Meta-Analysis Raises Questions Over Best Treatment for Refractory Schizophrenia

Despite adequate treatment with antipsychotics, an estimated one-third of all patients with schizophrenia experience persistent psychotic symptoms. For these patients, most experts agree clozapine is the best treatment option.

Jeffrey Lieberman, M.D.

But a study published February 3 in JAMA Psychiatry has raised a question as to whether clozapine is significantly better than several other antipsychotics when it comes to treating refractory schizophrenia.

In a meta-analysis led by Stefan Leucht, M.D., of the Department of Psychiatry and Psychotherapy at the Technical University of Munich, researchers included 40 randomized clinical trials of antipsychotics used to treat treatment-resistant schizophrenia in more than 5,000 patients.

The authors then developed a network model to compare the efficacy of these agents with each other. Nine antipsychotics were included in the comparison: chlorpromazine, clozapine, fluphenazine, haloperidol, olanzapine, quetiapine, risperidone, sertindole (not available in the United States), and ziprasidone.

With regard to both overall symptom improvement as well as improvements in just positive or negative symptoms, clozapine, olanzapine, and risperidone came out on top.

According to what former APA President Jeffrey Lieberman, M.D., chair of the Department of Psychiatry at Columbia University and New York Presbyterian Hospital and director of the New York State Psychiatric Institute, told Psychiatric News, these results agree with current medical practice.

The slight surprise was that the meta-analysis found little evidence that clozapine was significantly superior to the other two agents. For example, in comparison with sertindole, clozapine was found to reduce overall schizophrenia symptoms by an additional 0.40 points (measured by the Positive and Negative Syndrome Scale [PANSS]), falling in between risperidone (0.32 point reduction) and olanzapine (0.46 point reduction).

“That is an interesting result, but I don’t think intrinsic clinical practice will change because of these findings,” said Lieberman, who was not involved in this study. “For reasons we admittedly still don’t understand, clozapine does have a unique superiority in treating refractory patients, and that has been shown consistently across studies, both randomized trials and other types.”

Lieberman continued, “Instead, I think these results highlight the trend in clinical research wherein as the number of studies examining a particular medication increases, the population of enrolled patients becomes broader in their clinical criteria.” (In the meta-analysis by Leucht and colleagues, clozapine was featured in 20 of the 40 analyzed studies—more than any other antipsychotic agent.)

According to Lieberman, the result of this heterogeneity may be that a combination of lower symptom severity, heightened patient expectations, and other factors conspire to reduce drug response, increase placebo effect, and produce results that might be under-representative of the medication’s true efficacy.

This was an idea echoed by the authors of the meta-analysis, as well as Christoph Correll, M.D., and John Kane, M.D. (who led the landmark 1988 study that demonstrated the potential of clozapine and helped re-introduce its clinical use in the United States), who wrote an editorial response.

Both groups noted that the current meta-analysis excluded several open-label studies that showed a strong superiority for clozapine in treatment-resistant schizophrenia, such as the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) study. The question then is, what do these open-label findings mean?

“The fact that the studies with positive results were unblended and mostly unrandomized could be interpreted in one of two ways,” Kane and Correll wrote. “The positive findings are due to bias on the part of the treating clinicians, patients, and raters, or the patients in the open studies are more representative of the severely ill patients who benefit the most from clozapine but are less likely to enroll in complex and demanding randomized clinical trials.

“Thus, the biggest concern about the validity of the findings from [this] meta-analysis by Samara et al. is the question regarding the generalizability of the samples that were enrolled into the blinded RCTs,” they added.

When only trials that looked at clozapine therapy for first-episode psychosis were considered, the medication generally performs no better than other antipsychotics, Lieberman told Psychiatric News. “The superiority only manifests in severely ill, treatment-resistant patients.”

The study authors, editorial authors, and Lieberman all agreed that new, randomized clozapine studies that feature a large but representative population are crucial to address the current uncertainties.

“Our analysis suggests that more trials comparing clozapine with other [second-generation antipsychotics] in patients with more severe illness and using high clozapine doses are warranted,” Leucht and colleagues concluded.

Leucht’s meta-analysis was supported by a grant from the German Federal Ministry of Education and Research. ■