Abstract
Objective
Autoimmune epilepsy (AES) is increasingly recognized as a condition in patients with epilepsy of unknown etiology. Early immunotherapy improves outcomes; however, data on its prevalence and the frequency of anti‐neural/neuronal antibodies in Asian populations remain scarce. Herein, we aim to (1) evaluate the frequency of serum anti‐neuronal/neural antibodies, (2) compare the clinical characteristics between seropositive and seronegative patients, and (3) explore the clinical utility of the ACES (Antibodies Contributing to Focal Epilepsy Signs and Symptoms) score in diagnosing AES in a cohort of Singaporean patients with focal epilepsy of unknown etiology.
Methods
This was a cross‐sectional, observational, single‐center study performed at the National Neuroscience Institute (NNI), Singapore. Patients with focal epilepsy of unknown etiology were prospectively recruited. Serum samples were tested using both cell‐based (CBA) and immunohistochemistry (IHC)‐based tissue assays. Patients were classified into definite, probable, and possible AES based on antibody positivity. Clinical characteristics, MRI and EEG findings, and ACES scores were analyzed.
Results
One‐hundred‐and‐twenty patients were included. Nine patients (7.50%) were seropositive on IHC, while 2 patients (1.67%) were positive on CBA (1 with concomitant IHC positivity, 1 without). In total, 10 patients (8.33%) were seropositive on IHC and/or CBA. Seropositivity was associated with female sex, autoimmune and psychiatric comorbidities, behavioral changes, speech problems, abnormal MRI brain, medial temporal T2/FLAIR hyperintensities, and ACES ≥2. AUC of ACES for seropositivity was 0.878; sensitivity and specificity were 60.0% and 96.4% respectively at a cutoff of ≥2.
Significance
The frequency of specific, well‐characterized anti‐neural/neuronal antibodies in focal epilepsy of unknown etiology is low in Singapore. Distinct clinico‐radiological features, in particular those that comprise the ACES score, were useful in identifying AES. Their presence should prompt antibody testing to confirm the diagnosis of AES for earlier immunotherapy.
Plain Language Summary
This study performed in Singapore investigated autoimmune epilepsy in people who have focal epilepsy with no clear cause. Researchers tested blood samples for antibodies linked to autoimmune epilepsy and compared clinical features between patients with and without these antibodies. Out of 120 patients, 10 had positive antibody tests—they were more often female, more likely to have autoimmune or psychiatric conditions, behavior or speech problems, and abnormal brain MRI findings. An ACES score of 2 or more was useful for identifying patients with antibodies. Using these clinical features can help clinicians decide who should be tested and treated earlier.
Keywords: antibodies, autoimmune, epilepsy, neural, neuronal, singapore
Key points.
One hundred and twenty patients with focal epilepsy of unknown etiology were included.
Nine patients (7.50%) were seropositive on IHC; 2 (1.67%) were positive on CBA (1 anti‐GAD, 1 anti‐GABAb).
The frequency of specific, well‐characterized anti‐neural/neuronal antibodies in focal epilepsy of unknown etiology is low in Singapore.
Distinct clinico‐radiological features, in particular those that comprise the ACES score, were useful in identifying AES.
Their presence should prompt antibody testing to confirm the diagnosis of AES for earlier immunotherapy.
1. INTRODUCTION
Epilepsy is a common neurological disorder, affecting approximately 50 million people worldwide. 1 Among these, nearly 50% have epilepsy of unknown etiology, where an underlying cause remains unidentified despite extensive investigations. 2 , 3 , 4 , 5 In recent years, an autoimmune etiology has been identified for a subset of these cases, with the clinical term “autoimmune epilepsy (AES)” first introduced in 2002. 6 , 7 , 8 In 2017, the International League Against Epilepsy (ILAE) formally recognized autoimmunity as a distinct etiology, introducing the term autoimmune‐associated epilepsy (AAE) to describe patients with chronic epilepsy secondary to autoimmune brain diseases, differentiating them from acute symptomatic seizures from autoimmune encephalitis (AIE). 9 , 10 However, differentiating AAE from acute symptomatic seizures due to autoimmune encephalitis remains clinically challenging; therefore, the general term AES is still widely used. 10 , 11 , 12
In addition to the presence of epilepsy, the diagnosis of AES is supported by the identification of specific anti‐neuronal autoantibodies such as antibodies against N‐methyl‐D‐aspartate receptor (NMDAR), leucine‐rich glioma‐inactivated 1 (LGI1) protein, and glutamic acid decarboxylase 65 (GAD65). 13 , 14 , 15 The early detection of these autoantibodies is crucial, as prompt immunotherapy initiation improves outcomes, particularly for patients resistant to antiseizure medications (ASMs). 13 , 16 , 17 Indeed, patients with acute symptomatic seizures secondary to autoimmune encephalitis such as anti‐NMDAR‐ and anti‐LGI1‐antibody encephalitis have been shown to respond remarkably to immunotherapy rather than to ASMs alone. 16 , 18
Previous studies investigating AES have primarily relied on cell‐based assays (CBA)—these enable the detection of known specific anti‐neuronal autoantibodies but are only able to test for a limited set of antibodies. 16 , 19 Tissue‐based assays, either using immunohistochemistry (IHC) or indirect immunofluorescence (IIF), are not antigen specific but allow a broader spectrum of anti‐neural antibodies, both known and unknown, to be detected. Recent diagnostic algorithms have called for the use of IHC as an initial screening tool, and subsequently, confirming their presence on CBA. 20 Indeed, the combination of CBA and tissue‐based assay has been shown to enhance the diagnostic sensitivity of anti‐neuronal autoantibodies in people with epilepsy of unknown etiology, compared to using CBA alone. 19 , 21 , 22 , 23
In parallel, several scoring systems have been developed to aid in AES diagnosis. 24 The APE2 (Antibody Prevalence in Epilepsy and Encephalopathy) score incorporates clinical features and diagnostic data, with a score of ≥4 showing good sensitivity (82.6%) and specificity (82.0%) for detecting CNS autoantibodies in the original publication. 7 , 15 The ACES (Antibodies Contributing to Focal Epilepsy Signs and Symptoms) score was designed to identify focal epilepsy patients with a possible autoimmune basis by assigning 1 point for 6 clinical features—temporal MRI hyperintensities, autoimmune history, behavioral changes, autonomic symptoms, cognitive symptoms, and speech problems. A score of ≥2 is considered significant, with a reported sensitivity of 100.0% and specificity of 84.9% for antibody‐defined AES, and hence may be useful for identifying patients warranting antibody testing. 6
Despite the increasing recognition of AES, data on its prevalence and the frequency of anti‐neuronal antibodies in Asian populations remain scarce. 25 , 26 In this study on a cohort of people with focal epilepsy of unknown etiology from Singapore, we sought to (1) evaluate the frequency of serum anti‐neuronal/neural antibodies using CBA and IHC assays, (2) compare the clinical characteristics between seropositive and seronegative patients, and (3) explore the clinical utility of the ACES score in diagnosing AES.
2. MATERIALS AND METHODS
2.1. Study population
This was a cross‐sectional, single‐center study with prospective patient recruitment conducted at the National Neuroscience Institute (NNI) under local ethics approval (CIRB 2016/2469). People with focal epilepsy of unknown etiology were recruited from the NNI epilepsy service—this patient population was chosen as previous reports have demonstrated a higher frequency of anti‐neuronal autoantibodies in focal epilepsy. 6 , 27 The ILAE guidelines were used to define focal seizures and an unknown etiology of epilepsy. 28 Only patients who were able to provide written informed consent were included as they had epilepsy as the main clinical phenotype and were the most likely to have unrecognized AES due to the lack of fulminant encephalitic symptoms that are typically observed in AIE. Other inclusion criteria were age ≥18 years old and having had at least 1 seizure in the last 1 year prior to recruitment.
