Laurence A Goldberg
HPE Editorial Consultant
Pharmacogenetics and pharmacogenomics are related but subtly different topics. Pharmacogenetics has been defined as “the study of inter-individual variations in DNA sequence related to drug response”, and pharmacogenomics as “the study of the variability of the expression of individual genes related to disease susceptibility as well as drug response at cellular, tissue, individual or population level”, Geoff Tucker (Professor of Clinical Pharmacology, University of Sheffield, UK) told the audience. Pharmacogenetics is unlikely to lead to personalised treatment in the immediate future, except in the field of cancer chemotherapy; appropriate trials and professional cost analyses will be needed, he added.
Advances in molecular biology and gene sequencing have reinforced the belief that the genomic era will make precise, personalised prescriptions possible. A good example of this progress is the recent launch of the Amplichip CYP450 test – a pharmacogenomic microarray designed to test for variations in 2D6 and 2C19. This is now available for in-vitro testing in both the USA and the EU. There have been some notable success stories, although sometimes the likely benefits have been exaggerated, said Professor Tucker. One success story is that of trastuzamab (Herceptin(®)), which is now known to be effective only in the 15–20% of breast cancer patients who overexpress the human growth factor receptor HER-2. Abacavir represents another positive development; about 50% of white male HIV-positive patients have variations in the HLA-B gene that make adverse reactions to abacavir more likely. However, in general, the response to drug treatment will always be influenced to a greater extent by factors such as compliance, poor prescribing and drug interactions rather than by genetic variation alone. These factors are not sufficiently acknowledged outside clinical pharmacy, said Professor Tucker. About 3% of pharmacodynamic variance might be explained by genetic differences, but 30% variance could be caused by other factors and easily obscure the genetic effect, he said.
In order for a pharmacogenetic test to be cost-effective, a number of conditions need to be satisfied. They are:
- Severe clinical or economic consequences that could be avoided.
- Difficulty in monitoring drug response using current methods.
- Lack of alternative drug with equivalent therapeutic index and price.
- Well-established association between genotype and clinical phenotype.
- Availability of a rapid and inexpensive genetic test.
- Relatively high frequency of the variant gene(s).
Testing for genetic variations in TMTP (thiopurine methyl transferase), the enzyme responsible for the metabolism of azathioprine and 6-mercaptopurine, is a good example of a clinically useful application, said Professor Tucker. Routine testing can identify 90% of the “poor metabolisers” who are at risk of developing signs of toxicity. Direct measurement of red cell enzyme activity might be better for intermediate metabolisers, he added. Another example is metabolism of warfarin by CYP2C9. A dosing algorithm that takes into account the genetic variation in CYP2C9 activity has been developed and tested. The results showed that fewer patients were overdosed when it was used. However, the biggest danger with warfarin is when other drugs are added to established treatment. In this situation, there is no substitute for good clinical monitoring, said Professor Tucker.
Implementation in clinical practice
The implementation of pharmacogenomics in clinical practice is poor at present, and there is much work still to be done, said Henk-Jan Guchelaar (Professor of Clinical Pharmacy, Leiden University Medical Centre, The Netherlands). Currently, the technology is applied in diagnostic testing when there is an unexpected reaction to a drug, and it is used to investigate the side-effects of some drugs. It is not yet used in a preventive way (eg, to guide decisions about treatment), and it is not used to screen populations or to improve the efficacy of drug treatment.
The barriers to implementation include poor acceptance of new technologies and guidelines by clinicians, lack of evidence of improved patient care, overexpectations of diagnostic test criteria, limited usefulness of some tests and lack of data on cost-effectiveness and cost consequences. There is still a need for studies that demonstrate the benefits of pharmacogenetic testing and for cost- effectiveness analyses, he concluded.
Claims of “the right medicine for the right patient at the right dose” may be overoptimistic, according to Sandy Thomas (Director of the Nuffield Council on Bioethics, UK). It is important to discuss the ethical, legal and social issues, and we need to address these concerns now in order to ensure that the benefits of this technology are realised in future, she continued.
Assumptions have to be made about the context in which pharmacogenetics will develop, and this is not easy because there are still many areas of uncertainty.
