Just four years ago, only two people in the world had their genome sequenced: James D. Watson (co-discoverer of the structure of DNA) and J. Craig Venter (former President of the firm that mounted a private-sector rival to the Human Genome Project). There are now many thousands of such people. At genome meetings, scientists are talking about millions of fully sequenced genomes in coming years. And after that…?
It cost roughly a billion dollars to generate the first reference human genome in 2003; last year a company would charge $10,000 for this service. This year it costs a few thousand dollars. And in a few years we should be able to get our genomes sequenced for a few hundred dollars.
At some point, our genomic information will get cheap enough for most of us to take the plunge and “get our genomes done.” It may be curiosity, or concern about disease risk, or interest in ancestry and biological relationships in the context of social relationships. This seems big and incipient. In most cases, the revelations will not be apocalyptic or world-changing, but hey, a personal genome sequence might be worth a few hundred bucks. And then, for the rest of our lives, we’ll be figuring out what the genomic sequence means, re-interpreting that information in light of new science. But there is a catch. We need access to the data.
Much like the value of a single Facebook page, the value of a single individual genome information is limited. Deriving meaning from genomic data depends entirely on other people sharing their information too, and doing so in a way that links it to information about their health, their relatives, their diet and exercise, and their social and physical environment.
Some genomic pioneers are sharing their entire sequences on the Internet. Yet what our genomes reveal might return to haunt us later, and more to the point, our families. Some of us won’t want everyone knowing our risk of cancer or Alzheimer’s disease, or worse, guessing our kids’ health risks. Yet we’d like researchers to be able to study the data without harm coming to us. And we need them to have access to the genomes of others. We would like to be able to check our individual risk of cancer or Alzheimer’s disease when we choose, or trace our genetic lineages using genomic tools. Unless many, many people share data revealing their risk of cancer or Alzheimer’s disease, and opt into ancestry databases, there will be no way to interpret ours.
For this system to work, we need to be in control of the information that results from having a personal genome sequenced. So this right should not go exclusively to the company that does the sequencing, or the website that identifies how one genome differs from another, or the academic researchers who do the research. The company or university that does the sequencing provides a service, but it should not own or control the data about us.
We want others to use our information, so long as they take reasonable care of it and do not share it in ways that harm us or our relatives. But we damn sure want to be able to participate in research, contribute data to a doctor or genetic counselor, and study our personal and family history without asking permission or getting someone else to share our data.
Yet we do not have this right, and neither do you. By the time most Americans get around to having their genomes sequenced, the choice of how to manage personal genomic data may be well out of our hands. This is not because of deliberate policy, but inattention.
A Flawed Conceptual Framework For Thinking About Genomic Data
For very understandable reasons, mostly related to the past history of medical genetic testing, the debate is being framed in a way that will add regulation and complexity that could undermine the sharing of genomic information—the process that is vital in making the information useful. Three key terms are guiding most thinking about genomic data, and sending us down paths we may regret if we stay on them. Current thinking centers on medical risk, privacy, and ownership, and there are deep problems with how we’re thinking about each of them.
Because our genome can indicate vulnerability to certain diseases, current policy is focused on making sure we get that information from trained professionals. Interpreting genetic risk of disease requires real expertise. Women (and even men) in families with high inherited risk of breast and ovarian cancer can learn of inherited mutations that place them at high risk, for example; and the knowledge can help them decide whether to have surgery or medically manage their risk to reduce it significantly.
These tests cost $3400 at Utah-based Myriad Genetics (or less if done abroad). The concern is that someone might learn of that same risk for $499 by spitting in a tube and hitting the “breast cancer result” button at 23andMe, a company that will test saliva samples for diseases as well as for DNA ancestry. No one should take an action, such as prophylactic surgery without confirming results from 23andMe, which reports on only the three most common mutations. (By the way, both companies fully agree with the need for confirmation.) But it won’t be long before the cost of having your entire genome sequenced will be less than the current cost of commercial testing for just two genes. For the same price, you will get information about the twenty thousand genes in your genome.
