Evidence of a biological etiology for autism spectrum disorder and the effects on assessment and treatment

Contents

Abstract

1. Introduction

2. Literature review
2.1 Psychopathology
2.2 Physiological Psychology
2.3 Contemporary Issues in Psychology
2.4 Crisis and Emergency Management
2.5. Prefrontal cortex

3. Conclusion and Future Research

4. References

Abstract

The diagnosis rate of autism spectrum disorder (ASD) is increasing significantly, but assessment and treatment development is limited by an incomplete understanding of the disorder. This paper presents evidence to support at least in part a biological etiology for ASD supported by twin studies, and specifically implicating genetic mutations observed to increase with parental age. A complex model of disease is presented in which genetic mutations are suggested to impair critical neurodevelopmental processes in early childhood, resulting in the manifestation of both physiological and psychological symptoms. These results are supported by imaging studies indicating the presence of an abnormal pattern of brain growth in individuals with ASD. Current treatment options are presented and evaluated in order to determine the implications of a biological etiology as well as whether pharmacotherapy treatment is medically and psychologically justifiable for individuals with ASD. An animal model of ASD is presented using rats and mice exposed to valproic acid (VPA) in utero. Research is presented in which a VPA model is used in the assessment and successful treatment of several core ASD symptoms in an animal model. A biological etiology of ASD is concluded to be likely, and future research considerations are discussed in light of the research presented.

1. Introduction

Autism spectrum disorder (ASD) is a neurodevelopmental disorder affecting children and adults throughout the lifespan. ASD is characterized by the emergence of symptoms in early childhood including but not limited to deficits in communication and social skills, repetitive behaviors, developmental regression, hyperactivity, and cognitive deficits. ASD is measured on a spectrum in which diagnoses may be mild- resulting in an individual considered to be high-functioning, or severe- resulting in an individual considered to be low-functioning. Individuals with severe autism may be completely nonverbal, and often require full-time care throughout their lifetimes.

The diagnosis rate of ASD has been increasing in recent years at a significant enough rate that many are calling it an epidemic. Developing a treatment for ASD has become one of the top priorities for researchers in the field. However, the exact etiology of ASD remains a topic for debate among those in the mental health professions, making the pursuit of a treatment all the more difficult. Currently the outward symptoms of ASD are being treated with medications intended for other disorders, as not enough is understood about ASD to take a more targeted approach. It seems fitting then, that the symbol adopted by the ASD community is a puzzle piece, as the complexity of the disorder renders it an extraordinarily complicated puzzle not easily solved by our current breadth of knowledge.

In order to begin pursuing more targeted approach to the treatment of ASD it becomes necessary to break the problem into smaller pieces. The first of these is to gain an understanding of the disorder, and more specifically its etiology. This information will direct the future of professional efforts from research to assessment and treatment.

This paper will explore (i) the evidence for a biological etiology, (ii) the interaction of physiological and psychological symptoms stemming from the disorder, (iii) the implications of a biological etiology for treatment, and (iv) the implications of a biological etiology for assessment and treatment development. The overarching research question this paper will seek to answer is as follows: Is there enough evidence to support a biological etiology for ASD and what implications does this information have on assessment and treatment?

2. Literature review

2.1 Psychopathology

Twin studies. In a 2014 study, Danish researchers sought to explore the possibility of a genetic etiology of ASD. Participants in this study were recruited utilizing a national registry known as the Danish Twin Registry. Participants were same-sex twin pairs born in Denmark between the years of 1988 and 2000. All of the twin pairs were between the ages of 6-17. This study excluded opposite-sex twin pairs in order to eliminate gender-based variations in prevalence, and identified from the results probands with ASD (Nordenbæk, Jørgensen, Kyvik, & Bilenberg, 2014).

This study was conducted in three phases. The first phase involved the screening of all twins born between the target years in Denmark. Their parents were asked to complete a survey designed to include questions to determine zygosity as well as questions from the Child Behavior Checklist (CBCL). Twins screened positive if one or both twins scored at least three points on the CBCL (Nordenbæk, et al., 2014). In the second phase, the parents of positive probands were given questionnaires to evaluate social communication and screen for ASD.

Finally, in the third phase positive probands with questionnaire scores over the 90th percentile and twins with an existing diagnosis of ASD were invited for a clinical assessment. Assessments included psychological assessment tools as well as medical exams and DNA tests. Researchers found that the ASD probandwise concordance rate was as high as 95.2% in monozygotic twins, and as low as 4.3% in dizygotic twins (Nordenbæk, et al., 2014).

The heritability calculations of this study are limited by its sample size, but its results are strengthened by previous research confirming similar monozygotic concordance rates as well as the blinded status of clinicians to the zygosity of the twins studied. However, with the entirety of this study considered, the results of this research seem to support the existence of a genetic component to ASD. Although the sample size is limited, the results are supported by prior research and seem to indicate at biological etiology of ASD.

Related research by Bölte et al. is represented by an ongoing research project underway in Sweden (2014). Although this research takes a more specific interest in monozygotic twins with discordant diagnoses of ASD, there are many similarities in their research foci. This project seeks to explore discordant diagnoses of autism in monozygotic twins. Included in this research are fifty-five twin pairs between the ages of 9-23 with an average IQ of 93.3. Participants are being compared to typically developing twin pairs acting as control subjects.

Results so far indicate a neuronal-cortical abnormality that results in many of the symptoms associated with ASD (Bölte et al., 2014). However, this research is limited by its very nature as twins are notoriously of lower birth weight and commonly premature (Bölte et al., 2014). This makes generalizing this research to singletons difficult. However, as the collection of data is a continual process, in the next few years it should offer valuable insight.

Increasing parental age as a risk factor. A 2009 study published in the American Journal of Epidemiology utilized a system similar to the registry utilized in Denmark by Nordenbæk, et al. (2014). In this study researchers used data from the California Department of Developmental Services (DDS) to explore the effects of parental age on the risk of developing ASD (Grether, Anderson, Croen, Smith, & Windham, 2009). Unlike the previous two studies, this study specifically excluded twins to exclude the effects of multiple births (Grether, et al., 2009). Overall, 23,311 of the 7,550,026 children born during the target years of 1989-2002 were identified as having been diagnosed with ASD.

Researchers used logistical regression to analyze a possible correlation between increasing parent age and an increased risk for ASD. Although similar research had been conducted prior to this one, this was the original study to consider maternal and paternal age as independent variables. The results revealed that a ten-year increase in maternal age correlated with a 38% increase in risk for ASD, while a ten-year increase in paternal age correlated with a 22% increase in risk (Grether, et al., 2009). Of interest, researchers noted that the first-born children of older parents were also at a greater risk for ASD.

This study is limited by its use of a single source for collection of data, and by the fact that the data available through the source was limited. In addition, children registered with the California DDS must meet criteria for a substantial handicap (Grether, et al., 2009), perhaps understating the actual number of children who are affected as it does not account for high-functioning children. However, this study is strengthened by its very large population and the precision of the data collected (Grether, et al., 2009).

