Wendy Froehlich, MD

Autism spectrum disorders (ASDs) represent a heterogeneous set of neurodevelopmental disorders characterized by deficits in social communication and reciprocal interactions as well as stereotypic behaviors. The prevalence of ASDs has been increasing over the past 2 decades. According to the latest review of medical records in 14 selected sites in the United States conducted by the CDC, 1 in 88 children (1 in 54 boys and 1 in 252 girls) aged 8 years were identified as having ASDs.

 The reason for this rise in prevalence is not fully understood, but this increase clearly shows that ASDs are a significant public health issue.

ASDs are often accompanied by significant lifelong impairments. Individuals with ASDs often require intensive parental, school, and other social support. In addition to intensive behavioral therapies, services at school ranging from individualized education plans with combinations of speech therapy, occupational therapy, social skills training and physical therapy, to individualized aides, specialized classrooms, and sometimes even specialized schools are required for most children with ASDs. While services vary by region and can be challenging to obtain for children, supports for adults are even more limited. For many adults with ASDs and significant intellectual disability, psychiatric comorbidity, and/or medical comorbidity, independence is often not achievable. Some may be able to participate in vocational training and hold basic jobs but still require assistance with daily living. Many adults live either with parents, with other family members, or in group homes or residential facilities. Most individuals with typical autism require lifelong assistance with basic living skills, as well as financial, medical, social, and psychiatric support.

In individuals with ASD and without significant intellectual disability, high-functioning autism is often diagnosed. In those without language development delays, a diagnosis of Asperger disorder is made. These individuals tend to have relatively good outcomes and are able to live independently, although challenges still exist. High-functioning individuals with normal, above-average, or even superior intelligence may be able to attend college and graduate school, but they often struggle significantly with social demands. These demands may be too challenging for many individuals with ASDs to complete higher education without substantial psychosocial supports. Similarly, high-functioning adults with ASDs may be intellectually quite capable of performing in a wide variety of jobs and pursuing careers, but navigating the unwritten social maps may be too complex and significantly limit their abilities to obtain and hold positions in the workforce. These individuals are not only at greater risk for social isolation, but they may also be at greater risk for depression and anxiety as they recognize their impairments and have enough insight to be all too aware of their differences.

As with other disorders in psychiatry, psychoeducation is critically important for patients and their families. This article aims to provide the general psychiatric community with an update on the major findings on the biology of ASDs as well as the advances in diagnostic and interventional strategies.

The biology of ASDs has been investigated by state-of-the-art scientific methodologies in genetics, molecular biology, neurophysiology, neuroimaging, and neuropsychology. A consistent theme has been the heterogeneity of biological findings in the disorder. For example, to date, more than 100 disease genes and 44 genomic loci are reported in persons with ASD or autistic behavior.

 It is unlikely for a single anatomic abnormality or even a single physiologic process to explain the etiology of ASDs. Rather, it is more reasonable to anticipate discoveries of risk factors that when joined in various combinations result in ASDs. Nonetheless, careful investigations focusing on individual genes and protein products, anatomical structures and neurocircuits, endophenotypes, and the environment are important in identifying such risk factors. With this in mind, we focus on 3 categories of risk factors for ASDs: genetics, brain anatomy, and environment.

. One of the largest advancements in understanding the etiology of ASDs is the identification of a variety of genetic variations, and especially copy number variants, associated with ASDs. Copy number variants are segments of DNA that have been either deleted or duplicated from a person’s genome. Thus, while genes are typically inherited in pairs (one copy from the mother and one copy from the father), persons with copy number variants that consist of a deletion will only have a single copy of a gene(s), and those with a copy number variant consisting of a duplication will actually have 3 copies of the gene(s). Both deletions and duplications have been implicated in ASDs.

Copy number variants vary in size and may include several genes or only a few. They may be inherited, or they may arise spontaneously, in which case they are referred to as “de novo” mutations. Children with ASDs have been found to have higher rates of de novo copy number variants compared with siblings without ASD diagnoses.

 
 Selected copy number variants associated with autism spectrum disorders 

 lists some of the copy number variants that have been associated with ASDs and the respective genes thought to be involved. Several of these genes are involved with synaptic transmission and modulation of excitatory and inhibitory neurons. Of note, these variants are not unique to ASDs, and each may lead to diverse impairments, including other physical, cognitive, and/or psychiatric disorders. Similarly, the presence of a variant does not guarantee the development of ASDs. In addition to copy number variants, de novo single nucleotide variants have also been linked to autism through recent whole-exome sequencing studies and may help elucidate the protein networks involved in ASDs.

Many genes associated with ASDs are involved in synaptic transmission.

