CYP2D6 genotyping in daily psychiatric practice

It has been suggested that, on average, only 30% of all patients respond satisfactorily to drug treatment, 30% of all patients do not show efficacy, 10% only show adverse effects and 30% are non-compliant. Furthermore, recently it has been shown that 5.6% of all hospital admissions in The Netherlands are drug related.(1) These figures show the importance of research aimed at identifying patient- and drug-related factors that predict and explain treatment failure or unacceptable adverse effects in the individual patient.

Although other potential patient-related factors such as age, smoking behaviour, renal and liver function could also contribute to the variability in drug response, it has been suggested that pharmacogenetics can provide great benefits to public health and holds the promise of tailor-made or personalised medicine.

Pharmacogenetics is defined as the research area investigating whether, and to what extent, genetic variation can explain and predict the response to drugs of individual patients.

Psychiatric practice is suggested to be one of the first medical disciplines for the clinical use of pharmacogenetics testing. Antipsychotic and antidepressant drugs are used by the majority of psychiatric patients but the response to these drugs is variable. Approximately 20–40% of the psychiatric patients do not respond satisfactorily to pharmacotherapy. In psychiatric pharmacotherapy, there is usually a lag time of several weeks before the balance between the therapeutic response and the prevalence of adverse events can be evaluated.

This lag time, together with the high prevalence of clinically relevant adverse events during treatment with antipsychotic drugs (extrapyramidal syndromes, weight gain, lipid abnormalities, disturbed glucose homeostasis) and antidepressant drugs (sexual dysfunction, bleeding, sedation, gastrointestinal complaints), put psychiatric patients at risk of discontinuing pharmacotherapy because of unsatisfactory response. These complications make pharmacogenetics a promising pharmacological tool in optimising psychiatric pharmacotherapy by genotyping psychiatric patients before starting pharmacotherapy. If genotype information of clinically relevant polymorphisms were available before starting pharmacotherapy, an individualised advice for the choice of drug and dosage might be possible. This information could improve the response to antidepressant and antipsychotic drugs and prevent the occurrence of unacceptable adverse effects.(2,3)

At this moment, CYP2D6 genotyping is probably the most widely accepted application of pharmacogenetic testing in psychiatric practice. Approximately 5–10% of the Caucasian population can be classified as a poor metaboliser (PM) by lacking CYP2D6 activity and 1–10 % as an ultrarapid metaboliser (UM) by gene duplication resulting in high enzyme activity. Subjects with two active alleles are classified as extensive metabolisers (EM), and carriers of one active and one deficient allele are sometimes classified as intermediate metabolisers (IM). This group of IMs is expected to have subpopulation-specific clearance between EMs and PMs. As most antipsychotic and antidepressant drugs are at least partly metabolised by CYP2D6, this might implicate a high risk of unsatisfactory response to these drugs in up to 20% of psychiatric patients.

Figure 1 presents a classic example of the association between CYP2D6 genotype and dosage requirements of the antidepressant drug nortriptyline. The figure shows that most patients (EMs for CYP2D6) need a dose of 100–300mg of nortriptyline for a plasma concentration within the therapeutic range. However, PMs for CYP2D6 only need 20–30mg and UMs for CYP2D6 300–500mg nortriptyline in order to achieve a plasma concentration within the therapeutic range. Without knowledge of the CYP2D6 genotype, PMs for CYP2D6 will easily be overdosed (increased risk of adverse effects) and UMs for CYP2D6 will easily be under-dosed (increased risk of non-response). Together with PMs for CYP2D6, IMs for CYP2D6 might need a lower dose compared with EMs for CYP2D6 (as shown in Figure 1, as the group of patients requiring approximately 75–100mg nortriptyline for an adequate plasma concentration). As yet, prescribers in general do not have information on the CYP2D6 genotype of their patients, and drug prescriptions are therefore not individualised to the patient with respect to their CYP2D6 genotype.(4,5)

At first glance, it seems reasonable that we would like to know the exact CYP2D6 genotype of an individual patient. However, more than 80 mutations have been described in the gene coding for CYP2D6. An important question one might ask is how many mutations have to be determined in a CYP2D6 genotyping procedure in order to adequately predict the clinical phenotype. It is, for example, possible to determine 33 mutations with the microarray-based gene chip technology approved by the US Food and Drug Administration for clinical use (Amplichip®). The determination of these 33 mutations makes it possible to genotype the world’s population accurately for over 99%. One disadvantage of this chip is the high costs that accompany its use. These high costs limit cost effectiveness and thus the implementation of CYP2D6 genotyping in daily clinical practice.

