Case report of anti-vimentin IgG-associated autoimmune encephalitis: diagnostic challenges and the role of 18F-DPA714 PET/MRI

Abstract

Background

Anti-vimentin IgG-associated autoimmune encephalitis (AE) is a recently described autoimmune astrocytopathy characterised by relapsing-remitting cerebellar ataxia, brainstem involvement and pyramidal tract lesions. Nevertheless, several key aspects remain poorly elucidated, including the full clinical phenotypic spectrum (particularly peripheral nervous system (PNS) involvement), the interpretation of serological discrepancies and long-term management outcomes. Additionally, the imaging evidence for neuroinflammation and the evidence regarding evolution following treatment remain insufficient.

Case presentation

We reported a man in his late 40s who developed a core triad of cerebellar ataxia, pyramidal signs and peripheral neuropathy. Conventional MRI was unremarkable, but cerebrospinal fluid analysis revealed anti-vimentin IgG positivity (titre 1:10) with type III oligoclonal bands while serum was negative. ¹⁸F-DPA714 positron emission tomography/MRI showed elevated tracer uptake in the mesial temporal lobe, frontal lobe and brainstem, correlating with clinical signs. After intravenous immunoglobulin induction and ofatumumab maintenance therapy, the patient showed clinical improvement (modified Rankin Scale from 2 to 1) at 3-month follow-up, paralleled by reduced tracer uptake.

Conclusions

This first reported middle-aged male case expands the demographic and phenotypic spectrum of anti-vimentin IgG-associated AE to include PNS involvement. This case highlights the diagnostic value of antibody detection and advanced imaging modalities in atypical presentations of anti-vimentin IgG-associated AE.

Introduction

Anti-vimentin IgG-associated autoimmune encephalitis (AE) is a recently defined antibody-mediated central nervous system (CNS) disorder,¹ initially described predominantly in young females and typified by relapsing-remitting rhombencephalitis, cerebellar ataxia and symmetric corticospinal tract (CST) lesions on MRI.¹ However, full clinical spectrum—particularly peripheral nervous system (PNS) involvement—and long-term therapeutic outcomes remain incompletely characterised. Therefore, an imaging modality capable of reflecting intracerebral inflammation is valuable for diagnosing and monitoring.

Besides defining these demographic and clinical features, previous case series relied primarily on conventional MRI, which revealed bilateral symmetric CST lesions as characteristic findings but provided no longitudinal imaging follow-up data.¹ Treatment with corticosteroids and intravenous immunoglobulin (IVIG) mainly yielded suboptimal outcomes.

Building on this framework, our case expands the clinical, imaging and therapeutic landscape. We present the first middle-aged male patient, broadening the demographic spectrum. Distinct from previously reported CNS-limited cases, this patient exhibited prominent PNS involvement, with electromyography verifying sensory axonal injury and brachial plexopathy—extending the phenotypic spectrum. ¹⁸F-DPA714 positron emission tomography (PET)/MRI identified microglial activation undetectable on conventional MRI and provided the first longitudinal imaging data. Ofatumumab maintenance therapy achieved well-tolerated short-term functional improvement. These findings complement limited clinical data, refine clinical phenotypic and imaging characteristics and provide actionable evidence for diagnosis and targeted treatment of this rare autoimmune neurological disorder.

Subject and methods

Detailed descriptions of subject and methods, including data collection, hybrid ¹⁸F-DPA714 PET/MRI acquisition and antibody detection techniques, are provided in online supplemental material 1.

Case

3 years ago, a man in his late 40s developed diplopia and gait instability 6 weeks after receiving a COVID-19 vaccination, followed by progressive right-sided limb numbness (predominantly in the distal parts), with occasional choking episodes during swallowing (figure 1A). Initial cranial and cervical spine MRIs, performed 2 months after neurological symptom onset, revealed no abnormality (online supplemental figure S1). Based on significant cerebellar ataxia, diminished reflexes and peripheral neuropathy, Miller-Fisher syndrome was initially considered. Given the lack of key supportive features (no albuminocytological dissociation, negative anti-GQ1b antibodies, no ophthalmoplegia), immunosuppressive therapy was not pursued.

