THE FORUM ON TECHNOLOGY & INNOVATION

* * *

GENE CHIPS, PRIVACY AND NONDISCRIMINATION:

THE PROMISE AND PERIL OF

GENETIC TECHNOLOGY

* * *

WEDNESDAY,

OCTOBER 10, 2001

* * *

This transcript was produced from tape provided by the Council on Competitiveness.

I-N-D-E-X

PAGE

Introduction, Senator Jay Rockefeller 3

Presentation of Susan Siegel 9

Presentation of Dr. Hugh Rienhoff 19

Presentation of Sharon Terry 31

Presentation of Dr. Neil Holtzman 38

P-R-O-C-E-E-D-I-N-G-S

MR. ROONEY: Glad to see you all here today. Senator Frist is going to be here in a few minutes. He's coming back from Senator Mansfield's funeral, but we're going to go ahead with Senator Rockefeller, who will describe the topic today and introduce the speakers.

I just want to take a few minutes to take care of some housekeeping. First of all, I need to acknowledge that these briefings are brought to you by generous grants from the W.K. Kellogg Foundation and the Alfred P. Sloan Foundation. We couldn't do this without them, and we're very, very grateful for their support.

I want to direct your attention to the briefing packets that you picked up when you walked in today. On the back of the first item is the agenda for today's briefing, and we're going to follow the format that I think you're all familiar with.

Senator Rockefeller is going to briefly describe the topic and introduce the speakers all at once, and both Senators have brought you a terrific panel today. They're going to ask each of the speakers in turn to come here to the podium and give you a ten minute overview from their own perspective of their issue of genetic technology, privacy, and nondiscrimination.

And once that's done, then we're going to open the floor. The Senator is going to open the floor to your questions, and we have microphones on the floor, and we very much prefer if you come to the microphone. It's more interesting. It's also the only way to guarantee that your question gets asked, but we know that many of you are shy, and so the second item in your briefing packet are these green question cards.

If you're unwilling to come to the microphone, please go ahead and fill out your questions on a card, hold them up during the briefing, and our staff will collect them, and Senator Rockefeller and Senator Frist will ask questions on your behalf National Press Club style.

Last, but not least, the third item in your packet is a blue evaluation form. I would just ask that you please take a few moments at the end before you leave to fill this out and drop it off at the registration desk on your way out.

We care a lot about what you think. This is our principal means of feedback from you, and we pay a lot of attention to what you say. So we hope that you will do that, and I will turn it over to Senator Rockefeller.

SENATOR ROCKEFELLER: Thank you, Peter.

Bill Frist and I are playing kind of a tag team thing here today, which we apologize for and we apologize to our panelists for it, but it's necessary because of something that I have to do and the funeral that he's coming from. So sort of forgive the more than usual discombobulation. The panelists will be brilliant, and the discussion will be great, and the questions will all get answered.

You're good to be with us. Is Sharon here?

MS. TERRY: Yes.

SENATOR ROCKEFELLER: Here she is. I was just getting nervous because I didn't want to introduce somebody who didn't exist.

And we're going to talk about genetic technology, privacy, nondiscrimination. We've been on the edges of that before. In the title for today's briefing, we've used the phrase "gene chips," and what we mean by that is a suite of technologies that combine genetic sciences with the technology of the microchip, bringing to genetics the economies of scale and continuously improving price performance ratios that characterize information technology as a sector.

This is an overwhelmingly good thing: more efficient, lower cost genetic technology. It has generated lots of new research, and inevitably will improve people's lives.

But as this new technology makes it easier than every before to acquire and analyze and process genetic information, we increasingly will have to worry about what happens to this data, and it's what we're going through in many was on the anti-terrorism bill and all kinds of other things that we're taking up for other reasons. I mean, this is sort of a whole new everything that we're going through.

Case histories of genetic discrimination suggest that employers, insurers, and others will sometimes use genetic information in hiring and insurance decisions if they have access to the data. We probably all agree that there is a role for government in safeguarding confidential genetic data, but there's always a tradeoff between privacy protection and the benefits of free flows of information.

In the case of genetic information, we don't want to invoke protection so stringent that they restrict access to data for medical research purposes. That, after all, would be self-defeating. So how do we strike the right balance is a topic for our discussion today, and we have four very, very good people to do it whom I'm going to introduce in order, take my bottle of water, and get up and walk out and hope that Bill Frist gets back from his.

But they will each talk for about ten minutes. I will introduce all of them, and they'll each talk for about ten minutes, and then the usual procedure, the green cards, the microphones, all of that follows, plus the blue evaluation card at the very end.

Susan Siegel is President of Affymetrix, Inc. It's a company on the leading edge of -- I hate that word -- of the genetic technology revolution. Affymetrix is a pioneer in DNA chip technology, and its products are widely used in genetic research around the world. Its technology has allowed scientists to shorten the time required to conduct genetic tests and experiments from months to hours, greatly speeding up the investigative process.

Prior to joining Affymetrix, Ms. Siegel held a number of senior executive positions in the biotechnology industry. So she will be our first speaker.

Hugh Rienhoff founded DNA Sciences in 1998 and served as its chairman and CEO until September. DNA Sciences is also on that same cutting edge of the genetics revolution with its gene trust project that seeks to associate personal genetic data with medical histories in order to make new discoveries about human disease.

Dr. Rienhoff probably has a unique perspective on privacy because the patients that participate in his project willingly volunteer sensitive personal information in order to further medical research.

He is both a mathematician and a physician, and prior to founding DNA Sciences, he was a principal in an international venture capital firm.

Sharon Terry is Vice President of the Genetic Alliance, where she is the principal and advocate for consumers on genetic issues. She's also the founding Executive Director of PXE International, a lay advocacy group for people with PXE, a specific genetic condition.

