Neuropathology and Applied Neurobiology (2014), 40, 645–649
doi: 10.1111/nan.12132
Scientific correspondence
LGI1 autoantibodies associated with cerebellar degeneration We describe a young adult with marked cerebellar dysfunction and encephalopathy, in whom serum leucinerich glioma inactivated 1 (LGI1) autoantibodies were identified. The patchy, severe Purkinje and granular neuronal loss provides a unique pathological correlate for the clinical presentation of LGI1 autoantibody-associated encephalitis, and for the anatomic distribution of the LGI1 antigen within the human cerebellum. Voltage-gated potassium channel (VGKC) complex autoantibodies are described in association with limbic encephalitis and Morvan’s syndrome [1]. The recent demonstration of distinct target antigens within the VGKC complex has further clarified the immunemediated pathogenesis of these clinical syndromes, with autoantibodies against LGI1 [2] and contactin-associated protein related 2 (CASPR2) proteins reported in patients with limbic encephalitis, and CASPR2 autoantibodies identified more frequently in patients with Morvan’s syndrome [3]. Increased availability and application of serological testing has further broadened the disease spectrum, with VGKC complex autoantibodies described in patients with sleep disorders, movement disorders, Creutzfeld-Jakob Disease mimics, and a variety of neurocognitive symptoms [3]. Cerebellar involvement is rarely reported in association with limbic encephalitis [3,4], and has not been described as a prominent feature in patients with LGI1 autoantibodies. The case presented here provides evidence that LGI1 autoantibodies may be seen in association with cerebellar dysfunction, attributable to cerebellar degeneration. The unique pattern of cerebellar Purkinje and granular neuronal loss, together with the marked response to treatment with immunotherapy, distinguishes LGI1 autoantibody-associated cerebellar degeneration from other immune-mediated ataxias. A previously well 18-year-old Caucasian man presented to the emergency department with a two-month history of occipital headache, progressive clumsiness and difficulty walking. There was no history of infectious prodrome, toxic exposure or constitutional symptoms. The © 2014 British Neuropathological Society
neurological examination was remarkable for gazeevoked nystagmus, dysarthria, bilateral appendicular ataxia and truncal ataxia. Magnetic resonance imaging (MRI) of the brain revealed multiple T2-hyperintense lesions isolated to the cerebellar hemispheres (Figure 1A), without radiological evidence of limbic (Figure 1B) or brainstem involvement (Figure 1C). The cerebellar folia enhanced following gadolinium administration with a leptomeningeal pattern (Figure 1D). Cerebrospinal fluid (CSF) analysis revealed cellular pleocytosis [69 white blood cells per high powered field (WBC/hpf), lymphocytic predominance], with mild elevation of protein (0.468 g/L; normal range 0.15–0.45). Oligoclonal bands were detected in CSF as compared with serum. Microbiological studies were negative, including bacterial and fungal cultures and polymerase chain reaction for herpes simplex virus, varizella zoster virus, Epstein-Barr virus and cytomegalovirus. Flow cytometry demonstrated T-cell predominance with polyclonal B-cells; no malignant cells were identified on CSF cytology. Human immunodeficiency virus serology was negative. At the time, the patient’s presentation was felt to be most consistent with a parainfectious autoimmune cerebellitis, demyelination or a lymphoproliferative disorder. Paraneoplastic etiologies were also included in the differential. Over the ensuing 3 days the patient’s cognitive status worsened: he became apathetic with disinhibited behaviours and visual hallucinations. Sinus tachycardia was noted. Pulse methylprednisolone was empirically administered given the deterioration (1 g intravenous daily for 5 days). Appendicular ataxia improved marginally with treatment, however his cognitive symptoms progressed. A repeat MRI brain showed worsening of the cerebellar lesions, with persistent sparing of other brain structures (Figure 1E). A repeat lumbar puncture showed an increase in the lymphocytic pleocytosis (271 WBC/hpf), and an absence of malignant cells. Serum samples were sent for measurement of autoantibodies [NMDA (GluN1), VGKC (LGI1, CASPR2), AMPA-R, amphiphysin, GAD-65, CV2/ CRMP5, Recoverin, SOX1, Titin, Hu, Yo, Ri and PNMA2 (Ma2, Ta); Mitogen Advanced Diagnostics, Calgary, Alberta]. Computerized tomography scans of the chest, abdomen and pelvis, and testicular ultrasound were 645
646
Scientific correspondence
Figure 1. MRI brain. (A) Axial Fluid Attenuated Inversion Recovery (FLAIR) sequence shows multiple hyperintensities in cerebellar hemispheres, more prominent on the right. (B) Axial FLAIR shows normal appearance of the hippocampi. (C) Sagittal FLAIR highlights the normal brainstem in contrast with cerebellar hyperintensities. (D) T1 with gadolinium sequence shows leptomeningeal enhancement along cerebellar folia. (E) Evolution of cerebellar lesions on FLAIR 1 week following presentation, again accompanied by normal appearances of the temporal lobes and brainstem. Arrow points to the area of the cerebellum sampled on biopsy. (F) Resolution of lesions on FLAIR 2 days following treatment, 10 days following E.
