Archief Roodbont

With generic risperidone now approved, new trials of paliperidone, its proprietary offspring, were expected. This 6-week, randomized, controlled, double-blind, international study was industry-sponsored, -designed, and -reported. The 399 participants had chronic schizophrenia (mean illness duration, 10 years) and a recent exacerbation requiring rehospitalization. Patients with treatment-resistant disease were excluded.

One day after discontinuing other psychiatric medications, patients began extended-release paliperidone at 6 mg/day, quetiapine at 50 mg/day, or placebo. Target doses, achieved by day 5, were at the high end of recommended ranges (paliperidone, 9 mg/day; quetiapine, 600 mg/day); higher doses were optional. In the first 2 weeks, patients received monotherapy (except as needed for agitation or insomnia); during the following 4 weeks (additive phase), other medications could be added. The study’s primary endpoint was change in Positive and Negative Syndrome Scale (PANSS) total scores at week 2. Results were significantly better with paliperidone than with quetiapine at week 2.

During the additive phase, further PANSS changes were similar in the paliperidone and quetiapine groups. In intent-to-treat analyses, paliperidone was superior to quetiapine. Similar percentages of patients in the paliperidone and quetiapine groups took additional antipsychotics or other psychiatric medications (53%–55%). Mean prolactin levels were significantly higher with paliperidone than with quetiapine or placebo.

Comment: Skeptical readers will wonder about this study’s design and analysis. Why was quetiapine selected as the comparator (instead of olanzapine or haloperidol)? Was the study’s playing field level? Quetiapine requires 5 days to reach target doses, but paliperidone is often clinically effective at 3 days; it is not surprising that paliperidone separated from placebo at day 5 but quetiapine did not separate until day 9. Also, why are significance tests discussed when results favor paliperidone, but not when they might favor quetiapine (e.g., prolactin levels)? We need to be wary of industry-sponsored clinical trials showing benefit for the sponsor’s product (Am J Psychiatry 2006; 163:185).

— Joel Yager, MD

Published in Journal Watch Psychiatry May 22, 2009