Precision medicine: are we there?

Implementation of precision medicine requires a multidisciplinary and systematic approach

In his State of the Union address on 20 January 2015, United States President Barack Obama announced a new initiative in precision medicine, which aims to give “access to the personalised information we need to keep ourselves and our families healthier”.1 So, what is precision medicine? Previously referred to as personalised medicine, it can be defined as the correlation of innate and external factors at an individual level, to better understand the pattern of disease and its impact on the individual, and thus to tailor prevention, intervention and treatment. Precision medicine thus combines genomic and epigenomic data with environmental exposure and lifestyle factors. It has the potential not only to improve health outcomes but to save money by better targeting health interventions to those individuals most likely to benefit.

This research initiative in the US will provide funding through the National Institutes of Health and other partners, initially to cancer medicine. The longer term hope is to create better understanding of genomics, molecular biology and bioinformatics in a bid to improve health for all, not only people suffering with cancer.2

Genetics, the study of heredity, investigates the structure and function of a single gene. We are currently in transition from genetics to genomics, the study of all of an individual’s genes, their molecular structure and function, and their interrelationships and architecture. We are thus refocusing from the utility of single genes in monogenic disorders to the potential of genomics in many complex disorders.

Sequencing of the whole human genome, all 3.3 billion base pairs, has become cheaper, quicker and easier. In the 100 000 Genomes Project, Genomics England is sequencing the whole genome of 100 000 individuals with common cancers or rare inherited diseases, to facilitate the incorporation of genomic medicine into the National Health Service (http://www.genomicsengland.co.uk/the-100 000-genomes-project). Similarly, in the Melbourne Genomics Health Alliance, seven health and research organisations are working in partnership to incorporate genomics into health care and assess the health economics argument for genomics (http://www.melbournegenomics.org.au). This has been initiated with pilot projects in areas such as epilepsy, rare childhood diseases and colon cancer.

While genomics might underpin precision medicine, it is only part of the picture. The bioinformatic capacity to analyse the data will allow interpretation of the pinpoint accuracy of current genetic technology. The impetus needs to shift from technology to informatics in clinical practice and improving health outcomes. Benefits to patients will only occur when their clinical details can be linked to their specific genomic data and their treatment altered accordingly.

Implementation of precision medicine will require convergence of disciplines, involving not only clinicians and scientists, but also mathematicians, engineers and philosophers — rather than the siloed approach of previous decades. Governments internationally are now recognising the importance of this convergence.1,3 In Australia, we have much preliminary data already. The University of Melbourne and Cancer Council Victoria host PEDIGREE (Pathology, Epidemiology, DNA, Informatics & Genetics: a Research Enabling Enterprise) (http://www.cancervic.org.au/research/epidemiology/pedigree). PEDIGREE is a resource of 100 000 people, 20 000 families with cancer, 1 million biospecimens, data, researchers and community representatives, evolving through collaboration over two decades to enable studies of the genetic and environmental factors associated with the risk and prognosis of some of the common cancers that affect Australians. The aim is to develop risk management strategies which can be applied at a population level to those who have a genetic predisposition to these cancers and ultimately prevent the cancers from developing.

It is intuitive that a precise therapy, based on the biology of disease, would lead to more effective treatment. And it is intuitive that even if the precise therapy is expensive, the elimination of waste will lead to longer-term cost savings. But to ensure that this is the case, clinical trials must occur in parallel with economic modelling and a robust economic argument must be made to justify the use of precision medicine. Initial support for precision medicine has come through a few effective targeted treatments for cancer. The best known of these are trastuzumab, a monoclonal antibody therapy for HER2-positive breast cancer, and imatinib, a tyrosine kinase inhibitor, for the treatment of chronic myeloid leukaemia. Ward has cautioned that, apart from a few well known exemplars, little has changed in the treatment of cancer and that the “[s]ubstantive benefits of personalised medicine continue to elude us”.3 In the short term, it is the primary objective of the Obama initiative to tackle this problem.1

As a practical example of precision medicine, pharmacogenomics — the use of an individual’s genome to optimise medication prescribing — has promising early outcomes. The dictum to reduce medication errors and harm using the six “rights” — the right drug and right dose for the right person via the right route at the right time with the right documentation — should now expand to include a seventh, the right genotype. An integrated electronic medical record and a pre-emptive pharmacogenomic approach will facilitate this knowledge in advance of the need to use the medication. For example, in individuals of Han Chinese descent, the HLA-B*1502 genotype will predispose an individual taking the antiepileptic drug carbamazepine to severe, life-threatening Stevens–Johnson syndrome.4 It is simple and inexpensive to test individuals of this ethnicity for this allele before commencing carbamazepine. Pharmacogenomics will become more important over time, particularly in areas such as aged mental health, where polypharmacy in frail people complicates a number of comorbidities. Pharmacogenomics will assist in optimising drug selection for an individual, reducing adverse events and also the time lost when drugs are not effective.

To bring precision medicine into practice through genomics, we need a systematic approach. Australia will need to develop genomic literacy in the workforce through a combination of dedicated specialists (clinical geneticists and genetic counsellors), while upskilling all health professionals. In parallel, we require a workforce of genomic diagnosticians and clinical bioinformaticians to receive and translate genomic research discoveries for clinical use.

Precision medicine will be built on a foundation of evidence ranging from population studies to individualised approaches such as single-patient studies, where clinical trials are targeted to the disease process and the individual genotype.

Precision medicine initiatives raise a range of new ethical issues. Until now, most genetic tests have focused on the single gene under investigation, with recent expansion to small gene panels, specific for the clinical question. However, the genome can be sequenced in its entirety once, and then reinterrogated at multiple occasions over an individual’s lifetime. The large amount of data thus generated can be stored and used indefinitely. The issue of unintended findings has been debated for some time. Genomic data may inform individuals about susceptibility to one or many medical conditions which had not been anticipated by their personal or family history. The American College of Medical Genetics has issued a list of actionable genes, defined as genes for which information about mutations, even if found inadvertently, should be returned to the individuals who carry them.5 In the Australian context, in 2014, the National Health and Medical Research Council released a discussion document on the principles for the translation of “omics”-based tests from discovery to health care.6

While uncertainty is not a new concept in health care, the scale of the current uncertainty associated with the interpretation of genomic data is unprecedented. Without correct interpretation, clinical action based on a DNA sequence variation of uncertain significance is potentially harmful. Unless a sequence variation is classified as pathogenic, no clinical utility can be afforded to such a change. This relates to care of the patient, as well as implications for predictive testing and risk management in family members, and the facilitation of reproductive options, all of which need genetic and genomic certainty. Human research ethics and clinical ethics committees will need to be informed and equipped to deal with these issues as more findings from genomic studies are applied to clinical practice. Informed consent for participation warrants consideration in all quarters. It is imperative to gain and retain public trust.

Precision medicine offers a new dimension in prevention, diagnosis and treatment of human disease, one to be embraced and encouraged. In Australia, we have the potential to contribute to the worldwide knowledge base through our genomic capacity and mature research cohorts. The collection of robust, accessible and linked genotype–phenotype datasets, along with lifestyle and environmental data, will help us to realise the potential of precision medicine for the benefit of our communities.