Following previous validation in baboons, we have studied the characteristics of [18F]setoperone as a radioligand for investigating serotonergic 5-hydroxytryptamine2 (5-HT2) receptors in the normal, unmedicated human brain with positron emission tomography (PET); subjects orally pretreated with therapeutic amounts of ketanserin, sulpiride, or prazosin were also studied to evaluate the specificity and sensitivity of [18F]setoperone brain specific binding. In controls (n = 10), the tracer showed a clear-cut retention in both frontal cortex and striatum (known to contain a high density of 5-HT2 receptors) relative to cerebellum (known to be devoid of 5-HT2 receptors). In the seven young controls (20-39 years old), the frontal cortex/cerebellum and striatum/cerebellum ratios increased during the first hour to reach similar values of 2.53 +/- 0.12 and 2.38 +/- 0.11 (mean +/- SEM), respectively, and were essentially stable during the second hour. Pretreatment with ketanserin (a 5-HT2 blocker) significantly reduced the frontal cortex/cerebellum ratio to 0.7-1.0 at 65 min, whereas the striatum/cerebellum ratio was significantly, but only partially, reduced. During sulpiride treatment (a D2 blocker), the frontal cortex/cerebellum ratio was not altered, whereas the striatum/cerebellum ratio was significantly, but only partially, reduced. With prazosin pretreatment (an alpha 1-adrenergic blocker), neither the frontal cortex/cerebellum nor the striatum/cerebellum ratio was modified. These data in humans with PET demonstrate that [18F]setoperone labels with high sensitivity and selectivity 5-HT2 receptors in the frontal cortex; in striata, however, binding is to both 5-HT2 and D2 receptors. The deproteinated-to-whole plasma radio-activity concentration ratio increased with time following injection. The mean percentage of intact [18F]setoperone, in deproteinated plasma, was 82, 74, 53, 45, 30, and 22% at 5, 10, 20, 30, 60, and 110 min following injection, respectively. These data indicate that [18F]setoperone (a) is significantly bound to plasma proteins and (b) is significantly metabolized into several labeled metabolites that are much more hydrophilic than setoperone and, hence, presumably do not cross the blood-brain barrier. These results suggest the suitability of [18F]setoperone data for modeling of 5-HT2 receptor binding in brain.