BioAutism 2012 is launched

The next BioAutsim meeting is planned for late January at The Queensland Brain Institute. See the webpage: www.qbi.uq.edu.au/BioAutism-2012

Effective therapies for autism

A report prepared by Vanderbilt Evidence-based Practice Center in the US presents the results of a meta-analysis of the scientific literature regarding effective therapies for ASD. This study takes the published results of 159 studies and identifies which medical and behavioural therapies appear to be effective at treating the symptoms and core challenges. Of the 159 studies, the report finds only 13 well designed studies, with 90 being poorly designed. In terms of medical treatments, respiridone and aripiprazole are somewhat effective in reducing challenging behaviours, despite noteable side-effects. Intensive behavioural intervention also shows promise, and further research in this area for the age group of <2 year-olds is required. This is the most comprehensive assessment of the current state of ASD therapies, based on evidence-based practice. The report consists of a preliminary summary of the study followed by a 100+ pages of study details.
This report, free online, is essential reading for those in the ASD community. Given the large amount of research currently undertaken, new and revised therapies are sure to be additionally relevant in the very near future. 
Download the report yourself, here.
Randal

Twin spin

Autism spectrum disorder (ASD), while once considered to be induced by ‘bad’ parenting, is now considered a prototype complex genetic disorder on the basis of twin, family, and other genetic studies. Consistently, the heritable liability in ASD is estimated as between 80-90% (reviewed by Ronald & Hoekstra, 2011). However, not all ASD cases are inherited, and de novo (non-inherited or ‘sporadic’) mutations contribute to 5-20% of ASD cases (Marshall et al., 2008; Sebat J et al., 2007; Zhao X et al., 2007; Pinto et al., 2010; Levy D et al., 2011).  Factors such as parental age, multiple births, and foetal infection may increase the risk of de novo mutations (see Blog entry: Vaccination preventing autism).

            A recent twin study, which received a large amount of coverage in the popular press, has calculated a lower heritability estimate for ASD and a greater non-inherited contribution (Hallmayer J et al., 2011).  Hallmayer and colleagues (2011) therefore suggested that the non-inherited risk factors (environmental ‘triggers’) such as parental age, multiple births, and/or foetal infections, contribute to ASD aetiology.  Other perinatal risk factors for ASD have been suggested by recent data (see ARA Blog entry from Saturday, July 16, 2011), while previous studies highlighted a role for parental age and multiple births as factors increasing the risk of sporadic (rather than familial) ASD (Croen LA et al., 2002; Gardener H et al., 2009).

            However, the concordance data in the Hallmayer and coworkers study (2011) does not differ markedly from that of other ASD studies and, as a result, the methodology employed has been criticized by experts in the field (see SFARI comments at: http://sfari.org/news-and-opinion/news/2011/experts-critique-statistics-conclusion-of-autism-twin-study).  There is also concern that exclusion from their dataset of patients with some types of ASD may likewise affect their findings.

            This paper has been reviewed in detail on other blog sites:

http://neuroskeptic.blogspot.com/2011/07/autism-isnt-very-geneticor-is-it.html

andhttp://autismjabberwocky.blogspot.com/2011/07/study-genetic-heritability-and-shared.html

            Therefore, while ASD is a prototypic complex genetic disorder, and the exact percentage contribution of non-heritable prenatal factors to selected ASD populations (compared to the contribution of heritable factors) varies, it also appears that these values can also vary depending on the statistical methods employed to analyze the data.

by Naomi Bishop

Key Reference:

Genetic Heritability and Shared Environmental Factors Among Twin Pairs With Autism.

Hallmayer J, Cleveland S, Torres A, Phillips J, Cohen B, Torigoe T, Miller J, Fedele A, Collins J, Smith K, Lotspeich L, Croen LA, Ozonoff S, Lajonchere C, Grether JK, Risch N.

