A few days ago, NIH announced an agreement with the family of Henrietta Lacks. When she was being treated for an aggressive tumor, cells were taken without her consent or knowledge and used to create the HeLa cell line, which is tremendously widespread in biomedical research today – the story is well-chronicled in Rebecca Skloot’s The Immortal Life of Henrietta Lacks. The announcement raises some interesting questions about precedent and keeping information partitioned once it’s been released.After a German research group published the first full HeLa genome in March, there was an outcry among ethicists and commentators, well-reported here. The full genome was pulled from public databases shortly thereafter. The recent NIH announcement accompanied another publication of a specific HeLa line genome.
Under the new NIH policy, negotiated with the Lacks family, access to genomic data from HeLa lines will be controlled by the HeLa Genome Data Access Group, which will include two members of the Lacks family. Any research using NIH funding will be required to deposit any HeLa genomic information into that data archive, and other researchers are encouraged to do so as well. This option was preferred unanimously favored by the Lacks family over the possibilities of free access or no access at all.
The authors of the NIH announcement write:
It is important to note, however, that we are responding to an extraordinary situation here, not setting a precedent for research with previously stored, de-identified specimens. The approach we have developed through working with the Lacks family is unique because HeLa cells were taken and used without consent, and gave rise to the most widely used human cell line in the world, and because the family members are known by name to millions of people.
I’m not so sure this won’t be a precedent for other situations, especially given its high profile. The first unique characteristic – taken and used without consent – is not really unique, since it applies to lots of data from databases, and things like deanonymization/reidentification can create end-runs around limited consent (though, of course, that raises many separate issues). The second and third – wisely used and widely known – seem much more like issues of degree rather than of kind. This is not to argue that the NIH policy is a bad precedent—a question which needs quite a bit more analysis—just that I think it’s much more likely to serve as precedent than the authors suggest.
Practically speaking, it’s hard to argue these days that the information cat can be really put back in the bag. As Joe Pickrell at Harvard Medical school writes on Genome Unzipped, genomic data about HeLa cells have been readily available for days, and still are. Given that reality, do the restrictions on highest-quality genomic sequences archived by NIH, leaving just lower-quality and harder-to-access sequences available. The research community may follow the wishes of the Lacks family, which may help with attribution concerns, but with respect to privacy concerns, the information is certainly out there.