Monday, August 27, 2012

The Ethics of Genome Analysis

Lots of people are having their genomes sequenced or otherwise analyzed for specific alleles. Those people should get all the information that comes out of the analyses although, hopefully, it will be scientifically correct information and any medical relevance will be explained by experts.1

There's another group of people who submit their genomes for research purposes only and they usually sign consent forms indicating that their name will not be associated with the results. Under those circumstances, the researchers should never have access to the individual's name or any circumstances that are not relevant to the study.

Apparently that simple ethical rule is not always standard practice. Gina Kolatea writes about some ethical issues in the New York Times: Genes Now Tell Doctors Secrets They Can’t Utter.

Here's an example from her article ...
One of the first cases came a decade ago, just as the new age of genetics was beginning. A young woman with a strong family history of breast and ovarian cancer enrolled in a study trying to find cancer genes that, when mutated, greatly increase the risk of breast cancer. But the woman, terrified by her family history, also intended to have her breasts removed prophylactically.

Her consent form said she would not be contacted by the researchers. Consent forms are typically written this way because the purpose of such studies is not to provide medical care but to gain new insights. The researchers are not the patients’ doctors.

But in this case, the researchers happened to know about the woman’s plan, and they also knew that their study indicated that she did not have her family’s breast cancer gene. They were horrified.
This is a rather simple case of the researchers violating a standard protocol. They should not have known the identity of the patient and they should not have known what she intended to do.

Most of the "ethical problems" in the article are of this type. They involve researchers who are supposed to be concentrating on research and not on the treatment of individual patients. Those researchers have no idea whether the patients already know which alleles they carry or whether they are already undergoing medical treatment. That's just as it should be. If a DNA donor doesn't want to be contacted then it's ethically wrong for the researchers to violate that contract no matter how justified they think they are being. Furthermore, it should be impossible for them to find out the name and address of the donor so the issue should never come up.

John Hawks thinks this is an interesting ethical problem and he wants his students to discuss it in his classes [Grasping the genomic palantir].
That case is ethically straightforward compared to others, because the researchers could make a difference to an immediate medical decision. On the other hand, how many risk-free research participants went ahead with prophylactic mastectomies because researchers didn't know about their plans?

I think the article will be a good one for prompting student discussions in my courses, and I'll likely assign it widely. But I think the central ethical problem discussed in the article is temporary.
What will students learn from discussing issues like these? What controls are in place to make sure that students are informed about all the ethical issues? Will they be told that standard scientific protocols were violated once the researchers knew what the patient intended to do?

1. "Experts" do NOT include employees of any for-profit company that took money for sequencing the genome.


  1. Teaching students about human biology means teaching them about the limits of research, many of which are ethical limits.

    For my undergraduate students, they are far more likely someday to consider participating in research than to become researchers themselves. They should understand that participating in most research studies will not provide them any direct benefit, even if researchers may find something that would be directly relevant to their health decisions. As citizens, they should consider whether why researchers are regulated in this way and whether they support greater or less regulation of researchers.

    1. If the subject of the course is really human biology then I can think of dozens of important concepts and principles that have to be covered. None of them require knowledge about the "limits of research."

      We spend a good deal of time in one of my courses talking about ethics—what it is and why it is sometimes relevant and sometimes not. I've always had the advantage of teaching with someone who has substantial training in philosophy. This comes on very handy during class discussions when we inevitably end up talking about different opinions on ethical questions.

      It's hard to get students to think critically about these issues because they come into class loaded with all kinds of misconceptions about ethics. They often can't tell the difference between a scientific controversy and an ethical problem.

  2. Furthermore, it should be impossible for them to find out the name and address of the donor so the issue should never come up.

    Hang on, there's no evidence that the researchers knew this. The only information they had at the time was:

    1) The patient comes from a family with a segregating high-risk allele

    2) The patient does not bear the high-risk allele

    3) The patient intends to have a prophylactic double mastectomy.