Is it necessary to genotype so many mutations for a proper prediction of a patients’ clinical phenotype?

It has been shown that determination of five of these 33 CYP2D6 mutations (*3,*4,*6,*7,*8), as well as the presence of duplication, determines 98.5% of all phenotypes correctly in a Caucasian population. Inclusion of the rather frequent *5 (frequency 2–7%) and *41(frequency 8–20%) mutation will probably increase the reliability to over 99%. Because the *7 and *8 mutations have a low frequency of occurrence, the determination of a limited set of mutations (*3,*4,*5,*6,*41) at a cost of <€100 is more likely to be cost effective in daily clinical practice.

The disadvantage of this Caucasian set of mutations is the ethnic diversity in West European countries, resulting in a lower prediction of the correct phenotype when all patients would be genotyped with the Caucasian set of mutations. The most frequent alleles in a Caucasian population are the mutations *3 and *4, while the mutation *10 is more prevalent in Asian populations and the mutation *17 is more prevalent in African–American populations. Adding these two non-Caucasian mutations (*10,*17) will decrease the risk of missing a mutation in a patient of Asian or African descent. In conclusion, testing for seven mutations (*3,*4,*5,*6,*10,*17,*41) and the presence of duplications makes it possible to predict the correct phenotype in around 98-99% of all patients. Before implementing CYP2D6 genotyping in daily clinical practice, it is essential to know what mutations a specific laboratory determines.

The above shows that the classification of a CYP2D6 clinical phenotype from CYP2D6 genotyping results is not straightforward, and guidance of treating physicians with the interpretation of genotyping test results is therefore essential for a proper interpretation of CYP2D6 genotyping for patient care. Pharmacists might carry out the guidance of treating physicians because they possess the required pharmacological knowledge of drugs and they have a good overview of the medication history of genotyped patients. The pharmacist and the treating physician can evaluate the genotyping result together with other patients factors (environmental factors) like concomitantly used medications and smoking behaviour. The genotyping result can only be interpreted together with the environmental (patient) factors in order to interpret the clinical phenotype of an individual patient.(6,7)

CYP2D6 genotype provides information about the metabolic capacity to metabolise drugs metabolised by CYP2D6. Ideally, CYP2D6 genotyping will be applied only to drugs and active moieties that are eliminated or formed primarily by CYP2D6. However, what is primarily metabolised by CYP2D6? There are available data on the drugs that are metabolised by CYP2D6. However, these data do not always make a distinction between drugs primarily metabolised by CYP2D6 and drugs partly metabolised by CYP2D6. Therefore, debate will always remain regarding the question of which drugs for CYP2D6 genotyping has to be applied. To prevent these discussions, it is essential that evidence-based recommendations for the choice of drug and dosage for different CYP2D6 genotypes become available.

Therefore, in 2005, the Royal Dutch Association for the Advancement of Pharmacy (KNMP) established the Pharmacogenetics Working Group. In this multidisciplinary working group, (clinical) pharmacists, clinical pharmacologists, clinical chemists and physicians are represented. The primary goal of this group is to develop pharmacogenetics-based therapeutic (dose) recommendations and to integrate those into computerised systems for drug prescribing and automated medication surveillance. At present, 25 drugs at least partly metabolised by CYP2D6 have been evaluated and therapeutic (dose) recommendations have been specified for 76% (n=19) of these drugs. These recommendations provide guidelines for the application of CYP2D6 genotyping to the appropriate drugs and present a major step forward in the implementation of pharmacogenetics testing in daily clinical practice. However, it is essential that these guidelines be implemented by professionals with knowledge of CYP2D6 genotyping and the drugs mentioned in the guidelines.(8)

There are several challenges for making the implementation of pharmacogenetic testing in psychiatric practice a success. First, at this moment, prospective studies investigating the cost effectiveness of genotyping in daily clinical practice are not available. The lack of such studies hampers the implementation of pharmacogenetics in daily clinical practice. Without clear-cut studies showing clinical relevance of pharmacogenetic testing, the adoption by the medical community will remain slow. However, in psychiatry, the proof of concept for pharmacokinetic polymorphisms is widely accepted.

