Responsible and effective use of personal genomes

This is the final of three posts from panellists in the Race to the $1000 Genome session today at the Cheltenham Science Festival – this time by Genomes Unzipped’s own Caroline Wright.

As the previous posts from Clive Brown and Adam Rutherford have indicated, there has long been enormous hype and hope surrounding the human genome project and the prospect of a $1000 genome. But what do these developments really mean for the general public? What do we need to know – either as individuals or as health care providers – before we can decide whether it’s worth having a genome sequenced?

Before starting to unpick some of the issues involved in the responsible and effective use of personal genome sequences, it’s worth reviewing how, where and why someone might actually have their genome sequenced. There are currently essentially three different and nonequivalent contexts in which an individual could have their genome sequenced:

1. You can have your genome analysed privately through a personal genomics company (as the members of the Genomes Unzipped project have done), and will receive your results back directly, even to the point of receiving your entire genome sequence. There are various reasons people might want to do this, and the most common reason to date is probably personal interest either in genomics, health or ancestry. There is currently no requirement that these services provide useful information (though it should be a requirement that they provide accurate information within the limits of what is currently known).
2. You might have your genome sequenced as part of a research study, either as a healthy member of the public providing invaluable data on ‘normal’ genetic variation, or as part of a study focused on a specific disease. Usually participants in these studies do not get personal genetic information back, but are contributing altruistically to advancing scientific understanding and biomedical knowledge.
3. You may have had a genetic test through a your health care provider (e.g. the NHS in the UK), probably as a result of specific symptoms or a family history which put you in a substantially higher risk group for a particular disease. Here, there is a clear responsibility on behalf of the health care professional to ensure that the results you receive are accurate, understandable and useful.

It is clear that cheaper genome sequences have enormous potential for both the first two scenarios, but I would suggest that the third scenario is actually the main source of the continuing hype and hope surrounding genomics. How will cheaper genome sequencing affect medicine and public health in the near future? There are already a plethora of diagnostic tests used for inherited and heritable diseases (including inherited cancer predisposition), which are provided by clinical genetics services and can profoundly improve the management and treatment of individuals carrying highly predictive mutations. Because many of these diseases could be caused by mutations in one of many different genes, and the underlying cause of many more is still unknown, cheaper whole genome sequencing is likely to massively increase the number of families who can get a diagnosis and simultaneously reduce the time it takes. In addition, there are likely to be major benefits from sequencing tumour DNA, including targeting cancer treatments and improving disease prediction and prognosis. There may also be opportunities to offer screening tests to prevent disease in the future.

However, the critical question for any clinical test is whether the evidence associated with the test is good enough to make it worth doing. Like any other medical test, whole genome sequencing should be accurate, it must provide information that can distinguish between people who have (or will develop) a disease and those who don’t (and won’t), and it should provide useful information – either to allow individuals or families to plan for the future, or to inform clinicians about appropriate treatment or management decisions. These criteria are even more stringent in the context of population screening programmes, where the potential for doing harm to healthy individuals by making inaccurate predictions is very high. Although this may all sound like common sense, generating the right evidence to allow complex genomic data to be interpreted accurately in an individual is proving to be very challenging.

In addition to these weighty issues of evidence and utility, whole genome sequencing raises a whole range of wider ethical, legal and social issues. This is largely because of the sheer volume and breadth of information contained in a genome sequence, as well as it’s inherently shared familial nature which challenges our notions of ownership. Here are just a few of the big questions:

• Is getting informed consent – one of the cornerstones of contemporary ethical medical practice – practical or even possible in the context of whole genome sequencing?
• What should be done with incidental finding, i.e. results that are not relevant to the clinical question or research study at hand? Some findings may be personally sensitive or clinically actionable, but many will be uninterpretable or simply mundane. Do clinicians or researchers have a responsibility to communicate particular types of results to patients or research participants?
• How should we store genomic data in order to safeguard privacy and confidentiality, and who should have access an individual’s genome sequence?
• When should we allow whole genome sequencing of children, in order not to undermine their choices as future adults?
• How do we prevent unfair genetic discrimination, for example by insurers or employers, and dispel the continuing myth of genetic determinism?
• What regulatory framework should we use for direct-to-consumer testing services (see many previous posts on GenomesUnzipped!) and how will personal genomics affect clinical practice?

Despite these hurdles and outstanding questions, some things are certain: there is no doubt that whole genome sequencing will be fantastically useful for numerous specific medical applications, and these will increase as our ability to interpret the results matures. Nonetheless, for most people, DNA is not and never will be a crystal ball. We will gradually separate out effective uses from ‘snake oil’ applications through proper evaluation of the evidence and consideration of the implications. But knowing your genome sequence will neither cure all ills nor explain all behaviours. And we shouldn’t expect it to.