Background Autoantibodies to glial, myelin and neuronal antigens have been reported in a range of central demyelination syndromes and autoimmune encephalopathies in children, but there has not been a systematic evaluation across the range of central nervous system (CNS) autoantibodies in childhood-acquired demyelinating syndromes (ADS).
Methods Children under the age of 16 years with first-episode ADS were identified from a national prospective surveillance study; serum from 65 patients had been sent for a variety of diagnostic tests. Antibodies to astrocyte, myelin and neuronal antigens were tested or retested in all samples.
Results Fifteen patients (23%) were positive for at least one antibody (Ab): AQ4-Ab was detected in three; two presenting with neuromyelitis optica (NMO) and one with isolated optic neuritis (ON). Myelin oligodendrocyte glycoprotein (MOG)-Ab was detected in seven; two with acute disseminated encephalomyelitis (ADEM), two with ON, one with transverse myelitis (TM) and two with clinically isolated syndrome (CIS). N-Methyl-D-Aspartate receptor (NMDAR)-Ab was found in two; one presenting with ADEM and one with ON. Voltage-gated potassium channel (VGKC)-complex antibodies were positive in three; one presenting with ADEM, one with ON and one with CIS. GlyR-Ab was detected in one patient with TM. All patients were negative for the VGKC-complex-associated proteins LGI1, CASPR2 and contactin-2.
Conclusions A range of CNS-directed autoantibodies were found in association with childhood ADS. Although these antibodies are clinically relevant when associated with the specific neurological syndromes that have been described, further studies are required to evaluate their roles and clinical relevance in demyelinating diseases.
- Multiple Sclerosis
Statistics from Altmetric.com
The role of B cells and antibodies in central nervous system (CNS) demyelination is increasingly recognised.1 In multiple sclerosis (MS), where immunoglobulin and complement deposition in actively demyelinating lesions are reported in animal and human histopathological studies,1 immunoglobulins derived from a subset of MS patients appear to induce demyelination and axonal injury.2 Additionally, autoantibodies are consistently reported in a range of other demyelination syndromes, with antibodies against the astrocyte water channel protein, aquaporin-4 (AQP4), in neuromyelitis optica (NMO) or spectrum disorders (NMOSD); by contrast, antibodies to the myelin oligodendrocyte glycoprotein (MOG) are found in AQP4-negative NMO patients and also in a spectrum of acquired demyelinating syndromes (ADS).3 ,4
Brain-directed autoantibodies to ion channels, receptors and other synaptic proteins cause a number of distinct clinical-immunological syndromes, which are typically immunotherapy responsive.5 Antibodies to N-Methyl-D-Aspartate receptor (NMDAR) and the voltage-gated potassium channel (VGKC)-complex are the most commonly identified also in children with autoimmune encephalitis.6 As these antigenic targets may also be found in a variety of other functionally important sites, including the oligodendrocyte processes7 and the axoglial apparatus,8 these antibodies may also have a potential role in demyelinating syndromes as supported by recent case reports.9 ,10
Here, we have evaluated a paediatric cohort with a first episode of demyelination, identified from a prospective national UK surveillance study, for the presence of neuronal and glial autoantibodies. We reviewed the clinical and imaging phenotype of the positive patients and compared the antibody-positive with antibody-negative patients to determine whether antibody-positive patients represent a specific group of ADS.
Children under 16 years of age with a first episode of ADS evaluated with MRI (brain and/or spine) were ascertained from a national UK prospective surveillance study (Sep 2009–October 2011),11 where 1-year outcome was available. Demyelinating phenotypes were classified by the expert review panel based on International Paediatric Multiple Sclerosis Study Group (IPMSSG) criteria (described in ref 11); n=125: 40 acute disseminated encephalomyelitis (ADEM), 83 clinically isolated syndrome (CIS) and 2 NMO. Serum samples from 65/125 cases (52%; 14 ADEM, 49 CIS, 2 NMO) had been sent from 16 participating centres to the Clinical Neuroimmunology service at the Oxford Radcliffe Hospital Trust for testing for autoantibodies, as requested by the clinician for patient care, within 3 months of presentation. After initial testing, samples were stored at −20°C. Following identification from the surveillance study, samples were than tested/retested for the presence of antibodies to (1) AQP4; (2) MOG; (3) VGKC-complexes and the complexed proteins, leucine-rich glioma-inactivated 1 (LGI1), contactin-associated protein 2 (CASPR2) and contactin-2; (4)NMDAR and (5) glycine receptor (GlyR). The battery of antibody testing was performed systematically, non-selectively and at the time of testing, clinical information was not made available to the Oxford Laboratory. All patients with identified antibodies were managed according to the demyelination phenotype.
