Ziprasidone is a novel antipsychotic agent, with high affinity for dopamine D2 and serotonin (5-HT2) receptors in vitro and in animal models. The goal of this study was to determine the time course of 5-HT2 receptor occupancy (%RO) in healthy humans after a single p.o. dose. Positron emission tomography with the 5-HT2 ligand, [18F]setoperone, was performed in eight male volunteers, in the drug-naive, base-line (BL) state and 4 to 18 hr after ziprasidone (40 mg). Cerebral cortical binding potential [BP, maximum number of available receptors/KD or association rate for specific binding (k3)/dissociation rate for specific binding (k4)] was estimated using the cerebellum as reference. Transport rate from plasma to brain (K1), transport rate from brain to plasma (k2), association rate of nonspecific binding (k5) and dissociation rate of nonspecific binding (k6) were derived by fitting cerebellar time-activity curves to a three-compartment model. Fitting of cortical data to a 4-compartment model with K1/k2, k5 and k6, fixed at cerebellar values, was used to determine k3 and k4. %RO was calculated using the relation: %RO = [(BPBL-BPDRUG)/BPBL] x 100%. At BL, cortical parameter (mean +/- S.E.M) were: K1 = 0.121 +/- 0.0072 ml.min-1.g-1; k2 = 0.0581 +/- 0.004 min-1; k3 = 0.321 +/- 0.0026 min-1; k4 = 0.0957 +/- 0.0059 min-1; k5 = 0.0147 +/- 0.00066 min-1; and k6 = 0.0059 +/- 0.00042 min-1. Ziprasidone did not effect K1, k2, k5 or k6; however, k3 was reduced and k4 was elevated (P < .01). RO was nearly complete at 4 hr after dosing (98%) and remained elevated at 18 hr (46%). Plasma concentrations were well described by a biexponential function and decreased much more rapidly than RO. These results establish that ziprasidone has high potency for blocking 5-HT2 receptors in healthy humans; a potentially important characteristic of atypical antipsychotic agents.