Exclusion criteria were: known structural, metabolic or genetic explanation for epilepsy (with the exception of mesial temporal sclerosis as this might be a sequelae of AIE), 29 known epilepsy/electro‐clinical syndromes (e.g., Juvenile Myoclonic Epilepsy, Juvenile Absence Epilepsy, Lennox–Gastaut Syndrome), symptomatic seizures secondary to severe electrolyte disturbances, alcoholism, medication‐induced, central nervous system infections or severe sepsis, seizures as manifestation of progressive neurodegenerative disorders (e.g., advanced dementias, leukodystrophies), and psychogenic non‐epileptic seizures only without concomitant epileptic seizures. Individuals with active cancer were also excluded.
Demographic and clinical (age of epilepsy onset, semiology of seizures, frequency of seizures in the preceding 3 months and 6 months prior to recruitment, number of ASMs at time of recruitment, electroencephalogram [EEG], and MRI brain) data were collected. Information sources included participants and/or their caregivers, attending physicians, and electronic medical records. ACES scores were calculated from electronic medical records—this scoring system was selected because the study cohort comprised patients with focal epilepsy, consistent with the population in which the ACES score was originally developed.
2.2. Cell‐based assay testing for specific anti‐neuronal antibodies
Blood was collected from all participants, centrifuged at 1500g for 10 min, and serum aliquoted and stored at −80°C until antibody testing. Commercial cell‐based indirect immunofluorescence assays (Euroimmun AG, Lübeck, Germany) were used. The autoantibodies tested included: anti‐LGI1, anticontactin‐associated protein‐like 2 (CASPR2), anti‐NMDAR, anti‐α‐amino‐3‐hydroxy‐5‐methyl‐4‐isoxazolepropionic acid receptor (AMPAR), anti‐γ‐aminobutyric acid receptor B (GABAb), anti‐dipeptidyl‐peptidase‐like protein 6 (DPPX), and anti‐GAD65 antibodies. Testing was focused on neuronal surface antibodies (with the exception of anti‐GAD65) as these are likely to be pathogenic and their identification represents the potential for immunotherapy responsiveness, in contrast to the intracellular “onconeuronal” antibodies associated with paraneoplastic neurological syndromes that are usually refractory to immunotherapy with accompanying poor prognosis. All CBAs were performed by trained medical laboratory technologists according to the manufacturer's instructions and interpreted independently by two trained investigators (T.Y. and K.T.) who were blinded to clinical data. 30 Differing interpretations of the antibody assays were resolved by consensus after discussion and reviewing of the assay slides.
2.3. Immunohistochemistry tissue assay testing for anti‐neural antibodies
IHC tissue assay was performed according to previously published methods that have been optimized to detect neuronal surface antibodies. 21 , 31 , 32 , 33 Brains from non‐perfused adult rats were retrieved, fixed briefly in 4% paraformaldehyde at 4°C for 1 h, and then cryoprotected in 40% sucrose for 24 h. Following this, the rat brains were embedded in OCT cutting media and frozen at −80°C till sectioning. A cryostat was used for tissue sectioning to obtain 8 μm sagittal brain sections for immunostaining.
For immunostaining, slides were first left at room temperature for 1 h before use. A 5 min wash with 0.01 M phosphate buffer solution (PBS) was first done with subsequent blocking of endogenous peroxidases (using 0.03% hydrogen peroxide with no shaking and protection from light) for 10 min. After one 5 min wash with PBS, the slides were loaded onto Sequenza staining racks and washed 2 more times, 5 min each. The slides were then blocked with 5% normal goat serum (NGS) for 30 min. One‐hundred‐and‐thirty microliters of diluted serum (1:50) in blocking buffer (1% NGS) was then added to each slide and incubated overnight at 4°C. On the following day, three 5 min PBS washes were performed, following which, a biotinylated secondary antibody (goat antihuman IgG, BA‐3000, Vector Laboratories Inc., Newark, CA, USA) at 1:2000 in blocking buffer was added for 60 min. The last steps involved autoantibody visualization using an avidin‐biotin peroxidase (VECTASTAIN® Elite® ABC‐HRP kit, PK‐6100, Vector Laboratories Inc., Newark, CA, USA) method with diaminobenzidine (DAB) (Dako, Agilent Technologies, Santa Clara, CA, USA) as the chromogen. Finally, the slides were progressively dehydrated for 3 min each in 70% ethanol, 95% ethanol, 100% ethanol (3 times) and subsequently xylene, and were mounted and left for a day.
Sera of AIE patients with known specific anti‐neuronal autoantibodies (i.e., those detected on the CBA) available from the NNI neuroscience clinical laboratory were used as positive controls. Positive controls were also purchased commercially (Euroimmun AG, Lübeck, Germany) for this purpose. The slides were interpreted independently by 2 investigators (J.M.M.T. and T.Y.) and differing interpretations were resolved by consensus.
Patients were stratified into (1) definite AES if a known antibody was detected in CBA and confirmed on IHC, (2) probable AES if a known anti‐neuronal antibody was detected in CBA but negative on IHC, (3) possible AES if only IHC was positive, and (4) non‐AES if both CBA and IHC were negative. This classification was adapted from the study that established the ACES score. 6
2.4. Statistical analysis
A sample size of 110 participants was calculated using an estimated autoantibody prevalence of 8% (derived from previous studies using CBA), with a precision of 5% and finite population of 1000. 8 , 26 , 34 Descriptive comparative analysis of at least possible AES (i.e., seropositive on CBA and/or IHC) and non‐AES (i.e., seronegative on both assays) patients were performed with Mann–Whitney U test and Chi‐Square test where appropriate. ACES scores were calculated for all patients, and a ROC (receiver‐operating characteristics) curve was plotted with AUC (area under curve) calculation for at least possible AES. Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of the ACES score were calculated at different thresholds. Cases with missing data were excluded from analysis of the respective variables, and from ACES score analyses if the missing data was from any component of the score. All analyses were performed with R statistical software version 4.4.2 (R foundation for statistical computing, Vienna, Austria) and SPSS Statistics 30 (IBM, Armonk, NY, USA). Two‐tailed p values of <0.05 were considered statistically significant.
3. RESULTS
3.1. Demographic and clinical data
One‐hundred‐and‐twenty participants with focal epilepsy of unknown etiology were sequentially recruited from January 2017 to January 2020–113 (94.2%) from the outpatient clinics and 7 (5.8%) from the inpatient wards, reflecting the typical patient distribution in the NNI epilepsy service. Table 1 shows the demographic and clinical data of the study population. Sixty‐five (54.2%) were female and the median age at recruitment was 43.0 years‐old (Q1, Q3: 33.8, 53.3). The majority were Chinese (79.2%), and the remaining comprised Malay, Indian, Eurasian and other races. Four (3.3%) patients had a history of Type 2 Diabetes Mellitus, 9 (7.5%) had a history of autoimmune diseases, 7 (5.8%) had a history of malignancy, and 4 (3.3%) had psychiatric comorbidity. Ten (8.3%) patients had cognitive symptoms, 9 (7.5%) had behavioral changes, 6 (5.0%) had autonomic symptoms, and 3 (2.5%) had speech‐related symptoms.
TABLE 1.
