytical Chemistry.
Unique Odor Profile
“Our team has long been engaged in the detection of [VOCs] secreted by the human body. By chance, we came across reports on the detection of sebum VOCs for Parkinson’s,” Dong said.
Sebum, the oily substance secreted by the skin, may carry a distinct scent in individuals with PD. In a 2019 study cited by Dong, researchers noninvasively collected sebum samples from the upper backs of 64 participants. The findings suggested that samples from those with PD contained compounds associated with a unique odor profile.
Dong and his team began with a confirmatory experiment using sebum samples collected from the upper back, as in the original study. However, they found that earwax was easier to collect and had a more stable chemical composition. These findings led them to focus on earwax in the current study.
Ear wax also contains sebum. But unlike sebum on the surface of the skin, which is exposed to various factors that can degrade it. In contrast, sebum on skin inside the ear canal is protected.
Dong’s study included 209 participants, 108 of whom had a diagnosis of PD. Ear canal secretions were collected from all participants using swabs and analyzed using gas chromatography-mass spectrometry.
Results showed that ear canal secretions from participants with PD contained 196 distinct VOCs compared with 168 VOCs in those without PD. Interestingly, no two participants had identical VOC profiles.
A Disease ‘Fingerprint’?
“In this case, VOC components could be used as a ‘fingerprint’ for disease identification,” the researchers wrote.
Adjusted analyses identified four VOCs that significantly differed between participants with and without PD: ethylbenzene, 4-ethyltoluene, pentanal, and 2-pentadecyl-1,3-dioxolane.
The investigators trained the AIO system using VOC data. By combining gas chromatography-surface acoustic wave sensors with a convolutional neural network (CNN) model, the AIO system achieved up to 94.4% accuracy in distinguishing participants with PD from those without.
In addition, the CNN model demonstrated a high level of performance with an area under the curve of 0.98, well above the 0.8 threshold considered strong by the researchers.
“Further enhancements to the diagnostic model could pave the way for a promising new PD diagnostic solution and the clinical use of a bedside PD diagnostic device,” the investigators wrote.
For now, Dong said the study’s takeaway message for clinicians is that “the potential of volatile organic compounds secreted by the skin as biomarkers for Parkinson’s disease has been further verified.”
The study was funded by the National Natural Science Foundation of China, “Pioneer” and “Leading Goose” R&D Program of Zhejiang Province, and the Fundamental Research Funds for the Central Universities. The investigators reported having no relevant financial relationships.
