The ability to analyse people’s genetic constitutions will transform cancer care over the coming years. But to make the most of the technology, highly sensitive data will need to be shared. That presents problemsby Philip Ball / July 5, 2017 / Leave a comment
“It’s here already, and I believe it’s going to transform the NHS in the next 5-10 years.” Sally Davies, Chief Medical Officer (CMO) for England in the civil service, is talking about the “genomic revolution.” The annual report that she has just released—a regular component of the CMO’s advice to government—is focused on what genomic technologies can do for public health; it is called “Generation Genome.” She spoke about these prospects on 4th July at an event at the Wellcome Collection in London organised by the charity Progress Educational Trust (PET), which aims to promote the responsible application of science that affects people with genetic conditions and infertility.
Davies gave a persuasive pitch, which is backed up by a wealth of detail in Generation Genome, with each chapter written by other specialists. The ability to analyse people’s genetic constitutions—sometimes at the level of the entire genome (3.2bn “letters” of DNA’s molecular code), although that’s generally not necessary to screen for specific genetic conditions—will transform how some aspects of healthcare are done. It will improve cancer care, help to identify rare but often highly debilitating genetic diseases, and should ultimately lead to new medicines and treatments.
It will also raise questions about ownership, privacy and commercialisation of individuals’ genetic data, and will require changes in the way medicine is practiced that are going to be challenging for doctors, medical managers and other professionals to accommodate. One of the messages of the PET meeting, at which several other leading experts responded to and discussed Davies’s presentation of the report, is that it’s not just the general public who will need to be educated about the potential of genomic medicine; there may be a lack of understanding, even resistance, from those on whom we rely for our healthcare.
For some people, genetic testing and analysis sounds like an ominous technological intrusion. Some would, understandably, rather not know if their genes hold possible unpleasant surprises in store. Genetic counselors are already struggling to explain the complex issues about the risks (of cancer, for instance), associated with particular genetic profiles. With the flood of genetic data come difficult choices for healthcare professionals about how to prioritise choices with limited resources and to manage expectations—already, GPs are finding patients coming to them with the results of genetic testing by private companies, demanding treatments.
“The ability to analyse people’s genetic constitutions will improve cancer care—and help to identify highly debilitating diseases”
Since the roughly $2.7bn international Human Genome Project in 1990-2003 to decode the sequence of DNA “letters”—the molecules called nucleotide bases that are the constituents of DNA—the cost of full-genome analysis has dropped precipitously. The price is now down to around £700, and there is no reason to suppose it won’t continue falling. Understanding what all this coded information in our genes actually means depends in large part on searching for correlations between gene variants and traits—genotype and phenotype—across many individuals. Only about 1 per cent of the genome contains genes that encode the structures of the protein enzymes that drive and orchestrate our cells’ biochemistry. Some of the rest is “noncoding DNA,” which is involved in regulating genes: turning them on or off, or ramping protein production up or down. The remainder—the majority of our DNA—is mostly a mystery. Some might be useless “junk” accumulated through evolutionary history, but some might have a biological function.
Understanding how genes relate to health, wellbeing and illness is the objective of the 100,000 Genomes Project, an initiative led by the company Genomics England formed within the NHS, which seeks to sequence that number of individual genomes. So far, 31,000 have been “read,” mostly from patients with rare diseases and cancer.
Because the product of most genes is not a particular bodily function, trait or organ but a protein, it is often far from obvious how genes interact and collaborate to dictate health. In a few cases, diseases can arise from particular variants of a single gene—for example, because people who have inherited that variant can’t produce a functioning enzyme responsible for some vital process, as in the case of cystic fibrosis. Even then, identifying the gene responsible can be challenging if the condition is very rare, since there are then very few individuals whose genes can be compared to identify the “faulty” gene they share in common. Stephen O’Rahilly, a participant at the PET event and professor of clinical biochemistry and medicine at the University of Cambridge, said that we still underestimate the incidence of rare diseases and suspects there are many still to be discovered.
But more generally, common diseases with a genetic component (such as diabetes and heart disease) are associated with several, perhaps many (hundreds or more) genes. Then, extensive genome comparisons are needed to identify the associated genetic risk factors, and to assess the contributions of the respective genes.
