Some cases of Alzheimer’s disease progress quickly, mimicking prion-based Creutzfeldt-Jakob disease (CJD). Many people with this form of Alzheimer’s are misdiagnosed, because clinicians have no reliable way to distinguish between the two disorders. In the January 5 JAMA Neurology, researchers led by Isabelle Quadrio at Hospices Civils de Lyon, Bron, France, propose using levels of total prion protein (t-PrP) in cerebrospinal fluid (CSF) to differentiate CJD from AD. The authors found that people with prion disease had lower CSF levels of this protein than AD patients did. In a study of 209 patients with either disorder, t-PrP classified patients much more accurately than the currently accepted biomarker, 14-3-3 protein, they report. When they combined t-PrP with CSF tau, they correctly identified 96 percent of patients with atypical, fast-progressing AD in this study, as compared with 57 percent using 14-3-3 alone.

“This is a very promising result that should now be validated in independent replication studies,” wrote Henrik Zetterberg of the University of Gothenburg, Sweden, in an accompanying commentary. The findings complement recent data from the Gothenburg group highlighting synaptic markers in CSF as a way to sharpen the early diagnosis of AD (see Jan 2015 news). 

Currently, clinicians diagnose CJD by combining clinical symptoms, including dementia, involuntary muscle contractions, tremor, rigidity, and problems with balance and vision, with biomarkers such as MRI, EEG readings, and CSF 14-3-3. Together, these detect the disease with a specificity of about 71 percent (see Zerr et al., 2009). Protein 14-3-3, which escapes into interstitial fluid when neurons die, reflects rapid neurodegeneration (see Hsich et al., 1996). However, this marker appears in the CSF of people with other neurodegenerative conditions as well, including some cases of rapidly progressing AD. Likewise, CSF total tau rises steeply in CJD and in other disorders as well. Some researchers have suggested using the ratio of total tau to phosphorylated tau to improve differential diagnosis, since p-tau occurs mainly in AD (see Riemenschneider et al., 2003; Blennow et al., 2005).

Quadrio and colleagues wondered if prion protein, which makes up the underlying pathology in CJD, could be a more specific marker of this disease. In support, first author Aline Dorey found significantly lower levels of t-PrP in stored CSF samples from 52 autopsy-confirmed patients with CJD than in 30 patients with confirmed AD. Next, the authors turned to living patients, analyzing 55 people with probable AD and 26 with probable CJD, and obtaining similar t-PrP results. Diagnoses were based on a combination of several types of biomarker evidence and clinical features for each disease.

Comparing these findings with CSF from 26 age-matched controls revealed that t-PrP is lowest in prion disease and highest in AD, with healthy controls in the middle. Why might this be? The authors speculated that the high levels in Alzheimer’s patients reflect neuronal damage. In CJD, neurons die even more rapidly, but in this condition the presence of pathological prion may pull the normal protein into growing brain deposits, leading to less free protein in the CSF. This would be analogous to the drop in CSF Aβ42 in AD.

Could CSF t-PrP serve as a diagnostic? The authors pooled results for the autopsy and living cohorts and found that a cutoff of 263 μg/L t-PrP distinguished between AD and CJD with 82 percent sensitivity and specificity. The ratio of t-tau/p-tau differentiated the groups with better accuracy. But when the authors combined t-PrP and tau into a “CJ factor” with the formula t-tau / (p-tau x t-PrP), values of more than 0.054 distinguished CJD from AD with 96 percent sensitivity and 98 percent specificity, they reported.

“If CSF t-PrP indeed is a direct measure of PrP pathology in the brain, this would be a welcome complement to the more technically challenging methods that detect PrP seeding activity or misfolding in CSF or blood,” Zetterberg noted in his commentary (see Mar 2014 newsJun 2014 news).

