Recipient organisationGuilford Street Laboratories LTD
Funding£2.1M
PeriodDec 2025 — Nov 2028
In plain English
AI plain-English summary
A simple blood test could diagnose Parkinson’s disease up to seven years before symptoms appear, and with 100% accuracy in initial studies. Currently, Parkinson’s is diagnosed through clinical exams and brain scans, a process that can take years and still gets 10–20% of diagnoses wrong. This delay matters because new treatments that could slow or halt the disease are on the horizon—but they will only work if given early. The test, called PD Predict, measures eight proteins in the blood using targeted mass spectrometry and a machine-learning algorithm. It has already received FDA breakthrough designation. If validated in larger patient groups and approved by regulators, PD Predict could replace costly DaTscan imaging and cut diagnostic errors. For patients, that means faster access to symptomatic treatment and, eventually, to disease-modifying therapies at the stage when they can do the most good. The project will also standardise the assay for clinical labs and submit it for approval in the UK and US.
View original technical description
Research question: Can PD Predict, a new blood-based biomarker test, provide early, accurate diagnosis of Parkinson s disease? Background: Parkinson s disease (PD) is the second most common neurodegenerative condition. Currently, it is diagnosed solely on clinical grounds, based on the constellation of signs present on examination. Diagnosis is often delayed due to uncertain findings that require repeated assessments, coupled with additional investigations such as DaTscan imaging. Additionally, between 10-20% of diagnoses are wrong, and a similar number missed, due to other conditions that can resemble PD. These uncertainties directly impact patients and can result in delayed symptomatic treatment. Importantly, with emerging therapies poised to slow or halt PD progression, this represents a critical therapeutic window that must be addressed to enable early delivery of any future disease modifying treatments. In 2024, an unbiased proteomic study by UCL s translational mass spectrometry team identified eight plasma proteins that change in PD. These biomarkers were incorporated into PD Predict, a higher throughput targeted mass spectrometry assay, which, when combined with machine learning, correctly identified PD in 100% of cases in a larger patient cohort and correctly predicted disease up to seven years prior in high risk individuals with idiopathic REM sleep behaviour disorder (iRBD). To bring PD Predict into real-world clinical use, Guilford Street Laboratories (GSL), a UCL spin-out, has been established. PD Predict has received FDA breakthrough designation to accelerate regulatory approval.This project, in collaboration with neurologists and clinician scientists at UCL Queen Square Institute of Neurology and National Hospital for Neurology and Neurosurgery (NHNN), will advance PD Predict toward approval. Aims and Objectives: Over a three-year project, GSL aims to continue from the published work to: Develop PD Predict into a high-throughput, standardized, and fully quantitative assay. Validate its diagnostic accuracy in well-characterized PD cohorts and assess specificity against other causes of parkinsonism. Compare PD Predict with existing diagnostic methods such as DaTscan and the NuLiSA assay. Implement PD Predict into a laboratory clinical testing environment, meeting ISO15189 standards at NHNN Conduct real-world testing on patients with suspected PD to evaluate diagnostic impact Submit PD Predict for FDA and EMA regulatory approval for use in the UK and US. Methods: Targeted mass-spectrometry is an established clinical diagnostic platform, enabling the translation of multiple biomarkers with high specificity and scalability. The machine learning algorithm calculates a risk score, comparing patient results against known PD and non-PD profiles to generate a definitive positive or negative result. The assay s flexibility also allows for the addition of new biomarkers if required. Timelines for Delivery: Months 15-16: Confirm PD Predict s diagnostic capability in larger patient cohorts. Months 22-24: Compare with other biomarker platforms and develop to clinical standard. Month 36: Conduct real-world testing and submit for regulatory approval. Anticipated Impact & Dissemination: PD Predict has the potential to deliver earlier, more accurate PD diagnosis and reduce reliance on costly imaging tests. Improving early detection will enable individuals to access any future disease-modifying treatments at the earliest possible stage, helping to transform the trajectory of disease.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know