Can we keep our genomes quiet? Some suggestions from the US
18 October 2012
I have posted previously on the logistical difficulties in legislating against genetic discrimination.
The prospect that genetic information not only affects insurance and employment opportunities is alarming enough. But it has many other implications: it could be used to deny financial backing or loan approval, educational opportunities, sports eligibility, military accession, or adoption eligibility. At the moment, the number of documented cases of discrimination on the basis of genetic test results is small. This is probably due to the relatively few conditions for which there are currently definitive genetic tests, coupled with the expense and difficulty of conducting these tests. But genetic discrimination is a time bomb waiting to be triggered and the implications of whole genome sequencing (WGS) are considered in a very interesting and readable report by the US Presidential Commission for the Study of Bioethical Issues Privacy and Progress in Whole Genome Sequencing.
The report delves into two crucial questions: What information about an individual’s whole genome should remain private, and when should it remain private? The Commission explores how, when, and why genomic information should remain subject to clear rules of confidentiality, secrecy, information security, decisional autonomy, and freedom from unwanted intrusion out of respect for individuals. It concludes that policies and laws are needed to ensure protection of individuals’ privacy to secure the trust and participation of the public, and to help realise the full potential of WGS to advance clinical care and benefit society.
Whole Genome Sequencing
WGS determines the complete sequence of DNA in an individual’s cells, a sequence which is unique to every individual. There is an obvious substantial benefit in the increasing ability of geneticists to link variations in DNA with disease. But this process is still in its infancy; current clinical uses of DNA information are limited mostly to specific genetic tests. So for example if clinicians suspect a particular disease with a known genetic cause, such as Huntington’s disease, they can order a genetic test looking at one specific gene among the more than 20,000 genes in the human genome. But as the price of sequencing a whole genome continues to fall (in the USA from $2.5 in 2001 to a predicted $1,000 in the near future), it will soon be less expensive to sequence an entire genome than to perform a few individual genetic tests. Once this happens, whole genomes might be sequenced in lieu of discrete genetic tests, and such information can be stored in a patient’s medical records.
Any legislation that is devised to prevent the misuse of such information has to be agile enough to adapt to evolving technology and social norms around privacy without hampering medical progress in this vital field. The difficulty lies in balancing the majority of the benefits anticipated from whole genome sequencing research which accrue to society as a whole as against the associated risks which fall to the individuals sharing their data.To use whole genome sequencing to discover the changes in deoxyribonucleic acid (DNA) that underlie disease, scientists and clinicians must have access to whole genome sequence data from many individuals. Currently large amounts of patient data are being collected in the course of health, stripped of traditional identifiers, analyzed, and fed into research that might one day improve clinical care. But the sheer amount of information contained in our genomes is what makes whole genome sequence data different from other medical information. As more information about our genomes becomes available, variants that might be revealed by whole genome sequencing include:
- specific known disease variants
- variants of unknown significance (e.g., an unknown variant in the region that increases risk for heart disease)
- nonmedical genetic traits, including hair and eye colour
- carrier status variants, including variants that do not cause disease in the individual but could be passed on, such as mutations for hemophilia or cystic fibrosis
- susceptibility genes, such as those that slightly increase susceptibility to diabetes, heart disease, or some cancers
- genes for conditions with late onset that will not affect an individual until much later in life, such as Alzheimer’s disease or Parkinson’s
WGS also dramatically raises the privacy stakes because it necessarily involves examining and sharing large amounts of biological and medical information that is not only inherently unique to a single person but also has implications for blood relatives. Now that genetic data turns up so much more information than originally sought or even expected, the report recommends that it is even more essential that accessible WGS data should be stripped of traditional identifiers whenever possible to inhibit recognition or re-identification. Only in “exceptional circumstances” should entities such as law enforcement or defense and security have access to biospecimens or whole genome sequence data for non health-related purposes without consent.
Problems arising from disclosure of genetic information
Without such protections, individuals may not want to risk disclosing information about their genetic disposition to certain disorders in order to avail themselves of medical treatment.
Those who are willing to share some of the most intimate information about themselves for the sake of medical progress should be assured appropriate confidentiality, for example, about any discovered genetic variations that link to increased likelihood of certain diseases, such as Alzheimer’s, diabetes, heart disease and schizophrenia. Without such assurance in place, individuals are less likely to voluntarily supply the data that have the potential to benefit us all with life-saving treatments for genetic diseases. Everyone stands to gain immensely from our society taking the necessary steps to protect privacy in order to facilitate progress in this era of whole genome sequencing.
