Geneforum Interview with George Annas: Genetic Turf Wars: Whose DNA is it anyway?

An Interview with George J. Annas
Professor & Chair of the Health Law Department
Boston University School of Public Health

by Mark Compton, November 2000

Over the past decade, George Annas has come to be considered one of the nation's foremost experts on genetic privacy. That's largely because in 1995 he was the principal author of a model designed to serve as the basis for federal gene privacy legislation. Although Congress has since remained divided over the issue, the Annas model has been adopted - to greater or lesser degrees - by a number of states that have passed genetic privacy acts of their own.

George J. AnnasThe first state to take action was Oregon. But now Oregon's Genetic Privacy Act is being reviewed, largely at the urging of biotechnology industry leaders who believe the law inhibits their ability to collect the massive amounts of data required for disease association studies. Of particular interest to industry lobbyists is a clause in the statute which establishes that each individual owns his or her own DNA and thus is entitled to exercise autonomy over its use. Researchers would like to strike that clause. Annas wonders at their motivations.

[Mark Compton]: At present, Oregon's Genetic Privacy Act explicitly defines a DNA sample as being the property of the person from whom the sample was harvested. But various researchers, pharmaceutical companies and biotechnology organizations have objected to this provision, arguing that it unduly interferes with their ability to collect disease association data. Having heard these complaints, the Oregon legislature is currently considering removal of the property clause, substituting in its place some as-yet-unspecified legal remedies. If the legislators decide individuals should not be permitted to assert property rights over their own DNA samples, what implications would you expect?

[George J. Annas]: That property notion was developed on the basis of the common-sense notion that no one should have greater authority over your own body and DNA than you. The other underlying premise is that the simplest way to protect that autonomy is to say simply that you own your DNA and if anybody else wants to use it, they can only do so with your authorization. The idea that anyone else can own your DNA - like the biotech companies or the researchers - while you can't strikes me as nonsense. It can't possibly be that everyone else in the world can own my DNA, but I can't. So when researchers at biotech companies complain about the property notion potentially interfering with what they're doing, what they're really balking at is having to go to the trouble of getting explicit authorization from individuals before using their DNA for commercial or research purposes. But that's something they should be doing anyway. So basically what they're arguing is that they should be allowed to take DNA without consent. And I just don't find that persuasive and I hope that ultimately the Oregon legislature doesn't either.

Should researchers get out of line somehow, would you think they'd have less to fear if the only recourse was state-imposed "legal remedies" as opposed to the current property clause?

Well, without knowing what the legal remedies are, it's hard to say. But again, I think it's likely that someone who can assert ownership, especially in the US, will have a stronger claim than someone who cannot assert ownership to an asset that's been misused or exploited for commercial gain.

Since Oregon's Genetic Privacy Act currently includes no enforcement provisions, might it be argued that the substitution of legal remedies for property rights actually puts some teeth into the statute?

It's possible. But again, that depends on what the remedies are. Ultimately, what you want to protect is genetic privacy - to protect individuals from having their DNA taken without their consent and then used without their authorization. So if we can achieve that end through criminal penalties, that would be fine.

Of course, generally in such cases, industry can be expected to have a hand in helping the legislators define appropriate remedies.

I certainly would not be surprised at that - in which case, I think it's unlikely criminal penalties will be seriously considered.

To get closer to some of the implications, there's a topical issue I'd like to explore. Just recently, a California startup called DNA Sciences mounted a splashy, nationwide public relations campaign announcing its goal to seek out as many as 100,000 DNA donors for a so-called Gene Trust. Billing its effort as a search for knowledge with the potential to change medicine forever, DNA Sciences is directing its appeal entirely to altruism, promising no compensation apart from free diagnostic tests for contributors (assuming, of course, that any such tests are ever developed). To your way of thinking, does this sort of appeal rise to the level of informed consent?

It certainly does not rise to the level of informed consent and I would be surprised, actually, if very many people end up taking them up on this offer based on their vague promise to protect privacy even as they go about collecting personal information about each donor. This is the very type of thing that demonstrates the need for state and federal legislation to protect individuals. Because there is no enforcement whatsoever behind these promises. People are being asked to act strictly on faith. And what do we actually know about this "Gene Trust"? Who's the trustee? For whose benefit is this being done? And if DNA Sciences chooses not to live up to their promises, what can be done about it? The answer is: nothing.

Also, although DNA Sciences and other groups like it routinely promise to preserve donor anonymity, is that even possible in an Internet-powered culture where we're repeatedly reminded that information wants to be free?

I don't think the assurances are worth much so long as it's possible to link the DNA sample with identifiable information about the donor. But generally when people talk about anonymous use of DNA, what they really mean is that it's going to be coded. And my argument is that the person who has that code can break that code. So that's not good enough. I think if you're going to use anonymous DNA, you have to have DNA that's not linkable, which is to say there's no possibility of somebody going back through the records later to find that individual.

But of course, from the researchers' perspective, maintaining that all-important context is vital if they're trying to develop disease association data.