Although pharmacogenetics is likely to have a significant impact on both clinical trials and treatment, it is likely to be about 20 years before the impact is felt. In the meantime, it will be important to assess claims about what can be achieved and to consider ethical and policy issues associated with pharmacogenetics.
There is a widespread perception that the ethical issues associated with genetic information are fundamentally different from other medical information, but this is not the case. It is, after all, also possible to deduce information about response to medicines from blood tests, and testing for HIV and cholesterol can raise similar issues to genetic testing. It is the nature of the information that is key to considering its implications, not whether it is genetically derived, said Professor Thomas.
The ethical considerations that relate to pharmacogenetics fall into three major areas, namely research and development of medicines, public policy and the use and storage of genetic information.
It is likely that pharmacogenetic tests will be carried out by hospital doctors, GPs and pharmacists, and the results would be stored in medical records. In the UK, the Human Genetics Commission has said that special arrangements for storage of genetic information are not feasible. One interesting aspect of the use of genetic information is the questions that it raises around patient choice. For example, in future, it will be possible for medicines that are dangerous for some patients to be licensed because a genetic test could identify those at risk. In such circumstances, patients may have to take a test in order to receive the medicine. Should patients who refuse to take the test still be allowed to receive the medicines, asked Professor Thomas. If the test were to be part of the licence conditions, then doctors would have to make decisions about off-label prescribing in this context, she noted.
Towards personalised treatments?
Considering the practicality of personalised treatment in the postgenomic era, Kim Brøsen (Professor of Clinical Pharmacology, University of Southern Denmark, Odense) explained that, although drug response was always influenced by genetic factors, it was never determined by a single gene or group of genes alone. The eventual response was always the result of mutually interacting genes, modified by environmental and constitutional factors. Thus, genotyping before treatment could be useful, but only if drug response is determined mainly by a single gene (or a small number of genes) characterised for all clinically relevant single nucleotide polymorphisms (SNPs). In addition, all clinically relevant environmental and constitutional influences need to be known and measurable, both when treatment is started and during treatment.
In humans, there are 57 cytochrome P450 (CYP) genes and 33 pseudogenes that are organised into 18 families and 42 subfamilies. Those that metabolise drugs belong to families 1, 2 and 3. CYP2C9 is responsible for the metabolism of warfarin and is the root cause of most warfarin interactions because of two important SNPs. Individuals with the 2C9*2 and 2C9(*)3 SNPs require smaller doses of warfarin than normal in order to maintain an adequate international normalised ratio (INR). One study has shown that 70% of the population requires an average weekly dose of 39(±15)mg, fewer than 1% would need only 9(±4)mg per week, with the remainder of the population falling in between the two values.
The role of pharmacists
Anthonius de Boer (Director of the School of Pharmacy, University of Utrecht, The Netherlands) described how the training of pharmacists would need to change in order to accommodate the concept of personalised medicines. It will depend on the future roles of pharmacists, he said. They will need to be able to understand and judge the quality of publications of pharmacogenetic research and to determine the clinical relevance of pharmacogenetic research and translate these results into practice. In addition, they will have to be able to communicate this information to doctors and patients. They will need to be able to interpret the results of genetic tests and will have joint responsibility for using genotypic information to design individual treatment regimens. In future, pharmacists will still need to advise physicians about dose adjustments or alternative drugs, and they will need to be aware of and able to communicate the ethical, social and legal problems of genotyping.
Keynote lecture: space research
In the keynote lecture, Marc Heppener (Head of Scientific Research for the International Space Station, European Space Agency, The Netherlands) explained how research in space has numerous applications in healthcare on earth. For the past three years, Europe has been the main user of the international space station for scientific experiments. Experiments in biology, physiology and physics have been undertaken. For example, studies using transgenic mice have looked into the possibility of preventing muscle loss in space, and experiments designed to examine plant growth have established that signal transduction and gene expression are affected by the absence of gravity. These types of experiments have implications for the development of lunar colonies and also for long-distance space travel. Much of the work has important spinoff benefits; for example, microcomputed tomography equipment has been built to enable clear visualisation of bone changes in space. The development of such small, portable devices is of considerable interest for clinical practice on earth, said Dr Heppener.