If companies are going to give out information that leads to serious decisions about things like surgery, they’d better be right, and they’d better make sure their clients have the help of a health professional to make those decisions. Yet we know from a 2010 Government Accountability Office study that some of the results reporting is sloppy and poorly informed.
In these circumstances, the Food and Drug Administration (FDA) has no choice but to step in. Regulations are warranted. But policymakers will have to tread lightly. We know already that most of the valuable information in the genome will not in most cases be potent or revealing enough to require such dramatic levels of protection or regulation. It will be the rare person who newly discovers a deeply troubling medical risk.
Yes, it will happen and policies need to be in place to manage that situation. Genome sequencing can provide high-impact medical results, but most of the time, it will merely illuminate our genealogy and satisfy our curiosity, and give moderately interesting insights into disease risk. In the continuum of potential results, FDA has a clear role in the medical aspects, but at the other end of the continuum does not have any business blocking us from using a website to find folks we’re related to. Teasing apart what needs to be regulated from what does not is a work in progress, and frankly, we’re optimistic it can happen.
The Danger Of Medicalizing Access To Genomic Data
If regulators impose rules that allow us to obtain genomic data only as a medical service and through a health professional, however, access will never be cheap because it will always be bundled with expensive professional medical services. This is what Germany has done, and we should not follow their lead. By equating genome services to medical services, the German Government has reduced access, significantly limited the possible benefits to their citizens, and dramatically increased the costs.
Treating genomic services as if they were useful only for determining medical risk is like requiring you to review your investment portfolio through a professional financial advisor. By limiting who can provide such services, the German government is being short-sighted. It’s as if it had decided to cut off access to the World Wide Web in response to fears of pornography and copyright infringement. Imagine what the world would be like today if we had passed the Draconian Internet regulation bills proposed a decade ago when the Internet was as young as the nascent genomics sector is now. We would have sucked the water out of the pond that gave birth to the marvels of Google, Wikipedia and Facebook.
No, we need to be able to share our genomic data with whom we want when we want, and keeps us in control, not research institutions and genome companies as the gate-keepers.
What we need even more than protection and regulation is expert help to interpret the information. That will be a long slog. It will take decades of studying millions of people — including their clinical information and environments — to begin to understand what our individual genomes reveal. What’s needed is growth of a service sector that is currently tiny and vulnerable.
Our curiosity — and sometimes our health– will require ‘experts’ to do the science and to explain its implications. Some of these “genomicists” will be health professionals no doubt: genetic counselors, nurses, social workers, or physicians who learn the intricate math and follow the exploding technical literature. Others will be experts in genealogy, history, ethnicity, engineering and anthropology. New businesses are emerging to create interpretive software and interactive websites that walk us through tours of our genomic data. If you’re young and computer savvy; if you study genetics and read anthropology and history, you may have a job in a field that is just coming to life. That is, if we don’t screw it up by strangling this nascent field in its cradle.
A Call To Action
Your genome should belong to you, but it doesn’t always. Right now, for all intents and purposes, it belongs to the service or research institution that produces your sequence data. They may store your genomic information, and interpret it for you, but they have no obligation to provide you with the raw data in a form you can share with others.
That will change only if we collectively insist on it, and if we make sure that genome sequencing services adhere to standards and formats that will make our right to our genomic data meaningful. Genome data need to be interoperable. We need a button to push, or a place to mark “X” on the form we fill out when submitting our samples to be sequenced, and a way to download the information so that we get the results back in a form we can share. It makes sense to have strong privacy protections for those who store and use our data, but it makes no sense that we don’t get direct access to it ourselves, so we can choose what to do with it.
If we don’t move quickly to establish the principle of a consumer’s right to access their genomic information, our nation’s business model for providing services will soon resemble that of Germany. If we act now to recognize a right of access to our genomic data, the emergent services can easily accommodate it as they grow; yet failure to establish such practices from the get-go will mean we do not have access to our own genomic data. The policy debate will focus on the institutions that have the data. Power and choice should remain in the hands of the people the data came from and should belong to, not the institutions that store and control it.