Neural autoantibodies in psychiatric disorders are associated with antibodies against viral pathogens: a retrospective study of 619 patients

Abstract

A history of viral infection has been associated with a higher risk for psychiatric disorders. One potential underlying mechanism is that antiviral immunological responses could trigger cross-reactivity between viral and neural antigens, which would raise the co-occurrence of antiviral antibodies and anti-neural autoantibodies. We studied 619 patients' psychiatric diagnoses from the Department of Psychiatry and Psychotherapy, University Medical Center Göttingen, Germany. Anti-neural autoantibodies and antiviral antibody specific indices were measured in serum and/or cerebrospinal fluid (CSF) from all patients. Among these 619 patients, 115 tested positive for serum and/or CSF neural autoantibodies (18.6%), with the most often identified autoantibodies being anti-GAD65 in serum (2.2%) and CSF (1.6%), and anti-NMDA in serum (0.6%) and CSF (1.3%). The three main diagnostic groups presenting neural autoantibodies were patients with organic psychiatric disorders including dementia (81 of 377; 21.7%), those with psychotic disorders (9 of 66; 13.6%), and patients with affective disorders (19 of 138; 13.9%). Logistic regression analysis revealed a significant association between the varicella zoster virus (VZV) antibody-specific index and autoantibody positivity in patients with all diagnoses (F00-F79) (p < 0.005). Furthermore, the rubella antibody-specific index proved to be significantly associated with neural autoantibody positivity (p < 0.001) across all patients (F00-F79), and in those with affective disorders (p < 0.01). Our results show that VZV and rubella antiviral antibodies are associated with a higher propensity to develop anti-neural autoantibodies, suggesting that the known association between viral infection and later developing psychiatric disorders may be partly attributable to the development of anti-neural autoimmunity.

Keywords: Autoantibodies; Autoimmunity; Psychiatric disorder; Viral antibody indices.

Fig. 1

Higher antibody indices against VZV and rubella in psychiatric patients. We detected increased antibody specific indices for VZV (A) (p < 0.005) and the rubella antibody specific index (p < 0.001) in neural autoantibody-positive versus negative groups in all patients in diagnostic groups F00–F79. In addition, the VZV antibody specific index was higher (p < 0.001) in autoantibody-positive psychiatric patients diagnosed F00–F09 than in autoantibody-negative patients (B). Furthermore, in (D) we detected a higher rubella antibody specific index (p < 0.01) in autoantibody-positive psychiatric patients diagnosed F30–F39 than in autoantibody-negative patients. Patients diagnosed F20–F29 with neural autoantibodies revealed no increased viral pathogen antibody specific index (C). HSV herpes simplex virus, VZV varicella zoster virus, EBV Ebstein Barr virus, NAB+ neural autoantibody positive, NAB– neural autoantibody negative

Fig. 2

Logistic regression analysis of patients with antibody indices against viral antigens. The VZV antibody index showed a significant regression predicting neural autoantibody-positivity with an AUC of 0.60 (p < 0.005) for all diagnostic groups (F00–F79) (A). In addition, the rubella antibody specific index also showed relevant regression predicting autoantibody positivity in patients from all diagnostic groups with an AUC of 0.62 (F00–F79) (p < 0.001) (B). Interestingly, the rubella antibody specific index also showed an even higher regression with an AUC of 0.73 to distinguish autoantibody-positive from -negative patients (p < 0.01) (H). ASI antibody specific index, AUC area under the curve, HSV herpes simplex virus, VZV varizella zoster virus