 When the molecular and cellular machineries involved with neurotransmission become dysfunctional, neurological, psychiatric, and behavioral symptoms may occur. One proposed model for ASD suggests that the condition is a result of an imbalance of excitatory (E) and inhibitory (I) neurotransmission.

Glutamate and γ-aminobutyric acid (GABA) are the major excitatory and inhibitory neurotransmitters, respectively. Using optogenetic techniques, Yizhar and colleagues

 demonstrated that elevation, but not reduction, of cellular E/I balance within the mouse medial prefrontal cortex caused profound impairment in cellular information processing, associated with behavioral impairments resembling social withdrawal. Furthermore, compensatory elevation of inhibitory cell excitability partially rescued social deficits caused by E/I balance elevation in these mice.

Consistent with the E/I hypothesis, another study on a key ASD-associated gene, 

 has shown reduced number of GABAergic interneurons in 

 These results add to the accumulating evidence supporting the hypothesis of an elevated E:I ratio in ASD and other comorbid neuropsychiatric disease–related symptoms, such as seizures.

. To date, E/I imbalance in the human brain is difficult to demonstrate directly. However, numerous neuroimaging studies have been conducted to define the ASD brain and elucidate how anatomic regions are connected functionally and structurally. Rapid advances in neuroimaging technology have been helpful for neuroscience in general, but they also present challenges in establishing consistent phenomenology regarding neuroanatomic changes in ASD.

Other challenges include the heterogeneity of the ASD population and the methodological limitations related to the deviation in developmental trajectory of the brain. One consistent finding is an increased rate of brain growth in individuals with ASDs that begins during the first year of life but slows throughout later childhood and/or adolescence.

 Longitudinal studies provide a better understanding of the underlying neurodevelopmental trajectories.

In the first longitudinal structural MRI studies of young children (2 to 5 years) with autistic disorder, the frontal, temporal, and cingulate cortices as well as the amygdala were found to be disproportionately larger than those in typically developing children at baseline.

 Repeated scans in the following 3 years showed that almost all regions in the gray and white matter in children with ASDs grew almost linearly instead of along a curve and lacked the more apparent deceleration seen in neurotypical children.

 Similarly, the amygdala’s growth in children with ASDs a year later showed an increased rate, compared with controls.

 The increased growth rate in the cortices appears to dissipate at a later age (8 to 12 years).

 In addition to longitudinal studies, functional MRI, diffusion tensor imaging, pharmacological MRI, and imaging-genetics continue to advance our understanding of ASDs.

. ASDs were once thought to be largely genetic; heritability was previously estimated to be as high as 90%. However, a twin study indicates that genetic components account for only approximately 30% to 35% of the factors involved in the etiology of ASDs. In other words, environmental factors likely account for 60% to 65%.

 Several environmental factors have been hypothesized to be involved in the development of ASDs. However, it remains largely unknown what the specific environmental factors are, their impact, and their contributing mechanics to the development of ASDs.

Environmental factors associated with ASDs include perinatal events, such as abnormal presentation at birth (ie, breech presentation), umbilical cord complications, fetal distress, birth injury or trauma, multiple birth (ie, twin or triplet birth), bleeding during pregnancy, low birth weight, small for gestational age, birth defects, low Apgar scores, feeding difficulties, meconium aspiration, neonatal anemia, ABO or Rh incompatibility, and jaundice.

In addition, ASDs have been linked to maternal and, notably, paternal age (often defined as older than 35 or 40 or younger than 30), parity, preeclampsia, scheduled cesarean, and prematurity.

 The maternal immune system during pregnancy may also be involved in the development of ASDs as season of birth, maternal infection, and the presence of maternal antibodies/cytokines have also been associated with ASDs.

 Interestingly, E/I imbalance may be linked to some of the environmental factors potentially causing physiological abnormalities, such as oxidative stress

Recent advances in the diagnostic paradigms of ASD

. In an attempt to enhance diagnostic reliability and validity, the DSM-5 task force and work groups have proposed various changes in the nosology of autistic disorders and related disorders in DSM-5. First, “disorders usually first diagnosed in infancy, childhood, or adolescence” in DSM-IV will be renamed “neurodevelopmental disorders.”

Second, the previously discrete diagnoses under pervasive developmental disorders (including autistic disorder; Asperger disorder; childhood disintegrative disorder; and pervasive developmental disorder, not otherwise specified) will be consolidated into a single diagnostic category of ASD. This change is supported by a recent study demonstrating that clinical distinctions among categorical diagnostic subtypes of ASDs were not reliable across sites with well-documented fidelity using standardized diagnostic instruments.

A third change is that DSM-IV describes 3 defining dimensions of behavior:

• 	Presence of restricted, repetitive behaviors and interests

 
 Comparing nosology for autism spectrum disorders in DSM-IV and DSM-5ed 

, the ASD definition in DSM-5 contains only 2 domains: social communication/social interaction and restricted, repetitive behaviors and interests (RRBs).

