• Is there something in my genetic or family history that poses a risk for ASD?
• What environmental exposures pose risks for the development of ASD?
• How might genetics and the environment interact to influence the occurrence of ASD?

What Do We Know?

As with many complex disorders, causation is generally thought to involve some forms of genetic risk interacting with some forms of non-genetic environmental exposure. The balance of genetic risk and environmental exposure likely varies across the spectrum of ASD. The greatly increased concordance of strictly defined autism in monozygotic (identical) twins (70 - 90%) compared to dizygotic (fraternal) twins (0 - 10%) argues for the importance of genetic factors (Bailey et al., 1995; Steffenburg et al., 1989). Moreover, there are subpopulations of those diagnosed with ASD who have a known genetic mutation, often associated with a genetic disorder, such as fragile X syndrome, Rett syndrome, or tuberous sclerosis complex, the understanding of which has led to identification of possible pharmaceutical interventions. In many cases, the same genetic variation does not result in an ASD phenotype, suggesting possible genetic or environmental modifiers that could be important intervention targets. Using new technology that reveals gaps and extra copies in DNA sequences, researchers have found that some people with ASD have deletions and duplications of genetic material not found in their parents' DNA (Sebat et al., 2007). Recent genetics research has identified common genetic variations (e.g., Wang et al., 2009; Weiss et al., 2009), changes in chromosomal structure in specific genomic regions, (Marshall et al., 2008; Kumar et al., 2008; Weiss et al., 2008) and rare mutations in genes all associated with synaptic connectivity (Alarçon et al., 2008; Bakkaloglu et al., 2008; Durand et al., 2007; Jamain et al., 2003; Laumonnier et al., 2004; Strauss et al., 2006). Some of these findings have contributed to new hypotheses about the inheritance of ASD. In families with just one affected member, spontaneous deletions and duplications may be causal factors of ASD. However, what causes these spontaneous deletions and duplications is not clear and could be due to environmental exposures.

Taken together, rare genetic mutations, chromosomal abnormalities, and sub-microscopic deletions and duplications of genetic material are involved in at least 10% of ASD cases, yet individually each abnormality is found in no more than about 1 to 2% of cases (Abrahams & Geschwind, 2008). Since common genetic variations confer only a modest increase in risk, this suggests that the genetic factors in ASD may involve many different genes and interactions between genes and environment. Possible models include: many additional rare genetic mutations to be discovered; multiple common genetic variations each conferring a small increased risk; and many forms of ASD with different genetic contributions, both common and rare, in the population. There is growing recognition that the same genetic contributions can lead to a wide variety of different phenotypes across individuals. As one good example, deletions and duplications in chromosomal region 16p11 have been associated with a broad range of phenotypes, including disorders outside the autism spectrum. The factors responsible for this variability in disease phenotypes remain to be defined.

Researchers are working to better understand the interaction of genetic vulnerability with developmental experiences, such as a specific environmental exposure. While gene-environment interactions have been hypothesized to play a role in many medical disorders, these interactions have been difficult to prove or disprove beyond statistical tests showing that some genetic subgroups have a greater response to some environmental factors. Epigenetics is one mechanism by which it is thought that environmental factors may be influencing gene expression, and now molecular tools are allowing researchers to gain insight into epigenetic phenomena that may be contributing to a variety of disorders, including ASD (Baccarelli & Bollati, 2009; Nagarajan et al., 2008).

While genetics maps the sequence of DNA, epigenetics maps the modifications of the structure of DNA due to proteins or other factors that bind to the DNA helix. DNA is essentially linear text that gets "read" into RNA that in turn codes for proteins. Epigenetic modifications do not change the text, but they highlight or redact large sections of text, changing how it is read. Epigenetic modifications consist of biochemical "tags" that attach to the DNA in different places, leading to the "silencing" or "activation" of genes. The pattern of epigenetic silencing or activation of genes can differ between genders, between species or between generations, and can change during specific time windows in development or in response to environmental cues. It is thought that the addition or removal of epigenetic tags from DNA is one mechanism by which developmental experience (e.g., exposure to physical or emotional stimuli) can cause long-term biological and behavioral effects. In 2009, the first maps of the human epigenome provided the first comprehensive look at where and how nature and nurture may interact (Lister et al., 2009).

