Genetic prophecy and genetic privacy.

We like to think that knowledge is power and that the more knowledge we have about our health the more choices we will be able to make to enhance our quantity and quality of life. But is this always true?

Would you really want to know how (or when) you are likely to die or to lose your mind? And even if you do, would you want your family, your business clients, your law partners, or your life or health insurance company to have this information?

In a September 1995 cover story in the Sunday New York Times Magazine, Charles Siebert writes about his own confrontation with these questions. His family has a history of hypertrophic cardiomyopathy (HCM), an incurable type of heart disease characterized by a thickening of the heart walls that usually leads to an, early death. Geneticists cannot treat this condition, but by examining the DNA of family members (doing a pedigree), they can identify those who carry genes that indicate whether or not they are likely to suffer from this condition.

Siebert went to the National Institutes of Health in Washington, D.C., thinking he would probably volunteer to be part of the research project looking for this gene. But as he learned more and more about genetics, he began to reconsider. Genes predispose, but they do not dictate our futures. We are aU made up of genes, but genes are not all that we are.

Environmental factors may determine our future as much as our genes, and our interaction with our environment is unpredictable. Siebert's father, for example, died of HCM. But his cousin had been killed by a falling piece of scaffolding long before the disease could manifest itself.

Siebert wondered whether having a physician prepare a person's "genetic report card" is always a good thing, and he learned that genetic information can be "toxic" and undermine our ability to go on with our lives and plan our futures. In the end, he decided not to have a genetic test for the HCM--that he "didn't need to know." In his words:

I would drive all night to get back home--to get back, in a sense, to my own unknown. I had a right to that. And not simply for the privilege of hope but of pessimism as well--my own rendition, however misguided, of how and when I'm going to die.(1)

Charles Siebert's choice is still unusual, but it is one all of us will confront within the next few years as the new genetics produces bushel baskets of new genetic screening tests that will invade both medicine and public health. Indeed, the new genetics is the centerpiece of breathless predictions of both a new medicine and a new public health.

Molecular medicine, based on deciphering the genes of a patient instead of diagnosing the patient based on signs and symptoms, is said to be just around the corner. And using genetic screening in large populations to detect potentially preventable problems, such as breast cancer, is the dream of many public health professionals. But it will be some time, if ever, that either treatments or preventive measures for genetic predispositions are available.

In the meantime, most genetic information will be used primarily to identify and possibly stigmatize people like Charles Siebert, who carry genes that likely increase their risk of developing specific diseases, not only like HCM, but also cancer and dementia. This means that genetic screening and testing will cause harm and may open up new areas of litigation, especially in the area of individual privacy.

Genome as a Future Diary

Genetic information can be seen as uniquely private information, even more personal than other medical information such as HIV status or mental health, for at least three reasons: It can predict an individual's likely medical future; it divulges personal information about one's parents, siblings, and children; and it has a history of being used to stigmatize and victimize individuals.

The highly personal nature of the information in one's DNA can be illustrated by thinking of DNA as containing an individual's coded "future diary."(2) A diary is perhaps the most personal and private document a person can create. It contains a person's innermost thoughts and perceptions and is usually hidden and locked to ensure its secrecy.

Diaries describe the past. The information in one's DNA can be thought of as a coded probabilistic future diary because it describes an important part of a person's unique future and, as such, can affect and undermine our view of ourself and our life's possibilities. Unlike ordinary diaries that are created by the writer, the information in one's DNA, which is stable and can be stored for long periods of time, is largely unknown to the person.

Most of the code cannot now be broken, but parts are being deciphered almost daily. As decoding techniques get better, and if one's DNA is deciphered without permission, another person could learn intimate details of someone's likely future that even the individual does not know.

