Pharmacogenomics is the science of genetic influence on an individual’s responses to medication. It has the potential to optimise both drug development and drug therapy decision-making via enhancing the efficiency of clinical trials of new drugs and targeting medicines in clinical practice for improved health and economic outcomes.1
Research that underpins this science has focused on common genetic variations, including single nucleotide polymorphisms (SNPs), genomic insertions and deletions, and genetic copy number variations (CNVs). SNPs appear to be the most frequently inherited sequence variations, but CNVs account for larger regions of the genome. It is not clear which is more crucial in pharmacogenomics but it is likely that both play a role in the phenotypic outcomes and parameters. Research has shown that variations in genes contribute to an individual’s drug sensitivity, resistance and toxicity.2
Clinical practice boasts an increasing number of promising examples of pharmacogenomic tests, with most being applied in specialty care such as oncology. However, there has been little integration into primary care due to the complexity of the fundamental science, as well as clinical, economic and organisational obstacles to the effective delivery of the medicine.
A well known example is overexpression of the HER2 oncoprotein, which can predict the clinical response to the monoclonal antibody trastuzumab. This practice has facilitated the tailoring of therapy to individual breast cancer patients in order to avoid morbidity and costs associated with adverse drug reactions or lack of efficacy.1
One of the best established genotype-phenotype correlations is demonstrated by the thiopurine methyltransferase (TPMT) gene and its effect on thiopurine treatment for acute lymphoblastic leukaemia and immune modulation. TPMT is an enzyme that catalyses the metabolism of thiopurines, therefore regulating the balance between cytotoxic thioguanine nucleotides and inactive metabolites in haematopoietic stem cells (blood cells that give rise to the different blood cell types). Studies have shown that polymorphisms in the TPMT gene can lead to phenotypes that possess a high risk of haematopoietic toxicity after thiopurine therapy.
The hepatic enzyme cytochrome P450 CYP2D6 catalyses the metabolism of many drugs. The genotype or phenotype of CYP2D6 strongly influences the metabolism of codeine, a prodrug that requires bioactivation to morphine. As a consequence, the efficacy and safety of codeine have been shown to be associated with polymorphisms in this gene.
A particular inherited variant in a gene almost exclusively found in patients with south-east Asian ancestry has been strongly associated with increased risk in the serious adverse event Stevens-Johnson syndrome with the use of carbamazepine. The US Food and Drug Administration has included a warning in the drug’s labelling and information to prescreen patients with ancestry in genetically at-risk populations.3
Polymorphisms in genes which encode drug receptors, transporters and drug targets can also influence drug efficacy. VKORC1, a common promoter variant in the molecular target of warfarin, strongly affects the dosage levels required by individual patients. This gene encodes the vitamin K-epoxide reductase protein, the enzyme which breaks down warfarin. Variations in this gene are significantly associated with warfarin sensitivity and reduced dose requirements.
Particular transporters have been identified to possess pharmacogenomic relationships with drug pharmacokinetics or effects. For example, each patient has a variable response to digoxin. Mutations in the ABC1B1 gene has been shown to influence the maximum dose of digoxin that can be given before toxicity is experienced. Additionally, several phenotypes have been associated with polymorphisms in the SLCO1B1 transporter, such as an increased risk of simvastatin-induced myopathy, methotrexate-related gastrointestinal toxicity and disposition of the cyclin-dependent kinase inhibitor flavopiridol.2
Vaccines are a class of medicines that are being viewed under the pharmacogenomic microscope. Variations in genes involved in virus binding and cell entry, antigen recognition, processing and presentation, immune effector cell function and immunoregulation are all crucial in an individual’s ability to propagate a co-ordinated attack against an invading pathogen. Associations in response with genotype or phenotype have been recognised with vaccines against measles, mumps and rubella, influenza, HIV, Hepatitis B and smallpox.4
Oncology is, thus far and potentially will continue to be, the most promising field in pharmacogenomics since tumoural genetic variability is far more significant than that of our constitutional genome, multiplying the situations in which a response to a drug might be genetically determined. Also, in cancer therapeutics, there is a constant influx of novel targeted anticancer drugs released on the market since new technologies facilitate an exponential discovery of potent new tumoural drug targets.
A major example is the investigation into epidermal growth factor receptor pharmacogenomics to identify tumours, particularly non-small cell lung cancers that will respond to EGFR antagonists (tyrosine kinase inhibitors). The anticancer agent gefitinib does not work on tumours without EGFR-activating mutations but a slightly better outcome is achieved with erlotinib when compared with placebo. Both agents increase the median survival of patients with tumours that exhibit EGFR-activating mutations.
The activating mutation of the KRAS gene, a component of the EGF signalling pathway in colon cancers, is associated with resistance to cetuximab and panitumumab. The significance of the presence of KRAS mutations have warranted labelling by the European Medicines Agency that those tumours exhibiting these mutations should not be treated with anti-EGFR monoclonal antibodies.
The recent data collected from a sufficiently powered study of patients has demonstrated application of CYP2D6 pharmacogenetic and pharmacogenomic testing towards individualised endocrine treatment of postmenopausal early breast cancer. In the event of absent or decreased CYP2D6 activity, tamoxifen, a treatment option for oestrogen receptor positive breast cancer in postmenopausal women, is ineffective. Its bioactivation to the active metabolite endoxifen, which inhibits oestrogen receptors, is significantly reduced thus rendering a poor clinical response. Consequently, postmenopausal women with an oestrogen receptor positive breast tumour and decreased or absent CYP2D6 activity should be treated with aromatase inhibitors instead of tamoxifen.5
The processes of drug discovery and development have become increasingly expensive and inefficient. Fewer drugs are being approved and there are increased safety concerns of marketed drugs. Healthcare professionals well understand that drug response is often unpredictable and potentially attributed to individual genetic makeup. Pharmacogenomics, through the use of genomic-based testing, could address some of these concerns to better predict medication response. There is potential economic benefit in the immediate term through avoidance of potentially ineffective unsafe therapies in specific patients and in the longer term because of improved health outcomes.
The public health benefit of using pharmacogenomics to improve the risk-benefit profile of new and existing drugs is potentially significant. Pharmacogenomic testing has been shown to be cost-effective for therapies that are expensive, possess significant risks of serious adverse events, or for drugs that have a poor or highly variable drug response.
There is a growing interest in the pharmaceutical industry to demonstrate evidenced-based recommendations for adopting this testing as a tool to improve the efficiency of drug development over time.1
1 Deverka P, Vernon J, McLeod HL. Economic opportunities and challenges for pharmacogenomics. Annual Review of Pharmacology and Toxicology 2010;50:423–37.
2 Crews KR, Hicks JK, Pui CH. Pharmacogenomics and individualised medicine: translating science into practice. Clinical Pharmacology and Therapeutics 2012;92:467–75.
3 Lesko LJ, Zineh I. DNA, drugs and chariots: on a decade of pharmacogenomics at the US FDA. Pharmacogenomics 2010;11:507–12.
4 Poland GA, Ovsyannikova IG, JacobsonRM. Application of pharmacogenomics to vaccines. Pharmacogenomics 2009;10:837–52.
5 Becquemont L, Alfirevic A, Amstutz U et al. Practical recommendations for pharmacogenomics-based prescriptions: 2010 ESF-UB Conference on Pharmacogenetics and Pharmacogenomics. Pharmacogenomics 2011;12:113–24.