Progress in identifying environmental factors that increase autism risk has been made recently (Eskenazi et al., 2007; Palmer et al., 2006; Palmer, Blanchard & Wood, 2009; Rauh et al., 2006; Roberts et al., 2007; Windham et al., 2006), although this area of research has received less scientific attention and far fewer research dollars than genetic risk factors. Environmental factors may be pertinent not only to brain development, but also to chronic systemic features of at least some subgroups of ASD. An Institute of Medicine (IOM) workshop held in 2007 summarized what is known and what is needed in this field (Forum on Neuroscience and Nervous System Disorders, Institute of Medicine, 2008). Numerous epidemiological studies have found no relationship between ASD and vaccines containing the mercury based preservative thimerosal (Immunization Safety Review Committee, 2004). These data, as well as subsequent research, indicate that the link between autism and vaccines is unsupported by the epidemiological research literature. However, the IOM report acknowledged that the existing population-based studies were limited in their ability to detect small susceptible subpopulations that could be more genetically vulnerable to environmental exposures.

Of note, the Committee receives many public comments that reflect concerns about vaccines as a potential environmental factor in autism. Some members of the public are convinced that the current data are sufficient to demonstrate that vaccines do not play a causal role in autism and argue against using limited autism research funds to do additional vaccine studies when many other scientific avenues remain to be explored. At the same time, those who believe that prior studies of the possible role of vaccines in ASD have been insufficient argue that investigation of a possible vaccine/ASD link should be a high priority for research (e.g., a large-scale study comparing vaccinated and unvaccinated groups). A third view urges shifting focus away from vaccines and onto much-needed attention toward the development of effective treatments, services, and supports for those with ASD.

In addition, a number of other environmental factors are being explored through research because they are known or suspected to influence early development of the brain and nervous system. Recent studies suggest that factors such as parental age and exposure to infections, toxins, and other biological agents may confer environmental risk. These findings require further investigation and testing, some of which is ongoing through the CDC's Centers for Autism and Developmental Disabilities Research and Epidemiology (CADDRE) and Study to Explore Early Development (SEED) programs, as well as through several NIH-funded studies, including the Norwegian Autism Birth Cohort (ABC) study, the Childhood Autism Risks from Genetics and the Environment (CHARGE) study, the Early Autism Risk Longitudinal Investigation (EARLI) study, and the Centers for Children's Environmental Health and Disease Prevention, which is supported collaboratively by the National Institute of Environmental Health Sciences (NIEHS) and the Environmental Protection Agency (EPA).

What Do We Need?

Although most scientists believe that risk factors for ASD are both genetic and environmental, there is considerable debate about whether potential environmental causes, genetic precursors, or interactions between genes and environmental factors should be the highest priority for research aimed at identifying the causes of ASD. To date, few studies have ruled in or ruled out specific environmental factors. There are reports of associations of ASD with exposure to medications, maternal antibodies, toxicants, and infections prenatally or postnatally; however, these observations need to be the subject of additional study. It is still not known whether any specific factor is necessary or sufficient to cause ASD. Similar to other disease areas, advancing research on the potential role of environmental factors requires resources and the attraction of scientific expertise. Bringing this to bear on autism will help define the environmental factors to study, as well as the best approach for staging studies to examine environmental factors, interaction between factors, and between individual susceptibility and various environmental factors.

For example, some researchers believe that it is important to study a large number of exposures, or classes of exposure, that are known to affect brain development. Others support more tightly focused studies of one exposure or a limited number of exposures, with greatest biologic plausibility for interacting with known or suspected biologic or genetic ASD risk factors. In addition, it is also important to design studies that assess environmental exposure during the most relevant exposure windows: pregnancy and early development. In doing this research, it will be important for the field to develop sound standards for identifying and claiming that environmental factors contribute to ASD, as it is for genetics.