normal. Given the non-diagnostic investigations and rapidly worsening clinical condition, biopsy of the right cerebellum was undertaken to investigate for potentially treatable conditions. The site of biopsy is shown by the arrow in Figure 1E. Routinely stained sections showed scanty lymphocytic infiltrates in the meninges overlying cerebellar tissue (Figure 2A) and rare small clusters of reactive lymphocytes in the cerebellar molecular layer (Figure 2B), comprised of a mixture of CD20 immunopositive B-lymphocytes and more abundant CD3 immunopositive T-lymphocytes (Figure 2C). T-lymphocytes were scattered throughout the molecular layer (Figure 2C). The striking © 2014 British Neuropathological Society
abnormality in the sample was patchy cerebellar degeneration affecting both the crests and depths of the folia. The degeneration was characterized by widespread loss of Purkinje cell neurones with Bergmann gliosis (Figure 2A) and patchy, severe loss of the granular neurones, well demonstrated with NeuN immunolabelling (Figure 2D). No features of an infectious encephalitis (microglial nodules, viral inclusions), malignancy or demyelination (on LFB staining; not shown) were identified. SV40 immunolabeling for polyomaviruses was negative. The pathologic interpretation was of relatively widespread mixed lymphocytic infiltrates and Purkinje neuronal loss with patchy severe loss of granular neurones. NAN 2014; 40: 645–649
Scientific correspondence
647
Figure 2. Cerebellar biopsy. (A,B) H&E stained sections of cerebellum showing lymphocytes in the meninges and clustered in the molecular layer (yellow arrows). Cerebellar degeneration characterized by patchy Purkinje cell loss with Bergmann gliosis (blue arrow). Discrete areas of Purkinje (blue arrow) and granular neurone layers (red bar) depletion are seen, with intact white matter (black bar), and intact adjacent granular neurone layers (white bar). (C) CD3 immunolabeling highlighting T lymphocytes in the meninges and molecular layer where they are both clustered and more diffusely scattered. (D) NeuN immunolabelling demonstrates the severe focal granular neuronal loss.
Following the biopsy the patient continued to decline with depressed level of consciousness and persistent generalized myoclonus, with prominent involvement of orofacial musculature. EEG revealed a normal background rhythm with no epileptiform features. Five days after the biopsy, results from the serum autoantibody screen confirmed LGI1 autoantibodies, with medium titers. LGI1 autoantibodies were identified using a commercially available biochip immunoassay (Euroimmun GmbH; Lübeck, Germany). In the biochip assay, tissue sections of rat hippocampus and rat cerebellum were used as © 2014 British Neuropathological Society
a screen for autoantibodies against VGKC-associated proteins, including LGI1 and CASPR2. Monospecific antibody detection was confirmed through the parallel use of HEK293 cells transfected with complementary DNA encoding LGI1 channel constituents. Binding of patient’s autoantibodies was determined and semi-quantified by indirect immunofluorescence attended by appropriate positive and negative controls. Intravenous immunoglobulin (IVIg) was administered (2 g/kg over 2 days). Within 24 h of completing treatment the patient’s level of consciousness improved and his myoNAN 2014; 40: 645–649
648
Scientific correspondence
clonus resolved. Maintenance therapy was provided with prednisone (1 mg/kg/day). Repeat MRI brain 2 days after completing treatment showed dramatic improvement of the cerebellar lesions (Figure 1F). No suspicious metabolically active areas were seen with 18F-fluorodeoxyglucose positron emission tomography (PET) body scan. Electromyography and nerve conduction studies did not show neuromyotonia. Three weeks following diagnosis the patient was discharged with normal sensorium and near-complete resolution of ataxia, with plans to taper prednisone over 6 months. He remained normal at 9-month follow-up, with no clinical evidence of ataxia or cognitive dysfunction. This is a novel case of LGI1-associated encephalitis in a patient who presented with symptoms, signs and neuroimaging evidence of a cerebellar localization, and with neuropathological findings demonstrating selective Purkinje and granular neuronal loss. To our knowledge, this represents the first report of cerebellar degeneration associated with LGI1 autoantibodies. The