Arch Gen Psychiatry 2011, Jul 4 [Epub ahead of print]

Vaccination preventing autism

The Wakefield paper (Wakefield A et al., 1998), suggesting a link between the measles, mumps, rubella (MMR) vaccine and autism, was subsequently shown to be fraudulent and to have “killed children" (McBrien J et al., 2000; European Centre for Disease Prevention and Control (ECDC) - Surveillance and Communication Unit report, 2008; Godlee F et al., 2011; Poland & Jacobson, 2011).  Furthermore, many high-quality independent studies have confirmed there is no link between autism spectrum disorder (ASD) the MMR vaccine (Taylor B et al., 1999, 2002; Dales L et al., 2001; Farrington CP et al., 2001; Fombonne & Chakrabarti S, 2001; Kaye JA et al., 2001; Demicheli V et al., 2005; Fombonne E et al., 2006; Cox & Kirkham, 2007; Hornig M et al., 2008; Gerber & Offit, 2009) or, indeed, any other vaccine (Price CS et al., 2010). The media has been criticized for its role in the MMR debacle (Roger D, 2003; Speers & Lewis, 2004; Hilton S et al., 2007; Katelaris A, 2007), but lay readers need to be aware that one ‘scientific’ journal is continuing to publish misleading studies on this topic (for example, see the discussion at http://photoninthedarkness.com/?p=222).

            Multiple twin, family, and genetic-association studies have demonstrated that ASD is a classic complex genetic disorder. Most studies calculate heritability estimates of between 0.8-0.9 (80-90%) for ASD (reviewed by Ronald & Hoekstra, 2011). By contrast, non-inherited (referred to as de novo) mutations are calculated as contributing to between 5-20% of ASD cases, and contribute more to severe ASD than to milder ASD (Marshall et al., 2008; Sebat J et al., 2007; Zhao X et al., 2007; Pinto et al., 2010; Levy D et al., 2011).  ASD risk factors may include increased parental age (see ARA Blog entry from Saturday, July 9, 2011), which is known to lead to increases in chromosomal abnormalities.  Likewise, in utero infection has been suggested to increase the risk of ASD.  The evidence for increased risk of ASD associated with foetal rubella syndrome (FRS) is the strongest (Chess S, 1971, 1977; Chess S et al., 1978). This evidence is limited to the output of a single research lab, but has been supported by animal studies and case studies.  The link between other viral infections, such as CMV, and ASD remains weak.

            On the basis of the increased link between FRS and ASD, a recent study has calculated how many ASD cases have been prevented by the use of rubella vaccination in the USA (Berger BE et al., 2011). Rubella virus infection is known to cause chromosomal anomalies (Plotkin SA et al., 1965; Nusbacher J et al., 1967; Konishi S et al., 1970; Ansari & Mason, 1977) and, while the vast majority children with FRS will not have ASD, there may be an increased risk of disruption of ASD-implicated genes in early infection in utero.  Other mechanisms for increasing the risk of ASD are also possible. This recent study, therefore, indicates that rubella vaccination has prevented thousands of cases of ASD in the USA alone (Berger BE et al., 2011).

by Naomi Bishop

Key reference: Berger BE, Navar-Boggan AM, Omer SB. BMC Public Health 11: 340 (2011)

ASD in Down syndrome

Criteria of ASD are often applied to individuals with Down syndrome even though the validity of the criteria in this population is unclear. About 5% of individuals with Down syndrome exhibit repetitive stereotypic movements, but do they also have social communication deficits? There is also a greater incidence of intellectual disability in Down syndrome, that may cloud or confuse diagnosis of social communication skills. A recent report from the Kennedy Kreiger details the co-mobities. The importance of making these distinctions means that more appropriate interventions can be trialled. The other interesting observation comes from a consideration of the genetics. In Down syndrome, over 350 genes are present in excess on chromosome 21, compared to one gene in deficit in ASD, for example. This results in a reduced brain size in Down syndrome, but potentially increased brain size in ASD. What is particularly interesting is that 40% of individuals with Down syndrome have ASD (either PDD or autism). Therefore, despite (nearly all) individuals with Down syndrome having a defined and consistent genetic imbalance, not all of them develop ASD, motor stereotypies or display disruptive behaviour. The question now for researchers is Why? The challenge raised by the authors of the above study for educators is to develop a range of targetted interventions. 

