- modified on Feb 11, 2012 to include new information on PANS
- modified on Nov 11, 2011 to include new information on Murphy's clinical studies
- modified on Aug 12, 2010 to include Hornig's work on passive transfer
I was sorting through my papers on PANDAS and thought I’d package the chronology here and the short summary of each paper in case others are interested in some of the early insights by Husby in 1976, the foundational work by Swedo in 1994-1998, then the recent work on passive transfer in 2009, and now the documentation of what "sudden onset" means by Tanya Murphy in 2011. If you click on the underlined words, you'll be taken to the relevant papers.
Dicovery of anti-neural antibodies with GABHS
In 1976, Husby found that antibodies to GABHS bonded with neuronal tissue in the caudate nucleus (basal ganglia).
- He noted that this binding was found for strep of emm-type 6, 11, and 12.
- He also noted that the reaction did not occur in rabbit brains but only human neural tissue.
- 46% of sera from 30 children with rheumatic chorea showed IgG antibody reacting with neuronal cytoplasm of human caudate and subthalamic nuclei.
- The antibody was also detected in 14% of 50 children with active rheumatic carditis. 203 controls showed no such antibody response.
Acute Rheumatic Fever, Sydenham Chorea and OCD
In 1989 Swedo published her study looking at 70 children with OCD over a ten year period where she noted the incredible similarity in symptoms.
By 1993, Bronze and Dale published their findings that neural tissue had cross reactivity with antibodies to the M protein from strep emm-type 6. This was essentially a rediscovery of Husby but with the further isolation that the antibodies were to the M protein.
In 1994, Swedo published a fascinating paper entitled “Speculations on antineuronal antibody-mediated neuropsychiatric disorders of childhood” where she proposed the hypothesis that neurological abnormalities of childhood may be caused by antineuronal antibodies resulting from a GABHS infection. This seemed to combine Husby’s, Bronze, Dale theories together.
IVIG and Plasmaphersis
In 1995, Swedo and Allen found 4 children who exhibited sudden onset OCD symptoms coincident with infections. Two of the children had exacerbations coincident with GABHS infections and two with viral infections. Treatment with plasmapherisis, IVIG and prednisone were all found effective. They called this treatable subset of OCD, PITAND (pediatric infection-triggered auto-immune neuropsychiatric disorder).
In 1997, Swedo found that the D8/17 marker from Khanna (1989) work on Acute Rheumatic Fever seemed to correlate and support the theory of a distinct genetic pre-disposition for OCD and chorea. She labeled this distinct OCD subgroup PANDAS – in case you wondered where the term came in.
In 1998, Swedo published the landmark paper entitled “Pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections: clinical description of the first 50 cases.” In which they separated and defined a clinical subgroup of OCD patients.
Challenges to the definition came out almost immediately, most notably from Singer and Kurlan who had been studying Tourettes Syndrome and did not think that the definition properly separated the group from children with TS. This has essentially been the argument for the last decade. Singer/Kurlan testing TS kids and Swedo testing the PANDAS subgroup of OCD kids.
In 1999, Perlmutter and Swedo conducted a blinded placebo controlled study that demonstrated that IVIG and Plasmapherisis reduced symptoms by some 45 and 50% respectively for the PANDAS subgroup. A later study by Nicolson showed no improvement for OCD and TS patients who did not fit the PANDAS subgroup (i.e., no post-streptococcal exacerbations).
In 2003, Kirvan and Swedo published the landmark Nature paper entitled “Mimicry and autoantibody-mediated neuronal cell signaling in Sydenham chorea” which pulled together all the above papers into a finding that there were three distinct antibodies that cross-reacted with Lysogangliosides in the brain. In addition, they found one of these antibodies caused significant CaM Kinase II activation in sera.
During this time, Dale , Church and others were making similar observations regarding anti-basal ganglia-antibodies (ABGA).
However, not all thought the research sufficient. Kurlan, Kaplan and Singer wrote many editorials questioning whether the subgroup was sufficiently distinct. They were consistently unable to repeat Swedo's experiments and questioned therefore whether the diagnostic criteria was strong enough and whether causality was actually shown. Unfortunately, most of their studies were on kids with chronic tics and controlled OCD rather than on children who matched the lightning like onset described by Swedo. This difference in interpreting and applying the PANDAS criteria may explain the divergent findings. Those studying onset seemed to find one finding. Those studying long term non-remitting tics found another. While this should help definition, the unfortunate editorials have instead caused more confusion than clarity.
Antibody isolation and Cam Kinase II
In 2004, Swedo responded to Kurlan and Kaplan’s comments explaining the different presentation of PANDAS from traditional OCD in that PANDAS presented with sudden onset and distinct episodes unlike the Tourettes presentation from Kurlan.
Curiously, despite this clarification, several longitudinal studies continued to use acute onset or episodic course and used a definition of episodic more consistent with wax/wane than the sawtooth like remission and recurrence described by Swedo.
In 2006, Kirvan and Cunningham published their finding that children from the Swedo studies were distinct from Tourettes and traditional OCD/ADHD patients in that the Sydenham Chorea and PANDAS children had elevated CaM Kinase II activation in their sera.