Deciphering a genetic code also provides the reader of that code with probabilistic health information about the individual's family--especially parents, siblings, and children. Finally, genetic information and misinformation has been used by governments (U.S. immigration and sterilization policies and German Nazi racial hygiene policies, for example) to viciously discriminate against those perceived as genetically unfit and to restrict their reproductive decisions. The privacy issues involved in the new genetics can be illustrated by considering two genes identified within the last few months that code for early-onset Alzheimer's disease. People who carry this gene are not only very likely to develop the disease but are likely to begin to experience symptoms in their forties. There is currently no way to prevent the disease from manifesting itself and no cure for it. Of course, the hope is that once the genes responsible for creating the bodily conditions that permit or encourage Alzheimer's disease are identified, it will be easier to find treatments and prevention strategies. We would all welcome these developments. Meanwhile, of what use is the knowledge that one carries the early-onset Alzheimer's gene?

Also of great interest are genes that predispose one to cancers of various types. The first genes that are likely to be promoted for large-scale screening are the two breast cancer genes that have been identified to date, BRCA1 and BRCA2.

A woman who carries one of these genes will not necessarily develop breast cancer, but her risk is very high--perhaps over 80 percent--that she will. Since there is no way to prevent this disease, what good is knowing you will probably get it in the future? Charles Siebert decided against HCM testing. What should women do?

Science historian Robert Proctor highlights the psychological issues presented by screening for cancer genes:

No one knows what the psychological impact of such tests may be: there is a novel element, insofar as these are diagnostic technologies being developed in the absence of any appropriate therapy (short of watchful waiting or prophylactic coloectomy, ovariectomy, and mastectomy). Will women found not to have a breast cancer gene feel themselves not to be at risk? Will people not predisposed to colon cancer feel not obliged to exercise or cut down on fat and increase dietary fiber?(3)

Moreover, since genetic tests can be performed at any age (all one needs is a DNA sample), the question of fetuses, newborns, and minor children must also be faced. Disputes about the appropriateness of testing children for late-onset diseases that cannot be prevented have already broken out.

In one survey, for example, only 2 of 49 geneticists said they would test a five-year-old for the Huntington disease (HD) (an unpreventable degenerative disease that usually strikes people in their fifties) gene at the request of the child's parents, whereas more than 50 percent (100 of 189) of pediatricians would comply with this request.(4) There seems little to gain for the child (who will be labeled as already "sick" and lose his or her right to decide whether to be tested for the HD gene). Nonetheless, the pediatricians seem to believe that testing children is justified by the potential benefits to parents (enabling them to "adjust" to the implications of likely future disease in their offspring).

Genetic Privacy and the Public

The U.S. public is as confused as its medical professionals. In a May 1995 Harris poll, for example, although only 27 percent claimed to have read or heard "quite a lot" about genetic tests, 68 percent said they would be either "very likely" or "somewhat likely" to undergo genetic testing, even for diseases "for which there is presently no cure or treatment."

Perhaps most astonishing, 54 percent found it either "very" or "somewhat acceptable" to develop a governmental computerized DNA bank with samples taken from all newborns, and their names attached to the samples.(5) Although this survey reflects uninformed opinion, it raises serious privacy concerns that the public may ignore to its own peril.

The leaders,of the Human Genome Project, a worldwide scientific undertaking to map the 23 chromosomes of the human genome, were not so sanguine. At a 1991 congressional hearing, for example, National Institutes of Health Director Bernadine Healey testified that "like all powerful tools, genetic information can be misused and abused. Discrimination based on genotype must be prohibited as a matter of basic civil rights." At the same hearing, James Watson, the first head of the U.S. Human Genome Project, added, "The idea that there will be a huge databank of genetic information on millions of people is repulsive."(6)

What can be done to safeguard the genetic privacy of individuals in the genetic age?

One laissez-faire strategy is to let the market determine what genetic tests are done and have breaches of privacy remedied in civil lawsuits. The problems with this strategy are that lawsuits for breaches of privacy have not often been pursued because the private information is usually made known to even more people in the process) and that we should use the law to prevent predictable harms if we can. I believe that we can achieve consensus when these issues are understood and that we can and should act on this consensus to protect the genetic privacy of all Americans while it is still possible to do so.

One way to protect genetic privacy is to codify rules for "gene banks" and computerized records based on genetic analysis, but this is too narrow.(7) Although rules for gene banks are needed, it is very difficult to define what a "gene bank" is, and regulation at this level alone permits genetic analysis to be done, and information to be disclosed, in settings that do not involve the storage of any DNA samples.

