Application of Plasma Levels of Olanzapine and N-Desmethyl-Olanzapine to Monitor Clinical Efficacy in Patients with Schizophrenia

BACKGROUND: This therapeutic drug monitoring (TDM) study aimed to determine the role of olanzapine (OLZ) and N-desmethyl-OLZ (DMO) levels in the therapeutic efficacy of OLZ in patients with schizophrenia. METHOD: Plasma concentrations of OLZ (C(OLZ)) and DMO (C(DMO)) in schizophrenic patients 12 hours post-dose were assessed. The correlations of C(OLZ) and C(DMO) with the various scores of the Positive and Negative Syndrome Scale (PANSS) were evaluated. A receiver operating characteristic curve (ROC) was utilized to identify the threshold C(OLZ) and C(OLZ)/C(DMO) ratio for maintenance of satisfactory efficacy. RESULTS: A total of 151 samples from patients with schizophrenia were analyzed for individual C(OLZ) and C(DMO) levels. The mean C(OLZ) and C(DMO) levels were 37.0 ± 25.6 and 6.9 ± 4.7 ng/mL, respectively, and C(OLZ) was ~50% higher in female or nonsmokers (p<0.01). In all patients, the daily dose of OLZ was positively correlated with C(OLZ) and C(DMO). Linear relationships between C(OLZ) and OLZ dose were observed in both nonsmokers and smokers (r(s) = 0.306, 0.426, p<0.01), although C(DMO) was only correlated with OLZ dose in smokers (r(s) = 0.485, p<0.01) and not nonsmokers. In all patients, C(OLZ) was marginally negatively correlated with the total PANSS score. The total PANSS score was significantly negatively correlated with the C(OLZ)/C(DMO) ratio (p<0.005), except in smokers. The ROC analysis identified a C(OLZ)/C(DMO) ratio ≥2.99 or C(OLZ) ≥22.77 ng/mL as a predictor of maintenance of an at least mildly ill status (PANSS score ≤58) of schizophrenia in all patients. CONCLUSIONS: A significantly negative correlation between the steady-state C(OLZ)/C(DMO) ratio and total PANSS score was observed in Taiwanese schizophrenic patients. TDM of both OLZ and DMO levels could assist clinical practice when individualizing OLZ dosage adjustments for patients with schizophrenia.