First-generation antipsychotics (FGAs), resembling haloperidol and chlorpromazine, display high affinity for dopamine receptors, whereas second-generation antipsychotics (SGA) display preferential affinity for dopamine and serotonin receptors, amongst others. Numerous studies have investigated genetic variants in targeted neurotransmitters and transporters with varying results (see Table 1.2). It is important to note that Table 1.2 summarises only significant findings and that the many non-significant reports published to date are not included. However, given the difficulty of obtaining clinical samples for pharmacogenetic studies, and the complexity of response definition, when an association report is replicated in independent clinical settings, it may constitute a true finding.

The most replicated findings indicate that dopamine and serotonin genetic variants are involved in both the level of efficacy and the risk of adverse reactions. In particular, genetic variants in dopamine type 2 (D2), dopamine type 3 (D3) and serotonin type 2A (5-HT2A) are the most frequently associated with treatment efficacy [8, 14]. Dopaminergic variants are associated with response to FGA and SGA, whereas serotonergic variants are more likely to be associated with the level of efficacy of SGA, reflecting perhaps their pharmacological profiles. In general, those genetic variants associated with lower receptor expression (e.g. D2 -141-Del) or altered functioning (e.g. 5-HT2A 452Tyr) are associated with poorer antipsychotic efficacy [12, 14], indicating that the harbouring receptors are implicated, at least partially, in the antipsychotic mechanism of action. Genetic variants in dopamine receptors type 1 (D1) and 4 (D4) and in serotonin receptors type 1, 2C, 3 and 6 (5-HT1, 5-HT2C, 5-HT3 and 5-HT6, respectively) have also been associated with treatment efficacy, although the amount of evidence is limited [13, 14, 74, 76]. Their contribution to treatment efficacy needs clarification. Existing reports of other targeted receptors such as histamine type 2 (H2), histamine type 3 (H3) and histamine type 4 (H4) [68, 87, 88] require replication in independent cohorts to confirm their validity. Several neurotransmitter transporter variants have also been linked to treatment variability. The most significant finding indicates that the serotonin transporter (5-HTT) gene harbours polymorphic variants that influence the level of efficacy of the SGA clozapine, olanzapine and risperidone [10, 17, 20, 53, 86], providing further evidence of the important role of the serotonergic system in antipsychotic activity. Interestingly, studies have also linked variants in the dopamine transporter (DAT1) gene [17], but this finding needs confirmation.

Notwithstanding the numerous studies failing to replicate these findings, taken globally, these results indicate that the dopamine and serotonin systems play a major role in the therapeutic activity of currently available antipsychotics. This information may be useful to select patients who can benefit from available medications in a personalised manner. However, pharmacogenetic studies indicate that other targeted neurotransmitter pathways such as the adrenergic, glutamatergic, histaminic and muscarinic systems do not play a major role in the mechanism of action of current drugs. These receptor systems may be valid therapeutic targets, and novel drugs targeting glutamatergic receptors and other genes directly linked to risk of schizophrenia are under investigation [44].

The cytochrome P450 (CYP) group of metabolic enzymes are responsible for the biotransformation and clearance of more than 80 % of drugs, including antipsychotic medications. Table 1.1 summarises the main metabolic pathways of commonly used antipsychotics. It is well known that the genes encoding for these hepatic enzymes may harbour functional polymorphisms that render the enzymes inactive or poor metabolisers (PM), or induce higher metabolic rates (ultrarapid metabolisers, UM). These polymorphisms have consistently been associated with drug plasma concentrations [18, 23, 30, 81, 91], with individuals with one or more PM copies presenting higher plasma levels of substrate drugs than normal metabolisers (EM) and individuals with UM presenting lower plasma concentrations of drug metabolites. Alterations in genes controlling antipsychotics’ clearance can have a direct impact on treatment efficacy. Low drug plasma levels caused by the presence of UMs may lead to poor response. Additionally, the presence of PMs and, in some cases, UMs has been associated with toxic reactions, leading to side effects. This, in turn, may lead to poor compliance and lack of response. Several studies have reported associations between CYP polymorphisms and variability in the response to antidepressant medications [63, 77, 83]. However, little evidence relating CYP functional variants and level of antipsychotic efficacy can be found in the literature. CYP1A2 UM variants have been associated with low plasma levels and lack of therapeutic response to clozapine [33], whereas CYP2D6 and CYP3A4 variants were associated with response to SGA [23, 32]. The singularity of these findings and the moderate sample size warrants further investigation. The availability of different metabolic pathways may serve to explain the low impact that functional metabolic polymorphisms seem to have on antipsychotic efficacy.