neuronal loss and remarkable response to immunotherapy exemplifies the importance of early recognition and treatment of antibody-mediated brain disease. Additionally, the favourable outcome reported in this index case confirms that preferential cerebellar involvement may maintain a high potential for recovery with immunotherapy, despite the irreversible cerebellar neuronal loss. Similar findings are reported in descriptions of cerebellar infarction outcomes [5], which likely reflects plasticity within cerebellar circuits. A number of distinct features are useful in distinguishing LGI1 autoantibody-associated cerebellar degeneration from more common presentations of paraneoplastic cerebellar degeneration (PCD). First, no malignancy was found in our case, despite thorough investigations, including PET scan. Second, serial MRI demonstrated multiple evolving gadolinium-enhancing lesions, in contrast to descriptions of normal imaging in the majority of patients presenting with PCD [6]. Third, the dramatic response to immunotherapy in this case stands in contrast to the treatment-refractory progressive course typically described with PCD [7]. Finally, the pathology of this case further establishes LGI1 autoantibody-associated cerebellar degeneration as a unique entity. While granular neuronal loss is described in PCD, extensive Purkinje cell loss usually predominates [8] and inflammatory infiltrates are rare [9]. We suggest that LGI1 autoantibody testing should be considered in the differential diagnosis of © 2014 British Neuropathological Society
patients presenting with subacute ataxia, particularly young patients with encephalopathy. Staining of rat cerebellum against sera of patients with VGKC antibodies and LGI1 antibodies has previously revealed molecular layer [1] and cerebellar pinceau [4] binding, respectively. This contrasts in part with reported LGI1 receptor activity in the molecular layer and synaptic glomeruli of the granular layer in rat cerebellum [10]. The co-discovery of LGI1 autoantibodies and Purkinje and granular neuronal loss provides additional indirect evidence of LGI1 distribution within the human cerebellum. A limitation of our report is that patient serum was not shown to directly bind rat cerebellum. This limitation reflects the reality of practice within the clinical setting where elegant testing utilizing animal tissue to directly demonstrate antibody binding is not always available. Thus, while our findings are readily generalizable and our testing accessible (using commercially available autoantibody testing), further experiments are required to directly demonstrate LGI1-antigen distribution within granular layers of the cerebellum. Our patient initially presented with a cerebellar syndrome rather than symptoms of limbic encephalitis more commonly associated with LGI1 autoantibodies. Ataxia has been described in patients with mutations of the KCNA1 gene (encoding the Kv1.1 subunit of the VGKC) [11], and occasionally in patients with limbic encephalitis associated with VGKC antibodies [4], specifically CASPR2 antibodies [12]. Ataxia has not, however, been described as an early presentation of LGI1 autoantibodyassociated encephalitis. Isolated cerebellar dysfunction may represent a potential harbinger of limbic encephalitis,reminiscent of faciobrachial dystonic seizures [13] in its potential for providing a ‘window of opportunity’ for early diagnosis and initiation of immunotherapy in appropriate patients. On the other hand, there is evidence that the cerebellum is involved with more than motor control, with errors in cognitive processing and emotional control described following cerebellar injury [14]. It is possible that autoantibody-mediated cerebellar dysfunction may explain both the motor and the non-motor symptoms described in our patient, especially given the absence of temporal lobe hyperintensities typically (but not always) seen in limbic encephalitis (Figure 1B) [1]. In conclusion, we suggest that LGI1 autoantibodyassociated cerebellar degeneration may represent a distinct clinical entity that should be included in the differential NAN 2014; 40: 645–649
Scientific correspondence
diagnosis of subacute ataxia, recognizing the potential for neuronal loss and response to immunotherapy.
Acknowledgements We would like to thank the patient and his family for their willingness to participate in this scientific exchange.