Genetic implications of increasing parental age and ASD risk. In the above study, researchers hypothesized that an explanation for the increase in risk corresponding with increased paternal age lies in an increase of de novo mutations in the male germ cell (Grether, et al., 2009). Related research by Drake, Charlesworth, Charlesworth, & Crow measured the rates of spontaneous mutation in various cells (1998). Although this research is limited in its usefulness to this paper by the fact that it was not intended or pursued with the specific target of exploring ASD its results are extremely helpful in understanding some of the processes believed to be at the center of its etiology. In this study, researchers discovered that the rate of mutation in a male germ cell is much higher than that of a female germ cell. Specifically, researchers noted that the rate of genetic mutation is significantly higher for older males, as the cells have undergone much more division.

Researchers noted that a significant majority of genetic mutations occurring in humans are paternal in origin and passed through the male germ cells. It was suggested that an abnormality in the process of DNA methylation may be responsible for the larger male mutation rate (Drake, et al., 1998). Furthermore, researchers note that the increase from younger males to older males in cell division is not accounted for by cell-division dependent processes alone (Drake, et al., 1998). Although they note that human data is less reliable than other species, this research is strengthened by supporting evidence confirming similar findings.

Related research by Dong, et al. also implicates paternal origins in genetic mutations, but this time in the specific context of ASD (2014). In this study, researchers used whole exome sequencing to investigate exome data for 787 families with ASD. Using this method, researchers were able to explore the genetics of unrelated probands for similar genetic abnormalities. Researchers were specifically interested in the role of de novo insertions and deletions (indels) in the exomes of families with ASD. Using this information, researchers reported findings that suggest de novo indels (deletions) are inherited paternally 88% of the time, and that these indels are associated with ASD at a 1.6 odds ratio (Dong, et al., 2014). ‘

Furthermore, researchers implicated two genes as being associated with ASD, one of which is a chromatin regulator while the other is a regulator of synaptic vesicle release (Dong, et al., 2014). Researchers also associated de novo indels with a lower IQ. Due to a relatively small number of identified indels (n=119) researchers were unable to correlate an increased number of indels with increased paternal age. However, this research is strengthened by the results indicating a paternal origin for de novo mutations and its implication of genes involved in synaptic function as being associated with ASD.

While the previous study was unable to demonstrate a positive correlation between increasing paternal age and an increase in de novo mutations, a 2012 study by Kong, et al revealed evidence to support the hypothesis that the father’s age is a risk factor for ASD by demonstrating that the genetic mutations believed to contribute to ASD increase with age. The population studied was an Icelandic group of 78 parent-offspring trios. Similar to the research presented above, these researchers utilized genetic sequencing methods.

By studying these mutations in relation to parental age, researchers indicated that the mutation rate of these genes seems to be determined by the father’s age at the child’s conception, with an increase of two mutations per year (Kong, et al., 2012). However, these results are not linear as researchers observed that the mutation rate doubled every 16.5 years. These two findings alone suggest that there may be a significant risk to future generations if the current trend of postponing the age of having a first child continues.

Although the genetic recombination rate is higher for women, this study revealed results complementary to those found by Dong, et al. in that males were much more likely to transmit mutations. Researchers also revealed that paternal age at the time of conception seemed to be the primary determinant of the number of de novo mutations found in the child (Kong, et al., 2014). Rather than suggesting that parental age is the sole factor to consider in determining risk for ASD, researchers suggest that paternal age should be considered in research into the effects of other factors.

In summary, the purpose of this section was to evaluate the evidence in support of a biological etiology for ASD. Twin studies suggest that based on differing concordance rates between monozygotic and dizygotic twins that there may be a genetic component to ASD. Research above also suggests that increasing parental age is a risk factor for autism, and it is believed that genetic mutations are in part responsible. Specifically, researchers implicate de novo mutations as being associated with the development of ASD. These mutations have been observed to significantly increase with age, specifically those of paternal origin. With etiology evaluated and a potential cause in mind, the next step in this review is to explore the potential effects of the mutations implicated above.

2.2 Physiological Psychology

Model of disease. In a 2012 study, researchers in Seattle explored ASD from the perspective of a molecular neuroscientist. Their interest was specifically centered on de novo mutations, synaptic function, and chromatin pathways (Krumm, O’Roak, Shendure, & Eichler, 2014). This research also utilized exome sequencing methods. Researchers concluded that ASD does not fit within traditional Mendelian models of disease but appears to fit within a more complex/rare variant model. Rare variants are genes that have a minor allele frequency of less than 1%. The minor allele frequency is the frequency of the least common allele. The required combination of this extremely rare variant and de novo mutations would make the likelihood of developing ASD extremely low. However, the increased frequency of de novo mutations with age would also in theory increase these chances.

Researchers reported that there are significantly more de novo mutations found in probands than in unaffected siblings, and that certain synaptic pathways are associated with ASD (Krumm, et al., 2014). These observations are important for multiple reasons. The first, is perhaps obvious given the findings above. The latter observation is an important one as the discussion moves beyond etiological causes and enters the realm of effect. Researchers in this study identified multiple genes with roles in synaptic function as being potentially affected by de novo mutations.

This final observation is of the utmost importance as this discussion continues, because the mutated genes in question are responsible for major neurodevelopmental processes. Researchers identified several genes that form a network, the first of which is a scaffolding protein of significance to the development of signaling pathways. In addition, researchers identified several genes identified as being susceptible to de novo mutations as being responsible for activating neurodevelopmental genes with critical roles in plasticity, neuronal development, and processes such as synaptogenesis (Krumm, et al., 2014).

Deficits in synaptic function and autophagy. The identification of these effects is of the utmost importance, as related research by Tang, et al implicated impaired synaptic neurodevelopmental processes in the development of ASD (2014). In this study, the brains of subjects with ASD were examined postmortem. These were compared to the brains of typically developing control subjects for the purposes of neurological comparison. One of the findings was an increased density of the dendritic spine, suggesting a deficit in pruning.

Researchers compared the pruning rates of individuals with ASD to those of control subjects and found that while the former experienced a 41% decline in dendritic spine density, the participants with ASD only experienced a 16% decline (Tang, et al., 2014). The area of specific interest in this study is known as the Brodmann Area 21 and is involved in a number of social and communicative processes (Tang, et al., 2014). As social and communicative impairments are indicative of ASD, it would seem to suggest a correlation.

Upon further investigation researchers determined that the abnormal dendritic spine density was a result of pruning deficits, which were in turn apparently caused by a lack of autophagy (intentional destruction of cells) (Tang, et al., 2014). Researchers suggested that this loss of autophagy is caused by a hyperactive mTOR (mammalian target of rapamycin, mediates cell growth). This is a crucial concept as mTOR plays a critical role in the process of synaptogenesis and synaptic pruning, as neural networks are organized. Researchers warn that interruption of these neurodevelopmental processes would in fact cause immature or deviant neural circuits as commonly observed in individuals with ASD (Tang, et al., 2014). Researchers reported that rapamycin (discussed in the next section) helped to normalize these findings.