“It’s not just the general public who will need to be educated about the potential of genomic medicine; there may be resistance from those on whom we rely for our healthcare”
This sort of correlation analysis has been needed, for example, to identify some genes linked to increased risk of certain cancers. Not only does that work help to reveal which people stand at greater risk—and on whom therefore resources for monitoring the onset of cancer might be usefully focused; the genetic profile of a cancer patient might also show which treatments are more likely to be successful. Sometimes, potentially effective drugs (not just for cancer) can’t currently be used because they work on only a small subset of the population, or because they induce bad side effects in another subset. With genomic diagnosis, it should become possible to reinstate such medicines, targeting just those whose genetic make-up enables them to benefit. That’s a component of the much vaunted “personalised medicine” said to be on the horizon in the age of genomics.
“Generation Genome” offers a measured assessment of these and other possibilities. Davies wants to see genomic medicine “democratised”—to be used to improve healthcare for all, including some of the ethnic minority groups among whom take-up of genetic screening and diagnosis is currently low. Part of the challenge is to keep expectations realistic: Davies said that, undue emphasis on cancer care aside, she was fairly happy with the measured way in which the report had been covered in the media. But some others expressed worries that genomic medicine can be seen as a panacea.
In particular, while it has huge potential for diagnosis, the value for prevention, especially for developing new treatments and medicines, has yet to be proven. In principle, identifying new genes responsible for disease gives drug developers something to aim at—a drug that disables a rogue protein, say. But drug development is in something of a crisis and there are particular economic challenges associated with treatments for very rare conditions. Early diagnosis of cancer is one of the key contributors to a successful treatment; but in some respects genomic medicine might raise the prospect of an identification of risks, dangers and problems greatly outstripping our ability to find solutions.
At any rate, this approach to medicine will be data-driven, requiring the pooling of genomic data from large numbers of patients to improve the breadth and sensitivity of gene-based screening and diagnosis. There are obvious concerns about keeping personal data confidential and secure—Davies couldn’t suppress a wry smile as she insisted that the NHS’s methods are secure, but she indicated that the issues over healthcare insurance are largely resolved. But ownership of such data is complicated. Data about your genomics collected by the NHS won’t be your property: it will be “data about you,” not “your data,” Davies said. And as Michael Parker, chair of Genomics England’s independent ethics advisory committee, said, it may become necessary to think “beyond consent” about how such data is used.
That might sound shocking, but it’s actually an inevitable consequence of how genomic medicine would work. We would know rather little about what your personal genomic data says about your wellbeing and risk, unless we compare it with the genomics of many others. So unlocking the potential of genomics demands data-sharing—and trust. One might see it as a social contract: if everyone agrees to help, there’s a clear social benefit, and so enjoying that benefit entails a responsibility to contribute to it. Obviously, such a contract must include tight restrictions on proper use, and strong penalties for violating them (for commercial gain, say).
“People are understandably hesitant about giving up personal information in the vague service of ‘research’”
If people are going to accept a structure like this, they need to be well informed about what they are signing up to. There’s plenty of good evidence that folk are generous and tolerant when they know they can give something that will help others in need—but understandably more hesitant about giving up personal information or samples (of tissue, say) in the vaguer service of “research.” One thing that will need to feature high on this agenda of education and information is a clear indication of what genomic data imply. For me, a genomic profile is not so much more than any other medical data—much more information-rich than an X-ray or a brain scan, true, but basically of the same type.
Yet the hype around the Human Genome Project, and genetic research more generally, has grossly distorted that message. We have been told for years now that “genes’R’us”: that your genes are what make you you. To judge from the introductory material in “Generation Genome” (“Your genome is the instructions for making you”), scientists in this field still haven’t found a more realistic, more measured way of conveying it in an accessible form. And even if they do, the damage has been done: now private gene-sequencing companies such as 23andMe are cashing in—literally—on this notion that genes are a kind of manifest destiny, the secret code of everything we are. If the “genomic revolution” is going to run on cooperation, sharing and trust, we had better change the record right away.