Does early improvement predict response to the fast-dissociating D₂ receptor antagonist JNJ-37822681 in patients with acute schizophrenia?

Early predictability of sustained response to atypical antipsychotics in patients with schizophrenia has important implications for clinical decision making. In order to investigate whether early onset of efficacy correlates with week-6 response for the selective fast-dissociating D2 receptor antagonist JNJ-37822681, we analysed data from a 12-week placebo- and active-controlled (olanzapine) study designed to evaluate efficacy and safety of JNJ-37822681. Factors, including baseline Positive and Negative Syndrome Scale (PANSS) total score, waist circumference, weight, body mass index group, number of previous hospitalisations, age at diagnosis, race, sex and age at study entry, and relative (%) change from baseline on day 3 (early improvement) in PANSS total score, were analysed using logistic regression models and receiver operator characteristic (ROC) curve analysis, to predict the week-6 efficacy response (≥ 30% improvement in PANSS total score). Results showed that week-6 response with JNJ-37822681 30 mg bid treatment could be reliably predicted by improvement in PANSS total score on day 3, the number of previous hospitalisations, and race (80% accuracy [ROC area under curve]). Early improvement (day 3) in PANSS score had the highest predictive value as a single factor across all JNJ-37822681 doses. At a specificity of 70%, sensitivity for predicting week-6 response was: 0.60, 0.64, and 0.74 in the 10-, 20-, and 30 mg bid JNJ-37822681 groups, respectively; 0.40 in olanzapine group. Early improvement in PANSS may be a simple and reliable way to predict sustained response with JNJ-37822681 in patients with acute schizophrenia.