Positron emission tomography (PET) studies reveal that clozapine at clinically used doses occupies less than 60% of D2/D3 dopamine receptors in human striatum. Here, the occupancy of D2/D3 dopamine receptors by clozapine in patients with schizophrenia was determined to test the hypothesis that clozapine binds preferentially to extrastriatal dopamine receptors. A total of 15 clozapine-treated inpatients with schizophrenia underwent a [18F]fallypride PET scan. Receptor occupancy was calculated as percent reduction in binding potential relative to unblocked values measured in seven normal volunteers. Mean D2/D3 receptor occupancy was statistically significantly higher in cortical (inferior temporal cortex 55%) than in striatal regions (putamen 36%, caudate 43%, p<0.005). While the maximum attainable receptor occupancy Emax approached 100% both in the striatum and cortex, the plasma concentration at 50% of Emax (ED50) was much higher in the putamen (950 ng/ml) than in the inferior temporal cortex (333 ng/ml). Clozapine binds preferentially to cortical D2/D3 receptors over a wide range of plasma concentrations. This selectivity is lost at extremely high plasma levels. Occupancy of cortical receptors approaches 60% with plasma clozapine in the range 350-400 ng/ml, which corresponds to the threshold for antipsychotic efficacy of clozapine. Extrastriatal binding of clozapine may be more relevant to its antipsychotic actions than striatal. However, further studies with an intraindividual comparison of untreated vs treated state are desirable to confirm this finding.