April 21, 2000
Thank you for that kind introduction. I am delighted to be with you this morning. Regrettably, my schedule has not permitted me to attend much of this meeting. I am impressed, however, with the breadth and depth of the topics that you have addressed. Few programs offer coverage that ranges as broadly as this one, from technical scientific information to the legal and ethical implications of these advances.
Many of you probably are beginning to feel a bit saturated with this flood of new information. After years of teaching epidemiology to medical students, I have come to recognize the early warning signs of audience overload. There is the familiar glazed-eyed expression, complemented by the occasional bobbing of heads fading in and out of consciousness. As a group, you do not appear quite at that point, but I do not want to be held accountable for sending you over the edge.
I have chosen, therefore, to deliver a talk this morning that will not add to your burden of technical information. Rather, I would like to take a few minutes to reflect with you upon the broader societal implications of the advances that you are hearing about at this meeting. I do not pretend to have any great insight into these matters. My scientific career has involved little more than minor forays into the issues discussed today. Moreover, I am at best a casual reader of the evolving literature on these topics. If one were searching for a novice in the field, you could probably do no better than me.
There is a good chance, therefore, that nothing that I have to say to you will have any merit. If so, however, you will have wasted only a few minutes in an otherwise productive day. On the off chance that these remarks provide even a kernel of insight, I will feel completely satisfied.
Let me begin with an assertion. We are about to enter a period of time without any parallel in the advancement of knowledge concerning the human condition. It may, in fact, rival any period of time in the rate of scientific progress. I include here all branches of science — from the golden era of mathematics during the 17th century, to the dawn of chemistry in the 18th century, to the flowering of biology in the 19th century, to the explosion of physics in the early 20th century. Each of these periods was marked by a revolution that fundamentally created new paradigms to explain the world around us. These new paradigms, in turn, allowed us to manipulate the world in increasingly powerful ways. In so doing, we have reduced famine, prevented and treated disease, and improved the quality of life. At the same time, of course, we have polluted the air, land and sea, driven some animal species to extinction, and produced weapons of mass destruction. With each revolution, we have derived extraordinary benefits and, at the same time, exacted significant costs.
Please be assured that I am not arguing against scientific advancement. Indeed, I consider my entire life devoted to the cause of advancing science. My point is that the power of scientific knowledge is transforming — either for good or for harm. We, of course, hope that the benefits far outweigh the costs. Our experience suggests, however, that despite our best intentions, some level of harm should be anticipated.
The lessons of prior scientific revolutions find no greater relevance than today. For we stand on the threshold of a comparable, indeed I would say greater, transforming moment in science. The decoding of the human genome, one of the monumental achievements of molecular biology, will usher us into the so-called post-genomics era. In a matter of a few years, for the first time in human history, we will have the key to deciphering much of what makes us who we are as individuals. Where will this newfound knowledge take us? Will it lead us to healthier, happier, longer lives? Will it reduce human suffering and improve the quality of our existence? Or will it lead to an endless parade of moral dilemmas? Will it place at our disposal the ultimate tools for human malfeasance?
Let us start with the positive side of the equation. The decoding of the human genome will allow scientists to relate specific genes, or more likely specific combinations of genes, to predisposition to certain diseases. That is to say, we will be able to examine a person's DNA and estimate the likelihood that they will develop a variety of afflictions, from depression, to high blood pressure, to heart disease, to multiple sclerosis, to various cancers. For most conditions, these genetic patterns will not allow perfect prediction. In other words, we will not be able to establish with certainty that an individual will develop the disease in question. Rather, it will be a matter of probability„establishing relative likelihoods of disease development.
Even without a guarantee of disease development, however, this information will be incredibly powerful. For example, it will allow screening programs for the early detection of illness to be focused on those at unusually high risk, rather than applied inefficiently and with limited effectiveness in the general population. Such focused screenings might include mammography for breast cancer, sigmoidoscopies for colon cancer, and imaging of coronary arteries for heart disease. Detection through screening would allow treatment at an earlier stage of disease, affording less extensive therapy and a higher likelihood of cure.
Recognition of some of these predispositions might even occur prenatally, allowing the decision to be made about whether the pregnancy should be carried to term. At present, there are relatively few diseases, such as Down's Syndrome, for which prenatal screening is possible. In the post-genomics era, we can expect that list to grow substantially. One can imagine, for instance, that we will be able to detect those at risk for childhood cancer. Of course, establishing those conditions for which the postnatal pain and suffering is so great as to justify terminating a pregnancy raises serious questions. I will return to this issue later in the talk.
The ability to identify those persons at high risk of an illness serves further purposes. For instance, persons with an inherited predisposition to a disease might be counseled to avoid environmental exposures that further increase risk. As an illustration, persons with a genetic susceptibility to diabetes, might be the special focus of dietary interventions and efforts to control obesity. When the techniques of gene therapy improve, the susceptibility genes may become the targets of strategies to lower risk by altering genetic profiles. Among persons with high-risk alleles at gene sites that predispose to Alzheimer's disease, for example, we might develop strategies to change these alleles to low-risk alternatives. To date, gene therapy remains more promise than practicality, but there is every reason to believe that the technology will improve sufficiently to achieve the desired genetic modifications with minimal side effects.