Because of her strong involvement in consumer centered genetic research, she was recently appointed as an advisor to the ethical, legal and social implications research program at NIH.

She's an expert on those policy issues that lie in the interface between genetic technology, privacy, and consumer empowerment.

Neil Holtzman is Director of Genetics and Public Policy Studies at Johns Hopkins University, where he also holds appointments in pediatrics, epidemiology, health policy, and management, and probably more. Well, he's a pediatrician. So he does everything.

He's devoted a lot of his career to the study of new genetic technologies and their policy implications, including a stint as an analyst at the Congressional Office of Technology Assessment.

As I indicated, he's a pediatrician. He's into biophysics and genetics. He's offered a lot of books on genetic technology and public policy and has served as an advisor to the human genome project.

So those are the four, and, Susan, if we can start with you, I will gracefully disappear, leave my chocolate chip cookie, and wish you a brilliant discussion.

MS. SIEGEL: Good afternoon, and thank you, Senator Rockefeller, for that introduction and also for actually pulling this session together.

I'd also thank Senator Frist, who is going to be joining us.

And before we actually get started today on the specific policy issues, what I'd like to talk about is just take a minute to give some context to this discussion.

The last few years of the 20th century were known as the Information Age. We believe that we are now living in the Genetic Age. Now, the human genome is really a new frontier with regards to this genetic age, and with this exploration much about our human species over the next few years will be uncovered for the first time, and it will be used to improve the ability and the quality if not of our lives, certainly of our children's and our grandchildren's.

Now, as you all know, we now have the first draft of a human genome. It is a landscape waiting to reveal so much to us in studies ranging from a better understanding of known human diseases, such as Alzheimer's or Parkinson's to cancer, to more cognitive studies profiling addictive behavior, understanding depression, and most recently the link between our genes and speech.

Now, more and more we're seeing that whichever aspect of the human condition that we actually examine, be it physiological or psychological, the interplay between one's genes and one's environment is deeply intertwined.

At Affymetrix we developed gene chip technology. For those of you who haven't seen it before, you can see it here. This is the gene chip technology, and I have, just to give you a sense of scale, a gene chip.

Essentially this chip borrows very heavily from the processes in the manufacturing and photolithography from the semiconductor industry, and on this little piece of silicon on glass, in essence we have millions of pieces of DNA.

What it does, it's a very powerful tool for helping and enabling scientists to essentially unravel a lot of the mysteries of any genome, be it human or animal or plant or any living organism.

Now, we can do it through expression variability or sequence variability, and very briefly to give you what those are, expression variability is the upwards or downwards regulation or expression of certain RNA messages that essentially you can differentiate between either a normal or a sick cell, just as an example, normal or sick condition.

DNA variability really is the mutations or the changes, the single nucleotide polymorphisms which you probably hear an awful lot about out there that actually has a change in the base of our genetic chain, essentially our DNA.

Now, one of the things about these chips is that they can generate unprecedented amounts of information in one experiment, and they really do personify the information intensive nature of the genetic age.

Now, each chip generates approximately 50 megabytes of information. So just to give as a point of reference here, written out in text the genome of just one person, so three billion letters, those four letters, GATC; if you were actually to write those out, they'd fill 300 volumes of the Encyclopedia Britannica, again, just giving context.

So think about the analytical power that you're actually going to require to analyze not just one genome, but the interplay of each of the genes not only with one another, but with our environment, and with that, what it provides in terms of the complexity of the human being, but also the diversity of the populations.

Now, as impressive as the amount of data from a chip may be, just as impressive is its range of uses. For instance, in your home state, Senator Frist, actually, our technology is being used at St. Jude Children's Hospital for the study in new ways to differentially diagnose leukemia.

In West Virginia, in Senator Rockefeller's home state actually, at the National Institute of Occupational Safety and Health, gene chip technology is used to really link between genetics and work place toxicity.

Now, there are many applications of gene chip technology that I really just can't go into today because of time, but one of the vital interests to us certainly under the current circumstances, which it can also be used for, is national security, and obviously, we're working on developing chips that will allow for quick identification of the bacterial or viral agents that might be used in biological attack, the detection of quick identification of the bacterial and viral agents that might be used in biological attack, and we're working with the U.S. Air Force on this.

Now, any science that allows us to chart the innermost reaches of human existence, not just the human race as a whole, but each individual one in it also presents us with profound ethical questions, and none are more profound than those of privacy and nondiscrimination.

The human genome reaffirmed one of our most enduring truths. With regards to our genome -- you've all heard this -- we are all 99.9 percent the same. So each of us is inherently equal.

Now, look around you. Each one of us is certainly a different individual. So both sides of this equation, our inherent equality and our inherent individuality, underscore why a comprehensive approach to privacy and nondiscrimination is absolutely vital.

Now, at Affymetrix we began to explore these issues in 1997, and we created an ethics advisory committee. It serves as a forum for us for discussing a number of policy and really ethical questions, and it's under the guidance of Paul Berg, a Nobel Laureate professor at Stanford University, and he's certainly been one of the foremost thinkers in this field.

Now, at Affymetrix, there are three broad ethical principles that we believe should apply to the gathering and use of all personal genetic information. First and foremost, privacy, the assurance that all personal genetic information is treated as private and confidential and that all individuals have the right to decide who sees that information.

Second, nondiscrimination. Strict controls so that personal genetic information cannot be used as a basis for discrimination in employment, insurance, or for that matter, anything else.

And last, individual autonomy, the notion that people have a right to understand and provide consent for use of their genetic information and a right to decide whether or not they want to participate in genetic testing or treatment.

