Spring, 2005
In “Human Gene Banks” (Lahey Clinic Medical Ethics, Winter 2005), Dr. Garrath Williams raises a fundamental question about recent efforts to build general population human DNA banks for research: Is the money needed to develop these resources worth it? Looking at a broad sample of efforts around the world suggests that the answer is a resounding “no.”
There have been numerous efforts to develop population based DNA and data banks. In addition to those discussed by Dr. Williams in Iceland and the UK, projects were announced in Tonga, Sweden, Latvia, Singapore, Estonia and Sardinia, and several have been started in the United States.1 Most of these efforts combine public or quasi-public (e.g., nonprofit hospital) resources with commercial interests. In the US, numerous firms are in the business of procuring and storing human biosamples for research, including but by no means limited to Ardais, GeneLink, First Genetic Trust, GenomicsCollaborative, DNA Sciences, Integrated Laboratory Services, Zoion Diagnostics, Novagen, the National Disease Research Interchange and Phylogeny. With the exceptions of the MRC/ Wellcome project in the UK and the Marshfield Clinic’s Personalized Medicine Research Project (MPRP), the population efforts have been driven by commercial firms.
In most cases, these efforts have failed relatively early due to financial and political problems. A look at some examples of what has worked and not worked so far supports Dr. Williams’s contention that the best route to take for genetics research is smaller, narrowly focused repositories. Two successful efforts at building and operating a repository include that in Umeå, Sweden, and Marshfield’s PMRP. Uman Genomics is a company established to commercialize an existing tissue bank and related medical records held by the Medical Biobank of Umeå. The bank was created as part of a large population study of cardiovascular disease and diabetes, and is owned and operated by Umeå University and Västerbotten County Council.2 The company has exclusive commercial rights to the bank, which includes blood and plasma from more than 100,000 individuals and counting.
The second success is the Marshfield Clinic’s PMRP. This effort is focused on the prospective collection of DNA, with broad consent, from people in the Clinic’s mid-Wisconsin catchment area. The PMRP was funded by state and federal grants.3 Since its start in 2002, more than 20,000 individuals have agreed to participate.
In contrast, two efforts to build or operate a repository that failed may be compared and contrasted to the above. First is that of
Consent is generally seen as an ethical imperative for the compilation of human DNA and medical data banks.
Framingham Genomics, a company started to commercialize the extremely rich research data (including blood samples) on people in the 57-year-old Framingham Heart Study. This large epidemiological study has collected information over time on two generations of residents of Framingham, Massachusetts, encompassing over 10,000 individuals.4 The study, funded by the NIH’s National Heart, Lung, and Blood Institute (NHLBI), has a DNA repository including over 5,000 of the subjects. Framingham Genomics banked on securing exclusive commercial access to the project’s resources and data; the project was abandoned when the NHLBI refused to grant such broad rights, suggesting such exclusivity conflicted with their public mission.
Second is the population effort by Iceland’s deCODE Genetics. deCODE proposed legislation allowing the compilation of the population’s medical record data (going back some 15 years, plus comprehensive prospective collection) into a centralized Health Sector Database (HSD). An exclusive license was granted to deCODE allowing the company to build and operate the HSD, and, critically, to link the HSD with an extremely complete genealogy on the Icelandic population and with a proprietary dataset of genetic markers (this last developed with consent from subjects). This combined database, called the Genealogy Genotype Phenotype Resource (GGPR) could be used for research. The HSD has not been built, and it is not clear that it will be. Independent of its efforts to conceptualize the GGPR, deCODE has performed smaller studies targeted on a range of specific diseases. They have developed rich databases and networks of clinicians that enable them to do gene discovery as well as other types of research, now even including clinical trials. For example, deCODE’s genetic studies identified a gene, defects in which put patients at high risk of heart attacks. Knowing the mechanism of disease, they identified and licensed a drug candidate from Bayer, and are now running a clinical trial in Iceland on the drug.5 Given the success of their smaller, disease-focused research, the estimated $100 million it would have cost them to develop a computerized medical record system, install it in clinics and hospitals across the country, and pay for the past abstraction of medical records is not likely worth it.6
These four projects, perhaps exemplary of the kinds of human DNA banking activities that have been pursued, suggest several things. First, consent is generally seen as an ethical imperative for the compilation of human DNA and medical data banks. deCODE, which proceeded with a waiver of consent but an opportunity to opt out (totally or by individual doctor visit), is the lone exception to this. Even for some retrospective efforts (using tissues collected in years past), researchers have ensured that the scope of past consent would encompass the types of things proposed in the future, and, if not, required a new consent from subjects for new uses that exceed those disclosed in the original informed consents.
