- Email: [email protected]
Autoantibodies to glutamic acid decarboxylase (GAD) in focal and generalized epilepsy: A study on 233 patients
Journal of Neuroimmunology 211 (2009) 120–123
Contents lists available at ScienceDirect
Journal of Neuroimmunology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j n e u r o i m
Short communication
Autoantibodies to glutamic acid decarboxylase (GAD) in focal and generalized epilepsy: A study on 233 patients Luca Errichiello a,⁎, Giuseppe Perruolo b, Angelo Pascarella a, Pietro Formisano b, Carlo Minetti c, Salvatore Striano a, Federico Zara c, Pasquale Striano c a b c
Epilepsy Center, Department of Neurological Sciences, “Federico II” University, Napoli, Italy Department of Molecular Cell Biology and Pathology, “Federico II” University, Napoli, Italy Unit of Muscular and Neurodegenerative Diseases, Istituto “G. Gaslini”, University of Genova, Genova, Italy
a r t i c l e
i n f o
Article history: Received 18 December 2008 Received in revised form 15 April 2009 Accepted 17 April 2009 Keywords: Glutamic acid decarboxylase antibodies Epilepsy Diabetes Autoimmunity
a b s t r a c t Background: Autoantibodies to glutamic acid decarboxylase (GADA) have been associated to a wide range of neurologic conditions, including epilepsy. However, the spectrum of epileptic conditions associated with GADA is not completely established. We aimed to determine the occurrence of GADA in a large series of patients with different epilepsy types. Moreover, we assessed whether specific subgroups of patients are associated to GAD autoimmunity. Methods: GADA were measured by radioimmunoassay in a series of consecutive unselected epileptic patients observed over a 2-years-period. Patients with neuromuscular features, acute or subacute encephalopathic course, cognitive deterioration or psychiatric symptoms were excluded. Results: Two hundred thirty-three patients (121 women, mean age: 29.3 years; range: 6–78) were recruited. There were eighty-three (35.6%) patients with idiopathic (66 generalized, 17 focal) epilepsy; fifty-nine (25.3%) with cryptogenic (52 focal, 7 generalized) epilepsy, and ninety-one (39.0%) with symptomatic (75 focal, 16 generalized) epilepsy. GADA were detected in six (2.58%) patients. Two had idiopathic generalized epilepsy associated with diabetes mellitus type 1 (DM1); the other four patients suffered from cryptogenic temporal epilepsy and no history or signs of DM1. GADA positive patients could not be distinguished by seizure frequency or number of AEDs. However, in these cases, the mean epilepsy duration (8.5 ± 5.0 years) was shorter compared to the other 48 GADA-negative patients with cryptogenic focal epilepsy (17.3 ± 9.6) (p b 0.0001). Conclusions: We confirm that GAD autoimmunity may be associated with some forms of epilepsy. The preferential identification in patients with cryptogenic temporal epilepsy deserves particularly further investigation. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Autoantibodies to glutamic acid decarboxylase (GADA) have been associated with type 1 diabetes mellitus (DM1) and a large variety of neurologic conditions, including stiff-person syndrome, cerebellar ataxia, limbic encephalitis, and other paraneoplastic diseases (Saiz et al., 2008). Although GADA have been reported in up 4% of patients with epilepsy, more frequently exhibiting drugresistant focal seizures (Kwan et al., 2000; Peltola et al., 2000; Sokol et al., 2004; McKnight et al., 2005; Majoie et al., 2006), the spectrum of epileptic conditions associated with GADA is not completely established.
⁎ Corresponding author. Epilepsy Center Department of Neurological Sciences “Federico II” University Via Pansini 5, 80131 Napoli, Italy. E-mail address: [email protected] (L. Errichiello). 0165-5728/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jneuroim.2009.04.010
We aimed to determine the prevalence of GADA in a large series of patients with different epilepsy types, to assess whether specific subgroups of patients are associated to GAD autoimmunity. 2. Patients and methods GAD antibodies were analyzed in consecutive unselected epileptic patients attending the Epilepsy Center at “Federico II” University, Napoli, from April 2006 to April 2008. Informed consent was obtained from the patients or from parents. The study was approved by the local Ethical Committee. Patients showing additional neurological features (i.e., ataxia, cerebellar signs, rigidity, encephalopathic course, cognitive and psychiatric manifestations) indicative of other GADA-associated neurological conditions, were excluded. Epileptic syndromes were classified according to the international recommendations (Commission, 1989). Clinical records were
L. Errichiello et al. / Journal of Neuroimmunology 211 (2009) 120–123 Table 1 Clinical data of the 233 epileptic patients. General features