Now, you might say achieving consensus on these principles, which might be self-evident, actually becomes much more difficult when you start discussing how to implement them. In the public sector, 23 states have passed some form of privacy and nondiscrimination laws for genetic information, and some of them, like New Jersey's, are quite comprehensive. Others cover only specific areas like insurance or medical records.

Now, as laudable as these efforts might be by the states, privacy and nondiscrimination have their basis in constitutional law and have historically been cited here in Congress and before the Supreme Court. Most of us view these as fundamental rules of societal fair play that should apply regardless of where someone's zip code might be.

Now, since every American will actually feel the impact either now or in the future of the dramatic developments in genetic research, it follows that we actually need a standard of privacy and nondiscrimination that protects all Americans and not just those in the states I mentioned.

Now, at the federal level, we've seen a fragmented approach, looking mainly at questions surrounding medical care, providers, insurance coverage, and employment discrimination. The Health Insurance Portability and Accountability Act has provided some protections, both in the statute and in the follow-up regulations, but the insurance and employment discrimination issues, especially employment, are far from being resolved.

Now, as many of you know, Senator Frist, Senator Jeffords, Senator Daschle, and Congresswoman Slaughter have all introduced legislation in this area with a number of bipartisan co-sponsors.

The President has also indicated that he would sign a privacy/nondiscrimination bill just as he did when he served as governor in Texas.

Now, some are very concerned that legislation in this area can, in fact, impede important research by making it difficult for scientists to have access to aggregate genetic data. This is a very legitimate concern.

However, we believe it can be addressed by having the appropriate language regarding research imbedded in the legislation and by our ongoing efforts in the scientific arena to insure that we have informed consent from those who undergo genetic testing.

Now, no one I've spoken with wants to stymie scientific research, and I also suspect that virtually no one who undergoes a genetic test or participates in a clinical trial would, in fact, object in theory to his or her data being used in an anonymous aggregate manner for the good of the all.

But it's the obtaining of that data, the most personal data imaginable, by an employer, an insurer, a lender or even, God forbid, a telemarketer that frightens all of us.

So perhaps even more fundamentally, we have no basic prohibition on the theft of DNA, the way we protect other forms of privacy. For example, it's illegal for you to open your neighbor's mailbox and read his or her mail. It's illegal for you to tap someone's phone. It's illegal for you to spy into someone's bedroom. And it's illegal for you to do these things regardless of what you do with the information.

So shouldn't it also be illegal to take a cotton swab to someone's coffee cup, glass, or soda can, in fact, after this session and have your DNA?

So I do want to emphasize that as important as these pending bills are, we should view them not as an endpoint, but certainly as a starting point and as part of an ongoing process to insure that laws protecting our individual liberties keep pace with science and scientific discovery.

Now, for all of us who work in genetics, these are very exciting times. Genetics has absolutely gone mainstream, and you've seen this. You see it every day. You see it in the newspapers. You hear it on the TV. In fact, at the movies, you can see DNA in art galleries, and in fact, in museums.

Now, we're witnessing breakthroughs now in science that will forever change the ways in which we understand, manage and treat various forms of the human condition to improve the quality of life. So for those breakthroughs to continue, they will only happen with public acceptance and participation, and that acceptance will only happen if people, in fact, feel confident that both the private and public sectors have developed rules of fair play about how genetic information can be used.

Thank you for your attention.

SENATOR FRIST: Susan, thank you.

(Applause.)

SENATOR FRIST: We'll just proceed on down. This computer switched, and then we'll have questions after that.

Let me remind everybody. I'm not sure if Jay mentioned to fill out question cards or we always prefer you going to the microphones just to have a more lively interaction as we go forward, but feel free to fill out these question cards, and you can raise them in the air, and we'll have staff come through and pick those up.

DR. RIENHOFF: A slight technical problem here. I think I'll be all right.

While I'm rebooting, I want to thank the organizers, particularly the Senators for inviting me.

I started DNA Sciences really with the view of a physician. In fact, that's how I was originally trained, and most of the way I approach science is really starting from the bedside and moving back to the bench. That's how I think about the problems.

So DNA Sciences was conceived with a view towards enabling physicians to --

SENATOR FRIST: We can't hear you real well.

DR. RIENHOFF: Okay.

SENATOR FRIST: Do you know what I'm going to do? If you'd just step down for a moment, I'll remove this. Sorry about that.

MS. SIEGEL: It's genetics.

DR. RIENHOFF: Can you hear that? It could be the environment though.

(Laughter.)

DR. RIENHOFF: I mean, that's part of the theme here.

So the reason I started DNA Sciences was because a lot of the genetics that's relevant to clinical management is unknown, and one of the things I'd like to get across in this told is that the genetics that we're addressing from this point on is not like the genetics that was addressed, say, ten, 20, 30 years ago, which is called Mendelian genetics, where a single gene accounts for typically a single disease.

The kind of genetics that we're addressing now is where there's a complicated interplay between the environment and genes, and there are many genes that are involved in that process, as there are, as you can imagine, many environmental agents or elements.

And the important issue is that the information that we're going to understand about the genetics and complex disease and drug response, et cetera, et cetera, is more about a probabilistic picture.

For example, when a physician draws cholesterol, the cholesterol level assigns a certain risk that the patient will develop coronary artery disease. It's not an absolute, and in fact, many people who develop coronary artery disease do not have elevated cholesterols, and likewise, many people who have elevated cholesterols don't develop that disease.

So the issue is assigning risk to specific individuals and learning how to manage that risk.

Another important component to that is that for me, genetics is not really a different kind of test. If you think about assigning a fivefold increase probability of developing colon cancer to a given patient based on their certain genetic profile, that's no different than identifying that this particular patient has a cholesterol of 350.

So I like to think of genetics as just another analyte that we look at, another thing that we measure -- and you'll have to bear with me. Sorry about this -- that provides a higher resolution.