Demographic and clinical characteristics of the study population.
| n = 120 (%) | |
|---|---|
| Demographics | |
| Age (years‐old), median (Q1, Q3) | 43.0 (33.8, 53.3) |
| Female, no. (%) | 65 (54.2) |
| Educational level (years), median (Q1, Q3) | 12.0 (10, 15) |
| Race | |
| Chinese, no. (%) | 95 (79.2) |
| Malay, no. (%) | 9 (7.5) |
| Indian, no. (%) | 11 (9.2) |
| Eurasian, no. (%) | 1 (0.8) |
| Others, no. (%) | 4 (3.3) |
| Past medical history | |
| Type 2 diabetes mellitus, no. (%) | 4 (3.3) |
| Autoimmune condition(s), no. (%) | 9 (7.5) |
| Cancer, no. (%) | 7 (5.8) |
| Psychiatric condition(s), no. (%) | 4 (3.3) |
| Epilepsy history | |
| Epilepsy duration (years), median (Q1, Q3) | 7.5 (2.0, 14.0) |
| Age of seizure onset (years‐old), median (Q1, Q3) | 30.0 (17.0, 42.3) |
| Age diagnosed with epilepsy (years‐old), median (Q1, Q3) | 31.5 (17.8, 44.3) |
| Number of seizures in the last 6 months, median (Q1, Q3) | 3.0 (1.0, 11.3) |
| Number of seizures in the last 3 months, median (Q1, Q3) | 2.0 (1.0, 5.3) |
| Previous status epilepticus, no. (%) | 9 (7.5) |
| Medically refractory epilepsy, no. (%) | 28 (23.3) |
| Total no. of current antiepileptic medications, median (Q1, Q3) | 1.0 (1.0, 2.0) |
| Seizure semiology a | |
| Focal seizure without impairment of consciousness, no. (%) | 30 (25.0) |
| Focal seizure with impairment of consciousness, no. (%) | 89 (74.2) |
| Focal seizure evolving to bilateral convulsive seizure, no. (%) | 16 (13.3) |
| Faciobrachial dystonic seizure, no. (%) | 0 (0.0) |
| Others, no. (%) | 63 (52.5) |
| Other associated clinical features | |
| Cognitive symptoms, no. (%) | 10 (8.3) |
| Behavioral changes, no. (%) | 9 (7.5) |
| Autonomic symptoms, no. (%) | 6 (5.0) |
| Speech problems, no. (%) | 3 (2.5) |
| ACES score ≥2, no. (%) | 10/118 (8.5) b |
| MRI brain findings b | |
| Abnormal, no. (%) | 44/118 (37.3) |
| Medial temporal T2/FLAIR hyperintensities, no. (%) | 8/118 (6.8) |
| Right, no. (%) | 2/8 (25.0) |
| Left, no. (%) | 4/8 (50.0) |
| Bilateral, no. (%) | 2/8 (25.0) |
| Mesial temporal sclerosis, no. (%) | 8/118 (6.8) |
| Right, no. (%) | 4/8 (50.0) |
| Left, no. (%) | 4/8 (50.0) |
| Bilateral, no. (%) | 0/8 (0.0) |
| Parenchymal contrast enhancement, no. (%) | 0/118 (0.0) |
| Leptomeningeal enhancement, no. (%) | 3/118 (2.5) |
| Others, no. (%) | 29/118 (24.6) |
| EEG findings c | |
| Abnormal, no. (%) | 93/115 (80.9) |
| Localization of EEG abnormalities | |
| Left temporal, no. (%) | 32/93 (34.4) |
| Right temporal, no. (%) | 19/93 (20.4) |
| Bilateral temporal, no. (%) | 12/93 (12.9) |
| Extratemporal, no. (%) | 30/93 (32.3) |
| Epileptiform activity, no. (%) | 67/115 (58.3) |
| Electrographic seizure, no. (%) | 15/115 (13.0) |
| Slowing, no. (%) | 59/115 (51.3) |
| MRI‐EEG concordance | |
| Abnormal MRI, no. (%) | 18/44 (41.9) |
| Medial temporal T2/FLAIR hyperintensities, no. (%) | 4/8 (50.0) |
| Mesial temporal sclerosis, no. (%) | 5/8 (62.5) |
Note: Six factors contribute to the ACES score, consisting of autoimmune diseases, behavioral changes, cognitive symptoms, speech problems, autonomic symptoms, and temporal magnetic resonance imaging hyperintensities. A score ≥2 has been previously reported to favor an autoimmune etiology of epilepsy. 6
Numbers add up to more than 120 as some patients have more than one type of seizure semiology.
Two patients had missing MRI data, precluding the calculation of their ACES score.
Five patients had missing data for EEG.
The median duration of epilepsy was 7.5 years (Q1, Q3: 2.0, 14.0). Age of seizure onset was 30.0 years (Q1, Q3: 17.0, 42.3), and median age of epilepsy diagnosis was 31.5 years (Q1, Q3: 17.8, 44.3). The median number of seizures in the last 6 months before recruitment was 3.0 (Q1, Q3: 1.0, 11.3). Nine (7.5%) patients had a previous status epilepticus, and 28 (23.3%) had medically refractory epilepsy. Patients were treated with a median of 1.0 ASM (Q1, Q3: 1.0, 2.0) at recruitment. With regards to seizure semiology, 30 (25.0%) patients had focal seizure without impairment of consciousness, 89 (74.2%) had focal seizure with impairment of consciousness, 16 (13.3%) had focal seizure evolving to bilateral tonic–clonic seizures, and 63 (52.5%) had other concurrent presentations such as generalized tonic–clonic seizures. None had faciobrachial dystonic seizures.
Abnormal MRI findings were present in 44 (37.3%) patients—8 (6.8%) had medial temporal T2/FLAIR hyperintensities, 8 (6.8%) had mesial temporal sclerosis, and 3 (2.5%) had leptomeningeal enhancement. Twenty‐nine (24.6%) participants had non‐specific MRI findings. Among those with medial temporal T2/FLAIR hyperintensities (n = 8), 2 (25.0%) were right‐sided, 4 (50.0%) were left‐sided, and 2 (25.0%) were bilateral. For those with mesial temporal sclerosis (n = 8), 4 (50.0%) were right‐sided, 4 (50.0%) were left‐sided, and 0 (0.0%) were bilateral. Abnormal EEG findings were present in 93 (80.9%) of patients; 67 (58.3%) had epileptiform activity, 15 (13.0%) with electrographic seizures, and 59 (51.3%) had slowing. With regards to the lateralization of the EEG abnormalities, 32 (34.4%) were left temporal, 19 (20.4%) were right temporal, 12 (12.9%) were bilateral temporal, and 30 (32.3%) were extra‐temporal. Of note, 5 patients had no EEG data, and 2 patients had no MRI data.
Among patients with abnormal MRI findings (n = 44), 18 (41.9%) demonstrated MRI‐EEG concordance; in those with medial temporal T2/FLAIR hyperintensities (n = 8), 4 (50.0%) showed concordance, whereas 5 (62.5%) patients with mesial temporal sclerosis (n = 8) had concordance.
ACES score was derived from available clinical and investigational data. This was not calculated for 2 patients with missing MRI data. Of note, missing EEG data had no impact on the ACES score calculation as EEG findings do not contribute to the score. Ten (10/118, 8.5%) patients had an ACES score of ≥2, implying a high likelihood of definite AES. 6
3.2. Anti‐neuronal/neural autoantibodies testing with CBA and IHC assay
Table 2 shows the IHC and CBA test results. Nine of 120 (7.50%) sera samples were positive on the IHC tissue assay—1 (0.83%) showed a staining pattern consistent with anti‐GAD and 8 demonstrated positive reactivity comprising neuropil and neuronal patterns. Two of 120 (1.67%) were positive on CBA—1 (0.83%) anti‐GAD65 positive (also positive on IHC assay) and 1 (0.83%) anti‐GABAb positive (but negative on IHC). Illustrative results on IHC and CBA are shown in Figure 1. In total, 10 (8.33%) patients tested positive for either test—1 (0.83%) patient met the criteria for definite AES (anti‐GAD65 positive on both IHC and CBA); 1 (0.83%) patient was classified as probable AES (anti‐GABAb positive on CBA, negative on IHC); and 8 (6.67%) patients were categorized as possible AES (IHC staining demonstrated reactivity but CBA negative for known antibodies).