ERKsome findings

ERKs are enzymes in cells that perform numerous tasks. In brain cells, ERKs help the cell adapt to changing stimulus and help create new cells. Researchers have found that by reducing the level of ERK2 from midgestation in mice, these mice go on to show social differences, likened to ASD. However, we also  have reason to believe that a child with no ERK2 would not survive. Based on these pieces of information, we would expect that a child experiencing prolonged low ERK2 would likely develop ASD or another psychiatric disorder. Curiously, the opposite appears to be happening in ASD. ERKs are actually increased in the (postmortem) brains of some 8 year-olds with ASD. Particularly ERK5, which is responsible for brain cell death. However, we know how unique each child is with ASD. Whether these changes in ERK are consistent for ASD, selective for this group, or even a consequence of earlier developmental problems will not be known for some time. Animal studies still have some way to go in this regard also. What is clear is the delicate balance of ERK levels for learning and development. And because of this, perhaps now, ERK drug targets will migrate from cancer research towards ASD research.
Randal

ABS releases 2009 data on ASD

The report presents an overview of autism in Australia, including information on prevalence, education, disability, and need for assistance.

Read the free report here

Perinatal and neonatal risk factors

While ASD is not diagnosed until well after birth, there has been accumulating evidence that prenatal and neonatal factors may cause ASD. While genetic mutations remain a large risk factor it is important to keep in mind environmental risk factors. A recent meta-analysis (an analysis of published studies) identifies about 60 perinatal and neonatal risk factors. Factors associated with an increased risk of ASD were abnormal presentation, umbilical-cord complications, fetal distress, birth injury or trauma, multiple birth, maternal hemorrhage, summer birth, low birth weight, small for gestational age, congenital malformation, feeding difficulties, neonatal anemia, Rh incompatibility, and hyperbilirubinemia (jaundice). And those not associated with an increased risk were assisted vaginal delivery, postterm birth, high birth weight, and head circumference. When reading these results it is important to bear in mind a number of things: (1) most of the studies investigated by the group showed inconsistent results with only 30% of studies having sufficient power to allow risk factor correlation, (2) inconsistent results were often due to different study designs, (3) there was no evidence to suggest that only one factor was implicated in ASD, but instead, increased risk resulted from multiple factors, (4) a lot of the factors are related; for example cesarian delivery is often undertaken in cases of foetal distress, hyperbilirubinemia is associated with hypoxia, and feeding difficulties can be associated with any number of preceding factors, (5) these factors are not unique to ASD, but can result in other pervasive developmental disorders, children with lower IQ without ASD, etc, (6) similarly, it is not clear how many of the ASD cases reported also had intellectual disabilities, and importantly, co-morbidities such as Down syndrome, (7) the significant risk factors may be a result of an underlying cause of ASD, which leads to the shared-risk hypothesis, (8) ASD subtypes and symptoms were not distinguishable from the published studies and finally, (9) no maternal factors were considered such as smoking or medication, only foetal or neonatal events. Interestingly, head size was not a risk factor of ASD. This would appear to contradict reports that a subtype of ASD has an enlarged brain, but scientists currently believe that a large head size develops after birth, and hence would not necessarily be picked up at birth, and therefore not included in this study.
Read the full report here.
Randal

Advanced Paternal Age

There is a known increase in risk of chromosomal malformations, including Down syndrome, with advanced maternal age. But research in recent years suggests that the increase in risk of ASD is associated with advanced paternal age. The hypothesis is that there are more opportunities for copy number mutations to occur in the sperm of older fathers. Current research aims at modelling this in rodents. Australian researchers describe this attempt here.

Randal

Jaundice as a cause of ASD?

There has been recent interest in the possibility that jaundice causes autism. This interest has been stimulated by a publication identifying an association with autism and jaundice. Jaundice is the accumulation of blood cell break-down products (billirubin) in the body. In newborns, this is thought to be due to metabolic adjustments. In rare cases, neonatal jaundice can cause brain damage. The first striking aspect of this study is that despite jaundice prevalence of 60% in newborns, the autism prevalence in this study was about 5%. Also a disease called hemolytic disease causes an increase in billirubin but the incidence of ASD in this group was not investigated. Based on the jaundice hypothesis of ASD, it would be expected that there is a very high incidence of ASD in those with hemolytic disease. A further rebuttal of the jaundice hypothesis can be found here. If ASD is a multi-hit disorder - requiring at least two agents for one cause - then jaundice might be a candidate for one of the co-agents.
Randal

BioAutism 2011 Supplementary data and info

Final Program

Presentation by Katrina Williams

Clinical practice by Catherine Marraffa




Prospective cohorts - Andrew Whitehouse




The organisers:
Randal Moldrich, Elisa Hill, Naomi Bishop, Dennis Crowley