N (%)
Males Females Mean age, y (range) Median age, y Median age at onset, y (range) Mean antiepileptic drugs, n (range) Coexisting organ-specific autoimmunity Hypo/hyperthyroidism Rheumatoid arthritis Type 1 diabetes mellitus Mixed connettivitis
112 (48.07%) 121 (51.93%) 29.3 (6–78) 28.5 22.3 (3–51) 1.8 (1–4) 31 7 3 2
Epilepsy types Idiopathic epilepsy Focal Generalized Cryptogenic epilepsy Focal Generalized Symptomatic epilepsy Focal Generalized
83 (35.6%) 17 66 59 (25.5%) 52 7 91 (39.0%) 75 16
reviewed to determine age, duration and type of epilepsy, seizure frequency, therapy, and history of autoimmune disorders. Subjects with seizures occurring at least once in a month over a period of two years, despite the treatment with one or more antiepileptic drugs (AEDs), were considered as drug-resistant (Perucca, 1998). GAD65 antibodies were measured with a radioimmunoassay (RIA, CentAK anti-GAD65, Berlin; cut-off point for positivity: 0.9 U/ml). Statistical analysis was performed using Fisher's exact test with Yates' correction. 3. Results Two hundred thirty-three patients (121 women) with different epilepsies were recruited (mean age: 29.3 years, range: 6–78,; median age at epilepsy onset: 22.3 years, 3–51; mean antiepileptic drugs received: 1.8, range: 1–4). Concomitant autoimmune diseases were observed in 43 patients (Table 1). Eighty-three (35.6%) patients had idiopathic (66 generalized, 17 focal) epilepsy; fifty-nine (25.3%) cryptogenic (52 focal, 7 generalized) epilepsy, and ninety-one (39.0%) symptomatic (75 focal, 16 generalized) epilepsy (51 hippocampal sclerosis, 33 cortical dysplasia/dysembryoplastic tumours, 7 meningiomas, 7 haemorrhage and/or trauma). GADA were detected in six (2.58%) patients, four (7.69%) with cryptogenic focal epilepsy and two (3.03%) with idiopathic generalized epilepsy (IGE) (Table 2). These patients showed overall GADA
121
titres (28.2 ± 35.7 U/ml) lower to that observed in a series of 250 diabetic patients (48.1 ± 41.1 U/ml) (p b 0.0001). No elevated GADA titres were found in patients belonging to the other epilepsy types groups. GADA positive patients were negative for anti-islet cell-specific, anti-insulin, anti-protein tyrosine phosphataselike protein, anti-cardiolipin, anti-nuclear, anti-thyroid peroxidase, anti-gliadin and anti-GM1 antibodies. Both patients with IGE suffered also from DM1 requiring insulin treatment, as described elsewhere (Striano et al., 2008). The four GADA-positive patients with focal cryptogenic epilepsy showed electroclinical features of temporal lobe epilepsy (TLE) with normal brain magnetic resonance imaging (MRI) and no history or signs of DM1 or other autoimmune disorders. In addition, SPECT or PET study confirmed the temporal lobe involvement. The clinical features of these patients are reported in the Appendix and summarized in Table 2. GADA positive patients could not be distinguished by seizure frequency or number of AEDs (data not shown). However, in these cases, the mean epilepsy duration (8.5 ± 5.0 years) was shorter compared to the other 48 GADA-negative patients with cryptogenic focal epilepsy (17.3 ± 9.6) (p b 0.0001). 4. Discussion In the last few years, there has been increasing interest in the potential role of GADA in the pathogenesis of epilepsy. In this study, we observed GADA in 2.58% of patients with different epilepsy types, confirming that GAD autoimmunity may be associated with epilepsy. The background frequency of GADA in the general population has been scarcely investigated. However, despite considerable countryspecific variations, it is generally assessed to be around 0.4–1% of the normal population (Aanstoot et al., 1994; Batstra et al., 1999). It is not surprising that most of our patients displayed relatively low GADA titer (1.9–79 U/ml), since higher GADA levels are generally observed in other disease, such as stiff-person syndrome, limbic encephalitis, and other paraneoplastic syndromes (Saiz et al., 2008). Moreover, we used a type of assay (i.e., RIA) that has proved to be the most sensitive and specific for GADA detection (Aanstoot et al., 1994; Batstra et al., 1999), excluding the possibility of representing nonspecific background signals. In the two IGE patients from our series, the presence of GADA could be explained by the coexistence of DM1. Notably, a fourfold increase in DM1 has been recently suggested among IGE (McCorry et al., 2006) and GADA may represent the biological basis of this association, through co-existing autoimmune damage to the pancreas and a possible pancreatic source of the GAD autoantibodies (Striano et al., 2008). However, specific prospective studies should investigate this potential-epidemiologically relevant-association. Nevertheless, the present data confirm that GADA are not observed in patients with isolated IGE.