Figure 1. (A) Clinical timeline. The clinical timeline summarised key events from symptom onset through diagnosis, treatment and follow-up. (B) Tissue-based assay (TBA) for anti-vimentin IgG antibody. Immunostaining of rat cerebellar and hippocampal tissues with cerebrospinal fluid (CSF) and serum from the patient exhibited an astrocyte staining pattern. CSF and serum samples yielded positive results in the cerebellum, suggesting a potential antigen-antibody binding reaction (scale bar=50 µm). (C) Cell-based assay (CBA) for anti-vimentin IgG antibody. Dual-fluorescence CBA showed negative results in serum but weak positivity in CSF (titre 1:10) both at initial and follow-up testing (scale bar=50 µm). (D) IgG oligoclonal bands (OCBs). Isoelectric focusing demonstrated type III OCBs pattern in the CSF. (E) Brain imaging findings of the patient. Axial views of T2 FLAIR, T1WI, DWI and corresponding ¹⁸F-DPA714 PET/MRI are shown. Lesions are annotated by arrows. In the first visit, ¹⁸F-DPA714 PET/MRI suggested elevated tracer uptake in the mesial temporal lobe (SUVR_left=1.61, 25% above normal reference), frontal lobe (SUVR_left=1.43, SUVR_right=1.42, 14% and 11% above normal reference) and brainstem (SUVR=1.69, 9% above normal reference). At the follow-up scan 3 months later, the patient’s ¹⁸F-DPA714 uptake in the medial temporal lobe and brainstem had markedly improved (mesial temporal lobe SUVR_left=1.11; frontal lobe SUVR_left=1.21, SUVR_right=1.23; brainstem: SUVR=1.33). (F) Longitudinal changes in clinical parameters. Changes in CSF antibody titres and various scales (mRS, CASE, MoCA) from baseline to the 3-month follow-up. Pink bar indicates baseline values, blue indicates values at 3-month follow-up. To facilitate cross-parameter comparison, all values were standardised to their respective maximum at each time point. (G) Longitudinal changes in ¹⁸F-DPA714 PET/MRI uptake. Changes in SUVR values from baseline to 3-month follow-up in the mesial temporal lobe, frontal lobe and brainstem. Yellow indicates normal values for healthy controls, while pink and blue bar represent SUVR values at baseline and at the 3-month follow-up, respectively. The patient’s inflammatory burden in the mesial temporal lobe and brainstem has shown marked improvement. To facilitate cross-parameter comparison, all values were standardised to their respective maximum at each time point. CASE, Clinical Assessment Scale for Autoimmune Encephalitis; DWI, diffusion weighted imaging; FLAIR, fluid attenuated inversion recovery; IVIG, intravenous immunoglobulin; MoCA, Montreal Cognitive Assessment; mRS, modified Rankin Scale; PET, positron emission tomography; SUVR, standardised uptake value ratio; T1WI, T1-weighted imaging.

Following COVID-19 reinfection 1 year later, he developed right hand tremor, wrist tightness, worsening gait instability and subsequent right wrist dorsiflexion weakness. Electromyography (EMG) suggested right lower trunk brachial plexopathy (predominantly sensory). A diagnosis of dystonia was made but did not respond to titrated doses of oral clonazepam (initially 0.5 mg three times daily). Persistent limb numbness prompted a repeat EMG several months later, which indicated sensory axonal damage in upper limbs.

Cerebellar ataxia was demonstrated by bilaterally positive finger-to-nose and heel-to-shin tests (right side), a wide-based gait, scanning speech and involuntary movements. Meanwhile, positive Romberg’s sign suggested sensory ataxia. Pyramidal signs were also evident, featuring right hyperreflexia, a positive Babinski’s sign on the right and increased muscle tone in right upper limb. Peripheral neuropathy was confirmed by reduced pain and temperature sensation in right limb and abnormal proprioception. No psychiatric symptoms were observed. Clinical assessment scales yielded the following scores: modified Rankin Scale (mRS, 2), Clinical Assessment Scale for Autoimmune Encephalitis (CASE, 1) and Montreal Cognitive Assessment (MoCA, 29).

