|  The
pharmacotherapy of autistic disorder (autism) and other pervasive developmental
disorders (PDDs) currently involves the treatment of target symptoms including
hyperactivity, inattention, interfering repetitive thoughts and behavior,
and aggression toward self, others, or the environment. Although studies
in laboratory animals have identified neuropeptides, including oxytocin
and vasopressin, that mediate some elements of affiliative behavior, drugs
that have consistent, primary effects on the core social and communication
disability of autism have not yet been developed.
Elevated
levels of whole blood serotonin (5-HT) have long been known to be associated
with autism. Following reports that fenfluramine, an indirect 5-HT agonist,
decreased blood and brain 5-HT in animals, this drug received considerable
attention and some systematic investigation. Despite early enthusiasm,
most controlled studies found no consistent efficacy for fenfluramine.
Furthermore, the association of fenfluramine with primary pulmonary hypertension
and (in combination with phentermine) valvular heart disease has eliminated
its use as a safe agent.
Most
of the typical antipsychotic drugs have been studied in heterogeneous
groups of children that included autistic subjects. These trials were
often direct comparisons of two low-potency drugs and did not include
a placebo control. A number of these agents were found to be effective
for reducing hyperactivity, agitation, and stereotypies. Because of significant
sedation and adverse cognitive effects, however, studies of higher-potency
conventional antipsychotics were next pursued.
Several
well-designed controlled studies of haloperidol were conducted. In doses
of 1 to 2 mg/day, haloperidol was found to be more efficacious than placebo
for withdrawal, stereotypy, hyperactivity, affective lability, anger,
and temper outbursts. However, acute dystonic reactions along with withdrawal
and tardive dyskinesias were not infrequent.
Beginning
in the late 1980s, the opioid antagonist naltrexone was investigated as
a treatment for autism. Results from initial open-label reports and small
controlled studies suggested possible effectiveness, although more recent
large controlled investigations have failed to demonstrate improvement
in the majority of target symptoms or social behavior. The most consistent
findings from these studies were that naltrexone is well tolerated and
may be effective for reducing hyperactivity.
A
number of other drugs, including b-adrenergic
antagonists and the 5-HT1A partial agonist buspirone, have
been studied, although most of these trials were either uncontrolled or
contained a small number of subjects. Controlled investigations of mood
stabilizers, including lithium, valproic acid, carbamazepine, and gabapentin,
have not been reported in well-defined groups of autistic subjects.
The
pharmacological management of hyperactivity and impaired attention has
proven particularly challenging. These symptoms are most prominent in
younger autistic children when educational programming and interventions
are most critical. Anecdotal reports from physicians in clinical practice
and in academic centers commonly describe the onset or exacerbation of
irritability, insomnia, and aggression with psychostimulants. Early controlled
studies with these agents produced mixed results at best. In a recent
double-blind crossover study of methylphenidate and placebo in 10 autistic
children, modest improvement was seen on measures of hyperactivity and
irritability, and adverse effects were minimal. After completion of the
study, however, it was necessary to add haloperidol to the treatment of
2 of the 10 children because of persistent aggression. In a more recent
study, 8 of 13 children responded to methylphenidate in a double-blind,
placebo-controlled crossover study. Adverse effects were more frequent
at higher doses and included social withdrawal, dullness, sadness, and
irritability. A larger controlled study of methylphenidate is currently
being conducted by the Research Units on Pediatric Psychopharmacology
(RUPP) Autism Network to determine whether age, diagnosis, cognitive function,
or other factors are related to response to psychostimulants.
The
a2-adrenergic agonist clonidine
has been shown to be modestly effective for treating hyperactivity in
small samples of autistic children. A recent open-label, retrospective
study of guanfacine, an a2-adrenergic
agonist with a longer half-life than clonidine, which may be less sedating
and cause less pronounced hypotension, found that approximately one quarter
of children showed significant improvement in hyperactivity, inattention,
and insomnia. Those with PDD-not otherwise specified (NOS) and Asperger’s
disorder showed a greater rate of response than those with autism.
In
Kanner’s 1943 landmark description of 11 autistic children, the
repetitive natures of behavior, speech, and modes of social interaction
were designated as core clinical elements of the syndrome. Based on the
efficacy of serotonin reuptake inhibitors (SRIs) in the treatment of obsessive-compulsive
disorder, the high prevalence of interfering repetitive thoughts and behavior
in subjects with PDDs, and evidence for a dysregulation in 5-HT function
in autism, researchers have been studying the clinical response and side
effect profile of SRIs in this population.
