Neuronal intranuclear inclusion disease (NIID) is a slowly progressive neurodegenerative disorder characterized by eosinophilic hyaline intranuclear inclusions in the nervous system, visceral organs and skin.Reference Sone, Mori and Inagaki1 Recent genetic studies have shown an association between NIID and GGC repeat expansions in the NOTCH2NLC gene, thought to be more common in the East Asian population but rare in European-descent individuals.Reference Sone, Mitsuhashi and Fujita2
A 73-year-old woman of Ukrainian origin was hospitalized twice in 2 years in another hospital before being transferred to our institution. Initially, she presented with a generalized seizure and left-sided weakness. Electroencephalogram (EEG)indicated right hemispheric focal status epilepticus; brain MRI was interpreted as demonstrating white matter changes from chronic microangiopathy. The CSF analysis was unremarkable. She underwent treatment with anti-seizure medications and returned to her cognitive baseline within 2 weeks. EEG was normal after 3 months, and anti-seizure medication was discontinued.
Two years later, she presented again with sudden behavioral changes, altered awareness, agitation and verbal perseverations. A brain MRI reported new mild left parieto-occipital lobar edema with gyriform enhancement, suggesting encephalitis or neoplasm.
The initial EEG was inconclusive but suggested right hemispheric focal status epilepticus, prompting anti-seizure treatment. However, encephalopathy with agitation and aphasia persisted. Subsequent EEGs showed no seizures but bilateral slowing, more pronounced on the left, consistent with diffuse encephalopathy and superimposed left hemispheric cerebral dysfunction.
CSF analysis revealed mild pleocytosis (11 white cells/mm³) and elevated protein (0.64 mg/mL); cytology was negative for malignant cells. Extensive testing for infectious agents and autoimmune encephalitis, as well as systemic workup for primary malignancy, were negative.
Family history obtained from her sister noted that their mother had dementia, and their brother was diagnosed with dementia following a head injury.
She was again treated with anti-seizure medications, including phenytoin and levetiracetam. Given the MRI findings and suspicion of encephalitis, she also received empirical treatment with acyclovir and a short course with intravenous methylprednisolone. However, she remained encephalopathic, and a repeat brain MRI showed persistent cortical signal abnormalities, raising concern for a neoplastic etiology. Consequently, she was transferred to our institution for a brain biopsy. The brain histopathology and MRI findings were reviewed during interdepartmental tumor board meetings.
Imaging findings: Brain MRI without gadolinium during her first hospitalization showed symmetric confluent white matter T2-FLAIR hyperintensities in both cerebral hemispheres, especially in the frontal lobes (Figure 1A). High Diffusion-weighted imaging (DWI) signals were present along the corticomedullary junctions of the bilateral frontal lobes, predominantly on the right (Figure 1A). Mild lobar edema with cortical and subcortical T2-FLAIR hyperintensities was observed in the right temporoparietal lobe (Figure 1A). Infratentorially, small bilateral paravermal T2-FLAIR hyperintensities were noted (Figure 1A).
Brain MRI during her first and second hospitalizations. Brain MRI during her first hospitalization 2 years earlier (A). DWI (A.a–A.c) and T2-FLAIR (A.d–A.g) showed symmetric confluent white matter T2-FLAIR hyperintensities in bilateral deep cerebral white matter (blue arrows in A.d and A.e), high DWI signals along bifrontal corticomedullary junctions (white arrows in A.a and A.b) and T2-FLAIR hyperintensities in bilateral paravermal regions (red arrows in A.g), which are typical MRI findings of neuronal intranuclear inclusion disease. Mild lobar cerebral edema with cortical and subcortical T2-FLAIR hyperintensity involving the right temporoparietal lobe (yellow arrows in A.e and A.f) were also seen in retrospect. Brain MRI during her second hospitalization (B). DWI (B.a–B.c), T2-FLAIR (B.d–B.g), post-gadolinium T1W TSE (B.h–B.k) showed mild lobar cerebral edema and T2-FLAIR hyperintensities involving the left parieto-occipital cortical-subcortical region (yellow arrows in B.e and B.f) with gyriform enhancement (orange arrows in B.h–B.k). The pre-existing symmetric confluent T2-FLAIR hyperintensities in bilateral deep cerebral white matter (blue arrows in B.d and B.e), high DWI signals along the bifrontal corticomedullary junction (white arrows in B.a and B.b) and T2-FLAIR hyperintensities in bilateral paravermal regions (red arrows in B.g) were persistent. The previously seen right temporoparietal lobar edema associated with cortical-subcortical T2-FLAIR hyperintensities on brain MRI 2 years prior had resolved.

