From San Diego, at a meeting of the American Society of Human Genetics
The gold rush of human genetics is well under way. Now that tools for profiling genome activity are widely available, scientists have found more than 80 disease-related variations in human DNA, many of them in the past year.
As a result, ethical arguments about how to handle people’s genetic information—and related disease-risk information—are no longer academic. Much of the discussion this year among human-genetics researchers has revolved around the sticky issues posed by the windfall of new knowledge. For example, how can scientists keep genetic data from research studies truly anonymous when DNA is, by its nature, the ultimate fingerprint?
Furthermore, studies looking for genetic variations involved in a disease often include thousands of people as research subjects. Scientists use microchiplike wafers to test each person for hundreds of thousands of DNA variants simultaneously, usually under agreements of confidentially and nondisclosure.
What, then, should researchers do if they stumble across genetic variations in a study participant that put the person at increased risk for a life-threatening disease unrelated to the condition being studied? If the disease is treatable, saying nothing could be unethical, but telling the subject about the risk would require breaking the anonymity of the data and violating the study’s nondisclosure agreement.
“If the original agreement was that you wouldn’t tell them, then you shouldn’t tell them,” said Francis S. Collins, director of the National Human Genome Research Institute in Bethesda, Md., in an interview with reporters. “The triumph of discovering these variants should not be taken as evidence that we should necessarily offer this [information] to patients.”
After all, while a gene variant might double a person’s risk for a deadly disease, often the actual risk would remain low-8 percent rather than 4 percent, for example. “Would that change your behavior?” Collins asked. Knowledge of this greater risk could cause the person distress without being usefully “actionable,” he said.
The debate is far from settled, but Collins suggested that disclosure should be reserved for cases that meet certain criteria. The risk of disease would have to be high, and the disease would have to be preventable or curable. Such a case would raise important questions, Collins noted: “Is this the kind of scenario [the subject] imagined when they said they didn’t want to be notified? And if not, is it the benevolent thing to do to notify them anyway?”
Collins gave the example of a mutation in a gene called MutL homolog 1, colon cancer, nonpolyposis type 2 (MLH1) that’s known to increase a person’s lifetime risk of developing colon cancer from an average of about 6 percent to around 60 percent. The disease is often fatal, but frequent screenings starting before age 30 and removal of benign polyps can save a person’s life.
This would be a case in which Collins suggests that informing the person would be the correct ethical decision.