Antibodies to AQP4, MOG (cDNA courtesy of Dr K. O’Connor, Yale University, New Haven, Connecticut, USA),12 NMDAR, LGI1, CASPR2, Contactin-2 and GlyR were measured using cell-based assays in routine clinical use (sera at 1:20 dilution with the exception of CASPR2 and Contactin-2 where 1:100 dilutions was used) as previously described, where numerous healthy and disease controls were used to optimise the assays.12–16 For these cell-based assays, the binding of serum IgG to the surface of human embryonic kidney cells, transfected with cDNA encoding the autoantigens, was visualised using a fluorescence-labelled secondary antibody. All assays were assessed by two independent observers. VGKC-complex antibodies were measured by immunoprecipitation of rabbit brain VGKC-complexes labelled with radioactive dendrotoxin.
Non-parametric statistical tests (Kruskal–Wallis tests) were used for continuous distributions as appropriate given normality, and χ2 or Fisher exact tests were used for nominal data. Ethical approval for the surveillance study was from the UK multicentre Research Ethics Committee (09/H1202/92), and the ethical approval for further antibody testing on referred samples was from the Oxfordshire Regional Ethical Committee A (07/Q1604/28).
Sera from 65 patients were available for antibody testing. Fifteen patients (six males; median age 9 years, range 2.3–15) were positive (23%) for at least one antibody: AQ4 (n=3/64); MOG (n=7/59); NMDAR (n=2/57); VGKC-complex (n=3/57); CASPR2 (n=0/55); LGI1 (n=0/55); contactin-2 (n=0/55) and GlyR (n=1/55). Patients with ADEM presentation were less likely to have serum sent for antibody testing (14/40; 35%) compared with CIS (transverse myelitis (TM), optic neuritis (ON) and other, 49/83; 59%) (p=0.02; see table 1).
Antibodies to MOG were found in seven patients: two developed MS following CIS onset (one following a clinical relapse and one with new lesions on imaging), two with monophasic ADEM, two with ON and one with TM. AQP4 antibodies were detected in two patients with NMO and an additional patient with ON (no relapse at 1 year). NMDAR antibodies were found in one patient with monophasic ADEM and one additional patient (also positive for VGKC-complex antibodies) with ON. VGKC-complex antibodies were also found in a patient with monophasic ADEM and a patient with CIS later developing relapsing-remitting multiple sclerosis (RRMS). One patient with TM had GlyR antibodies. Further demographic, clinical and paraclinical features of antibody-positive patients are summarised in table 2, with the neuroimaging from nine patients, to illustrate the range of antibodies, are presented in figure 1A–I.
When comparing antibody-positive with antibody-negative patients, there were no significant differences in the patients’ demographics clinical and paraclinical features (see table 1). Repeat serum samples were only available in 5/65 patients; two with AQP4 positive NMO where antibody titres were used for disease monitoring and management decisions, and three patients with monophasic ADEM, with raised VGKC-complex antibodies (1630 pmol) in one and MOG antibodies in two; these antibodies were negative at repeat testing at 3 months after onset. In 8/15 positive patients, the antibody positivity was only identified retrospectively, at the time of this study (as the serum had been sent for other antibodies) thus likely to reflect the low number of repeats.
