Researchers have developed a simple blood test using artificial intelligence to predict Parkinson’s disease years before symptoms appear. They aim for it to become an affordable, finger-prick test for early diagnosis and to aid in finding treatments to slow the disease’s progression. Parkinson’s UK called it “a major step forward,” but larger trials are needed to verify its accuracy.

Parkinson’s affects nearly 10 million people worldwide, including over 150,000 in the UK. Many are diagnosed only after symptoms like tremors and movement issues have emerged, caused by the death of nerve cells in the brain. Currently, there is no cure or treatment to halt the disease, though therapies can manage symptoms.

Researchers from University College London and University Medical Center Goettingen in Germany studied blood samples from people with and without Parkinson’s. They identified eight key proteins linked to inflammation and protein degradation that could predict the disease. These markers could help develop new treatments.

The team tested blood from 72 patients at risk for brain disorders and followed them over ten years. Using the AI tool, they accurately predicted that 16 would develop Parkinson’s, sometimes up to seven years before symptoms began. Overall, the test predicted 79% of cases, with ongoing follow-ups for further confirmation.

Senior author Prof. Kevin Mills emphasized the need for early intervention: “We need to start experimental treatments before patients develop symptoms.” Co-author Dr. Jenny Hällqvist added, “We need to protect neurons before they are lost.”

Parkinson’s symptoms typically begin mildly and worsen gradually, affecting movement with shaking, slow steps, and muscle stiffness. The disease results from nerve cells losing their ability to produce dopamine due to a build-up of the protein alpha-synuclein.