No doubt, my list of benefits—early detection of risk to allow enhanced prevention, conventional treatment and gene therapy—is incomplete. As impressive as this list may be—as dramatic as these societal benefits might be—surely there will be applications that cannot even be imagined today. My goal here is not to enumerate a complete inventory of the advantages to be expected, but rather to suggest the types of enhancements that might result. I emphasize that most of these advances remain theoretical, yet the distance between what is conceivable and what is achievable in this realm is narrowing every day.
Now, as we speculate about the positive outcomes that will emanate from this work, so must we anticipate the negative as well. In some respects, the dark side of this scientific revolution is simply the mirror image of the bright side. Consider for example, the ability to predict with some accuracy the risk of developing some fatal disease. An immediate question that arises is who should have access to such information? Perhaps the most obvious recipient is the individual involved. Even this is not so clear-cut, as the information might be disturbing to some people, especially if no preventive or curative treatment is known. Although such a circumstance might seem speculative, it already is a present-day reality.
Huntington's chorea is a fatal hereditary disease of the nervous system. It is the condition perhaps most widely recognized as the killer of Woody Guthrie. The gene defect for this condition is now known, and asymptomatic family members of affected individuals can be tested for the defect. At the same time, there is no effective treatment for this disease. Someone who tests positive for the abnormal allele, therefore, knows that they will eventually develop the disease, but is powerless to do anything about it. Some individuals might prefer to have such information to plan accordingly. Others would view the information as simply extending the period of time that the disease rules their lives and thus would prefer not to have such information. Huntington's chorea, where a single gene is involved, and the presence of the abnormal allele is synonymous with eventual disease occurrence, presents clear answers. For most diseases, however, the information will not be so clear-cut. Multiple different genes will be involved in these diseases and they will influence, but not completely determine, risk. The risk of disease occurrence will not be measured in absolutes; it will be assessed in likelihoods. How will people process this information? For example, suppose the presence of an abnormal gene increases the lifetime risk of colon cancer from three percent to ten percent. A person with the affected gene clearly is at increased risk by a factor of three. Still, there is only a one-in-ten chance that this individual will actually develop the disease. Is this level of absolute risk sufficient to increase screening efforts, with the associated costs, discomfort and chance of false positive findings? This is a subjective question and we might reasonably expect rational people to come to differing answers. Imagine how much more complicated the situation will become when there are literally scores if not hundreds of abnormal genes that can be screened simultaneously. We can imagine the situation in which virtually every person is found to be at risk for some affliction, and in many instances, for multiple adverse outcomes.
Even more problematic will be the question of who else has access to this information. For example, will life and health insurance providers have such access? Might health insurers treat specific gene profiles as pre-existing conditions and deny coverage? Will life insurance companies rate applicants' risks based upon genetics, in the same manner that they now rate persons by age, or smoking status, or other attributes? After all, these genetic profiles will translate into probabilities that are the fundamental tools of actuarial science. How does one balance the applicant's right to privacy with the company's need to determine risk and assign cost fairly?
What about others? Will employers have access to these genetic profiles? Just as many employers now require psychological testing and drug screening, it is not far-fetched to believe that they will want to have access to certain genetic data. Such information might be used by them to predict anticipated disabilities, longevity of employment and health care costs. How does one balance the employee's right to privacy with the company's right to optimize its workforce? What about others? Should a prospective bride or bridegroom have access to this information about their prospective mate? When one can define "for better or for worse," at least from a health point of view, with reasonable accuracy, should marital vows be so informed? If such information is easily obtained, can one anticipate a powerful new form of natural selection? That is, will persons with favorable genetic profiles become "hot commodities"? If so, these individuals will have a selective mating advantage, and over generations we would anticipate powerful selection toward their traits. In contrast to natural selection—where the environmentally adapted survive to breed—here we would observe a much more direct, rapid, and unnatural selection.
What about cost? Advances in technology may ultimately produce such genetic screening at modest cost. This, of course, could be affected greatly by whether private companies are permitted to patent any of this information. If cost does become prohibitive, the divide between the "haves" and the "have-nots"—forever present in relation to health status—will reach unprecedented dimensions. Those who can afford to obtain this genetic information and act upon it would be able to detect, prevent and treat diseases, but the poor might be closed out of any benefit from the technology. Economic gaps in quality and longevity of life would become even larger than they are today. Since ethnicity and race in our society are so strongly correlated to economics, the racial and ethnic disparities in health would be expected to widen over time.
Although one can imagine other untoward consequences of this technology, I will stop here in the interests of time. Again, my purpose is not to focus on either the positive or negative dimensions, but to provide a balanced picture. To report only the benefits or only the dangers would not do justice to the topic. As we are swept uneasily by the tide of the genetic revolution, we must anticipate all of the possible consequences and try to be as prepared as possible for them.
In the final analysis, I am an optimist. I believe that as we better understand our world and ourselves, we can make more informed decisions. Even if one was not such an optimist, however, one must face the future with a heavy dose of realism. As our genetic knowledge base grows, we will confront possibilities never before imagined in human history. As we celebrate the benefits, let us also be vigilant to the potential hazards. Thank you very much.