Open access has become a buzzword in the age of free information. But bioethicists at the WTCHG have had to develop new ways of regulating access to genome data in order to protect research participants and communities in the developing world whilst promoting the growth of scientific understanding.
Rapid distribution of information via the internet is one of the key features of all science today. In the decade defined by fast-growing knowledge distributors like Google and Wikipedia, internet users have become accustomed to the comfort of being able to access whatever information they need fast and often for free. This has also changed the ways in which scientific results are published - nowadays they are often distributed in an "open access" mode, i.e. free to the reader, and possibly at the expense of the authors. New journals like "Public Library of Science" (PLOS) are thriving on this new publishing model, and even if scientific results are still published behind paywalls, there are rules prescribing that certain kinds of datasets, e.g. protein structures or gene sequences, have to be deposited in public databases, accessible to the wider scientific community, before they can be used in a publication.
In the context of this trend towards more open distribution of data, however, there are ethical concerns attached to the publication of DNA information relating to individual humans, which have required researchers to think carefully about the access philosophy. At the WTCHG, a team of three dedicated ethicists, working "embedded" with the genetics research groups, has come up with new ways of granting restricted access to sensitive data concerning individuals especially from indigenous populations in the developing world.
The issue first came up in 2005, when the Centre secured funding for MalariaGEN (Malaria Genomic Epidemiology Network), a study aimed at trawling human genomes for genetic traits that affect resistance or susceptibility to malaria. "We had a policy discussion at the start of the project," remembers Jantina de Vries, one of the three ethicists associated with MalariaGEN, "and we realised that, contrary to standard policy, uncontrolled data release was probably not going to be appropriate." As malaria mainly affects sub-Saharan Africa, a genomic study of malaria susceptibility would necessarily involve DNA samples from individuals with limited experience or understanding of medical research, who might be unable to comprehend the kinds of use to which their genetic information might be put in the future were it to be made publicly accessible.
Within limits, researchers can work with local communities to engage and educate the study subjects about the potential implications of research they are doing, sometimes drawing upon traditional understanding of genetics and inheritance. But what others might do with the data, and indeed, what uses might become possible in decades to come, is beyond the scope of such efforts, and often beyond what researchers in developed countries could forecast themselves. To take one high profile example: Who knows whether in 50 years from now somebody will go and produce clones of James Watson or Craig Venter, using their published genome sequences? Thus, the notion of "informed consent" from DNA donors becomes highly problematic.
In fact, as we share part of our DNA with our parents, siblings, and children, one might argue that informed consent from close relatives of the DNA donor would also be necessary. In 2007, James Watson triggered a debate when it transpired that he had not consulted his children before allowing his personal genome sequence to be published.
"Because we are aware of the limits of consent in such cases, we have decided not to release data without regulation," explains De Vries. In close consultation between researchers, ethics committees, funding bodies and others, MalariaGEN set up a dedicated review panel to control access to the genomic data generated by the project, the IDAC (Independent Data Access Committee). The six members of IDAC, all experts in relevant disciplines, review all requests for access to MalariaGEN data. An application form for people interested in the data is to be found online, and they will have to justify their request with a description of their research interest, typically about a paragraph or two long.
So far, says De Vries, the committee has received some 15 to 20 applications, all from academic researchers. If commercial companies were to request access to data, this would draw additional complications, as the terms of the original consent would have to be verified, and the ethics committees of the relevant countries may have to be consulted.
A second reason to control data access, says Michael Parker, who chairs the Ethox Centre for bioethics and leads the ethics research attached to the MalariaGEN project, is to protect the emerging capacity of African researchers. If raw data were to be released immediately, researchers in the wealthier nations with their more powerful IT resources might be able to produce publishable results much more rapidly than the scientists who gathered the data in the first place. Therefore, MalariaGEN's data release policy includes the option of delaying access for nine months after the date when the researchers who created the data first had access to it. Parker is convinced that the malaria work will provide a boost for science in Africa, specifically because the future research building on the DNA finding will depend on local knowledge of disease patterns.
Thus, the bioethics work is aimed at supporting the researchers, rather than hindering their progress. "Scientists often see ethics as an obstacle to what they are trying to do, but we try to help them do their work more successfully," Parker says. In teaching medical ethics to Oxford University students from around the world, Parker aims to build capacity for such bioethics support in the developing countries as well. Apart from the close collaboration with the scientists, he describes global outreach as the second defining feature of the work he leads at Ethox.
If and when the IDAC grants access to biomedical datasets, the data are released only to specifically named applicants, who have to sign a legally binding contract banning them from passing the data on to others, and forcing them to delete the datasets when they are no longer needed. While this mechanism isn't 100% secure against misuse of data, it is a strong and legally enforceable deterrent, standing in sharp contrast to the prevailing philosophy of open access to all data.
Some other researchers and institutions appear to think that this much protection is unnecessary. "This area is currently open to debate," admits De Vries. There are many prominent genomics researchers and research funders who argue that unrestricted open-access to genomic data - with consent taken to imply that the DNA donors understand the full implications of their participation - is the best way to promote scientific progress in the pursuit of better understanding of serious diseases. The MalariaGEN and Ethox team however believe that a managed approach to data-release is a more sustainable and appropriate approach to the promotion of science in a global context.
"We see our approach as a contribution to the debate," says De Vries. "It shows that such research can be done in an ethical way." Parker, De Vries, and their colleagues have recently published an account of their data release policy, and are observing an increasing amount of interest from other research groups. "There is a genuine interest in our approach, and we observe an increasing role of ethics in science," Parker concludes. "People are beginning to see it as an important part of the scientific process."
Michael Gross is a science writer based at Oxford.
For more information, see:
M. Parker et al., PLOS Medicine, 2009, 6, e1000143.
www.malariagen.net
www.ethox.org.uk