Hermann P et al. 2021, Biomarkers and diagnostic guidelines for sporadic Creutzfeldt-Jakob disease (Lancet Neurol 20(3):235-246, PMID 33609480, doi 10.1016/S1474-4422(20)30477-4)

An expert review in The Lancet Neurology, March 2021, led from the German national reference centre in Goettingen with 19 authors from surveillance and research centres in Europe, the USA, Australia and Japan. It surveys diagnostic and supportive tests for sporadic Creutzfeldt-Jakob disease (sCJD), covering RT-QuIC seed amplification, CSF 14-3-3, tau and neurofilaments, MRI, EEG, PRNP sequencing and blood candidates, and recommends amending the WHO-based clinical criteria for probable sCJD to include RT-QuIC. Read depth: the free PubMed Central author manuscript (PMC8285036) was read in full as text; the publisher version is behind a subscription and was not read, Figure 2 (the criteria graphic) and the supplement were not read. Information only, not medical advice. This entity records aggregate findings of the review and no individual patient data.

Key facts

  • The review covers literature from January 2015 to 15 November 2020 found through Google Scholar and PubMed, plus older work chosen by the authors, so it is an expert narrative review and not a systematic review or a formal guideline process.
  • Its own aim is an overview of biomarker-based sCJD diagnosis and suggested guidelines for clinicians dealing with rapidly progressive dementia.
  • Across 21 retrospective and prospective study rows in its Table 1, CSF RT-QuIC specificity was 99 to 100 percent; sensitivity ranged from 73 to 89 percent with first generation protocols and 86 to 97 percent with second generation protocols (the text quotes 92 to 97 percent for second generation tests in the best studies).
  • RT-QuIC sensitivity is described as very high for the common MM1, MV1 and VV2 subtypes, lower for MV2 (75 to 93 percent), and much lower and based on small numbers for VV1 and MM2 cases and for sporadic fatal insomnia and fatal familial insomnia.
  • RT-QuIC on olfactory mucosa and skin biopsies is reported with sensitivities of 89 to 100 percent, but the review says their value against relevant mimics still needs validation.
  • CSF 14-3-3 by Western blot is reported with a pooled sensitivity of 92 percent and specificity of 80 percent, with lower sensitivity in early disease and in MV2 and MM2, and specificity falling when controls include acute brain injury or inflammation.
  • CSF total tau is described as useful with sensitivity and specificity around 90 percent in most studies but not formally part of the case definition then, with no agreed assay-specific cut-off; the p-Tau/t-Tau ratio is described as a better composite.
  • On MRI, restricted diffusion on DWI is described as typical, with reported sensitivity from about 80 up to 98 percent and specificity from 74 to 98 percent, and a 2020 index in 770 definite cases gave 92 percent sensitivity and 97 percent specificity.
  • The authors state that EEG periodic complexes have fallen to 39 to 45 percent sensitivity in recent CSF biomarker comparison studies, likely because cases are recognised earlier.
  • Most of the authors recommend amending the WHO clinical criteria for probable sCJD by adding RT-QuIC; Figure 2 is adapted from the criteria of the UK National CJD Research and Surveillance Unit, which were based on WHO criteria, with refined imaging criteria. A positive RT-QuIC lets a case be classed as probable at an early stage, even with a single cardinal symptom.
  • The authors advise that RT-QuIC should be run, where available, in every suspected prion disease case, while 14-3-3, EEG and MRI remain routine; they recommend PRNP analysis in all suspected cases and tell clinicians to contact national surveillance units for RT-QuIC access.
  • They warn that even a false positive rate under 1 percent will produce some wrong diagnoses as testing spreads, and that RT-QuIC cannot reliably separate forms of human prion disease.
  • On blood, the review sees no immediate prospect of a highly specific test comparable to CSF RT-QuIC, but suggests plasma tau and NfL as triage or trial-monitoring tools, and suggests RT-QuIC at trial enrolment with blood markers repeated during trials.
  • Bibliographic details agree across PubMed, Europe PMC and Crossref: Lancet Neurol 2021 March, volume 20, issue 3, pages 235-246, review article.

Timeline

  1. 1998WHO diagnostic criteria for sCJD combine symptoms, EEG and CSF 14-3-3.
  2. 2009MRI patterns (DWI and FLAIR) suggested for the WHO criteria.
  3. 2015Second generation RT-QuIC introduced, shortening assay time and raising sensitivity.
  4. 2020Literature search cut-off for the review (15 November).
  5. 2021Review published in The Lancet Neurology 20(3).

People

Last checked 1 November 2026

Sources for this page