EARLY PHARMACOLOGIC INTERVENTION IN AUTISM: FLUOXETINE IN PRESCHOOL CHILDREN

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. Autism is a severe neurodevelopmental disorder characterized by qualitative differences in three behavioral areas: social reciprocity, communication, and the breadth of interests manifest by repetitive behaviors or restricted interests. Very few individuals with autism are able to live independently as adults. Speech by age five years is the single best predictor of good outcome. This is a revised protocol examining the impact of sustained early pharmacologic intervention in preschool children with autism. Young (30-58 months) children with narrowly defined autism will be randomly assigned to yearlong treatment with placebo or fluoxetine, a selective serotonin reuptake inhibitor (SSRI) which is approved for other indications in children as young as six years. The primary outcome measure will be the pattern of development in behavioral domains that appear particularly compromised in autism: specifically reciprocal social interaction, pragmatic communication, and restricted interests and/or repetitive behaviors. Development in these core areas will be assessed using a composite measure based on relevant subscales from the Pervasive Developmental Disorders Behavior Inventory parent report version (PDDBIp) and supplemented with other more widely used instruments that either do not have age-based standards or are not specifically designed for use in autism. Each subject may participate in four study intervals as follows. 1. The screening interval will allow for confirmation of the diagnosis and determination of eligibility for randomized treatment. No study treatment will be provided during this interval. It is expected that three to four subjects will be withdrawn after this interval for every subject that receives randomized treatment in the study. 2. The acute treatment interval will span between the baseline to Week 12/Month 3 assessments. 3. The extended treatment interval will span the period following the Month 3 assessment through the Month 12 assessment, which is the primary endpoint of the study. 4. The long-term follow up interval will examine the ten-year period, which follows the Month 12 assessment. This will be an observational period with no interventions provided by the study, but no limitation placed on the number or type of interventions that the subject experiences. Currently, early intensive behavioral interventions are viewed as among the most effective and promising means to improve the overall prognosis for persons with autism. However, many of these interventions are difficult to implement broadly because of their cost and requirement for highly skilled personnel. This study has great significance both because of the importance of determining whether readily implemented interventions can significantly improve the functioning of some children with autism and because of the importance of rigorously determining whether interventions offered to young children are safe and tolerable. Hypothesis: 1. Individuals treated with fluoxetine will demonstrate a significantly more rapid rate of improvement in the composite measure over the course of one year than individuals treated with placebo. We estimate an effect size (standardized difference in the composite score between groups) of 0.35 at 3 months, 0.4 at 7 months and 0.5 at 12 months. 2. 60% of the fluoxetine group and 20% of the placebo group will show significant improvement in overall functioning as defined by a CGI-I of "1 - very much improved" or "2 - much improved" after 12 weeks of treatment. 3. Individuals who show clinically significant behavior changes after 12 weeks of treatment will have greater cumulative developmental gains as larger increases in the PDDBIp composite measure than those who do not show clinically significant acute behavioral changes. 4. Individuals whose parents report at least one 1st degree relative with diagnosed or treated bipolar affective disorder will show more robust changes in the PDDBIp composite than individuals whose parents do not report such a 1st degree relative. We also hypothesize that individuals with two copies of the short allele of serotonin transporter protein (SERT) gene will demonstrate slower and less robust changes in the PDDBIp composite than individuals with at least one copy of the long allele. Further, we hypothesize that individuals with the low activity form of the cytochrome P450 gene, CYP2D6, will have more activation and a greater number of total side effects than individuals with the high activity form of the gene.