Cerebral cortex swelling in V180I genetic Creutzfeldt–Jakob disease: comparative imaging study between sporadic and V180I genetic Creutzfeldt–Jakob disease in the early stage

ABSTRACT

The most common genetic Creutzfeldt-Jakob disease (gCJD) in Japan is caused by a point mutation in which isoleucine replaces valine at codon 180 of the prion protein (PrP) gene (V180I gCJD). Evidence suggests that cerebral cortex swelling, which appears as abnormal hyperintensities on diffusion-weighted imaging (DWI), is a characteristic magnetic resonance imaging (MRI) finding of V180I gCJD. However, no study has directly compared the MRI findings between V180I gCJD and sporadic CJD (sCJD). The current study, therefore, aims to clarify the imaging features of V180I gCJD, which would lead to prompt genetic counselling and analysis of the PrP gene, particularly focusing on cerebral cortex swelling. We included 35 patients with sCJD (n = 23) or V180I gCJD (n = 12). Cerebral cortex swelling on T2-weighted imaging (T2WI) or fluid-attenuated inversion recovery (FLAIR) wherein abnormal cortical hyperintensities were observed on DWI, and the distribution of grey matter hyperintensities on DWI were visually evaluated. V180I gCJD patients had significantly more cerebral cortex swelling (100% vs. 13.0%, p < 0.001), an overall correct classification of 91.4%, and parahippocampal gyrus hyperintensities on DWI (100% vs. 39.1%, q = 0.019) than sCJD patients. Cerebral cortical hyperintensities on DWI with swelling on T2WI or FLAIR are characteristic imaging findings of V180I gCJD and are useful for differentiating it from sCJD.

KEYWORDS:

• Cruetzfeldt–Jakob disease
• diagnosis
• diffusion-weighted imaging
• magnetic resonance imaging
• swelling
• V180I

Introduction

Creutzfeldt-Jakob disease (CJD) is a fatal and transmissible neurodegenerative disease that can be classified as sporadic, genetic, and acquired [1]. Genetic CJD (gCJD) can be further classified based on the presence of mutations in the prion protein (PrP) gene, with individual genetic mutations showing varying geographic distributions and frequencies [2,3]. The most common gCJD in Japan is caused by a point mutation in which isoleucine replaces valine at codon 180 of the PrP gene (V180I gCJD) [3]. The clinical features of V180I gCJD are relatively uniform but significantly different from those of typical sporadic CJD (sCJD) as evidenced by their older age of onset, longer disease duration with slower progression, and lower positive rate of periodic sharp wave complexes (PSWC) on electroencephalography (EEG) [3–5]. Furthermore, patients with V180I gCJD rarely present with a family history of the disease [3,5]. Therefore, diagnosing V180I gCJD at its early stage merely based on clinical features is difficult, often leading its misdiagnosis as other types of dementia, such as Alzheimer’s disease [6,7].

Diffusion-weighted brain magnetic resonance imaging (MRI) has been reported to be a useful tool for the early diagnosis of CJD [8,9]. Hyperintensities along the cortices with or without basal ganglia involvement on diffusion-weighted imaging (DWI) have been observed with high sensitivity, as is the case in sCJD [5]. Moreover, the absence of DWI abnormalities in the occipital lobes and cortical swelling at the site of the DWI abnormalities have been reported to be characteristic MRI findings of V180I gCJD [4,10]. However, studies comparing the MRI findings between V180I and sCJD have been scarce, and the usefulness of these radiological features in discriminating between V180I and sCJD remains unclear.

The current study therefore sought to clarify the imaging features of V180I gCJD that would help discriminate it from sCJD and promote prompt genetic counselling and analysis of PrP gene. For this purpose, we compared the distribution of grey matter hyperintensities on DWI and cerebral cortex swelling on T2WI or FLAIR wherein abnormal hyperintensities were observed on DWI.

Results

The demographic characteristics of the patients are summarized in Table 1. The age upon MRI was significantly higher in V180I gCJD patients than in sCJD patients (p = 0.015). The frequency of methionine and valine heterozygosity at codon 129 was also higher in V180I gCJD patients than in sCJD patients (p = 0.015). No significant differences in symptoms and signs on MRI and field strength of MRI were noted between sCJD and V180I gCJD patients. PrP^Sc positivity rates and frequency of PSWC on EEG were significantly lower in V180I gCJD patients than in sCJD patients (p = 0.013 and 0.001, respectively).

Table 1. Comparison of the clinical features between V180I and sporadic Creutzfeldt-Jakob disease (sCJD).

