A team of researchers, led by scientists from UCL and University Medical Center Goettingen, Germany, has developed a blood test using artificial intelligence (AI) to predict Parkinson's disease up to seven years before symptoms appear. Parkinson's, the fastest-growing neurodegenerative disorder, currently affects nearly 10 million people worldwide. It is caused by the death of nerve cells in the brain's substantia nigra, leading to a loss of dopamine production due to the buildup of the protein alpha-synuclein.
Current treatments involve dopamine replacement therapy after symptoms manifest, such as tremors and memory problems. Early prediction could allow for treatments that protect dopamine-producing cells, potentially slowing or stopping the disease's progression. The study, published in Nature Communications, utilized machine learning to analyze eight blood-based biomarkers, achieving a diagnosis accuracy of 100%.
The test was further validated on 72 patients with Rapid Eye Movement Behaviour Disorder (iRBD), a condition preceding synucleinopathies like Parkinson's. The AI accurately predicted 79% of iRBD patients would develop Parkinson's, matching the clinical conversion rate over ten years.
Co-first-author Dr. Michael Bartl and Dr. Jenny Hällqvist highlighted that identifying these proteins in blood could enable earlier drug therapies, potentially slowing disease progression. The team, including Professor Kailash Bhatia, is further testing the accuracy on high-risk individuals and aims to develop a simpler blood spot test for even earlier detection.
The research, funded by several organizations including Parkinson's UK, represents a significant step toward a definitive diagnostic test for Parkinson's. Professor David Dexter of Parkinson's UK emphasized the potential for this less invasive blood test to distinguish Parkinson's from other similar conditions, contributing to recent advances in the field. | BGNES
