Feature:
Human Gene Banks

Winter, 2005

Garrath Williams,PhD

Lancaster University, Lancaster, UK

The past five years have seen a new wave of interest in the storage of human genetic materials. This interest partly reflects increased knowledge of genetics and partly the hope of further understanding and new applications.

On the one hand, states - and especially their police forces - have seen that quite basic existing genetic knowledge is of great use to them. The ability to identify individuals in terms of a near unique "genetic fingerprint" has already prove forensically very valuable. As the number of samples held in police databases increases, so too does the usefulness of these databases for linking samples found at crime scenes with individual identities. On the other hand, medical and scientific interest in genetics begins with the fact that we have barely begun to decipher the meaning of this identifying data. We can identify some mutations of single genes that result in rare and extreme diseases - a good example is Huntington's disease. But in most cases the connections between human genetic variability and health remain hidden, and we have hardly any knowledge concerning the interaction of large numbers of genetic variations. Given our poor understanding of gene expression, the best weapon to study these is sheer statistical force - to compare health outcomes across large numbers of people with their genetic variations. To do this, new human gene banks orders of magnitude larger than the many well-established gene banks dedicated to the study of particular diseases are needed.

Not only is banking "genes" important - we also need information about the health or phenotype of the sample donor. In the large-scale research gene banks that have attracted much recent attention, this personal information is meant to be continually updated. This has the significant implication that samples cannot be made fully anonymous; otherwise, we could never link new data back to the original sample. Our ability to identify individuals on the basis of genetic samples means that gene banks pose significant privacy issues that will only increase as we become better able to decipher genetic information.

Bioethics has given the most attention to human gene banks for medical research. This is natural, in that bioethics is especially concerned with research and human health. Also much more publicity has surrounded these projects. But something is odd about this emphasis, because the biggest, best funded and most effective human gene banks are forensic ones. the UK National Police Database, for instance, includes samples from 2.75 million people (September 2004) - twice as many as it did two years ago - and is still growing.¹

By contrast, the gene banks that have attracted recent attention remain more promise than reality. The two best known examples are the icelandic Health Sector Database² and UK Biobank,³ neither of which has yet gathered any genetic samples. In the US, an interesting third example is First Genetic Trust, ⁴ a private company that essentially mediates between researchers and individual donors.

The Icelandic database has been controversial for several reasons. Meant to include the entire population of Iceland - 270,000 people - it features medical data on an "opt-out" basis: that is, individuals are not asked for their consent, but may register their dissent and withdraw from the database. So far, only a small minority of the Icelandic population has opted out. Moreover, the database has been licensed, by act of Parliament, for a twelve-year period to the company deCode Genetics, ⁵ closely linked with Hoffman La Roche. as many see it, the genetic heritage and health data of the Icelandic people have been sold to a private company at a low price and with scant regard for the principle of informed consent. However, although the license was granted in 1998, deCode has yet to gather any genetic samples.

By contrast UK Biobank is intended to include more samples - 500,000 - from only a subset of the British population - 45- to 69-year-olds (the age range when many common diseases of Western societies set in). Research subjects, who must specifically consent, will complete lifestyle and health questionnaires as well as grant access to their health care records. Funded by the UK government and its Medical Research Council as well as the Wellcome trust (the world's largest medical research charity), UK Biobank will be open to both public and private researchers on terms yet to be announced.⁶

Two obvious issues for human genetic banking are posed by these cases: commercial exploitation and the consent of individual research subjects. Samples and information have often been spoken of as gifts by the donor (the reference is to Titmuss's work on blood donation in the context of a publicly funded health care system).⁷ "Gift" is ambiguous, though. Donation is a free and consensual act; the sample is donated supposedly with no thought of return or recompense. Yet while people may be happy to give to public health research, this may be because they receive ongoing benefits from that system - so that gifts are not quite so one-sided as may be thought. It is less clear that people are, or should be, so willing to "give" to profit-making entities.

