Behavioural pharmacology of the new generation of antipsychotic agents

In conclusion, the newer agents all have behavioural profiles which can be clearly differentiated from those of the older, classical agents. Behavioural data indicate that to a greater or lesser extent, all the newer antipsychotics will produce fewer acute EPS than the older agents. However, the new 'atypical' agents all have distinct profiles. Olanzapine has a profile similar to that of clozapine, albeit somewhat more potent. Olanzapine, like clozapine, displays a wide margin between the doses predictive of efficacy and those which induce EPS. The compound also substitutes for clozapine in drug discrimination assays and increases punished responding in a conflict paradigm. Quetiapine also has a clozapine-like profile, although it lacks the cholinergic receptor affinity and is relatively weak in most behavioural assays. Quetiapine, like olanzapine, also reverses PCP-induced deficits, and substitutes for clozapine in drug discrimination assays. However, no data are currently available regarding quetiapine's action in anxiolytic tests. Risperidone, sertindole and ziprasidone have profiles of activity different from those of older agents, predominantly due to their 5-HT2a affinity. All these agents possess some properties similar to those of clozapine, but there are some differences: for example, risperidone and sertindole fail to substitute for clozapine in drug discrimination assays and are inactive in classical conflict models of anxiety. It is more difficult to make an accurate assessment of the behavioural profile of ziprasidone, due to a lack of published data. Given the behavioural differences exhibited by animals receiving the new antipsychotic agents, one would predict that these drugs will have distinct clinical profiles. All the agents display activity indicative of agents with a reduced propensity to induce EPS. However, significant differences may be observed in their efficacy against negative and cognitive symptoms. It will be important to assess the clinical profiles of these agents carefully if the predictive value of the pre-clinical 'models' is to be improved.