The report cites as an example the story of a woman whose sisters suffered from a disabling genetic disorder “alpha-1 antitrypsin deficiency”. The genetic illness meant that her sisters’ bodies did not make enough of a protein that protected their lungs and liver from damage, which could lead to emphysema and liver disease. The woman in question signed up for a research study into the disease and found that she tested positive, but did not dare to disclose this to her doctor for fear that her insurance company would remove cover. When she finally succumbed to pneumonia she yielded this information, reluctantly, and is now being treated for the condition. But she cannot persuade her brother or son to be tested, for the same reason. Neither of them want any information in their medical records that could jeopardise their jobs or access to health insurance. The existing federal laws prohibiting discrimination of this sort are insufficient.
There are other hazards, of course. A key finding in the report is that current governance and oversight of genomic information varies widely according to location and the circumstances under which it was obtained. For example, only around half of US States offer legal protection against surreptitious testing of an individual without their consent. Such a scenario can arise very easily, as the report suggest:
For example, in many states someone could legally pick up a discarded coffee cup and send a saliva sample to a commercial sequencing entity in an attempt to discover an individual’s predisposition to neurodegenerative disease. The information might then be misused, for example, by a contentious spouse as evidence of unfitness to parent in a custody case. Or, the information might be publicized by a malicious stranger or acquaintance without the individual’s knowledge or consent in a social networking space, which could adversely affect that individual’s chance of finding a spouse, achieving standing in a community, or pursuing a desired career path.
And then there is the problem of incidental information that is revealed by WGS. One example that illustrates this dilemma is the Alzheimer’s risk associated with certain variants of the ApoE gene. Individuals who carry the ApoE4 variant have a higher risk of developing Alzheimer’s disease, but not everyone with this variant will develop Alzheimer’s. Suppose that whole genome sequencing is being performed on a young adult for a breast cancer research study he or she is involved in, and the ApoE4 variant is discovered. Should this finding be returned? The finding is not clinically actionable—meaning that there is not an effective treatment or cure—and it is not certain that individuals with the ApoE4 variant will develop Alzheimer’s disease. Some argue that the only acceptable reason to return an incidental finding is that the finding is clinically relevant and actionable, and the ApoE4 variant’s association with Alzheimer’s disease fails to cleanly meet these criteria. Others argue that it should be completely up to the individual whose whole genome is sequenced to make this decision.
The Commission therefore recommends that researchers, clinicians, and commercial whole genome sequencing entities must make individuals aware that incidental findings are likely to be discovered in the course of whole genome sequencing. The consent process should convey whether these findings will be communicated, the scope of communicated findings, and to whom the findings will be communicated. But the whole consent issue is fraught with problems, not least of all because the standard medical consent form has become so over-legalised and technical it is beyond the comprehension of the average research participant. The report quotes the following reflection from biomedical professor Daniel Masys at the University of Washington School of Medicine:
How does consent change when a person lacks genetic health literacy, [or] when the health condition does not yet exist, but is a future probability, and some of those may be non-treatable conditions? When a health condition does not have implications for you, but it does for your offspring, what are the terms of consent there, especially if your offspring have different views about what they want to know about genetics, and then lastly, for these incidental findings versus disease specific testing..? I’ll just leave you with those questions, as the first of many that you will engage.
In the light of this the Commission recommends that informed consent forms should:
1) briefly describe whole genome sequencing and analysis;
2) state how the data will be used in the present study, and state, to the extent feasible, how the data might be used in the future;
3) explain the extent to which the individual will have control over future data use;
4) define benefits, potential risks, and state that there might be unknown future risks; and
5) state what data and information, if any, might be returned to the individual.
The report is entitled “Privacy and Progress” despite the fact that the word “privacy” does not appear in the U.S. Constitution. Of course the value of privacy is implicitly recognised in the Bill of Rights through provisions guaranteeing anonymity as part of First Amendment freedom of speech, freedom from government appropriation of one’s home (Third Amendment); freedom from unreasonable search and seizure of one’s body and property (Fourth Amendment); freedom from compulsory self-incrimination (Fifth Amendment) and so on. The common law of some states includes a breach of confidentiality tort. Most states recognize one or more right to privacy torts, first proposed in the 1890 article “The Right to Privacy” by Samuel Warren and Louis Brandeis. Nevertheless the report describes this approach to privacy as highly sectoral, and contrasts it with the European regulation of privacy which provides protections that are consistent across different types of data or information. In this report, the Commission focuses on “informational” and “decisional” privacy as they pertain to whole genome sequencing:
We use the term “privacy” in reference to both limited access to genetic information and data, and to the absence of interference with decisions about the collection, use, and sharing of genetic information. A person whose whole genome is sequenced might have both decisional privacy concerns (about who is permitted to decide whether whole genome sequencing data are shared) and informational privacy concerns about whether such data will be shared in confidence, securely, or in de-identified form.