Conflict of interest disclosures Claude Steriade has nothing to disclose. Gregory S Day serves as the Clinical Director for the Anti-NMDA-Receptor Encephalitis Foundation (Inc., Canada). The Foundation is supported by private donations. He is the recipient of a Future Leaders in Dementia award, including support for travel (Pfizer Canada). Marvin J Fritzler is the director of Mitogen Advanced Diagnostics Laboratory that performed the autoantibody testing and has received gifts in kind from Euroimmun GmbH, manufacturer of the anti-LGI1 biochip immunoassay. Brian J Murray has nothing to disclose. Julia Keith has nothing to disclose. C. Steriade† G. S. Day† L. Lee† B. J. Murray† M. J. Fritzler‡ J. Keith§ †Division of Neurology, Department of Medicine, Sunnybrook Health Sciences Centre, University of Toronto, §Department of Anatomical Pathology, Sunnybrook Health Sciences Center, University of Toronto, Toronto, Ontario, and ‡Departments of Medicine and Biochemistry and Molecular Biology, University of Calgary, Calgary, Alberta, Canada
References 1 Vincent A, Buckley C, Schott JM, Baker I, Dewar BK, Detert N, Clover L, Parkinson A, Bien CG, Omer S, Lang B, Rossor MN, Palace J. Potassium channel antibody-associated encephalopathy: a potentially immunotherapyresponsive form of limbic encephalitis. Brain 2004; 127: 701–12 2 Lai M, Huijbers MG, Lancaster E, Graus F, Bataller L, Balice-Gordon R, Cowell JK, Dalmau J. Investigation of LGI1 as the antigen in limbic encephalitis previously attributed to potassium channels: a case series. Lancet Neurol 2010; 9: 776–85
© 2014 British Neuropathological Society
649
3 Klein CJ, Lennon VA, Aston PA, McKeon A, O’Toole O, Quek A, Pittock SJ. Insights from LGI1 and CASPR2 potassium channel complex autoantibody subtyping. JAMA Neurol 2013; 70: 229–34 4 Irani SR, Alexander S, Waters P, Kleopa KA, Pettingill P, Zuliani L, Peles E, buckely C, Lang B, Vincent A. Antibodies to Kv1 potassium channel-complex proteins leucine-rich, glioma inactivated 1 protein and contactinassociated protein-2 in limbic encephalitis, Mrovan’s syndrome and acquired neuromyotonia. Brain 2010; 133: 2734–48 5 Kelly PJ, Stein J, Shafqat S, Eskey C, Doherty D, Chang Y, Kurina A, Furiet KL. Functional recovery after rehabilitation for cerebellar stroke. Stroke 2001; 32: 530–4 6 Dalmau J, Rosenfeld MR. Paraneoplastic syndromes of the CNS. Lancet Neurol 2008; 7: 327–40 7 Shams’ili S, Grefkens J, de Leeuw B, van den Bent M, Hooijkaas H, van der Holt B, Vecht C, Sillevis Smitt P. Paraneoplastic cerebellar degeneration associated with antineuronal antibodies: analysis of 50 cases. Brain 2003; 126: 1409–18 8 Verschuuren J, Chuang L, Rosenblum MK, Lieberman F, Pryor A, Posner JB, Dalmau J. Inflammatory infiltrates and complete absence of Purkinje cells in anti-Yoassociated paraneoplastic cerebelllar degeneration. Acta Neuropathol 1996; 91: 519–25 9 Scaravilli F, An SF, Groves M, Thom M. The neuropathology of paraneoplastic syndromes. Brain Pathol 1999; 9: 251–60 10 Fukata Y, Adesnik H, Iwanaga T, Bredt DS, Nicoll RA, Fukata M. Epilepsy-related ligand/receptor complex LGI1 and ADAM22 regulate synaptic transmission. Science 2006; 313: 1792–5 11 Browne DL, Gancher ST, Nutt JG. Episodic ataxia/ myokymia syndrome is associated with point mutations in the human potassium channel gene, KCNA1. Nat Genet 1994; 8: 136–40 12 Becker EB, Zuliani L, Pettingill R, Lang B, Waters P, Dulneva A, Sobott F, Wardle M, Graus F, Bataller L, Robertson NP, Vincent A. Contactin-associated protein-2 antibodies in non-paraneoplastic cerebellar ataxia. J Neurol Neurosurg Psychiatry 2012; 83: 437–40 13 Irani SR, Michell AW, Lang B, Pettingill P, Waters P, Johnson MR, Schott JM, Armstrong RJ, S Zagami A, Bleasel A, Somerville ER, Smith SM, Vincent A. Faciobrachial dystonic seizures precede Lgi1 antibody limbic encephalitis. Ann Neurol 2011; 69: 892–900 14 Schmahmann JD, Sherman JC. The cerebellar cognitive affective syndrome. Brain 1998; 121: 561–79
Received 15 October 2013 Accepted after revision 23 February 2014 Published online Article Accepted on 10 March 2014
NAN 2014; 40: 645–649
doi: 10.1111/nan.12132
Scientific correspondence
