Highly Relapsing Anti-GABAA Receptor Encephalitis: A Case with Four Immunotherapy-Responsive Episodes

Sir, Case series have revealed a link between GABAA (gamma-aminobutyric acid) receptor autoantibodies and specific autoimmune encephalitis.[1,2] Anti-GABAA receptor encephalitis is a rare disease with an unknown natural course. Here, we report a case of anti-GABAA receptor encephalitis with a high seizure burden, distinctive brain MRI features, a high tendency to relapse, and a positive response to immunotherapy each time. A 61-year-old man presented with a generalized tonic-clinic seizure on October 28, 2021. Brain Magnetic resonance imaging (MRI) revealed bifrontal asymmetrical cortical and subcortical T2/Fluid-attenuated inversion recovery (FLAIR) hyperintense, non-enhancing lesions, with magnetic resonance spectroscopy (MRS) showing a high choline peak and a reduced N-acetyl aspartate (NAA) peak. There was no diffusion restriction and no evidence of either hemorrhage or calcification [Figure 1a]. The primary physician considered the possibility of a low-grade neoplasm and prescribed anti-seizure medications.Figure 1: Serial MRI brain findings: Axial Flair MRI on October 28, 2021 with bifrontal asymmetric cortical and subcortical lesions (a); On November 18, 2021 with new lesions in the bilateral cingulate gyrus and right medial occipital lobe (b); On May 5, 2022 with multiple new bifrontal and temporal lobe lesions (c); On August 31, 2022 with a new lesion in the left temporal lobe (d); On December 17, 2022 with complete remission of all lesions (e); On March 6, 2024 with three new lesions in both temporal and left frontal lobes (f)The patient had altered behavior and complex partial seizures on 18th November. Follow-up MRI showed an enlargement in the size of the pre-existing lesions and new lesions in the right medial occipital lobe and bilateral cingulate gyrus [Figure 1b]. There was minimal partial diffusion restriction in the left frontal lobe lesion. The lesions were non-enhancing, and the spine MRI was normal. The rapid lesion growth within 20 days and the multifocal nature of the lesions suggested autoimmune pathogenesis as a more likely possibility than a low-grade neoplastic lesion. CSF analysis, including cytology, was normal. Investigations, including myelin oligodendrocyte glycoprotein (MOG), aquaporin-4 (AQP4), and thyroid peroxidase (TPO) antibodies, were negative. The patient was treated with pulse methylprednisolone, resulting in good recovery. Tapering doses of oral steroids were continued. After 5 months, on May 5, 2022, he experienced focal seizures, dysarthria, and choreiform movements. A repeat MRI showed multiple new cortical and subcortical non-enhancing and non-diffusion-restricting lesions in the bifrontal and bitemporal lobes [Figure 1c]. An EEG showed bitemporal sharp waves. Available autoantibodies associated with autoimmune encephalitis, including NMDA receptor, AMPA, LGI-1, CASPR2, GABAB, and GAD, yielded negative results. Paraneoplastic neuronal antibodies were negative. A repeat CSF analysis, including the IgG index and virology screening, was negative. He received Methylprednisolone pulse therapy and Intravenous Immunoglobulin and improved remarkably. Immunotherapy-responsive waxing and waning brain lesions further suggested a diagnosis of autoimmune encephalitis. GABAA receptor autoimmunity was suspected at this time based on distinctive MRI features, but testing was not available during this period. He was started on second-line immunotherapy with Rituximab; however, he opted to discontinue treatment. This precipitated a seizure on August 31st. MRI showed a single new lesion in the left temporal lobe with resolution of previously seen lesions [Figure 1d]. Rituximab was restarted. An MRI of the brain on December 17, 2022, showed complete resolution of all lesions [Figure 1e]. He again discontinued Rituximab and remained symptom-free for several months. However, on March 6, 2024, he developed word-finding difficulty. A repeat MRI showed three new lesions in both the temporal and left frontal lobes [Figure 1f]. At this point, as GABAA Receptor antibody testing was now available, both serum and CSF samples were analyzed. GABAA receptor antibodies were detected in the CSF and serum of the patient by a tissue-based indirect immunofluorescence assay performed in the Neuroimmunology laboratory of Amrita Institute of Medical Sciences, Kochi, Kerala, India. Serum specimens were positive at a dilution of 1:40 [Figure 2], and CSF specimens were positive at a dilution of 1:4 [Figure 3]. Rituximab was restarted and he was completely asymptomatic at his last follow-up. Thus, he had four relapses, each responding well to immunotherapy.Figure 2: GABAA Receptor staining characteristics by tissue-based Immunofluorescence assay (serum): Distribution of immune reactivity of patient's serum 1: 40 dilution on fresh frozen cryostat mouse brain sections at 10 pm. Neuropil staining observed as "green" fluorescence when incubated with anti-Human IgG (Fab)2 conjugated with FITC by indirect immunofluorescence assay (IFA). (a-f). GABAA Receptor Ab reaction in Mouse Frozen section in Serum (1:40 dilution) sample stained by Anti Human FITC. Magnification 10x OB: Olfactory Bulb, BG: Basal Ganglia, HI: Hippocampus, CX: Cerebral cortex GL: Granular Layer, PC: Purkinje Cell Layer, MFigure 3: GABAA Receptor staining characteristics by tissue-based Immunofluorescence assay (CSF): Distribution of immune reactivity of patient's CSF 1: 4 dilution on fresh frozen cryostat mouse brain sections at 10pm. Neuropil staining observed as "green" fluorescence when incubated with anti-Human IgG (Fab)2 conjugated with FITC by indirect immunofluorescence assay (IFA). (a-f): GABAA Receptor Ab reaction in Mouse Frozen section in CSF (1:4) sample stained by Anti Human FITC Magnification 10x OB: Olfactory Bulb, BG: Basal Ganglia, HI: Hippocampus, CX: Cerebral cortex GL: Granular Layer, PC: Purkinje Cell Layer, MAnti-GABAA Receptor encephalitis has a characteristic imaging pattern. It features multifocal hyperintensities in the grey and white matter on FLAIR and T2 MRI scans, without mass effect, diffusion restriction, or contrast enhancement.[3] Another study showed multifocal lesions in 77% of cases (20/26). Diffusion restriction was seen in only one out of 26 patients, and only two out of 26 patients showed contrast enhancement.[2] Our case corroborated previously reported imaging findings. While relapses in Anti-GABAA receptor encephalitis were reported to be 12% in the review by Guo CY et al.[4], Saptola M et al.[2] observed no relapses in their analysis of 26 patients. Shorter follow-up and uninterrupted immunotherapy might explain the lack of relapses. O’Connor et al.[5] reported multiple relapses in one out of four cases. Our patient had a total of four relapses over two and a half years, each responding well to immunotherapy. Three relapses occurred when immunotherapy was discontinued due to either physician discretion or patient preference. One severe relapse necessitated ICU admission. The availability of antibody testing for rare diseases like Anti-GABAA receptor encephalitis is limited, especially in resource-poor areas. In our case, diagnosis was delayed due to the initial unavailability of testing. To the best of our knowledge, this is the only reported case of Anti-GABAA receptor encephalitis in India. In conclusion, any patient with suspected autoimmune encephalitis having such a unique imaging phenotype should be directly investigated for GABAA Receptor antibodies. The clinician must be watchful for relapses if immunotherapy is withdrawn. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.