ArticlesAntibody titres at diagnosis and during follow-up of anti-NMDA receptor encephalitis: a retrospective study
Introduction
Anti-NMDA receptor encephalitis is an immune-mediated disorder associated with IgG antibodies to the GluN1 subunit of the NMDA receptor.1 The antibody reactivity depends on the conformation of GluN1 and it is detectable by use of immunohistochemistry of rat brain, cultures of dissociated rat hippocampal neurons,1 and HEK293 cells (cell-based assay)2 expressing GluN1 alone or in combination with GluN2 (GluN1/N2).1 These techniques differentiate these antibodies from those against linear GluN1 or GluN2 epitopes that have been described in various disorders and have little or no syndrome specificity.3, 4, 5 Patients with antibodies detectable by all three techniques—brain immunohistochemistry, neuronal cultures, and CBA with GluN1/N2—develop a highly predictable syndrome that we termed anti-NMDA receptor encephalitis.1 Subsequent studies with 1500 samples from consecutive patients with suspected autoimmune encephalitis, 100 of them with NMDA receptor antibodies,1 showed that all samples (either serum or CSF) that were positive by brain immunohistochemistry and CBA showed also reactivity with the cell surface of cultured neurons (unpublished). This cell surface reactivity, although useful for basic research studies,6 was similar to that of any synaptic antibody and therefore we excluded it for routine testing. By contrast, brain immunohistochemistry (which produces a highly characteristic pattern of reactivity7, 8 and might reveal additional antibodies) combined with CBA with fixed cells expressing GluN1/N2 (which confirms the identity of the antigen)1 are a set of techniques that, since 2007, we have routinely used in combination to test the presence of antibodies in serum and CSF samples of all patients referred to us.
Over the years we have studied patients who had a clinical picture characteristic of anti-NMDA receptor encephalitis but whose serum tested negative in other laboratories. When we examined samples from these patients (about 20 patients), we confirmed that the serum samples were negative but the CSF samples (not examined in other laboratories) were positive with both immunohistochemistry and CBA techniques. These results suggested that in some patients with anti-NMDA receptor encephalitis the antibodies were detectable only in the CSF,9, 10, 11, 12 and that examining both serum and CSF improved sensitivity.12 Despite these studies and the few cases comparing paired serum and CSF samples reported by other investigators (eg, 14 cases in one study,13 20 in another12), serum testing using only CBA has been suggested to be sufficient for the identification of NMDA receptor antibodies and the diagnosis of anti-NMDA receptor encephalitis.14 Serum testing with CBA has identified NMDA receptor antibodies in patients with several neuropsychiatric disorders (schizophrenia, Creutzfeldt-Jakob disease, and Parkinson's disease),15, 16, 17, 18 and in healthy individuals,18 but these findings have not been reproduced in studies using more comprehensive testing (CBA and brain immunohistochemistry) in serum and CSF samples.1, 19, 20 These discrepancies are not trivial: serum CBA testing might miss the diagnosis of a disease that is severe but potentially treatable, such as anti-NMDA receptor encephalitis.9, 10, 12, 21 Moreover, serum CBA testing without confirmation with a second technique or demonstration of antibodies in CSF might mislead the initial diagnosis of an unrelated disease, such as schizophrenia, Creutzfeldt-Jakob disease, or Parkinson's disease, by suggesting an immune-mediated disorder. Therefore, clarification of appropriate antibody testing is crucial, but no studies have directly compared the sensitivity and specificity of different NMDA receptor antibody techniques in paired serum and CSF samples.
After a patient is diagnosed with anti-NMDA receptor encephalitis, an important question in clinical practice is the utility of following up antibody titres during the disease. Some investigators assume there is a correlation between serum titres and the course of the disease that is as good as that of CSF titres and therefore serum titres can be used to make clinical decisions.13, 22 However, the data supporting this notion are not convincing, since they only come from case reports and small series, often affected by bias (cases with good outcome, in whom antibodies were assessed in serum only after plasma exchange or intravenous immune globulin). These treatments might transiently decrease any type of serum antibodies, but have little effect on CSF titres.23 Additionally, scarce information exists about serum titres in patients with poor outcome or CSF titres in any clinical scenario.24 Recent studies25, 26 suggest that, after clinical recovery, patients might still have detectable antibodies in serum and CSF. Therefore, the degree of correlation between serum and CSF antibody titres, relapses, and outcome is unclear.
To address these questions, we determined the sensitivity and specificity of NMDA receptor antibody testing in paired serum-CSF samples of patients with anti-NMDA receptor encephalitis and controls. We then focused on a subset of patients who had clinical relapses, good outcome, or poor outcome to investigate whether serum or CSF titres, or both, were associated with outcome. Finally, we investigated whether the pattern of epitope recognition could potentially help in the initial diagnosis or predict outcome.
