Evaluation of a New Criterion for Detecting Prion Disease With Diffusion Magnetic Resonance Imaging

Key Points

Question  Is diffusion magnetic resonance imaging (MRI) using a new criterion of at least 1 positive brain region for diagnosis of sporadic Creutzfeldt-Jakob disease (sCJD) more accurate than currently used criteria and 3 commonly used cerebrospinal fluid tests?

Findings  In this diagnostic study of 1387 patients with suspected sCJD, the new criterion increased the sensitivity of MRI for autopsy-confirmed sCJD compared with current applied criteria. The diagnostic performance of diffusion MRI was comparable with the improved version of the real-time quaking-induced conversion test in cerebrospinal fluid.

Meaning  The findings suggest that diffusion MRI is an accurate test for establishing a diagnosis of prion diseases in the appropriate clinical context.

Importance  Early diagnosis is a requirement for future treatment of prion diseases. Magnetic resonance imaging (MRI) with diffusion-weighted images and improved real-time quaking-induced conversion (RT-QuIC) in cerebrospinal fluid (CSF) have emerged as reliable tests.

Objectives  To assess the sensitivity and specificity of diffusion MRI for the diagnosis of sporadic Creutzfeldt-Jakob disease (sCJD) with a new criterion (index test) of at least 1 positive brain region among the cortex of the frontal, parietal, temporal, and occipital lobes; the caudate; the putamen; and the thalamus.

Design, Setting, and Participants  This diagnostic study with a prospective and a retrospective arm was performed from January 1, 2003, to October 31, 2018. MRIs were collected from 1387 patients with suspected sCJD consecutively referred to the National Prion Disease Pathology Surveillance Center as part of a consultation service.

Intervention  Magnetic resonance imaging. Four neuroradiologists blinded to the diagnosis scored the MRIs of 200 randomly selected patients. One neuroradiologist scored the MRIs of all patients.

Main Outcomes and Measures  Sensitivity and specificity of the index test compared with currently used criteria and CSF diagnostic (improved RT-QuIC, 14-3-3, and tau CSF tests).

Results  A total of 872 patients matched the inclusion criteria (diffusion MRI and autopsy-confirmed diagnosis), with 619 having sCJD, 102 having other prion diseases, and 151 having nonprion disease. The primary analysis included 200 patients (mean [SD] age, 63.6 [12.9] years; 100 [50.0%] male). Sensitivity of the index test of 4 neuroradiologists was 90% to 95% and superior to sensitivity of current MRI criteria (69%-76%), whereas specificity was 90% to 100% and unchanged. Interrater reliability of the 4 neuroradiologists was high (κ = 0.81), and individual intrarater reliability was excellent (κ ≥0.87). The sensitivity of the index test of 1 neuroradiologist for 770 patients was 92.1% (95% CI, 89.7%-94.1%) and the specificity was 97.4% (95% CI, 93.4%-99.3%) compared with a sensitivity of 69.8% (95% CI, 66.0%-73.4%; P < .001) and a specificity of 98.0% (95% CI, 94.3%-99.6%; P > .99) according to the current criteria. For 88 patients, index test sensitivity (94.9%; 95% CI, 87.5%-98.6%) and specificity (100%; 95% CI, 66.4%-100%) were similar to those of improved RT-QuIC (86.1% [95% CI, 76.5%-92.8%] and 100% [95% CI, 66.4%-100%], respectively). Lower specificities were found for 14-3-3 and tau CSF tests in 452 patients.

Conclusions and Relevance  In this study, the diagnostic performance of diffusion MRI with the new criterion was superior to that of current standard criteria and similar to that of improved RT-QuIC. These results may have important clinical implications because MRI is noninvasive and typically prescribed at disease presentation.

Considerable progress has been made in the diagnosis of sporadic Creutzfeldt-Jakob disease (sCJD)^1-5 using magnetic resonance imaging (MRI).^6-13 Initially, CJD diagnostic criteria comprised only the presence of abnormal signal in the caudate and putamen because of the low sensitivity of conventional MRI sequences in detecting cortical abnormalities.^14,15 Subsequently, Zerr et al¹⁶ updated sCJD diagnostic criteria, adding MRI signal hyperintensities in the cortex of at least 2 lobes, excluding the frontal lobes on diffusion-weighted imaging (DWI) and fluid-attenuated inversion recovery (FLAIR). These criteria were adopted for national surveillance purposes by many countries, including the US; they are referred to in this article as the current standard.¹⁷ However, these criteria might be too conservative, missing the diagnosis of patients with frontal cortex and/or only 1 lobe involvement or with only the caudate or putamen affected.

