From genomic revolution to healthcare revolution

Genomic testing will inevitably become common medical practice says Richard Holland – but in order to give the best care whilst safeguarding personal information   we need to develop the technical and ethical framework for this…and we need to do it now

The human genome project, although a hugely significant achievement at the time, is old news now having been completed over a decade ago. The 1,000 genomes project is also complete, and now Genomics England is well on the way to delivering a dataset containing 100,000 genomes. These numbers are impressive, as is the raft of cutting-edge technology behind them, but what is it all for?

The variations from one person’s genome to the next (their genotype) drive a lot of their individuality from a physical perspective (their phenotype), but those genetic variations can also affect their susceptibility to and recovery from particular diseases and conditions. Not all illnesses have a genetic connection, but enough major afflictions demonstrate a link between genetics and outcomes that it is worth investigating in more detail. Cancer is a commonly cited example, but even in infectious diseases such as Ebola, resistance or even immunity, can be predicted by the presence or absence of key genetic markers.

The main outcome of genetics research in healthcare is simply a better understanding of how these diseases operate and how the body fights them. Alongside this arises the possibility of developing and prescribing treatments based on their efficacy against a specific genotype, or in other words, the application of personalised medicine.

To date, as with the 100,000 genomes project, the bulk of attention has been focused on building up a big enough collection of individual genomes so that the variations between them may be properly understood. Given that so much variation between individuals appears to have no real effect, it is important to be able to identify which variants are useful indicators and which can safely be disregarded. Much effort has also been spent on developing the tools that can carry out this detailed analysis on such large volumes of data.

There have already been some significant achievements in the understanding of the genetics of disease and the development of personalised medicines, notably in cancer research, and numerous drugs are already on the market which rely on companion diagnostics – a form of personalised medicine to match drug or dose to genotype. Genetic screening at hospitals is also commonplace for assisting with the diagnosis of cancer.

The ultimate conclusion of these efforts should see genetic testing becoming a routine part of the healthcare experience, with general practitioners administering tests to personalise the diagnosis and treatment of the majority of conditions. But this hasn’t happened yet, despite the technology being available, so it is interesting to ask what is stopping this from being reality today, instead of tomorrow?

The technology may be available, but it is slow, cumbersome, and expensive. Sequencing an individual’s genome still takes a few hours and costs more than a thousand dollars. While comparable to some of the more specialist blood tests already carried out in pathology labs, the data it produces is bulky and hard to manage. An individual genome can be moved around and analysed with relative ease, but the clinical-grade data analysis technology we have today does not scale well and when asked to handle hundreds of genomes per day it would struggle to deliver. Having said that, this will probably not be the case for much longer as the speed and momentum behind technological improvements in this space is unstoppable.

Another hurdle is a regulatory one. Enough has been said elsewhere about the relative pros and cons of organisations like 23andMe, but what they are offering is the ability for individuals to make their own assessment of their genetic health, which in any clinical setting would be considered a diagnostic if used by a medical professional. Regulation already covers genetic diagnostics in the clinic, but the idea of delivering this information to individuals to make their own interpretation is a completely new space that 23andMe and the FDA are effectively challenging each other to resolve. Resolved it will have to be as well, as it is unthinkable that if we were to make individual genomes available to patients, then we would not expect them to draw their own conclusions about their genetics using the same tools as their doctors.

The biggest hurdle of all though is an ethical one, and on two fronts at that. The first front is the dilemma of knowledge. With genetic diagnostics available that can make entire genomes available and predict our chances of contracting or failing to recover from a range of terrible diseases, do we really want to be told this information in advance and spend our life worrying about it, or would we prefer to leave this to our doctors so that we can continue our lives in ignorant bliss? And if the doctors know our chances but we don’t, especially from having run a whole-genome test while looking for something else entirely, is it ethical for them not to tell us?

The second front is trust. Recent attempts to set up national identity cards in the UK failed because of the worry that it would give the ability for agencies to snoop on individuals. Similarly, the NHS has delayed setting up a national computerised records system over patient fears that sensitive medical data might fall into the wrong hands. Whether those public backlashes were founded on facts or truth is irrelevant, the important thing is that the authorities in question failed to develop trust in them that they would do the right thing. Having your genome on a national database has proved highly controversial once already on the basis of lack of trust in the authorities. There has been controversy in the UK police as a result of DNA records collected from everyone that has had contact with the justice system, so it is unlikely an attempt to build an even bigger database any time soon is likely to succeed.

The 100,000 genomes that Genomics England is collecting is potentially the start of an NHS-wide genome database but it will be many years until public approval is won to implement a national scheme. The possibilities that arise from having this data available in the clinic and (suitably anonymised) to researchers are huge, as it would allow a reliable picture to be built up of the UK’s genetic condition, our national disposition to certain diseases, and the ability to set up national public health schemes to minimise the risk factors before an outbreak can occur. Individual variations could be checked quickly by every healthcare provider to select the best treatment for a patient regardless of which clinic they present to, and early-warning systems could be put in place to monitor changes in genetics and disease research so that doctors could inform patients of previously unknown risks.

Still, with all these dreams of medicinal nirvana to come, the technology has yet to catch up. The key technological bottleneck is simply deciding what information is important, how to move it to where it is needed, and how to present it in a way that doctors can interpret under pressure in a clinical setting with speed, accuracy, and regulatory compliance.

It may be years until the ethical and regulatory issues can be overcome, and until the public develops sufficient trust in the technology to permit national schemes to be developed. But if organisations such as the Pistoia Alliance can help make sure that the technology is ready to go when the patient is ready to use it then we will be ahead of the curve when the time comes, and there will be no delay in making the best possible medicine available to everyone.

The author:

 Richard Holland is Executive Director, Operations at the Pistoia Alliance – A group that focuses on solving common informatics challenges in a collaborative manner.