Table 2 Clinical data of GADA positive patients. Epilepsy type
Idiopathic generalized epilepsy
Cryptogenic focal epilepsy
GADA positive cases Patient ID GADA titre (U/ml) Epilepsy onset (years) Drug-resistance Functional imaging findings Antiepileptic therapy Autoimmune associated disease
2/66 (3.03%) Patient 1 7.12 12 Yes –
Patient 2 6.53 8 Not –
VPA, BDZ, LEV, LTG DM1
VPA DM1
4/52 (7.69%) Patient 1 69.73 13 Yes R temporal lobe hypoperfusion (SPECT) (CBZ), OXC, LEV, TPM Not
Patient 2 78.79 49 Yes L temporal lobe hypometabolism (PET) (PB), CBZ Not
Patient 3 5.65 19 Not L temporal lobe hypometabolism (PET) OXC Not
Patient 4 1.93 10 Yes R temporal hypermetabolism (PET) OXC, LEV Not
SPECT: single photon emission computed tomography; PET: positron emission tomography; DM1: Diabetes mellitus type 1; CBZ: carbamazepine; OXC: oxcarbazepine; PB: phenobarbital; VPA: valproate; BDZ: benzodiazepines; LEV: levetiracetam; LTG: lamotrigine; TPM: topiramate. In parenthesis, past therapy.
122
L. Errichiello et al. / Journal of Neuroimmunology 211 (2009) 120–123
We did not identify GADA in patients with symptomatic epilepsy, including mesial temporal sclerosis, in accordance with previous reports (Kwan et al., 2000; Sokol et al., 2004). In addition, we did not observe increased serum GADA in idiopathic focal and cryptogenic generalized epilepsy; however, the size of these latter groups was too small to draw any definitive conclusions. Indeed, the main finding of this study was the identification of GADA in four (7.69%) subjects within the cryptogenic focal epilepsy group. These patients showed the electroclinical features of TLE. None of them showed acute or subacute onset of epilepsy or evidence of MRI abnormalities, as observed in limbic encephalitis. In addition, none of them showed evidence of autoimmune disorders or other organ-specific autoantibodies. Overall, these data therefore suggest that a subgroup of patients with cryptogenic TLE may elicit an immune response against GAD. In our series, GADA positive patients could not be distinguished by seizure frequency or number of AEDs. However, when analyzing the clinical characteristics of the cryptogenic focal epilepsy group, we observed a shorter disease duration in the GADA positive patients respect the patients with no evidence of GADA autoimmunity. This discrepancy could further emphasize the possible link between autoimmunity and epilepsy in this subgroup of patients. We could not demonstrate the intrathecal synthesis of GADA in cerebrospinal fluid (CSF) of our patients, which would have provided proof of pathogenicity of these antibodies. However, oligoclonal CSF IgG bands are found in most but not all cases with GADA and neurological disorders (Peltola et al., 2000). Moreover, lack of intrathecal synthesis of an antibody does not completely exclude autoimmunity (Vincent, 2008). The clinical significance of raised GADA in epilepsy remains to be still elucidated. As GAD catalyzes the conversion of L-glutamic acid to GABA, the indirect suppression of GABA via GADA could theoretically favour the development of seizures. However, GAD is a cytoplasmic protein, and it is unclear how the antibodies might cause the associated phenotypes. Nevertheless, it has been shown that GAD becomes membrane-associated to synaptic vesicles, through complex formation with the heat shock protein 70 family (Hsu et al., 2000). If GAD is inhibited at the synapses, GABA-concentration may be affected because of the functional link between GABA synthesis and vesicular GABA transportation at the nerve terminals (Hsu et al., 2000). Alternatively, the antibodies may be endocytosed and transported internally to their cytoplasmic antigen (McKnight et al., 2005). It is also presently difficult to explain the preferential occurrence of GADA in specific epilepsy types. A differential susceptibility of certain subgroups of GABAergic neurons to an immune-mediated disturbance, depending on immunodominant epitope specificity, might account for different clinical phenotypes. Additional research should investigate the characterization of immunodominant epitope(s) on the GAD molecule and studies on cell-mediated GAD autoimmunity are needed to compare GAD autoimmunity in epilepsy to other GADAassociated disorders. However, the possibility that GADA are only an unspecific marker of an underlying immune response cannot be completely ruled out at the present. In summary, we confirm that GAD autoimmunity may be associated with some epilepsy types. The preferential identification of GADA within the subgroup of patients with cryptogenic TLE deserves particular attention and further investigation. Disclosure The authors have no conflicts of interest to declare. Acknowledgements We wish to thank Dr. Adriana Dato for kindly contributing to the blood sampling.