Progress has also been made with cerebrospinal fluid (CSF) tests. Satisfactory sensitivity (70%-85%) but low specificities (40%-74%) have been reported for 14-3-3 and tau.^18,19 Recently, an improved version of real-time quaking-induced conversion^20,21 (RT-QuIC, also known as IQ-CSF or second-generation RT-QuIC) showed a remarkable increase in sensitivity with no loss of specificity, reaching a sensitivity of 95% and a specificity of 100%.^22-24 This study assessed the sensitivity and specificity of diffusion MRI for the diagnosis of sCJD with a new criterion (index test) of at least 1 positive brain region among the cortex of the frontal, parietal, temporal, and occipital lobes; the caudate; the putamen; and the thalamus and compared the index test’s performance with that of currently used criteria. We also compared diffusion MRI diagnostic performance with CSF test diagnostic performance.

This diagnostic study used MRIs from 1387 patients with suspected sCJD that were consecutively collected by the National Prion Disease Pathology Surveillance Center (NPDPSC) in Cleveland, Ohio, from January 1, 2003, to October 31, 2018, following prospective and retrospective approaches. The prospective approach included living patients whose MRIs were collected through an MRI consultation service starting from 2009. The retrospective approach included the MRI retrieval of autopsied individuals referred to the NPDPSC. Inclusion criteria were MRI study with DWI and apparent diffusion coefficient (ADC) maps available and autopsy diagnosis made at the NPDPSC. Clinical records of patients without autopsy performed at the NPDPSC were not available. Written informed consent was provided by the patient or legal next of kin for autopsy through the Centers for Disease Control and Prevention–funded NPDPSC Autopsy Program. All imaging and laboratory data were anonymized. This study was approved by the University Hospitals Institutional Review Board, and informed consent was waived by the institutional review board on deceased individuals for the purpose of this study.

Brain examination was performed by histopathologic analysis, including prion protein immunohistochemistry and Western blot, and subtype diagnosis of prion disease was established or excluded.^2,3,25 Details on the procedures for the identification of pure and mixed sCJD subtypes are reported in the eMethods in the Supplement.

CSF tests of IQ-CSF, 14-3-3, and total tau were performed and reported by the NPDPSC as previously described.^18,26 Diagnostic reliability analysis of 14-3-3 did not include patients with indeterminate results, and the total tau test result was considered positive for values greater than 1150 pg/mL.

Diffusion MRI lesion profiles were generated for each patient by grading the signal hyperintensities of 12 brain regions on a 4-point ordinal scale: 0 for no hyperintensity, 1 for questionable, 2 for sCJD-related hyperintensity associated with low diffusivity on ADC maps, and 3 for presence of extensive sCJD-related hyperintensity with low diffusivity in that specific region. FLAIR images were systematically inspected before DWI in all patients. Both hemispheres were inspected, and the highest score was recorded for each brain region. Symmetry or asymmetry of the signal hyperintensities also was recorded. The score was assigned to 5 cortical regions (frontal and parietal, including the precuneus, temporal, and occipital lobes), the caudate, the putamen, the thalamus, limbic structures (cingulate, insula, and hippocampus), and the cerebellum. MRI results were considered positive when at least 1 of 8 regions, excluding 3 limbic structures and cerebellum, was scored 2 or higher. The cingulate, insula, hippocampus, and cerebellum were excluded because they can be spontaneously hyperintense on DWI of healthy individuals.^27,28 Diagnostic values of the proposed criterion (index test) were computed considering neuropathologic findings as the reference standard and were compared with current standard criteria: involvement of both the caudate and the putamen or at least 2 cortical regions among parietal, temporal, and occipital lobes.^16,17

Three neuroradiologists each with 15 years of experience (A.B., M.G., and R.L.) and 1 neuroradiologist with 1 year of experience (M.E.M.M.) blinded to clinical data and diagnosis scored the diffusion MRIs of 200 individuals with autopsy-confirmed diagnosis. The interrater reliability (IRR) among the neuroradiologists was measured. The intrarater reliability for each neuroradiologist was measured in a subset of MRI studies from 60 individuals that each neuroradiologist read with an interval of at least 1 year. One neuroradiologist (A.B.) read and scored all MRIs in electronic format.