LGI1 autoantibodies associated with cerebellar degeneration We describe a young adult with marked cerebellar dysfunction and encephalopathy, in whom serum leucinerich glioma inactivated 1 (LGI1) autoantibodies were identified. The patchy, severe Purkinje and granular neuronal loss provides a unique pathological correlate for the clinical presentation of LGI1 autoantibody-associated encephalitis, and for the anatomic distribution of the LGI1 antigen within the human cerebellum. Voltage-gated potassium channel (VGKC) complex autoantibodies are described in association with limbic encephalitis and Morvan’s syndrome [1]. The recent demonstration of distinct target antigens within the VGKC complex has further clarified the immunemediated pathogenesis of these clinical syndromes, with autoantibodies against LGI1 [2] and contactin-associated protein related 2 (CASPR2) proteins reported in patients with limbic encephalitis, and CASPR2 autoantibodies identified more frequently in patients with Morvan’s syndrome [3]. Increased availability and application of serological testing has further broadened the disease spectrum, with VGKC complex autoantibodies described in patients with sleep disorders, movement disorders, Creutzfeld-Jakob Disease mimics, and a variety of neurocognitive symptoms [3]. Cerebellar involvement is rarely reported in association with limbic encephalitis [3,4], and has not been described as a prominent feature in patients with LGI1 autoantibodies. The case presented here provides evidence that LGI1 autoantibodies may be seen in association with cerebellar dysfunction, attributable to cerebellar degeneration. The unique pattern of cerebellar Purkinje and granular neuronal loss, together with the marked response to treatment with immunotherapy, distinguishes LGI1 autoantibody-associated cerebellar degeneration from other immune-mediated ataxias. A previously well 18-year-old Caucasian man presented to the emergency department with a two-month history of occipital headache, progressive clumsiness and difficulty walking. There was no history of infectious prodrome, toxic exposure or constitutional symptoms. The © 2014 British Neuropathological Society
neurological examination was remarkable for gazeevoked nystagmus, dysarthria, bilateral appendicular ataxia and truncal ataxia. Magnetic resonance imaging (MRI) of the brain revealed multiple T2-hyperintense lesions isolated to the cerebellar hemispheres (Figure 1A), without radiological evidence of limbic (Figure 1B) or brainstem involvement (Figure 1C). The cerebellar folia enhanced following gadolinium administration with a leptomeningeal pattern (Figure 1D). Cerebrospinal fluid (CSF) analysis revealed cellular pleocytosis [69 white blood cells per high powered field (WBC/hpf), lymphocytic predominance], with mild elevation of protein (0.468 g/L; normal range 0.15–0.45). Oligoclonal bands were detected in CSF as compared with serum. Microbiological studies were negative, including bacterial and fungal cultures and polymerase chain reaction for herpes simplex virus, varizella zoster virus, Epstein-Barr virus and cytomegalovirus. Flow cytometry demonstrated T-cell predominance with polyclonal B-cells; no malignant cells were identified on CSF cytology. Human immunodeficiency virus serology was negative. At the time, the patient’s presentation was felt to be most consistent with a parainfectious autoimmune cerebellitis, demyelination or a lymphoproliferative disorder. Paraneoplastic etiologies were also included in the differential. Over the ensuing 3 days the patient’s cognitive status worsened: he became apathetic with disinhibited behaviours and visual hallucinations. Sinus tachycardia was noted. Pulse methylprednisolone was empirically administered given the deterioration (1 g intravenous daily for 5 days). Appendicular ataxia improved marginally with treatment, however his cognitive symptoms progressed. A repeat MRI brain showed worsening of the cerebellar lesions, with persistent sparing of other brain structures (Figure 1E). A repeat lumbar puncture showed an increase in the lymphocytic pleocytosis (271 WBC/hpf), and an absence of malignant cells. Serum samples were sent for measurement of autoantibodies [NMDA (GluN1), VGKC (LGI1, CASPR2), AMPA-R, amphiphysin, GAD-65, CV2/ CRMP5, Recoverin, SOX1, Titin, Hu, Yo, Ri and PNMA2 (Ma2, Ta); Mitogen Advanced Diagnostics, Calgary, Alberta]. Computerized tomography scans of the chest, abdomen and pelvis, and testicular ultrasound were 645
646
Scientific correspondence
Figure 1. MRI brain. (A) Axial Fluid Attenuated Inversion Recovery (FLAIR) sequence shows multiple hyperintensities in cerebellar hemispheres, more prominent on the right. (B) Axial FLAIR shows normal appearance of the hippocampi. (C) Sagittal FLAIR highlights the normal brainstem in contrast with cerebellar hyperintensities. (D) T1 with gadolinium sequence shows leptomeningeal enhancement along cerebellar folia. (E) Evolution of cerebellar lesions on FLAIR 1 week following presentation, again accompanied by normal appearances of the temporal lobes and brainstem. Arrow points to the area of the cerebellum sampled on biopsy. (F) Resolution of lesions on FLAIR 2 days following treatment, 10 days following E.