But I really want to emphasize the point that genetics is not deterministic. It is more probabilistic, and it's much more today than it was 20 years ago when we're dealing with single gene disorders like sickle cell, cystic fibrosis.

The reason we discovered those is that they're easy, relatively easy to find, and the reason we haven't discovered the genetics for common heart disease, common cancer, et cetera, is because they're hard to find, and one of the things that DNA Sciences set out to do was to devise a strategy using the current tools that we have, the new tools like the sequence of the human genome and other things, to identify those genetic elements.

So I think we're on the road here. I probably used up five of my ten minutes here. I don't know what happened. Okay. Here we go.

So I think I've already stated this first slide, but the point is that the genetics of common disease -- and a lot of that is predisposition, but not entirely -- and the genetics of drug response is out there to be discovered. Most of that has not been discovered, and that's what we've set out to do, and many other people, both in commercial organizations, as well as the academic world, are trying to do the same thing.

There are many reasons why one would want to do this, and I think participants that we've encountered are interested purely in accelerating the discovery of research. Most of these diseases that we're focused on, like diabetes and heart disease, et cetera, are known to run in families, and people are highly motivated typically to want to participate in studies if they think there's some good that's going to come out of it.

Obviously, the researchers are keen to understand it, and that includes the people who are developing new therapies, et cetera, et cetera. Physicians and patients want better treatments. They want more refined diagnostic tools and better predictive tools, which is a lot of what genetics can provide.

And I would extend this to say that employers would also want their patient or their employee populations to be in better health, and particularly ones who are self-insured.

I don't have on here insurance companies because I can't really figure out what they have to gain from this other than identifying where the risk is in a particular group of people and assigning higher premiums to those at risk. So I don't think insurance companies have anything to gain from this as far as I can tell, and I'll have more comments on that.

From a clinical and scientific point of view, genetics can contribute an understanding about the predisposition to a disease, but predisposition or the reason somebody gets a disease often determines the path that that disease follows as well.

There are many, many examples of that where a particular subset of patients has a different course even though clinically they have the same diagnosis. So not only will genetics provide insight into why people have disease and who's going to get it on a probabilistic basis, but what course that's going to follow.

And ultimately that leads into what is the appropriate therapy for that particular patient.

I think the other advantages to understanding the genetic predisposition to disease come in the form of developing prophylactic agents. I think once we understand what the risks are, then we can identify how to mitigate those risks through drugs, for example.

But we can't do those kinds of studies unless we know who's at risk, and I think they are very difficult studies to do if we take all comers of whom only a subset actually have a particular predisposition. Statistically it's too hard to do.

And finally, I think epidemiology, which is focused on identifying general risk in populations, is going to be able to be much more precise in identifying the environmental causes of disease if we can hold fast as a constant, if you will, the genetic elements in a particular population.

So I think epidemiology will jump forward once we understand the genetic component.

It is true that genetics is probably one of the most unbiased ways of doing a study, meaning that when you look for a gene in the genome, you're making the assumption that anywhere in the genome that gene can lie, and you have no prejudice as to what gene might be participating, and we're still in the infancy of understanding human biology. In many cases we have no clue as to what the underlying cause of a disease is, and genetics gives us that first glimpse of what pathway or what particular tissue might be amiss.

And I made the point earlier, but I'll reemphasize it, that genetics is just another kind of test. Yes, it is DNA, which is heritable, but I remind everybody that more than 80 percent of our cholesterol levels are genetically determined independent of your environment.

And so in many ways, I think of cholesterol as a genetic test, and many of the genetic tests that will be coming out that assign risk for common disease, as I've said, will have the same kind of elements that cholesterol does, namely, it will assign a certain relative risk of disease. It's a probabilistic statement.

So, as I said, we're in a new era. The genome is sequenced or at least a rough draft is there. I would emphasize that there's an enormous amount of variation in the genome that's below the surface that probably is accounting for a lot of the variability that we see. It's not the common stuff. It's not the common variances. It's probably the relatively uncommon things.

And I think that the deeper that we dig into the genome actually the more variation we find. It's really astounding.

We also have large scale sequencing and genotyping. Sue talked about some of that technology. There are two other things which I think are prominent, which is Internet connectivity, and that's really a communication tool that allows people to communicate, people to aggregate, and the content to be zooming around, which wasn't the case previously.

And that's layered on top of a general movement over the last 15 or 20 years of consumerism in health care. When I originally trained as a physician, the patient really had very little information, and when my grandfather, who was a physician, when a patient came to them, that patient was in the thrall of the physician. There's no question about it. That was in some ways the Golden Age of medicine from a physician's point of view.

But now the physician and the patient are partners in information sharing, and that's driven a lot of other things.

So the new genetics, if you will, or the genetics of common disease and drug response needs to take advantage of the new tools that are in place, and one of the things that's very important, as I said earlier, is that the genetics of common disease is about variations that are found in populations, not in extremely rare families or extremely rare diseases.

And so a population based approach needs to be taken, and what I say here is that it's geographically diverse and genetically diverse.

The other thing that we at DNA Sciences felt was important is that we wanted to have a direct relationship with the person who was in the study. We felt that by having intermediaries was a mistake, and there are lots of reasons for that, but one of them is having long-term contact with the patient so that we could continue to elaborate on their clinical history as their clinical history evolved, as it always does.

Another thing that we're capable of doing now and we find ourselves increasingly sequencing every individual at specific loci because the variation is so great; this previously notion that one particular variant might account for all of the patients, sickle cell anemia is a classic example of that. Cystic fibrosis, one particular variant accounts for 67 percent of the patients.

