FDA Considers Pediatric Indications for Antipsychotics

March 6, 2009

The FDA Pediatric Advisory Committee met in November to review drug trials and safety data for several medications under consideration for pediatric-specific labeling. Drugs included the antipsychotics olanzapine (Zyprexa) and risperidone (Risperdal). Although not yet finding sufficient evidence of safety and efficacy in this population, the committee specified additional information that could be submitted for the applications to be reconsidered.

The FDA Pediatric Advisory Committee met in November to review drug trials and safety data for several medications under consideration for pediatric-specific labeling. Drugs included the antipsychotics olanzapine (Zyprexa) and risperidone (Risperdal). Although not yet finding sufficient evidence of safety and efficacy in this population, the committee specified additional information that could be submitted for the applications to be reconsidered. FDA approval of the supplemental New Drug Applications (sNDAs) could occur without the advisory committee support. If approved, product patent and “exclusivity” from generics is extended for the pediatric indications under the Best Pharmaceuticals for Children Act (BPCA) reauthorization, passed September 2007 to encourage pediatric drug testing for age-specific labeling.

Thomas Laughren, MD, director of the FDA Division of Psychiatric Products, summarized the agency’s evaluation of the olanzapine studies for the committee. He indicated that the data generally support the sNDA for acute mania of bipolar disorder and for schizophrenia in pediatric patients.

A 3-week randomized, placebo- controlled, flexible-dose study was conducted in multiple sites in the United States and Puerto Rico in adolescents 13 to 17 years of age with acute mania in bipolar disorder type I. A 6-week trial was also conducted at sites in the US and in Russia with patients of that age-group who had schizophrenia.

In his summary assessment, Laughren indicated that results in those with bipolar disorder were “highly favorable” to olanzapine. The study was characterized as positive in demonstrating efficacy: active drug was associated with significantly greater improvement of symptoms from baseline, measured in total score on an Adolescent Structured Young Mania Rating Scale (YMRS). There were fewer dropouts attributed to lack of efficacy (20% with olanzapine vs 35% with placebo).

The number of patients who discontinued therapy because of a lack of efficacy was even greater in the study on schizophrenia: 14% of the olanzapine recipients and 51% of placebo recipients left the study. “This finding by itself is almost enough, in my view, to convince one of the benefits in this condition,” Laughren said.

Although olanzapine treatment was also superior to placebo in improving symptoms of schizophrenia, measured with a children’s version of the Brief Psychiatric Rating Scale (BPRS-C), there was a disparity in results between the US and Russian sites. Treatment effect of olanzapine was comparable at sites in both countries; however, the placebo response was much larger in the US, and the statistical superiority of olanzapine was largely derived from the Russian data. In addition, the Russian sites appeared to be more successful than those in the US in recruiting study participants.

These disparities-and reliance on non-US sites for a statistically favorable outcome-had been of particular concern to one FDA analyst, Laughren noted, and were considered a possible basis for finding the application “unapprovable.”

Laughren acknowledged that the concern raised questions about study conduct at the Russian sites but assured the committee that the sites had been inspected and that there was no evidence of misconduct.

“It is also important to point out,” Laughren indicated, “that there are alternative explanations for more successful recruitment at the Russian sites and also a more successful outcome.”

Among the factors that could lead to the geographic disparity, Laughren explained, is the greater catchment area and less competition for recruitment in Russian sites. “If difficulty in recruitment in the US sites led to enrollment of a more heterogeneous group of subjects,” he noted, “this could have led to a higher placebo response rate.”

In concluding his report, Laughren pointed to the demonstrated efficacy of olanzapine in adults and argued that the geographic disparity in the pediatric data was not sufficient to find the sNDA nonapprovable. He indicated, however, that the agency would ask the sponsoring manufacturer“to further address this concern.”

The pharmacokinetic data for olanzapine did not indicate significant differences between adolescents and adults, although the younger population tended to be exposed to higher dosages in relation to body weight. The adverse event profile and other safety parameters were also similar to those in adults but with some greater magnitude in the younger patients. Laughren indicated that these distinctions could be reflected in modified labeling.

“In addition,” Laughren informed the committee, “we have recently asked the sponsor to provide more complete information generally with regard to effects on weight, glucose regulation, and lipid levels, so that labeling for olanzapine can be enhanced with regard to these risks.”

The committee voted unanimously to recommend that the agency continue to evaluate the safety of olanzapine in the pediatric population and to implement additional risk-management regulatory actions concerning the monitoring of metabolic changes.

Assessing Risperdal, drug class The FDA staff analysis of the Risperdal studies was prepared by Mitchell Mathis, MD, deputy director of the Division of Psychiatry Products. Two 6-week randomized, placebo-controlled trials were conducted in children and adolescents who had schizophrenia, and one 3-week trial was conducted in patients who had bipolar disorder. Active treatment groups in the schizophrenia study received daily doses of either 1 to 3 mg or 4 to 6 mg. Groups in bipolar disorder study received either 0.5 to 2.5 mg or 3 to 6 mg/d.

The data supported the applications. The most useful dosages were 3 mg/d for schizophrenia and 2.5 mg/d for bipolar disorder. Higher dosage was not associated with additional benefit in these studies but with an increase in adverse effects. Mathis suggested that the optimal dosages in the studies should be considered as targets rather than ceilings in pediatric populations.

“While I believe we should certainly label the drug with the information learned from the clinical trials, and even identify target dose . . . ,” he indicated, “I think it would be too restrictive to the prescriber to limit the dose to a maximum when we know that doses up to 6 mg/d were also shown to be efficacious in the same studies that demonstrated efficacy for the lower dose ranges.”

The review of risperidone adverse events from the pediatric studies, presented by Felicia Collins, MD, medical officer at the FDA, indicated an adverse-effect profile similar to that in adults. As to whether the standard ongoing safety monitoring is adequate for the expanded indication to pediatrics, the committee unanimously recommended several additional measures for the use of risperidone, olanzapine, and this class of antipsychotics:

• Assess on-label and off-label product use with specific attention to age and indication and concomitant drug use.
• Follow up on metabolic syndrome, growth, sexual maturation, hyperprolactinemia, and extrapyramidalside effects.
• Conduct studies-possibly with the NIH-on long-term effects in the pediatric population.

The committee voted unanimously to withhold recommendation on labeling until this additional information is provided. It also recommended that no public communications be issued on the sNDAs for pediatric populations before the committee has received and discussed data from these measures.