Fourth, “play and imagination” and “stereotyped and repetitive use of language” are no longer in DSM-5. However, sensory abnormalities are now taken into account in DSM-5 under RRBs. Lastly, the age of onset of symptoms (before age 3) required in DSM-IV is replaced by a more flexible criterion of “early childhood” in DSM-5.

 
 Symptom domains used in defining autism spectrum disorders in DSM-IV and DSM-5 

These changes have received many reactions from the media, families, and clinical and scientific communities. Using data archived from a field trials study for DSM-IV, a recent study reported that the proposed DSM-5 criteria could substantially alter the composition of the autism spectrum and exclude a significant portion of high-functioning individuals and those with ASD other than autistic disorder.

 The DSM-5 work group for neurodevelopmental disorders pointed out that this study used data from DSM-IV field trials and therefore the findings could not be generalized to apply to DSM-5.

 Data from phase 1 field trials of DSM-5 suggest that the criteria adequately captured those with clinical and subthreshold autistic presentations and that the diagnostic fit was stable across age and sex.

 Another study found that relaxing the DSM-5 criteria by requiring one less symptom criterion in RRB increased sensitivity with minimal reduction in specificity.

Whether the proposed changes will affect both diagnosis and services covered by insurance remains to be seen. The question of whether services should be based on specific diagnoses as opposed to clinical needs is raised. Other proposed changes in the nosology of neurodevelopmental disorders in DSM-5 are summarized in 

 
 Neurodevelopmental disorders in DSM-IV and DSM-5 

Diagnostic assessment and genetic testing

. Diagnosing ASDs using clinical criteria is only part of a full diagnostic assessment. ASDs may be diagnosed in conjunction with a neurodevelopmental syndrome or independently as idiopathic autism. The presence of a neurodevelopmental or genetic syndrome may prompt clinical testing for autism, and similarly, the diagnosis of an ASD may prompt genetic testing for an underlying genetic syndrome.

In 2010, the International Standard Cytogenomic Array Consortium published a consensus article recommending array comparative genomic hybridization (aCGH), also known as chromosomal microarray, as the first-line tier of testing for any individual with an ASD.

 This recommendation was supported shortly after by a paper published by the Autism Consortium.

 Currently, studies suggest that genetic testing with aCGH can uncover genetic variations in up to 20% of cases of autism.

Variations are more likely to be found in those with intellectual delay, physical dysmorphisms, and other congenital anomalies or medical disorders. It should be noted that aCGH alone will not identify all genetic variations associated with ASDs. For example, testing for fragile X syndrome, sequencing for certain genes, metabolic testing for metabolic disorders, and other laboratory tests may be considered in conjunction with aCGH.

. Because early behavioral interventions are a key aspect of treatment in ASDs, early diagnosis is an important area of research. Developing screening tools and tests to improve early detection is an ongoing area of investigation in the field. Currently, the American Academy of Pediatrics recommends screening all toddlers at 18 and 24 months.

A brief parent questionnaire, the Communication and Symbolic Behavior Scales Developmental Profile Infant-Toddler Checklist, accurately identifies autism or other developmental delays in approximately 75% of 1-year-olds at their 1-year pediatric checkup.

 Other investigators are seeking to identify unique patterns to identify ASDs using biological tests such as MRI and electroencephalography.

 Questionnaires or screens such as these may improve the rate of identification and referral and lead to earlier treatments.

. Behavioral modalities are first-line treatment interventions for ASDs. They improve language skills, cognitive abilities, adaptive behaviors, and social skills, and reduce aggression and anxiety.

 The earlier the initiation of behavioral interventions, the better. Increased brain plasticity in younger children may maximize treatment benefits; early altering of developmental trajectories may best improve outcomes and decrease the likelihood of severe problematic behaviors.

The 2009 National Standards Report reviewed a wide range of nonpharmacological treatments for autism.

 Of 38 treatments investigated, 11 were found to have enough supporting evidence to qualify as “established” treatments substantiating favorable outcomes. These established treatments were all behavioral interventions, and several fell under the umbrella category of applied behavior analytic (ABA) therapy. ABA therapy is based on the science of understanding the laws by which environmental events influence and change behavior.

Discrete trial training (DTT) is perhaps the most common type of ABA therapy. It involves breaking down complex skills and teaching each subskill through a series of highly structured, massed teaching trials. Each trial consists of a precise and consistent instruction designed to elicit a specific response. Often, the sought-after response is an imitation of the therapist’s model or compliance with a verbal request. The response is shaped and reinforced with rewards contingent on the child’s production of the target response. One criticism of DTT is that because of the highly structured trials through which desired responses are learned, it may be difficult for individuals with ASDs to generalize the responses to the natural environment. In response, some therapists recommend other ABA techniques endorsed by the National Standards Project, including pivotal response training (PRT).

