Cerebrospinal fluid real‐time quaking‐induced conversion is a robust and reliable test for sporadic creutzfeldt–jakob disease: An international study

Real‐time quaking‐induced conversion (RT‐QuIC) has been proposed as a sensitive diagnostic test for sporadic Creutzfeldt–Jakob disease; however, before this assay can be introduced into clinical practice, its reliability and reproducibility need to be demonstrated. Two international ring trials were undertaken in which a set of 25 cerebrospinal fluid samples were analyzed by a total of 11 different centers using a range of recombinant prion protein substrates and instrumentation. The results show almost complete concordance between the centers and demonstrate that RT‐QuIC is a suitably reliable and robust technique for clinical practice. Ann Neurol 2016;80:160–165

Real-time quaking-induced conversion (RT-QuIC) has been proposed as a sensitive diagnostic test for sporadic Creutzfeldt-Jakob disease; however, before this assay can be introduced into clinical practice, its reliability and reproducibility need to be demonstrated. Two international ring trials were undertaken in which a set of 25 cerebrospinal fluid samples were analyzed by a total of 11 different centers using a range of recombinant prion protein substrates and instrumentation. The results show almost complete concordance between the centers and demonstrate that RT-QuIC is a suitably reliable and robust technique for clinical practice. ANN NEUROL 2016;80:160-165 C reutzfeldt-Jakob disease (CJD) belongs to a family of fatal neurodegenerative diseases known as transmissible spongiform encephalopathies (TSEs). TSEs are characterized by the post-translational conformational change of a normally expressed protein called prion protein (PrP) into a disease-associated form known as PrP Sc . Once formed, PrP Sc can induce PrP to undergo a conformational change and produce more PrP Sc in a selfpropagating manner. The PrP Sc aggregates, becomes protease resistant, and deposits throughout the brain, leading to spongiform change and neuronal loss.
Patients with sporadic CJD (sCJD) present with a rapidly progressing dementia, and death usually occurs within 6 months. Current diagnostic criteria for sCJD rely on clinical features, the results of electroencephalography and magnetic resonance imaging, and the presence of 14-3-3 protein in the cerebrospinal fluid (CSF). 1,2 These tests are not specific for CJD, and none is able to detect all forms of CJD. 3,4 A new approach to the premortem diagnosis of sCJD has been to exploit the ability of small amounts of CSF PrP Sc to convert native PrP into PrP Sc in a newly described protein aggregation assay known as real-time quaking-induced conversion (RT-QuIC). This technique uses recombinant PrP (rPrP) as a substrate, which is induced to aggregate by the addition of CSF containing PrP Sc . Thioflavin T (ThT) in the reaction binds to the aggregated PrP Sc , causing a change in the ThT emission spectrum, enabling the reaction to be monitored in real time. 5,6 CSF RT-QuIC has been shown to be an accurate diagnostic test for sCJD, with a high degree of sensitivity (85-87%) and specificity (99-100%). 7,8 An increasing number of laboratories have established RT-QuIC analysis, and more are interested in doing so. Before RT-QuIC is fully accepted into clinical practice as a reliable premortem diagnostic test, studies are required to ensure that laboratories performing this assay using different forms of recombinant PrP as a substrate and a variety of instrumentation are producing comparable results.
This study reports the findings of 2 international ring trials that were undertaken over a 2-year period. The initial ring trial comprised European participants, whereas the second ring trial was wider and included participants from Australia, Canada, and Japan.

CSF Samples
CSF samples were provided by the National CJD Research & Surveillance Unit, United Kingdom. These samples are stored at &minus;80 8C, and consent was obtained from the next of kin for their use in research (05/MRE00/67). CSF samples were selected on the basis of having appropriate ethical consent and sufficient volume to ensure that each participant had an adequate volume of CSF for analysis. Each set of CSF samples was sent to each laboratory on dry ice and was analyzed blind to the final diagnosis. The performance of the RT-QuIC assay was evaluated in an interlaboratory ring trial, where identical CSF samples were analyzed by each of the participating laboratories. Using this approach, it may be possible to identify analytical procedures that affect overall assay performance.

First Ring Trial
Seven laboratories (UK, It-Rome, It-Mil, It-Bol, Fra, Ger-Mun, Bel) participated in the first ring trial, and each received 10 CSF samples sent on dry ice. Of these, 1 was from a patient with neuropathologically confirmed sCJD, 4 were from patients with probable sCJD, diagnosed according to the World Health Organization criteria, 2 patients improved and the diagnosis of sCJD was excluded on clinical grounds, 1 patient had a steroidresponsive encephalopathy, 1 had no neuropathological evidence of CJD at postmortem, and 1 had mixed Alzheimer disease and vascular dementia at postmortem. The age of the patients ranged from 48 to 86 years and included 5 females and 5 males. The disease duration of the patients with sCJD ranged from 2 to 12 months, and the CSF samples were taken between 53% and 94% of the disease duration.