TABLE 2.
Antineural/neuronal antibody testing.
| Test type | IHC | CBA |
|---|---|---|
| Total positive, no. (%) | 9 (7.50) | 2 (1.67) |
| Anti‐NMDAR, no. (%) | 0 (0) | 0 (0) |
| Anti‐CASPR2, no. (%) | 0 (0) | 0 (0) |
| Anti‐LGI1, no. (%) | 0 (0) | 0 (0) |
| Anti‐AMPAR, no. (%) | 0 (0) | 0 (0) |
| Anti‐GABAb, no. (%) | 0 (0) | 1 (0.83) a |
| Anti‐DPPX, no. (%) | 0 (0) | 0 (0) |
| Anti‐GAD65, no. (%) | 1 (0.83) b | 1 (0.83) b |
| Others, no. (%) | 8 (5.83) c | – |
| Negative, no. (%) | 111 (92.50) | 118 (98.30) |
The sample was positive on CBA but not on IHC.
The same sample was positive on both IHC and CBA.
Eight samples demonstrated reactivity on IHC but not on CBA.
FIGURE 1.
Representative IHC and CBA immunostaining patterns. (A) to (H) show staining patterns (top panel: Rat hippocampus; lower panel: Rat cerebellum) on in‐house IHC tissue assay. (A) Serum negative control. (B) serum anti‐NMDAR, (C) serum anti‐LGI1, and (D) serum anti‐GABAb positive controls showing faint neuropil pattern. Sera from (E) Patient 3 and (F) Patient 5 (refer to Table 3 for patient ID) demonstrating neuronal staining pattern, especially of the cerebellar granular layer. (G) Patient 2's serum showing no reactivity on IHC (but positive for anti‐GABAb on CBA). (H) Serum from Patient 1 demonstrates anti‐GAD65 pattern, with typical staining surrounding the cerebellar Purkinje cells. (I) to (L) demonstrate indirect immunofluorescence staining using commercial CBA and primate cerebellar tissue. (I) Serum‐negative control on CBA. (J) Patient 2's serum showing anti‐GABAb positivity on CBA (but negative on IHC). (K) Serum‐negative control on tissue IIF. (L) Patient 1's serum demonstrates blotchy immunofluorescence signal in the cerebellar granular layer and (M) on positivity on CBA.
3.3. Clinical details of patients with definite, probable, and possible AES
The only patient classified as definite AES was a 57‐year‐old Chinese female with MRI negative medically refractory focal epilepsy diagnosed at 32 years old (Table 3, patient 1). Four months prior to enrolling in this study, she presented with dizziness and was noted to have down‐beating nystagmus on examination. She tested positive for anti‐GAD antibodies on both IHC and CBA (Figure 1, panels H and M). Anti‐GAD positivity was further confirmed with commercial immunoblot and immunofluorescence tissue assay (Euroimmun AG, Lübeck, Germany) (Figure 1, panel L) as well as on ELISA with levels of >2000 U/mL (normal: <5 U/mL) (RSR Ltd., Cardiff, UK). This confirmed the diagnosis of anti‐GAD autoimmunity manifesting with epilepsy and cerebellar ataxia—the ages at which she developed these clinical syndromes were consistent with previous reports, that is, 20 to 30s for anti‐GAD epilepsy and 50s for anti‐GAD cerebellar ataxia. 35 , 36 Anti‐GAD antibodies remained positive on CBA 6 months after initial testing. She was treated with pulsed intravenous methylprednisolone followed by oral prednisolone taper which led to an improvement of her ataxia but not her seizures. Her cerebellar ataxia recurred 4 months after cessation of prednisolone; this responded well to repeat pulsed intravenous methylprednisolone, followed by maintenance oral prednisolone and intravenous rituximab. Seizure frequency remained unchanged at ~2 seizures every 6 months.
TABLE 3.
Clinical characteristics of the 10 seropositive patients.
| ID | Age | Gender | Race | Past medical history | Age diagnosed with epilepsy | Semiology | Frequency of seizures in the last 6 months | Medically refractory epilepsy | Total no. of antiseizure medications at inclusion | MRI findings | EEG findings | CBA | IHC staining | ACES score | Positive ACES criteria | Diagnostic group |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 57 | Female | C |
Autoimmune Psychiatric |
32 | FwIC | 2 | Yes | 2 | Others | Normal | Anti‐GAD65 | Anti‐GAD65 pattern | 2 |
Autoimmune Behavioral |
Definite AES |
| 2 | 35 | Female | C | None | 35 | FwoIC, Others | 3 | No | 1 | Normal | Normal | Anti‐GABAb | Negative | 1 | Autonomic |
Probable AES |
| 3 | 31 | Female | C |
Autoimmune Psychiatric Cancer |
30 | FwoIC | 1 | No | 1 | Others | EA | Negative |
DGML CBGL |
2 |
Autoimmune Behavioral |
Possible AES |
| 4 | 37 | Female | C | None | 28 | FwIC, Others | 1 | No | 1 | Normal | Normal | Negative |
DGML CBML |
1 | Autonomic |
Possible AES |
| 5 | 33 | Female | C | Autoimmune | 20 | FwIC, Others | 4 | No | 0 | Others | S | Negative |
DGML CBGL |
1 | Autoimmune |
Possible AES |
| 6 | 48 | Female | C | Psychiatric | 48 | FwIC, Others | 8 | No | 1 | MTH | Normal | Negative |
DGML CBGL |
2 |
MTT Behavioral |
Possible AES |
| 7 | 58 | Female | C | None | 6 | FwoIC, Others | 3 | No | 1 | Others | S | Negative |
DGML CBML CBGL |
0 | None |
Possible AES |
| 8 | 41 | Female | C | None | 35 | FwIC, Others | 6 | No | 1 | Normal | EA, S | Negative |
DGML CBML CBGL |
3 |
Cognitive Behavioral Speech |
Possible AES |
| 9 | 24 | Male | C | None | 23 | FwIC, Others | >100 | Yes | 1 | MTH | EA, S | Negative |
DGML CBGL |
2 |
MTTH Behavioral |
Possible AES |
| 10 | 21 | Female | C | None | 17 | FwIC, Others | 16 | Yes | 3 | MTH, LE |
EA, S, ES |
Negative |
DGML CBGL |
2 |
MTTH Speech |
Possible AES |
Abbreviations: C, Chinese; CBGL, cerebellar granular layer; CBML, cerebellar molecular layer.; DGML, dentate gyrus molecular layer; EA, epileptiform activity; ES, electrographic seizure; FwIC, focal seizure with impairment of consciousness; FwoIC, focal seizure without impairment of consciousness; LE, leptomeningeal enhancement; MTH, medial temporal T2/FLAIR hyperintensities; S, slowing.
The patient with probable AES was a 35‐year‐old Chinese female with no notable past medical history who first presented with seizures 9 months before study recruitment that was well controlled with sodium valproate (Table 3, Patient 2). She tested positive for anti‐GABAb on CBA (but not on IHC) (Figure 1, panels G and J). She declined immunotherapy following discussion with her managing neurologist. Anti‐GABAb antibodies remained positive on CBA 6 months after initial testing. She remained seizure‐free on sodium valproate.
The clinical characteristics of all 10 seropositive patients are summarized in Table 3.