Appendix A Here it follows a clinical description of the four patients with cryptogenic temporal lobe epilepsy and increased GADA titres. Patient 1 is a 27-year-old woman (GADA titres: 69.73 U/ml) with drug-resistant TLE from the age of 13. Her seizures were characterized by gastric aura followed by consciousness alteration, paraesthesia of left hemisoma, verbal and oral-chewing automatisms. The initial frequency of the episodes was daily. Her electroencephalograms (EEGs) showed a right temporal focus. MRI was normal. An interictal single photon emission computed tomography (SPECT) revealed right mesiotemporal temporal lobe hypoperfusion. She was initially treated with carbamazepine with poor benefits. Currently, the patient experiences up to 5 episodes per month despite therapy with levetiracetam (3000 mg/day) topiramate (200 mg/day) and oxcarbazepine (1800 mg/day). Patient 2 is a 57-year-old woman (titres of GADA: 78.79 U/ml) with cryptogenic epilepsy started at the age of 49 years with a first nocturnal tonic-clonic seizure. At that time, EEG showed left temporal paroxysmal activity. MRI was unremarkable. A positron emission tomography (PET) study revealed left posterior temporal hypometabolism. Treatment with phenobarbital was ineffective. To date, she reports about two seizures per year despite treatment with carbamazepine, 800 mg/day. Patient 3 is a 20-year-old boy (titres of GADA: 5.65 U/ml) diagnosed with epilepsy at the age of 19 years due to complex partial seizures of temporal lobe origin sometimes followed by secondarily generalization. EEGs showed a left temporal focus and a PET study revealed left temporal lobe hypometabolism. MRI was normal. The patient was controlled with relatively low oxcarbazepine doses (1200 mg/day). Patient 4 is a 20-year-old woman (titres of GADA: 1.93 U/ml) with a history of therapy-resistant TLE from the age of 10. Her seizures were characterized by sudden loss of contact, incomprehensible verbal production, gestual automatisms, dystonic hand movements, and scialorrhea. Seizures were about daily and could sometimes become secondarily generalized. The EEG showed a left temporal focus. MRI was unremarkable. A PET revealed right temporal hypermetabolism. Therapy with oxcarbazepine (1800 mg/day) and levetiracetam (3000 mg/day) resulted in partial seizure control (5–10 seizures/month).
References Aanstoot, H.J., Sigurdsson, E., Jaffe, M., Shi, Y., Christgau, S., Grobbee, D., Bruining, G.J., Molenaar, J.L., Hofman, A., Baekkeskov, S., 1994. Value of antibodies to GAD65 combined with islet cell cytoplasmic antibodies for predicting IDDM in a childhood population. Diabetologia 37, 917–924. Batstra, M.R., van Driel, A., Petersen, J.S., van Donselaar, C.A., van Tol, M.J., Bruining, G.J., Grobbee, D.E., Dyrberg, T., Aanstoot, H.J., 1999. Glutamic acid decarboxylase antibodies in screening for autoimmune diabetes: influence of comorbidity, age, and sex on specificity and threshold values. Clin. Chem. 45, 2269–2272. Commission on Classification and Terminology of the International League Against Epilepsy, 1989. Proposal for revised classification of epilepsies and epileptic syndromes. Epilepsia 30, 389–399. Hsu, C.C., Davis, K.M., Jin, H., Foos, T., Floor, E., Chen, W., Tyburski, J.B., Yang, C.Y., Schloss, J.V., Wu, J.Y., 2000. Association of L-glutamic acid decarboxylase to the 70-kDa heat shock protein as a potential anchoring mechanism to synaptic vesicles. J. Biol. Chem. 275 (27), 20822–20828. Kwan, P., Sills, G.J., Kelly, K., Butler, E., Brodie, M.J., 2000. Glutamic acid decarboxylase autoantibodies in controlled and uncontrolled epilepsy: a pilot study. Epilepsy Res. 42 (2–3), 191–195. Majoie, H.J., de Baets, M., Renier, W., Lang, B., Vincent, A., 2006. Antibodies to voltagegated potassium and calcium channels in epilepsy. Epilepsy Res. 71 (2–3), 135–141. McCorry, D., Nicolson, A., Smith, D., Marson, A., Feltbower, R.G., Chadwick, D.W., 2006. An association between type 1 diabetes and idiopathic generalized epilepsy. Ann. Neurol. 59 (1), 204–206. McKnight, K., Jiang, Y., Hart, Y., Cavey, A., Wroe, S., Blank, M., Shoenfeld, Y., Vincent, A., Palace, J., Lang, B., 2005. Serum antibodies in epilepsy and seizure-associated disorders. Neurology 65, 1730–1736. Peltola, J., Kulmala, P., Isojärvi, J., Saiz, A., Latvala, K., Palmio, J., Savola, K., Knip, M., Keränen, T., Graus, F., 2000. Autoantibodies to glutamic acid decarboxylase in patients with therapy-resistant epilepsy. Neurology 55, 46–50. Perucca, E., 1998. Pharmacoresistance in epilepsy: how should it be defined? CNS Drugs 10 (3), 171–179.
L. Errichiello et al. / Journal of Neuroimmunology 211 (2009) 120–123 Saiz, A., Blanco, Y., Sabater, L., González, F., Bataller, L., Casamitjana, R., Ramió-Torrentà, L., Graus, F., 2008. Spectrum of neurological syndromes associated with glutamic acid decarboxylase antibodies: diagnostic clues for this association. Brain 131, 2553–2563. Sokol, D.K., McIntyre, J.A., Wagenknecht, D.R., Dropcho, E.J., Patel, H., Salanova, V., da Costa, G., 2004. Antiphospholipid and glutamic acid decarboxylase antibodies in patients with focal epilepsy. Neurology 62, 517–518.
123
Striano, P., Perruolo, G., Errichiello, L., Formisano, P., Beguinot, F., Zara, F., Striano, S., 2008. Glutamic acid decarboxylase antibodies in idiopathic generalized epilepsy and type 1 diabetes. Ann. Neurol. 67, 127–128. Vincent, A., 2008. Stiff, twitchy or wobbly: are GAD antibodies pathogenic? Brain 131, 2536–2537.