The IRR analysis was performed for 200 individuals (150 with sCJD and 50 with nonprion diseases) randomly selected from the 619 individuals with sCJD by stratified sampling (strata were disease subtypes) and from 151 individuals with nonprion disease by simple random sampling. The Fleiss κ and intraclass correlation coefficients (ICCs) were used to measure interrater agreement in the diagnosis and in scoring each region, respectively. Intrarater analysis was assessed with the Cohen κ coefficient and percentage agreement in a subsample of 60 individuals (30 with sCJD and 30 with nonprion diseases) selected from the 200 individuals through the same sampling procedure described above. Further details on these statistical analyses are reported in the eMethods in the Supplement.

The McNemar test was used to compare diagnostic parameters (ie, accuracy, sensitivity, and specificity) of the proposed and current diffusion MRI criteria for each neuroradiologist from the subsample of 200 individuals and for 1 neuroradiologist (A.B.) from all 619 individuals with sCJD and 151 individuals with nonprion disease. Sensitivities of these criteria were also compared in 102 individuals with other prion diseases (A.B.). Diagnostic parameters of diffusion MRI scored by this neuroradiologist (A.B.) were compared with those of IQ-CSF, 14-3-3, and total tau tests performed on the same individuals using the McNemar test.

The Fisher exact test was used to compare sensitivity values of the same diagnostic test between sCJD subtypes. Statistical significance level was set at a 2-sided P < .05. Statistical analyses were performed using R software, version 3.6.0 (R Foundation for Statistical Computing).

Of the 1387 patients, 872 had a DWI sequence available and autopsy diagnosis performed at the NPDPSC. The following diagnoses were made: sCJD (619 cases), other prion diseases (102 cases), and nonprion diseases (151 cases). Patients with other prion diseases included familial CJD, variably protease-sensitive prionopathy, sporadic fatal insomnia, Gerstmann-Sträussler-Scheinker disease, fatal familial insomnia, and variant CJD. The diagnostic performance of 4 neuroradiologists applying the proposed and current MRI criteria was computed in a subsample of 200 individuals (150 with sCJD and 50 with nonprion diseases) (mean [SD] age, 63.6 [12.9] years; 100 [50.0%] male). The flowchart of case selection and patient demographics is given in Table 1 and the eFigure and eTable 1 in the Supplement. Comparison between demographic data of the 200 individuals with those of the larger cohort is given in eTable 2 in the Supplement.

Sensitivity values of 3 of the neuroradiologists (A.B., M.G., and R.L.) in 200 individuals using the proposed criteria were 92%, 95%, and 94%, and corresponding specificities were 100%, 90%, and 96%; sensitivity and specificity of the other neuroradiologist (M.E.M.M.) were 90% and 96%, respectively (Table 2). Sensitivities were significantly greater than those obtained using current criteria (new criterion vs current criteria: 94.7% vs 76.0% for neuroradiologist 1, 94.0% vs 73.3% for neuroradiologist 2, 90.0% vs 69.3% for neuroradiologist 3, and 92.0% vs 74.0% for neuroradiologist 4; all P < .001), whereas specificities were unchanged. Sensitivity and specificity were not significantly different between the neuroradiologists (sensitivities: 94.7% for neuroradiologist 1, 94.0% for neuroradiologist 2, 90.0% for neuroradiologist 3, and 92.0% for neuroradiologist 4; specificities: 90.0% for neuroradiologist 1, 96.0% for neuroradiologist 2, 96.0% for neuroradiologist 3, and 100.0% for neuroradiologist 4). Thus, neuroradiologist experience did not affect use of new diffusion MRI criteria.

Interrater agreement in the diagnosis was good (mean [SE] κ = 0.814 [0.041]) with the proposed criteria, and it was significantly higher (P < .001) when compared with that obtained by using the current criteria (mean [SE] κ = 0.646 [0.041]). The agreement in the scores was excellent, with ICCs between 0.86 and 0.93 in 6 of 12 brain regions. It was good for temporal cortex (ICC, 0.83), cingulate and insula (ICC, 0.81), occipital cortex (ICC, 0.77), and cerebellum (ICC, 0.70) and fair for the hippocampus (ICC, 0.58) (eTable 3 in the Supplement). Intrarater reliability in 60 individuals was excellent for all neuroradiologists; for each of the neuroradiologists, κ was 1.0, 0.90, 0.93, and 0.87, with percentage agreements of 100%, 95%, 97%, and 93%.

