There has been enormous growth over the past 10-15 years in research attempting to identify effective medications for children and youth with ASD. A few classes of medications have been shown to be effective for treating specific symptoms associated with autism. There is accumulating data to support the use of stimulants and non-stimulants for hyperactivity and attention deficits, and of atypical antipsychotics (risperidone, aripiprazole) for irritability and impulsive aggression. There is disappointing data regarding the use of SSRIs (citalopram, fluoxetine) for repetitive behaviours but several issues related to possible subgrouping may not have been addressed adequately. However, the work done to date is based on the premise that if a symptom domain is shared by multiple neurodevelopmental disorders, then its biology must be common across such disorders. As such we “borrowed” medications from other such disorders (e.g. ADHD, OCD) and tested them in autism. Although the results have been sometimes fruitful, the predictability and degree of clinical benefit has been lower among those with autism. In addition, this approach has not adequately addressed either impairments in core social and communication development or in key learning and cognitive domains essential for skill development. Previous studies have focused primarily on behavioural symptoms shared by other disorders, rather than on weaknesses in cognitive and motor domains of functioning (e.g., fine and gross motor impairments, poor motor coordination, memory weaknesses, sensory issues).
As our understanding of the molecular genetics and basic neurosciences of ASD grows, the field is considering new treatment approaches. We are for the first time in a position to identify novel molecular targets based on molecular genetics, neuropathology and animal model work. This cutting edge research facilitates the translation of basic science findings into treatments that target core ASD symptoms and potentially facilitate learning. There are currently several brain pathways and neurochemical mediators identified as potentially important targets for treatment in ASD. In this article, we focus on the glutamate signalling system, oxytocin and immune function, although other models arising from single gene disorders such as tuberous sclerosis or Shank3 mutations are also of great interest.
The Glutamate Signalling System
The glutamate system has recently become the focus of potentially translational work. Research in the Fragile X model of autism has demonstrated that manipulating aspects of this system can produce autism-like features. In addition to the mGluR5 pathway, there are other possible molecular targets within the glutamate signalling system that may be modified by medication. Several different avenues of research have found impairments in the glutamate system in people with ASD. Studies of glutamate levels in blood suggest that individuals with ASD have increased levels of glutamate compared to typically developing controls (Rolf et al 1993, Moreno-Fuenmayor et al 1996, Aldred et al 2003, Shinohe et al 2006). Multiple pathology studies have documented aberrations in enzymes and receptor density important in glutamate related systems (Fatemi at al 2002, Purcell et al 2001). Emerging genetic findings have either implicated glutamate related genes or genes involved in the structure and function of synapses important for glutamate signalling (Phillipe et al, 1999, Serajee et al, 2003, Barnaby et al 2005, ADD).
A few medications targeting glutamate pathways have preliminary data to support the notion that manipulation of glutamate/GABA related systems offer promise for novel therapeutics. Both a new GABA agonist and mGluR5 inhibitors are currently in trials. Medications with effects relevant to the NMDA receptor (a receptor involved in the glutamate signalling system) also suggest promise (e.g. Amantadine, King et al 2001; dextromethorphan, Woodard et al, 2005, Welch et al 1992, Phillips et al 1999). Of particular interest is memantine, a non-competitive NMDA inhibitor. In a series of four open label studies involving 186 children with autism, memantine was suggested to improve irritability, social withdrawal, hyperactivity, inappropriate speech, stereotypy, memory, and language (Owley et al 2006, Chez et al 2007, Niederhofer et al 2007, Erickson et al 2007). Current studies of memantine include a large randomized trial looking at core symptom domains sponsored by Forest Labs and a trial targeting motor functioning, expressive language, and memory sponsored by Autism Speaks.
There is also accumulating data to suggest that oxytocin related systems should be explored for potential therapeutics in ASD. This and related peptides have been documented to be involved in aspects of social perception and cognition as well as the development of repetitive behaviors in animal models. Insel et al. suggested that abnormalities in the neural pathway for oxytocin could explain some deficits shared by autism, such as the early onset, high prevalence in males, genetic loading, and neuroanatomical differences. Although the data implicating oxytocin in the pathophysiology of ASD is still limited, early work suggests lower blood level of oxytocin in children with ASD compared to typically developing peers (Modahl et al 1998), and aberrant maturation of this system (Green et al 2001). Recent genetic studies suggest a genetic association between the OXT receptor gene and autism (Wu et al., 2005; Ylisaukko-oja et al., 2006; Jacob et al., 2007) and intravenous administration of oxytocin may facilitate aspects of social perception in ASD (Hollander et al 2007) in ASD. Single dose studies of intranasal oxytocin in both typically developed adult volunteers and youth with ASD have also supported the role of oxytocin in social perception/cognition and randomized trials of this compound are currently underway.