Consent is problematic because these proposed gene banks promise a new form of health research, one that is essentially collective and peculiarly open-ended - aiming to learn more about the relations between genetic variation and health. This seems to render informed consent impossible: researchers cannot tell donors in advance what their sample and data will be used to research. The sheer scale of the projects also makes asking donors about each individual study quite impractical - something one might anyhow think burdensome to individuals and problematic from a privacy point of view. commentators and policy documents, therefore, tend to speak of "blanket" or "broad" consent.

Beyond this, it is not clear that these massive new biobanks represent a sensible priority for research. Gathering high-quality data for such a large number of people seems impractical. Perhaps the best that the large gene banks can do is to facilitate the identification of a much smaller pool suitable for intensive investigation of a particular drug or condition. In this case, how important the gene bank was in the first place is open to question; the many much smaller disease-specific banks would serve just as well. Likewise, the promise of technologies for population genetic screening is unlikely to be realized or to represent a sound public health investment: We already know that most chronic ill-health and premature death in Western societies owes to factors such as diet and exercise patterns, not to mention wealth inequalities. Testing individual genetic predispositions can only be the most minor supplement to addressing those factors.

Doubts of this sort may have slowed private investment in large scale, openended biobanks over the past couple of years - resulting in the failure of deCode to begin collecting genetic samples. Markets and private companies are unlikely to sink large sums into projects whose pay-offs are so uncertain. This leaves biobanking in the hands of public and charitable concerns - where one may think ensuring that investments are directed to public health benefits is much more important. But because a project like UK Biobank is so enormous - a £60 million budget ($120 million), which many suspect is only the beginning - it has escaped the usual processes of scientific peer review. Yet if we are going to place less weight on informed consent than we do for standard medical research - as it seems we must if such projects are to be practicable - then clearly we must place much more weight on scientific and public Gene Banks continued from page ¹ scrutiny. That is, we must be sure that the projects are both scientifically well conceived and reflect proper public priorities for research. My own view, in line with the doubts sketched in the preceding paragraphs, is that such a case has simply not been made.

Interesting parallels and illuminating differences exist between forensic and medical biobanks. Costly as forensic biobanks are, they are certainly much cheaper overall, because they do not require the same sort of detailed data collation as research biobanks. Their benefits, in terms of crime detection if not crime prevention, are much clearer and more immediate. consent has not been thought an important principle in the case of forensic banks - it would be plainly absurd for police forces to have to require consent from someone they have good basis to suspect of a serious crime. At the same time, it is disturbing that forensic databases have attracted even less debate than research biobanks - despite the fact that they pose serious privacy issues and represent a huge potential growth in state power.

While few bioethicists are willing to endorse the icelandic example and bypass consent entirely, it now seems clear that individual consent is not the most important issue posed by this new form of collective and prospective medical research. Much more central is whether research biobanks represent a well-conceived priority. ⁸ This question may be even more difficult to debate than in the case of the forensic databases; yet it surely deserves to be debated more vigorously and more widely than it has so far.

Footnotes

¹ Williams R, Johnson P, Martin P. Genetic Information & Crime Investigation. London: Wellcome Trust; 2004. Available at http://www.dur.ac.uk/p.j.johnson/

² http://www.wellcome.ac.uk/doc_WTD003280.html

³ http://www.ukbiobank.ac.uk/

⁴ http://www.firstgenetic.net/

⁵ http://www.decode.com/

⁶ See the excellent briefing papers on UK Biobank by Genewatch: http://www.genewatch.org/

⁷ Titmuss, RM The Gift Relationship. Oakley A, Ashton J, eds. Rev ed. New York: The New Press; 1997.

⁸ See also Ruth Chadwick "The Icelandic Database - Do Modern Times Need Modern Sagas?" Br Med J 1999;319:441-444.

in collaboration with
Dartmouth-Hitchcock Medical Center

The opinions expressed in the journal, Lahey Clinic Medical Ethics,
belong to the individual contributors and do not represent the institutional position
of Lahey Clinic on any subject matters discussed.