
YOU are a living data record. Every cell in your body contains genetic information that, if combined with other genomes into a giant database, could form part of one of the most powerful medical diagnostic tools ever created.
England’s National Health Service is preparing to do just that. “We need to welcome the genomic era and deliver the genomic dream,” the country’s chief medical officer, Sally Davies, wrote in a this year calling for the expansion of genomic healthcare.
“By bringing together lots of people’s genomes and their health data, and following them long term over a life course we will be able to build up insights that we can’t get today,” says , chief scientist at the . Backed by the UK’s Department of Health, this organisation aims to pool genetic information from 100,000 people so researchers can find new diagnostic indicators and ways of targeting therapies.
Advertisement
Medical researchers promise that the benefits of gathering our DNA data will be immense, but your genome is the ultimate personal information – it is literally a description of you. Should we be concerned about giving it away, not knowing who might see it in the future? Or should the potential lives saved outweigh any privacy worries?
The 100,000 Genomes Project is already benefiting people with certain cancers or rare diseases who share their genome – or their cancer’s genome. Although only 32,000 genomes have been sequenced since the initiative started in 2012, it has delivered diagnoses to people who had spent decades without one. Whether that leads to better treatments remains to be seen.
“This is not a database for the sake of a database. This is a database for direct healthcare”
“We use that information to feed back to the NHS and the clinicians looking after that person,” says Caulfield. “This is not a database for the sake of a database. This is a database for direct healthcare.”
But soon the government will make the 100,000 Genomes Project a more concrete part of the NHS in England. From March 2018, NHS England will offer whole-genome sequencing to anyone with specific cancers or diseases, with the option to contribute their genomes to the project’s database. By the end of 2018, the 100,000 Genomes Project will hand over the responsibility for sequencing genomes to NHS England, moving the database into everyday use.
Donate your genome
The study of many genomes, yours included, may benefit others even if it doesn’t help you personally. That makes giving the NHS your genome a bit like blood donation, but unlike blood, sharing DNA might have its downsides.
“The level of interest in that data is going to be far-reaching,” says bioethicist . He is concerned about NHS England granting access to commercial companies, including drug firms. There is a risk your data could end up being used for marketing purposes, for example.
Private companies have an important role to play in unlocking the benefits of genomic medicine by using genomic data to develop new tests and therapies, says Caulfield. In March 2015, the 100,000 Genomes Project set up a that could access some genomic data and health information from its database, stripped of identifiers such as a patient’s name and NHS number.
Companies who were obliged to commit researchers and other staff to the consortium, which wrapped up recently after operating for two years. The idea is that a successor organisation will work with NHS England to continue this work, says Caulfield.
But who will have access to the database in the future is unclear (see “Your data, your choice“). “There needs to be much more transparency on the part of the Department of Health and the government,” says Hockings.
The NHS in England has a history of mishandling health data deals. Last year, 91av found that the Royal Free London NHS Trust hadn’t disclosed that it had given Google’s artificial intelligence branch DeepMind access to detailed information on millions of patients, as part of the development of an app to monitor kidney disease. The Information Commissioner’s Office, the UK’s data watchdog, later ruled the Trust had “failed to comply with data protection law” and hadn’t adequately informed patients as to how their data was being used.
More widely, in July 2016, NHS England finally pulled the plug on its disastrous care.data programme, which aimed to transfer anonymised patient data from doctors’ practices into a central database. It was accused of failing to properly inform people about how their records would be handled, leading to a massive public backlash.
But the benefits of a genomic database strongly outweigh the privacy risks, says , former chair of the , an independent body that advises policy-makers on medical ethics. “It needs to be part of the infrastructure of the NHS,” he says. “You can’t have personalised medicine without pooling all that information.”
For now, the scope of the planned database is limited, but at the University of Southampton, UK, says that unlocking the real benefits of genomic medicine might require much wider national or international genetic databases.
For example, certain genomic mutations have been reclassified through similar large-scale initiatives, she says: families who thought they had a disease-causing mutation in one of the BRCA genes, linked to hereditary breast and ovarian cancer, were told that their specific variant wasn’t disease-causing after all.
“Such reclassification can have important clinical consequences,” Lucassen says, such as telling individuals they don’t need surgery or extra screening.
People contributing to such a database should ideally know how their data will end up being used, says Montgomery. Traditional informed consent – like the kind patients give before an operation – assumes that an individual has understood all the risks and implications of what they are consenting to. Getting informed consent from someone who donates their DNA is much trickier, as it isn’t possible to know how an individual’s data will be tested or used in the future, says Montgomery.
“Brits may have a particular social obligation to donate genomic data, since the whole of the UK benefits”
Although it may be impossible to give people assurances about how their data will be used, Montgomery advocates setting clear boundaries and strong sanctions when it comes to misuse, and the European Union agrees. Its General Data Protection Regulation, which comes into effect in the UK in May next year, anyone from processing genetic data in order to identify an individual. It is likely to remain part of UK law after Brexit.
Getting these rules right and explaining them to the public is crucial. Montgomery says that Brits may have a particular social obligation to donate genomic data as part of the NHS’s social contract. Individuals should be willing to contribute since the whole of UK society stands to benefit from discoveries made through NHS England’s genomic database, he says, since its healthcare provision is open to all.
But this can only happen if NHS England’s genomic mission isn’t derailed by a data-sharing fiasco. Although often overhyped, the benefits of genomic medicine are real and can only be uncovered with vast numbers of genomes from different populations. The NHS is uniquely placed to do exactly that, says Montgomery. “If we don’t get ahead on this, I think we are wasting science.”
Your data, your choice
The safest place to store genomic data might not be a monolithic database. at Stanford University in California is working on a system that allows individuals to share their genomes with researchers, while keeping full control of their data.
With his scheme, individuals would hold their genetic data on their smartphone and choose to share it through an app. “Everybody holds their own genome,” says Bejerano. “There is no centralised database.”
The system works by converting a DNA sequence into numbers and encrypting them, making it impossible to retrieve the original DNA sequence without the decryption key. By comparing encrypted numbers from different people, researchers can identify particular gene variants as interesting for further investigation. They could then ask the people those variants belong to to share their medical records to see if they have disorders in common.
Researchers can’t use the encrypted numbers to access raw genetic information, ensuring privacy is maintained (Science, ).
The system could one day replace centralised databases like the 100,000 Genomes Project (see main story). But as it stands, Bejerano’s method can be used only to share genomic information anonymously, while the most useful genomic research requires fully linking a person’s genome with their medical records.
This article appeared in print under the headline “National genome service”
Article amended on 31 August 2017
We have corrected Jonathan Montgomery’s affiliation