Imaging studies. Related research by Carper & Courchense also examined the brains of individuals affected by ASD (2005). Using a magnetic resonance imaging scanner researchers compared images of the brains of twenty-five children diagnosed with ASD to a group of typically developing controls. They reported that the frontal cortex grows very quickly in the early childhood of those affected by ASD, and although growth slows later in childhood the effects may still be apparent as neural connectivity and responsivity can be affected permanently (Carper & Courchense, 2005).

The images revealed that the dorsolateral prefrontal cortex and the medial frontal cortex regions of the frontal lobe are enlarged in children with ASD under the age of five, but that there is less dorsolateral prefrontal cortex growth with age observed in older children with ASD (Carper & Courchense, 2005). These findings complement theories that suggest that ASD is characterized by a period of overgrowth followed by a period of slowed growth. To support this theory researchers, point out that between 6 and 14 months, the head circumference of a child with ASD is around the 95th percentile on average (Carper & Courchense, 2005).

Related research by Lange, et al. also utilized imaging studies to compare the brains of children affected by ASD to typically developing controls (2015). However, the results of this study revealed longitudinal volumetric changes in the whole brain as opposed to focusing on regional differences. Participants included 100 male subjects diagnosed with ASD, all of whom received between one and three scans. Researchers observed an atypical pattern of volumetric brain changes over the lifetime as compared to control subjects.

Specifically, imaging revealed an increased volume in early childhood followed by a decrease in later childhood (meeting the control curve between ten and fifteen years) (Lange, et al., 2015). These results support above findings that denote a similar pattern. Of significant difference is a major abnormality observed in which the individuals with ASD demonstrated a volumetric decline throughout adulthood.

Synaptic pruning abnormalities. A 2014 study by Piochon, et al. sought to explore a link between motor learning and cerebellar plasticity in individuals with ASD. These researchers also observed a synaptic pruning deficit, stemming from what appeared to be abnormal neurodevelopment in a mouse model. These are the same findings described by Tang, et al. found as a result of postmortem examinations of human brains (2014). Researchers caution that synaptic pruning is a crucial component of forming neural circuits; in fact, abnormalities in this process could indeed result in many of the symptoms we associate with ASD (Piochon, et al., 2014).

In this study, researchers used a mouse model of a specific genetic copy-number variant mutation often observed in individuals with ASD. They reported pruning deficits, observed in the mouse model through deficits in a natural process known as the elimination of extra climbing fibers which is used as a model for autophagy and synapse elimination in humans (Piochon, et al., 2014). Researchers caution again in this study that impairments in such a critical process in neural development could have lifelong implications for neural circuitry. Researchers also note that these impairments could also explain motor learning deficits observed in ASD, stemming from an observed deficit in cerebellar plasticity (Piochon, et al., 2014). The limitations of this study include an inability to establish causation, and the use of an animal model.

Research by Thomas, Knowland, & Karmiloff-Smith also investigated abnormalities in the pruning process as being implicated in ASD (2011). However, these researchers hypothesized that the abnormalities were the result of overaggressive pruning as opposed to a lack thereof. Researchers created a neural network model to explore developmental regression- a symptom almost unique to ASD. Certainly, the type of regression observed in children with ASD is unique to the disorder. Researchers hypothesized that this regression was the result of overaggressive pruning processes (Thomas, et al., 2011). Developmental regression is considered indicative of ASD in addition to being unique to the disorder and seems to be linked to atypical synaptic pruning processes.

An alternative theory concerning the etiology of ASD suggests that causation can be attributed to environmental factors. This theory may stem from a phenomenon mentioned by Thomas, et al. in which children in Romanian orphanages displayed a form of “quasi-autism” (2011). These children displayed several symptoms of ASD, but the symptoms were milder and yielded a better outcome than those with true autism (Thomas, et al., 2011). In this study, researchers evaluated this possibility, but found that the more severe the developmental regression (and by extension the ASD) there was less of a discrepancy between high or low environmental risk.

The limitations of this study include the fact that the model used was specifically designed to reflect results for developmental regression as opposed to ASD in general. However, as developmental regression is a symptom observed almost uniquely in ASD, the information remains of use to the study of the disorder. However, the results of this study are strengthened by subsequent studies also suggesting that atypical pruning patterns may play a role in the development of ASD.

In summary, this section presented evidence for the interaction of psychological and physiological symptoms observed in ASD. A complex model of disease was presented, as well as research implicating synaptogenesis, a hyperactive mTOR pathway, and pruning deficits as possibly stemming from genetic mutations. Imaging studies suggested atypical volumetric brain development as well as an alternative growth pattern has been observed in individuals with ASD. With developed hypotheses established as to the effects of genetic mutations with proposed causal effects, this next section will explore treatment options in light of this information.

2.3 Contemporary Issues in Psychology

In consideration of at least a partial biological etiology and the interaction of psychological and physiological symptoms, the third object of this paper was to determine how this knowledge should affect treatment. One of the major ongoing debates in the mental health community concerns the pharmaceutical treatment of mental health disorders in children. This question has once again been raised in recent years due to the increase in autism diagnoses. Some professionals do not believe that pharmacotherapy is a viable or necessary treatment for children with autism, while others disagree, citing a reduction of symptoms as evidence of success. However, following a presentation of the evidence for a biological etiology, it seems prudent to also present evidence in consideration of treatment options.

Alternative treatment with micronutrients. While research on this topic has been negligible, many parents are turning to vitamins and supplements for their children’s ASD treatment. While empirical evidence is extremely limited, a 2011 study claimed to find success (Mehl-Madrona, Leung, Kennedy, Paul, & Kaplan). In this study, researchers recruited 88 children with autism and their families, and offered them their choice in treatment. Researchers recruited 44 children whose families requested a non-pharmaceutical treatment and matched them with 44 children whose families requested conventional treatment. The former group was provided with micronutrient treatment while the latter received conventional pharmaceuticals.

The results of this study revealed a reduction in symptoms for both groups, but the micronutrient group’s reduction was significantly greater (Mehl-Madrona, et al., 2010). However, there are several flaws in the research methods of this study which severely compromise its results. First, the children were not randomly assigned to their groups. The parents who chose the micronutrient group were required to administer up to eighteen pills a day, requiring more commitment and involvement on their part.

In addition, this study did not control for treatments or therapies outside of the study parameters. This is a concern, because the micronutrient group has already requested a treatment requiring a higher level of motivation. It should be considered that these families were more highly motivated than the families who chose conventional treatments as the micronutrient group was more than three times as likely to follow additional recommended treatment options (Mehl-Madrona, et al., 2010). The micronutrient families are therefore more likely to adhere more strictly to administration schedules, as well as participate in treatments and therapies outside of the study parameters.