The performance of the index test in 770 individuals (619 with sCJD and 151 with nonprion disease) scored by 1 neuroradiologist (A.B.) is reported in eTable 4 in the Supplement. Accuracy of diffusion MRI applying the new criterion was 93.1% (95% CI, 91.1%-94.8%) and sensitivity was 92.1% (95% CI, 89.7%-94.1%), and these values were significantly higher than applying the currently used criteria¹⁶ (75.3% [95% CI, 72.1%-78.3%] and 69.8% [432 of 619; 95% CI, 66.0%-73.4%]; P < .001). Specificity values were not significantly different between criteria (97.4% [95% CI, 93.4%-99.3%] and 98.0% [95% CI, 94.3%-99.6%]; P > .99). These results are equivalent to those obtained by the same neuroradiologist (A.B.) in the subsample of 200 individuals. The new criterion correctly identified a significantly higher number of patients with sCJD with respect to current criteria in VV1, VV2, MM1, and MM2 pure subtypes, as well as for MM1 + 2 mixed form (eTable 5 in the Supplement). Other pure and mixed subtypes did not show any significant difference.

An accurate positive MRI diagnosis was made in 138 patients who would have had a false-negative MRI result applying the current criteria: 53 with only 1 positive region, 61 with 2 positive regions, and 24 patients with 3 positive regions. In particular, patients with the following combinations of diffusion MRI signal abnormalities would have missed the diagnosis with the current criteria: 27 patients with abnormality only in the parietal lobes, 12 patients with abnormality only in the caudate, 42 patients with abnormalities in the parietal and frontal cortices, 9 patients with abnormalities in the caudate and thalami, and 19 patients with abnormalities in the parietal and frontal cortices and caudate. Five demonstrative cases are shown in Figure 1 and Figure 2.

A follow-up diffusion MRI was available in 24 of 49 patients with initially false-negative results, and the result was positive in 16 of 24 (66.7%). Accuracy and sensitivity increased to 95% when all MRI studies were considered. Diagnostic sensitivity for pure and mixed sCJD subtypes ranged from 83% to 100% for the new criterion with no statistically significant differences between subtypes (93% for MM1, 95% for MM2, 93% for MV1, 100% for MV2C, 92% for MV2K, 90% for VV1, 84% for VV2, 96% for MM1 + 2, 92% for MV1 + 2C, 83% for MV1 + 2K, 100% for MV1 + 2K + C, 92% for MV2K + C, 67% for VV1 + 2, and 100% for not otherwise specified; P = .09) (eTable 5 in the Supplement). Sensitivity of MRI in 102 patients with other prion diseases was 63.7% (65 of 102) (eTable 6 in the Supplement). Diffusion-weighted imaging signal hyperintensities were rated more prominent than FLAIR in 490 of 536 patients (91.4%), equal in 43 of 536 patients (8.0%), and less prominent in 3 of 536 patients (0.6%).

Diagnostic parameters of diffusion MRI provided by the neuroradiologist who scored all 770 images were compared with the results of CSF tests obtained from the same patients during the same visit. Data on both diffusion MRI and IQ-CSF were available in 79 patients with sCJD and 9 with nonprion disease. Diagnostic accuracy was 95%, sensitivity was 95%, and specificity was 100% for MRI, and diagnostic accuracy was 88%, sensitivity was 86%, and specificity was 100% for IQ-CSF, with no significant differences in sensitivity (P = .12) and perfect agreement for specificity (Table 3). No statistically significant differences were observed among sCJD subtypes (eTable 7 in the Supplement). Sensitivity was 100% when both MRI and IQ-CSF results were considered.

New MRI criterion and CSF total tau and 14-3-3 proteins were compared in 452 patients (391 patients with sCJD and 61 with nonprion disease) who had undergone all 3 tests. In this cohort, diffusion MRI achieved 93% accuracy, 92% sensitivity, and 100% specificity, surpassing both total tau (83% accuracy, 87% sensitivity, and 61% specificity) and 14-3-3 (84% accuracy, 92% sensitivity, and 23% specificity) (Table 3). Diffusion MRI sensitivity (92% vs 87%, P = .03) and specificity (100% vs 61%, P < .001) significantly exceeded those of total tau, whereas only specificity was significantly improved (100% vs 23%, P < .001) compared with 14-3-3. However, the 14-3-3 test was nondiagnostic in 92 of 452 patients (20.4%).