Contents lists available at ScienceDirect
Journal of Neuroimmunology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j n e u r o i m
Short communication
Autoantibodies to glutamic acid decarboxylase (GAD) in focal and generalized epilepsy: A study on 233 patients Luca Errichiello a,⁎, Giuseppe Perruolo b, Angelo Pascarella a, Pietro Formisano b, Carlo Minetti c, Salvatore Striano a, Federico Zara c, Pasquale Striano c a b c
Epilepsy Center, Department of Neurological Sciences, “Federico II” University, Napoli, Italy Department of Molecular Cell Biology and Pathology, “Federico II” University, Napoli, Italy Unit of Muscular and Neurodegenerative Diseases, Istituto “G. Gaslini”, University of Genova, Genova, Italy
a r t i c l e
i n f o
Article history: Received 18 December 2008 Received in revised form 15 April 2009 Accepted 17 April 2009 Keywords: Glutamic acid decarboxylase antibodies Epilepsy Diabetes Autoimmunity
a b s t r a c t Background: Autoantibodies to glutamic acid decarboxylase (GADA) have been associated to a wide range of neurologic conditions, including epilepsy. However, the spectrum of epileptic conditions associated with GADA is not completely established. We aimed to determine the occurrence of GADA in a large series of patients with different epilepsy types. Moreover, we assessed whether specific subgroups of patients are associated to GAD autoimmunity. Methods: GADA were measured by radioimmunoassay in a series of consecutive unselected epileptic patients observed over a 2-years-period. Patients with neuromuscular features, acute or subacute encephalopathic course, cognitive deterioration or psychiatric symptoms were excluded. Results: Two hundred thirty-three patients (121 women, mean age: 29.3 years; range: 6–78) were recruited. There were eighty-three (35.6%) patients with idiopathic (66 generalized, 17 focal) epilepsy; fifty-nine (25.3%) with cryptogenic (52 focal, 7 generalized) epilepsy, and ninety-one (39.0%) with symptomatic (75 focal, 16 generalized) epilepsy. GADA were detected in six (2.58%) patients. Two had idiopathic generalized epilepsy associated with diabetes mellitus type 1 (DM1); the other four patients suffered from cryptogenic temporal epilepsy and no history or signs of DM1. GADA positive patients could not be distinguished by seizure frequency or number of AEDs. However, in these cases, the mean epilepsy duration (8.5 ± 5.0 years) was shorter compared to the other 48 GADA-negative patients with cryptogenic focal epilepsy (17.3 ± 9.6) (p b 0.0001). Conclusions: We confirm that GAD autoimmunity may be associated with some forms of epilepsy. The preferential identification in patients with cryptogenic temporal epilepsy deserves particularly further investigation. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Autoantibodies to glutamic acid decarboxylase (GADA) have been associated with type 1 diabetes mellitus (DM1) and a large variety of neurologic conditions, including stiff-person syndrome, cerebellar ataxia, limbic encephalitis, and other paraneoplastic diseases (Saiz et al., 2008). Although GADA have been reported in up 4% of patients with epilepsy, more frequently exhibiting drugresistant focal seizures (Kwan et al., 2000; Peltola et al., 2000; Sokol et al., 2004; McKnight et al., 2005; Majoie et al., 2006), the spectrum of epileptic conditions associated with GADA is not completely established.
⁎ Corresponding author. Epilepsy Center Department of Neurological Sciences “Federico II” University Via Pansini 5, 80131 Napoli, Italy. E-mail address: [email protected] (L. Errichiello). 0165-5728/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jneuroim.2009.04.010
We aimed to determine the prevalence of GADA in a large series of patients with different epilepsy types, to assess whether specific subgroups of patients are associated to GAD autoimmunity. 2. Patients and methods GAD antibodies were analyzed in consecutive unselected epileptic patients attending the Epilepsy Center at “Federico II” University, Napoli, from April 2006 to April 2008. Informed consent was obtained from the patients or from parents. The study was approved by the local Ethical Committee. Patients showing additional neurological features (i.e., ataxia, cerebellar signs, rigidity, encephalopathic course, cognitive and psychiatric manifestations) indicative of other GADA-associated neurological conditions, were excluded. Epileptic syndromes were classified according to the international recommendations (Commission, 1989). Clinical records were
L. Errichiello et al. / Journal of Neuroimmunology 211 (2009) 120–123 Table 1 Clinical data of the 233 epileptic patients. General features
N (%)
Males Females Mean age, y (range) Median age, y Median age at onset, y (range) Mean antiepileptic drugs, n (range) Coexisting organ-specific autoimmunity Hypo/hyperthyroidism Rheumatoid arthritis Type 1 diabetes mellitus Mixed connettivitis
112 (48.07%) 121 (51.93%) 29.3 (6–78) 28.5 22.3 (3–51) 1.8 (1–4) 31 7 3 2
Epilepsy types Idiopathic epilepsy Focal Generalized Cryptogenic epilepsy Focal Generalized Symptomatic epilepsy Focal Generalized
83 (35.6%) 17 66 59 (25.5%) 52 7 91 (39.0%) 75 16