Finally, the sole clinician whom all participants were under the care of was not blind to the study conditions. While the population size was larger, there remain significant limitations to this research and to the conclusions that micronutrients are an alternative treatment to pharmacotherapy. However, it does suggest that treatment with conventional pharmacotherapy resulted in an improvement in symptoms to at least some degree. Due to the nature and purpose of this research there is a significant risk of researcher bias. These limitations coupled with the uncontrolled variable of familial motivation call into question the reliability of the study findings.

Rapamycin. Rapamycin is an immunosuppressant drug that is currently the subject of many experiments in which researchers are using it to treat an animal model of ASD in rats and mice. As explained by Tang, et al. above, a hyperactive mTOR is believed to contribute to the development of ASD (2014). To review, mTOR is the mammalian target of rapamycin and it mediates cell growth, assisting in proliferation. Rapamycin targets this hyperactive mTOR and block some of its functions. As an up-regulated mTOR is also observed in some cancers, rapamycin is being introduced into some cancer treatment protocols.

Tuberous sclerosis is a condition that has a co-morbidity with ASD and causes the growth of benign tumors in organs all over the body. It is also caused by excess cell proliferation. This, coupled with its co-morbidity with ASD makes it an ideal model for testing the effects of rapamycin. In a 2012 study, researchers published the results of a study that tested the effects of rapamycin on a mouse model of tuberous sclerosis complex (TSC) (Sato, et al.). The mice used were between 6-7 months old at the time of the study. These researchers again implicated an uninhibited mTOR in the association between ASD and tuberous sclerosis. In fact, researchers point out that about half of all patients with TSC also have ASD (Sato, et al., 2012).

The results of this study revealed that rapamycin reversed the impaired social interactions observed in the mouse model. In addition, researchers found enhanced transcription in several genes involved in mTOR signaling (Sato, et al., 2011). Researchers noted that the abnormal mTOR signaling in the mice with social deficiencies was of relevance. Specifically, researchers stated that their results suggest that mTOR inhibitors may be a useful pharmacotherapy treatment for ASD, even in adults (Sato, et al., 2012). This study is limited by the simple fact that it is an animal model, and by the fact that the specific model utilized was for TSC instead of ASD. However, it is strengthened b/y a strong correlation between the disorders, as well as the following studies confirming similar results.

Related research by Burket, Benson, Tang, & Deutsch also sought to observe the effects of rapamycin on a mouse model, but this study used an ASD model (2014). In this study, the mice were injected with rapamycin once daily for four consecutive days and retested on the fourth day. The results indicated that rapamycin improved the sociability of the mice affected by the ASD model. Researchers measured this by allowing the mice to spend time in an environment in which they were allowed to explore either an inverted cup or a stimulus mouse (Burket, et al., 2014). In addition to spending more time in the social versus the nonsocial compartment, the number of social approaches made by the ASD-model mouse increased.

Initially, the ASD-model mice spent more time exploring the cup than the stimulus mouse. However, after administration of rapamycin the ASD-model mice significantly increased the time they spent in active exploration of the stimulus mouse. Unlike the previous study, the mice used in this research were only four weeks old, suggesting that mTOR inhibitors such as rapamycin may be effective at a range of ages, as mice are not yet developed at this age. Although this study again represents an animal model, it does confirm previous findings that a hyperactive mTOR may be responsible for many of the core ASD symptoms, as well as the findings that rapamycin may help to improve these symptoms.

Despite the fact that these were animal models, rapamycin appeared to be effective at improving sociability in mice from the age of four weeks (Burket, et al., 2014) through six-seven months (Sato, et al., 2012), at which point a mouse will have attained its peak social and sexual maturity. These findings appear to suggest that mTOR inhibitors may be effective at improving sociability even if they are not administered until adulthood. If proven in humans, such findings may indicate the possibility of a treatment that can be administered throughout the lifetime.

A 2013 study sought to investigate the effects of rapamycin in humans, specifically for the purposes of symptom reduction in TSC. As mentioned above, some evidence suggests that both TSC and ASD may be associated with a hyperactive mTOR and share a fifty-percent co-morbidity rate. The present study included 86 children and rendered positive results. Although it took about six months to reach maximum effectiveness, the administration of rapamycin gave one hundred percent of participants complete control of their convulsions within one year of beginning treatment (Canpolat, et al., 2014).

Of even more interest, researchers reported that not only was rapamycin effective, it was also tolerated without any significant side effects. The limitations of this research include the sample size. While the study enrolled 86 children, only seven were administered rapamycin. As it is a relatively new treatment, the ideal dosage and duration of treatment has not been empirically proven (Canpolat, et al., 2014). However, this study is strengthened by the wealth of pertinent medical information available for each patient as well as the researcher’s two-year follow up. However, rapamycin has not been evaluated to determine long-term effects, requiring that providers exercise extreme caution when administering to developing children.

All of the above considered, the treatment of ASD with rapamycin at this time is not approved. This is due to the fact that rapamycin is a toxic immunosuppressant with dangerous side effects. While it has been used in the treatment of cancers and other potentially fatal disorders, it has not been deemed appropriate at this time for the treatment of ASD. However, variations of the drug may be a viable treatment option in the future and observing its effects in animal models helps to paint a more accurate picture of ASD. In the exploration of future treatment options, medications that act in a similar manner (inhibiting cell proliferation) may warrant consideration.

Risperidone. Risperidone is an atypical antipsychotic medication that has been FDA-approved to treat some of the behavioral symptoms of ASD. In previous generations, conventional neuroleptics were used for this purpose, but were largely discontinued due to severe extrapyramidal symptoms (EPS). Pharmacotherapy treatment with conventional antipsychotics has largely given way to treatment with atypical neuroleptics due to reports of increased efficacy and EPS reduction.

A 2008 study by Miral, et al. sought to compare the efficacy of risperidone with that of haloperidol, a conventional antipsychotic previously used in the treatment of ASD. In this study, researchers randomly assigned 30 child and adolescent participants to either treatment for twelve weeks as part of a double-blind study. The dosages of both drugs were kept at comparable doses between the ranges of 0.01-0.08 mg/kg/day (Miral, et al., 2008). While the use of both drugs led to improvements, the participants taking risperidone had a significantly greater reduction in symptoms as measured by two different rating scales. In addition, the participants taking risperidone improved in all subscale scores, while the participants taking haloperidol did not improve in the language subscale (Miral, et al., 2008).

As with all antipsychotic medications, atypical or conventional, both groups reported side effects. Both groups reported constipation, enuresis nocturna, and upper respiratory tract infections, but the haloperidol group also reported blunted affect, rigidity, difficulty sleeping, and increased appetite (Miral, et al., 2008). In addition, the participants in the haloperidol group demonstrated an increase in extrapyramidal symptoms during the trial. This was not observed in the group taking risperidone.