Early neuropathology studies have implicated neuroinflammatory processes in the brain of children with ASD. (Vargas et al 2005). Activation of neuroglia cells can produce brain changes consistent with those seen in autism. Reported findings (Vargas et al., 2005; Pardo et al., 2005; Zimmerman et al., 2006) suggest a dysregulated immune response. Many studies have documented either elevated levels of proinflammatory cytokines such as TNF-alpha/IL-12, tumor necrosis factor receptor II, interferon – gamma among others (Jyonouchi et al, 2005a, 2005b, 2001(Croonenberghs et al., 2002a), Croonenberghs et al., 2002b), or lower levels of IL-10 which is a counter regulatory molecule. There is also evidence of immunogenetic differences (Pardo et al 2005, Torres et al 2006 and Ashwood et al 2006) and a family history of autoimmune disorders in individuals with ASD than in the general population (Sweeten et al., 2003). There are currently multiple compounds that have the potential to modulate the central nervous system immune system in favourable ways; however, their side effect profiles vary. It is of note that nutritional supplements may have a role to play. For example, omega 3 fatty acids at high enough doses have been noted to decrease microglia activation in the brain (ADD). Still the need for large randomized trials is urgent.
The advancement of research practices, from borrowing medications from other conditions, to a focus on translating findings from molecular genetics and neuroscience into treatment may finally give us therapeutic options for treating core symptom domains and facilitating learning. The great advantage of such an approach is the potential of these medication treatments to facilitate progress in psychoeducational and behavioural interventions. To date, behavioural interventions are the mainstay of intervention programs for children with ASDs, but are faced with the problem of variable treatment responsiveness and high cost structures. Translational treatment options such as oxytocin, memantine and other glutamate/GABA modifying agents as well as potentially immunomodulators have the potential to modify neurobiological impairments that may impede progress in behavioural interventions, and thus, may facilitate skill acquisition. For example, drugs that manipulate Glutamate/GABA to facilitate memory and learning may be used in children making slow or stagnant progress in behavioural interventions. Oxytocin has also been suggested as a potential treatment for improving the social attention and motivation to potentially facilitate response to socialization interventions. Both early intensive behavioral interventions and computerized technologies targeting cognitive skill acquisition are potentially great models for such an approach as such interventions facilitate the formation of new connections. The hope is that such medications may improve the neural substrate for skill acquisition in ASD.
In summary, recent findings from neuroscience and molecular genetics are giving us the opportunity to start translating such data to novel therapeutics. In addition, one should not underestimate the potential that clinical trials have to inform our understanding of the disorder. For example, a trial of a compound manipulating the glutamate system has the potential to also inform our understanding of what is different in this system in ASD. Thus, it is now imperative that these novel treatments are tested in clinical trials and that the autism community supports such initiatives to help advance our knowledge of neurobiological impairments and translate this knowledge into novel effective treatments.
Evdokia Anagnostou, MD, is a Clinician Scientist at the Bloorview Research Institute University of Toronto. Joel Bregman, MD is Executive Director of the Thompson Center for Autism and Neurodevelopmental Disorders at the University of Missouri, and is a Professor at the Department of Psychiatry/Department of Child Health at the University of Missouri-Columbia. Latha V. Soorya, PhD, BCBA, is an Assistant Professor at the Mount Sinai School of Medicine, and is Chief Psychologist at the Seaver Autism Center.
Additional collaborating authors for this article include: David Q. Beversdorf, MD, Associate Professor, Department of Radiology and Neurology, School of Medicine, University of Missouri-Columbia; Elena H. Drewel, Clinical Assistant Professor, Department of Health Psychology, University of Missouri-Columbia; David Grodberg, MD, Assistant Professor, Department of Psychiatry, Mount Sinai School of Medicine; Raghuram Prasad, Assistant Professor, Clinical Psychiatry, School of Medicine, University of Missouri-Columbia; Ellen Drum, BA, Clinical Research Assistant, Bloorview Research Institute, University of Toronto; Holly E. Rice, MEd, Project Development Specialist, Thompson Center for Autism and Neurodevelopmental Disorders, University of Missouri-Columbia; and Lily Schwartz, BA, Clinical Research Coordinator, Mount Sinai School of Medicine.