3.4. Comparison of seropositive and seronegative patients
Comparative analysis of patients with at least possible AES (i.e., seropositive on CBA and/or IHC) (n = 10) and non‐AES patients (n = 110) is shown in Table 4. Patients with at least possible AES were found to be more often female (p = 0.0205), had a history of autoimmune conditions (p = 0.0272) and psychiatric conditions (p = 0.00163), behavioral changes (p = 0.000144), speech problems (p = 0.0181), an ACES score of ≥2 (p = 0.0000118), abnormal MRI brain findings (p = 0.0383), and medial temporal T2/FLAIR hyperintensities on MRI (p = 0.0199). No statistically significant differences in age, race, EEG findings and other characteristics were observed. Multivariable analysis was not performed due to small samples sizes. Likewise, statistical analysis was not undertaken using definite AES (n = 1) and at least probable AES (n = 2) as distinct categories due to the small number of patients.
TABLE 4.
Comparison between at least possible AES patients vs. non‐AES patients.
| Variables | Patient groups | p‐value | |
|---|---|---|---|
| At least possible AES (n = 10) | Non‐AES (n = 110) | ||
| Demographics | |||
| Age, median (Q1, Q3) | 36.0 (31.5, 46.3) | 44.0 (34.0, 53.8) | 0.185 |
| Female, no. (%) | 9 (90.0) | 56 (50.9) | 0.0205 |
| Educational level, median (Q1, Q3) | 12.5 (11.25, 15.75) | 12.0 (10.0, 15.0) | 0.419 |
| Race | |||
| Chinese, no. (%) | 10 (100.0) | 85 (77.3) | 0.119 |
| Malay, no. (%) | 0 (0.0) | 9 (8.2) | 1.00 |
| Indian, no. (%) | 0 (0.0) | 11 (10.0) | 1.00 |
| Eurasian, no. (%) | 0 (0.0) | 1 (0.91) | 1.00 |
| Others, no. (%) | 0 (0.0) | 4 (3.6) | 1.00 |
| Past medical history | |||
| Type 2 diabetes mellitus, no. (%) | 0 (0.0) | 4 (3.6) | 1.00 |
| Autoimmune condition(s), no. (%) | 3 (30.0) | 6 (5.5) | 0.0272 |
| Cancer, no. (%) | 1 (10.0) | 6 (5.5) | 0.465 |
| Psychiatric condition(s), no. (%) | 3 (30.0) | 1 (0.9) | 0.00163 |
| Epilepsy history | |||
| Epilepsy duration in years, median (Q1, Q3) | 5.0 (1.0, 12.0) | 8.0 (2.0, 14.0) | 0.517 |
| Age of seizure onset, median (Q1, Q3) | 21.0 (16.3, 30.8) | 30.0 (17.0, 44.0) | 0.149 |
| Age diagnosed with epilepsy, median (Q1, Q3) | 29.0 (20.8, 34.3) | 32.0 (17.3, 45.8) | 0.401 |
| Number of seizures in the last 6 months, median (Q1, Q3) | 3.5 (2.3, 7.5) | 3.0 (1.0, 11.0) | 0.622 |
| Number of seizures in the last 3 months, median (Q1, Q3) | 2.0 (0.3, 4.0) | 2.0 (1.0, 5.8) | 0.916 |
| Previous status epilepticus, no. (%) | 1 (10.0) | 8 (7.3) | 0.556 |
| Medically refractory epilepsy, no. (%) | 3 (30.0) | 25 (22.7) | 0.697 |
| Total no. of current antiepileptic medications, median (Q1, Q3) | 1.0 (1.0, 1.0) | 1.0 (1.0, 2.0) | 0.226 |
| Seizure semiology | |||
| Focal seizure without impairment of consciousness, no. (%) | 4 (40.0) | 26 (23.6) | 0.266 |
| Focal seizure with impairment of consciousness, no. (%) | 7 (70.0) | 82 (74.5) | 0.717 |
| Focal seizure evolving to bilateral convulsive seizure, no. (%) | 0 (0.0) | 16 (14.5) | 0.355 |
| Faciobrachial dystonic seizure, no. (%) | 0 (0.0) | 0 (0.0) | ‐ |
| Others, no. (%) | 8 (80.0) | 55 (50.0) | 0.0990 |
| Other associated clinical features | |||
| Cognitive symptoms, no. (%) | 1 (10.0) | 9 (8.2) | 0.596 |
| Behavioral changes, no. (%) | 5 (50.0) | 4 (3.6) | 0.000144 |
| Autonomic symptoms, no. (%) | 2 (20.0) | 4 (3.6) | 0.0786 |
| Speech problems, no. (%) | 2 (20.0) | 1 (0.9) | 0.0181 |
| ACES score ≥2 a , no. (%) | 6 (60.0) | 4/108 (3.7) | 0.0000118 |
| MRI brain findings a | |||
| Abnormal, no. (%) | 7 (70.0) | 37/108 (34.3) | 0.0383 |
| Medial temporal T2/FLAIR hyperintensities, no. (%) | 3 (30.0) | 5/108 (4.6) | 0.0199 |
| Right, no. (%) | 0/3 (0.0) | 2/5 (40.0) | 0.464 |
| Left, no. (%) | 2/3 (66.7) | 2/5 (40.0) | 1.000 |
| Bilateral, no. (%) | 1/3 (33.3) | 1/5 (20.0) | 1.000 |
| Mesial temporal sclerosis, no. (%) | 0 (0.0) | 8/108 (7.4) | 1.00 |
| Right, no. (%) | 0/0 (0.0) | 4/8 (50.0) | – |
| Left, no. (%) | 0/0 (0.0) | 4/8 (50.0) | – |
| Bilateral, no. (%) | 0/0 (0.0) | 0/8 (0.0) | – |
| Parenchymal contrast enhancement, no. (%) | 0 (0.0) | 0/108 (0.0) | – |
| Leptomeningeal enhancement, no. (%) | 1 (10.0) | 2/108 (1.9) | 0.235 |
| Others, no. (%) | 4 (40.0) | 25/108 (23.1) | 0.258 |
| EEG findings b | |||
| Abnormal, no. (%) | 6 (60.0) | 87/105 (82.9) | 0.0964 |
| Localization of EEG abnormalities | |||
| Left temporal, no. (%) | 3/6 (50.0) | 29/87 (33.3) | 0.411 |
| Right temporal, no. (%) | 0/6 (0.0) | 19/87 (21.8) | 0.340 |
| Bilateral temporal, no. (%) | 2/6 (33.3) | 10/87 (11.5) | 0.171 |
| Extratemporal, no. (%) | 1/6 (16.7) | 29/87 (33.3) | 0.660 |
| Epileptiform activity, no. (%) | 4 (40.0) | 63/105 (60.0) | 0.316 |
| Electrographic seizure, no. (%) | 1 (10.0) | 14/105 (13.3) | 1.00 |
| Slowing, no. (%) | 5 (50.0) | 54/105 (51.4) | 1.00 |
| MRI‐EEG concordance | |||
| Abnormal MRI, no. (%) | 3/7 (42.9) | 15/37 (40.5) | 1.000 |
| Medial temporal/T2 FLAIR hyperintensities, no. (%) | 1/3 (33.0) | 3/5 (60.0) | 1.000 |
| MTS, no. (%) | 0/0 (0.0) | 5/8 (62.5) | – |
Note: Bold values imply statistical significance.
Two patients in the seronegative group had missing MRI data, precluding the calculation of their ACES score.
Five patients had missing data for EEG.
There were no significant differences in MRI or EEG localization between seropositive and seronegative patients. Likewise, there was no significant difference in the proportion of patients with MRI‐EEG concordance between seropositive patients with abnormal MRI findings (3/7, 42.9%) (Patients 3, 7, and 9 in Table 3) versus seronegative patients with abnormal MRI (15/37, 40.5%) (p = 1.000).
