What Is New in This Research Area, and What Have We Learned in the Past Two Years?

While many treatments and interventions for ASD are still in the early stages of development, several new findings on potentially efficacious interventions for children and adults with ASD have arisen during the past two years.

Early Behavioral Interventions

Evidence for the benefits of early behavioral intervention continues to mount, with researchers now focusing on testing interventions for infants and toddlers, identifying the most effective aspects of treatments, and disseminating these interventions in community settings. While gains have been made in this area of research, the effects of these interventions as measured to date are modest. Positive results of implementation of targeted interventions in community settings have recently been reported (Kaale, Smith & Sponheim, 2012; Lawton & Kasari, 2012).

The past two years have seen the first randomized controlled trials (RCTs) of parent-mediated interventions, an important milestone. However, these interventions have not yet been found to be efficacious (e.g., Carter et al., 2011). Progress has been made in developing social interventions. For example, a longitudinal followup of a targeted, joint attention intervention found that joint attention and play are important targets for interventions aimed at enhancing language acquisition and that these functional gains may persist over the long term (Kasari et al., 2012a).

A study of early intensive intervention using the Early Start Denver Model (ESDM) with toddlers, a method that was studied in an earlier RCT described in the 2011 Strategic Plan, was found to result in changes in electrophysiological brain activity. This biological marker correlated with positive changes in behavior (Dawson et al., 2012).

In a different RCT, children who received a 12-week, parent-delivered ESDM intervention were compared to a control group of children receiving typical community interventions (Rogers et al., 2012). Both groups of children showed developmental gains and reduced core autism symptoms, although there was no clear advantage of one intervention over the other. However, the degree of improvement across both community and ESDM groups was higher in children that received more hours of intervention, and younger children (14 months) made more developmental gains than older children (24 months).

Behavioral and Psychosocial Interventions for School-Age Children and Adults

There is a general paucity of research on interventions for adolescents and adults, as underscored by a recent systematic review (Taylor et al., 2012). This review noted that evidence-based approaches to support transition to adulthood and employment are particularly lacking.

Although the evidence base for behavioral interventions in older children and adults is less well-developed than early interventions, new research points to a number of promising efficacious interventions. An RCT of a brief social skills intervention documented improvements in peer relationships in the classroom that persisted over time (Kasari et al., 2012b). In another pair of studies that built on earlier findings, cognitive behavioral therapy (CBT) and social skills training were useful for decreasing anxiety in some children with mild ASD, and many children who were re-evaluated after six months showed sustained improvements (Antshel et al., 2011; Reaven et al., 2012). Social skills training may also have positive effects on core social symptoms, as suggested by a recent study of a social skills group intervention in children with ASD (DeRosier et al., 2011). Additionally, one RCT that addressed common psychiatric concerns in ASD found that mindfulness-based therapy was efficacious for reducing anxiety, depression, and rumination in adults (Spek, van Ham & Nyklí?ek, 2012).

Medications

To date, the only medications approved by the Food and Drug Administration (FDA) for the treatment of any aspect of ASD are the antipsychotic drugs aripiprazole and risperidone, both of which are used for the treatment of irritability associated with ASD. A recent meta-analysis (a study that contrasts and combines results from different studies to identify patterns among the results) of these medications for adolescents and children with ASD confirmed that, while these drugs are effective in the treatment of behavioral disturbances, there are frequently adverse side effects including sedation, involuntary muscle spasms, and weight gain (Cohen et al., 2013). In a study combining treatment and intervention, the combined effects of risperidone and parent training were more positive than medication alone for improving adaptive behavior (Scahill et al., 2012b). Another group of researchers reported potential usefulness of N-acetylcysteine, a glutamate modulator, for treating irritability (Hardan et al., 2012).

Clinical trials for selective serotonin reuptake inhibitors (SSRIs) for reducing repetitive behaviors have produced conflicting results for adults and children, indicating that there may be age-associated drug effects or effects intrinsic to this particular class of drugs (Hollander et al., 2012; Scahill et al., 2012a). No uniform guidance has yet emerged from these studies.

