Background: Autoimmune encephalitis has become increasingly important in neuropediatrics since a treatable autoimmune disease has been identified in a growing number of children with encephalopathic, epileptic, and psychiatric symptoms. Despite increasing numbers of cases and the growing list of identified autoantibodies, the underlying pathomechanisms remain unclear.
A functional link between infection and autoimmune encephalitis has been demonstrated for herpes simplex infections and anti- N-methyl-D-aspartate (NMDA) receptor encephalitis, and molecular mimicry has been used as an explanation.
At the same time, there are reports of different viral infections triggering anti-NMDA receptor encephalitis. These indicate unspecific immunological reactions.
Finally, recent data show that in patients with anti-NMDA receptor encephalitis, various neuronal autoantibodies exist at the same time, of which only a fraction is actually directed against NMDA receptors. This raises questions about the relevance of neuronal autoantibodies.
Aims: Currently, the search for neuronal autoantibodies is usually restricted to encephalitis patients in clinical practice. Data about the frequency in other disease groups are limited. The aim of our work is to map the prevalence of neuronal autoantibodies in children with various neurological disorders get a deeper understanding of their relevance.
Problem: Do neural autoantibodies exist in neuropediatric diseases apart from encephalitis cases? Are there indications for other associations similar to those between HSV infection and NMDA receptor autoimmunity? Are these antibodies primarily nonspecific epiphenomena or crucial for pathogenesis?
Methods: Patient cohort: Currently, there are 152 patients (95 females) and the average age is 8.5 years (2 months–17 years). Diseases include headache, facial nerve palsy, meningitis, epilepsy, multiple sclerosis, movement disorders, syndromal disorders, psychosis, and psychosomatic illnesses. Immunohistochemical screening of cerebrospinal fluid (CSF) samples using tissue-based assay: incubation on native mouse brain tissue. Indirect immunofluorescence staining using antihuman immunoglobulin G. Identification of a surface signal based on the staining pattern. Classification according to signal intensity and specificity in groups 0 to 3. Secondary analysis to identify known autoantibodies: positive testing of positive samples by means of a commercial assay (cell-based assay).
Results: In the current cohort, 11% (17/152) of CSF samples showed a strong surface signal (groups 2–3), another 15% (23/152) a weak positive to unclear signal. Seventy-four per cent (112/152) were negative or showed an intracellular signal. Disease entities showed a clear distribution of the 26% positive findings within infectious-inflammatory diseases (17/152), psychiatric-psychosomatic entities (6/152), and other diagnoses (7/152).
Conclusion: Comparative data, especially from CSF samples, are currently missing. Previous study usually analyzed serum from adult psychiatric and epilepsy cohorts. Here, the prevalence varies (7–23%). However, serum autoantibody frequency is already high in healthy controls. CSF results therefore may have a higher significance for a possible disease relevance. This initial mapping across various neuropediatric conditions shows that neuronal autoantibodies exist apart from encephalitis. A pathogenetic relevance does not always exist. The accumulation among infectious-inflammatory diseases though strengthens the hypothesis of a connection between infection triggers and the formation of neuronal autoantibodies. In a further step, the identification of unknown autoantibodies (positive in screening vs. negative in commercial assays) will be our focus.
In addition, the immunohistochemical screening used here already provides a valuable supplement to clinical routine diagnostics and was already crucial for therapeutic attempts in individual cases beyond the context of this project.