Research studies on risk factors can be pursued through several means. Smaller, focused studies are needed for hypothesis testing and to provide insight for replication studies. Similar to other health outcomes research for relatively rare conditions, case-control studies can be an effective first line of inquiry. The NIH-supported CHARGE and CDC-supported CADDRE/SEED studies are good examples of this approach, in which environmental exposures and biological pathways, along with genetics, are being examined. Other existing cohorts could also be identified and used for epigenomic as well as traditional genomic and environmental studies.

To address public concerns regarding a possible vaccine/ASD link, it will be important for the IACC to continue to coordinate with the National Vaccine Advisory Committee (NVAC), a Federal advisory committee chartered to advise and make recommendations regarding the National Vaccine Program.

Epigenomics provides a ready mechanism for understanding how genes and environment may act jointly to affect autism risk. Studies are needed to investigate whether candidate environmental exposures alter epigenetic mechanisms that modify the expression of suspected autism susceptibility genes or genomic regions. Such studies should incorporate examination of time or stage of development as an important factor determining the impact of environmental agents on epigenetic programming. Finally, studies are needed to understand how changes in epigenetic tags in response to environmental stimuli could lead to specific phenotypic characteristics associated with autism.

Another approach for studying risk factors for ASD requires large sample sizes to disentangle the many possible genetic and environmental factors that contribute to and help explain ASD and the frequently co-occurring conditions. For other complex disorders, large DNA collections (i.e., >20,000 samples) have been necessary to detect the full genetic risk architecture. There are no genetic repositories of this size for ASD. Similarly, large birth cohort studies, in which biological samples have been collected throughout pregnancy and early postnatal life, may be essential for detecting the interplay of environmental exposures and genetic factors that lead to ASD. As a complement to these large-scale studies, research on critical subpopulations that may be at higher risk could provide leverage in identifying genetic and environmental risk factors.

2011 Addendum To Question 3: What Caused This To Happen And Can It Be Prevented?

What Is New in This Research Area, and What Have We Learned This Past Year?

A variety of discoveries have advanced knowledge of the biological underpinnings of autism. It was previously found that, among individuals with ASD, copy number variants, which are submicroscopic deletions and duplications in the genome occur more frequently in areas containing ASD risk genes (Cook & Scherer, 2008; Sebat et al., 2007; Weiss et al., 2008). A 2010 study showed that some of these copy number variants involved genes previously found to be associated with autism and some involved new rare mutations (Pinto et al., 2010). In addition, it was reported that neurodevelopmental disorders are more common in infants born prematurely and that preterm infants are at increased risk of developing autism (Johnson et al., 2010b). A study of blood mercury levels in 452 children in the NIH-funded Childhood Autism Risk from Genetics and the Environment (CHARGE) study showed that total mercury in blood was neither elevated nor reduced in preschool children with ASD (Hertz-Picciotto et al., 2010). In a separate study, no link was found between the exposure to thimerosal, a mercury-containing preservative used in vaccines, and increased risk for ASD (Price et al., 2010).

New data based on the Autism Speaks–supported Autism Treatment Network (ATN) patient registry and studies of high-risk infants indicate that autism is associated with high rates of several medical conditions, including gastrointestinal dysfunction, sleep disturbance, psychiatric conditions, and seizures (Presentation to IACC on the Autism Treatment Network, 2010). These co-occurring conditions are poorly studied, yet investigating them may reveal unexpected clues to environmental risk factors. For example, nonmotor features associated with Parkinson's disease (e.g., gastrointestinal problems, olfactory deficits, autonomic abnormalities) have yielded information about Parkinson's disease etiology and can serve as harbingers of the condition (Tolosa et al., 2009).