3.5. Utility of ACES score for identifying at least positive AES
AUC of the ACES score for at least possible AES patients (i.e., seropositive on CBA and/or IHC) (n = 10) was 0.878. Using the previously published ACES score threshold of ≥2, this was found to be 60.0% sensitive and 96.4% specific, with a PPV of 60.0% and an NPV of 96.3% for at least possible AES patients (Table 5). 6 A threshold of ACES ≥3 was 10.0% (95% CI 0.3–44.5%) sensitive and 100.0% (95% CI 96.6–100.0%) specific, with a PPV of 100.0% (2.5–100.0%) and an NPV of 92.3 (90.7–93.7%), while a threshold of ACES ≥1 was 90.0% (95% CI 55.5–99.8%) sensitive and 71.6% (95% CI 62.1–79.8%) specific, with a PPV of 22.5% (95% CI 16.8–29.4%) and an NPV of 98.7% (95% CI 92.4–99.8%). Diagnostic performance of ACES score ≥2 for definite AES and for at least probable AES are also shown in Table 5.
TABLE 5.
Diagnostic properties of ACES score stratified by definite, at least probable, and at least possible AES.
| ACES ≥2 | Definite AES | At least probable AES | At least possible AES |
|---|---|---|---|
| Sensitivity (%) (95% CI) | 100.0 (2.5–100.0) | 50.0 (1.3–98.8) | 60.0 (26.2–87.8) |
| Specificity (%) (95% CI) | 92.3 (86.0–96.4) | 92.2 (85.8–96.4) | 96.4 (90.8–99.0) |
| PPV (%) (95% CI) | 10.0 (5.6–17.2) | 10.0 (2.4–33.7) | 60.0 (33.6–81.6) |
| NPV (%) (95% CI) | 100.0 (96.7–100.0) | 99.1 (96.4–99.8) | 96.3 (92.4–98.2) |
Abbreviation: CI, confidence interval.
4. DISCUSSION
In this study, we used a combination of IHC and CBA techniques to evaluate the frequency of serum anti‐neuronal/neural autoantibodies in a cohort of patients with focal epilepsy of unknown etiology. We found that the frequency of seropositivity (on CBA and/or IHC), that is, at least possible AES, was 8.3% (10/120), with a low frequency of definite (CBA and IHC positive) (0.83%, 1/120) and probable (CBA positive but IHC negative) (0.83%, 1/120) AES. The 1 definite AES patient was positive for anti‐GAD65 and 1 probable AES patient was positive for anti‐GABAb, which are both well‐established autoantibodies in AES pathogenesis. 37 The absence of anti‐GABAb immunostaining on IHC could be explained by lower autoantibody titers in serum, which is a known cause of false negativity with IHC. 22 Indeed, anti‐GABAb has been reported to be more difficult to detect in serum compared to cerebrospinal fluid (CSF) regardless of method. 19
Our findings are comparable with recent studies employing IHC alongside confirmatory methods, reporting prevalence rates between 3.4% and 6.8%. 6 , 38 , 39 This low frequency is further contextualized by Nóbrega‐Jr et al., who found neuronal autoantibodies in only 3 of 100 patients with mesial temporal lobe epilepsy with hippocampal sclerosis (MTLE‐HS), a subgroup suspected to have higher autoimmune burden, with no statistically significant difference from controls, suggesting that autoantibody positivity remains uncommon even in epilepsy populations with structural substrates conventionally linked to autoimmune etiology. 40 However, other studies have reported higher frequencies. A study by Duan et al. utilizing IHC screening on rat brain with confirmatory CBA in both serum and CSF reported a prevalence of 26.6%, 25 and Bozzeti et al. demonstrated a prevalence of 31.5% utilizing IHC with rat brain, CBA and immunoblot. 41 This discordance may be due to lack of CSF testing as well as the strict case selection in our cohort where patients had epilepsy as the main clinical phenotype and were able to provide written informed consent (hence unlikely to have significant cognitive impairment), rather than having epilepsy as one of the manifestations of AIE. Sakamoto et al. reported a positivity rate of 23.3% (14 out of 60 patients) with initial IHC screening (although only 10 of these patients had subsequent confirmed antibodies), which is higher than the frequency (at least possible AES, 8.33%) observed in our cohort—this again may be due to differences in case selection criteria. 20 Prior meta‐analyses have estimated AES prevalence among epilepsy patients of unknown etiology to be approximately 4.0% and 7.5%—our findings reinforces that while the frequency of AES is low, identifying cases is still beneficial considering the potential for reversibility with immunotherapy. 26 , 42 Future research assessing the value of immunotherapy not just in definite AES but also in probable and possible AES patients will be useful to clarify if such a treatment approach is justified.
The interpretation of isolated serum antibody positivity however requires caution, particularly in patients lacking additional clinical, radiological, or electrophysical features suggestive of AES. While neuronal surface antibodies detected by CBA may support an autoimmune etiology, isolated seropositivity especially on IHC may occasionally represent non‐pathogenic or incidental immune reactivity. Clinical context therefore remains essential when determining the significance of antibody findings and the potential role of immunotherapy.
We found several clinical characteristics that were associated with at least possible AES compared to non‐AES patients: female sex, history of autoimmune and psychiatric conditions, behavioral changes and speech problems, abnormal MRI brain findings, and medial temporal T2/FLAIR hyperintensities on MRI. Indeed, 4 of these features overlapped with those contributing to the ACES score, thus it is unsurprising that an ACES score of ≥2 was also a discriminatory variable. Our findings are consistent with previous reports, notwithstanding some minor differences such as female sex not being significant in previous studies, 6 , 7 although Smith et al. reported a significant association between female sex and anti‐LGI1 antibody‐associated AES. 43 In summary, the presence of these clinic‐radiological features in patients with focal epilepsy, especially in combination, should raise suspicion of AES and antibody testing should be performed to confirm the diagnosis.
We also assessed the diagnostic performance of the ACES score for AES in our cohort. Given the small number of definite and probable AES patients in our cohort, interpretation of sensitivity indices for these 2 diagnostic groups should be performed with caution. However, we observed that the specificity of ACES ≥2 remained high across definite, at least probable, and at least possible AES, suggesting that this threshold can help to reliably rule in AES in patients with focal epilepsy of unknown etiology. It is of note that tests with high specificity are usually associated with high PPV; however, it was not observed in our cohort, which is likely due to the low frequency of AES. AUC of the ACES score for at least possible AES was 0.878, indicating good discriminative ability. We found that a threshold of ACES ≥1 had the highest sensitivity of 90.0% (and specificity of 72.2%) for at least possible AES, indicating a potential role for this additional threshold to select patients for further antibody testing in our patient population.
Our study has several limitations. First, only serum was tested instead of both serum and CSF. This could have resulted in a lower frequency of antibody positivity. It is known that while autoantibodies can be present in both CSF and serum, CSF testing offers higher sensitivity for some antibodies, especially anti‐NMDAR. 19 However, our cohort comprised patients with focal epilepsy (without overt encephalitis) who would not have undergone lumbar puncture for clinical purposes, representative of patients managed in a predominantly outpatient‐based epilepsy service. Indeed, in the study that developed the ACES score, only 8% of patients had available CSF for analysis. 6 Second, using rodent brain as the IHC substrate, antibodies that recognize only human epitopes may not be detected or visualized. 19 There is also a possibility of false positivity on IHC due to non‐specific binding—the use of another testing method such as neuronal culture may have increase specificity. Third, while none of the patients were encephalitic at the time of recruitment as reflected by the intact mental capacity to provide informed consent, formal cognitive assessment was not done in this study, which could have led to under‐reporting of cognitive symptoms. Recall bias could also have been a factor during data extraction for parameters such as seizure frequency and history of behavioral or speech symptoms.