Well, some have certainly made that argument. And, in some studies, identifiable DNA is undeniably important. But if that's the case, they should just go ahead and get the individual's authorization before moving forward. And that's not really all that hard to do.

So long as the authorization rises to the level of informed consent.

Exactly right. If the researchers decide to do a new project, they should be required to go back and get additional consent. But again, that shouldn't be hard because the only reason for keeping an identifier in the first place is to allow them to go back to the donor. So the argument that it's impossible to go back or cumbersome to do so is not particularly compelling - especially when they're arguing at the same time that they need to keep the identifier so they can go back and learn more whenever they feel the need to do so.

So let's say a donor's privacy actually is broached. What potential implications might that individual face?

It depends on what's in their DNA. Suppose there's an indication that they're at higher risk for a form of cancer or for early Alzheimers disease. If their employer finds out about that, they might not get promoted or they may even lose their job. If a life insurer finds out, they may not be able to get life insurance. Or they may only be able to obtain health insurance policies at prohibitively high rates. More important to me, the subjects themselves may end up finding out something about themselves they really don't want to know. For example, we haven't found the early Alzheimers gene yet, but assuming we do, researchers might find that a person is at a higher risk for getting early Alzheimers, and learning this may discourage that person from getting advanced training or going to medical school or doing any number of other things they might otherwise do - thinking they're not going to live all that long anyway. Or their spouse might start to think differently of them. Or their children might think differently of them.

So there's an element of self-fulfilling prophecy here?

No question about it. We know how much people agonize already over their health and get screened all the time for various forms of cancer, high blood pressure, cholesterol … you name it. You can magnify that 100 times with genes because we think of them as being that much more predictive. They're not really as deterministic as all that, but we tend to treat them almost as if they were magic. And in this case, perception is reality since there's a distinct possibility we would discriminate against ourselves should we suddenly learn we're at extremely high risk for various lethal disorders.

And might not the anxiety thus engendered be sufficient in and of itself to induce various adverse health effects?

That's very possible as well and there are many respected professionals who believe that's the case. Certainly, before we get into using DNA for clinical diagnostics, we need to have at least a couple hundred thousand genetic counselors who are able to talk to people intelligently about what genetics means. Because, at this point, it's certainly clear that very few people understand anything at all about the field - including physicians.

And apart from that education, is there some special informed consent that should accompany those kinds of diagnostics?

Absolutely. And we're just in the infancy of trying to figure out how to get informed consent prior to applying a wide panel of genetic tests, for example. It's one thing to get informed consent prior to a very narrow test, say, for the cystic fibrosis gene or even the breast cancer gene, but it's quite another thing to get informed consent prior to administering a battery of 1000 tests. And there are DNA chips being developed even now that will be capable of doing just that. In the future, it's conceivable that 10,000 - or even 100,000 - genetic tests could be run at the same time. I personally don't think it's possible to give informed consent to any and all DNA-based tests a researcher might ever choose to run. That would require something more like a waiver of consent, saying you really don't care what the researchers do. But I think that with genetic tests, which are potentially so life-changing, you'd be well advised to be quite specific with your consents.

Because once the veil of privacy has been pierced, is there really anything that can be done to make a person whole?

I think not, which is one reason why I've suggested that parents should not be allowed to test their children for diseases that aren't going to manifest until adulthood unless there's some way the disease can be prevented or mitigated during childhood. Otherwise, you take away the child's right not to know about their genetic predispositions. It's very important that decisions about testing be left for the individual to make once they're able to grasp all the implications.

And if the privacy of family members of a DNA donor should be compromised, what sort of recourse would currently be available to them? Perhaps more tellingly, what sort of recourse should be available to them, given the assumption that they provided no consent whatsoever - either informed or otherwise - to have their genetic material examined?

Right. It would be difficult under current law for them to show that they've been harmed. Usually, in malpractice law, we talk about specific harms - for example, a physical harm, medical bills or loss of earnings. So unless you can demonstrate that you were fired because somebody else in your family notified your employer of a genetic concern, it would be very difficult to get any satisfaction whatsoever under current law. The physician could always say he passed word to other family members about the genetic predisposition because he thought they might be in danger. And that's exactly why I think we need specific legislation that says doctors can never tell relatives about what they've found - only you can. And that, in turn, would protect doctors from malpractice actions for keeping those secrets. Because if we had a federal privacy law, everybody would know exactly what the rules were and would be able to act accordingly.

As for protecting individuals from unwanted information divulged by family members, I don't suppose we can really write legislation for that.

No, I don't think we can. That's part of being in a family. We often disclose and know a lot more about our family members than we may really want to know. But that's all part of the deal. Still, that can present problems. Take the example of twins who are both at risk for Huntington's Disease. If they're identical twins, that means either they both have the Huntington's gene or neither one of them has it. So if one wants to get tested and the other one doesn't, who should win in that case? Historically, the policy has been that a geneticist shouldn't test one unless both want to know. Personally, I think that's wrong. I think the right to know probably should trump the right not to know. So, in that case, the other one who doesn't want to know is probably going to have to move and not come in contact with their sibling. But I think the people who want to be tested should have a right to that information.

