I recently came across a review article by Henry T. Greely, a Professor of Law, Professor (by courtesy) of Genetics, and Director of the Center for Law and Bioethics at Stanford. The article is entitled “The Uneasy Ethical and Legal Underpinnings of Large-Scale Genomic Biobanks (pdf)” and was recently published in the Annual Review of Genomics and Human Genetics.
According to Mr. Greely, the identity of participants in large-scale genomic biobanks cannot effectively protected. A biobank is defined as a database of genotypic and phenotypic data. Using genetic information, physical information, or a combination of the two, people can identify an individual in such a large database:
“Someone really interested could get a DNA sample from me - from a licked stamp, a drinking glass, or some tissue - and have it genotyped for a few hundred dollars, but few will have to go to the genomic data; the phenotypic and demographic data will often be sufficient.”
“Eliminating name, mailing address, and social security number does not eliminate identifiers; it just eliminates the easiest identifiers, making the search somewhat more difficult and expensive.”
Unfortunately, it is impossible to remove all the data one could use to identify biobank participants. As Mr. Greely opines, “[t]he more the data is removed or obscured, the more scientific value is lost; the more data is kept, the less real the anonymity.”
So what is the answer? First, consent forms must reveal the fact that while biobanks will attempt to provide anonymity, they simply will not be able to guarantee it. They must also reveal that they cannot inform subjects of all the risks and benefits because many future research topics haven’t even been suggested as of yet. Second, biobanks must prevent participants from being upset by unexpected uses of their materials, either through a thorough consent form, or through general communication with research subjects (such as a mailing list or online community). Third, researchers have a moral (and perhaps legal) duty to inform participants of potentially harmful information uncovered by research. This raises a whole host of questions, including how significant the correlation between a gene and a disease must be to require a participant’s knowledge, how long the biobank should monitor the participant’s genetic information, and whether the biobank should be responsible for genetic counseling.
Mr. Greely raises a number of interesting questions that will have to be answered by governments and companies around the world as the need for biobanks increases and the relative ease of biobank creation decreases.