In a recent study that we did of 122 patients that had an identical clinical syndrome, we found that only two of them shared the same genetic variation. Every genetic variation in all of the others was unique, all in the same gene, but different genetic variations.

So there's a huge amount of sequencing that needs to get done in individuals, and also you need the ability to go back because we're not smart enough when we start these studies to know exactly what we're looking for, and as I said, not only do you need to go back to the patient and get additional history, but you need to go back to their genome and do additional analyses.

So I will jump forward and just talk 30 seconds on what we did, the Gene Trust. It's basically a genetic epidemiological study based on the Web. It's self-reported data. Patients provide that data. Typically they are very good historians when it comes to certain kinds of diseases.

Most people know they have diabetes. Most people know if they've had coronary artery bypass surgery. They don't lie about that.

It was an IRB approved protocol with proper informed consent, and we did a nationwide kind of sampling for this. So we got all 50 states basically.

I won't go into the details of the implementation, but I think you can imagine that security of the information was paramount, and we spent a huge amount of time just teaching ourselves and devising the kind of software and the infrastructure to allow us to collect that information in a secure fashion.

I think we can have a discussion on this, but all of us are aware of the issues with respect to health care information and what the threats are to its dissemination or to its leaking, and I think this is maybe the heart of the discussion we're having today.

And I'll close with just a few ideas that I think are important to consider in any sort of legislative initiative. One is that the IRBs need to be professional, meaning no conflicts of interest, and they need to be composed of people who have the time to properly review these protocols.

I think a uniform informed consent is a desirable thing. We dealt with this in a number of different ways ourselves.

One of the biggest problems, it seems to me, with respect to genetic information now and predisposition is this notion of decreasing the risk pool. In fact, I think from a societal point of view, we need to increase the risk pool and find ways to do that.

And I think when you really take it to its logical conclusion, some kind of national health insurance always comes up in that kind of discussion because that's the ultimate risk pool in the United States.

I'd like to leave with this notion, again, that genes are just another analyte. DNA is just another thing that we can measure. It doesn't tell the whole story. There are lots of environmental interactions, interactions between genes, et cetera, et cetera, and it's a very complicated story that I think in the short term it's getting more complicated, not simpler.

So I'll conclude there.

SENATOR FRIST: Thank you.

(Applause.)

SENATOR FRIST: Again, I'll add any questions or cards or comments, go ahead and rise them up in the air and we'll be happy to pick them up from you as we proceed ahead.

MS. TERRY: Thank you.

I'm going to present today a consumer perspective on some of the issues we've been hearing about, balancing the promise and the peril. I'm going to give you my perspective, which is informed by my experience. I don't come here today as a scientist at all, but I come here as the parent of two affected children, affected by a rare condition, but now having worked several years with the Genetic Alliance and many common conditions working throughout genetics, you know, in a number of ways.

I'm a participant in research. Two days after my kids were diagnoses, my blood was taken, as was the blood of my children, without informed consent and all the other things that one would hope would be in place, but I was naive at the time, of course.

I'm now a researcher. I run a blood and tissue bank in a foundation for the disorder. I help conduct research. I published two articles back to back in Nature Genetics, probably the only person with a Master's in Religious Studies who's done so.

I administrate PXE International, which is a disease specific organization for the condition my children have, and I'm a board member of the Genetic Alliance.

My kids, Elizabeth and Ian, were diagnosed in 1995 with pseudo xanthoma elasticum, which is a disease I, of course, had never heard of. It was two days before Christmas, and we were very much devastated. The condition can lead to central vision loss, legal blindness, gastrointestinal bleeding, cardiovascular problems. We read death as early as 13 in some cases. So we were incredibly overwhelmed with this diagnosis of what appeared to be two healthy children.

In 1996, there was limited, poor quality information. There was no prognosis, and there was no treatment for this condition.

We went to the Genetic Alliance, which is an international coalition of families, researchers, health professionals, industry, organizations, corporations, all promoting healthier lives for those living with genetic conditions.

The Genetic Alliance helped us with mentoring and resources, networking and support, and we founded PXE International.

I'm going to talk about the balance because I think the idea of genetics and genetic discrimination and the issues around it really lie on a balance of which the crux is the ethical, legal, and social issues that we keep hearing so much about.

And I want to also say that although we often focus on the risks and the benefits, what we really need to think about, too, are the nuances, that there is no black and white in these issues, and that, in fact, some of the continuums in terms of the types of research, length of research, whether diseases are common or rare, the morbidity, et cetera, are really critical in terms of the balance.

I think policies are needed, and I'm going to talk about privacy, confidentiality, oversight, access, and disparity with regard to policies.

With regard to privacy, I'm going to ask: is it possible anymore? And while in some cases that may seem outrageous, what I think we as consumers feel is that it really isn't possible the way we would want it to be possible, and then we start to ask ourselves the question: do we really want privacy or do we really want to prevent misuse of information?

Ultimately genetics or our genetic identity is the ultimate identifier. Eventually just knowing somebody's DNA, and as Sue pointed out, picking it up from a glass we'll be able to tell someone your genome, and so it's the ultimate identifier, more than your Social Security number, your name, your address.

And so we want misuse of that information to be prevented, and that's a more critical issue for us as consumers.

Confidentiality is critical, and as Hugh pointed out, it's not really much different from other kinds of medical records with regard to genetics, but we're looking at research, clinical trials, genetic testing, and those sorts of areas need to be protected so that people do participate.

Consumers very much want research to go forward and need the protections in place to be able to do that.

Oversight is going to be critical with regard to research, genetic tests, clinical trials, and drugs, and we have the beginnings of some of those discussions happening in our society. A number of federal committees, like a Secretary's Advisory Committee on Genetic Testing, are looking at those issues and need to continue to do that both with consumer involvement, industry involvement, academia, all the players around the table together to make decisions about these critical issues of oversight.