On September 8, 2010, NIEHS and Autism Speaks co-sponsored a meeting of scientists from both inside and outside the field of autism to identify novel opportunities and mechanisms to accelerate research on environmental factors and autism (Agenda from Autism and the Environment Workshop, 2010). Environmental factors considered included all factors affecting health that are external to the individual (such as physical, biological, chemical, dietary, social, and cultural), as well as the non-genetic characteristics of an individual (such as age, nutritional status, physical functioning, and medical history). As noted during the meeting, understanding environmental influences to autism will require both agnostic, discovery-based science as well as hypothesis-driven science in parallel. The set of recommendations developed from the meeting cite the need for strong interdisciplinary teams to move findings back and forth from clinical and epidemiologic settings to mechanistic studies. Research needs and opportunities identified included expansion of epidemiology investigations to capitalize on existing resources, development of a range of model systems that can address the complexity of autism, exploration of bioinformatics and screening approaches to identify environmental chemicals of interest, increased emphasis on neuropathology, enhancement of capacity for measurement of environmental compounds, harmonization of exposure assessment instruments, and mechanisms for expanding the workforce.

Technical advances in the past year increase traction for finding genetic and environmental risk factors. Novel bioinformatics platforms can be used to map genes to specific signaling pathways and to explore what environmental exposures are most likely to influence those pathways. Toxicogenomics data such as those produced by the EPA's National Center for Computational Toxicology could be mined to determine which environmental compounds act on the genes of interest. An important finding this year revealed the extent of "parent of origin" effects—i.e., for many genetic variations, risk depends on whether this variation was inherited from the maternal or paternal genome. And recent studies have revealed the importance of epigenetic mechanisms in disease etiology, bringing together genetic and environmental factors for the first time.

Information on the utility of induced pluripotent stem (iPS) cells and mesenchymal stem cells for exploring the biological bases of ASD is rapidly developing, pointing to the opportunity to use these tools as molecular assays for understanding genetic variation as well as for translational toxicology. Although research this year revealed several differences between these iPS cells, which can be easily and non-invasively derived from a person's skin cells, and embryonic stem cells, iPS cells continue to be one of the most promising new frontiers for understanding risk for ASD.

In a 2009 report by the National Vaccine Advisory Committee (NVAC), it was recommended that, in the context of immunization research, the ASD clinical subset of particular interest is regressive autism (National Vaccine Advisory Committee, 2009 (PDF – 630 KB)). Although the NVAC stressed that the temporal occurrence of this regression and the immunization schedule is not evidence of a causal relationship, regressive autism warrants further research in rigorously defined subsets of ASD. The NVAC noted that studies in this subpopulation might involve comparison of immune cytokine profiles between regressive and non-regressive ASD to screen for differential immune system profiles, or prospective immunization responsive profiling in siblings of children with regressive ASD. In addition, the NVAC recommended that studies assess whether adverse events following immunization (e.g., fever and seizures) correlate with risk of ASD, and that immune response profiles be examined in ASD cases with a history of adverse events following immunization.

The 2009 IACC ASD Research Portfolio Analysis indicated that about one-third of autism research studies funded by the Federal government and private organizations corresponded to risk factors/Strategic Plan Question 3, with the majority of this funding directed toward the identification of genetic risk factors and less funding and attention toward environmental research (IACC, 2010 (PDF – 1.7 MB)). This analysis suggests that environmental research is an understudied area that has been given insufficient attention and requires a heightened priority. Based on this, the Committee made several specific recommendations for research objectives and needed resources, which are reflected in the new objectives added to the Plan in 2011.

What Gap Areas Have Emerged Since Last Year?

Suitable model systems and those that offer better high-throughput capabilities for the study of environmental risk factors and their interaction with genetic susceptibility are needed. For example, models such as Drosophila melanogaster (fruit flies) and Danio rerio (zebrafish) have been extremely useful in identifying environmental contributors to other conditions, such as Parkinson's and Alzheimer's disease. The genetics and biology of synapse formation and function are increasingly well understood, underscoring the potential utility of vertebrate and invertebrate models for exploring how environmental exposure can affect brain function at the cellular and molecular levels.