normal. Given the non-diagnostic investigations and rapidly worsening clinical condition, biopsy of the right cerebellum was undertaken to investigate for potentially treatable conditions. The site of biopsy is shown by the arrow in Figure 1E. Routinely stained sections showed scanty lymphocytic infiltrates in the meninges overlying cerebellar tissue (Figure 2A) and rare small clusters of reactive lymphocytes in the cerebellar molecular layer (Figure 2B), comprised of a mixture of CD20 immunopositive B-lymphocytes and more abundant CD3 immunopositive T-lymphocytes (Figure 2C). T-lymphocytes were scattered throughout the molecular layer (Figure 2C). The striking © 2014 British Neuropathological Society
abnormality in the sample was patchy cerebellar degeneration affecting both the crests and depths of the folia. The degeneration was characterized by widespread loss of Purkinje cell neurones with Bergmann gliosis (Figure 2A) and patchy, severe loss of the granular neurones, well demonstrated with NeuN immunolabelling (Figure 2D). No features of an infectious encephalitis (microglial nodules, viral inclusions), malignancy or demyelination (on LFB staining; not shown) were identified. SV40 immunolabeling for polyomaviruses was negative. The pathologic interpretation was of relatively widespread mixed lymphocytic infiltrates and Purkinje neuronal loss with patchy severe loss of granular neurones. NAN 2014; 40: 645–649
Scientific correspondence
647
Figure 2. Cerebellar biopsy. (A,B) H&E stained sections of cerebellum showing lymphocytes in the meninges and clustered in the molecular layer (yellow arrows). Cerebellar degeneration characterized by patchy Purkinje cell loss with Bergmann gliosis (blue arrow). Discrete areas of Purkinje (blue arrow) and granular neurone layers (red bar) depletion are seen, with intact white matter (black bar), and intact adjacent granular neurone layers (white bar). (C) CD3 immunolabeling highlighting T lymphocytes in the meninges and molecular layer where they are both clustered and more diffusely scattered. (D) NeuN immunolabelling demonstrates the severe focal granular neuronal loss.
Following the biopsy the patient continued to decline with depressed level of consciousness and persistent generalized myoclonus, with prominent involvement of orofacial musculature. EEG revealed a normal background rhythm with no epileptiform features. Five days after the biopsy, results from the serum autoantibody screen confirmed LGI1 autoantibodies, with medium titers. LGI1 autoantibodies were identified using a commercially available biochip immunoassay (Euroimmun GmbH; Lübeck, Germany). In the biochip assay, tissue sections of rat hippocampus and rat cerebellum were used as © 2014 British Neuropathological Society
a screen for autoantibodies against VGKC-associated proteins, including LGI1 and CASPR2. Monospecific antibody detection was confirmed through the parallel use of HEK293 cells transfected with complementary DNA encoding LGI1 channel constituents. Binding of patient’s autoantibodies was determined and semi-quantified by indirect immunofluorescence attended by appropriate positive and negative controls. Intravenous immunoglobulin (IVIg) was administered (2 g/kg over 2 days). Within 24 h of completing treatment the patient’s level of consciousness improved and his myoNAN 2014; 40: 645–649
648
Scientific correspondence
clonus resolved. Maintenance therapy was provided with prednisone (1 mg/kg/day). Repeat MRI brain 2 days after completing treatment showed dramatic improvement of the cerebellar lesions (Figure 1F). No suspicious metabolically active areas were seen with 18F-fluorodeoxyglucose positron emission tomography (PET) body scan. Electromyography and nerve conduction studies did not show neuromyotonia. Three weeks following diagnosis the patient was discharged with normal sensorium and near-complete resolution of ataxia, with plans to taper prednisone over 6 months. He remained normal at 9-month follow-up, with no clinical evidence of ataxia or cognitive dysfunction. This is a novel case of LGI1-associated encephalitis in a patient who presented with symptoms, signs and neuroimaging evidence of a cerebellar localization, and with neuropathological findings demonstrating selective Purkinje and granular neuronal loss. To our knowledge, this represents the first report of cerebellar degeneration associated with LGI1 autoantibodies. The