Older Antipsychotic Drug Found Safer than New Treatments

September 17, 2008
Pharmaceutical Executive

Volume 0, Issue 0

A new study from the NIMH shows that next-generation treatments for schizophrenia might not be any better than older, first generation drugs-and in some cases might be worst.

A new study of children receiving antipsychotic treatments for schizophrenia has caused a stir after it was revealed that patients taking second generation antipsychotics such as olanzapine (the active ingredient in Lilly’s Zyprexa) and risperidone (Janssen-Cilag’s Risperdal) were more prone to adverse reactions than children using older treatments.

The study, funded by The National Institute of Mental Health and published online in the American Journal of Psychiatry, was conducted to see if the efficacy of the new drugs in children and adolescents was comparable to their efficacy in adults. The investigators noted that not much pediatric data was available on newer drugs, and they wanted to compare the drugs to a treatment that had a fair amount of data-such as molindone.

“At the time we started this study, there were essentially no data on children [for the drugs],” said lead investigator Linmarie Sikich, a children’s psychiatrist and assistant professor at the University of North Carolina. “We had thought that there might be advantages in atypical versus traditional agents. However, most of the existing work was suppositions from adult work.”

Children were randomized to receive a series of medications; if they didn’t respond to one medicine, their second line drug would be pre-established, so all the treatment could be double-blinded.

If children tolerated their regime at eight weeks, the treatment continued for a year to look at longer term side effects. “We really expected when we started the study that the two atypical agents-risperidone and olanzapine-would be much better tolerated and more efficacious than the molindone, though we did expect that the molindone would cause less weight gain,” Sikich said.

That was not the case, however. In the first eight weeks, there was no detectible difference in the efficacy of any of the medications, according to the report. They all reduced symptoms of hallucinations and delusions at the same speed, and worked on the same proportion of kids. The bad news is that none of the treatments worked in a large number of kids. Overall response rates at eight weeks were quite low; about 50 percent responded to molindone, 34 percent to olanzapine, and 46 percent to risperidone-though the differences may not have been significant, because a large percentage of the children had stopped taking their medications by that point.

“We weren’t surprised that olanzapine caused more weight gain-16 pounds average weight gain in the first eight weeks of the study-but we were surprised that olanzapine was associated in changes in LDL cholesterol and increases in insulin that may predispose to the development of diabetes,” Sikich said. Molindone was associated with more occurrences of restlessness.

Olanzapine seemed to come out worst in the comparison. Sikich said that it might be difficult to recommend new starts on the that treatment for children with schizophrenia. “Right now, our practice parameters say that traditional antipsychotics shouldn’t be the first line of treatment, but I think this study raises some pretty serious questions about that,” Sikich said.

Molindone has been off-patent for years, but no one has made a generic version. Lilly and Jansen donated the olanzapine and risperidone, respectively; they haven’t, however, reviewed the results or commented on the study.

Asked if she would pull her patients off any of these treatments based on this study, Sikich was hesitant. “If you switch them off a treatment, it must be done slowly,” she said. “If someone tried five other treatments, and olanzapine was the only one working, I would leave them on it, even if they had significant side effects. If it was the first or second one, and they had significant side effects, I would switch them. But you have to individualize it. There is no ‘one answer’ for everyone.”