Two radioligands (McN5652 and C-DASB) used to label the serotonin transporter reuptake site (SERT) were assessed in baboons. Apparent ligand binding was compared in 1 control and in 1 baboon given 2 regimens of 4 daily doses of 5 mg/kg MDMA 17 and 7 mo before imaging (assumed to induce serotonergic neurotoxicity). Control imaging sessions were also compared to sessions after paroxetine pre-treatment (10 mg/kg given 60 min before session or 5 mg 2 times on night before session and 2 times 60 min before sessions. Input function and plasma protein binding were measured during each session. Baboons were killed after imaging, and autoradiography was performed on brain tissue. 3 saline-treated baboons were killed without having undergone any imaging, and their brains were examined for amounts of 5-HT and 5-HIAA via HPLC. Radiation indicating McN5652 was highest in pons, midbrain, hypothalamus, striatum and thalamus, and these findings were highly correlated with in vitro distribution of the SERT. McN5625 presence was also correlated with in vitro measures of 5-HT and 5HIAA. C-DASB distribution correlated with McN5625 distribution, but with greater contrast between areas of high (e.g. hypothalamus) and low (e.g. cerebellum) SERT density. McN5652 concentration was reduced after MDMA in 4 of 10 regions (pons, putamen, parietal cortex and cerebellum), whereas C-DASB concentration was reduced after MDMA in frontal, occipital, and parietal cortex, and in the putamen. Correlation between radioligand distribution and measures of brain 5-HT were also high for both ligands in MDMA-treated baboons. Paroxetine pre-treatment produced earlier uptake peaks than at baseline or in MDMA-treated animals, and ligand metabolism was inhibited for both McN5652 and C-DASB. After Bonferroni test, paroxetine produced no significant effects on McN5652 distribution, though before Bonferroni test applied, differences were found in cortical regions and cerebellum. Ligand concentration for both McN5652 and C-DASB was lower in MDMA-treated animals than in paroxetine-pre-treated controls in some areas )frontal cortex, cingulate), while ligand concentration was higher in MDMA-treated animals in other areas, such as pons, hypothalamus and midbrain. Study findings suggest that while both McN5652 and C-DASB may be binding to SERT, C-DASB may produce a better approximation of SERT distribution and density than McN5652. Findings also suggest that there was no single variable that could serve as a reference for calculating non-specific binding for SERT. It is notable that despite having previously used comparisons between McN5652 enantiomers as a means of assessing non-specific binding, the study authors did not use this procedure in this investigation. The study sample size is small, and the study design is unusual in that only 1 baboon received MDMA, and none of the animals used for neurochemical analysis were given MDMA or paroxetine. This means that all study comparisons relating to MDMA relate to findings in 1 of 5 animals. Lower ligand distribution in some brain areas in MDMA-treated baboon than in the paroxetine-pretreated animal suggests that decreased concentration may be altered by factors other than reduced SERT. For these reasons, findings from human studies using these ligands should be treated with caution. In addition, a recent review indicates that reduced SERT density may be determined by several factors (Kish 2002), and hence even if it could be accurately measured via radioligands, it cannot serve as an unambiguous measure of serotonergic neurotoxicity. However, study finding does suggest that serotonergic neurotoxicity might be better detected in areas where ligand binding was higher after MDMA than after paroxetine (such as pons or hypothalamus).