30 March 2015
One of the legally and ethically problematic issues regularly debated in the context of biobanks and tissue repositories is that of its potential for forensic use. When Anna Lindh (the Swedish foreign minister) was murdered in 2003, her killer was subsequently identified by way of matching DNA traces found at the crime scene with data contained on the killer's Guthrie card (an archived heel blood test done on every child born in Sweden). This was an elegant and inspired forensic move by the prosecuting authorities in Stockholm, but it led to frantic debate in the relevant scientific communities about whether mechanisms ought to be developed that restricted such use in the future.
The rationale for this discussion was not what one might first suspect it to be: it was not driven by a desire to strengthen individuals' informational self-determination, or a sign that genetic information was in some way instantly recognised as particularly volatile and needing additional protection (though the jury is still out on that particular question). The driver behind this discussion is essentially the same as that in the context of medical confidentiality taken by the Court in X v Y , succinctly summarised in that judgment by Rose J:
[i]n the long run, preservation of confidentiality is the only way of securing public health; otherwise doctors will be discredited as a source of education, for future individual patients 'will not come forward if doctors are going to squeal on them'. (my emphasis). (1)
This is, in essence, a consequentialist public health argument. It is not about protecting the privacy or augmenting informational self-control of individuals, but about providing stability and coherence in the system. If the information is not safe in the system, I will not give my information to the system. This would have disastrous consequences for the provision of clinical care to the benefit of everyone.
The same can be said about the post-Lindh discussion about forensic access to research infrastructures storing genetic information. If it becomes commonly known that biological material and genetic information entrusted to researchers might be subsequently used to match me, or a close relative, to the scene of a crime, it is clear that less material would be provided to research. This would have disastrous consequences for the generation of research knowledge to the benefit of everyone.
Since then, this conundrum has featured heavily in the governance design of large-scale infrastructures dealing with identifiable genetic information. A small number of approaches have emerged: some biobanks design their infrastructure in a way which prevents the efficient linking of identifiable information (killing the problem). In some cases, biobanks endeavour to inform participants appropriately about this risk during the course of the consent procedures (seeking exculpation). Some other banks have developed policies to resist, through the courts, every attempt to access the resource for forensic purposes (using process as resistance). It seems clear that such a challenge is ultimately futile, but the idea is to make the exercise so expensive in terms of time and resources as to make it virtually meaningless.
In some jurisdictions (Germany is one), there has been extensive debate about creating an absolute statutory duty to keep genetic information provided for research purposes confidential. Such a statutory duty would mean that even a court could not compel a researcher or an institution to hand over such information. In the meantime, the scholarly debate in relation to this issue has very much focused on the extent of the consent provided when the genetic material was first obtained. In close analogy to, and ultimately in overlap with, general principles of data protection law, DNA information about an individual should only be used for the narrow purposes for which consent was originally given - the focus therefore lies strongly on the collection point.
We can see a similar set of issues developing in relation to genetic information from other contexts. In the recent decision in X v Z  (2), the Court of Appeal dealt with the question whether DNA profiles derived during the course of a criminal investigation should be made available for other purposes. In short, X murdered his wife in 2013. DNA evidence was taken at the scene of the crime. Further DNA evidence was taken, post mortem, from the victim and the residual DNA evidence, which could not be explained using existing samples, was subsequently matched to X by way of a sample given whilst he was in custody. X was convicted and is serving his sentence.
In the meantime, care proceedings in relation to his children (collectively Z) had begun. X expressed the wish to take a role in their lives and asserted biological fatherhood. At the same time he refused to give a DNA sample to underpin this assertion. The President of the Family Division, Sir James Munby, had, in the Court below, given permission for the crime scene profiles to be released to Z for the purposes of a paternity test. This decision was appealed by X and the Metropolitan Police, and the Court of Appeal allowed the appeal.
Without going into the substance of the decision, the Court placed significant weight on the context and purpose of the original collection of material. Having decided that it would be arbitrary and irrational to make a significant normative difference between genetic information derived from a place (the scene of the crime) and that derived from an individual (the suspect at the police station), the Court underscored the particular protection that ought to be given to genetic information. The Court quoted directly from the decision of the ECtHR in Marper v UK (2009) (3):
While the information contained in the profiles may be considered objective and irrefutable in the sense submitted by the Government, their processing through automated means allows the authorities to go well beyond neutral identification. The Court notes in this regard that the Government accepted that DNA profiles could be, and indeed had in some cases been, used for familial searching with a view to identifying a possible genetic relationship between individuals. They also accepted the highly sensitive nature of such searching and the need for very strict controls in this respect. In the Court's view, the DNA profiles' capacity to provide a means of identifying genetic relationships between individuals [...] is in itself sufficient to conclude that their retention interferes with the right to the private life of the individuals concerned. (my emphasis). (4)
The issue of secondary uses of genetic information (including samples but also, significantly, identifying profiles derived from the samples) will play a significant role in the governance of genetic banking in the short term. Whether infrastructures are intended for treatment purposes (such as fertility clinics), lifestyle purposes (such as 23andMe and other direct-to-consumer products), research (UK Biobank and others) or forensic purposes (such as the national database criticised in Marper), the linking and secondary uses of this genetic information will give rise to significant ethical and legal issues which should be the subject of intense debate not just in the academic setting.