During the second hospitalization, a brain MRI with gadolinium revealed new mild lobar edema and T2-FLAIR hyperintensities involving the cortex and subcortical white matter of the left parieto-occipital lobe (Figure 1B) with gyriform enhancement (Figure 1B). The previously seen right temporoparietal abnormalities have resolved. Symmetric confluent deep white matter T2-FLAIR hyperintensities, corticomedullary high DWI signals and paravermal T2-FLAIR hyperintensities persisted (Figure 1B).
Brain biopsy: Brain biopsy, performed at the left parietal lobe at the region of subcortical FLAIR hyperintensity and cortical enhancement on MRI, revealed that the cerebral cortex and underlying white matter showed relatively mild reactive astrogliosis confirmed on Glial fibrillary acidic protein (GFAP) immunostains (Figure 2C). Occasional macrophages were present in the Virchow–Robin spaces around blood vessels. Numerous neurons, astrocytes and oligodendrocytes contained an eosinophilic intranuclear inclusion body, distinct from the nucleolus and often adjacent to it. These inclusions were intensely stained by p62 (Figure 2F-H). p62 also labeled inclusions in endothelial cells. A sparse population of CD3 positive T lymphocytes was scattered in the neuropil and around blood vessels. There were no cells labeled for CD20. Luxol fast blue stain showed a normal myelination pattern. Neither spongiosis nor infarcts were identified. All microbiologic studies were negative.
Cerebral and skin biopsies. Cerebral biopsy (A–H), showing slight hypercellularity, due to the presence of T lymphocytes (B, CD3 immunostain) and gliosis (C, GFAP immunostain). Intranuclear inclusion bodies are present in neurons (arrows in D, H&E) and oligodendrocytes (inset in D, H&E). Intranuclear inclusions in three astrocytes (arrows in E, H&E) with eosinophilic cytoplasm. P62 immunostain (F–H) demonstrating inclusion bodies in a neuron (arrow in F), a perineuronal satellite oligodendrocyte (arrow in G) and endothelial cells (and arrow in H). Skin biopsy, P62 immunostain (I), showing intranuclear inclusion bodies in sweat glands (arrow in I) and arrector pili smooth muscle (inset in I). Magnifications bars: (A) 25 μm, (B) 250 μm, (C) 75 μm, (D, E, F) G 7.5 μm, (H, I) 25 μm.

Skin biopsy: Skin biopsy, performed at the right axilla, showed eosinophilic intranuclear inclusions in the sweat gland on Hematoxylin and Eosin (H&E), confirmed on p-62 immunochemistry, which also stained intranuclear inclusions in arrector pili smooth muscles (Figure 2I).
Genetic studies: Long-read sequencing revealed 9 and 110 GGC repeats in the NOTCH2NLC 5’-UTR, indicating one normal allele and one pathogenic expansion. (Supplementary Figure) Testing for CGG repeat expansions in the FMR1 gene was negative.
The patient’s mental status recovered slowly over several weeks. The mental status, cranial nerves and motor and sensory examinations were normal at follow-up 2 months after discharge except for mildly impaired short-term memory, decreased perception of the right visual hemifield, truncal ataxia and fine finger postural tremors.
NIID was first reported in the 1960s, initially in children and juvenile European cases discovered postmortem.Reference Sone, Mori and Inagaki1 Since the introduction of skin biopsy to facilitate the diagnosis of NIID and the identification of typical MRI findings in the early 2010s, the number of reported cases has been rapidly increasing in the East Asian population, particularly in older adults.Reference Sone, Mori and Inagaki1,Reference Suthiphosuwan, Sasikumar, Munoz, Chan, Montanera and Bharatha3
Typical MRI features of NIID include high DWI signals along the corticomedullary junctions, especially in the frontal lobes, and confluent subcortical and deep white matter T2-FLAIR hyperintensities.Reference Sone, Mori and Inagaki1,Reference Suthiphosuwan, Sasikumar, Munoz, Chan, Montanera and Bharatha3 Infratentorial findings often include T2-FLAIR paravermal hyperintensities.Reference Suthiphosuwan, Sasikumar, Munoz, Chan, Montanera and Bharatha3 Rarely, a cortical encephalitis-like MRI pattern, characterized by transient unilateral focal cerebral edema with cortical T2-FLAIR hyperintensities and gyriform enhancement, has been reported in approximately 10% of East Asian NIID cases.Reference Sone, Mori and Inagaki1
Recently, the discovery of GGC repeat expansions in the NOTCH2NLC gene has been found as a causative factor of NIID pathogenesis.Reference Sone, Mitsuhashi and Fujita2,Reference Boivin, Deng and Pfister4 These expansions lead to the production of pathogenic polyglycine (polyG) containing proteins, resulting in intranuclear inclusions and neurodegeneration.