Despite the interest in the novel neuronal antibodies and their significance in specific forms of demyelinating disease or autoimmune encephalitis, and individual case reports on their presence in less typical syndromes, there have been no systematic studies. Within the limitation of the study, where not all samples from the national cohort were sent for antibody testing (higher rate of CIS compared with ADEM, see table 1) and that the follow-up period was for 1 year, we have identified glial antibodies in 14% of the patients. Interestingly, neuronal antibodies against NMDAR, VGKC-complex and GlyR were also detected, albeit at a lower frequency (8%). Nevertheless, since the antibody-positive group could not be distinguished from the negative group based on the demyelinating phenotype, and the frequency of any individual antibody was not high, the clinical relevance of these findings remains to be established.
Antibodies to MOG were seen in seven patients (12% of those tested). These antibodies have previously been detected in 20–40% of patients with ADS. They can be present in patients with ADEM, CIS and in persistent levels in paediatric MS.3 It remains unclear why such differences of positivity exist in the reported literature, although in cohorts where the MOG full-length protein is used, compared with the extracellular domain principally as here, a higher incidence of positivity appears to be reported, as reviewed elsewhere.4 However, the sensitivity and specificity of different MOG constructs in different assays in predicting a subtype-specific demyelination requires further evaluation by direct comparative studies of these assays in large well-characterised cohort/s, as previously done for AQP4.13 At this stage, based on these findings and the reported literature (reviewed in Refs. 3 and 4), antibody positivity cannot predict a subtype-specific demyelination, the risk of relapse, or the risk of developing MS. Three patients had antibodies to AQP4 (3/64; 4.5%), which was a higher incidence than detected in a large Canadian cohort of first episode of demyelination (2/279; 0.7%) where an indirect immunofluorescence assay was used to identify NMO IgG positivity.17 In two of the three patients with AQP4 antibodies, the clinical phenotype at presentation was NMO. Interestingly, the third patient, who presented with ON and poor visual recovery was initially positive for AQP4, did not receive any immunotherapy, and the antibodies became negative on repeat sample at 3 months. AQP4 antibodies have been detected in patients who do not fulfil the current clinical criteria for NMO, and it has recently been suggested that AQP4-positive adults presenting with a single or recurrent attacks of ON, myelitis, or brain/brainstem disease should also be included in the NMO spectrum and managed accordingly.18–21 This has not been evaluated in a paediatric cohort.
Five patients were positive for antibodies to NMDAR (n=2), VGKC-complex (n=3) and GlyR (n=1), antibodies that are reported with well-characterised distinct encephalopathy phenotypes usually including seizures, cognitive dysfunction and psychiatric symptoms as in limbic involvement, with additional movement disorders in NMDAR encephalitis, or stiff person syndrome/progressive encephalomyelitis with rigidity and myoclonus,5 features that were not seen in these ADS patients. However, association between these antibodies and demyelination has previously been reported; NMDAR antibodies in a 15-year-old girl presenting with encephalopathy and multiple relapses with TM and ON,9 antibodies to contactin-2 in 7% of patients with MS22 and, recently, antibodies to a different potassium channel, the ATP-sensitive inward rectifying KIR4.1 in 46.9% of patients with MS; in the latter case, the antibodies were shown to alter the expression of Glial fibrillary acidic protein (GFAP) in astrocytes.23 Taken together, these studies suggest that perturbation of function of these receptors and channels, which are found on myelinated processes of oligodendrocytes (NMDAR),7 the juxtaparanode of myelinated axons (VGKC-complexes)8 and at inhibitory synapses in the spinal cord (GlyR), can result in demyelination, or contribute to the accompanying disability, and since the antibodies are identified by binding to the antigen on live cells they have the potential to be pathogenic. Nevertheless, the low prevalence of these antibodies in the different demyelination phenotypes does question their clinical relevance. Although low-level VGKC antibodies have been reported in other diseases, such as neurodegenerative and genetic disorders,24 our group did not identify NMDAR or GlyR antibodies when evaluating 20 immune and 98 neurological controls.25
It remains unclear whether testing for a broader panel of neuronal surface autoantibodies should be done routinely, and it is possible that evaluating a larger cohort of patients may reveal novel and clinically relevant antibody-specific demyelination phenotype. Further studies are now required to evaluate whether CNS-directed autoantibodies are directly pathogenic and clinically relevant or whether they simply reflect a secondary response to demyelination, with some other pathogenesis, in the CNS. Importantly, identifying these processes precisely may influence diagnosis and management decisions in these patients.
