Rationale: CNS-active drugs produce specific electroencephalographic changes. It is known that olanzapine produces an increase in the slow (delta and theta) and a decrease in fast (all beta) frequency bands, but no relationship between plasma concentrations-effect has been studied.

Objective: To elucidate the concentration-effect relationship of olanzapine after oral administration of a single dose in healthy young volunteers.

Methods: Twenty volunteers (10 males, 10 females) received a 5mg dose of olanzapine according to a placebo controlled crossover design. Plasma levels of olanzapine were measured before and at different times up to 96h after drug administration by a validated liquid chromatography tandem mass spectrometry method. Quantitative EEG parameters were obtained at different times (0-12h). Nineteen-lead, 3-minute vigilance-controlled EEG recordings with eyes-closed were obtained. Relative theta activity (3.5-7.5 Hz) from 5 seconds free of artefact epochs after applying Fast-Fourier Transform (0.3-35 Hz) were used as the endpoint for the pharmacodynamics.

Data analysis was performed in NONMEM. First, the PK model was developed and covariate inclusion was tested. Subsequently individual PK parameter estimates were fixed and the PD model was derived.

Results: The PK of olanzapine was best described with use of a two-compartment model with first-order absorption and elimination from the central compartment. To describe the absorption process it was necessary to include a lag time. An allometric model was used to describe part of the interindividual variability (IIV) in V/F and CL/F. V/F was estimated to be 798L and CL/F was 24.9L/h for a 70kg volunteer. The overall interindividual variability in V/F was 30% and decreased to 21% when differences in weight were accounted for; and the IIV in CL/F was reduced from 29% to 27%. The other parameters (IIV, expressed as coefficient of variation) estimated by the model were: ka 0.804h^-1 (80%), tlag 0.8h (19%) , V2/F 267L, CLd/F 26.6L/h.

When the values of relative theta activity vs time were plotted a clear placebo effect was observed. This effect was best described by a cosine function. Once the structure of the placebo effect was known, placebo and drug effects were analysed simultaneously. A direct Emax model best described the effect of olanzapine. Estimates for the parameters (IIV) describing the placebo effect were: Baseline 17.1% (33%), Phase-shift 3.66 h (11%), Amplitude 2.08%. The estimates (IIV) for olanzapine effect were: Emax 43.9%, C50 13.5 microg/L (46%).

Conclusions: The pharmacokinetic/pharmacodynamic relationship of olanzapine when relative theta activity is used as the endpoint was well characterised by population PK/PD modelling. A cosine function described adequately the placebo effect on the theta frequency band, and permitted to differentiate adequately olanzapine effect from the placebo effect.

* M. Valle is supported by FIS through contract CP04/00121 from the Spanish Health Department in collaboration with Institut de Recerca de l'Hospital de la Santa Creu i Sant Pau. Barcelona. This study was supported in part by a grant from the Fundació La Marató de TV3 (Catalonia Television, Spain).