During the study, no serious extrapyramidal effects were observed, but this may be due to the relatively short duration of treatment. As antipsychotics must be administered daily for their effects to continue (likely meaning children with ASD would require them daily for the duration of their lives), a twelve-week study is not sufficient to measure the risk of severe side effects with either drug. However, the results do suggest that there is a reduced risk of EPS with risperidone, as expected.

This study is limited by its small sample size, as well as a lack of a homologous group of patients. However, it does provide evidence supporting the general safety and efficacy of risperidone for ASD. In particular, it suggests that providers are correct in their efforts to make the shift from using conventional antipsychotics to atypical ones. It does not appear based on this evidence that doctors or parents must compromise by sacrificing effectiveness for safety.

Related research by Nagaraj, Pratibha, & Prahbjot also studied the effects of risperidone in the treatment of ASD, but this time in comparison to a placebo (2005). The length of the study was six months and included 40 children between the ages of 2 and 9. The majority of children in the treatment group demonstrated improvement in two different rating scales, as well as improvement in communication and social responsiveness alongside a reduction in hyperactivity and aggression (Nagaraj, et al., 2005).

In addition, researchers reported improvements in overall functioning as well as a smoother pattern of daily activity, especially eating and sleeping (Nagaraj, et al., 2005). Researchers also present a theory that ASD can be compared to schizophrenia in terms of treatment considerations. It is believed that the mix of serotonin and dopamine antagonists founds in risperidone act to treat the positive and negative symptoms of schizophrenia. If considered in a similar manner, researchers suggest the symptoms of ASD can also be divided into these categories. For example, the positive symptoms of ASD would be aggression, irritability, tantrums, etc., while the negative symptoms would be social responsiveness or communication deficits (Nagaraj, et al., 2005).

This study was limited by a rather small sample size, as the actual treatment group consisted of only nineteen participants. However, it was strengthened by the fixed dosage utilized. Parents reported an increase in both appetite and weight as well as mild dyskinesia in three children, but none of the symptoms were severe enough to warrant the child’s withdrawal (Nagaraj, et al., 2005). Researchers hypothesized that the lower dose utilized contributed to the relatively low number of dyskinesia and EPS in comparison to previous studies.

While risperidone has been a helpful tool for individuals with ASD and their families, it is not a cure. Risperidone does not treat the underlying cause of ASD. There isn’t sufficient evidence available to argue for or against the long-term safety of the drug in children, nor are its potential effects on neurodevelopment known. At this time, it remains an effective tool at reducing certain symptoms, but this is no more a long-term solution for individuals with ASD than is taking an aspirin for an individual suffering from chronic migraines.

The object of this section was to discuss the implications of a biological etiology for treatment options. Specifically, this section sought to explore evidence presented to address questions relative to the safety and efficacy of pharmacotherapy in the treatment of ASD. Research was presented evaluating current treatment options in order to explore whether pharmacotherapy is a medically and psychologically justifiable treatment option for individuals with ASD. Included in this section were several pharmacotherapy or alternative treatment options that work to treat either the outward symptoms or a suspected underlying cause of ASD by creating chemical or physical differences in the brain.

2.4 Crisis and Emergency Management

Assessment. The final section of this paper concerns itself with the induction of ASD symptoms in rats and mice, and the findings discovered as a result. This section explores the implications of a biological etiology for the assessment of ASD. Assessment is an invaluable tool in the treatment of any model of disease, but particularly for a complex model such as ASD. The development of a reliable model of disease is imperative before treatment efforts can proceed, or a complete understanding of the disease can be gained. Presented below is a review of research utilizing an animal model created for the assessment of physiological and psychological symptoms of ASD induced in rats and mice.

Valproic acid (VPA) is a known teratogen that can cause a range of birth defects. VPA has been used in rats and mice to model ASD, as exposure to VPA in early embryonic development causes symptoms of ASD in humans. These symptoms are also observed in rat and mouse models exposed to VPA. In addition to these symptoms, there are physical abnormalities observed in these VPA-exposed animal models such as abnormalities in brain volume, genetic abnormalities, abnormal synapse formation, and abnormal pruning (Nicolini, Ahn, Michalski, Rho, & Fahnestock, 2015).

The importance of this model will be discussed in order to highlight the importance of assessment. Each of the VPA models below was designed to assess ASD from a different perspective. Several models were used to assess the neurological symptoms of ASD, while others were used to assess behavioral symptoms. In either case, the creation of a model for the purposes of assessment was very important as it allows researchers to confirm or question previous findings in humans with ASD that were tentatively associated with the disorder.

VPA in the assessment of neurological symptoms. VPA is utilized in several studies to model neurological symptoms commonly observed in individuals with ASD. The role of mTOR signaling pathways is further explored in a VPA animal model of ASD. In addition, the process of neurogenesis is explored through investigations into the gene responsible for its regulation. Finally, abnormalities in the prefrontal cortex are examined for their suspected role in ASD.

Role of mTOR and neurogenesis. In a 2015 study, researchers in Canada sought to explore the association between ASD and mTOR signaling pathways (Nicolini, et al.). Researchers collected brain tissue samples from eleven postmortem patients with ASD as well as from VPA-exposed rats. It was observed that in these patients as well as the VPA-exposed rats, the mTOR pathways were actually down-regulated. Researchers stated that the evidence suggests that abnormalities in either direction of this pathway could be associated with ASD (Nicolini, et al., 2015). This is in contrast to previous research implicating an up-regulation of mTOR.

The limitations of this study include a small sample size, as well as the fact that cause of death could not be matched for the human subjects. Researchers caution that a disruption in the mTOR pathway could not only create problems in the establishment of neural networks, but also in their long-term function (Nicolini, et al., 2015). The strengths of this study include the fact that similar findings were noted between VPA-induced rats and humans with ASD. In addition to modeling behavioral symptoms, the VPA-exposed rats demonstrated physical differences in brain structure. These findings suggest that the VPA-exposure model in rats could be a valid model of ASD in humans.

Related research by Almeida, Roby, & Krueger also studied the neurological symptoms of VPA-exposed rats (2014). Researchers in this study sought to explore the effects on neurogenesis by studying the gene that regulates it: BDNF. The results of the study revealed that the expression of BDNF was increased in the brains of VPA-exposed rats. Researchers warned that due to the major role BDNF plays in the process of neurogenesis, disruption of the natural process could result in permanent consequences resulting from deficits in connectivity (Almeida, et al., 2014).

It was revealed that when a fetus is injected with BDNF the process of neurogenesis occurs early, and proliferating cells exit the cell cycle prematurely (Almeida, et al., 2014). Of note, in comparison to the previous study the rats were injected at E12.5 as opposed to E9. Interestingly enough, the maternal brain did not display similar increased in BDNF expression after administration, suggesting a major difference between the effects of VPA on developed adult versus developing fetal brains (Almeida, et al., 2014).

