The changes in aminergic receptors elicited by antidepressant treatments have been extensively examined in the brain of experimental animals using radioligand and molecular techniques. However, there is a very limited direct information regarding the changes effected by such treatments in the human brain, as well as its relationship to clinical improvement. Using positron emission tomography (PET) scanning, the authors examined the cortical 5-Hydroxytryptamine-2A (5-HT2A) receptor binding of [18F]fluoro-ethyl-spiperone after a 4-week treatment with the selective serotonin reuptake inhibitor paroxetine. [18F]fluoro-ethyl-spiperone labels 5-HT2A receptors in the cortex and dopamine D2 receptors in the basal ganglia. A binding index (BI) was calculated in the frontal cortex and the basal ganglia (mostly caudate-putamen) by reference to cerebellum. Thirty-seven inpatients with major depression with a mean +/- SD score on the 21-item Hamilton Rating Scale for Depression (HAM-D-21) of 26.3 +/- 4.3 at admission were treated with paroxetine 40 mg/day. After 4 weeks of treatment, the BI in the frontal cortex of remitted patients (HAM-D-21 score = 4.7 +/- 4.0; N = 20) was significantly greater than the score in nonresponder patients (HAM-D-21 score = 21.2 +/- 4.0; N = 17) (BI = 0.54 +/- 0.15 and 0.41 +/- 0.17, respectively; p < 0.02). No such difference was observed in the basal ganglia (5.45 +/- 1.11 and 5.39 +/- 0.82, respectively; p = 0.85). The significant difference in cortical BI persisted when age was used as covariate (p < 0.016). These data suggest that clinical improvement in patients treated with paroxetine is associated with an increase in the density of 5-HT2A receptors in the frontal cortex.