reviewed to determine age, duration and type of epilepsy, seizure frequency, therapy, and history of autoimmune disorders. Subjects with seizures occurring at least once in a month over a period of two years, despite the treatment with one or more antiepileptic drugs (AEDs), were considered as drug-resistant (Perucca, 1998). GAD65 antibodies were measured with a radioimmunoassay (RIA, CentAK anti-GAD65, Berlin; cut-off point for positivity: 0.9 U/ml). Statistical analysis was performed using Fisher's exact test with Yates' correction. 3. Results Two hundred thirty-three patients (121 women) with different epilepsies were recruited (mean age: 29.3 years, range: 6–78,; median age at epilepsy onset: 22.3 years, 3–51; mean antiepileptic drugs received: 1.8, range: 1–4). Concomitant autoimmune diseases were observed in 43 patients (Table 1). Eighty-three (35.6%) patients had idiopathic (66 generalized, 17 focal) epilepsy; fifty-nine (25.3%) cryptogenic (52 focal, 7 generalized) epilepsy, and ninety-one (39.0%) symptomatic (75 focal, 16 generalized) epilepsy (51 hippocampal sclerosis, 33 cortical dysplasia/dysembryoplastic tumours, 7 meningiomas, 7 haemorrhage and/or trauma). GADA were detected in six (2.58%) patients, four (7.69%) with cryptogenic focal epilepsy and two (3.03%) with idiopathic generalized epilepsy (IGE) (Table 2). These patients showed overall GADA
121
titres (28.2 ± 35.7 U/ml) lower to that observed in a series of 250 diabetic patients (48.1 ± 41.1 U/ml) (p b 0.0001). No elevated GADA titres were found in patients belonging to the other epilepsy types groups. GADA positive patients were negative for anti-islet cell-specific, anti-insulin, anti-protein tyrosine phosphataselike protein, anti-cardiolipin, anti-nuclear, anti-thyroid peroxidase, anti-gliadin and anti-GM1 antibodies. Both patients with IGE suffered also from DM1 requiring insulin treatment, as described elsewhere (Striano et al., 2008). The four GADA-positive patients with focal cryptogenic epilepsy showed electroclinical features of temporal lobe epilepsy (TLE) with normal brain magnetic resonance imaging (MRI) and no history or signs of DM1 or other autoimmune disorders. In addition, SPECT or PET study confirmed the temporal lobe involvement. The clinical features of these patients are reported in the Appendix and summarized in Table 2. GADA positive patients could not be distinguished by seizure frequency or number of AEDs (data not shown). However, in these cases, the mean epilepsy duration (8.5 ± 5.0 years) was shorter compared to the other 48 GADA-negative patients with cryptogenic focal epilepsy (17.3 ± 9.6) (p b 0.0001). 4. Discussion In the last few years, there has been increasing interest in the potential role of GADA in the pathogenesis of epilepsy. In this study, we observed GADA in 2.58% of patients with different epilepsy types, confirming that GAD autoimmunity may be associated with epilepsy. The background frequency of GADA in the general population has been scarcely investigated. However, despite considerable countryspecific variations, it is generally assessed to be around 0.4–1% of the normal population (Aanstoot et al., 1994; Batstra et al., 1999). It is not surprising that most of our patients displayed relatively low GADA titer (1.9–79 U/ml), since higher GADA levels are generally observed in other disease, such as stiff-person syndrome, limbic encephalitis, and other paraneoplastic syndromes (Saiz et al., 2008). Moreover, we used a type of assay (i.e., RIA) that has proved to be the most sensitive and specific for GADA detection (Aanstoot et al., 1994; Batstra et al., 1999), excluding the possibility of representing nonspecific background signals. In the two IGE patients from our series, the presence of GADA could be explained by the coexistence of DM1. Notably, a fourfold increase in DM1 has been recently suggested among IGE (McCorry et al., 2006) and GADA may represent the biological basis of this association, through co-existing autoimmune damage to the pancreas and a possible pancreatic source of the GAD autoantibodies (Striano et al., 2008). However, specific prospective studies should investigate this potential-epidemiologically relevant-association. Nevertheless, the present data confirm that GADA are not observed in patients with isolated IGE.
Table 2 Clinical data of GADA positive patients. Epilepsy type
Idiopathic generalized epilepsy
Cryptogenic focal epilepsy
GADA positive cases Patient ID GADA titre (U/ml) Epilepsy onset (years) Drug-resistance Functional imaging findings Antiepileptic therapy Autoimmune associated disease
2/66 (3.03%) Patient 1 7.12 12 Yes –
Patient 2 6.53 8 Not –
VPA, BDZ, LEV, LTG DM1
VPA DM1
4/52 (7.69%) Patient 1 69.73 13 Yes R temporal lobe hypoperfusion (SPECT) (CBZ), OXC, LEV, TPM Not
Patient 2 78.79 49 Yes L temporal lobe hypometabolism (PET) (PB), CBZ Not
Patient 3 5.65 19 Not L temporal lobe hypometabolism (PET) OXC Not
Patient 4 1.93 10 Yes R temporal hypermetabolism (PET) OXC, LEV Not
SPECT: single photon emission computed tomography; PET: positron emission tomography; DM1: Diabetes mellitus type 1; CBZ: carbamazepine; OXC: oxcarbazepine; PB: phenobarbital; VPA: valproate; BDZ: benzodiazepines; LEV: levetiracetam; LTG: lamotrigine; TPM: topiramate. In parenthesis, past therapy.