Features	V180I	sCJD	p value
Codon 129 (MM/MV)^a	5/7	20/3	0.015
Sex (male/female)^b	8/4	11/12	0.288
Age upon MRI (years, mean ± SD)^c	80.0 ± 3.7	73.2 ± 10.8	0.015
Disease duration (months, median, range)^d	3.0 (1–31)	2.0 (1–26)	0.400
Symptoms and signs upon MRI
Akinetic mutism (n, %)^a	1/12 (8.3)	2/23 (8.7)	0.971
Myoclonus (n, %)^b	3/12 (25.0)	12/23 (52.2)	0.123
Cognitive impairment (n, %)^a	11/12 (91.7)	23/23 (100)	0.343
Cerebellar dysfunction (n, %)^a	2/12 (16.7)	11/23 (47.8)	0.139
Pyramidal signs (n, %)^a	2/12 (16.7)	6/23 (26.1)	0.685
Extrapyramidal signs (n, %)^a	1/12 (8.3)	5/23 (21.7)	0.640
Cerebrospinal fluid findings
Positivity rate of 14-3-3 protein (n, %)^a	11/12 (91.7)	19/23 (82.6)	0.640
Positivity rate of t-τ protein (n, %)^a	11/12 (91.7)	22/23 (95.7)	1.000
Positivity rate of PrP^Sc (n, %)^a	1/8 (12.5)	13/19 (68.4)	0.013
PSWC in EEG (n, %)^a	0/12 (0)	14/23 (60.9)	0.001
Codon 129 is presented as the total cases of the indicated polymorphism type where MM indicates methionine homozygous and MV indicates methionine and valine heterozygous.
In V180I, the field strength of the MRI was 1.5 T in 10 cases and 3 T in 2 cases, whereas in sCJD, it was 1.5 T in 15 cases and 3 T in 8 cases.

SD, standard deviation; MRI, magnetic resonance imaging; t-τ, total τ; PrP^Sc, scrapie-type prion protein; PSWC, periodic sharp wave complexes; EEG, electroencephalogram.

^aFisher’s exact probability test. ^bChi-squared test. ^cWelch’s test. ^dMann – Whitney U test.

The κ values for interobserver variability between the two examiners were 0.765 and 0.612 for cerebral cortex swelling and distribution of grey matter hyperintensities on DWI.

Cerebral cortex swelling was found more frequently in V180I gCJD patients (12/12, 100%) than in sCJD patients (3/23, 13.0%; p < 0.001) (Figure 1). To differentiate between V180I gCJD and sCJD, cerebral cortex swelling was associated with a sensitivity of 100% (95% CI, 64.0–100), specificity of 87.0% (95% CI, 66.4–97.2), positive predictive value of 80.0% (95% CI, 51.9–95.7), negative predictive value of 100% (95% CI, 76.2–100), and overall correct classification of 91.4% (95% CI, 76.9–98.2). All three patients with sCJD who showed cerebral cortex swelling had methionine homozygosity at codon 129. Age upon MRI did not significantly differ between sCJD patients with and without cerebral cortex swelling (72.3 ± 10.3 vs. 73.4 ± 11.1, p = 0.883). The frequency of cerebral cortex swelling did not differ significantly between 1.5- and 3-T MRI systems (p = 0.458).

Figure 1. Magnetic resonance imaging (MRI) of an 80-year-old male with V180I (a – d) and an 84-year-old male with sporadic Creutzfeldt – Jakob disease (sCJD) (e, f). Brain MRI revealed increased signal intensity in the cerebral cortices on diffusion-weighted imaging in patients with both V180I CJD (a, c) and sCJD (e). Increased signal intensity in the cerebral cortices was also observed on T2-weighted images, with swelling in one patient with V180I CJD (b, d) and no swelling in one patient with sCJD (f).

Gray matter hyperintensities in the inferior frontal, orbitofrontal, mesial frontal, precentral, postcentral, inferior temporal, and parahippocampal gyri were more common in patients with V180I gCJD (Table 2). After correcting for multiple comparisons, hyperintensities on DWI were significantly more frequent in V180I gCJD patients than in sCJD patients only in the parahippocampal gyrus (q = 0.019) (Figure 1, Table 2). Gray matter hyperintensities in the occipital lobe, especially in the lingual gyrus and superolateral occipital gyrus, on DWI were observed less frequently in V180I gCJD patients than in sCJD patients, but no significant difference was observed after correcting for multiple comparisons (Figure 1, Table 2). The frequencies of grey matter hyperintensities on DWI at all regions evaluated are listed in Supplementary table S1. To differentiate between V180I gCJD and sCJD, hyperintensities on DWI in the parahippocampal gyrus were associated with a sensitivity of 100% (95% CI, 64.0–100), specificity of 60.9% (95% CI, 38.5–80.3), positive predictive value of 57.1% (95% CI, 34.0–78.2), negative predictive value of 100% (95% CI, 68.1–100), and overall correct classification of 74.3% (95% CI, 56.7–87.5).