We thank Drs B Lang and L Jacobson for confirming antibody results for VGKC-complex and NMDAR; and members of the UK & Ireland Childhood CNS Inflammatory Demyelination Working Group (Drs K Chong, K Forrest, K Foster, R Gunny, N Hussain, M Likeman, J Livingstone, S Mordekar, K Nischal, N Sibtain G Vassallo and S West) for their help in reporting the patients, data collection and MRI interpretation as part of the UK surveillance study.
Contributors YH: study concept and design, acquisition of data, analysis and interpretation of data, drafting and revising the manuscript. MA: study concept and design, acquisition of data, statistical analysis, analysis and interpretation of data, drafting and revising the manuscript. MW: acquisition of data, analysis and interpretation of data. CC: statistical analysis, analysis and interpretation of data, revising the manuscript. CGDG: acquisition of data, revising the manuscript. CH: acquisition of data, revising the manuscript. PEJ: acquisition of data, revising the manuscript. Rachel Kneen: acquisition of data, revising the manuscript. MGP: acquisition of data, revising the manuscript. WPW: acquisition of data, revising the manuscript. EW: acquisition of data, revising the manuscript. PW: study concept and design, drafting and revising the manuscript. AV: study concept and design, analysis and interpretation of data, drafting and revising the manuscript, study supervision. ML: study concept and design, analysis and interpretation of data, drafting and revising the manuscript, study supervision.
Funding This work was supported by the National Institute for Health Research (NIHR) Oxford Biomedical Research Centre based at Oxford University Hospitals NHS Trust and the University of Oxford (YH, PW and AV) and by the NHS specialised services for rare diseases (Neuromyelitis optica; PW and AV). The surveillance study was funded by the Multiple Sclerosis Society UK (grant number 893/08) and Action Medical Research (grant number SP4472).
Competing interests MA is funded by the MS Society and Action Medical Research charities. CC has been an investigator on research with unrestricted research funding from Eli Lilly, Wyeth and SHS. CH has received travel grants from Merck Serono and Bayer. MGP has received a meeting support grant from Euroimmun. EW has received travel grants from UCB, Shire and Biogen Idec, educational grants to organise meetings from Merck Sereno, Novartis, Bayer and Biogen Idec, speaker's fees from Merck Sereno and consultancy fees from Genzyme. AV serves/has served on scientific advisory boards for the Patrick Berthoud Trust, the Brian Research Trust and the Myasthenia Gravis Foundation of America; has received funding for travel and speaker honoraria from Baxter International Inc and Biogen Inc; serves as an Associate Editor for Brain; receives royalties from the publication of Clinical Neuroimmunology (Blackwell Publishing, 2005) and Inflammatory and Autoimmune Disorders of the Nervous System in Children (Mac Keith Press, 2010); receives/has received research support from the European Union, NIHR Biomedical Research Centre Oxford, Euroimmun AG and the Sir Halley Stewart Trust; and has received Musk antibody royalties and consulting fees from Athena Diagnostics Inc. AV and PW and the University of Oxford, hold patents and receive royalties and payments for antibody assays in neurologic diseases. ML receives research grants from Action Medical Research and MS Society; receives research support grants from the London Clinical Research Network and Evelina Appeal; received travel grants from Merck Serono; and awarded educational grants to organise meetings by Novartis, Biogen Idec, Merck Serono and Bayer. All other authors have no financial disclosures involving either stock ownership in medically related fields, consultancies, advisory boards, partnerships, honoraria, grants, intellectual property rights, expert testimony, employment, contracts or royalties from the preceding 24 months.
Ethics approval Ethical approval for the surveillance study was from the UK multicentre Research Ethics Committee (09/H1202/92) and the ethical approval for the antibody testing was from the Oxfordshire Regional Ethical Committee A (07/Q1604/28).
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.