Regulatory approaches that have been adopted to date include an Equal Employment Opportunity Commission (EEOC) guideline on genetic discrimination and a handful of state statutes that seek to limit the use of genetic information. In March 1995, the EEOC issued a clarification of the definition of "disability" that included discrimination on the basis of genetic information. The example the EEOC gives is informative:

CP's genetic profile reveals an increased susceptibility to colon cancer. After making CP a conditional offer of employment, R learns about CP's increased susceptibility, to colon cancer. R then withdraws the job offer because of concerns about matters such as CP's productivity, insurance costs, and attendance. R is treating CP as having an impairment that substantially limits a major life activity. Accordingly, CP is covered by the third part of the definition of "disabifity."(8)

The most recent state to adopt legislation relating to genetic privacy is Oregon. That legislation requires, among other things,

* that informed consent be obtained before genetic testing,

* that genetic information not be retained without authorization,

* that employers not use genetic information to discriminate against individuals (consistent with EEOC guidelines), and

* that insurance companies not use genetic information to "reject, deny, limit, cancel . . ." policies for medical expenses.(9)

Genetic Privacy Act

For the short term, a state-by-state approach is likely. But genetic information cannot be confined by state borders. We will need federal legislation to effectively protect genetic privacy.

My colleagues Leonard Glantz and Winnie Poche and I believe that we not only need uniform rules for the collection, analysis, and storage of DNA samples, but we also need rules for storing and disclosing information derived from DNA samples. These rules must cover adults, children, incompetent persons, dead bodies, and fetuses and embryos. To codify rules for genetic privacy and make them uniform throughout the United States, we have drafted a proposed federal law, The Genetic Privacy Act of 1995.(10) The core of this act protects private genetic information," which is defined as

any information about an identifiable individual that is derived from the presence, absence, alteration, or mutation of a gene or genes, ,or the presence or absence of a specific DNA marker or markers, and which has been obtained:

(1) from an analysis of the individual's DNA; or

(2) from an analysis of the DNA of a person to whom the individual is related.

The act prohibits individuals from analyzing DNA samples unless they have verified that written authorization for the analysis has been given by the individual or his or her representative.

The individual has the right to

* determine who may collect and analyze DNA,

* determine the purposes for which a DNA sample can be analyzed, know what information can reasonably be expected to be derived from the genetic analysis,

* order the destruction of the DNA sample,

* delegate authority to another individual to order the destruction of the sample after death,

* refuse to permit the use of the sample for research or for commercial activities, and

* inspect and obtain copies of records containing information derived from genetic analysis of the sample.

A written summary of these principles (and other requirements under the act) must be supplied to the individual by the person who collects the DNA sample. The act requires that the person who holds private genetic information in the ordinary course of business keep such information confidential, and the act prohibits disclosure of private genetic information, unless the individual has authorized disclosure in writing or disclosure is limited to access by specified researchers for compiling data.

The act deals explicit with three areas likely to breed the most controversy and consequently the most litigation: genetic testing of children, research using identifiable stored DNA samples, and informing relatives of their increased risk of genetic disease based on diagnosis of a genetically related person.

The act protects the privacy of children by not permitting genetic analysis of a child under age 16 unless

* there is an effective intervention that will prevent or delay the onset, or ameliorate the severity, of the disease; the intervention must be initiated before the age of 16 to be effective; and

* the minor's parent or guardian has received the disclosures required and signed the authorization.

Minors aged 16 and 17 must themselves consent to any analysis of their DNA samples and must be "accompanied by a parent or other adult family member" when they are counseled regarding the DNA analysis and its possible consequences.

Research on identifiable DNA samples stored in a "genetic databank" or "gene bank" "often no more than a collection of samples in a clinic) cannot be conducted without specific authorization of the person if the sample can be traced back to that person. "If the sample has been stripped of all identifiers so that it is impossible to link it with a person, research can be conducted on the sample without specific authorization.

