Is there a blood test for Parkinson's yet?

Can a doctor run a blood test today to tell you if you have Parkinson's?

No routine blood test exists today to diagnose Parkinson's disease. If you visit a general practitioner or a clinic, they cannot order a blood test to tell you if you have the condition. Currently, the only way to obtain a 100% certain diagnosis is neuropathology, which is a brain tissue examination performed after death.

The closest scientists have come are experimental tests still confined to research laboratories. A meta-analysis of 36 studies found that alpha-synuclein, which is the key toxic protein that clumps inside the brain, was higher inside exosomes, which are tiny bubbles released by cells into the blood. To try and validate this, an active trial registration (NCT06941012) is currently recruiting participants to measure these bubbles, while a separate preprint, which is a research paper that has not been peer-reviewed, by Long and colleagues in August 2026 evaluated a blood test using an ultra-sensitive clumping technique.

Current clinical diagnosis is entirely subjective and highly prone to error. Neurologists must rely on physical symptoms like bradykinesia, which is a slowness of movement, alongside muscle stiffness or a resting tremor. A systematic review and meta-analysis of 20 studies by Rizzo and colleagues found that clinical diagnosis accuracy is only 80.6% compared to autopsies, and this drops to 73.8% when made by non-experts.

What are researchers actually finding in blood samples?

Researchers are tracing several distinct proteins in blood, but their diagnostic meaning remains unsettled. A primary target is alpha-synuclein, the protein that clumps in the brain. A meta-analysis of 32 studies with 2,683 patients and 1,838 controls found that total alpha-synuclein is elevated in blood, but meta-analyses directly contradict each other on whether these levels correlate with milder or more severe symptoms.

Other experimental markers are used to estimate ongoing brain damage. A network meta-analysis of 49 studies with 7,787 participants found that neurofilament light chain, which is a structural protein that leaks into blood when nerve fibres are damaged, is significantly higher in atypical parkinsonian syndromes like progressive supranuclear palsy than in Parkinson's disease. Additionally, a preprint, which is a research paper that has not been peer-reviewed, found that brain cell inflammatory markers like glial fibrillary acidic protein remain indistinguishable from healthy controls.

Scientists are registering trials to test combinations of these blood markers. A trial registration, which is a formal public record of a trial before it begins, details plans for study NCT07480187 to measure blood plasma ratios of amyloid-beta 42 to amyloid-beta 40 alongside tau proteins to build a molecular profile. Another trial registration, NCT06941012, is tracking four specific alpha-synuclein modifications in brain-derived blood vesicles, but neither trial has published any results.

Why does a test that works in spinal fluid struggle in blood?

The blood-brain barrier severely restricts the passage of brain proteins into the bloodstream. This barrier is a tightly packed cellular border that protects the brain but also prevents potential disease markers from leaking out. Consequently, proteins are highly concentrated in cerebrospinal fluid, which is the liquid surrounding the brain, but are diluted to extremely low levels in blood draws.

Blood is a highly complex environment containing interfering materials. A preprint, which is a research paper that has not been peer-reviewed, points out that red blood cells contain over 99% of blood-based alpha-synuclein. If these cells burst during collection, they flood the sample and completely hide the tiny amounts of protein that actually came from the brain.

Pre-analytical factors make blood-based measurements highly inconsistent. A systematic review of 18 studies involving 1,695 patients and 1,288 healthy controls found that testing these brain-derived packages is highly sensitive to the needle size used, patient fasting status, and the time of day of the draw. Furthermore, nobody has developed capturing molecules that can reliably separate vesicles originating in the brain from those produced by the rest of the body.

Can these blood markers tell Parkinson's apart from similar conditions?

Experimental blood tests cannot yet reliably distinguish Parkinson's disease from atypical syndromes. Although some markers show statistical differences between groups, they fail to provide clear diagnoses for individual patients. These tests suffer from small study sizes, inconsistent results, and significant publication biases that exaggerate their accuracy on paper.

Neurofilament light chain can signal that a more aggressive disease is present but cannot identify which one. A 2025 network meta-analysis of 49 studies with 7,787 participants found that blood levels of this nerve-damage protein were higher in multiple system atrophy and progressive supranuclear palsy compared to Parkinson's disease. However, the test cannot reliably tell these atypical syndromes apart from one another.

Measuring alpha-synuclein in blood-borne packages has failed to differentiate these conditions. A 2023 meta-analysis of 13 studies with 1,565 patients with Parkinson's and 206 with multiple system atrophy found that alpha-synuclein levels within these vesicles did not differ significantly. A systematic review of 18 studies found that diagnostic sensitivity ranged from a weak 0.53 to 0.96, and noted that small studies with negative results were frequently left unpublished.

What still needs to happen before a test reaches an ordinary clinic?

Regulators require a blood test to pass strict analytical and clinical validation before general clinical use. Analytical validation proves the test is consistent, requiring a coefficient of variation, which measures results variability, to remain below 15% to 20% under European Medicines Agency guidelines. Most diagnostic studies use weak case-control designs that compare established patients to healthy controls, which ratings under the QUADAS-2 bias-assessment tool show heavily exaggerates accuracy.

Diagnostic developers cannot assume blood tests will automatically replicate what is seen in spinal fluid. A July 2026 preprint, which is a research paper that has not been peer-reviewed, analysed 130 patients and found that brain, blood, and cerebrospinal fluid vascular markers reflect entirely distinct biology, resulting in a low blood diagnostic accuracy of 0.54. Blood markers must therefore be validated directly in large, independent patient groups.

Large, ongoing studies are trying to track patients over time to gather this evidence. The VαMPiRE study, which is a registered trial that has not yet reported results, plans to follow 600 patients with Parkinson's and 600 controls over 24 months to see if blood markers can detect the disease early. Another registered trial with no published results, the BIOPARK study (NCT05385315), is recruiting patients from a new geographical region to ensure their metabolic blood marker works across different populations.

What this does not show

A significant portion of the published evidence in this field has been retracted, corrected, or remains unreported. A randomised trial of 160 patients by Huang and colleagues investigating blood inflammatory markers was retracted in 2024, and a trial of 60 patients by Tamtaji and colleagues claiming probiotics lowered blood inflammatory markers was retracted in 2025. A systematic review on integrated medicine by Wang and colleagues was retracted in 2023, while a 2026 preprint trying to perform a genomic screen on 755 patients was retracted in 2026 to protect patient privacy. Additionally, a trial of 50 participants by Alrafiah and colleagues assessing inflammatory biomarkers was corrected in 2020.

Major chemical markers used to capture brain packages in the blood are fundamentally flawed. Researchers have relied on L1CAM to isolate brain-derived vesicles, but biochemical studies show L1CAM is not exclusive to the brain and behaves as a soluble blood protein, leading to non-specific binding. Consequently, a systematic review of 18 studies found that blood vesicle measurements provide only moderate and unreliable diagnostic accuracy. Furthermore, several highly publicised claims rely entirely on registered trials that have not yet reported any results, including the BIOPARK study (NCT05385315), the VαMPiRE study, and clinical trials NCT06846658, NCT07480187, and NCT06941012.

While this research highlights why a simple blood test is not yet available in clinics, it provides you with the specific questions to ask about how your current clinical diagnosis was reached. Your neurologist can explain how they weighed your physical symptoms and ruled out alternative conditions without relying on experimental laboratory tests.