    None of those factors violate ethical gudelines, in fact (1) and (2) are inherent in the fact that it was a mapping study. One can argue that the researchers shouldn't have known point (3) - it really depends on whether it was necessary for the study. Or maybe their entire study sample was people planning to have a mastectomy, we just don't know.

    On the basis of (1) to (3), they approached their ethics board for guidance. Clearly you feel that the ethics board should have made a different decision. I don't however think the researchers were wrong to check.

    1. The researchers should not know what the patient plans to do.

      They should also know that they have no right to contact the patient. There should be a strong barrier of protection around the identity of the subjects. We have plenty of experience with double blind experiments. We know how to do this.

    2. This is not as simple as you make it out to be, since there are potential people involved other than the research subjects and the researcher. For example, suppose the researchers discover the patient is HIV+? Better consent forms will help. In studies I coordinated, the consent form explicitly states something along the line of, "although we are not directly looking for clinically relevant information in this study, if a result of potential clinical relevance is found, you will be informed".

      In many cases (such as above) the study is carried out with fully-identified individuals, for example, in order to correlate molecular analysis with future clinical outcomes. The people on the bench doing the technical work only know sample codes, but there is an individual that can match those codes back to patient identity. Almost all double blind experiments are eventually unblinded so the results can be interpreted.

    3. You are free to conduct your research in this manner if you choose.

      As long as the patients know that the sequence of their genome can easily be traced back to them by anyone on the research team, then there's no ethical problem.

      How do you decide which information is "clinically relevant"? Is it clinically relevant to let a patient know that she should have no fear of inheriting something from her father because that's not her father? Is it clinically relevant to inform a patient that they have a 10% higher risk of getting Alzheimer's when they are 80 years old?

      Do you have experts on your research team (physicians , philosophers, scientists) who have drawn up a list of alleles that pass the "clinically relevant" threshold? Can you share that list with us?

    4. How do you decide which information is "clinically relevant"?
      We have a clinical bioethics review board and they get to make all of these tricky judgement calls. The board is composed of, exactly as you suggest, physicians, philosophers (specializing in bioethics) and scientists. The possibility of detecting non-paternity is also explicitly written into the consent form.

      As long as the patients know that the sequence of their genome can easily be traced back to them by anyone on the research team, then there's no ethical problem.
      This is not the case. In fully-identified studies there is a database that links coded study numbers to patient identifiers. The database is encrypted with generally only one or two specific individuals with the access key. In non-identified studies, there is no such database and no one can trace patient samples back to individuals.

      Note that in the United States (where this research is conducted) HIPAA regulations protecting patient privacy are very strict, with significant penalties associated with unauthorized disclosure. The list of HIPAA protected health identifiers (PHI) is:
      1) Names
      2) Geographic subdivisions smaller than a State
      3) Dates (except year) directly related to patient
      4) Telephone numbers
      5) Fax numbers
      6) E-mail addresses
      7) Social security numbers
      8) Medical record numbers
      9) Health plan beneficiary numbers
      10) Account numbers
      11) Certificate/license numbers
      12) Vehicle identifiers and serial numbers
      13) Device identifiers and serial numbers
      14) Web URLs
      15) Internet Protocol (IP) address numbers
      16) Biometric identifiers, including finger and voice prints
      17) Full face photographic images and any comparable images
      18) Any other unique identifying number, characteristic, or code, except as permitted under HIPAA to re-identify data

      Clinical/Translational research is not trivial to do correctly. There should be an office at your institution to help researchers navigate the requirements and regulations successfully.

  3. The ethical thing may or may not be the correct moral thing. All ethical rules are heuristics, subject to stronger moral demands from other obligations. This case is interesting because the competition is between 2 high priority morality values.

    1. Like you, I tend to make a distinction between "ethics" and "morals." In my case, "morals" is a null set—there's no such thing in my worldview.