Table 2. Frequency of grey matter hyperintensities at the gyral and nuclear levels in patients with V180I and sporadic Creutzfeldt-Jakob disease (sCJD) (regions with p < 0.1).

	V180I, (%) n = 12	sCJD, (%) n = 23	p value	q value
Inferior frontal	8 (66.7)	7 (30.4)	0.040	0.176
Orbitofrontal	9 (75.0)	7 (30.4)	0.012	0.186
Mesial frontal	10 (83.3)	10 (43.5)	0.024	0.184
Precentral	5 (41.7)	2 (8.7)	0.033	0.171
Postcentral	7 (58.3)	4 (17.4)	0.022	0.230
Inferior temporal	10 (83.3)	11 (47.8)	0.070	0.270
Parahippocampal	12 (100)	9 (39.1)	<0.001	0.019
Lingual	0 (0)	8 (34.8)	0.032	0.196
Superolateral occipital	3 (25.0)	13 (56.5)	0.076	0.260

A false discovery rate was applied as a post-hoc analysis to correct for multiple comparisons. A q value of<0.05 (displayed in bold) was considered significant.

Discussion

Our results showed that all patients with V180I gCJD exhibited cerebral cortex swelling, which was associated with high sensitivity and specificity for discriminating between V180I gCJD and sCJD patients. The distribution of grey matter hyperintensities on DWI also differed between V180I gCJD and sCJD patients, with the former showing significantly more hyperintensities in the parahippocampal gyrus. A previous study on abnormal DWI signals reported that occipital lobe sparing in the early stages is a characteristic of V180I gCJD. Consistent with this, the frequency of abnormal DWI signals was relatively low in the occipital lobe of V180I gCJD patients but did not significantly different from that of sCJD patients.

Cerebral cortex swelling is a useful MRI finding to distinguish between V180I gCJD and sCJD and may also help in the early diagnosis of V180I gCJD. In our study, cerebral cortex swelling showed excellent ability to discriminate between V180I gCJD and sCJD patients. Previous studies have also described cerebral cortex swelling as a radiological feature of patients with V180I gCJD [4,10–12]. Moreover, reports have shown that V180I patients with abnormal DWI signals show pathological findings in the cortex that differ from those in sCJD patients such that they exhibit greater spongiform changes, relatively better preserved neuronal numbers, and lesser deposition of abnormal PrP and astrocytic gliosis counts [6], which may explain the differences in imaging findings between V180I gCJD and sCJD patients. There have been several cases involving patients with both V180I gCJD and sCJD wherein abnormal cortical hyperintensities on DWI had been detected before neurologic symptoms [13–19]. In some of these cases, cerebrospinal fluid (CSF) biomarkers, including total τ, 14-3-3 protein, and RT-QuIC, were negative at the preclinical stage, while DWI signal abnormalities were observed [13,14,17,18]. Moreover, consistent with our results, lower positivity rates for RT-QuIC and PSWC in EEG have been reported in patients with V180I gCJD [5,20]. Therefore, clinicians should be aware that abnormal cortical signals on DWI with swelling on T2WI or FLAIR are a characteristic finding of V180I and that genetic counselling and analysis of the PrP gene when such findings are detected may promote an early diagnosis. Considering future disease-modifying therapy, early diagnosis is important for early intervention and inclusion in clinical trials.

V180I gCJD shares several common features with Alzheimer’s disease. Accordingly, our study showed that hyperintensities in the parahippocampal gyrus on DWI were significantly more frequent in V180I gCJD patients than in sCJD patients. Reports have shown that the earliest neuropathological changes in Alzheimer’s disease appear in the entorhinal cortex, which is the anterior part of the parahippocampal gyrus [21]. Moreover, atrophy of the parahippocampal gyrus has been reported as an important early imaging marker of Alzheimer’s disease [22]. V180I gCJD is often misdiagnosed as Alzheimer’s disease in its early stages owing to its older onset and relatively slower progression compared to sCJD [6,7]. At the early stages, patients with both V180I gCJD and Alzheimer’s disease have been reported to exhibit hypoperfusion in the posterior cingulate gyrus and precuneus on brain perfusion single-photon emission computed tomography [15,23]. In an autopsy case involving a V180I gCJD patient with coexisting Alzheimer’s disease-type pathology, the most patchy PrP deposits were colocalized with amyloid β plaques, leading the authors to speculate that amyloid β plaques may have acted as a facilitating factor for PrP deposition [24].

