If you allow researchers to do brain imaging on you for some research purpose, and they incidentally discover a tumor, or a blood vessel with thin walls, you probably want them to inform you about this finding. There are no doubts about the finding; the risks are well-known; it is actionable.

Suppose instead that you donate a blood sample to a biobank. Suppose that researchers studying the sample discover a genetic variant that, depending on a number of interacting factors, might result in disease in three years’ time, or in thirty years, or not at all. It is difficult to predict! Do you still want to know?

How should these incidental findings be handled that increasingly often will be made in genetic biobank research? We are all different, so finding variants with some statistical relation to disease is more or less expected.

A common approach to this question within attempts to develop a policy for incidental biobank findings is to formulate general conditions for when researchers should inform participants. Like: if the finding is analytically valid; if it has clinical significance; if it is actionable – then participants should be informed.

The problem is: we already knew that. We know what these conditions mean in imaging studies when a tumor or a damaged blood vessel is discovered. In these cases, the conditions can be assessed and they make it reasonable to inform. But what about genetic risk information, which often is more multidimensional and has unclear predictive value?

This question is discussed in a recent article in the European Journal of Human Genetics, written by Jennifer Viberg together with Mats G. Hansson, Sophie Langenskiöld, and me:

Viberg argues when we enter this new and more complex domain, we cannot rely on analogies to what is already known in a simpler domain. Nor can we rely on surveys of participants’ preferences, if these surveys employ the same analogies and describe the findings in terms of the same general conditions.

Time is not yet ripe for a policy for incidental genetic findings, Viberg and colleagues conclude. Formulating a policy through analogies to what is already known is to cover up what we do not know. The issue requires a different form of elucidation.

That form of elucidation remains to be developed.

Pär Segerdahl

We participate in debates - the Ethics Blog