To our knowledge, this study was the largest diffusion MRI study in individuals with definitive prion disease.^1,16,29-31 With the development of new potential therapies to alter the natural history of prion disease, early diagnosis is essential to maximize the chance of salvaging brain tissue.³² It is important that practitioners are aware of the diagnostic potential of diffusion MRI because it is the most widely available diagnostic test. In a cohort of 200 patients with autopsy-confirmed diagnoses scored independently by 4 neuroradiologists, we found that the results of newly proposed diffusion MRI criterion matched or exceeded values previously reported for currently used MRI sCJD diagnostic criteria.^1,8,9 Moreover, sensitivity values of the 4 neuroradiologists applying the current criteria were similar to those obtained by Zerr et al¹⁶ in 233 individuals DWI. We found that the use of our at least 1 region criterion significantly improved the sensitivity, especially for patients with sCJD with MM and VV genotypes, whereas specificity remained unchanged. Of note, similar diagnostic performances were achieved by the 4 neuroradiologists regardless of their years of experience. Diagnostic sensitivity reached 100% when both MRI and IQ-CSF were considered, suggesting that the 2 tests are complementary, and they should be used together in the diagnostic workup of patients with suspected sCJD.

The new criterion is simple, robust, and practical to use. On the contrary, the requirement of 2 or more positive brain regions by current standard criteria is likely to limit the detection of positive cases more so at early stages of the disease. This study showed that in the appropriate clinical context the probable diagnosis of sCJD should be raised with MRI in patients presenting with a single positive region showing DWI abnormality in the parietal cortex or in the caudate and in patients with 2 positive regions in the cortical ribbon of the parietal and frontal lobes. The sensitivity of MRI with inclusion also of limbic structures and cerebellum was even higher than with 8 brain regions. Notwithstanding, we excluded limbic structures and cerebellum to avoid false-positive readings by less experienced neuroradiologists. Furthermore, our study supported the superiority of the DWI over FLAIR, and the findings suggest that DWI should be acquired for all patients with rapidly progressive encephalopathy.^27,33,34

Few MRI studies^16,35,36 have analyzed individual sCJD subtypes; moreover, they examined only the main subtypes because of the small sample size of the autopsy-confirmed subtypes. Our study found high diagnostic performance of diffusion MRI in all subtypes, including mixed subtypes and, for the first time to our knowledge, the recently distinguished MV2C and MV2K. The excellent sensitivity of our criterion for almost all sCJD subtypes may inform in vivo identification of sCJD subtypes by MRI, which is the focus of a forthcoming study.

MRI sensitivity in other prion diseases characterized by spongiform changes (familial CJD, sporadic fatal insomnia, and variant CJD) was above 75%. As expected, sensitivity was lower in Gerstmann-Sträussler-Scheinker disease, variably protease-sensitive prionopathy, and fatal familial insomnia, which rarely have vacuoles on histopathologic analysis. These results suggest that DWI signal hyperintensity may be associated with the spongiform change rather than other histopathologic hallmarks of prion diseases and confirm the results of previous studies.^13,34

The topography of the diffusion signal abnormality is heterogeneous in sCJD because it reflects the multiple configurations in which key brain regions are affected.^36-38 Furthermore, the involvement of the cerebral cortex is often asymmetrical. This heterogeneity of signal abnormalities at presentation, which is mostly associated with the distinct subtypes, may be a confounder in the interpretation of the findings by the neuroradiologists.³⁶ Despite the good diagnostic values reported in the literature, clinical use of MRI may be hampered by the rarity of the disease and variations among sCJD subtypes. Diffusion MRI signal hyperintensity was often not correctly interpreted on the initial MRI at referring hospitals in a study of 103 patients with CJD.³⁹ Retrospective reading of the original MRI reports found that the diffusion MRI signal abnormalities were rarely missed. On the contrary, often the neuroradiologist failed to raise the possibility of sCJD diagnosis because these findings were usually considered as an improbable presentation of a common disease rather than a typical presentation of a rare one. The most frequent sCJD misdiagnoses on MRI were acute cortical ischemic infarct, venous thrombosis, encephalitis, or metabolic disorders.³⁹ All the above diseases are typically associated with diffusion MRI signal abnormalities and clinical signs that are different from those found in sCJD. Nevertheless, any diagnostic test must be interpreted in the appropriate clinical context, which includes ruling out treatable illnesses and potential mimickers. When a patient presents with signs of rapidly progressive encephalopathy and sCJD is suspected, even if the initial MRI findings do not suggest the diagnosis, review of the study by an experienced neuroradiologist should be encouraged to facilitate early diagnosis. The MRI consultation program was started at the NPDPSC to address this need.