Expansion and integration of epidemiological studies using different designs and types of data are needed. Combining data from multiple studies will be necessary to enhance statistical power, requiring standardization of protocols, instrument development, and data harmonization methods. This should also include standardized protocols on biological specimen collection, storage, and analysis. International studies offer unique opportunities to examine populations with different genetic and environmental exposure backgrounds. It would be helpful to create an autism "atlas" to examine differences in autism prevalence as a function of geography. Such analysis has proved useful in both cancer and asthma research.

There is a need for greater collaboration between genetic and environmental science investigators. Studies collecting genetic information should include data on environmental exposures and vice versa; large data sets are needed to allow mapping of detailed genetic, environment, and phenotypic information, including co-occurring medical conditions, inflammatory markers, pattern of onset, developmental course, and family history.

To accelerate our understanding of the role of epigenetics in autism etiology, further development and application of sensitive assays to measure DNA methylation, histone modification, and other epigenetic marks are needed. Studies are also needed to examine how exposures may act on maternal or paternal genomes via epigenetic mechanisms to influence risk for ASD.

The lack of adequate postmortem brain tissue continues to be a major barrier to progress in understanding the neurobiology of ASD, including the potential influence of environmental factors on the functional pathways involved in ASD.

Efforts to increase analytical capacity and core facilities are needed. For example, adding an environmental, immune, or animal models core to an already existing multidisciplinary team that studies autism would be beneficial. Access to these core facilities and services could encourage individual scientists to expand the scope of their studies to address environmental hypotheses.

Aspirational Goal: Causes Of ASD Will Be Discovered That Inform Prognosis And Treatments And Lead To Prevention/Preemption Of The Challenges And Disabilities Of ASD.

Research Opportunities

• Genetic and epigenetic variations in ASD and the symptom profiles associated with these variations.
• Environmental influences in ASD and the symptom profiles associated with these influences.
• Family studies of the broader autism phenotype that can inform and define the heritability of ASD.
• Studies in simplex families that inform and define de novo genetic differences and focus on what role the environment might play in inducing these differences.
• Standardized methods for collecting and storing biospecimen resources from well-characterized people with ASD as well as a comparison group for use in biologic, environmental, and genetic studies of ASD.
• Case-control studies of unique subpopulations of people with ASD that identify novel risk factors.
• Monitor the scientific literature regarding possible associations of vaccines and other environmental factors (e.g., ultrasound, pesticides, pollutants) with ASD to identify emerging opportunities for research and indicated studies.
• Better understanding environmental and biological risk factors during prenatal and early postnatal development in "at risk" samples.
• Cross-disciplinary collaborative efforts to identify and analyze biological mechanisms that underlie the interplay of genetic and environmental factors relevant to the risk and development of ASD, including co-occurring conditions.
• Convene ASD researchers on a regular basis to develop strategies and approaches for improving data standards and sharing, understanding gene-environment interactions, improving the speed of replication of findings, and enhancing the translation of research on potential causative factors to prevention and treatment studies.
• Measures of key exposures for use in population- and clinic-based studies and standards for sample collection, storage, and analysis of biological materials.
• Studies of behavioral, developmental, and medical variations across those with ASD who share common genetic factors.
• Studies of clinically meaningful subgroups to examine common genetic and environmental factors, as well as unique epigenomic signatures.

Short-Term Objectives

Note: Dates that appear next to the objectives indicate the year that the objective was added to the Strategic Plan. If the objective was revised in subsequent editions of the Plan, the revision date is also noted.