Our study has several limitations. First, this study did not include gCJDs other than V180I gCJD; hence, it remains unclear whether cerebral cortex swelling is highly specific to V180I gCJD. Second, given that coronal planes were not available in most cases, only axial planes of diffusion-weighted images were evaluated, which may not accurately distinguish between the anterior and posterior parts of the cingulate gyrus and insula. Finally, V180I gCJD patients were older upon MRI and had a higher frequency of methionine and valine heterozygosity in the codon 129 polymorphism than did sCJD patients, which may have promoted differences in imaging findings. However, all three patients with sCJD who showed cerebral cortex swelling were methionine homozygous at codon 129. Moreover, no significant difference in age upon MRI was noted between sCJD patients with and without cerebral cortex swelling.

In conclusion, cerebral cortical hyperintensities on DWI with swelling on T2WI or FLAIR are characteristic imaging findings of V180I gCJD and can be useful for differentiating it from sCJD. Clinicians should be aware of this finding and recognize that prompting genetic counselling and analysis of the PrP gene when this finding is detected may lead to the early diagnosis of V180I gCJD.

Patients and methods

Patients

This retrospective study was approved by the Institutional Review Board of Chiba University Graduate School of Medicine, who waived the need for informed consent. Patients with CJD who visited Chiba University Hospital between February 2009 and February 2022 were identified from our database. The inclusion criteria for patients with sCJD were as follows: satisfying the probable or definite category in a set of diagnostic criteria [25] and the absence of pathogenic mutations during PrP gene analysis. The inclusion criteria for patients with V180I gCJD were as follows: satisfying the probable or definite category in a set of diagnostic criteria [25] and confirmation of V180I mutation during PrP gene analysis. Four patients without PrP gene analysis were excluded. Based on the aforementioned criteria, a total of 23 sCJD (definite 2, probable 21) and 12 V180I gCJD patients (probable 12) were included in the present study. Two V180I gCJD patients and one sCJD patient had already been described in previous reports [10,11,13].

The medical records of the patients were reviewed for polymorphisms at codon 129 of the PrP gene, sex, age upon MRI, disease duration (time from onset to MRI), symptoms, and signs upon MRI (akinetic mutism, myoclonus, cognitive impairment, cerebellar dysfunction, pyramidal signs, and extrapyramidal signs), CSF markers (total τ protein, 14-3-3, and PrP^Sc proteins), and the appearance of PSWC on EEG. The open reading frame and polymorphisms in codon 129 of the PrP gene were analysed after genomic DNA was extracted from the blood of the patients, as previously described [26]. The CSF markers of all patients were assessed at Nagasaki University [27]. Total τ and 14-3-3 proteins in the CSF were evaluated via Western blotting as previously described [20]. PrP^Sc in the CSF was detected using real-time quaking-induced conversion (RT-QuIC), as previously described [28].

Image interpretation

In cases requiring several MRI procedures, MRI performed at the earliest time after disease onset was used for evaluation. All MRI studies of the patients were performed during routine clinical care using 1.5-T (n = 25) and 3-T (n = 10) MR scanners. DWI axial and T2WI or FLAIR axial images were available for all patients. MRI data for each patient were independently evaluated by two examiners who were blinded to the clinical data (A.S. and H.M). Differences in interpretations were ultimately settled by a third examiner who was blinded to the clinical data (J.H.). The examiner judged cerebral cortex swelling on T2WI and/or FLAIR images to be positive or negative when abnormal hyperintensities on DWI were observed. The examiner also determined the distribution of grey matter hyperintensities on DWI according to 26 cortical and 5 subcortical subdivisions, as previously described [9].

Statistical analysis

All statistical analyses were performed using SPSS software version 25.0 (SPSS Japan, Tokyo, Japan). Demographic data of the sCJD and V180I gCJD patients were compared using Welch’s test and the Mann – Whitney U test for continuous variables and Fisher’s exact probability test and the χ² test for categorical variables. Age upon MRI in sCJD patients with and without cerebral cortex swelling were compared using Student’s t-test. The frequency of cerebral cortex swelling was compared between 1.5- and 3-T MRI systems using Fisher’s exact probability test. The frequency of grey matter involvement on DWI in each subdivision between sCJD and V180I gCJD patients was compared using Fisher’s exact probability test and the χ² test. A p value of <0.05 was considered statistically significant. False discovery rate correction was performed using the Benjamini – Hochberg procedure and adjusted p values (hereafter presented as q values) were calculated such that q values of <0.05 were considered statistically significant. Interrater agreements were tested using Cohen’s κ statistic, which was interpreted as follows: <0.2, poor; 0.21–0.4, fair; 0.41–0.6, moderate; 0.61–0.8, very good; and 0.81–1, excellent.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/19336896.2023.2197809

Additional information

Funding

No targeted funding reported.