As to the issue of the geneticist's "duty to protect" other family members by informing them of a condition discovered in a patient, the act opts for protecting privacy by prohibiting disclosures to others. The policy reasons for this provision of the act are both philosophical and practical.

Philosophically, the genetic information is and should remain private. People in this sense have what Charles Siebert exercised--a "right" not to know as well as a right to be the only one who knows. Moreover, carrying a particular gene does not put anyone else at risk.

Family members do have a moral obligation to other family members and should be encouraged to share genetic information that may have an impact on other family members. But their obligation is not legal and should not be made so by statute.

As a practical matter, defining the scope of such a "duty" would be virtually impossible. Would it only be related to reproductive decisions? Would it only include genetic conditions for which there were treatments? How would it work, for example, in the area of the breast cancer and colon cancer genes? In regard to the gene for early Alzheimer's?

Ultimately, we believe it is good social policy to assume that family members will act benevolently toward each other in these circumstances, even if there will be some instances where this is not the case. There is, and should be, no professional obligation to disclose genetic information to relatives.

Regarding a broader social policy issue, the act itself does not prohibit the use of genetic information by employers and insurance companies (because this is a separate problem from privacy). It is, however, reasonable public policy to prohibit both employers and health insurance companies from using genetic information in making hiring and coverage decisions.

Congress should act now to protect genetic privacy. While we wait for congressional action, states should act, and private companies and practitioners should voluntarily adopt these privacy rules as their own.

The gene has become more than a piece of information; it has become "a cultural icon, a symbol, almost a magical force. DNA seems to be the locus of the true self, therefore relevant to the problems of personal authenticity posed by a culture in which the `fashioned self' is the body manipulated and adorned with the intent to mislead."(11) To the extent that we accord special status to our genes and what they reveal, genetic information is uniquely powerful and uniquely personal, and thus merits unique privacy protection.

The Human Genome Project has devoted approximately $3 million a year for the past five years to explore the legal, ethical, and social policy issues raised by the project. The Genetic Privacy Act is one of the products of this funding.

In addition, the Institute of Medicine's Committee on Assessing Genetic Risks has made more than 225 specific recommendations dealing with genetic screening and testing, virtually all of which are reasonable.(12)

We know the complex privacy and policy issues that come with the new genetics. The challenge is to act now to try to maximize the benefits and minimize the "toxic" harm that will come to all of us from our newly acquired genetic knowledge.

Notes

(1) Charles Siebert, The DNA We've Been Dealt, N.Y. TIMES AUG., Sept. 17,1995, at 50. (2) See generally GENE MAPPING: USING LAW & ETHICS AS GUIDES (George J. Annas & Sherman Elias eds., 1992). (3) ROBERT PROCTOR, THE CANCER WARS 239 (1995). (4) Tony Smith, Genes & Ethics, 311 BRIT. MED. J. 574 (1995). (5) Harris Poll #34, May 29,1995. (6) Christopher Anderson, Evolution of a Gadfly, 353 NATURE 686 (1991). (7) George J. Annas, Privacy Rules for DNA Databanks: Protecting Coded "Future Diaries," 270 JAMA 2346 (1993). (8) EEOC COMPLIANCE MANUAL [sections]902.8 (Mar. 1995). (9) S. 276, 68th Leg., Or. (1995). (10) George J. Annas et al., The Genetic Privacy Act and Commentary, Boston University School of Public Health, Boston, MA (1995) (available at http://www-busph.bu.edu/Depts/ HealthLaw/). (11) DOROTHY NELKIN & M. SUSAN LINDEE, THE DNA MYSTIQUE: THE GENE AS A CULTURAL ICON 2 (1995). (12) COMMITTEE ON ASSESSING GENETIC RISKS, INST. OF MEDICINE, ASSESSING GENETLC RISKS: IMPLICATIONS FOR HEALTH AND SOCIAL POLICY (National Academy Press, Washington, DC, 1994). (13) George J. Annas, Editorial, Genetic Prophecy and Genetic Privacy: Can We Prevent the Dream from Becoming a Nightmare? 85 AM. J. PUB. HEALTH 1196 (1995).