Diagnosing sporadic Creutzfeldt-Jakob (sCJD) disease continues to be a very difficult task given the relative rarity of the disorder, the rapid progression of the disease, and the lack of prospective studies on potential biomarkers. A new evidence-based guideline by the AAN Guideline Development Subcommittee, led by Taim Muayqil, MBBS, of the King Saud University in Riyadh, Saudi Arabia, has concluded that testing for 14–3-3 protein in cerebrospinal spinal fluid (CSF) with available assays can be “moderately” helpful for some patients.
The incidence of sCJD is approximately one case per one million persons. Patients experience rapidly advancing dementia that usually results in death in just six months. Because symptoms can include cognitive, emotional, sleep, motor, balance, and sensory disturbances, inconsistent clinical presentations can delay diagnosis and social and therapeutic interventions when there is little time to spare.
Neurology Today asked the AAN's evidence-based medicine methodologist Gary Gronseth, MD, professor and vice chair of neurology at the University of Kansas Hospital in Kansas City, to discuss the new guideline in a telephone interview with Neurology Today.
HOW IS SPORADIC CJD CURRENTLY DIAGNOSED?
Because positive confirmation can only be made through a brain autopsy or, sometimes, a biopsy, we need better tools to identify it early. While neuroimaging, especially diffusion weighted imaging (DWI), can detect changes in the brain that are characteristic of the disease, having a reliable blood or spinal fluid biomarker would be a great help.
WHY WAS A GUIDELINE NECESSARY?
Initially, studies showed that CSF 14–3-3 was very sensitive, but this sensitivity has been disputed and newer research has shown variations in sensitivity, depending on the duration of illness, CJD subtype, and different laboratory screening techniques. Enzyme-linked immunosorbent assay and Western Blot techniques are the most common laboratory tests for detecting CSF 14–3-3.
TELL US ABOUT THE PANEL's REVIEW OF THE LITERATURE.
The panel systematically reviewed available studies from 1995 through 2010 to identify research involving patients suspected of having sCJD who underwent CSF 14–3-3 testing. Nine Class II studies involving 1,849 patients formed the basis for the guideline, most comprising cohort studies with mixed retrospective and prospective data collection. No Class I studies were found.
WHAT DID YOU FIND ABOUT THE SENSITIVITY AND SPECIFICITY OF TESTS?
We would like to have found that the test was 100 percent sensitive and specific, but tests with that accuracy are rare. We found that both Western Blot and ELISA are probably moderately accurate, with an overall sensitivity of 92 percent and 80 percent specificity for sCJD, with a positive ratio of 4.7 and negative likelihood ratio of 0.10. That said, the CSF 14–3-3 assay is an imperfect test because it lacks the diagnostic accuracy to either rule-in or rule-out CJD diagnosis in most cases. Our analysis found that despite varying case definitions, the sensitivity of CSF 14–3-3 remained relatively stable, while the specificity was a bit more variable. We concluded that the estimate of the sensitivity of CSF 14–3-3 assays in our review is more reliable than the specificity.
It is important to note that there is significant variation in the way the CSF 14–3-3 assays are performed by laboratories. Earlier, most studies relied on the Western Blot test, which is subjective and interpreted qualitatively, while many later studies used the quantitative ELISA technique, where sensitivity and specificity may vary according to the laboratory's determined cutoff.
WHAT WERE THE MAIN CONCLUSIONS?
We found that CSF14–3-3 can be useful in helping to distinguish sCJD from other rapidly progressive dementing disorders in the right circumstances. It is important to realize that the usefulness of the test, given its moderate diagnostic accuracy, will be highly dependent on the clinician's judgment as to the likelihood that the patient actually has CJD.
For patients with a low likelihood of sCJD before the test, for example if they have had symptoms for more than a year, the test is almost useless. A positive result in this circumstance would most likely represent a false positive.
However if a patient has rapidly progressive dementia and is strongly suspected of having CJD — with a pretest probability of approximately 20 percent to 90 percent — clinicians should order a CSF 14–3-3 assay because test results will importantly reduce uncertainty.
Because most patients with sCJD die within six months of dementia onset, testing nine months after the onset of symptoms is probably unnecessary. Most patients with non-CJD dementia fall into that category. But for rapidly progressing dementia of five months or less, testing is potentially useful.
Clinicians' judgments regarding whether or not to order a CSF 14–3-3 assay should consider the rarity of the disease, the practice setting (community hospital versus tertiary referral center), the patient's clinical presentation, and the results of already obtained ancillary tests such MRI of the head.
WHAT RECOMMENDATIONS DID THE PANEL MAKE REGARDING FUTURE RESEARCH DIRECTIONS?
We made several suggestions, including the need for prospective studies with good follow-up, as well as standardization of the definition of a positive CSF 14–3-3 result and validation in well-designed cohort studies. These would be very useful. Other issues that need to be resolved include which specific isoform of 14–3-3 (beta or gamma) is most useful for screening purposes, as well as standardization of how laboratories use and interpret ELISA and Western Blot results.
There is also a need for studies comparing the use of other biomarkers, or a combination of multiple biomarkers, in addition or in lieu of CSF 14–3-3. These include total tau, p-tau, S-100 protein, or neuron-specific enolase (NSE) in the CSF.
Also helpful would be more research on the usefulness of MRI, including DWI and fluid attenuated inversion recovery (FLAIR) sequences, as well as further study of the utility of CSF biomarkers in subgroups of patients based on such factors as patient demographics, time of presentation, duration of illness, specific etiologies of prion disease, and genetic factors.