Steady-state concentrations of S12024, a novel compound for treatment of Alzheimer's disease, were studied to determine the pharmacokinetic parameters of S12024 in Phase IIa patients and to assess the effect of patient characteristics on those pharmacokinetics. A prospective sparse sampling strategy was used to obtain oral repeated data (n = 285) from 89 patients, which were analyzed using a one-compartment model and the NONMEM computer program. The model suggested that apparent clearance of S12024 was influenced by the study and by patient age. In the Spanish study, apparent clearance was increased by 68% and 26% for doses of 100 mg and 300 mg, respectively, and patient age decreased oral clearance by approximately 10% per decade in the patient age range (50 to 90 years). Data from only a few patients in the Spanish study were probably responsible for the observed study influence on apparent clearance of S12024, and no measured covariates could explain this effect. The model provided an excellent characterization of the observed data and it predicted correctly the plasma concentrations from an earlier Phase I trial and a subsequent Phase IIb study. The present model, built from Phase IIa data, provides a basis for examining the influence of patient covariates and the magnitude of their effects on the pharmacokinetics of S12024. The study effect is probably an artefact that will disappear by further expanding of the population model in the future.