Blood test may reveal early signs of Parkinson's disease.

Researchers at Chalmers University used machine learning to identify short-lived blood markers tied to early Parkinson's, and the team says the findings could lead to a clinical test within about five years.

Researchers at Chalmers University of Technology report finding short-lived signals in blood that are linked to the early phase of Parkinson's disease. They examined biological processes thought to begin long before motor symptoms arise. Using machine learning, the team identified a genetic pattern in people without Parkinson's symptoms. Authors say the work could help detect the disease earlier and support drug research. Key findings: - The study found short-lived blood patterns that appear between the incubation period and symptom onset of Parkinson's. - Researchers focused on two cellular processes: DNA repair and the cellular response to stress. - Machine learning was used to identify a genetic pattern in people who were not yet showing symptoms. - The team notes that motor symptoms often appear after 50 to 80 percent of relevant brain cells are already damaged or lost. - Investigators say the results could lead to a clinical test within about five years and could inform drug development and repurposing efforts. Summary: These results point to a potential window in which Parkinson's-related processes leave detectable signals in blood well before motor symptoms. Researchers report the findings could be developed into a screening test in roughly five years and could contribute to efforts to understand and target early disease mechanisms.