Of course, consent can be problematic, because it adds to the cost of research, it can impute biases if some people are more or less likely to participate (for example, in Iceland, it was reported that psychiatric patients were more likely to opt out; in other cases, women are more likely to opt out of having their medical records included in research), and it can result in low participation rates. Nonetheless, consent is the best way to show respect for persons and their rights to privacy and to choose whether and in what ways they will participate in research.
Second, privacy is universally acknowledged as requiring protection, particularly for research involving human genetics. Methods of protecting privacy range from anonymization (stripping all links and identifying information), use of firewalls to prevent subject identification, data smoothing to help mask identities and data (or linking code) encryption. In Europe, researchers are required to use Data Protection authorities, which are governmental boards that ensure adequate protections are in place for use of many different types of private information. Third, commercial pressures and intellectual property ownership issues can determine whether a project succeeds or not. Firms that have been highly involved in official gene banking efforts often demand exclusive commercial rights to the resources (the DNA banks and related databases) developed at their expense. Exclusivity is exclusionary, and it cuts against the grain for public institutions and for academic or clinical researchers to agree to such conditions. The exclusivity demanded by Framingham Genomics was a deal breaker, and the exclusivity granted to deCODE was a sore point for many critics.
As suggested by the numerous efforts described here, many believe there is money in human genes, but few efforts have succeeded thus far in developing generalized population based research human gene banks. Patents on human genes, while arguably necessary to foster downstream research on therapies, are far removed in technology and time from a profitable product. Thus, the investments have largely proved thus far to be commercially too risky and may only proceed if development costs are supported by public or foundation grants, as seen with both Uman Genomics and the Marshfield Clinic. This does not imply that no commercial entities are or will be successful, but it is too soon to predict whether firms—such as Ardais7—will be around for the long haul.
Jon F. Merz, MBA, JD, MD
Department of Medical Ethics University of Pennsylvania School of Medicine
Editor’s note: Dr. Merz received research support from deCODE Genetics and serves on the Ethics and Security Advisory Board for the Marshfield PMRP. The opinions expressed here are his own.
Footnotes
1 Merz JF. On the intersection of privacy, consent, commerce and genetics research. In:Knoppers BM, ed., Populations and Genetics:Legal Socio-Ethical Perspectives. New York: KluwerLegal Int’l, 2003.
2 http://www.umangenomics.com/.
3 http://www.mfldclin.edu/pmrp/.4
4 http://www.framingham.com/heart/.5
5 http://www.decode.com/.
6 Merz JF, McGee GE, Sankar P. “Iceland Inc.”? on the ethics of commercial population genomics. Soc. Sci. Med. 2004;58:1201–1209.
7 http://www.ardais.com/
It is too soon to know how important a role commercial activity will play in developing even these smaller scale gene banks and in delivering the oft-made promises of beneficial products and technologies.
Response: Professor Merz’s reply usefully widens our discussion of genetic banking to smaller, disease-specific collections. As he argues, these are much more practicable and more likely to bring about results than the much hyped national collections. It is too soon to know how important a role commercial activity will play in developing even these smaller scale gene banks and in delivering the oft-made promises of beneficial products and technologies. His examples suggest that the projects most likely to succeed rest on large scale public investment.
Professor Merz rightly highlights the difficult question of how, and on what terms, such resources should be commercialized. A great deal will hinge on the particular nature and circumstances of the bank. While Merz asks how likely is it that commercial investment will make the research possible, my emphasis was upon how far public and charitable resources ought to be given over to such research.
Ruth Chadwick has recently stressed an interesting disjunction raised by the promised applications of genetics to health.1 A rhetoric of personalized medicine promises drugs tailored to a person’s genetic makeup and dietary and lifestyle advice to enhance the sphere of personal choice. Genetic science heralds, then, a greater individualization of health. At the same time, such technologies can only be developed by large, collective endeavors that rely on extensive support from public bodies.
It is a testament to our sense of health as an individual rather than a social matter, that we are happy to nurse the prospect of individualized benefits, so marginal compared to those that might be gained from addressing, say, the “toxic food environment” facing our children (to take just one example). Of course, these are not exclusive alternatives: We can surely stress the collective as well as the individual aspect. And one important way to do both, I believe, is to continue pressing the question of our priorities for publicly funded research.
Garrath Williams, PhD
Lancaster University, Lancaster, UK
Footnotes
1 Chadwick R. Nutrigenomics, individualism and public health. Proc Nutr Soc. 2004; 63:161–166.
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.