122
L. Errichiello et al. / Journal of Neuroimmunology 211 (2009) 120–123
We did not identify GADA in patients with symptomatic epilepsy, including mesial temporal sclerosis, in accordance with previous reports (Kwan et al., 2000; Sokol et al., 2004). In addition, we did not observe increased serum GADA in idiopathic focal and cryptogenic generalized epilepsy; however, the size of these latter groups was too small to draw any definitive conclusions. Indeed, the main finding of this study was the identification of GADA in four (7.69%) subjects within the cryptogenic focal epilepsy group. These patients showed the electroclinical features of TLE. None of them showed acute or subacute onset of epilepsy or evidence of MRI abnormalities, as observed in limbic encephalitis. In addition, none of them showed evidence of autoimmune disorders or other organ-specific autoantibodies. Overall, these data therefore suggest that a subgroup of patients with cryptogenic TLE may elicit an immune response against GAD. In our series, GADA positive patients could not be distinguished by seizure frequency or number of AEDs. However, when analyzing the clinical characteristics of the cryptogenic focal epilepsy group, we observed a shorter disease duration in the GADA positive patients respect the patients with no evidence of GADA autoimmunity. This discrepancy could further emphasize the possible link between autoimmunity and epilepsy in this subgroup of patients. We could not demonstrate the intrathecal synthesis of GADA in cerebrospinal fluid (CSF) of our patients, which would have provided proof of pathogenicity of these antibodies. However, oligoclonal CSF IgG bands are found in most but not all cases with GADA and neurological disorders (Peltola et al., 2000). Moreover, lack of intrathecal synthesis of an antibody does not completely exclude autoimmunity (Vincent, 2008). The clinical significance of raised GADA in epilepsy remains to be still elucidated. As GAD catalyzes the conversion of L-glutamic acid to GABA, the indirect suppression of GABA via GADA could theoretically favour the development of seizures. However, GAD is a cytoplasmic protein, and it is unclear how the antibodies might cause the associated phenotypes. Nevertheless, it has been shown that GAD becomes membrane-associated to synaptic vesicles, through complex formation with the heat shock protein 70 family (Hsu et al., 2000). If GAD is inhibited at the synapses, GABA-concentration may be affected because of the functional link between GABA synthesis and vesicular GABA transportation at the nerve terminals (Hsu et al., 2000). Alternatively, the antibodies may be endocytosed and transported internally to their cytoplasmic antigen (McKnight et al., 2005). It is also presently difficult to explain the preferential occurrence of GADA in specific epilepsy types. A differential susceptibility of certain subgroups of GABAergic neurons to an immune-mediated disturbance, depending on immunodominant epitope specificity, might account for different clinical phenotypes. Additional research should investigate the characterization of immunodominant epitope(s) on the GAD molecule and studies on cell-mediated GAD autoimmunity are needed to compare GAD autoimmunity in epilepsy to other GADAassociated disorders. However, the possibility that GADA are only an unspecific marker of an underlying immune response cannot be completely ruled out at the present. In summary, we confirm that GAD autoimmunity may be associated with some epilepsy types. The preferential identification of GADA within the subgroup of patients with cryptogenic TLE deserves particular attention and further investigation. Disclosure The authors have no conflicts of interest to declare. Acknowledgements We wish to thank Dr. Adriana Dato for kindly contributing to the blood sampling.