2.5. Prefrontal cortex

Related research by Martin & Manzoni used the VPA rat model to study the effects of VPA-exposure in adulthood (2014). These researchers also studied the neurological effects of VPA exposure, noting the effects of abnormal synaptic pruning patterns. This research was interesting in itself because the researchers chose to conduct a longitudinal study in order to observe the long-term effects of VPA exposure. This study was conducted in response to several studies published previously suggesting that synaptic abnormalities self-corrected in adolescence. In contrast, these researchers discovered that synaptic function is reduced in adult VPA-exposed rats (Martin & Manzoni, 2014).

Of interest, these findings are reminiscent of the volumetric studies described above, particularly those in which a volumetric growth curve was described (Lange, et al., 2015; Carper & Courchense, 2005). In particular, the findings of Lange, et al. support a specific growth curve initiated by the well-documented explosion of synaptic growth that characterizes the early childhood neural development of children with ASD, followed by an atypical decline in adulthood (2015). Of the most relevance to the current study is the period described by Lange et al. as occurring between the years of 10-15 (adolescence) in which growth appears to slow to meet the rate of typically developing controls before entering the decline of adulthood (2015).

Considering this curve, the findings by Martin & Manzoni and their predecessors seem less mysterious. In fact, they indicate what one might expect to find in a longitudinal study of neurodevelopment in individuals with ASD based on previous research. Rather than self-correcting, adolescence appears to merely characterize a period of unusually slow growth, which follows a period of overgrowth. These two stages combined would likely cause a brain to appear during adolescence as though it was developing normally.

Related research by Hara, et al sought to investigate the effects of VPA on monoamine levels in a mouse model (2015). In particular, these researchers were interested in the effects of VPA exposure on the dopaminergic system. Researchers used methamphetamine (METH) injections to further explore these effects. Of note, the study did not reveal abnormalities in any of the monoamine levels. However, these results should be interpreted with caution, as the VPA was injected at E12.5 instead of E9, creating a problem as previous research studies have cited E9 as the critical period for rat and mouse models in order to most accurately model the sensitive period in human development.

Of interest, the results stemming from the portion of the study involving METH were intriguing. Researchers observed that prenatal exposure to VPA reduced METH-induced hyperlocomotion, but only in male mice (Hara, et al., 2015). As mentioned above, this study is limited by the day of in utero exposure. In addition to perhaps being a less representative model, this condition presents another problem, as it becomes difficult to compare these results to those of previous studies. However, the results stemming from the METH portion of the experiment suggest that hypofunction of the dopaminergic system may be an effect of prenatal VPA-exposure (Hara, et al., 2015).

VPA in the assessment of behavioral symptoms. In this section a VPA model is used to assess several behavioral symptoms of ASD. These behaviors include obvious observable symptoms such as hyperactivity and repetitive behaviors as well as more subtle symptoms such as social responsiveness, nonexploratory movement, and sociability deficits. Finally, research is presented in which the VPA animal model was used in the assessment process, leading to the development and administration of a treatment.

A 2012 study by Japanese researchers Narita, et al. sought to observe the behavioral effects of VPA on rats (2010). The rats were divided into three groups of 7-8 rats each. On embryonic day 9, pregnant rats were injected with either VPA, thalidomide (THAL), or a placebo. This day was chosen specifically to model the time at which a human embryo would be most susceptible if exposed to VPA. Researchers observed that both rats injected with VPA or THAL displayed behavioral symptoms as well as neurological symptoms.

Specifically, researchers observed that the teratogen exposed rats displayed more nonexploratory movement than the controls, as well as deficits in achievement of learning tasks (Narita et al., 2010). Researchers did not note any social responsiveness deficits in the teratogen groups, although it was suggested that sample groups of 7-8 may have been too small to reflect these differences. The sample size is of concern, and certainly presents a limitation to this research. The use of thalidomide is a novel concept in this paper, but the sample size makes these results difficult to generalize. However, this study is strengthened by the observation that serotonergic neural development is abnormal in the teratogen-exposed groups. The latter seems to complement the notes of Nagaraj, et al., as it was suggested that many of the social responsiveness and communication symptoms observed on autism were explained by abnormalities in serotonin (2005).

The final study to be presented in this paper seems fitting, as it provides a conclusion to both this section and this paper with a glance towards the future of ASD research. In this model, researchers utilized an animal model including both rats and mice, the first study presented here to do so. Researchers Kim, et al. sought to investigate a possible treatment for abnormalities observed in these models (2014). In this study, researchers studied the prefrontal cortex due to its implications in ASD and its involvement in social processes and behaviors.

Specifically, they investigated a hypothesis that acetylcholine (ACh) could be implicated in ASD. This hypothesis was based on the associations between dysregulated ACh and other neurological disorders such as Alzheimer’s disease and schizophrenia (Kim et al., 2014). In light of the many similarities drawn between schizophrenia and ASD, this seems to reflect a plausible theory. For example, ASD and schizophrenia share the existence of symptoms that can be categorized as positive and negative, and these symptoms are improved with neuroleptic medications (conventional or atypical) designed to affect dopamine and serotonin.

Researchers observed that compared to controls, the VPA-exposed models demonstrated increased levels of AChE, the primary metabolic enzyme of ACh and decreased levels of ChAT, the rate-limiting enzyme for ACh (Kim et al., 2014). The former statistic is of the most interest, as an elevated AChE level can negatively affect the amount of ACh in the brain (Kim et al., 2014). With this information, researchers elected to test the effects of donepezil, an AChE inhibitor used in the treatment of Alzheimer’s in the hopes of raising the ACh levels to a more normal state.

Researchers used a marble-burying test to measure repetitive behaviors typically found in ASD. While the VPA-exposed mice displayed more repetitive behaviors at the baseline but following treatment with donepezil the VPA-exposed mice performed at the same level as controls (Kim et al., 2014). Researchers noted a similar pattern in other measures. For example, in addition to improvements in repetitive behavior researchers observed that VPA-exposed mice demonstrated deficits in hyperactive behavior, cognitive flexibility, and several measures of sociability, all of which improved following treatment with donepezil without exception.

In conclusion, researchers noted that following prenatal exposure to VPA, both animal models demonstrated increased levels of AChE and poorer scores on a number of scales relating to core symptoms of ASD. Following treatment with donepezil, the affected animals improved on all measures. Although it is still early in the research process for this drug, there are several potential advantages to these findings if proven effective. First, donepezil is already FDA-approved and demonstrated safe for the treatment of other neurological disorders. In addition, it is associated with fewer serious side effects than rapamycin and risperidone. Although many more studies are warranted before this theory can be tested, for now it provides a sense of direction for the future.