At least twelve medication trials have been launched for the core domains of ASD or neurodevelopmental disorders associated with ASD, such as fragile X syndrome. A Phase 2 RCT of Arbaclofen, a selective GABA-B agonist, with children and adults with fragile X showed positive effects for reducing social avoidance (Berry-Kravis et al., 2012). In addition, at least 10 trials of the pro-social neuropeptide oxytocin are underway or were recently completed in children and adolescents with ASD (ClinicalTrials.gov website).

Treatment of co-occurring medical conditions continues to be an important area of study. A Pediatrics supplement (Perrin & Coury, ed., 2012) based on the work of the U.S. Health Resources and Services Administration (HRSA) Autism Intervention Research Network on Physical Health (AIR-P) and the Autism Speaks Autism Treatment Network (ATN) provided empirically-based physician guidelines for the management of gastrointestinal (GI) issues, sleep, and attention deficit hyperactivity disorder (ADHD), as well as descriptive information on the prevalence and nature of a wide range of co-occurring medical conditions. Studies showed that melatonin, a hormone produced by the brain to regulate sleep, is useful for treating insomnia in ASD (Malow et al., 2012) and that controlled-release melatonin with CBT may be useful for treating night awakening (Cortesi et al., 2012). An open label trial of donepezil, a drug that enhances the function of the neurotransmitter acetylcholine in the brain, was found to increase rapid eye movement (REM) sleep and decrease REM latency in children with ASD (Buckley et al., 2011).

Epilepsy—which commonly co-occurs with ASD—and interictal epileptiform discharges (patterns of electrical activity that resemble those during seizures but are present between seizures) are also associated with sleep disruption. Recently, the antiepileptic drug levetiracetam was shown to reduce nocturnal epileptiform activity (which is present in many individuals that lack a clinical diagnosis of epilepsy) during non-REM sleep (Larsson et al., 2012).

In the broader context of biomedical research, a 2011 National Academy of Sciences report on precision medicine is changing the culture of treatment development (National Research Council, 2011). Noting that most common diseases include cases caused by rare syndromes requiring different treatments, precision medicine argues for the development of targeted treatments based on more precise diagnostics, defined by biomarkers. Thus, biomarkers may play an important role not only in the development of diagnostic tools, but also in advancing the autism treatment field.

Other Treatments and Interventions

A number of other treatments and interventions for ASD have been studied during the past two years. For example, a meta-analysis concluded that exercise is beneficial for social skills and motor performance for people with ASD, with individual interventions more effective than group interventions (Sowa & Meulenbroek, 2012 ). Additionally, an RCT found that a movement-based yoga, dance, and music therapy program was effective for improving behavior (Rosenblatt et al., 2011). Finally, a 12-week RCT of repetitive transcranial magnetic stimulation (rTMS) in mildly affected individuals with ASD resulted in improved error-related negativity (a proxy measure of executive functioning) and improved error monitoring and correction (Sokhadze et al., 2012).

What Gaps Have Emerged in the Past Two Years?

Improved Trial Design and Treatment Biomarkers

Although some helpful interventions for ASD have been identified, a precision medicine approach is lacking. Specifically, no biomarkers or clinical features have been identified to indicate which interventions are most likely to be helpful for a given individual. Trial designs need to be improved by: including larger, more diverse samples (e.g., greater ethnic diversity and additional individuals with more severe disability challenges, such as individuals with limited or no verbal communication); examining the length and intensity of interventions; developing objective/physiologic measures of autonomic response or physical activity; and testing measures of how well new skills are maintained and generalized in real-world settings. More sensitive outcome measures and biomarkers need to be developed so that clinical trials may identify mediators and moderators of treatment response. Thus, better characterization of biomarkers andendophenotypes (observable characteristics that are thought to be strongly genetic in origin) may aid in the development of customized, targeted interventions.