Therefore, in The Netherlands, genotyping of CYP2D6 has already been implemented in daily psychiatric practice in a number of mental healthcare institutions. CYP2D6 genotyping is mainly used to investigate patients with an expected compromised metabolism by CYP2D6, for example, on the basis of unexpected results of therapeutic drug monitoring. Second, implementation of CYP2D6 genotyping in daily psychiatric practice was complicated because recommendations for the choice of drug and the dosage were not available. The guidelines provided the Pharmacogenetic Working Group of the Royal Dutch Association for the Advancement of Pharmacy are a major step forwards for the implementation of pharmacogenetics in daily clinical practice.

The recommendations of the Pharmacogenetics Working Group can support the third challenge for implementation of pharmacogenetics testing in daily clinical practice; that is, knowledge and education. Efforts should be made to educate medical communities to interpret and understand the advantages and limitations of pharmacogenetic testing. Therapeutic (dose) recommendations according to a patient’s genotype, and even drug labels with pharmacogenetics information, are meaningless if we fail to educate and train medical communities to interpret pharmacogenetics information appropriately.

Pharmacists could become a key players in recommending pharmacogenetic testing and interpreting pharmacogenetic testing results. To make this happen, it is essential that pharmacy schools start training future pharmacists in the interpretation, management, application and delivery of pharmacogenetic information. These educational efforts could result in pharmacogenetics becoming the pharmaceutical care of the future. Finally, interpreting pharmacogenetic results in daily clinical practice demands a considerable amount of time. Because of the lack of knowledge of most physicians in applying genotyping results to their patients appropriately, it is essential that drug prescribing to genotyped patients be guided by someone who has an extensive knowledge of pharmacogenetics.

 This will become increasingly important as the field expands to pharmacodynamic polymorphisms of drug-targeted receptors and drug transporters, in addition to pharmacokinetic polymorphisms, such as CYP2D6. The amount of time needed for pharmacogenetic counselling demands that the pharmacy profession decide whether these activities are considered a primary ‘front office’ activity for the profession. This might imply that other, more traditional activities, such as logistics and manufacturing, are becoming less important.

• CYP2D6 genotyping is probably the most widely accepted application of pharmacogenetic testing in psychiatric practice.
• Testing for seven mutations (*3,*4,*5,*6,*10,*17,*41) and the presence of duplications makes it possible to predict the correct CYP2D6 phenotype in approximately 98–99% of all patients;
• Guidance for treating physicians with the interpretation of genotyping test results is essential for a proper interpretation of CYP2D6 genotyping for patient care.
• In The Netherlands, pharmacogenetics-based therapeutic (dose) recommendations integrated into computerised systems for drug prescribing and automated medication surveillance are available.
• Pharmacists could become key players in recommending pharmacogenetic testing and interpreting pharmacogenetic testing results. To make this happen, it is essential that pharmacy schools start training future pharmacists in the management, application and delivery of pharmacogenetic information.

1. Leendertse AJ et al; HARM Study Group. Frequency of and risk factors for preventable medication-related hospital admissions in the Netherlands. Arch Intern Med 2008;168(17):1890–6.
2. Fava M. Management of nonresponse and intolerance: switching strategies. J Clin Psychiatry 2000;61(suppl 2):10–12.
3. Hugenholtz GW et al. Reasons for switching between antipsychotics in daily clinical practice. Pharmacopsychiatry 2005;38(3):122–4.
4. Mulder H et al. Prevalence of patients using drugs metabolized by cytochrome P450-2D6 in different patient populations: a cross-sectional study. Ann Pharmacother 2007;41(3):408–13.
5. Bertilsson L, Dahl ML, Tybring G. Pharmacogenetics of antidepressants. Clinical aspects. Acta Psychiatr Scand 1997;96(suppl 391):14–21.
6. Bradford LD. CYP2D6 allele frequency in European Caucasians, Asians, Africans and their descendants. Pharmacogenomics 2002;3(2):229–43.
7. Heller T et al. Amplichip CYP450 GeneChip: a new gene chip that allows rapid and accurate CYP2D6 genotyping. Ther Drug Monit 2006;28(5):673–7.
8. Swen JJ et al. Pharmacogenetics: from bench to byte; an update of guidelines. Clin Pharmacol Ther 2011;89(5):662–72.