Access is really important. Availability and affordability are going to be the bottom line for anyone with any disease, whether you're diagnosed with breast cancer or zerodermapigmentosa, another rare disorder. What you want is accessible and affordable and available treatments, and so those are going to be really critical in our decision making around how we protect information and how we use it.

Disparities are going to be very important in asking will genetics widen the gap in health care disparities. It certainly has the ability to do that. The new technologies at first, as we keep saying, may be cutting edge and so may be only available to the affluent. We have the opportunity to have that not happen and to plan ahead so that it doesn't happen, and that will be critical.

The Genetic Alliance did a genetic discrimination study feeling that this was a critical issue that wasn't well documented and that there was an increasing potential for unauthorized disclosure and misuse of information. We contacted 332 volunteer consumers who came forward in telephone interviews and looked at areas of discrimination that they were experiencing.

And we saw similar to what Sue mentioned, that health insurance at work, life insurance, disability, those things were really critical to people, and about 75 percent of those who came forward reported some sort of discrimination. It's certainly a critical issue and one that we need to really be looking at quite seriously.

The bottom line for most people is informed decision making; that whatever decisions are made with regard to their participation in research, their employment, their health insurance be informed decisions, and those informed decisions can be placed in a context that's really a critical one for making them in a way that's relevant and realistic.

Supportive policies in a community base relevant to the community in which they are operative in the context of the family, the individuals able to look at those issues with the research enterprise partnering with both, and the health care enterprise being part of it as well; if all of those entities can partner together, all within a context that's supported by good public policy, then research can go forward and people can feel comfortable in the decisions that they have to make.

For the future, I would ask: do genomics research and technology have the vehicle they need to move into the mainstream with the appropriate protections in place? And I think the mandate is that we must stimulate, accelerate, and integrate genetics research into health care with balance, focusing on the promise, while recognizing the peril.

Thank you.

(Applause.)

SENATOR FRIST: Thank you, Sharon.

Neil.

DR. HOLTZMAN: Senator Frist, ladies and gentlemen, when Peter Rooney asked me to do this, I said that I would rather not prepare formal remarks and, difficult as it might be, and it's proving difficult to respond or to comment on the remarks of the previous speakers.

So I've been furiously taking notes, and most of these will fly out of my head, and there's a lot more that I can say, but let me begin with Hugh Rienhoff's comment, which he mentioned a number of times in one form or another, that genetic tests are not different than others.

The question that faces you, many of you staff who are wrestling with genetic legislation is if genetic tests are not different from other tests, why do we need special legislation.

Now, I've come down on both sides of that issue. Let me first indicate some of the reasons why we might need special legislation for genetics.

First, there is a tremendous amount of public concern, and Sue Siegel mentioned this. The press is filled with genetics. Genetics is the mainstream, as she said, and in addition to the excitement and the hope that has been generated, there have been a lot of fears.

So subjectively there is a concern and a perceived need for special legislation regarding genetics.

Now, that's subjective. I think that we still have to recognize that there are differences in genetics, and the point I want to make is that research issues should be separated from clinical use, and we need to think about special considerations in both areas.

In terms of research, we have to remember that in involving usually a subject or a patient in research, looking for genetic factors in disease, we're not only involving that patient him or herself many times, but also his or her relatives, which puts it in a category that is different from many other types of research.

In addition, once we collect a specimen for any kind of research and from which DNA can be obtained, we're immediately presenting to ourselves and to other researchers a source that can be used for a wide range of studies, for instance, for looking at any of the 30,000 or so genes that each of us possess.

So there's a concern about the use of specimens, and certainly the question of how informed consent should deal with these questions of contacting relatives, of future use of specimens, and et cetera.

So in the area of research, these are quite different.

Now, when we turn to practice, there is this concern that Sharon Terry mentioned about maybe it's not privacy as much as having the results fall into wrong hands, and how legislation is crafted to prevent that seems to me to be an enormous challenge.

Now, on the other hand, so these are areas where genetics is different. On the other hand, there are other situations, particularly in the use of medical information, where privacy violations and misuse of genetic information are also prominent.

I worked at OTA, as you've heard, when it was still in existence, at a time when HIV concerns were coming up, and certainly before we even thought about, before the public was usually even aware of the possibilities of insurance discrimination for genetic tests that this was a major concern in terms of AIDS and HIV, where not only is it a matter of the person, him or herself, but it's the matter of partners and possibly relatives as well.

So that many of the concerns that have risen about genetics apply to other areas of medical investigation in the clinical practice as well, and in fact, some critics of genetics legislation have raised the question that they label genetic exceptionalism as to whether we need special legislation for genetics, and I think there are two ways of looking at it, and you folks are really in an important position for doing this.

Number one is if we codify protections, if we can, of genetic information, whether it's tests alone or tests plus family history, do we then leave untouched the enormous number of violations that don't involve genetics per se in terms of intrusion into the privacy of one's medical records?

Another way of looking at this, however, is whether by for the first time grappling with a subset of the problems of medical information, namely, genetic tests or family histories, by putting special protections around that subset, do we establish a precedent that we can then use for other broader areas of medical information?

I don't know the answers to these questions, but I think as you craft legislation or consider which legislation to craft that you need to think about these problems. I like to put it in terms of are we putting a foot in the door by starting on special genetic legislation or are we slamming a door shut to protections of other violations of medical privacy?

Now, it sounds -- let me turn and raise sort of a second theme, and I'll begin by talking about what Sue Siegel pointed out, that we're 99.9 percent identical, suggesting that there really aren't very many differences.

