Pharmacogenomics and Drug Prescribing (Genetically Guided Prescribing) in Psychiatry – The Current State of Evidence

PSYCHIATRIC GENETICS

While the genetic architecture of many psychiatric disorders is slowly being identified—primarily through family and heritability studies – pharmacogenomic testing instead has focussed on the involvement of single nucleotide polymorphisms (SNPs) and copy number variants (CNVs) in genes encoding the human cytochrome P450 (CYP) drug metabolising enzymes [Foulds et al. 2016] and the blood-brain-barrier. [Singh et al. 2014], [Eap et al. 2021]

Read more on the epigenetics in psychiatric disorders.

In particular, for psychotropics, the CYP2C19, CYP2D6, CYP2B6, CYP3A4 enzymes and ABC family of the transporter (blood-brain-barrier) are highly polymorphic. They are responsible for the metabolism and CNS bioavailability of more than 50% of medications currently in clinical use. [Wang et al. 2009]; [Foulds et al. 2016]; [Eap et al. 2021]

For instance, bupropion is largely a prodrug that requires activation by CYP2B6, and pharmacogenomic variation of CYP2B6 significantly impacts optimal dosing. [Eap et al. 2021]

Another example of the interindividual variability of SNPs affecting the metabolic capacity of CYP2C19 is with the administration of citalopram and escitalopram. [Mrazek et al 2011]; [Ji et al 2014], [Hicks et al 2015]

Metabolism by CYP2C19 reduces drug activity. Therefore, CYP2C19 ultra-rapid metabolisers have low plasma drug concentrations (i.e. impacts remission), whereas CYP2C19 poor metabolisers have higher plasma concentrations (i.e. impacts tolerance, and potentially safety – QTc). [Eap et al. 2021]

Thus, pharmacokinetic pharmacogenetic genes can influence optimal dosing. Hepatic and blood-brain-barrier variants, when considered together, appear to provide the best dose predictive dose guidance reports. [Singh, 2015], [Bousman et al. 2018]

CYP ENZYMES AND DRUG METABOLISM - KEY POINTS

The enzyme superfamily, cytochrome P450 (CYP450), plays a key role in the metabolism of various drugs and xenobiotics. [Zanger and Schwab, 2013]

The CYP450 enzyme family consists of over 50 different isoenzymes and is the most important group of phase I drug metabolizers.

It is important to note that each CYP isoenzyme has a broad range of substrates, and for many substrates, a range of isoenzymes can metabolize them. However, CYP isoenzyme activity is heavily influenced by genetic polymorphisms that can increase or decrease activity or cause a particular isoenzyme to have no activity.

Genetic variants have been classified into four phenotypic classes: [Ingelman-Sundberg et al., 2007]

• Poor metabolizers (PM) – Lack a completely functional copy of a gene for one CYP450 isoenzyme.
• Intermediate metabolizers (IM) – Have a functional and a defective copy of a CYP450 isoenzyme gene or instead may have two partially defective copies.
• Extensive metabolizes (EM) – Two active genes for a particular CYP450 isoenzyme.
• Ultra-rapid metabolizers (UM) – More than two active genes for a particular CYP450 isoenzyme.

CYP2D6:

Metaboliser classification: 

Ultra-rapid metabolisers:

• Estimated to be present in up to 28% of North Africans, Ethiopians, and Arabs
• Up to 10% in Caucasians
• 3% in African Americans
• Up to 1% in Hispanics, Chinese, and Japanese

Poor metabolisers:

• More commonly found in European Caucasians and their descendants

Intermediate metabolisers:

• Sub-Saharan African and East Asian populations

Normal metabolisers:

• 43–67% of the population.

Medications metabolised by CYP2D6.

