Placebo response trajectories in short-term and long-term antipsychotic trials in schizophrenia

Increasing rates of placebo response have eroded placebo-control group differences in randomized controlled trials, although the reasons for this trend remain unclear. Data were extracted from the placebo arms in two identically designed 6-week studies and one 52-week study in the ziprasidone clinical trial database. The objective of this analysis was to identify distinct patterns of placebo response trajectories that could capture individual variability in the time course of change during a 1-year trial using growth mixture latent class analyses. These long-term placebo response patterns were contrasted with two 6-week schizophrenia studies. The placebo response trajectory analysis that showed 58% (Group 4) had gradual improvement in the PANSS negative subscale score (p<0.05), fewer dropouts (p<0.05) and improvement in abnormal movements, contrasted with 3 other trajectory groups that showed worsening on these measures. Almost all subjects (98%) in this symptom improvement group were treated with conventional antipsychotics just prior to placebo treatment. In contrast, the trajectory analyses showed worsening of symptoms based on PANSS total score in the 1-year trial (+15.5, SEM 2.6). Some gradual improvement of symptoms (-14.0, SEM 1.6) was also noted in 67% (n=114) of patients in the 6-week short term trials. Our findings indicate that substantial heterogeneity in placebo response occurs in both short-term and long-term trials. The placebo response trajectories appeared to depend on the efficacy measure of symptom reduction chosen, prior antipsychotic use profile, and trial durations. Further research is warranted to examine the trajectory patterns of placebo responses in independent patient populations.