Well, if you do the arithmetic, it turns out that with three billion base pairs, that it means that we differ if it's 1.1 percent in about three million base pairs, and if you do the further arithmetic that we have about 30,000 genes, then it means that in each gene there are 100 differences, the potential of 100 differences on average from one to the other.

So there is still an enormous amount of variation from one individual to the next, and most of that is within a given culture or what has been called race rather than between them.

Now, the reason I mention that is -- and Hugh Rienhoff touched on this a lot -- is that because there's so much complexity, so much opportunity for variation, that the possibilities of being able to discover genetic differences that can be translated into tests that have high predictive value for common complex diseases is extremely difficult.

And you and I were talking a little bit; Hugh Rienhoff and I were talking a little bit before about whether this is possible, and he characterized himself as an optimist. I'm not so sure.

If one looks at the common diseases that have been classified so far for which genes have been identified, notably breast cancer and colon cancer, that one has to recognize that where a single gene plays a prominent role, that that gene accounts for less than five percent of cases, and in the ensuing ten or so years since those discoveries have been made, enormous searches and enormous expenditures of money have been spent looking for similar kinds of genes to explain other common diseases and have succeeded, again, in only a few of those diseases in identifying highly predictive genes in a very small percentage of all the people who suffer from those diseases.

And efforts to look for more for the genes that underlie the more common forms of those diseases have failed, and I think it's not a technological failure. We've got wonderful technology, and it's going to get better. It's the failure that's due to the complexity of the genetics.

So that while there will be claims made for predictive tests or diagnostic tests, what we need even before we or in parallel at least to considering issues of privacy is the question of the goodness of those tests. Are we going to let tests get out there where the people using them are unaware that they really have limited predictive value or limited diagnostic capabilities?

And for that we do need oversight. The Secretary's Advisory Committee on Genetic Testing is grappling with this, and I would suggest to you that if we're really going to protect the public, in addition to the problems of privacy and confidentiality, that a major concern is assuring the quality and the validity of these genetic tests.

This is something that FDA does if a manufacturer wants to market a test as a kit, but right now most genetic tests are being marketed as laboratory services. The Secretary's advisory committee has recommended that FDA regulate these tests as well, and FDA is beginning to do that.

But I point out to you that unless that happens, we can have a lot of tests out there that will be very poor predictors and that will be very badly misused.

So, finally, let me again turn to the last thing that Sharon Terry talked about, and that was the matter of disparities, and she raised the question of whether there was too much emphasis on research.

Let me raise what I think is a question of disparities in a broader area, and that is are we paying too much attention to genetics. I pointed out to you that, yes, there are the rare diseases and the rare forms of common diseases where genetics makes a contribution, but for the vast majority of diseases that people, you and I, suffer from or will suffer from today, the opportunity to make genetic prognostications or genetic predictions is very limited, and the question is: if we're going to solve the problems of those diseases, is the genetic approach the right approach? Do we need to pay more attention to the environment, the social factors, et cetera?

So with those comments, I will conclude. Thank you.

SENATOR FRIST: Thank you, Tony.

(Applause.)

SENATOR FRIST: We've got one microphone. Do we have to microphones? We have two microphones here and would like for people to come to microphones if you have certain questions or comments.

Let me open just by asking the panel a pretty basic question, and that is one that I had to address very early on when we first wrote our legislation, and that is -- and, Hugh, I should probably address it to you first -- in the basic history and physical exam review of symptoms that you do as a physician, by long standing tradition you have the past medical history, and in that you ask questions of family history very early on. You know, what is your chief complaint, and then very soon after that, you go to the family history for obvious reasons. What did your mother die of? What did you father die of? How sick were they? Did they have any other diseases, and the like?

What is the principal difference between giving an insurance company that information and allowing them access to that information as compared to where we are today in terms of a genetic profile? What is the difference between the two?

DR. RIENHOFF: I would say that it's on a continuum in the sense that, first of all, a lot of physicians actually don't take a family history, strangely enough. Maybe if they train at Hopkins they do because there's such an emphasis on that there, but I've been told that a lot don't.

But there are a lot of issues there. The family history is a fairly blunt tool in terms of diagnosis, and a lot of people don't know exactly what their grandfather died of or their grandmother, et cetera, and a genetic test that might identify predisposition, which is, I think, where this question is, from my point of view, it really again just increases the probability of having that particular disease.

So if your mother -- well, I'll use my family as an example. My sister, my mother, and my grandmother have breast cancer. It seems like cancer seems to run in the family. What implications does that have for me as a male on cancer?

Well, actually I don't know the answer to that. That's one I'd love to know the answer. Whether there's a predisposition, I would act on that predisposition. Whether an insurance company would consider me a higher risk for cancer at a younger age, I don't know. It gets back to Tony's question of how good that information is, how predictive it is.

But I think the answer is that the family history is such a poor tool unless there's 25 affected individuals in the family or unless it's a Mendelian disease of the type Sharon was talking about, it probably is very different.

SENATOR FRIST: Other comments on that? It's a pretty basic question, but it's one that a lot of us still struggle with.

Sue.

MS. SIEGEL: Yes, I would agree that it's a sense of gradation, as you were saying, with regards to family histories versus genetic tests, but I have to say that with regards to genetic tests, when you actually have the marker, it's very confirmatory, and therefore, the understanding of how it might pass on to future generations is a little bit more explicit than just the suggestive nature of family histories.

So hence, that subjective portion of it that you were mentioning, Tony, with regards to how the public perceives it and genetic testing, especially when you have the marker is very confirmatory.

DR. RIENHOFF: Well, can I just make one quick comment there?

If you have a family history of sickle cell disease, that's very different than having a positive family history for breast cancer. Most of the common diseases, as Tony said, probably have genes that predispose, but most people don't get the disease that have the predisposing allele. So that's the way they're going to behave.