A
large double-blind comparison of the potent but nonselective SRI clomipramine,
the relatively selective norepinephrine reuptake inhibitor desipramine,
and placebo was conducted in children and adolescents. Clomipramine was
superior to both desipramine and placebo on ratings of autistic symptoms,
including stereotypies, anger, and ritualized behaviors, with no difference
between desipramine and placebo. Clomipramine was equal to desipramine
and both drugs were superior to placebo for reducing hyperactivity. One
child developed QT prolongation on electrocardiogram, another developed
severe tachycardia, and a third had a grand mal seizure. Subsequent open-label
trials of clomipramine have suggested that younger children may tolerate
clomipramine less well and show a decreased response compared with adolescents
and adults with PDDs.
Because
of their better side effect profile compared with clomipramine, selective
SRIs (SSRIs) have been receiving increasing attention for the treatment
of interfering repetitive behavior and other symptoms. To date, only one
double-blind, placebo-controlled study of an SSRI has been published.
Fluvoxamine (mean dose, 276.7 mg daily) or placebo was given to 30 autistic
adults for 12 weeks. Eight of 15 subjects who received fluvoxamine versus
none given placebo were responders. Fluvoxamine was effective for reducing
repetitive thoughts and behavior, aggression, and inappropriate repetitive
language use. Adverse effects included transient nausea and sedation.
In
contrast, a study of similar design in children and adolescents found
the drug (mean dose, 106.9 mg/day) to be poorly tolerated with only limited
efficacy. Only 1 of 18 of the fluvoxamine-treated children improved with
the drug. Fourteen of the children randomly assigned to fluvoxamine demonstrated
adverse effects, typically behavioral activation.
To
date, there have been no published controlled studies of fluoxetine, sertraline,
paroxetine, or citalopram. Results from open-label studies have been mixed.
Trials conducted in adolescents and adults have generally yielded more
positive findings than those in children.
The
results from many of these studies suggest that SRIs may be less well
tolerated and less effective in younger (prepubertal) autistic
subjects compared with autistic adolescents and adults (postpubertal).
Recent data indicate that significant changes in measures of 5-HT function
occur during puberty in autistic individuals. For example, it was found
that mean platelet 5-HT levels were significantly higher in prepubertal
autistic children than prepubertal normal controls, but no difference
was found between postpubertal male autistic subjects and postpubertal
normal controls. Furthermore, results from a positron emission tomography
brain imaging study showed that changes in brain 5-HT synthesis capacity
which normally occur in developing humans are disrupted in autistic children.
Thus, pre- and postpubertal autistic subjects may have significant differences
in brain 5-HT function which influence their ability to tolerate and respond
to SRIs. Pharmacogenetic differences among autistic individuals, which
may affect SRI tolerability and responsivity, will also require more investigation.
Considerable
interest has been generated by the introduction of the atypical antipsychotics.
These agents appear to be better tolerated and have a lower risk of acute
and tardive dyskinesias compared with conventional antipsychotics. In
addition, these drugs have been shown to improve both the “positive”
and “negative” symptoms of schizophrenia. A number of investigators
have suggested that the negative symptoms of schizophrenia are comparable
with those that characterize the social impairment of autism. These agents
have largely been targeted toward the treatment of aggression in subjects
with PDDs.
There
have been only two reports describing the use of clozapine in autism.
Three children with significant hyperactivity or aggression were given
clozapine after they had not responded to typical antipsychotics. Improvement
was observed in the three subjects after 3 months’ treatment at
dosages up to 200 mg/day. More recently, the case of a 17-year-old male
with autism who showed a significant reduction in signs of “overt
tension,” hyperactivity, and repetitive motions in response to clozapine
275 mg/day was described. The scarcity of reports on the use of clozapine
might reflect concern regarding the risks of agranulocytosis and seizures
associated with the drug. Because autistic individuals typically have
an impaired ability to communicate and often a high pain threshold, infections
secondary to a decreased white blood cell count may not be identified
in a timely manner. In addition, clozapine can lower the seizure threshold
and up to one third of individuals with autism have seizures. Furthermore,
the necessary frequent blood draws are not ideal for children, particularly
those with autism.
A
number of open-label reports with risperidone describing improvement in
aggression, self-injury, ritualistic behavior, irritability, impulsivity,
hyperactivity, and social relatedness have appeared. In the only published
controlled study to date, 8 of 14 adults with autism or PDD-NOS treated
with risperidone for 12 weeks responded compared with none of 16 given
placebo. Risperidone reduced interfering repetitive behavior, as well
as aggression toward self, others, and property. In general, the drug
was well tolerated. The weight gain that has been observed with risperidone
and other atypical antipsychotics in the treatment of some children and
adolescents with PDDs did not occur to the same degree in this study of
adults.