Section snippets
Patients and samples
We randomly selected 250 patients (a sample size calculation is provided in the appendix), using an online random integer generator, from a cohort of 600 worldwide cases with anti-NMDA receptor encephalitis whose serum and CSF were studied in JD's laboratory at the Hospital of the University of Pennsylvania, Philadelphia, PA, USA (from February, 2006, until February, 2011) or at the Hospital Clínic, University of Barcelona, Spain (from March, 2011, until December, 2011). We defined anti-NMDA
Results
All 250 patients with anti-NMDA receptor encephalitis had antibodies in the CSF detectable with both rat brain immunohistochemistry and CBA using fixed cells expressing GluN1/N2 (sensitivity 100·0%, 95% CI 98·5–100·0). By contrast, only 214 (85·6%, 80·7–89·4, p<0·0001) serum samples were positive with both techniques; 18 (7%) were negative with one of the techniques, and the other 18 (7%) were negative with both techniques (table 1). Antibodies were more frequently detected in the serum of
Discussion
This study provides several novel findings that are relevant for the diagnosis and interpretation of antibody titres during the course of anti-NMDA receptor encephalitis: (1) by the time of diagnosis of the disease, NMDA receptor antibodies are always present in CSF but 13·2% (95% CI 9·6–18·0) of the patients do not have serum antibodies detectable with CBA; (2) high antibody titres are associated with poor outcome or the presence of a teratoma, or both; (3) change of titres in CSF correlates
References (40)
- et al.
Anti-NMDA-receptor encephalitis: case series and analysis of the effects of antibodies
Lancet Neurol
(2008) - et al.
A patient with epilepsia partialis continua with anti-glutamate receptor epsilon 2 antibodies
Pediatr Neurol
(2003) - et al.
Anti-NMDA receptor encephalitis presenting with imaging findings and clinical features mimicking Rasmussen syndrome
Seizure
(2011) - et al.
Anti-NMDA-receptor antibody encephalitis: performance evaluation and laboratory experience with the anti-NMDA-receptor IgG assay
Clin Chim Acta
(2013) - et al.
Autoantibodies associated with diseases of the CNS: new developments and future challenges
Lancet Neurol
(2011) - et al.
Outcome of children with opsoclonus-myoclonus regardless of etiology
Pediatr Neurol
(1995) - et al.
Treatment and prognostic factors for long-term outcome in patients with anti-NMDA receptor encephalitis: an observational cohort study
Lancet Neurol
(2013) - et al.
Anti-NMDA-receptor encephalitis: a cause of psychiatric, seizure, and movement disorders in young adults
Lancet Neurol
(2008) - et al.
Clinical experience and laboratory investigations in patients with anti-NMDAR encephalitis
Lancet Neurol
(2011) - et al.
Anti-NMDA Receptor encephalitis antibody binding is dependent on amino acid identity of a small region within the GluN1 amino terminal domain
J Neurosci
(2012)
Autoantibodies and cell-mediated autoimmunity to NMDA-type GluRepsilon2 in patients with Rasmussen's encephalitis and chronic progressive epilepsia partialis continua
Epilepsia
A subset of lupus anti-DNA antibodies cross-reacts with the NR2 glutamate receptor in systemic lupus erythematosus
Nat Med
Cellular and synaptic mechanisms of anti-NMDA receptor encephalitis
J Neurosci
Paraneoplastic anti-N-methyl-D-aspartate receptor encephalitis associated with ovarian teratoma
Ann Neurol
A patient with encephalitis associated with NMDA receptor antibodies
Nat Clin Pract Neurol
Retrospective analysis of NMDA receptor antibodies in encephalitis of unknown origin
Neurology
Anti-N-methyl-D-aspartate receptor (anti-NMDAR) encephalitis: an etiology worth considering in the differential diagnosis of delirium
Clin Toxicol (Phila)
N-methyl-D-aspartate antibody encephalitis: temporal progression of clinical and paraclinical observations in a predominantly non-paraneoplastic disorder of both sexes
Brain
Disease-relevant autoantibodies in first episode schizophrenia
J Neurol
NMDA receptor autoantibodies in sporadic Creutzfeldt-Jakob disease
J Neurol
Cited by (730)
Autoimmune encephalitis: Early and late findings on serial MR imaging and correlation to treatment timepoints
2024, European Journal of Radiology OpenAutoimmune central nervous system disorders: Antibody testing and its clinical utility
2024, Clinical BiochemistryParaneoplastic NMDA encephalitis, a case report and an extensive review of available literature
2024, Radiology Case ReportsBlood and CSF findings of cellular immunity in anti-NMDAR encephalitis
2024, International ImmunopharmacologyAutoimmune and paraneoplastic seizures
2024, Handbook of Clinical Neurology
- †
These two authors have contributed equally for this report