neuronal loss and remarkable response to immunotherapy exemplifies the importance of early recognition and treatment of antibody-mediated brain disease. Additionally, the favourable outcome reported in this index case confirms that preferential cerebellar involvement may maintain a high potential for recovery with immunotherapy, despite the irreversible cerebellar neuronal loss. Similar findings are reported in descriptions of cerebellar infarction outcomes [5], which likely reflects plasticity within cerebellar circuits. A number of distinct features are useful in distinguishing LGI1 autoantibody-associated cerebellar degeneration from more common presentations of paraneoplastic cerebellar degeneration (PCD). First, no malignancy was found in our case, despite thorough investigations, including PET scan. Second, serial MRI demonstrated multiple evolving gadolinium-enhancing lesions, in contrast to descriptions of normal imaging in the majority of patients presenting with PCD [6]. Third, the dramatic response to immunotherapy in this case stands in contrast to the treatment-refractory progressive course typically described with PCD [7]. Finally, the pathology of this case further establishes LGI1 autoantibody-associated cerebellar degeneration as a unique entity. While granular neuronal loss is described in PCD, extensive Purkinje cell loss usually predominates [8] and inflammatory infiltrates are rare [9]. We suggest that LGI1 autoantibody testing should be considered in the differential diagnosis of © 2014 British Neuropathological Society
patients presenting with subacute ataxia, particularly young patients with encephalopathy. Staining of rat cerebellum against sera of patients with VGKC antibodies and LGI1 antibodies has previously revealed molecular layer [1] and cerebellar pinceau [4] binding, respectively. This contrasts in part with reported LGI1 receptor activity in the molecular layer and synaptic glomeruli of the granular layer in rat cerebellum [10]. The co-discovery of LGI1 autoantibodies and Purkinje and granular neuronal loss provides additional indirect evidence of LGI1 distribution within the human cerebellum. A limitation of our report is that patient serum was not shown to directly bind rat cerebellum. This limitation reflects the reality of practice within the clinical setting where elegant testing utilizing animal tissue to directly demonstrate antibody binding is not always available. Thus, while our findings are readily generalizable and our testing accessible (using commercially available autoantibody testing), further experiments are required to directly demonstrate LGI1-antigen distribution within granular layers of the cerebellum. Our patient initially presented with a cerebellar syndrome rather than symptoms of limbic encephalitis more commonly associated with LGI1 autoantibodies. Ataxia has been described in patients with mutations of the KCNA1 gene (encoding the Kv1.1 subunit of the VGKC) [11], and occasionally in patients with limbic encephalitis associated with VGKC antibodies [4], specifically CASPR2 antibodies [12]. Ataxia has not, however, been described as an early presentation of LGI1 autoantibodyassociated encephalitis. Isolated cerebellar dysfunction may represent a potential harbinger of limbic encephalitis,reminiscent of faciobrachial dystonic seizures [13] in its potential for providing a ‘window of opportunity’ for early diagnosis and initiation of immunotherapy in appropriate patients. On the other hand, there is evidence that the cerebellum is involved with more than motor control, with errors in cognitive processing and emotional control described following cerebellar injury [14]. It is possible that autoantibody-mediated cerebellar dysfunction may explain both the motor and the non-motor symptoms described in our patient, especially given the absence of temporal lobe hyperintensities typically (but not always) seen in limbic encephalitis (Figure 1B) [1]. In conclusion, we suggest that LGI1 autoantibodyassociated cerebellar degeneration may represent a distinct clinical entity that should be included in the differential NAN 2014; 40: 645–649
Scientific correspondence
diagnosis of subacute ataxia, recognizing the potential for neuronal loss and response to immunotherapy.
Acknowledgements We would like to thank the patient and his family for their willingness to participate in this scientific exchange.