3. Conclusion and Future Research

The research on this topic has generated a global interest. In fact, the research presented above stems from thirteen countries around the world. It includes participants from a variety of backgrounds, ethnicities, cultures, and socioeconomic statuses. Given the variety, one might expect to see a wide range of results. However, that has not been the case. In fact, the above results have been generally consistent, enough to generate a picture of ASD as it is currently understood. From this research, it is possible to extrapolate a few key points as well as foci for future research development.

In the first section, the object was to examine the evidence for a biological etiology for ASD. Based on the twin studies, the significant difference between monozygotic and dizygotic concordance rates suggests a genetic component to ASD. As mentioned above, research on this subject is ongoing, and larger sample sizes may further clarify these correlations. Following these, evidence was presented suggesting that increasing parental age was associated with an increased risk of ASD. In explanation of this increase, research demonstrating rates of spontaneous mutation was presented in which it was revealed that mutation rates increased with age in male germ cells. In light of this, the specific type of mutations implicated were explored, leading to the revelation that de novo mutations appeared to be associated with ASD, and seemed to be inherited predominately from paternal genes.

The second section concerned itself with the physiological factors that may be affecting psychological symptoms in light of a biological etiology. This section built off the previous, beginning with the reports that de novo mutations were increased in probands with ASD, and implicating several neurodevelopmental processes as being susceptible to de novo mutations. These processes were later identified and expanded upon to include the processes of synaptogenesis and pruning. Imaging studies were presented in which volumetric brain patterns were identified as appearing to be atypical in individuals with ASD. Specifically, an atypical growth curve was identified.

In the third section, treatment options were explored in light of a biological etiology and in consideration of the interactions between physiological and psychological symptoms. Micronutrient, antipsychotic, and immunosuppressant treatment options were presented in detail. These treatment options were presented in comparative as well as placebo-controlled studies to evaluate their effectiveness.

In the final section, the importance of assessment was illustrated. Specifically, the implications of a biological etiology for ASD were explored in light of the interaction between physiological and psychological symptoms and in consideration of the revelations gained from the results of research into current treatment options. An animal model using VPA to induce ASD symptoms was utilized in the assessment portion of each of the studies presented in this section and both neurological and behavioral symptoms were explored. This section and the literature review concluded with a study in which VPA-exposed mice and rats were assessed. Based on these findings, a novel treatment was developed and tested yielding positive results.

There are several conclusions that can be draw from this paper, although it seems that they raise as many questions as they answer. The first of these is that there seems to be sufficient evidence to support at least in part a biological etiology, as evidenced by twin concordance rates suggesting a genetic component to the disorder. However, this twin research is in progress and the present evidence is not comprehensive enough to be considered definitive. While in and of itself increasing parental age may not be indicative of a genetic component, the increasing mutation rate with age coupled with evidence correlating ASD with de novo mutations suggests that further investigation into the exact mechanism and resulting effects of mutation is warranted. Specifically, paternal age at conception is implicated as the most significant effects of genetic mutations seem to occur in the male sperm cell.

However, research has not been conducted to determine whether modern technology such as cryopreservation, in-vitro fertilization, or similar methods would allow older parents to reduce their risks of transmitting mutations to their children. As the age of sperm cell at conception has been implicated in these studies, it seems plausible that the use of cryopreservation to allow for conception with younger sperm may circumvent these risks. However, this research also requires additional research to determine whether alternative methods of conception pose risks of their own.

The de novo mutations mentioned above were implicated in disrupting the process of synaptogenesis as well as pruning patterns. This initial period of rapid growth is supported by the presence of macroencephalopathy in children with ASD, imaging studies documenting both regional and overall volumetric brain enlargements, a lack of autophagy, and documented abnormal patterns of brain growth. However, more longitudinal studies are required in order to confirm this pattern of development. The process of pruning is a critical phase in early neurodevelopment, as it helps to organize neural circuitry. An interruption in this process could permanently alter the synaptic connections of the brain.

These abnormalities could in fact lead to many of the core symptoms of ASD, although longitudinal studies would be required in order to examine these effects over the course of a lifetime. While mTOR has been implicated in ASD, studies are not consistent in determining whether down or up-regulation is at fault. Some studies suggest a disruption in either direction could cause symptoms of ASD. Future research is needed in this area to explore why some studies produced results suggesting an underactive mTOR as being implicated in ASD in contrast to the majority of research implicating an up-regulation.

While the use of rapamycin has been demonstrated to have the potential to reverse social impairments associated with ASD, as well as allow the pruning process to resume, it is not a viable treatment option for ASD. While ASD is certainly life-altering and a serious disorder, it is not fatal. As rapamycin is toxic, it is currently reserved for life-threatening diseases such as cancer and epilepsy. However, the research presented above suggests that mTOR inhibiting treatments warrant further research, and the development of less toxic variants should be a future research target.

Risperidone was included as it remains a current treatment option, but it is not a cure. It has the benefit of improving certain symptoms, but there is not sufficient evidence to support the long-term safety of risperidone in children. EPS and dyskinesia have been reported in children, but there is a need for longitudinal studies to determine how antipsychotic use in childhood might affect neurodevelopment throughout the lifespan. As VPA seemed to affect a developing brain but not a developed one, research to examine the effects of risperidone use in childhood on adults is important.

The symptom improvements observed suggest that the functions of risperidone on the serotonergic and dopaminergic systems warrant further research. The effects or presence of monoamine abnormalities is unclear and should be a target of future investigation. Only one study was presented to examine the effects of a micronutrient treatment plan, but due to significant issues with the research methods utilized it is impossible at this time to evaluate the treatment. Furthermore, there is no evidence to suggest that micronutrient therapy would have the potential improve symptoms of ASD.

Finally, the use of an animal VPA model offered support for several of the previous findings. However, one of these studies offered novel findings. The use of an animal model to induce ASD, observe abnormalities, formulate, and test a treatment based on those observations is an excellent example for the importance of assessment. In future research, prenatal exposure to VPA during the critical period appears to be a viable model for ASD. While the use of donepezil to treat abnormalities in ACh levels is a novel concept, the present VPA models allow for comparative studies. Future research into the ACh levels of humans with ACh is warranted to determine if they are also elevated. In addition, future research into treatment of ASD provides a general direction and the hope of a safer treatment.

Overall, ASD remains a puzzle. Although several of the pieces have been added, they have in many ways served to create an even larger puzzle. However, the information gained from each new study helps researchers gain a better understanding of a disorder quickly increasing in prevalence around the world. The etiology, symptoms, treatment options, and current models of assessment included in this paper are of importance to not only researchers but also to psychologists and mental health professionals working in the field.

Most graduating young professionals entering the mental health field this year will encounter an individual with ASD at some point in their careers. When they do, the same process this paper has taken applies. Before one can begin to assess and treat, they must establish an understanding of the disorder- beginning with its etiology, followed by its symptoms, available treatments, and then establishing a model for assessment and future treatment considerations.