Although PRT is also based on a system of contingency rewards, it aims to provide interventions in the natural environment with the goal of shaping behavioral improvements that may be generalized across a variety of settings. Social skills groups and training (which may be employed using a wide variety of strategies, including PRT as well as peer mentoring, social stories, modeling, social problem solving, scripting procedures, computer-based interventions, self-monitoring, and others) are also widely used and accepted with varying levels of supporting research.

Unfortunately, in contrast to the behavioral interventions mentioned earlier, there remain a significant number of treatments advertised to parents that have little peer-reviewed evidence to support their efficacy. These include hyperbaric oxygen treatment, chelation therapy, and stem cell therapies offered outside the United States. These treatments cannot only be extremely expensive, but they may also carry significant medical risks that parents and care providers may not always be informed of before initiating treatment.

. Drugs may also play a role in treating individuals with ASDs and overlapping symptoms of other psychiatric disorders. For example, children with ASDs may have significant symptoms that overlap with ADHD, anxiety, obsessive-compulsive disorder, and/or mood disorders. Psychopharmacologic agents aimed at treating these conditions may also be used for individuals with ASDs.

 Among these agents, risperidone and aripiprazole are the only two FDA-approved for the treatment of ASD in children. Specifically, these two drugs are approved for the treatment of irritability and associated symptoms, such as aggression, self-injurious behaviors, and temper tantrums. However, serious adverse effects (weight gain, metabolic abnormalities, and tardive dyskinesia) are associated with these medications, so they should be used with caution. Furthermore, persons with ASDs or other developmental disabilities may be more sensitive to adverse effects of medications and may require slower titration schedules with lower target doses.

In the area of new and emerging treatments, 

-acetylcysteine (NAC) is a safe, orally bioavailable prodrug of cysteine that is known for its role as an antidote against acetaminophen overdose.

 Cysteine supplied by NAC treatment can also be oxidized to cystine, a substrate for the glutamate-cystine antiporter that causes the reverse transport of nonvesicular glutamate into the extracellular space. This process ultimately stimulates the type 2/3 metabotropic glutamate receptors, inhibiting the vesicular release of glutamate and resulting in a decrease in glutamatergic neurotransmission and a reduced E:I ratio.

The results from a pilot study of NAC (900 mg daily for 4 weeks, then 900 mg twice daily for 4 weeks, and finally 900 mg 3 times daily for 4 weeks) showed significant improvements on irritability and associated symptoms in children with ASD.

 On the basis of these data, a larger study will be conducted in hopes of replicating this result while examining the effect of NAC on glutamatergic transmission and glutathione metabolism.

While medications currently available for ASD do not treat the core symptoms of social and communication deficits or restricted interests and repetitive behaviors, ongoing research is aimed toward discovering such therapeutics. Oxytocin is an endogenous hormone that may increase the saliency of social stimuli and link the encoding of these stimuli to social reward and reinforcement.

 Intravenous oxytocin may reduce repetitive behaviors and increase retention of social cognition in patients with ASD.

 However, the intravenous route is not clinically attractive, and oxytocin’s poor blood-brain barrier penetration has limited its use. Researchers are attempting to circumvent this problem by administering the compound intranasally. It is hypothesized that intranasal administration of peptides allows for passage through clefts in the nasal epithelium into the cerebrospinal fluid.

 In a double-blind, randomized, placebo-controlled, crossover design, a single dose of oxytocin nasal spray was shown to improve recognition of others’ emotions by participants with ASD.

 In another study, participants with ASD given a single dose of intranasal oxytocin responded more strongly to others and exhibited more appropriate social behavior and affect.

 A single dose of intranasal oxytocin also significantly improved eye gaze frequency in a randomized, double-blind, placebo-controlled trial in teenagers with fragile X syndrome, a genetic syndrome known to be associated with an elevated prevalence (20% to 60%) of autistic symptoms.

With recent advancements in the understanding of cellular mechanisms and biological pathways involved in specific genetic syndromes associated with ASDs, novel therapeutics are also being developed and targeted for specific genetic disorders. For example, the affected gene in fragile X syndrome codes for a negative regulator of the metabotropic glutamate receptor mGluR5. Thus, disturbance of the gene leads to enhanced excitatory activity at this receptor. mGluR5 antagonists such as AFQ056 are showing promising results in preliminary clinical trials of individuals with full mutations of 

Articles

 This article aims to provide the general psychiatric community with an update on the major findings on the biology of ASDs as well as the advances in diagnostic and interventional strategies.

Wendy Froehlich, MD

Articles

Autism Spectrum and Neurodevelopmental Disorders