CAFE Study Shows Varying Benefits Among Atypical Antipsychotics

Paula Moyer, MA

October 25, 2005

Oct. 25, 2005 (Amsterdam) — Although all of the atypical antipsychotic agents preserve neurocognitive function, they vary in terms of the specific domains they influence, said Richard S. E. Keefe, PhD, who presented study findings here at the European College of Neuropsychopharmacology 18th Congress.

Because of this intraclass variation, "it's important to take these data...and target individualized medicine for these patients, and make sure patients have access to all of these medications, because each patient will respond differently," Dr. Keefe said at the late-breaking presentation. Dr. Keefe is an associate professor of psychiatry and behavioral sciences at Duke University Medical Center in Durham, North Carolina.

The study, Comparison of Atypicals for First-Episode Psychosis (CAFE), involved 26 centers in the U.S. and Canada, and it recruited patients who were experiencing their first psychotic episode. The investigators attempted to determine whether there were any advantages among the antipsychotic agents for first-episode patients because such patients typically respond well to treatment, but they often discontinue treatment and therefore have recurrent episodes that are less easy to treat.

In this study, the investigators compared quetiapine (Seroquel), risperidone (Risperdal), and olanzapine (Zyprexa) in these patients. There were 134 patients in the quetiapine group and 133 each in the olanzapine and risperidone groups. The study was designed to evaluate overall effectiveness, as measured by all-cause treatment discontinuation, of the three agents.

Eligible patients were 16 to 40 years old, had a diagnosis of first-episode schizophrenia, and had psychotic symptoms for one month to five years, with scores greater than 4 on the Positive and Negative Symptoms Score (PANSS) psychosis subscale. The patients could not have had a previous psychotic episode from which they recovered; nor could they have a lifetime history of more than 16 weeks of antipsychotic treatment.

Discontinuation of treatment could be due to an inadequate effect, adverse effects, and "patient decision," which consisted of failure to keep appointments, desire to stop medication for personal reasons, and other patient-specific reasons for stopping treatment. Secondary outcomes included PANSS and Clinical Global Impression (CGI) scores, as well as the Calgary Depression Scale for Schizophrenia, two neurocognitive batteries, and adverse effect characterizations. The average doses were 11.7 mg for olanzapine, 506 mg for quetiapine, and 2.4 mg for risperidone.

"There was no difference among the drugs in terms of the all-cause discontinuation rate," Dr. Keefe said. "After one year of treatment, all three of the medications had a pretty high rate of discontinuation, about 70% in all of them. There was no difference in subcategories of reasons for discontinuation, whether it was patient decision or inadequate effect."

There were also no statistically significant differences in the medications' effects on PANSS scores, Dr. Keefe said. However, the analysis of the improvement of positive symptoms showed that patients on olanzapine improved more than those on quetiapine. "That was the only difference among the drugs," he said. The agents were associated with similar results on the CGI and the Calgary Depression Scale.

Similarly, the agents were associated with similar rates of adverse effects. Dr. Keefe noted that patients receiving quetiapine were more likely to report dry mouth and somnolence, and those receiving risperidone were more likely to complain of drooling. Patients receiving olanzapine gained an average of7 kg over their baseline weight after 12 weeks and approximately 11.5 kg at one year. The other two agents were associated with similar rates of weight gain, which was less than that seen with olanzapine, as well as fewer numbers experiencing significant weight gain. Of those treated with olanzapine, 80% gained 7% or more over their baseline weight compared with 57.6% of those receiving risperidone and 50% of those receiving quetiapine.

Patients underwent the neurocognitive battery at baseline, week 12, and week 52. The investigators assessed the cognitive domains of processing speed, reasoning and problem solving, verbal memory, working memory, and vigilance. The results showed no statistically significant difference among the drugs, with a modest improvement in all. However, when the investigators assessed the effect of the medications on different cognitive domains, they saw that quetiapine had a greater effect on verbal fluency and processing speed. Risperidone was associated with a greater improvement in working memory.

The results show the importance of tailoring the treatment to the individual patient's needs, Dr. Keefe said. "All patients should have access to all of the medications, so that the physician can make the best choice for the individual patient," he said.

CAFE was funded by AstraZeneca, the maker of Seroquel.

ECNP 18th Congress: Late-breaking session, presentation 5. Presented Oct. 23, 2005.

Reviewed by Gary D. Vogin, MD