Second Ring Trial
Eleven laboratories (UK, It-Rome, It-Mil, It-Bol, Fra, Ger-Mun, Ger-Goet, Jpn, Aust, Can, and Bel) participated in the second ring trial, and each received 15 CSF samples sent on dry ice. One laboratory analyzed the CSF samples twice using a BMG LABTECH (Ortenberg, Germany) Optima and a BMG LABTECH Omega. Of the 15 CSF samples, 5 were from patients with probable sCJD, 3 were from patients with neuropathologically confirmed sCJD, 1 patient improved and the diagnosis of sCJD was excluded on clinical grounds, 1 had seizures, 1 had anti-immune encephalopathy, 1 had a psychiatric disorder, 1 had Huntington disease, 1 had a non-CJD dementia, and 1 had normal pressure hydrocephalus. The age of the patients ranged from 55 to 87 years and included 6 females and 9 males. The disease duration of the patients with sCJD ranged from 1 to 26 months and the CSF samples were taken between 57% and 90% of the disease duration.

Results
The results of the ring trials are given in Tables 2 and 3. In the first ring trial 6, of 7 laboratories obtained positive RT-QuIC responses in the CSF samples from all 5 sCJD cases; the remaining laboratory obtained positive RT-QuIC responses in the CSF of 4 of the 5 sCJD cases. A negative RT-QuIC result was obtained in the CSF of 1 sCJD case, which had a disease duration of 12 months. This laboratory was using a BMG LABTECH Optima instrument, with full-length hamster rPrP as substrate and 30 ml of CSF. These conditions were similar to other laboratories in the study that obtained positive results for this particular CSF sample. A limited volume of CSF was sent to each laboratory, and this meant that repeating the analysis of this particular CSF by the individual laboratory was not possible.
None of the laboratories obtained positive RT-QuIC responses in any of the 5 CSF samples from patients without sCJD. Four of these laboratories used a BMG LABTECH Optima, and the remaining 3 used a BMG LABTECH Omega. However, all the laboratories used the same full-length hamster rPrP.
The second ring trial included a larger number of laboratories with a wider geographical distribution and included a wider range of instrumentation and type of rPrP. Despite this, the results showed complete concordance (see Table 3). Of the 11 laboratories that participated in this ring trial, 11 obtained positive RT-QuIC responses in all 8 CSF samples from sCJD patients. All 11 laboratories obtained a positive RT-QuIC response in the CSF from the sCJD patient, with a disease duration of 26 months. This CSF sample was taken 21 months after the onset of symptoms. The laboratory that failed to obtain a positive CSF RT-QuIC response in an sCJD case in the first ring trial correctly identified all sCJD cases in the second ring trial. None of the analytical parameters had been changed by the laboratory in question. Importantly, the laboratory that used a Tecan Infinite shaker with a human rPrP as a substrate also correctly identified all sCJD cases. One laboratory analyzed the ring trial CSF samples using both the BMG LAB-TECH Optima or a BMG LABTECH Omega and obtained similar results using both instruments despite having different cutoff criteria. None of the laboratories obtained positive RT-QuIC responses in CSF samples from patients with non-CJD disorders.

Discussion
In the first ring trial, 6 of 7 laboratories correctly identified all sCJD cases; however, 1 laboratory obtained a negative CSF RT-QuIC result from an sCJD patient with a disease duration of 12 months. It has been reported that CSF samples from sCJD patients with longer disease durations may have lower seeding efficiency 10 ; however, this effect was not seen in the second ring trial, where the CSF from an sCJD patient with a disease duration of 26 months was identified by RT-QuIC by all laboratories. All the laboratories that participated in the first ring trial used the same source and type of rPrP and either Optima or Omega BMG LABTECH instrumentation, and demonstrated high levels of accuracy and agreement between laboratories. The second ring trial took place 18 months later and was expanded to include other European laboratories and participants from Australia, Canada, and Japan. Some of the additional laboratories used alternative types of rPrP and other forms of instrumentation. This enabled a more rigorous assessment of the robustness and transferability of the RT-QuIC technique.
All 11 laboratories accurately identified the 8 CSF samples from sCJD patients, and none detected a positive RT-QuIC in any of the non-CJD cases. The agreement between laboratories using different rPrP as a substrate and different forms of instrumentation is encouraging. The accuracy of the results obtained was identical for each of the rPrP substrates used. From this limited number of CSF samples, we have achieved an overall sensitivity of between 85.7% and 100% and a specificity of 100%. This compares well with a previous intralaboratory study, which had fewer participants and reported a sensitivity of 85% and a specificity of 99%. 10 The complete concordance between laboratories demonstrates that CSF RT-QuIC is adaptable to different laboratory instrumentation and different types of rPrP. The high level of accuracy and agreement between laboratories using CSF RT-QuIC is supportive of this technique being introduced into clinical practice.