Investigators are also beginning to actively link genetic makeup with drug effects—both therapeutic as well as side effects, such as weight gain with risperidone (Adkins et al., 2011; Correia et al., 2010). This area of research has been termed "pharmacogenomics," and its goal within the context of ASD is to both define biomarkers of drug responsiveness among people with ASD as well as to identify those who may be most prone to adverse effects.

Co-occurring Conditions

Understanding the pathophysiology of ASD will be paramount in the quest for new treatments and delineating the mechanisms behind treatment response. The effect of co-occurring conditions on ASD, and vice versa, is unclear and needs to be better defined. Research needs to be conducted to determine if co-occurring conditions such as anxiety, depression, gastrointestinal disturbances, epilepsy, or atypical immune response involve similar mechanisms in individuals with and without ASD. It is possible that anxiety and/or affective disorders in individuals with ASD stem from core aspects of atypical neurodevelopment in ASD; alternatively, these co-occurring conditions may actually represent reactive compensatory mechanisms in those with ASD. It is also important to understand whether treatments for co-occurring conditions have similar effects in individuals with and without ASD.

Alternatives to pharmacological treatments should also continue to be explored. Likewise, as the effects of behavioral interventions become more apparent, better information regarding the most critical components of treatment is needed. The effectiveness and longevity of treatment effects in real-world settings must also be established.

Outcome Measures

Additionally, outcome measures that can monitor changes in brain connectivity and activity and correlate those changes with behavioral and social therapies should be developed. For example, more research could be conducted using magnetic resonance imaging (MRI) or electroencephalography (EEG) as a tool to measure the physiological effects of a variety of treatments, including behavioral, pharmacological, and rTMS. These measures could then be compared to concurrent changes in behavior. This approach has already been successfully used to study the effect of behavioral interventions using EEG (see study described above, Dawson et al., 2012) and functional MRI (Voos et al., 2013). Likewise, standardized measures of sensory processing should be established to measure changes in sensory processing in response to interventions, such as occupational therapy-based interventions.

Relating Intervention Success to Clinical Phenotype

Clinically, determination of which individuals are most likely to be responsive to particular social and behavioral interventions is needed, as well as efforts to better understand the mechanisms behind efficacy in those individuals. In addition, there is a need for further research on the many cognitive, educational, and computer-based programs used to remediate learning deficits in individuals with ASD to determine which are most effective. These should be compared in well-characterized ASD cohorts. Better understanding of which interventions are likely to be effective for different individuals will help individuals to select the interventions that will best help them address their needs and meet their cognitive and learning potential. Effectiveness studies for relatively inexpensive community-based interventions (e.g., exercise, yoga, acupuncture, mindfulness) should also be conducted.

Other Concerns

Some interventions are widely used despite a limited evidence base. These include complementary and alternative treatment approaches, among others. There is a need for more research on these interventions so that consumers can make informed choices about their use. The needed research includes studies to identify potential principles and mechanisms of action and to evaluate safety and efficacy. In addition, research that explores the potential treatment implications of novel biological findings, such as differences in immune functioning in individuals with autism, is needed.

While many interventions (pharmacological, behavioral, and otherwise) successfully change targeted behaviors or symptoms, there is a need to measure the global, and not merely the specific, effects of all interventions. These effects may include the unintentional development of alternate problem behaviors, as well as effects on stress levels, unique talents, and overall life satisfaction, including satisfaction with social and peer groups. When possible, studies should include self-report measures from people with autism who serve as study participants. For example, data from self-report measures might help to understand if the benefits of interventions targeting "harmless" autistic behaviors (e.g., "stimming") outweigh the risk of impacts such as provoking anxiety or confusion. A worthy goal for any intervention is to help individuals with ASD understand and utilize their strengths.

All research involving human participants must be conducted in accordance with high ethical standards, minimizing harms and risks and maximizing benefits; respecting human dignity, privacy, and autonomy; taking special precautions with vulnerable populations; and striving to distribute the benefits and burdens of research fairly. The described research will play an important role in distinguishing effective interventions from ineffective ones and in identifying risks and benefits that should be taken into consideration before using these interventions.

References