Patients with rapidly progressing AD present a particular diagnostic challenge, since they may have the motor problems and high CSF t-tau and 14-3-3 that CJD patients also have. Brain amyloid does not clearly demarcate these cases, because amyloid deposits are frequently found in the brains of people with sporadic and genetic CJD as well (see Kovacs et al., 2011). At present, many people with this type of AD end up being treated at prion centers, and are only correctly diagnosed after death (see Sep 2011 newsFeb 2013 news). The CJ factor performed equally well in a cohort of 46 patients with this type of AD, with values below 0.054 distinguishing them from CJD patients with 100 percent sensitivity and 96 percent specificity. In further analysis, the authors found that the CJ factor misclassified only 4 percent of rapid AD patients, as compared to 43 percent misclassified by the presence of protein 14-3-3 in the CSF. The combination of PrP and the tau ratio worked better than either biomarker alone, with the former misclassifying 9 percent of rapid AD patients, and the latter 13 percent.

“Determination of t-PrP in CSF as a new biomarker may help clinicians to exclude CJD in patients with unusual AD phenotypes,” Kurt Jellinger at the Institute of Clinical Neurobiology, Vienna, wrote to Alzforum (see full comment below).—Madolyn Bowman Rogers


  1. This international study group (Belgium, France, Germany, Austria) retrospectively determined CSF t-PrP and the ratio of total (t)tau and phospho (p)tau 181 in a cohort of 232 patients (30 autopsy-confirmed AD; 52 autopsy-confirmed CJD; 101 probable AD, including 46 atypical AD phenotypes; 26 probable CJD; and 23 non-demented controls). In CJD patients, CSF t-PrP concentrations were decreased compared to controls and AD patients. Differential diagnosis of CJD vs. atypical AD reached 82.1 percent sensitivity and 91.3 percent specificity. Misclassification rate of atypical AD, when using CSF 14-3-3 protein determination alone (43.5 percent) was reduced to 4.3 percent when calculating CSF ratio t-tau / (p-tau181 x t-PrP), with the correct classification reaching 98.4 percent of the patients.

    This study provided evidence that determination of t-PrP in CSF as a new biomarker may help clinicians to exclude CJD in patients with unusual AD phenotype, for whom CJD may be a differential diagnosis, while currently available non-specific biomarkers such as 14-3-3 protein or t-tau diagnostically do not contribute enough. The clinical benefits of using CSR t-PrP in addition to current classic biomarkers should be confirmed by further studies of larger cohorts of confirmed CJD and (atypical) AD patients. 

    View all comments by Kurt A. Jellinger

Make a Comment

To make a comment you must login or register.


News Citations

  1. New Biomarkers? Synaptic Proteins in Spinal Fluid Predict Cognitive Decline
  2. Test Uses 'Seeding' to Detect Aβ Oligomers in Cerebrospinal Fluid
  3. A Blood Test for Preclinical Prion Disease, But Will It Be Used?
  4. Research Brief: Does Alzheimer’s Progress Fast in Some People?
  5. Mistaken Identity—Prion Disease or Alzheimer’s on Fast Forward?

Paper Citations

  1. . Updated clinical diagnostic criteria for sporadic Creutzfeldt-Jakob disease. Brain. 2009 Oct;132(Pt 10):2659-68. Epub 2009 Sep 22 PubMed.
  2. . The 14-3-3 brain protein in cerebrospinal fluid as a marker for transmissible spongiform encephalopathies. N Engl J Med. 1996 Sep 26;335(13):924-30. PubMed.
  3. . Phospho-tau/total tau ratio in cerebrospinal fluid discriminates Creutzfeldt-Jakob disease from other dementias. Mol Psychiatry. 2003 Mar;8(3):343-7. PubMed.
  4. . Diagnostic value of 14-3-3beta immunoblot and T-tau/P-tau ratio in clinically suspected Creutzfeldt-Jakob disease. Int J Mol Med. 2005 Dec;16(6):1147-9. PubMed.
  5. . Genetic Creutzfeldt-Jakob disease associated with the E200K mutation: characterization of a complex proteinopathy. Acta Neuropathol. 2011 Jan;121(1):39-57. Epub 2010 Jul 1 PubMed.

Further Reading

Primary Papers

  1. . Association of cerebrospinal fluid prion protein levels and the distinction between Alzheimer disease and Creutzfeldt-Jakob disease. JAMA Neurol. 2015 Mar;72(3):267-75. PubMed.
  2. . Cerebrospinal fluid total prion protein: a potential in vivo marker of cerebral prion pathology. JAMA Neurol. 2015 Mar;72(3):261-3. PubMed.