Last week, The New England Journal of Medicine released the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE). It tested whether an old standby drug once popular for treating schizophrenia–perphenazine, a relative of thorazine–is as good as the newer antipsychotics.
”Surprisingly,” in the words of the study’s sponsor, the National Institute on Mental Health, it was. “The older, less expensive medication used in the study generally performed as well as the newer medications.”
With costs of up to $6,000 a year for the new drugs, compared to only several hundred for drugs like perphenazine, state Medicaid officials who devise the drug formularies are salivating to limit prescribing options. Not so fast. Doctors must retain flexibility.
Some background. The new generation of antipsychotics (called “atypicals” because of their novel action on brain receptors) received FDA approval for schizophrenia and bipolar illness within the last 15 years. There were exciting case reports of Lazarus-like awakenings, where patients who had been unresponsive to the old medications, finally were able to leave their bedrooms after years of torment by hallucinations.
Within a few years of their introduction, the atypicals became the first-line treatment for psychotic illnesses in the U.S., accounting for about 90 percent of the market share.
Then, as clinical experience accumulated, problems appeared. Severe weight gain was often accompanied by type-2 diabetes and elevated cholesterol. The cost-benefit calculus would tip in the favor of the drugs if symptom relief were dramatically greater. But given the risk of serious complications were they worth it?
This was the question before CATIE researchers. They recruited nearly 1,500 patients with schizophrenia and assigned them randomly to traditional perphenazine and four atypicals (Geodon, Risperdal, Seroquel, Zyprexa).
The main outcome was gauged by the rate of medication switching, a measure intended to tap both the patients’ and the doctors’ judgment of how well a medication worked, including whether patient could tolerate side effects. Zyprexa got the best results (about a third of patients remained on the drug for the full 18 months of the study compared to roughly a quarter of the other patients and there were fewer hospitalizations in the group), yet this was offset by greater weight gain and increases in measures of glucose and lipid metabolism.
The study, unprecedented in its scope, has much more data to yield. Yet to be published, for example, are specific results about the effect of the medications on so-called negative symptoms such as social withdrawal and emotional flatness–symptoms considered better treated by atypicals and key to success both in the community and at work. Another issue is the dose. For example, Seroqual and Geodon were used at lower doses than they are now used in the clinical setting, because the lowwer doses were considered appropriate at the time CATIE was designed. Perhaps the drugs would have been more effective at higher doses.
Two important conclusions can be drawn at this time. First, older drugs such as perphenazine deserve respect. Contrary to the notion among many mental health workers, they are not pharmaceutical relics. Some patients do very well on them. And the price can’t be beat (a boon to the use of these drugs in the developing world).
Second, we must remember that research is about averages while clinical practice is always about the individual in the doctor’s office. Balancing symptom relief with side effects is a delicate and idiosyncratic process. More data to come from CATIE will likely help guide it. Regardless, physicians must retain the freedom to discover what is best for their patients.