Key facts
- Published 21 April 2021 in Alzheimer's Research & Therapy, volume 13, article 86, open access under CC BY; received 29 September 2020, accepted 23 March 2021.
- Ten authors led by Inga Zerr (Gottingen) with Franc Llorens (IDIBELL and CIBERNED), and Gothenburg and UCL co-authors including Zetterberg and Blennow.
- Cohort: 309 plasma samples, including 83 CJD cases (15 probable, 68 definite) and 70 healthy blood-donor controls, plus Alzheimer, Lewy body, frontotemporal, vascular and other neurological groups. Cases came from the German TSE reference centre biobank.
- CJD had the highest plasma levels, about 38-fold for NfL and 18-fold for t-tau versus controls; both markers separated CJD from non-CJD dementias with AUC 0.93.
- Cutoffs reported: NfL 70 pg/mL gave 79 percent sensitivity and 96 percent specificity versus non-CJD dementias; t-tau about 6 pg/mL gave 95 percent sensitivity and 84 percent specificity. Measured on the Quanterix Simoa HD-1.
- Plasma t-tau, but not NfL, related to PRNP codon 129 genotype, subtype and disease duration, and gave moderate survival prediction.
- Plasma and CSF values correlated, and diagnostic accuracy was similar (plasma t-tau AUC 0.93 versus CSF 0.97).
- Stated limits: no serial samples, no external validation cohort, retrospective selection, and blood drawn up to 15 days after CSF in some cases.
- Funded by multiple public and charitable sources including the Robert Koch Institute, Instituto de Salud Carlos III and the European Research Council; two authors declare industry biomarker ties.
Timeline
- 2020Manuscript received by the journal.
- 2021Accepted.
- 2021Published online.
People
Last checked 1 November 2026