Blood seed amplification

The research record

This is the project's file for this landmark, as written and checked. Sources are listed at the end; anything not yet confirmed is marked "Not yet verified".

Who

Ayami Okuzumi, Taku Hatano, Nobutaka Hattori (senior author) and colleagues (Juntendo University, Tokyo), with partners in Luxembourg, Nature Medicine 29:1448-1455, 2023; funded by Japanese government and AMED grants. Annika Kluge, Daniela Berg (University Hospital Kiel) and Friederike Zunke (last author, University Hospital Erlangen), Brain 145:3058-3071, 2022; funded by the German Research Foundation and ParkinsonFonds Germany.

What happened

The spinal-fluid assay (landmark 9) needs a lumbar puncture. Blood is much easier to draw, but blood has little synuclein seed and many interfering proteins.

• Okuzumi 2023. The team added an immunoprecipitation step (an antibody pulls synuclein out of serum first) before the amplification reaction (IP/RT-QuIC). The first internal cohort had 279 people with a synucleinopathy (221 with Parkinson's); the second had 40 patients (34 with Parkinson's, 6 with MSA) and 9 non-neurodegenerative controls; the blinded external cohort had 35 patients (20 Parkinson's, 15 MSA), 20 controls and 6 people with a tau disease. The test was positive in 210 of 221 (95%), 34 of 34 and 15 of 20 (75%) Parkinson's patients in the three cohorts. The area under the curve (AUC, where 1.0 is perfect) separating Parkinson's from controls was 0.96 and 0.93 in the two internal cohorts. In the blinded external cohort it was 0.86 for Parkinson's and 0.80 for multiple system atrophy. Separation of MSA from controls inside the internal cohorts was weaker (AUC 0.64 and 0.73). The amplified seeds kept disease-specific properties, which allowed PD and MSA to be told apart.
• Kluge 2022. Neuron-derived extracellular vesicles (tiny packets released by nerve cells) were isolated from plasma. Misfolded synuclein in them was detected by immunoblot under native conditions and was raised in all Parkinson's samples. The seeding assay could amplify it. The study was small (30 patients and 50 controls, with 15 per group in several analyses) and not blinded.

Why it moved us toward a cure

A cheap blood test would make it possible to screen large numbers of at-risk people for prevention trials (landmark 225) and to check target biology in ordinary clinics.

Outcome

Open. Both studies are first reports with limited cohorts; neither was shown here to work as a stand-alone test across sites. The spinal-fluid assay remains the validated tool. A June 2025 network meta-analysis put blood-based seed assays among the better-performing sample types (pooled sensitivity and specificity near or above 90% for several sample types), but a 2024 study said no blood or saliva assay had yet been rigorously verified. Corrections: Okuzumi 2023 has an author correction (Nat Med 2025;31:698) replacing one image in Extended Data Fig. 3c; Kluge 2022 has a correction (published online 21 Oct 2022, in the issue of Brain 2023;146:e6) because the publisher had supplied incomplete supplementary files. Neither changes the results. No retraction is recorded.

Sources

• Okuzumi et al., Nat Med 2023;29:1448–1455, DOI 10.1038/s41591-023-02358-9, PMID 37248302
• Kluge et al., Brain 2022;145:3058–3071, DOI 10.1093/brain/awac115, PMID 35722765
• Okuzumi author correction, Nat Med 2025, DOI 10.1038/s41591-025-03521-0 · Kluge correction, Brain 2023, DOI 10.1093/brain/awac370
• Systematic review and network meta-analysis of seed amplification assays, 2025 (PMC12192484) (search summary; I did not read the paper)
• Not yet verified: independent replication of either study; any 2024–26 blood-based assay approved or qualified by a regulator; sensitivity and specificity in prodromal people; whether the Kluge result has been replicated.