Dates 2009 Revised in 2010	A.	Coordinate and implement the inclusion of approximately 20,000 subjects for genome-wide association studies, as well as a sample of 1,200 for sequencing studies to examine more than 50 candidate genes by 2011. Studies should investigate factors contributing to phenotypic variation across individuals who share an identified genetic variant and stratify subjects according to behavioral, cognitive, and clinical features. IACC Recommended Budget: $43,700,000 over 4 years.
Dates 2009	B.	Within the highest-priority categories of exposures for ASD, identify and standardize at least three measures for identifying markers of environmental exposure in biospecimens by 2011. IACC Recommended Budget: $3,500,000 over 3 years.
Dates 2009	C.	Initiate efforts to expand existing large case-control and other studies to enhance capabilities for targeted gene-environment research by 2011. IACC Recommended Budget: $27,800,000 over 5 years.
Dates 2009	D.	Enhance existing case-control studies to enroll racially and ethnically diverse populations affected by ASD by 2011. IACC Recommended Budget: $3,300,000 over 5 years.
Dates 2010	E.	Support at least two studies to determine if there are subpopulations that are more susceptible to environmental exposures (e.g., immune challenges related to infections, vaccinations, or underlying autoimmune problems) by 2012. IACC Recommended Budget: $8,000,000 over 2 years.
Dates 2010	F.	Initiate studies on at least 10 environmental factors identified in the recommendations from the 2007 IOM report "Autism and the Environment: Challenges and Opportunities for Research" as potential causes of ASD by 2012. IACC Recommended Budget: $56,000,000 over 2 years.
Dates 2011	G.	Convene a workshop that explores the usefulness of bioinformatic approaches to identify environmental risks for ASD by 2011. IACC Recommended Budget: $35,000 over 1 year.
Dates 2011	H.	Support at least three studies of special populations or use existing databases to inform our understanding of environmental risk factors for ASD in pregnancy and the early postnatal period by 2012. Such studies could include: Comparisons of populations differing in geography, gender, ethnic background, exposure history (e.g., prematurity, maternal infection, nutritional deficiencies, toxins), and migration patterns; and Comparisons of phenotype (e.g., cytokine profiles), in children with and without a history of autistic regression, adverse events following immunization (such as fever and seizures), and mitochondrial impairment. These studies may also include comparisons of phenotype between children with regressive ASD and their siblings. Emphasis on environmental factors that influence prenatal and early postnatal development is particularly of high priority. Epidemiological studies should pay special attention to include racially and ethnically diverse populations. IACC Recommended Budget: $12,000,000 over 5 years.
Dates 2011	I.	Support at least two studies that examine potential differences in the microbiome of individuals with ASD versus comparison groups by 2012. IACC Recommended Budget: $1,000,000 over 2 years.
Dates 2011	J.	Support at least three studies that focus on the role of epigenetics in the etiology of ASD, including studies that include assays to measure DNA methylations and histone modifications and those exploring how exposures may act on maternal or paternal genomes via epigenetic mechanisms to alter gene expression, by 2012. IACC Recommended Budget: $20,000,000 over 5 years.
Dates 2011	K.	Support two studies and a workshop that facilitate the development of vertebrate and invertebrate model systems for the exploration of environmental risks and their interaction with gender and genetic susceptibilities for ASD by 2012. IACC Recommended Budget: $1,535,000 over 3 years.

Long-Term Objectives

Note: Dates that appear next to the objectives indicate the year that the objective was added to the Strategic Plan. If the objective was revised in subsequent editions of the Plan, the revision date is also noted.

Dates 2009	A.	Conduct a multi-site study of the subsequent pregnancies of 1,000 women with a child with ASD to assess the impact of environmental factors in a period most relevant to the progression of ASD by 2014. IACC Recommended Budget: $11,100,000 over 5 years.
Dates 2009	B.	Identify genetic risk factors in at least 50% of people with ASD by 2014. IACC Recommended Budget: $33,900,000 over 6 years.
Dates 2009	C.	Determine the effect of at least five environmental factors on the risk for subtypes of ASD in the prenatal and early postnatal period of development by 2015. IACC Recommended Budget: $25,100,000 over 7 years.
Dates 2009	D.	Support ancillary studies within one or more large-scale, population-based surveillance and epidemiological studies, including U.S. populations, to collect data on environmental factors during preconception, and during prenatal and early postnatal development, as well as genetic data, that could be pooled (as needed) to analyze targets for potential gene/environment interactions by 2015. IACC Recommended Budget: $44,400,000 over 5 years.