We thank Dr Kopeček et al for their interest in our paper (Reference Xiberas, Martinot and MalletXiberas et al, 2001b ). They conclude that atypical antipsychotics do not exert special temporal or limbic selectivity, which depends instead on drug dosages. First, we believe that generalisations drawn from data obtained from five patients, each one treated with a different antipsychotic drug, are not sound, because of intersubject variability. For instance, should Dr Kopeček et al have considered plasma drug concentrations and patient H2 of our article, their conclusion would have been modified. In our article, we drew conclusions from the statistical comparisons of [⁷⁶Br]-FLB457 measures obtained with positron emission tomography (PET) in subgroups of patients, receiving the usual dosage recommended by the pharmaceutical firms for each antipsychotic drug, for treating psychotic episodes.

Second, we have already reported the importance of dosage when interpreting neuroimaging measures of regional D₂ dopamine receptor blockade by antipsychotic drugs (Reference Xiberas, Martinot and MalletXiberas et al, 2001a ). Inspection of the table that Kopeček et al draw from our article suggests that for a striatal D₂ receptor binding index approaching 65-70%, the atypical antipsychotics induce extrastriatal/striatal indices comparable with that induced by the lowest oral dosage of haloperidol reported. This is consistent with our previous publication (Reference Xiberas, Martinot and MalletXiberas et al, 2001a ) where we specifically highlighted the dose-dependence of extrastriatal/striatal D₂ blockade, from a study in a larger sample of patients treated with an atypical antipsychotic. We demonstrated that plasma concentrations were more accurately related than daily oral doses to the different regional binding profiles determined with PET. Clearly, two binding profiles could be distinguished depending on the plasma concentration of the drug: low striatal binding associated with marked extrastriatal binding for low plasma concentrations, or marked binding in both striatal and extrastriatal regions for higher plasma concentrations. This may be applicable to both atypical and typical compounds, if very low doses of typical neuroleptics (i.e. below the recommended therapeutic dose range) are considered, but this is a speculation. Therefore, having previously highlighted the effect of dosage (Reference Xiberas, Martinot and MalletXiberas et al, 2001a ), we chose to highlight in our second article (Reference Xiberas, Martinot and MalletXiberas et al, 2001b ) that, at plasma concentrations obtained in actual clinical practice, and compared with haloperidol, various atypical antipsychotic drugs have a regional binding profile that is higher in mesocorticolimbic regions than in striatum.