Antibrain Antibodies in Children with Autism Spectrum Disorder and in Mothers Are Associated with More Severe Cognitive and Behavioural Profiles

Background: Previous work reported the detection of antibrain autoantibodies in the peripheral blood of children affected with ASD, specifically reacting with the cerebellum. Presence of these autoantibodies has been associated with lower adaptive and cognitive function, as well as with core symptoms of autism. Additionally, subsets of mothers of affected children produce IgG directed against fetal proteins, which during pregnancy most likely pass into the fetal circulation and reach the brain, since prenatally the blood-brain barrier is still permeable. These maternal antibodies have been associated with autism, lower expressive language scores and increased irritability using the Aberrant Behavioral Checklist.

Objectives: To assess: (1) the prevalence of 45 and 62 kDa autoantibodies among autistic children and unaffected siblings, (2) the prevalence of 37, 39, and 73 kDa antibrain antibodies in mothers of affected children, and (3) the clinical, cognitive and behavioural correlates of positive maternal or patient autoantibody status.

Methods: Antibrain autoantibodies were measured in plasma samples collected from 355 Italian autistic patients, 142 unaffected siblings and 333 mothers using Western Blot technology. Data were analyzed using the chi-squared test or the non-parametric Mann-Whitney U test. Nominal P<0.05 was considered significant.

Results: Antibrain autoantibodies were detected in 48/355 (13.5%) ASD and 26/142 (18.3%) unaffected siblings (P=0.200, n.s.). Among autistic children, the 45 kDa autoantibody is significantly associated with cognitive impairment and lower scores in all four Vineland Adaptive Behavior Scales (both P<0.05). The 62 kDa autoantibody is specifically associated with motor stereotipies and greater birth order (P<0.05). Maternal antibrain antibodies (37, 39 and 73 kDa) were detected in 137/355 (41.1%) mothers of ASD children. The presence of maternal antibodies, either alone or in combination, is correlated with impaired verbal and non-verbal language development, with sleep/wake cycle disturbances in their autistic children (P<0.05), and with neurodevelopment delay (P<0.01), as defined by principal component 3 in Sacco et al., Autism Res 3:237-252, 2010. Finally, we observed that antibrain autoantibody production is familial: presence of the 62 kDa Ab in the child is significantly associated with presence of the 39 and/or 73 kDa antibodies in their mother. Furthermore, these two maternal antibodies are in turn associated with more frequent autoimmune disease among first-degree relatives (P<0.05 and 0.01, respectively).

Conclusions: Our results confirm and extend previous observations in an ethnically distinct sample, providing further evidence of a strong pathomorphic contribution of antibrain antibodies to autism. The prevalence of antibrain autoantibodies does not differ between autistic and unaffected siblings, indicating that within the same sibship they do not confer increased risk of developing autism. However, they negatively influence autism severity, resulting in cognitive impairment and lower VABS scores in patients positive for antibrain antibodies; lower verbal and non-verbal language development, as well as wake/cycle disturbances in autistic children exposed to maternal antibrain antibodies. Antibrain antibodies are part of a broader familial liability to autoimmunity. Genetic factors conferring this vulnerability are currently under investigation.