Appendix A Here it follows a clinical description of the four patients with cryptogenic temporal lobe epilepsy and increased GADA titres. Patient 1 is a 27-year-old woman (GADA titres: 69.73 U/ml) with drug-resistant TLE from the age of 13. Her seizures were characterized by gastric aura followed by consciousness alteration, paraesthesia of left hemisoma, verbal and oral-chewing automatisms. The initial frequency of the episodes was daily. Her electroencephalograms (EEGs) showed a right temporal focus. MRI was normal. An interictal single photon emission computed tomography (SPECT) revealed right mesiotemporal temporal lobe hypoperfusion. She was initially treated with carbamazepine with poor benefits. Currently, the patient experiences up to 5 episodes per month despite therapy with levetiracetam (3000 mg/day) topiramate (200 mg/day) and oxcarbazepine (1800 mg/day). Patient 2 is a 57-year-old woman (titres of GADA: 78.79 U/ml) with cryptogenic epilepsy started at the age of 49 years with a first nocturnal tonic-clonic seizure. At that time, EEG showed left temporal paroxysmal activity. MRI was unremarkable. A positron emission tomography (PET) study revealed left posterior temporal hypometabolism. Treatment with phenobarbital was ineffective. To date, she reports about two seizures per year despite treatment with carbamazepine, 800 mg/day. Patient 3 is a 20-year-old boy (titres of GADA: 5.65 U/ml) diagnosed with epilepsy at the age of 19 years due to complex partial seizures of temporal lobe origin sometimes followed by secondarily generalization. EEGs showed a left temporal focus and a PET study revealed left temporal lobe hypometabolism. MRI was normal. The patient was controlled with relatively low oxcarbazepine doses (1200 mg/day). Patient 4 is a 20-year-old woman (titres of GADA: 1.93 U/ml) with a history of therapy-resistant TLE from the age of 10. Her seizures were characterized by sudden loss of contact, incomprehensible verbal production, gestual automatisms, dystonic hand movements, and scialorrhea. Seizures were about daily and could sometimes become secondarily generalized. The EEG showed a left temporal focus. MRI was unremarkable. A PET revealed right temporal hypermetabolism. Therapy with oxcarbazepine (1800 mg/day) and levetiracetam (3000 mg/day) resulted in partial seizure control (5–10 seizures/month).
References Aanstoot, H.J., Sigurdsson, E., Jaffe, M., Shi, Y., Christgau, S., Grobbee, D., Bruining, G.J., Molenaar, J.L., Hofman, A., Baekkeskov, S., 1994. Value of antibodies to GAD65 combined with islet cell cytoplasmic antibodies for predicting IDDM in a childhood population. Diabetologia 37, 917–924. Batstra, M.R., van Driel, A., Petersen, J.S., van Donselaar, C.A., van Tol, M.J., Bruining, G.J., Grobbee, D.E., Dyrberg, T., Aanstoot, H.J., 1999. Glutamic acid decarboxylase antibodies in screening for autoimmune diabetes: influence of comorbidity, age, and sex on specificity and threshold values. Clin. Chem. 45, 2269–2272. Commission on Classification and Terminology of the International League Against Epilepsy, 1989. Proposal for revised classification of epilepsies and epileptic syndromes. Epilepsia 30, 389–399. Hsu, C.C., Davis, K.M., Jin, H., Foos, T., Floor, E., Chen, W., Tyburski, J.B., Yang, C.Y., Schloss, J.V., Wu, J.Y., 2000. Association of L-glutamic acid decarboxylase to the 70-kDa heat shock protein as a potential anchoring mechanism to synaptic vesicles. J. Biol. Chem. 275 (27), 20822–20828. Kwan, P., Sills, G.J., Kelly, K., Butler, E., Brodie, M.J., 2000. Glutamic acid decarboxylase autoantibodies in controlled and uncontrolled epilepsy: a pilot study. Epilepsy Res. 42 (2–3), 191–195. Majoie, H.J., de Baets, M., Renier, W., Lang, B., Vincent, A., 2006. Antibodies to voltagegated potassium and calcium channels in epilepsy. Epilepsy Res. 71 (2–3), 135–141. McCorry, D., Nicolson, A., Smith, D., Marson, A., Feltbower, R.G., Chadwick, D.W., 2006. An association between type 1 diabetes and idiopathic generalized epilepsy. Ann. Neurol. 59 (1), 204–206. McKnight, K., Jiang, Y., Hart, Y., Cavey, A., Wroe, S., Blank, M., Shoenfeld, Y., Vincent, A., Palace, J., Lang, B., 2005. Serum antibodies in epilepsy and seizure-associated disorders. Neurology 65, 1730–1736. Peltola, J., Kulmala, P., Isojärvi, J., Saiz, A., Latvala, K., Palmio, J., Savola, K., Knip, M., Keränen, T., Graus, F., 2000. Autoantibodies to glutamic acid decarboxylase in patients with therapy-resistant epilepsy. Neurology 55, 46–50. Perucca, E., 1998. Pharmacoresistance in epilepsy: how should it be defined? CNS Drugs 10 (3), 171–179.
L. Errichiello et al. / Journal of Neuroimmunology 211 (2009) 120–123 Saiz, A., Blanco, Y., Sabater, L., González, F., Bataller, L., Casamitjana, R., Ramió-Torrentà, L., Graus, F., 2008. Spectrum of neurological syndromes associated with glutamic acid decarboxylase antibodies: diagnostic clues for this association. Brain 131, 2553–2563. Sokol, D.K., McIntyre, J.A., Wagenknecht, D.R., Dropcho, E.J., Patel, H., Salanova, V., da Costa, G., 2004. Antiphospholipid and glutamic acid decarboxylase antibodies in patients with focal epilepsy. Neurology 62, 517–518.
123
Striano, P., Perruolo, G., Errichiello, L., Formisano, P., Beguinot, F., Zara, F., Striano, S., 2008. Glutamic acid decarboxylase antibodies in idiopathic generalized epilepsy and type 1 diabetes. Ann. Neurol. 67, 127–128. Vincent, A., 2008. Stiff, twitchy or wobbly: are GAD antibodies pathogenic? Brain 131, 2536–2537.