4. References

Almeida, L. E. F., Roby, C. D., & Krueger, B. K. (2014).Molecular and cellular neurosciences: Increased BDNF expression in fetal brain in the valproic acid model of autism Academic Press. doi:10.1016/j.mcn.2014.01.007

Bölte, S., Willfors, C., Berggren, S., Norberg, J., Poltrago, L., Mevel, K., & ... Lichtenstein, P. (2014). The roots of autism and ADHD twin study in Sweden (RATSS ).Twin Research and Human Genetics: The Official Journal Of The International Society For Twin Studies,17(3), 164-176. doi:10.1017/thg.2014.12

Burket, J. A., Benson, A. D., Tang, A. H., & Deutsch, S. I. (2014). Rapamycin improves sociability in the BTBR T(+)Itpr3(tf)/J mouse model of autism spectrum disorders. Brain Research Bulletin, 100, 70.

Canpolat, M., Per, H., Gumus, H., Yikilmaz, A., Unal, E., Patiroglu, T.. . Kumandas, S. (2014). Rapamycin has a beneficial effect on controlling epilepsy in children with tuberous sclerosis complex: Results of 7 children from a cohort of 86. Child's Nervous System,30(2), 227-240. doi:10.1007/s00381-013-2185-6

Carper, R. A., Courchesne, E. (2005). Localized enlargement of the frontal cortex in early autism, Biological Psychiatry, 57(2), 126-133. http://dx.doi.org/10.1016/j.biopsych.2004.11.005.

Dong, S., Walker, M. F., Carriero, N. J., DiCola, M., Willsey, A. J., Ye, A. Y... Sanders, S. J. (2014). De novo insertions and deletions of predominantly paternal origin are associated with autism spectrum disorder. Cell Reports,9(1), 16-23. doi:10.1016/j.celrep.2014.08.068

Drake, J. W., Charlesworth, B., Charlesworth, D., & Crow, J. F. (1998). Rates of spontaneous mutation. Genetics,148(4), 1667-1686.

Grether, K., Anderson, M., Croen, L., Smith, D., & Windham, G. (2009). Risk of autism and increasing maternal and paternal age in a large North American population. Am. J. Epidemiol. 170(9): 1118-1126. doi:10.1093/aje/kwp247

Hara, Y., Takuma, K., Takano, E., Katashiba, K., Taruta, A., Higashino, K... Matsuda, T. (2015). Reduced prefrontal dopaminergic activity in valproic acid-treated mouse autism model. Behavioural Brain Research,289, 39-47. doi:10.1016/j.bbr.2015.04.022

Kim, J., Seung, H., Kwon, K. J., Ko, M. J., Lee, E. J., Oh, H. A... Bahn, G. H. (2014). Subchronic treatment of donepezil rescues impaired social, hyperactive, and stereotypic behavior in valproic acid-induced animal model of autism. PloS One,9(8), e104927.

Kong, A., Frigge, M. L., Masson, G., Besenbacher, S., Sulem, P., Magnusson, G.,…Stefansson, K. (2012). Rate of de novo mutations and the importance of father's age to disease risk.Nature,488(7412), 471-5.

Krumm, N., O’Roak, B. J., Shendure, J. & Eichler, E. E. (2014). A de novo convergence of autism genetics and molecular neuroscience. Trends in Neurosciences, 37(2), 95-105.

Lange, N., Travers, B. G., Bigler, E. D., Prigge, M. B.D., Froehlich, A. L., Nielsen, J. A., Cariello, A. N., Zielinski, B. A., Anderson, J. S., Fletcher, P. T., Alexander, A. A. and Lainhart, J. E. (2015). Longitudinal volumetric brain changes in autism spectrum disorder ages 6–35 Years . Autism Res, 8:82–93. doi:10.1002/aur.1427

Martin, H. G. S., & Manzoni, O. J. (2014). Late onset deficits in synaptic plasticity in the valproic acid rat model of autism. Frontiers in Cellular Neuroscience,8, 23-NA. doi:10.3389/fncel.2014.00023

Mehl-Madrona, L., Leung, B., Kennedy, C., Paul, S., & Kaplan, B. J. (2010). Micronutrients versus standard medication management in autism: A naturalistic case-control study.Journal of Child and Adolescent Psychopharmacology,20(2), 95-103. doi:http://dx.doi.org/10.1089/cap.2009.0011

Miral, S., Gencer, O., Inal-Emiroglu, F. N., Baykara, B., Baykara, A., & Dirik, E. (2008). Risperidone versus haloperidol in children and adolescents with AD: A randomized, controlled, double-blind trial. European Child & Adolescent Psychiatry,17(1), 1-8. doi:10.1007/s00787-007-0620-5

Nagaraj, R., Singhi, P., & Malhi, P. (2006). Risperidone in children with autism: Randomized, placebo-controlled, double-blind study. Journal of Child Neurology,21(6), 450.

Narita, M., Narita, N., Oyabu, A., Imura, Y., Kamada, N., Yokoyama, T... Uchida, A. (2010). Nonexploratory movement and behavioral alterations in a thalidomide or valproic acid-induced autism model rat . Neuroscience Research,66(1), 2-6. doi:10.1016/j.neures.2009.09.001

Nicolini, C., Ahn, Y., Michalski, B., Rho, J. M., & Fahnestock, M. (2015). Decreased mTOR signaling pathway in human idiopathic autism and in rats exposed to valproic acid. Acta Neuropathologica Communications,3(1), 3-3. doi:10.1186/s40478-015-0184-4

Nordenbæk, C., Jørgensen, M., Kyvik, K., & Bilenberg, N. (2014). A Danish population-based twin study on autism spectrum disorders. European Child & Adolescent Psychiatry, 23(1), 35-43. doi: http://dx.doi.org/10.1007/s00787-013-0419-5

Piochon, C., Kloth, A. D., Grasselli, G., Titley, H. K., Nakayama, H., Hashimoto, K Hansel, C. (2014). Cerebellar plasticity and motor learning deficits in a copy-number variation mouse model of autism. Nature Communications,5, 5586. doi:http://dx.doi.org/10.1038/ncomms6586

Sato, A., Kasai, S., Kobayashi, T., Takamatsu, Y., Hino, O., Ikeda, K., & Mizuguchi, M. (2012). Rapamycin reverses impaired social interaction in mouse models of tuberous sclerosis complex. Nature Communications,3, 1292. doi:http://dx.doi.org/10.1038/ncomms2295

Tang, G., Gudsnuk, K., Kuo, S., Cotrina, M. L., Rosoklija, G., Sosunov, A. Sulzer, D. (2014). Loss of mTOR-dependent macroautophagy causes autistic-like synaptic pruning deficits. Neuron,83(5), 1131-1143. doi:10.1016/j.neuron.2014.07.040

Thomas, M. S. C., Knowland, V. C. P., & Karmiloff-Smith, A. (2011). Mechanisms of developmental regression in autism and the broader phenotype: A neural network modeling approach. Psychological Review,118(4), 637.

[...]