Placebo-Controlled Study of Risperidone for the Treatment of Children and Adolescents With Autism and Negative Behavioral Symptoms

The primary purpose of this study is to compare the relative safety and efficacy of risperidone and placebo in the treatment of children and adolescents with autistic disorder. HYPOTHESES: (1) Risperidone will be more effective than placebo in reducing impulsive aggression, agitation, self-injurious behavior, and troublesome repetitive behavior associated with autism. (2) Risperidone will result in more sedation (transient) and weight gain than placebo. (3) Patients continued on risperidone will be significantly less likely to experience exacerbation of symptoms of irritability, aggression, agitation, and stereotypy than those randomized to placebo, as measured by the Aberrant Behavior Checklist (ABC), the Ritvo-Freeman Real Life Rating Scale, and the compulsions scale from the Children's Yale-Brown Obsessive Compulsive Scale (CY-BOCS). (4) Patients continued on risperidone would show superior adjustment and functioning at the end of the trial, as evidenced by lower Clinical Global Impression ratings, when compared to patients randomized to placebo. Design Phase I: Double-Blind Phase - Randomized, double-blind, placebo-controlled, parallel groups design. Eight-week double-blind treatment with risperidone or placebo. Eight-week open trial with risperidone for placebo non-responders (patients who were randomized to placebo and showed no improvement). (Risperidone responders will be eligible to enter the four-month extension study. Placebo responders and risperidone non-responders will be managed as clinically appropriate by each research site.) Phase II: Extension Study - Four-month, open treatment with risperidone. Dose adjustment permitted according to clinical assessment (efficacy or adverse events). Two-month, randomized, double-blind, placebo-controlled discontinuation, parallel group design. Completers of four-month Extension Phase protocol who have maintained significantly improved status (decrease greater than 25% in ABC from Protocol I Baseline ratings and CGI of much or very much improved) will be randomized at the end of four months to placebo substitution or risperidone continuation. Group assigned to placebo substitution will undergo weekly blinded reductions of entry dose (dosage at end of Phase I) by 25% per week over three consecutive weeks. After full placebo substitution, placebo group will remain on placebo for total of up to 5 weeks (three-week taper, five-week remaining on placebo). Group assigned to continued active treatment will be maintained on entry dose level for full 8 weeks of Phase II, assuming no behavioral deterioration. Active treatment patients may have dose reduced for treatment emergent effects. Randomization - Balanced within site by Tanner Stage (pre-pubertal: Tanner I or II as measured by the absence of pubic hair; post-pubertal: Tanner III or greater), gender, and anticonvulsant use.