Conflict of interest disclosures Claude Steriade has nothing to disclose. Gregory S Day serves as the Clinical Director for the Anti-NMDA-Receptor Encephalitis Foundation (Inc., Canada). The Foundation is supported by private donations. He is the recipient of a Future Leaders in Dementia award, including support for travel (Pfizer Canada). Marvin J Fritzler is the director of Mitogen Advanced Diagnostics Laboratory that performed the autoantibody testing and has received gifts in kind from Euroimmun GmbH, manufacturer of the anti-LGI1 biochip immunoassay. Brian J Murray has nothing to disclose. Julia Keith has nothing to disclose. C. Steriade† G. S. Day† L. Lee† B. J. Murray† M. J. Fritzler‡ J. Keith§ †Division of Neurology, Department of Medicine, Sunnybrook Health Sciences Centre, University of Toronto, §Department of Anatomical Pathology, Sunnybrook Health Sciences Center, University of Toronto, Toronto, Ontario, and ‡Departments of Medicine and Biochemistry and Molecular Biology, University of Calgary, Calgary, Alberta, Canada
References 1 Vincent A, Buckley C, Schott JM, Baker I, Dewar BK, Detert N, Clover L, Parkinson A, Bien CG, Omer S, Lang B, Rossor MN, Palace J. Potassium channel antibody-associated encephalopathy: a potentially immunotherapyresponsive form of limbic encephalitis. Brain 2004; 127: 701–12 2 Lai M, Huijbers MG, Lancaster E, Graus F, Bataller L, Balice-Gordon R, Cowell JK, Dalmau J. Investigation of LGI1 as the antigen in limbic encephalitis previously attributed to potassium channels: a case series. Lancet Neurol 2010; 9: 776–85
© 2014 British Neuropathological Society
649
3 Klein CJ, Lennon VA, Aston PA, McKeon A, O’Toole O, Quek A, Pittock SJ. Insights from LGI1 and CASPR2 potassium channel complex autoantibody subtyping. JAMA Neurol 2013; 70: 229–34 4 Irani SR, Alexander S, Waters P, Kleopa KA, Pettingill P, Zuliani L, Peles E, buckely C, Lang B, Vincent A. Antibodies to Kv1 potassium channel-complex proteins leucine-rich, glioma inactivated 1 protein and contactinassociated protein-2 in limbic encephalitis, Mrovan’s syndrome and acquired neuromyotonia. Brain 2010; 133: 2734–48 5 Kelly PJ, Stein J, Shafqat S, Eskey C, Doherty D, Chang Y, Kurina A, Furiet KL. Functional recovery after rehabilitation for cerebellar stroke. Stroke 2001; 32: 530–4 6 Dalmau J, Rosenfeld MR. Paraneoplastic syndromes of the CNS. Lancet Neurol 2008; 7: 327–40 7 Shams’ili S, Grefkens J, de Leeuw B, van den Bent M, Hooijkaas H, van der Holt B, Vecht C, Sillevis Smitt P. Paraneoplastic cerebellar degeneration associated with antineuronal antibodies: analysis of 50 cases. Brain 2003; 126: 1409–18 8 Verschuuren J, Chuang L, Rosenblum MK, Lieberman F, Pryor A, Posner JB, Dalmau J. Inflammatory infiltrates and complete absence of Purkinje cells in anti-Yoassociated paraneoplastic cerebelllar degeneration. Acta Neuropathol 1996; 91: 519–25 9 Scaravilli F, An SF, Groves M, Thom M. The neuropathology of paraneoplastic syndromes. Brain Pathol 1999; 9: 251–60 10 Fukata Y, Adesnik H, Iwanaga T, Bredt DS, Nicoll RA, Fukata M. Epilepsy-related ligand/receptor complex LGI1 and ADAM22 regulate synaptic transmission. Science 2006; 313: 1792–5 11 Browne DL, Gancher ST, Nutt JG. Episodic ataxia/ myokymia syndrome is associated with point mutations in the human potassium channel gene, KCNA1. Nat Genet 1994; 8: 136–40 12 Becker EB, Zuliani L, Pettingill R, Lang B, Waters P, Dulneva A, Sobott F, Wardle M, Graus F, Bataller L, Robertson NP, Vincent A. Contactin-associated protein-2 antibodies in non-paraneoplastic cerebellar ataxia. J Neurol Neurosurg Psychiatry 2012; 83: 437–40 13 Irani SR, Michell AW, Lang B, Pettingill P, Waters P, Johnson MR, Schott JM, Armstrong RJ, S Zagami A, Bleasel A, Somerville ER, Smith SM, Vincent A. Faciobrachial dystonic seizures precede Lgi1 antibody limbic encephalitis. Ann Neurol 2011; 69: 892–900 14 Schmahmann JD, Sherman JC. The cerebellar cognitive affective syndrome. Brain 1998; 121: 561–79
Received 15 October 2013 Accepted after revision 23 February 2014 Published online Article Accepted on 10 March 2014
NAN 2014; 40: 645–649