Cerebrospinal fluid (CSF) analyses revealed unremarkable cell counts and protein concentrations, but an elevated IgG level of 58.9 mg/L. Comprehensive autoimmune panels for antibodies associated with autoimmune cerebellar ataxia, paraneoplastic syndromes and demyelinating disorders were negative in both serum and CSF (see online supplemental material 2). Although serum anti-vimentin IgG antibodies were negative, CSF was positive at a titre of 1:10 (figure 1B,C). Isoelectric focusing electrophoresis revealed type III oligoclonal bands (OCBs) (figure 1D).

Cranial MRI showed only bilateral white matter lesions, without typical CST involvement or infratentorial/spinal cord inflammation. However, ¹⁸F-DPA714 PET/MRI demonstrated significantly elevated ¹⁸F-DPA714 uptake in mesial temporal lobe (standardised uptake value ratio (SUVR)_left=1.61; normally 1.29), frontal lobe (SUVR_left=1.43, SUVR_right=1.42; normally 1.26 and 1.28) and brainstem (SUVR=1.69, normally 1.55), with increases of 25%, 35% and 9% above normal, respectively (figure 1E).

The patient received IVIG at 0.4 g/kg/day for 5 days, followed by monthly subcutaneous injections of Ofatumumab‌ as maintenance therapy (total of four vials administered by the 3-month follow-up). The patient was fully informed of the potential benefits, risks, and alternative treatment options and provided written informed consent before initiation. At 3-month follow-up, gait instability markedly improved, with partial resolution of limb numbness. The mRS (from 2 to 1) and MoCA score (from 29 to 30) improved, while CASE score remained unchanged (figure 1F). Follow-up MRI revealed no new or active lesions.

¹⁸F-DPA714 PET/MRI uptake values (figure 1G) in mesial temporal lobe, frontal lobe, and brainstem have all decreased, with values now below the normal reference ranges (mesial temporal lobe: SUVR_left from 1.61 to 1.11, −31%; frontal lobe: SUVR_left from 1.43 to 1.21, −15%, SUVR_right from 1.42 to 1.23, −13%; brainstem SUVR: from 1.69 to 1.33, −21%). Repeat cell-based assay (CBA) testing revealed remained positive CSF anti-vimentin IgG (1:10 dilution). Subsequent management consisted of ongoing monthly Ofatumumab‌ maintenance therapy and long-term neurotrophic support. No significant adverse reactions were reported during this period.

Discussion

Epidemiology and clinical features

This recently defined disease, diagnosed via CSF anti-vimentin IgG testing (CBA), currently lacks large-scale epidemiological data. Its incidence and prevalence await future prospective studies.^{1 2} Reported patients had a median age of 33 years old (range: 23–41), with a female predominance (10/14, 71%).¹ Preceding symptoms such as infections or vaccinations were observed in 79% of patients (11/14).¹ Neurologically, all reported cases exhibit cerebellar ataxia, cranial nerve palsies and pyramidal tract signs. Sensory abnormalities (12/14, 86%) and muscle weakness (10/14, 71%) are also frequently observed.¹

EMG and nerve conduction studies provided the first objective evidence of peripheral nerve involvement—sensory axonal damage and brachial plexopathy. Comprehensive antibody screening—including panels for paraneoplastic syndromes and peripheral neuropathies (eg, anti-GQ1b, GM1, GD1b)—was negative in both serum and CSF. While current testing panels cannot definitively exclude all potential PNS-directed autoantibodies,² this extensive negative screening substantially reduces the likelihood of a distinct co-existing antibody-mediated process. Furthermore, the improvement of PNS symptoms following immunotherapy supports an immune-mediated pathological process.³ Nevertheless, whether anti-vimentin IgG antibodies directly contribute to peripheral nerve injury or whether PNS symptoms represent an indirect consequence of CNS inflammation remains unresolved.

