Genomic technologies are in the news. This summer the UK Prime Minister announced a £300m investment package for the 100,000 Genomes Project, which aims to sequence 100,000 genomes by 2017. So what’s all the fuss about, and what does it mean for lawyers who practise in healthcare, information technology or both?
A (Very Short) Genomics Primer
Genetic information is encoded in molecules of deoxyribonucleic acid, or DNA. Each such molecule consists of two enormously long chains of smaller molecules called nucleotides that come in four ‘flavours’ distinguished by their base—adenine (A), thymine (T), cytosine (C) and guanine (G). If one were to liken a DNA chain to a necklace then each base would be a bead on that necklace. Two DNA chains are arranged in the famous double helix.
The complete set of information in a person’s (or any other organism’s) DNA is called its genome. A human genome consists of two chains of 3 billion base ‘beads’ on a ‘necklace’ distributed over 23 (or 24, in the case of a man) different chromosomes. The genome contains all the information for the RNA molecules and via them the proteins—the stuff that makes up a human body—that person will ever synthesise.
Each genome contains a huge amount of data: a genome, after all, has to include all the information about not only how to build a human body but also to enable it to function and repair itself for a lifetime. In computer storage terms a fully sequenced human genome contains about 1.5GB of information. But this is an ideal. The output of a modern sequencing machine is a data file per genome more in the region of 100GB, before any phenotypic or clinical annotation. Storage and interrogation of genomic data in any quantity is therefore a complex and costly matter.
Genomic medicine builds upon three distinct technologies:
- Modern sequencing methods, in which genetic information in human tissue is converted into digital form, base strings (sequences) of As, Ts, Cs and Gs;
- Translational technologies, by which genomic information is associated with ‘phenotypes’ (how something appears in the physical world) and clinical data, such as medical records;
- The implementation of clinical insights derived from the above in the treatment of an individual patient.
The key breakthrough for genomic medicine has been a (continuing) revolution in sequencing technologies. The very first sequencing of a human genome – the subject of the Human Genome Project – took 13 years and cost about £2 billion. Today, via ‘massively parallel sequencing’ methods, a human genome can be sequenced in about a day for less than £1,000.
The great promise of genomic medicine is that it will facilitate a much wider variety of effective treatments and preventative measures that are available today. Its applications include:
- Swifter diagnoses for rare diseases. Many sufferers of rare diseases have conditions that their clinicians have never encountered. There have been many sad cases where patients have been subjected to years of futile testing at great personal and financial cost—a ‘diagnostic odyssey’—before a diagnosis could be made. Being able to compare a patient’s genome against a database of known genetic conditions can lead to a diagnosis and to effective therapies much more quickly and without consuming other scarce healthcare resources.
- Treating cancers. Cancer, fundamentally, is a disease of the genome: a change in a cell’s DNA can upset its normal working state, making it divide in an uncontrolled manner and damage neighbouring cells. Genomic analysis can help with the diagnosis of cancers and help clinicians devise tailored therapies for cancer patients.
- Reproductive health. Some prospective parents are both carriers for rare genetic diseases. Genomic technologies enable clinicians to help those parents have children who are unaffected by the disease, for example via IVF and embryo screening.
There are many circumstances where an understanding of your genome will alert you to health risks—dispositions to acquire certain illnesses—as well as indicate how you should be treated. At some stage in the near future a baby’s genome could be sequenced to reveal a disposition to acquire a particular disease later in life. If it were known that, say, dietary or lifestyle factors affected the likelihood of developing that disease, that baby’s parents could act to instill beneficial habits from a very early age.
Some drugs are prescribed for certain medical conditions on a ‘one size fits all’ basis across the entire population. But not all people respond to all drug treatments the same way. The field of pharmacogenomics enables clinicians to distinguish between the people who will benefit from certain drug treatments and those for whom prescribing that drug would be a waste of time and resources or, worse still, harmful.
What E-Health Lawyers Need to Know About Genomic Medicine
Genomic medicine raises a number of legal, ethical and policy issues and to illustrate how they’ll arise we’ll first outline how the 100,000 Genomes Project – the UK’s flagship genomics programme – will operate. We’ll then briefly examine some of the most significant challenges in more depth.
How Will the 100,000 Genomes Project Work?
The 100,000 Genomes Project is directed by a state-owned company, Genomics England. Its goal is to generate 100,000 whole genome sequences by the end of 2017. Those genomes will be sequenced from human tissue (typically a blood sample) collected from participants. The participants will all be NHS patients, with an emphasis on those with certain cancers and rare genetic diseases.
Collection and consents. A number of NHS institutions will act as ‘NHS Genomic Medicine Centres’ (GMCs). Their jobs will be (i) to identify and recruit patients from whom samples can be taken; (ii) to ensure that the participants provide the necessary consents; and (iii) to collect samples from, and relevant clinical information about, those patients and then send those samples for sequencing. A number of NHS Trusts, including Guys and St Thomas’ and University College London Partners, have already participated in the early stages of the project and can be expected to continue their involvement until its conclusion. The samples will be stored at a DNA storage centre.
