14 Jun 2017
by Allison Marin
In the largest comparative analysis of antipsychotic medications in schizophrenia to date, clozapine and long-acting injectable (LAI) antipsychotics come out on top in terms of relapse prevention, reports a new study published online June 7 in JAMA Psychiatry. The nationwide, register-based study, led by Jari Tiihonen at the Karolinska Institutet in Stockholm, Sweden, included nearly 30,000 patients.
“This is a landmark study,” wrote John Kane, of Zucker Hillside Hospital in Glen Oaks, New York, in an email to SRF. “It is a very important report with several methodological strengths. The fact that it draws on a nationwide database is important in eliminating missing data, and the fact that each subject serves as his or her own control eliminates the selection bias seen in some cohort or naturalistic studies.”
The multitude of antipsychotic drugs on the market, each with its own set of adverse effects, can make choosing the right one for a patient very difficult. “Unfortunately, despite significant research, methodological issues have precluded a clear understanding of whether there are clinically meaningful differences between antipsychotic treatments in terms of relapse prevention of schizophrenia,” Tiihonen wrote in an email to SRF.
An assessment of adverse events and long-term outcomes of antipsychotic treatment requires large numbers of patients that are not practical to achieve in randomized clinical trials (RCTs). In addition, the patient populations included in RCTs are highly selected because of both patient refusal as well as necessary exclusions due to substance abuse and other comorbidities. Thus, as few as 15 percent of schizophrenia patients are ultimately eligible for RCTs (Hofer et al., 2000).
While observational, population-based studies can overcome some of these issues and better assess the real-world effectiveness of the drugs, they, too, have selection biases due to the confounding differences among individuals in factors such as illness severity and treatment adherence, explained Tiihonen. In the new study, the researchers overcame this limitation by performing within-subject analyses, a first for comparative studies of antipsychotic effectiveness, he added.
The new analysis included all Swedish individuals age 16 to 64 in 2006 who received or maintained a diagnosis of schizophrenia or schizoaffective disorder between 2006 and 2013, a total of 29,823 patients with an average age of 44 years.
Patients were followed for an average of six years. During the follow-up period, 44 percent were rehospitalized at least once, and 72 percent experienced at least one treatment failure, defined as rehospitalization, medication discontinuation or a switch to a different antipsychotic, or death.
Clozapine, as well as monthly LAI formulations of paliperidone, zuclopenthixol, perphenazine, and olanzapine, were associated with the lowest risk of psychiatric rehospitalization, while flupentixol and quetiapine were associated with the highest risk.
LAIs overall were associated with a 22 percent lower risk of rehospitalization compared to oral formulations of the same drugs.
When the researchers compared each antipsychotic only to the most commonly prescribed one—oral olanzapine—clozapine and LAI zuclopenthixol were associated with a significantly lower risk of rehospitalization. Compared to oral olanzapine, clozapine and LAIs as a group had the lowest risk of treatment failure.
Kane told SRF he was not surprised by the results. “We have believed for some time that clozapine and LAIs are associated with lower relapse/rehospitalization rates, but that is not easy to demonstrate in RCTs,” he said.
While perhaps not surprising, the results have important implications for treatment. “Our results demonstrate that there are substantial differences in antipsychotic effectiveness, between both individual drugs and routes of administration, that clinicians should consider when making treatment recommendations,” concluded Tiihonen. The team is currently conducting a similar study in a large cohort of first-episode patients.