Another property consideration we haven't touched on yet has to do with the matter of compensation. Some believe DNA donors should be compensated in some way if lucrative discoveries are made on the basis of their DNA material. But without provisions that recognize a property right to one's own genetic material, are claims of that sort likely to gain any traction?

It's hard to see on what basis people could be compensated if the law doesn't believe they own the DNA in the first place. I guess if the researchers had some sort of agreement with the donors from the outset, then there might be a basis. But those kinds of agreements are all very problematic. It's unlikely in any event that any one individual's DNA is going to be all that valuable. Most studies are going to involve DNA from tens of thousands of people. But in the rare case where one person's DNA turns out to be valuable - like the John Moore case - yes, I think those individuals should be compensated.

John Moore?

Yes. There's a celebrated California case in which a man named John Moore had been diagnosed as having hairy cell leukemia. His spleen was subsequently removed, but the doctors didn't tell him they were going to give the spleen to a biotech company, which in turn used the tissue to develop an immortal cell line that could be used for therapeutic purposes. The patent for that cell line turned out to be worth a lot of money. And when John Moore finally found out what had happened a couple of years later, he sued to try and get some of the proceeds. Ultimately, the California Supreme Court held that the biotech company owned the patent, along with Mr. Moore's cell line, which led to the conclusion that he had no rights to it himself. That meant the only person he could sue was his physician for not explaining that his spleen might be used commercially. And today biotech companies continue to rely very heavily on that case [John Moore v. The Regents of the University of California]. That's exactly the sort of arrangement they want, because that case essentially says everybody can own your DNA except for you.

Also, above and beyond property protections, doesn't it make sense to have precautions in place that preclude employers and insurance providers from essentially implementing red-lining policies?

Yes, absolutely. That's what I call a fifth-line issue. The first issue for legislation is to prevent people from taking any tissues or samples or blood from you for the purpose of doing DNA analysis without your consent. The second level is to prohibit people from performing DNA tests using tissue or blood samples they already have without first obtaining informed consent. The third level is to prevent people from storing your DNA samples without your consent. The fourth level is to prevent people who have obtained results from disclosing them to third parties, such as employers, without your authorization. And the fifth level - the anti-discrimination level - prevents people who get information about you from using it against you. So you're exactly right. You need both privacy legislation and anti-discrimination legislation.

On the whole, does it seem to you that we're plunging headlong into the collection and analysis of personal DNA samples without having paid adequate attention to the rules, ethics and social norms that come into play?

I think the answer to that is "yes". It's all happening very quickly now. Just in the last year there've been dozens of little biotech companies that have started to go around collecting all the DNA samples they can get their hands on. And I've talked to a lot of representatives from these companies and they all say, "Gee, we don't know what the rules are. We're trying to kind of make them up as we go along." And there's some truth to that, although there has been work on these rules. There are published guidelines and things people can look at. But as a general matter, there's certainly no national standards, no federal legislation. Even our institutional review board's guidelines don't deal very well with stored tissue samples.

So from the biotech industry's perspective, it's better to ask forgiveness than permission?

That's where we are right now.

And who should be party to shaping public policy? Should that be left to government-commissioned panels of worthies or should these be matters that are thrown open to general dialogue?

I think they should be open to general dialogue. After all, what we're doing is asking average citizens to contribute their DNA or get involved in efforts to help biotech companies find something that may cure diseases. Ultimately, if any of this pans out, we're going to get more diagnostic tests and hopefully some effective treatments - all of which are going to be used by the public, so it makes no sense not to have the public involved in setting policies right from the start.

And there's also the possibility of backlash, isn't there, if the public is kept out of the discussion? Witness the current brouhaha over genetically modified foods.

Very true. There are people who are very worried that even the word "gene" is going to become like the word "nuclear". I don't want that. And certainly the example of genetically modified food is shaking up a lot of people in the biotech industry.

As a practical matter, what can be done to facilitate public dialogue? Eric Lander, Director of the Whitehead MIT Center for Genome Research, has been quoted as saying the public is "deeply, deeply uneducated" about the issues pertaining to biotechnology. If that's true, is there a way the public can ever be drawn into meaningful dialogue?

I think it's difficult, but the real question is: Do you try? I think most of the industry is looking for excuses not to even try. But the idea that these concepts can't be explained to the public strikes me as nonsense. It may be that researchers don't want to get informed consent because they don't think they can come up with palatable explanations for the research they're doing. But, believe me, once they're been required to obtain informed consent, they'll figure out how to do it somehow.

Editorial Note: Geneforum has come to understand that at the time of this interview with George Annas, DNA Sciences, Inc. had already adopted and implemented an independent IRB-approved consent process, applicable to all participants in the company's "Gene Trust" research project. According to DNA Sciences, their privacy policy and terms and conditions constitute legally binding contracts enforceable in a court of law.

About the Interviewer

Mark Compton monitors trends in information technology and biotechnology from a comfortable perch midway between the Silicon Valley and Oregon's Silicon Forest.

printer-friendly version