OVERVIEW

Investigators from the University of Oklahoma Health Sciences Center utilized samples obtained by colleagues at Yale/MGH and the National Institute of Mental Health to investigate the role of autoantibodies in Sydenham chorea (SC) and Pediatric Autoimmune Neuropsychiatric Disorder Associated with Streptococcal infection (PANDAS.)

SUBJECTS

Subjects included 8 children with Sydenham chorea (SC) and two cohorts of patients with PANDAS (N= 25 and N = 35; the latter group were participants in the Yale-NIMH IVIG trial), as well as three groups of psychiatric controls – 18 children with Tourette disorder, 25 with (non-PANDAS) obsessive compulsive disorder (OCD) and 18 with attention deficit hyperactivity disorder (ADHD) and 28 healthy controls (free of psychiatric and neurologic illness).  Serum samples were obtained from the SC and PANDAS groups during acute illness and convalescence and from the psychiatric contrast groups during symptomatic periods.  Additionally, cerebrospinal fluid (CSF) samples were obtained during symptomatic illness from the 35 children in the Yale-NIMH IVIG trial cohort.

METHODS

Direct enzyme-linked immunosorbent assay (ELISA) was used to measure IgG autoantibodies against four human neuronal antigens (tubulin, lysoganglioside Gm1, and dopamine receptors D1 and D2).   To assess bioactivity of the sera samples, calcium/calmodulin-dependent protein kinase II (CaMKII) activation was assayed using techniques previously employed.

RESULTS 

Figure 1 shows the results for the first PANDAS cohort, the SC patients and psychiatric contrast groups – as shown, PANDAS and SC patients had similarly high levels of anti-D1R, anti-D2R, anti-lysoganglioside Gm1 and anti-tubulin autoantibodies, as well as striking activations of CaM Kinase II.  Autoantibody titers and CaMKII activity in the SC and PANDAS groups was significantly higher than that of the psychiatric contrast groups or healthy controls.  Serum and CSF samples from the more recent PANDAS cohort (n=35) were used to replicate and extend these findings.  Autoantibody titers were elevated and CaM KII showed signs of activation in PANDAS sera obtained during acute illness; these levels had normalized in the convalescent sera.  Importantly, CSF autoantibodies were found in 32 of 35 PANDAS CSF samples (91.4%) and included anti-D1R (10 of 34 samples), anti-D2R (22/34), anti-lysoganglioside Gm1 (2/27) and anti-tubulin (4/32), as well as CamKII activation in 24 of 34 samples tested.  Comparisons of serum and CSF samples showed agreement for 33 of 35 subjects, with 31 subjects displaying one or more positive results in both serum and CSF.  At 6 months follow-up, symptom severity was decreased for all 35 PANDAS patients with 80% reported to be “much improved” or “very much improved”.  Although anti-tubulin titers were not significantly decreased, all other titers showed significant decreases (anti-D1R, p = 0.002; anti-D2R, p = 0.039 and anti-lysoganglioside Gm1, p = 0.0005) and CaMKII activation had normalized (p = 0.0001 for change from baseline.)

DISCUSSION

The discussion section interprets the study’s findings in light of background knowledge and historical information, as well as recent clinical and laboratory findings.  This study is additional support for post-infectious autoimmune basis for PANDAS (and SC); suggesting that PANDAS and SC share disease mechanisms.  This is an exciting development when considered in light of recent evidence that recurrent Group A streptococcal (in mice) produce GAS-specific T cells that provoke neuroinflammatory changes and Dr. Dritan Agalliu’s study showing disruption of blood-brain barrier (BBB).  With a leaky BBB, these circulating autoantibodies can get into the CNS and produce damage – positive CSF findings support that hypothesis.