There is no comprehensive federal law in the USA that protects genetic privacy. The Genetic Information Nondiscrimination Act (GINA) prohibits discrimination by employers and health insurers based on the results of genetic tests, but does not provide privacy protections. In addition, GINA does not address the complexity of large-scale genomic data. Many states have laws governing genetic information and some of these laws provide privacy protections, but the laws vary greatly from state to state. Of the all the recently enacted privacy laws, the 1996 Health Insurance Portability and Accountability Act is the federal law most relevant to medical privacy. The law restricts the use or disclosure of health information by health care providers. But it is not clear whether genetic or genomic information on its own is protected health information under the Act. Nor has the U.S. Supreme Court yet established a constitutional right to informational privacy applicable to whole genome sequence data. Although the Supreme Court has addressed privacy rights of biomedical information in the context of the Fourth and Fourteenth Amendments (Sorrell v. IMS Health Inc., 131 S. Ct. 2653 (2011)) there is no case law addressing informational privacy in the context of whole genome sequencing. As far as individual states are concerned, the Supreme Court of California has gone furthest in this area by ruling that individuals are entitled to informed consent, but do not maintain property or privacy rights over cells after they have been removed from their body (Moore v. Regents of the University of California).
“…and Progress in Whole Genome Sequencing”
Inadequate though the current law is, one should not be too alarmist about the potential spread of genomic information. After all, while some experts might be able to determine an individual’s hair color or specific cancer risk from whole genome sequence data (a file of 6 billion nucleotides), these data are not interpretable by the vast majority of individuals. Even if we know that a whole genome sequence is from one individual, we cannot know which of the over 7 billion people on Earth that person is without a key linking the whole genome sequence information with a single person or their close relative. The Commission’s recommendations are therefore conservative on the whole and seek to strike the appropriate balance between privacy interests and the importance of keeping this tremendous resource open for biomedical research. Any legislation or regulatory regime put in place must be flexible enough to accommodate the rapid development of genomic technology. To demonstrate how rapidly things are changing, we only need to look at the understanding of the functionality of each of the genes that make up the 20,000 in the whole genome. A recent major recent advance was made by the National Institute of Health’s Encyclopedia of DNA Elements (ENCODE) project, finding that 80 percent of the genome has a “biochemical function.” For years, that number had stood at 10 percent.
The ENCODE Project is a major step toward demonstrating the function of the whole genome sequence that was determined in the Human Genome Project, much like Google Maps can refine a snapshot of the Earth by showing traffic, alternate routes, and the location of landmarks.
Researchers recently determined a fetal whole genome sequence using a blood sample from the mother, an innovation which would have been unthinkable a few years ago, but will soon be routine in antenatal care. In the light of all of this, the Commission is at pains to preserve what it calls “intellectual freedom” by emphasising that respect for persons implies not only respecting individual privacy, but also respecting research participants as autonomous persons who might choose to share their own data.
Regulatory parsimony recommends only as much oversight as is truly necessary and effective in ensuring an adequate degree of privacy, justice and fairness, and security and safety while pursuing the public benefits of whole genome sequencing. Therefore, existing privacy protections and those being contemplated should be parsimonious and not impose high barriers to data sharing.
Without this counterbalancing regard for the importance of information sharing, vital discoveries risk being shut down by individual privacy interests. The report cites the 2010 case of Beleno v. Texas Department of State Health Services. Parents sued the state the health services, which had collected and stored newborn blood samples for research purposes, without seeking parental consent. The parents argued that the lack of proper consent was a violation of privacy. The out-of-court settlement that was reached resulted in the destruction of 4 million similar specimens that had been collected without parental consent.
Key specific recommendations include:
- Holding all those who come into contact with WGS data accountable by law for its security.
- Developing standardised and fully informed consent procedures.
- Integration of WGS data into health records.
- Federal and State legislation to protect WGS data, regardless of how it was obtained.
- Prohibition of WGS without the consent of the individual concerned.
The report has been presented to the Obama Administration.
Sign up to free human rights updates by email, Facebook, Twitter or RSS