Aside from targeted receptors and hepatic enzymes, a number of genes including metabolic enzymes, transporters and genes directly linked to schizophrenia have been associated to treatment response. The most biologically relevant findings associate polymorphisms in catechol-o-methyltransferase (COMT) and multidrug resistance 1 (MDR1) genes with antipsychotic response. COMT is an enzyme involved in the metabolic degradation of catecholamines, including dopamine catabolism, and is located in a region linked to mental disorders. The COMT gene harbours a well-investigated functional polymorphism, Val158Met. The COMT Met158 variant displays lower enzymatic activity which leads to higher dopamine availability. Interestingly, this variant is associated with higher improvement in response to SGA treatments [7, 16, 71, 89, 97], suggesting that control of dopamine activity is part of their antipsychotic action. MDR1, also known as ABCB1, is a transmembrane protein that regulates blood-brain barrier transport. Genetic variants in this transporter have been associated with the level of efficacy of several antipsychotic drugs [19, 62, 73, 92, 93] and may reflect drug availability in the brain. BDNF is another protein which has been linked to schizophrenia risk and recent studies have also been related to response levels [56, 95]. Finally, the evidence supporting the association between glutamate metabotropic receptor type 3 (GRM3), methylenetetrahydrofolate reductase (MTHFR) and regulator of G-protein signalling 4 (RGS4) genetic variants with level of efficacy is limited and needs replication [48, 58].

In summary, several genes coding for targeted receptors, transporters and enzymes have been shown to contain genetic variants that significantly influence the level of antipsychotic efficacy. However, the magnitude of these associations is moderate and therefore their clinical value limited. Single individual genes or variants cannot be used for the personalisation of antipsychotic treatment, given the low genetic effects observed. Attempts at combining information in several genes, and with clinical and environmental data, have not produced the clear results required for the application of this information into clinical practice [10, 14, 57]. Standardisation of treatment response definition and further studies including detailed clinical and environmental data are required to move the field forward before using this knowledge for the prediction of the level of efficacy of currently available antipsychotic treatments.

Whereas research protocols for the measurement of clinical efficacy are still in need of standardisation, side effects constitute less complex phenotypes which are relatively easier to determine (e.g. amount of weight gain, presence/absence of movement disorders). Given the severity of the side effects associated with antipsychotic treatment, it is not surprising that in recent years relatively more effort has been put into identifying side-effect biomarkers. As a result, pharmacogenetic studies have been relatively more successful in finding genetic factors contributing to adverse reactions than in finding response-related variants. As in the case of level of efficacy, genes involved in pharmacodynamics and pharmacokinetic processes, and genes previously linked with schizophrenia risk, have been related to a variety of antipsychotic-induced side effects. The most significant findings are summarised in the following subsections.

Genetic advances in the field of psychiatry have been boosted by the development of high-throughput methodologies such as GWAS. Although genomic strategies are relatively new in psychiatry, recent GWAS have yielded increasing and unequivocal evidence for common SNPs contributing to schizophrenia risk [44]. Unfortunately, only a limited number of GWAS have been conducted in order to explore the genetic variants involved in treatment failure or success. Difficulties in obtaining large enough samples with detailed information on response phenotypes are one of the main explanations for the lack of studies [13].