In conclusion, we elucidated the frequency and clinical characteristics of AES in patients with focal epilepsy of unknown etiology in Singapore, emphasizing the importance of recognizing AES in this population. Our findings reinforce the diagnostic value of distinct clinical and radiological features (in particular those that form the ACES score) in identifying AES—the presence of these characteristics should prompt antibody testing to confirm the diagnosis of AES for earlier immunotherapeutic intervention. Future research focusing on standardized case definition, detailed clinical phenotyping, comprehensive antibody testing in prospective, multi‐center cohorts will be crucial in refining the diagnosis and management of AES.
FUNDING INFORMATION
This study was funded by an AM‐ETHOS grant awarded by Duke‐NUS Medical School to Seong Jin Park, an NNI RIE2025 CG Pilot grant awarded to Jeanne May May Tan, and a National Medical Research Council (Singapore) Research Training Fellowship grant awarded to Tianrong Yeo. Jeanne May May Tan is funded by the SingHealth Duke‐NUS Clinician Investigator Advancement Award (06/2024/CIVA/04‐A04). Tianrong Yeo is funded by the National Medical Research Council (Singapore) Clinician Scientist Award (CSAINV24jul‐0004).
CONFLICT OF INTEREST STATEMENT
Jeanne May May Tan has received honoraria from AstraZeneca, Merck, and Roche for speaker's fees and travel grants, and research grants from the National Medical Research Council (Singapore) and SingHealth. Nigel Choon Kiat Tan owns stock in Pfizer, Johnson & Johnson, Roche, and Zimmer. Kevin Tan has received travel grants and compensation from Merck, Eisai, Roche, AstraZeneca, and Alexion for consulting services and educational activities. Tianrong Yeo has received honoraria from Argenx, ASNA, AstraZeneca, decodeMR, Edanz Pharma, Euroimmun AG, Merck, Novartis, Roche, and Terumo BCT for consulting services and speaker's fees, and research grants from the National Medical Research Council (Singapore), AstraZeneca, and Roche. He has also received travel grants and awards from PACTRIMS, ACTRIMS, ECTRIMS, Orebro University, UCB, and Merck. The other co‐authors declare no competing interest. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
ETHICS STATEMENT
This study was conducted in Singapore under local ethics approval (CIRB 2016/2469). Written informed consent was obtained from every study participant.
ACKNOWLEDGMENTS
The authors would like to thank all patients who participated in this study.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study were obtained from National Neuroscience Institute (NNI), Singapore and are not publicly available due to institutional privacy policies and ethical restrictions. Data may be made available from the corresponding author upon reasonable request and with permission from NNI.
REFERENCES
- 1. Global, regional, and national burden of epilepsy, 1990–2016: a systematic analysis for the global burden of disease study 2016. Lancet Neurol. 2019;18(4):357–375. 10.1016/s1474-4422(18)30454-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Fiest KM, Sauro KM, Wiebe S, Patten SB, Kwon CS, Dykeman J, et al. Prevalence and incidence of epilepsy: a systematic review and meta‐analysis of international studies. Neurology. 2017;88(3):296–303. 10.1212/wnl.0000000000003509 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Beghi E. The epidemiology of epilepsy. Neuroepidemiology. 2020;54(2):185–191. 10.1159/000503831 [DOI] [PubMed] [Google Scholar]
- 4. Delaney S, Fitzsimons M, White M, Power K, O' Donoghue S, Kilbride R, et al. Analysis of the aetiology of epilepsy in 3,216 adult patients attending a tertiary referral center enabled by an electronic patient record. Seizure. 2020;81:332–337. 10.1016/j.seizure.2020.08.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Symonds JD, Elliott KS, Shetty J, Armstrong M, Brunklaus A, Cutcutache I, et al. Early childhood epilepsies: epidemiology, classification, aetiology, and socio‐economic determinants. Brain. 2021;144(9):2879–2891. 10.1093/brain/awab162 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. de Bruijn M, Bastiaansen AEM, Mojzisova H, et al. Antibodies contributing to focal epilepsy signs and symptoms score. Ann Neurol. 2021;89(4):698–710. 10.1002/ana.26013 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Dubey D, Alqallaf A, Hays R, Freeman M, Chen K, Ding K, et al. Neurological autoantibody prevalence in epilepsy of unknown etiology. JAMA Neurol. 2017;74(4):397–402. 10.1001/jamaneurol.2016.5429 [DOI] [PubMed] [Google Scholar]
- 8. Brenner T, Sills GJ, Hart Y, Howell S, Waters P, Brodie MJ, et al. Prevalence of neurologic autoantibodies in cohorts of patients with new and established epilepsy. Epilepsia. 2013;54(6):1028–1035. 10.1111/epi.12127 [DOI] [PubMed] [Google Scholar]
- 9. Scheffer IE, Berkovic S, Capovilla G, Connolly MB, French J, Guilhoto L, et al. ILAE classification of the epilepsies: position paper of the ILAE Commission for Classification and Terminology. Epilepsia. 2017;58(4):512–521. 10.1111/epi.13709 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Steriade C, Britton J, Dale RC, Gadoth A, Irani SR, Linnoila J, et al. Acute symptomatic seizures secondary to autoimmune encephalitis and autoimmune‐associated epilepsy: conceptual definitions. Epilepsia. 2020;61(7):1341–1351. 10.1111/epi.16571 [DOI] [PubMed] [Google Scholar]
- 11. Rada A, Bien CG. What is autoimmune encephalitis‐associated epilepsy? Proposal of a practical definition. Epilepsia. 2023;64(9):2249–2255. 10.1111/epi.17699 [DOI] [PubMed] [Google Scholar]
- 12. Melzer N, Rosenow F. Autoimmune‐associated epilepsy ‐ a challenging concept. Seizure. 2024;128:20–23. 10.1016/j.seizure.2024.05.017 [DOI] [PubMed] [Google Scholar]
- 13. Dubey D, Singh J, Britton JW, Pittock SJ, Flanagan EP, Lennon VA, et al. Predictive models in the diagnosis and treatment of autoimmune epilepsy. Epilepsia. 2017;58(7):1181–1189. 10.1111/epi.13797 [DOI] [PubMed] [Google Scholar]
- 14. Daif A, Lukas RV, Issa NP, Javed A, VanHaerents S, Reder AT, et al. Antiglutamic acid decarboxylase 65 (GAD65) antibody‐associated epilepsy. Epilepsy Behav. 2018;80:331–336. 10.1016/j.yebeh.2018.01.021 [DOI] [PubMed] [Google Scholar]
- 15. Husari KS, Dubey D. Autoimmune Epilepsy. Neurotherapeutics. 2019;16(3):685–702. 10.1007/s13311-019-00750-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Toledano M, Britton JW, McKeon A, Shin C, Lennon VA, Quek AML, et al. Utility of an immunotherapy trial in evaluating patients with presumed autoimmune epilepsy. Neurology. 2014;82(18):1578–1586. 10.1212/wnl.0000000000000383 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Iorio R, Assenza G, Tombini M, Colicchio G, Della Marca G, Benvenga A, et al. The detection of neural autoantibodies in patients with antiepileptic‐drug‐resistant epilepsy predicts response to immunotherapy. Eur J Neurol. 2015;22(1):70–78. 10.1111/ene.12529 [DOI] [PubMed] [Google Scholar]