A remarkable finding of our study was that diffusion MRI diagnostic parameters were comparable to those of the IQ-CSF. As expected, diffusion MRI exceeded the diagnostic values of older CSF tests, such as total tau and 14-3-3. These findings were not affected by patient population, including whether the procedures were performed on the same patients, as we did in this study, or by comparing data from separate cohorts, as reported previously.^18,26 The IQ-CSF sensitivity for all sCJD cases (86%) was lower than the sensitivity values reported in the literature (92%-95%).^24,26 A possible contributing factor to this discrepancy was the different representation of individual subtypes in this sCJD cohort compared with previous studies.^24,26,40

IQ-CSF sensitivities differed by sCJD subtype. Although sCJD MM1 and VV2, the 2 most common subtypes, have similar sensitivities, MM2 sensitivity was 100% for diffusion MRI and 67% in this study and 80% in previous studies for IQ-CSF.^24,26,40 Although more of the rare sCJD subtypes need to be analyzed, these results suggest that the 2 tests are complementary and should both be performed.

There are a few limitations in this study. The diffusion MRI lesion profile score was assigned semiquantitatively by neuroradiologists from 2 different medical centers with expertise in evaluating patients with prion disease. The high prevalence of prion disease (3:1) in the cohort of 200 individuals may overestimate test performance. Despite the high mean IRR values of the neuroradiologists, the reliability of the scoring system may be lower when used by less expert radiologists or neurologists. In the future, a strategy to extract the MRI signal from diffusion MRI and automatically generate a lesion profile score may improve the accuracy and consistency of CJD diagnosis.

The diagnostic performance of diffusion MRI with the new criterion was superior to that of current standard criteria and equal to that of IQ-CSF in this study performed in, to our knowledge, the largest autopsied cohort with suspected CJD to date. When both diagnostic tests were used, the sensitivity reached 100%. MRI appear to offer several advantages over IQ-CSF in terms of an early diagnostic marker because it is noninvasive and a more accessible test. Moreover, positive MRI findings may suggest sCJD diagnosis even when it is not suspected yet by the physician. Future studies should examine the capability of MRI to identify sCJD subtypes using the brain regions scoring procedure proposed in this study. The effective use of DWI and ADC maps in the diagnosis of sCJD requires that general radiologists become aware of the importance of correctly interpreting the diffusion abnormalities that are characteristic of this rare disease.

Accepted for Publication: March 6, 2020.

Corresponding Author: Alberto Bizzi, MD, Neuroradiology Unit, Fondazione IRCCS Istituto Neurologico Carlo Besta, Via Celoria 11, 20133 Milan, Italy (alberto.bizzi@istituto-besta.it).

Published Online: June 1, 2020. doi:10.1001/jamaneurol.2020.1319

Author Contributions: Dr Bizzi had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Bizzi, Lodi, Safar, Gambetti.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Bizzi, Pascuzzo, Blevins, Lodi, Gambetti.

Critical revision of the manuscript for important intellectual content: Bizzi, Pascuzzo, Grisoli, Lodi, Moscatelli, Castelli, Cohen, Schonberger, Foutz, Safar, Appleby, Gambetti.

Statistical analysis: Pascuzzo, Castelli, Safar.

Obtained funding: Bizzi, Schonberger, Safar, Appleby.

Administrative, technical, or material support: Blevins, Schonberger, Foutz, Safar, Appleby, Gambetti.

Supervision: Bizzi, Lodi, Safar, Appleby.

Conflict of Interest Disclosures: Dr Blevins reported receiving personal fees from Fondazione IRCCS Istituto Neurologico C. Besta during the conduct of the study. Dr Appleby reported receiving grants from the Centers for Disease Control and Prevention during the conduct of the study. No other disclosures were reported.

Funding/Support: This study was supported by grant 666992 from the European Union’s Horizon 2020 Research and Innovation Programme (Drs Bizzi and Pascuzzo) and grants P01AI106705, R01NS083687, and P01A107774 from the National Institutes of Health (Dr Gambetti), and the Charles S. Britton Fund (Dr Gambetti). The National Prion Disease Pathology Surveillance Center (Ms Blevins, Dr Cohen, Mr Foutz, and Dr Appleby) is funded by grant NU2GCK000434 from the Centers for Disease Control and Prevention.

Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.