In 2007, Kirvan further showed that Tubulin is a target of the anti-neural antibodies in patients with sydenham chorea.
Creation of a mouse model of PANDAS (EAE) and passive transfer
In August of 2009 Yaddanapudi showed behavioral abnormalities in a set of mice after innoculation with GABHS. These mice were especially bred to have high T-cell rates and be prone to blood-brain barrier disruption. Yaddanapudi showed that IgG transfered from innoculated mice to non-innoculated mice transferred the behavioral abnormalities. This is known as passive transferance and a key finding for proving auto-antibody effects.Explanation of how the Blood-brain barrier is crossed
In November 2009, Bartholomäus et al unlocked a key part of explaining how the blood-brain-barrier can be breached. Using mice similar to Yaddanapudi (i.e., bred to have high T-cell rates and prone to blood-brain barrier disruption), they were able to watch individual T-cells cross the blood-brain-barrier. Once across, the T-cells produced inflammation recruiting other T-cells to the site of the breach. This could explain how antibodies in the blood stream cross the blood-brain barrier which has been the missing element since Husby's initial findings over three decades ago.
Th17, Autoimmunity and the blood brain barrier
Excaberations due to GABHS in acute and chronic cases
Recently (2010), Wang et al published a remarkable finding that repeated nasal innoculation with live (or dead) GABHS produced Th17 cell response in mice. What was interesting was that mice innoculated with GABHS produced Th1 response, whereas those with just nasal coloniztion produced Th17. Th17 was recently identified in 2006 (see Annunziato et al ) and is highly implicated in auto-immune disorders.
In 2007, Kebir et al published the paper "Human TH17 lymphocytes promote blood-brain barrier disruption and central nervous system inflammation" where it was shown that these Th17 cells are highly pro-inflammatory and promote blood-brain barrier disruption. Perhaps this is the missing link in how auto-antibodies are crossing the blood brain barrier (i.e., GABHS carriage triggers Th17, GABHS infection triggers anti-neuronal antibodies, Th17 disrupts the BBB allowing anti-neuronal antibodies (or B/T-cells) to cross).
While this is a mouse model and may not apply to humans, the study is noteworthy because only carriage was needed. This may explain why an apparent "allergic" reaction is seen in some children -- i.e., that only colonization is necessary on subsequent exposure and not infection.
- the PANDAS criteria was suffienctly narrow to exclude Tourettes patients who just happened to have a coincident strep infection.
- whether chronic conditions had the same association with GABHS as acute onset
Trying to reconcile the camps
In July 2010, a multi-disciplinary workshop on "PANDAS" was held at the National Institude of Health to bring the research perspectives together and compare findings. Clinicians reported a highly coherent set of symptoms that appeared at onset. These findings were compared with the symptoms in the two longitudinal studies on Tourettes children and raised the question of whether the PANDAS criteria was suffienctly narrow to separate patients from Tourette controls.
The result of this workshop was the recommendation to:
- separate the clinical presentation from the etiology/pathogensis (cause)
- lower the importance of tic symptoms in the defining criteria
- require acute onset of OCD symptoms in the primary citerion
The term "Pediatric Acute-Onset Neuropsychiatric Syndrome (PANS)" was chosen to be an umbrella term covering a coherent set of symptoms that could triggered by infection (PITAND and PANDAS) or non-infection triggers. Essentially PANDAS is within PITAND is within PANS.
This has the implication that PANDAS criteria would be adjusted to the PANS criteria plus an association with GABHS.
Continued clinical studies
Several followup studies have focused on the clinical presentation of PANDAS. Murphy in 2011 published a study of 109 subjects (68 controls). Looking only at the clinical presentation, Dr. Murphy concluded that those in the PANDAS group were
statistically more likely to:
- have had definite remissions in neuropsychiatric symptoms;
- have dramatic onset of symptoms;
- have definite remissions;
- show remissions of neuropsychiatric symptoms during antibiotic therapy;
- have elevated streptococcal
- have episodes of fever/sore throat at onset/flare up;
- show positive GAS culture results with symptom onset/flare up; and
- present with clumsiness
In addition, the NIH is repeating the 1998 Perlmutter study and has opened a clinical trial for new patients who exhibit dramatic and rapid onset of symptoms.
What I find rather fascinating about all the above is the nice trail of good science leading up to the discovery of the antibodies. The strongest counter evidence to PANDAS is the longitudinal studies that selected cases of chronic tic/Tourettes. One study found a correlation and one did not. While I'm personally frustrated that the reasonable debate in the research community causes confusion and denial of treatment in the clinical setting, it is impressive how quickly progress has been made.
Just think, how long from ulcers being thought purely the result of type A behavior/stress and then the finding of Helicobacter pylori? In 1586, Donatus decribes gastic ulcers. In 1913, Rosenow suggests that streptococci produce ulcers. In 1951, Allende notes that penicillin seems to help ulcers. In 1984, NY Times publishes summary of research. Altman states "I've never seen the medical community more defensive or more critical of a story." It's not until 2005 that Warren and Marshall are awarded the Nobel Prize. I appreciate these are different but it reminds me to have patience.
Quite a saga.
Edited by Buster, 12 February 2012 - 12:15 PM.