For many years, following the introduction of chlorpromazine in the 1950s, little progress was made in the discovery of new drugs for schizophrenia (Reynolds, 1992). Dopamine D2 receptor blockade was recognised as the only therapeutic target for antipsychotics (Creese et al, 1976) and the inevitable consequences of striatal blockade remained problematic. However, the strategies and stimuli for discovery of new drugs changed with the introduction of new, atypical antipsychotics in the 1990s. These include clozapine, remoxipride (now withdrawn), olanzapine, risperidone and sertindole (Kerwin & Taylor, 1996). The goal of antipsychotic drug development has always been to widen the therapeutic ratio between efficacy and adverse effects. These new drugs have in the main achieved this. However, which therapeutic targets these drugs employ remains a mystery, and this information is clearly important for future research into more selectively targeted agents.

Antipsychotic agents have been the mainstay in the management of schizophrenia for a number of years. Their therapeutic efficacy is primarily attributed to dopamine receptor antagonism (Creese et al, 1976), leading to a reduction in the positive symptoms of schizophrenia such as paranoia and hallucinations. Unfortunately, they have little effect on the negative symptoms (such as flattened affect, poverty of speech, anhedonia and social withdrawal) or cognitive deficits. The blockade of central dopamine receptors by classical antipsychotic agents also leads to the development of both acute and chronic motor disturbances (extrapyramidal side-effects) (EPS) (Meltzer, 1992).

What mechanisms make an antipsychotic atypical? A common view is that the different therapeutic and side-effect profile of atypical compared to typical antipsychotics is due to their high affinity for specific dopamine and/or 5-HT (serotonin) receptors. Dopamine D4 and 5-HT2a receptors are particularly implicated, though many others have been proposed as well, including dopamine D1 and D3, 5-HT1a, 5-HT2c, 5-HT6, 5-HT7 and α2 adrenergic receptors (for review, see Deutch et al, 1991; Ashby & Wang, 1996; Kinon & Lieberman, 1996).

Since the introduction of chlorpromazine in the 1950s, followed by the discovery (with in vitro receptor binding assays), in the mid-1970s, that antipsychotic drugs block a subtype of dopamine receptor (D2/D2-like) (Creese et al, 1976) and that affinity for these receptors appears to correlate directly with clinical potency for antipsychotics (Peroutka & Synder, 1980), the study of neurotransmitters and receptors has been a major target of schizophrenia research (Owens, 1996). In 1983, the first visualisation, by positron emission tomography (PET), of the binding of D2 dopamine receptors in the brain of a living human subject was reported (Wagner et al, 1983). Following this, the number of research studies using PET and single photon emission tomography (SPET) has increased enormously.

The conventional antipsychotic drugs are associated with a wide range of unwanted effects (Barnes & Guy Edwards, 1993; Guy Edwards & Barnes, 1993). These side-effects represent part of the burden of patients given antipsychotic medication, affect compliance with treatment, and contribute to the prescribing clinician's weigh-ing-up of the risks and benefits of a particular antipsychotic in an individual case. The conventional drugs show differences in their side-effect profiles, particularly with regard to sedation and extrapyramidal problems, and it is probably a reasonable generalisation to state that these have largely governed clinicians' choice. The newer ‘atypical’ antipsychotics (such as clozapine, risperidone, sertindole, olanzapine, quetiapine and amisulpride) are less liable to cause extrapyramidal symptoms than the conventional antipsychotics. However, this advantage is greater for some of the new drugs than for others, and their safety profiles also differ in other key respects. Thus, in terms of side-effects and safety, each of these newer drugs must be evaluated individually.

Cognitive dysfunction has long been recognised as a cardinal feature of schizophrenia, dating back to Kraepelin's description of dementia praecox (1919). Cognitive dysfunction in schizophrenia is seen early in the illness and is initially mild, increasing in severity in some chronic patients. It is recognised as a primary and pervasive deficit that is independent of positive symptoms in that it persists after resolution of acute symptoms. It is not restricted to a subgroup and is seen to a greater or lesser extent in almost all patients suffering from schizophrenia (Goldberg et al, 1990).

The development of atypical antipsychotics has revolutionised the treatment of schizophrenia, as well as providing new insights into its cause. The archetypal atypical antipsychotic is clozapine, which has therapeutic advantages over traditional antipsychotics, as well as a low potential for producing extrapyramidal side-effects (EPS) (Kane et al, 1988). However, clozapine causes agranulocytosis in 1% of patients, and there has consequently been a search for novel atypical antipsychotics, as efficacious as clozapine, but without the need for intensive blood monitoring. There has been much discussion of the definition and characteristics of an atypical antipsychotic drug, and an operational understanding seems to have been agreed upon, that atypical drugs have therapeutic efficacy in treating schizophrenia, without producing EPS (Deutch et al, 1991; Kerwin, 1994).