The
RUPP Autism Network recently completed an 8-week, double-blind, placebo-controlled
study of risperidone in 101 children with autism. Risperidone resulted
in a significant decrease in self-injury, aggression, and agitation compared
with placebo. Nearly 70% of children given risperidone responded compared
with a placebo response rate of 11.5%. Increased appetite with associated
weight gain, transient sedation, tremor, and drooling were more common
with risperidone than placebo. Results from a longer-term treatment extension
phase of this study will be forthcoming.
Case
reports, an open-label case series, and a prospective comparison with
haloperidol have described positive responses to olanzapine. In a recently
published study using a parallel-groups design, 12 children with autism
were randomly assigned to 6 weeks of open-label treatment with olanzapine
or haloperidol. Both groups showed symptom reduction. Five of six subjects
in the olanzapine group and three of six in the haloperidol group were
rated as responders. Weight gain was significantly greater in the olanzapine
group, whereas extrapyramidal symptoms occurred more frequently in children
treated with haloperidol.
Only
one report of quetiapine in the treatment of autism has been published.
Two subjects completed the entire 16-week trial and both were considered
responders. However, only one of these two subjects continued to benefit
from longer-term treatment. Three subjects dropped out because of lack
of response and sedation and a fourth because of a seizure. Other significant
side effects included behavioral activation, increased appetite, and weight
gain.
New
directions in drug development are beginning to emerge. The N-methyl-d-aspartate
(NMDA) subtype of glutamate receptor, which is central to developmental
processes including neuronal migration, differentiation, and plasticity,
is receiving increasing attention. Preliminary results from studies of
drugs that modulate glutamate neurotransmission have been published. In
an open-label study, 8 of 13 subjects given lamotrigine for intractable
epilepsy showed a decrease in “autistic symptoms.” Lamotrigine
attenuates some forms of cortical glutamate release via inhibition of
sodium channels, P- and N-type calcium channels, and potassium channels.
In a controlled study of the NMDA receptor antagonist amantadine, no difference
was found between drug and placebo on parent ratings, although clinician-rated
measures of hyperactivity and inappropriate speech showed significant
improvement. A trend toward greater percentage of responders was observed
in the amantadine group and the drug was well tolerated.
On
the basis of these preliminary results, and reports that the “negative”
symptoms of schizophrenia improve with drugs active at the NMDA receptor,
additional research with agents affecting the glutamatergic systems appears
warranted. The group II/III metabotropic-glutamate receptor agonists and
modulators of AMPA receptors may hold promise in this regard. Of interest,
one mechanism of action underlying the relative efficacy of atypical antipsychotics,
such as risperidone, for autism may be the suppression of glutamate release
via 5-HT2A antagonism. A model describing the interaction of
5-HT2A receptors and thalamocortical glutamatergic neurons,
which may underlie this effect, is presented in Figure
1.
Neuroimmunological
dysfunction has also been implicated in the pathophysiology of autism.
Insults to the immune system can lead to increased production of catabolites
of tryptophan, including quinolinate and kynurenate, which can cause significant
neurotoxicity via activity at the NMDA receptor complex. Thus neuroimmune
dysregulation in autism would not be inconsistent with altered glutamatergic
function, as described above. Results from small open-label studies of
intravenous immunoglobulin and oral vancomycin have suggested that these
interventions may be helpful in some subjects. Controlled studies of agents
with direct effects on immune function, however, have not been conducted.
Significant
progress has been made in the psychopharmacology of autism. As in more
recently conducted studies, investigators should continue to include subjects
who demonstrate severity in a particular symptom domain as it is unlikely
that any one drug will benefit the wide range of maladaptive behaviors
seen in autism. Future research should include longer-term studies of
atypical antipsychotics to gather longitudinal efficacy and safety data.
Larger controlled trials of SRIs in pre- versus postpubertal individuals,
as well as studies designed to determine the effects of these drugs in
subjects with different subtypes of PDD, including Asperger’s disorder,
are also needed. In these studies, the optimal dosage for age and developmental
level and the duration of adequate treatment should be determined. In
addition, genetic predictors of treatment response, such as 5-HT transporter
protein genotype, should be sought. Finally, exploration of promising
therapeutic strategies, such as those affecting glutamatergic and neuroimmune
function, may provide new insights into the neurobiology and treatment
of this devastating group of disorders.
top of page
|