• CYP2D6 mainly metabolises risperidone to 9-OH-Risperidone.
• Other psychiatric medications predominantly metabolised by CYP2D6 include Aripiprazole and Brexpiprazole, Clomipramine, Fluoxetine, Olanzapine, Venlafaxine, Vortioxetine, Nortriptyline and Zuclopenthixol.
• Codeine is metabolised through this pathway and converted to morphine; hence, greater intoxication can occur in North Africans.
• It also plays a role in many other drugs, including antiemetics, antiarrhythmics, beta-blockers, tamoxifen, and atomoxetine.
• Tamoxifen used in the treatment of breast cancer is metabolised by CYP2D6. There is consistent evidence that paroxetine and fluoxetine have a large effect on the metabolism of tamoxifen and should not be used. Indirect evidence indicates that bupropion may also have a large effect on the metabolism of tamoxifen. Venlafaxine has little or no effect on the metabolism of tamoxifen and may be considered the safest choice of antidepressants. Desvenlafaxine is not metabolized by the P450 system and may consequently be another option. Mirtazapine has not been extensively studied, but existing research suggests a minimal effect on CYP2D6. [Desmarais and Looper, 2009]

Inhibitors of CYP2D6: Fluoxetine, Paroxetine, Bupropion, Duloxetine

CYP1A2: 

• This enzyme is involved in the metabolism of clozapine and olanzapine mainly.
• Fluvoxamine is a potent inhibitor along with grapefruit juice.
• Caffeine increases levels of clozapine through competitive inhibition of CYP1A2.
• Tobacco smoke contains polycyclic aromatic hydrocarbons that induce CYP1A2 and can thus reduce levels of many medications by up to 50% (Duloxetine, Clozapine, Olanzapine, Benzodiazepines and TCA’s)
• Ciprofloxacin also inhibits the enzyme and can cause an increase in levels of clozapine.

CYP 3A4: 

• This enzyme is involved in the metabolism of quetiapine and ziprasidone.
• It is also responsible for 35% of the metabolism of clozapine.
• It is induced by carbamazepine, phenytoin, barbiturates and rifampicin. It is inhibited by erythromycin, ketoconazole and protease inhibitors.

CYP2C19:[Sistonen J et al, 2007], [Desta Z et al. 2002]

• CYP2C19 is involved in the metabolism of many TCAs (e.g. amitriptyline, imipramine, and clomipramine) and SSRIs (e.g. citalopram, escitalopram and sertraline) as well as some benzodiazepines (e.g. diazepam and clobazam).
• Found to be highly polymorphic (over 35 variants) with large ethnic differences for different alleles such as the loss of function alleles, CYP2C19*2 and *3.
• CYP2C19*2 is present in 15% of Africans, 29-35% of Asians, 12-15% Caucasians, and 61% of Oceanians. In contrast, the CYP2C19*3 allele was found mainly in Asian populations (5-9%).

GENETIC TESTING IN PSYCHIATRIC CARE

Through genome-wide association studies (GWAS), the contribution of common and rare genetic variants, as well as the integration of knowledge on epigenetic events, has provided substantial evidence of the relationship between genetic changes, their interaction with the environment, and a patient’s risk of developing a psychiatric disorder.

Overall, risk scores and polygene pathway analyses of GWAS data provides clinically useful information on the brain and behavioural phenotypes.

As such, there is a clear effort to complement the considerable progress in genetic linkage studies with the real-world application of diagnostic genetic testing to individualise drug prescriptions.

However, genetic biomarkers to guide treatment (pharmacogenetics) is much further along the translational utility path than for diagnosis.

However, although there is good evidence that genetic testing can inform clinical decision making, there has been some controversy as the evidence base evolves:

• Analytic Validity – Does the genetic test accurately identify the variants of concern? Regulators require minimum ‘read depth’ and variant coverage to help ensure quality genotyping.
• Clinical Validity – Is there sufficient evidence of a correlation between the genetic variant and the psychiatric disorder? The Clinical Pharmacogenetics Implementation Consortium (CPIC) is an NIH funded body curating candidate gene association studies. [Hicks et al. 2015]
• Clinical Utility – Is the result replicable, and will it improve patient outcomes by identifying clinically useful genetic markers that can guide optimal treatment? Meta-analysis of RCTs (comparing genetically guided to unguided care) has suggested clinical utility, but large independent studies are in progress currently to confirm these findings. [Bouseman et al 2018], [Sebastian 2021]

To answer these criticisms, the International Society of Psychiatric Genetics (ISPG), which was established in 1992, aims to promote and facilitate education and research in the genetics of psychiatric disorders. [ISPG 2019]

The committee recognises that the implementation of pharmacogenomics requires education and training and the development of approved guidelines.