- 18. de Bruijn M, van Sonderen A, van Coevorden‐Hameete MH, et al. Evaluation of seizure treatment in anti‐LGI1, anti‐NMDAR, and anti‐GABA(B)R encephalitis. Neurology. 2019;92(19):e2185‐e2196. 10.1212/wnl.0000000000007475 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Ricken G, Schwaiger C, De Simoni D, et al. Detection methods for autoantibodies in suspected autoimmune encephalitis. Front Neurol. 2018;9:841. 10.3389/fneur.2018.00841 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Sakamoto M, Matsumoto R, Shimotake A, Togawa J, Takeyama H, Kobayashi K, et al. Diagnostic value of an algorithm for autoimmune epilepsy in a retrospective cohort. Front Neurol. 2022;13:902157. 10.3389/fneur.2022.902157 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Höftberger R, Sabater L, Marignier R, Aboul‐Enein F, Bernard‐Valnet R, Rauschka H, et al. An optimized immunohistochemistry technique improves NMO‐IgG detection: study comparison with cell‐based assays. PLoS One. 2013;8(11):e79083. 10.1371/journal.pone.0079083 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Nagata N, Kanazawa N, Mitsuhata T, Iizuka M, Nagashima M, Nakamura M, et al. Neuronal surface antigen‐specific immunostaining pattern on a rat brain immunohistochemistry in autoimmune encephalitis. Front Immunol. 2022;13:1066830. 10.3389/fimmu.2022.1066830 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Bien CG, Holtkamp M. “Autoimmune epilepsy”: encephalitis with autoantibodies for epileptologists. Epilepsy Curr. 2017;17(3):134–141. 10.5698/1535-7511.17.3.134 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Karakis I. Merging art with science: a detection score for autoimmune epilepsy. Epilepsy Curr. 2022;22(3):170–172. 10.1177/15357597221086114 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Duan Y, Zhang Z, Zhao X, Cheng H, Zhang S, Guo S, et al. Validation of different predictive scoring scales in patients with new‐onset epileptic seizures or epilepsy related to neuronal surface antibody‐mediated autoimmune encephalitis. Epilepsy Behav. 2023;147:109403. 10.1016/j.yebeh.2023.109403 [DOI] [PubMed] [Google Scholar]
- 26. Cabezudo‐García P, Mena‐Vázquez N, Ciano‐Petersen NL, García‐Martín G, Estivill‐Torrús G, Serrano‐Castro PJ. Prevalence of neural autoantibodies in epilepsy of unknown etiology: systematic review and meta‐analysis. Brain Sci. 2021;11(3):392. 10.3390/brainsci11030392 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Steriade C, Gillinder L, Rickett K, Hartel G, Higdon L, Britton J, et al. Discerning the role of autoimmunity and autoantibodies in epilepsy: a review. JAMA Neurol. 2021;78(11):1383–1390. 10.1001/jamaneurol.2021.3113 [DOI] [PubMed] [Google Scholar]
- 28. Fisher RS, Cross JH, French JA, Higurashi N, Hirsch E, Jansen FE, et al. Operational classification of seizure types by the international league against epilepsy: position paper of the ILAE Commission for Classification and Terminology. Epilepsia. 2017;58(4):522–530. 10.1111/epi.13670 [DOI] [PubMed] [Google Scholar]
- 29. van Sonderen A, Thijs RD, Coenders EC, et al. Anti‐LGI1 encephalitis: clinical syndrome and long‐term follow‐up. Neurology. 2016;87(14):1449–1456. 10.1212/wnl.0000000000003173 [DOI] [PubMed] [Google Scholar]
- 30. Yeo T, Chen Z, Yong KP, Wong PYW, Chai JYH, Tan K. Distinction between anti‐VGKC‐complex seropositive patients with and without anti‐LGI1/CASPR2 antibodies. J Neurol Sci. 2018;391:64–71. 10.1016/j.jns.2018.05.020 [DOI] [PubMed] [Google Scholar]
- 31. Ramberger M, Berretta A, Tan JMM, Sun B, Michael S, Yeo T, et al. Distinctive binding properties of human monoclonal LGI1 autoantibodies determine pathogenic mechanisms. Brain. 2020;143(6):1731–1745. 10.1093/brain/awaa104 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Irani SR, Alexander S, Waters P, Kleopa KA, Pettingill P, Zuliani L, et al. Antibodies to Kv1 potassium channel‐complex proteins leucine‐rich, glioma inactivated 1 protein and contactin‐associated protein‐2 in limbic encephalitis, Morvan's syndrome and acquired neuromyotonia. Brain. 2010;133(9):2734–2748. 10.1093/brain/awq213 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Lai M, Huijbers MG, Lancaster E, Graus F, Bataller L, Balice‐Gordon R, et al. Investigation of LGI1 as the antigen in limbic encephalitis previously attributed to potassium channels: a case series. Lancet Neurol. 2010;9(8):776–785. 10.1016/s1474-4422(10)70137-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Gozubatik‐Celik G, Ozkara C, Ulusoy C, Gunduz A, Delil S, Yeni N, et al. Anti‐neuronal autoantibodies in both drug responsive and resistant focal seizures with unknown cause. Epilepsy Res. 2017;135:131–136. 10.1016/j.eplepsyres.2017.06.008 [DOI] [PubMed] [Google Scholar]
- 35. Tröscher AR, Mair KM, Verdú de Juan L, et al. Temporal lobe epilepsy with GAD antibodies: neurons killed by T cells not by complement membrane attack complex. Brain. 2023;146(4):1436–1452. 10.1093/brain/awac404 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Hadjivassiliou M, Sarrigiannis PG, Shanmugarajah PD, et al. Clinical characteristics and management of 50 patients with anti‐GAD ataxia: gluten‐free diet has a major impact. Cerebellum. 2021;20(2):179–185. 10.1007/s12311-020-01203-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. McGinty RN, Handel A, Moloney T, et al. Clinical features which predict neuronal surface autoantibodies in new‐onset focal epilepsy: implications for immunotherapies. J Neurol Neurosurg Psychiatry. 2021;92(3):291–294. 10.1136/jnnp-2020-325011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Schulz P, Lütt A, Stöcker W, Teegen B, Holtkamp M, Prüss H. High frequency of cerebrospinal fluid autoantibodies in patients with seizures or epilepsies of unknown etiology. Front Neurol. 2023;14:1211812. 10.3389/fneur.2023.1211812 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Zhou K, Zhang L, Shen S, Lin JF, Wang JR, Zhou D, et al. Neurological autoantibody prevalence in chronic epilepsy: clinical and neuropathologic findings. Seizure. 2024;115:28–35. 10.1016/j.seizure.2023.12.018 [DOI] [PubMed] [Google Scholar]
- 40. Nobrega‐Jr AW, Gregory CP, Schlindwein‐Zanini R, Neves FS, Wolf P, Walz R, et al. Mesial temporal lobe epilepsy with hippocampal sclerosis is infrequently associated with neuronal autoantibodies. Epilepsia. 2018;59(9):e152–e156. 10.1111/epi.14534 [DOI] [PubMed] [Google Scholar]
- 41. Bozzetti S, Rossini F, Ferrari S, Delogu R, Cantalupo G, Marchioretto F, et al. Epileptic seizures of suspected autoimmune origin: a multicentre retrospective study. J Neurol Neurosurg Psychiatry. 2020;91(11):1145–1153. 10.1136/jnnp-2020-323841 [DOI] [PubMed] [Google Scholar]
- 42. Chen L, Zhu L, Lu D, Dai S, Han Y, Wu Z, et al. Association between autoimmune encephalitis and epilepsy: systematic review and meta‐analysis. Seizure. 2021;91:346–359. 10.1016/j.seizure.2021.07.005 [DOI] [PubMed] [Google Scholar]
- 43. Smith KM, Dubey D, Liebo GB, Flanagan EP, Britton JW. Clinical course and features of seizures associated with LGI1‐antibody encephalitis. Neurology. 2021;97(11):e1141–e1149. 10.1212/wnl.0000000000012465 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study were obtained from National Neuroscience Institute (NNI), Singapore and are not publicly available due to institutional privacy policies and ethical restrictions. Data may be made available from the corresponding author upon reasonable request and with permission from NNI.