Sequencing. A leading US sequencing company, Illumina, Inc., has been chosen to perform the actual sequencing services, via a pair of deals worth about £240m. The work will be carried out at a new £27m sequencing centre to be built by the Wellcome Trust at its Genome Campus in Hinxton, near Cambridge.
Translation. The consents obtained by the Genomic Medicine Centres will allow Genomics England access not only to the ‘pure’ genomic information but also to the relevant patient’s clinical information. Genomics England will combine that information (via a process that is yet to be explained in detail) and send it back to the patient’s clinicians for use in the clinic. The clinical team caring for that patient will interpret those results and take appropriate steps for that patient’s care.
Research. Genomics England will also create an anonymous data set of that information and make it available upon application by clinicians, scientists, people in training and by industry, although its use will be limited to hypothesis testing using tools provided by Genomics England itself. More extensive research programmes will be facilitated by ‘Genomics England Clinical Partnerships,’ which will provide opportunities for academic and clinical communities, as well as commercial organisations, access to data held by Genomics England. It is not yet clear how Genomics England and, ultimately, the NHS will share in any commercial successes developed by industry as a result of having access to that data, but Genomics England are understood to be open to proposals.
Legal, Ethical & Policy Issues
This overview illustrates a number of legal and policy areas typically raised by genomic medicine. These issues, and more besides, will all have to be addressed as genomic medicine becomes ever more integrated with mainstream medical practice.
Healthcare Resources. Even if it does nothing else the 100,000 Genomes Project will create huge extra demand for people who can interpret genomic data and, which is just as important, explain its significance to patients. This will be especially true in cases of so-called ‘incidental’ findings, cases where a patient’s genome is sequenced because he or she is being treated for rare disease X, but it turns out from the sequencing that he or she is susceptible to, or a carrier for, rare disease Y — which will in turn create demand for more tests to confirm that diagnosis or to test other members of the patient’s family. It’s a good time to train as a clinical geneticist.
Bioinformatic Challenges. As we noted earlier genomic data takes up serious amounts of disk space. Compiling both genomic data and phenotypic/clinical data is a very complex undertaking; establishing an information management system that can link that information to a patient record is even more difficult. If the ultimate goal is to establish such systems at a population level – which it surely must be – then there will need to be substantial fresh investment in national healthcare IT infrastructure to link genomic information with electronic patient records. Security concerns will be paramount: arguably there is nothing more personal than one’s whole genome sequence, and any leak or unauthorised use will easily fail the ‘Daily Mail’ test.
Public Trust and Consents. This brings us on to the broader issue of public trust. The care.data fiasco in Britain in early 2014 illustrates what can happen when a large-scale bioinformatics project loses the trust of the community whose interests it is designed to serve. Similar challenges probably now affect all population-level ‘big data’ projects in a post-Snowden world. So it will be critically important to ensure that government clearly makes the case for genomic medicine and that it be conducted transparently. Genomic medicine is too important to be damned as yet another failed government IT project: if there are hiccups in implementation, as there are bound to be, then those in charge must be honest about them with the public. Similarly, the scope of the consents granted by patients must be broad enough to encompass the use of their genomic data for that patient but, as we will see below, also for wider research purposes. These distinctions must be explained carefully and clearly.
Research Implications. In this branch of science, as with so many others, the best research work will be done by those who have access to the largest sets of data. Genomic researchers will want access to as many human genomes (with detailed phenotypic annotations) as possible: you can do much better science if you have access to 500,000 annotated genomes than you can with 50,000. But these data sets come with strings attached: you may only use them if your proposed use is within the terms of the associated consent. We can probably assume that most people would not like their genome to be used without their consent: we feel that we ‘own’ our genetic information, that we have a legitimate interest in keeping it private, that we’re more than just a datum. Yet much of the very best science in this field, science that may directly benefit us or someone we love dearly, can only be done in cases where researchers have access to massive data collections, drawn from a number of countries, where each human is represented as a (large) file of data.
New Therapies and their Commercialisation. If a life sciences company creates a breakthrough therapy for a cancer after obtaining access to the 100,000 Genomes Project’s anonymous research data set what, if anything, should it pay by way of royalty to the NHS? Bringing a new therapy such as a targeted drug to market is a notoriously expensive, difficult and risky business. Those who most stand to benefit will be the patients suffering from the relevant condition: they will care much more about the availability of a new therapy than NHS royalties. Is it enough that the NHS stimulates private sector activity, or should it also seek to reap commercial rewards, even if that behaviour might deter the very work it hoped to promote?
Each of these topics merits much longer discussion than presented here and there are even more that we could mention, such as the implications for health insurance, the role for cloud-based data storage and access, the development of ‘direct to consumer’ genomic kits, whether genes can fairly be the subject of patent rights and the regulatory arrangements that apply to ‘biobanks.’
One should always be wary of claims that a new technology will ‘revolutionise’ any aspect of society or to ‘transform’ an industry: these expressions have lost much of their impact through overuse. I do hope it’s clear, however, from this brief introduction that the field of genomics can fairly make such claims. But there’s a lot of challenging work ahead to help it realise its potential.
This article first appeared in the September 2014 edition of e-Health law & policy.