Only two European NIID casesReference Podar, Gutmann and Harmuth5,Reference De Santis, Politi, Valente and Albanese6 with typical MRI findings and NOTCH2NLC GGC repeat expansions have been reported so far, but neither exhibited the cortical encephalitis-like MRI pattern. To our knowledge, this is the first reported case of recurrent cortical encephalitis-like presentation of NIID in a patient of European descent.
The brain biopsy revealed astrogliosis and numerous p62-positive intranuclear inclusions in astrocytes and oligodendrocytes, consistent with NIID. There was evidence of smoldering encephalitis, indicated by CD3-positive T-cell infiltrates and perivascular macrophages, suggesting an immune response and neuroinflammation corresponding with the cerebral edema and cortical MRI abnormalities. Unlike in a previous report,Reference Mori, Yagishita, Funata, Yamada, Takaki and Miura7 we did not find evidence of cerebral infarction. There is limited literature describing histopathological correlation, and the biological substrate of episodic attacks in NIID remains to be elucidated.
Clinically, NIID may be indistinguishable from fragile X-associated tremor/ataxia syndrome (FXTAS). Both conditions can show similar MRI findings and p62-positive intranuclear inclusions in skin and brain biopsies.Reference Zhou, Jankovic, Ashizawa and Tan8 Therefore, testing for CGG repeat expansions in the FMR1 gene is recommended to distinguish FXTAS from NIID, which was negative in our patient. Additionally, the cortical encephalitis-like MRI pattern has not been reported in FXTAS, and the presence of p62-positive intranuclear inclusions in oligodendrocytes on brain biopsy indicates NIID rather than FXTAS.Reference Zhou, Jankovic, Ashizawa and Tan8
Commercial antibodies targeting specific poly-amino acids, such as 1C2 for polyglutamine (polyQ), have been widely used in the pathological diagnosis of various repeat expansion disorders. While immunohistochemistry using these antibodies was not performed in this case, their potential utility in identifying pathogenic repeat-associated proteins in NIID and other expansion disorders is important. Future studies incorporating such markers may enhance diagnostic accuracy and further elucidate the pathological mechanisms underlying these conditions.
This case highlights that NOTCH2NLC-related NIID can occur in European-descent patients, mirroring features reported in East Asia. Diagnosis requires skin biopsy and NOTCH2NLC testing, plus FMR1 testing given overlap with FXTAS. In this case, cortical encephalitis-like MRI lesions, consistent with smoldering encephalitis on histopathology, reflect an immune response and neuroinflammation.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/cjn.2025.41.
Acknowledgments
The authors thank Jing Xian Teo for technical support.
Author contributions
Conceptualization: SS, AS
Methodology: SS, AS, ASLN
Formal analysis and investigation: WKL, EGYC, JNF, YJT, DGM
Writing – original draft preparation: SS, TRL, AS
Writing – review and editing: All authors
Resources: SS, AS, DGM, ASLN
Supervision: SS, AS
Funding statement
The NOTCH2NLC genetic testing was supported by a grant awarded to Jia Nee Foo from the Singapore Ministry of Health’s National Medical Research Council Open Fund Individual Research Grant (MOH-000559) and the Singapore Ministry of Education Academic Research Fund Tier 2 (MOE-T2EP30220–0005) and Tier 3 (MOEMOET32020-0004), Clinician-Scientist Award (MOH-CSAINV21-0005) and Transition Award (MOH-TA18may-0003) to Adeline Su Lyn Ng and National Precision Medicine program PHASE II FUNDING (MOH-000588) to Weng Khong Lim.
Competing interests
On behalf of all authors, the corresponding author states that there is no conflict of interest.
Ethical standards
This study has been approved by our Institutional Review Ethics Board. Informed consent for publication was obtained from the patient.