OBJECTIVE: Central D1, D2, and 5-HT2 receptor occupancy in schizophrenic patients treated with clozapine was determined and related to clozapine serum concentrations. METHOD: Seventeen patients treated with clozapine (125-600 mg/day) were examined with positron emission tomography (PET) and one to three of the following selective radioligands: [11C]SCH23390 (N = 11), [11C]raclopride (N = 16), and [11C]N-methylspiperone (N = 5). Clozapine concentration in serum was determined by gas chromatography/mass spectrometry. RESULTS: D2 receptor occupancy (20%-67%) was lower than that previously determined in patients treated with classical neuroleptics (70%-90%). D1 receptor occupancy (36%-59%) was higher than that induced by classical neuroleptics (0%-44%). 5-HT2 receptor occupancy was very high (84%- 94%), even at low clozapine doses. Despite a 20-fold range in clozapine serum concentration (105-2121 ng/ml) at the time of PET examination, D2 receptor occupancy was low in all patients and was not described by the curvilinear relationship between serum drug concentration and receptor occupancy that has been demonstrated for classical antipsychotics. CONCLUSIONS: The results confirm in an extended series of patients that clozapine is atypical with regard to degree of D2 receptor occupancy, a finding that may explain the lack of extrapyramidal side effects. The combination of relatively high D1, low D2, and very high 5-HT2 receptor occupancy values is unique to clozapine. Clozapine serum concentrations have not been unequivocally shown to predict clinical effects. In this study, concentration did not predict degree of occupancy in brain. Thus, careful clinical titration cannot be replaced by monitoring of drug concentrations for optimization of clozapine treatment in individual patients.