Blood test could spot Parkinson’s years earlier

AI-powered blood test

An artificial intelligence-powered blood test that may be able to predict Parkinson’s disease several years before symptoms start to develop could allow earlier treatment and help guide the search for new treatments.

Researchers at University College London and University Medical Centre Goettingen in Germany report in the journal Nature Communications that the test was 100% accurate in identifying Parkinson’s when comparing blood samples from 99 people recently diagnosed with the neurodegenerative disease and 36 people without it.

It looks at a panel of eight protein biomarkers circulating in the blood that have been found to correlate with the severity of Parkinson’s symptoms.

A simple blood test would be a major step forward, as Parkinson’s is often diagnosed much later, when symptoms have developed and substantial numbers of dopamine-producing neurons in the brain have been lost.

Brain scans can be helpful, but are not yet good enough to make a diagnosis, and it isn’t feasible to take samples of cells from the brain for analysis. Meanwhile, progress has recently been made with tests that measure a protein called alpha-synuclein in cerebrospinal fluid (CSF), although, they require an invasive lumbar puncture and a peripheral blood test would be much simpler.

Patient advocacy group Parkinson’s UK, which co-funded the study, called the findings a “major step forward,” although it cautioned that the test would need to be tested in larger studies before it could be ready for routine use.

At the moment, the journey to a Parkinson’s diagnosis can be a long one and often means that around one in five people diagnosed with the disease might not have it.

“With more work, it may be possible that this blood-based test could distinguish between Parkinson’s and other conditions that have some early similarities, such as multiple systems atrophy or dementia with Lewy bodies,” said the charity’s head of research, Professor David Dexter.

The study also looked at blood samples from people with idiopathic REM sleep behaviour disorder (iRBD), a group which is known to be at a elevated risk of going on to develop Parkinson’s.

The test identified that 79% of 54 iRBD samples had a similar readout to the samples from people already diagnosed with Parkinson’s, so may predict who might go on to develop the disease up to seven years earlier.

One of the lead researchers, Professor Kevin Mills of UCL – whose mother had iRBD and went on to develop Parkinson’s – said: “At present we are shutting the stable door after the horse has bolted and we need to start experimental treatments before patients develop symptoms.”

He added: “As new therapies become available to treat Parkinson’s, we need to diagnose patients before they have developed the symptoms. We cannot regrow our brain cells and therefore we need to protect those that we have.”

A note of caution was sounded by Professor Ray Chaudhuri of King’s College London, an expert in movement disorders and neurology, who said that, while the test could be a step forward, “questions, however, remain about the ethics of predictive diagnosis in relation to proper counselling, as well as absence, currently, of any disease-modifying treatment.”

Potentially disease-modifying drugs are in clinical trials, however, including alpha-synuclein-targeting therapies from Lundbeck, Novartis/UCB, AstraZeneca, and Sanofi/ABL Bio, amongst others.

Photo by on Unsplash