SENATOR FRIST: Sharon.

MS. TERRY: I would say that ten to 20 years from now this question won't be a question because I think both are going to have tremendous impact. As Neil pointed out, if tests are not valid or efficacious, then they're not as informative.

Even when they are, and even when you do have that mutation for that particular disorder, environment plays a tremendous role. So I think a good family history and good genetic tests will have the same weight and should be protected the same way.

SENATOR FRIST: Tony.

DR. HOLTZMAN: Well, I really have very little to add. For those of you who are confused, I'm Neil and Tony. So it's --

SENATOR FRIST: That's right.

DR. HOLTZMAN: I think the problem of markers that Sue mentioned, one has to, again, reiterate what I said, is it depends on the disease, and sickle cell anemia is much different, let's say, than its use even in a family where there's a history of breast cancer.

I mean if you've got a brother with sickle cell anemia, you have a 50 percent chance of having sickle cell, 25 percent chance of having sickle cell anemia.

For breast cancer with family history, only a very small proportion of all those women with a family history of breast cancer are going to have BRCA-1, and even if they had a BRCA-1 or BRCA-1 mutation, their chance of getting breast cancer is not 100 percent. It's in the order of 50 to 85 percent when you do relatively unbiased studies.

Another point and the last point to make about family history is, again, we've begun to look at how frequently is there a positive family history for common diseases, and seldom does it get over 20 or 25 percent of all the people who get disease, heart disease, cancer, et cetera. So one, again, has to remember that family history, a crude clue, and I think it's cruder than the kinds of genetic tests that should be out on the market and that will be developed, but we still have this large pool of people who are going to get these diseases where family history is not of help.

SENATOR FRIST: Thank you.

The Health Insurance Portability Act of 1996, as most of you know, prohibits a group health plan or an issuer of a group health plan from using genetic information to establish rules of eligibility or continued eligibility and also says that genetic information shall not be treated as a preexisting condition if you don't have a diagnosis of the condition that's related to that.

I say that all as background to ask our panelists as to whether or not they feel that the current privacy regulations being used at HHS, being implemented by HHS are sufficient or should they be changed?

Any comments?

MS. SIEGEL: I'll think about that for a sec.

SENATOR FRIST: Any comments?

All right. Questions, I've got a whole series here I'm going to go through.

Hugh?

DR. RIENHOFF: One thing I would say is that there's some wisdom in that in that sense that the -- and this is to pick up on Tony's comment -- you may inherit a predisposition, but the probability of getting it is so low if you don't have the right environmental conditions, et cetera, et cetera, that it's unfair to discriminate against that person. You should wait until you have the diagnosis.

So I think this underscores what's going to be the genetic nature of these common diseases.

SENATOR FRIST: And this whole idea of predisposition versus current diagnosis or indications of the disease, and this comes from a question, in prohibiting genetic discrimination, it is important to allow insurers access to current health information for purposes of health insurance and prohibit discrimination based on predictive genetic information.

How can you distinguish between a predisposition towards disease and indications of current illness?

To the panel.

DR. HOLTZMAN: Well, maybe I'm missing something in the question. I mean, if there's present illness, then the patient is symptomatic, and one could then use genetic tests to confirm the physician's suspicion of this specific cause or whether genes contribute or are the specific cause of the disease.

When it comes to predisposition, again, one is dealing with healthy people. Can you repeat the second part of the question?

SENATOR FRIST: That's basically it, predisposition toward versus current indications of disease. Is that a clear-cut dividing line to everybody?

MS. SIEGEL: Yes.

DR. HOLTZMAN: Yeah, and I would add one other thing, is that there's predisposition. There's early detection before you're symptomatic, and then there's the point at which you actually have symptoms and you could say you have the disease.

SENATOR FRIST: All right. It sounds pretty clear-cut, but it is difficult in terms of this evolution of disease because you get like infectious disease. You're colonized before you actually have manifestations of the disease. This whole genetic framework is a little bit different, and it's hard; at least it's hard to write in legislation this sharp dividing line.

DR. RIENHOFF: Yeah, and I'll add another layer of complexity. Tony said that these tests could be used in a confirmatory way. If you took cancer, for example, breast cancer, probably the best estimate is that about 30 percent of the cases out there are caused by genes that cause the disease or predispose you, and the other, you know, the remaining 70 percent or so, there's no measurable genetic effect. We shouldn't say it's zero.

So if you get the diagnosis of breast cancer and it turns out you have a variation of a gene that is known to predispose, it doesn't necessarily mean you got breast cancer because of that. So, therefore, the implications for your family members are different. It doesn't mean that particular gene is going to cause disease or is the cause of disease in other family members, including the patient.

SENATOR FRIST: Thank you.

Let's go to the microphone.

AUDIENCE MEMBER: One of the -- is the mic on?

SENATOR FRIST: Is that mic on or has anybody got the volume? If not, yell it out.

AUDIENCE MEMBER: One of the cases that comes up is you ask for life insurance, and when you do, they then give you a physical, and during the course of the physical, one of the analytes might be what is your cholesterol level, and they use that as an indicator to determine how you're at risk.

And Hugh mentioned that this is like that analyte, that is, a genetic profile of you that indicated heart disease or cancer or whatever, which is predictive to some extent, would be a likely indicator that an insurance company would like to use.

Now, you usually waive any privacy you have in order to get the insurance, and I presume absent legislation that barred an insurer from asking for that information, the person who wanted the insurance would not be protected to avoid it, but my question is:

(a) Is that a policy we should have? Because (b) the tests that I'm hearing and the predictive value is so questionable at this state of the art of science that it should not be used because the discrimination would be unfair in that case rather than it would be scientific.

I would welcome to have my premises corrected obviously, but I --