In 2019, the ISPG updated their genetic testing statement in 2019, which provides guidance on how pharmacogenomic evidence and clinical genetic testing can or may be successfully used to screen and identify individuals at risk and guide treatment. [ISPG 2019]

As well as the promotion of educational programs for trainees and mental health professionals, the ISPG outlined the following:

• Pharmacogenomic testing should only be used as a decision-support tool to assist in the thoughtful implementation of good clinical care.
• For example, the identification of HLA alleles (HLA-A and HLA-B) testing before carbamazepine and oxcarbazepine use can help prevent severe cutaneous adverse drug reactions.
• The identification of common genetic variants that correlate with a major adult psychiatric disorder should not be deemed as having sufficient clinical value because, individually, some variants have little to no clinical significance.
• Alternatively, the use of polygenic risk scores or risk allele burden testing, which provide a cumulative genetic risk, may have more clinical value. However, these are currently not recommended for clinical use.
• As such, direct-to-consumer genetic tests for health and risk assessment are routinely wrong or misinterpreted. More research is needed before these tests can inform healthy individuals about the risk of developing a psychiatric disorder or for treatment purposes in patients.
• The detection of de novo or inherited copy number variants (CNV) in adult-onset psychiatric disorders may have some clinical value. However, these are often only applicable for the diagnosis of a rare or atypical psychiatric condition.

The ISPG also state that familial risk counselling and consultation with a medical geneticist can prove useful. This is because their professional expertise can help interpret and understand the health implications associated with a genetic test result.

Furthermore, the ISPG rigorously state that genetic testing results should remain private and that patient autonomy should be respected regarding incidental or secondary findings.

Other practice guidelines have emerged more recently as pharmacogenetics becomes more widely embraced. [Hicks et al 2015], [van Westrhenen et al 2021]

In short, these guidelines recommend pharmacogenetic testing when there are difficulties with efficacy and tolerability with psychotropics. The FDA now also endorses the use of such guidance and has there  guidance based on medication filings upon initial registration with the FDA: Table of Pharmacogenomic Biomarkers in Drug Labeling | FDA

At this stage, HLA-B testing to gauge the risk of Steven Johnson Syndrome before initiating carbamazepine in patients of Asian ethnicity is the only mandated pharmacogenetic test, with other guidance reports considered optional – to augment prescribers’ decisions.

A statistical association between HLA-B*1502 and lamotrigine-induced SJS/TEN in Han Chinese subjects exists. [Zeng et al., 2014]

Importantly, all guidelines recommend that clinical acumen can over-ride pharmacogenetic reports given limitations in full knowledge of relevant variants at this stage of the field’s evolution.

At present, CYP genotyping is integrated as follows : [Swen J et al., 2011],[Crettol, 2014]

• Define environmental factors such as concomitant medications that are inhibitors or inducers of different CYP isoenzymes.
• Define all personal factors such as age, sex, and disease state.
• Therapeutic drug monitoring using drug selection and dosing based on the physician’s experience and standard laboratory test results (liver, renal, blood, ECG, etc.)
• In non-response or ADRs, phenotype and genotype testing can be used to present data on personalizing drug and dose selection.

SUMMARY

Through the correlation of genotype with clinical phenotype, pharmacogenomic profiling aims to provide personalised therapy with improved efficacy and reduce adverse drug reactions.

Pharmacokinetic pharmacogenetic tests combing hepatic and blood-brain-barrier variants are showing the most promise for clinical utility.

These tools are designed to augment prescribers deliberations rather than replace them, with clinical acumen still central in decision making.

Pharmacogenetics will likely enhance shared-decision making and risk-benefit analyses in medication choices.

Learn more: 

1. DNA Interpretation- A Complex Process – Genetically Guided Prescribing By A/Prof Ajeet Singh
2. Evolution of Pharmacogenetic Prescribing – Genetically Guided Prescribing By A/Prof Ajeet Singh
3. The Basics of Pharmacogenetics – Genetically Guided Prescribing by A/Prof Ajeet Singh
4. Genetics, Epigenetics, and Pharmacogenomics – Genetically Guided Prescribing by A/Prof Ajeet Singh