The CSF-restricted antibody pattern in our patient aligns with recognised pathophysiology of anti-vimentin IgG-associated AE—intrathecal production of anti-vimentin IgG by plasma cells and plasmablasts within the CNS. Several lines of evidence support this interpretation. First, type III OCBs on isoelectric focusing provides direct evidence of compartmentalised immunoglobulin production within the CNS.⁴ Second, the original discovery study identified clonally expanded B cells in the CSF expressing IgG autoantibodies targeting vimentin.¹ Third, a recently published case documented the exact same phenomenon—CSF vimentin IgG positive while serum was negative, suggesting this pattern reflects the biology of this disease rather than an isolated finding.⁵

Neuroimaging

Current neuroimaging findings showed most patients had bilateral symmetric CST lesions on MRI.¹ In our case, diagnosis remained robust despite conventional MRI showing only bilateral white matter lesions without characteristic CST involvement, highlighting its limitation in detecting active neuroinflammation during remission or in subtle stages.² Faced with this diagnostic challenge, we employed ¹⁸F-DPA714 PET/MRI, a translocator protein (TSPO)-targeted PET tracer that visualises microglial activation as a sensitive marker of neuroinflammation.^{6 7} Our previous studies demonstrated that ¹⁸F-DPA714 PET detected neuroinflammation in a substantial proportion of MRI-negative AE patients, with tracer uptake correlating with disease severity on the mRS.⁸ Furthermore, ¹⁸F-DPA714 PET/MRI allowed us to quantify neuroinflammatory burden at baseline and assess changes following treatment.^{9 10} In our case, it demonstrated significantly elevated tracer uptake in clinically relevant regions: brainstem (SUVR=1.69, +9%), aligned with prominent cerebellar ataxia and balance impairment; frontal lobe (SUVR_left=1.43, +14%; SUVR_right=1.42, +11%), correlated with observed pyramidal signs and motor dysfunction; and elevated uptake in mesial temporal lobe (SUVR_left=1.61, +25%). At 3-month follow-up, reduced tracer uptake in the mesial temporal lobe and brainstem paralleled clinical improvement (mRS from 2 to 1). ¹⁸F-DPA714 PET can detect inflammatory activity when conventional MRI is unremarkable, supporting early diagnosis and immunotherapy,⁷ and offers objective monitoring of treatment response. This multimodal integrated structural-molecular imaging strategy holds promise for tracking disease progression and evaluating treatment efficacy.

Clinical outcome and prognosis

Currently, no established maintenance strategy exists for this disease, and relapse rates are high following corticosteroid tapering.¹ In our patient, after IVIG induction resulted in partial clinical improvement, we considered ofatumumab as a potential maintenance option. This decision was based on the rationale that anti-vimentin IgG-associated AE is an antibody-mediated disorder with clonally expanded B cells in the CSF,¹ and on the emerging evidence supporting B-cell depletion in AE.^11–14 Ofatumumab is a fully human anti-CD20 monoclonal antibody that induces B-cell depletion.¹¹ Recent studies have explored its use in AE, including case series and cohort studies in anti-N-methyl-D-aspartate receptor (NMDAR) encephalitis, antibody-negative AE, and anti-KLHL11 encephalitis, showing favourable efficacy and safety profiles.^{12 13} The patient was informed of the potential benefits, risks, and alternatives, and provided written informed consent for this exploratory use.

In our case, following IVIG and ofatumumab, mRS improved from 2 to 1, with resolution of gait instability and sensory symptoms. Slight numbness of right fingertip remained. He maintained stable without relapse during 3-month follow-up, suggesting a more favourable short-term prognosis than previously reported. Treatment was well tolerated, with no adverse events recorded during this period. These findings complement the limited clinical data, refine its clinical phenotypic and imaging characteristics, and provide actionable evidence for diagnosis and targeted treatment of this rare autoimmune neurological disorder. Continuous monitoring remains essential to determine long-term prognosis and evaluate risk of disease recurrence.

Conclusions

This case report of a middle-aged male presenting with cerebellar ataxia and peripheral neuropathy expands the clinical spectrum. Notably, ¹⁸F-DPA714 PET imaging plays a crucial role in evaluating intracerebral inflammation in such patients and exhibits great potential as a sensitive tool for diagnosis and treatment monitoring.