Tip:
Highlight text to annotate it
X
We're now going to hear from Amy McGuire at the Baylor from the Baylor College of Medicine.
Amy is going to be talking about ethical, legal and social issues in genomics reflecting
back and planning ahead.
Amy McGuire: OK, good afternoon. And thank you very much
for inviting me to be here today. It's really exciting to be part of this celebratory event
and I'm very happy to be here.
So, I was asked today to talk about the ethical, legal and social issues in genomics and study
of the ethical, legal and social implications of genomics and the human genome project has
really been tightly integrated with the entire and NHGRI and the human genome project since
its inception. And so the LC program and the Division of Extramural Research at NHGRI was
established in 1990 and it has always been structured as with a multiple disciplinary
focus. Its original function and purpose was to identify and examine key ethical, legal
and social issues associated with mapping the human genome, to stimulate public discussion
about the human genome project, to develop policy options and to expand public education.
And as NHGRI has moved beyond the human genome project, so has the LC program and so it is
still dedicated to these four kinds of missions but now related to the broader issues in genomics.
The LC program at NHGRI has organized their funding of extramural research around four
general topical areas. The first is privacy and confidentiality and the fair use of genetic
information including genetic discrimination; the clinical integration and impact of genetic
testing; issues related to genetic research including things like research design and
issues around informed consent and human subjects research protection; and education in the
development of resources to stimulate public education around genetics and the ethical,
legal and social issues associated with genomics.
And so, there has been a tremendous amount of success within the LC program and the research
that's been funded by the LC program and all four of these areas of research over the years.
And many of my colleagues who have set the groundwork for the research that we're doing
today have been successful at engaging with the scientific community and conducting research
that's been useful and impactful on science practice and policy.
And I'm not going to really go into any great detail about sort of the projects that have
come before us, but I want to point out that one of the most, I think, impactful things
that the LC program has done is really to sort of lay out a new area of research. And
LC has really come to represent and it's kind of a representation of a broader area of research
that deals with ethical, legal and social issues in science more generally, that takes
a multi disciplinary approach. And so this has been accepted at an international level
and I think the importance of studying these issues in all its areas of science has really
gained a lot of traction. And I think that can be really traced back to the beginning
of the LC program at NHGRI and NHGRI's dedication to looking at those issues in the context
of the human genome project.
So I'm going to talk about this broader LC research area, when I talk about how we can
plan for the future and some of the things that we might be thinking about in conducting
research as we move forward in plotting our course over the next 10 years. So if we look
at the strategic plans that we have been looking at all day, there's a text box that's dedicated
to identifying some the salient and critical ethical, legal and social issues that are
going to be implicated over the next decade and these are organized into four major categories:
Ethical, legal and social issues related to economic research; things like human subjects
research protection; insuring that we have diversity in our cohorts and our research
cohorts; making sure that we have effective community engagement in getting the community
involved in our research projects.
There are a lot of issues in the area of genomic medicine. As we see this translation into
generic medicine, I think some of the most important ones are making sure that we are
effectively communicating with patients about the uncertainty and evolving nature of predictions
based on genetic information and also effectively evaluating the effectiveness of genomic services
as they get integrated into clinical care.
There are legal and policy issues that I am particularly interested in and I think the
strategic plan really highlights three of the most import ones which are, issues around
intellectual property, issues around insurance reimbursement for genomic services, and issues
around the regulation of genetic test.
And finally, there are broader societal issues that are largely conceptual nature that have
to do with implications of how to genomic knowledge changes that we think about human
health and disease, how we think about self identity, how we think about things like free
will and individual responsibility.
So all four of these areas are extremely important and there's very good research going on in
each of these areas. And what I'm going to do today is not talk specifically about the
research topics themselves, but rather to try to identify a couple of major themes that
can help affect the ways in which we do the research that can make the most impactful
as we move forward in the area of genomics.
And so in 2008 there was an assessment panel that looked at the LC program at NHGRI and
it published a report and kind of strewn throughout the report are three major themes that I think
come up charting the course for the future of LC research more broadly. I think these
three themes are integrated in some ways into the strategic plan was published yesterday
as we move forward to how this area of research will move will progress.
So, I'm going to talk about each of these three: Integration, collaboration and expansion.
And I'm going to use my own experience and my own research to try to illuminate each
of these domains or each of these themes. So it's a rather narrow lens through which
we'll be looking at these themes. I recognize that there's a lot of research going on that
can help inform how we think about these three domains, but I want to use my experience to
try to show you my perspective on how these things have come into play in my research.
So, with integration, here I think that the main idea is that there has been a tension
in sort of LC research since its beginning between LC researchers being responsive to
the issues that are coming up for scientists as they're coming up and helping them to manage
those issues in real time, versus LC researchers trying to identify the larger, broader issues
that might be implicated in different areas of genomics and to study those broader issues
slightly detached from the research that's going on at the time. And I think that this
tension is very healthy and both kinds of researcher are very important to the LC field
to advance affectively. But my personal research has been more focused on the first model.
I've been very informed in the kind of research that I do by the scientists that are doing
the research and the issues that are coming up for them in real time. And this is just
one example of several that I could tell you about.
So in 2007 Jim Watson was up here earlier and was telling you about getting his personal
genome sequence. He mentioned that the sequence was done by 454 Life Sciences and it was analyzed
at Baylor College of Medicine. And I can say that in that project the scientists both at
454 Life Sciences and Baylor College of Medicine were very conscious and conscientious about
the ethical, legal and social issues that were going to come up in the context of that
project and the broader implications that that project would have for sort of how we
integrate whole genome sequencing into our research practice and clinical care.
And so as soon as the Baylor College of Medicine got involved in this project, I was brought
on board by our genome sequencing center and Richard Gibbs and was intimately involved
in the entire project. I went all the scientific meetings and participated in the planning
of the project, helped them identify the ethical, legal and social issues that were coming up
and to manage those the best that we could. And even participated as we started to disseminate
the results of the project through different press releases and publications and you can
see here we had a paper in "Nature" that was a letter that was a four page letter describing
our results. An entire page of that was dedicated to identifying what the ethical, legal and
social issues were in the project and providing a justification for our management plan in
the context of this particular project.
So, I think that's a good illustration of how we had a very close integration between
the science and the LC research and how they both sort of served to foster the other and
it really resulted in, I would say, better research on both sides. This project got me
really interested in some of the issues and so I worked with some of my colleagues in
the LC community, Tim Caulfield and Mildred Cho, and we thought about the issues that
had come up in the context of this research and thought more broadly about the policy
implications of that, when you have research participants who aren't Jim Watson. Frankly,
some of the ways in which we managed the project were specific to the fact that we had a very
unique individual before us who was the research participant.
So we thought about these issues a bit more broadly and wrote about them and put out some
preliminary recommendations for managing them in research with other research participants'
involvement. And this also led to an international consensus working group where we got a bunch
of people together and we talked about these issues and we provided we went back and forth
and we looked at some analysis and looked at data and we came up with a more solidified
set of recommendations that was published in “PLoS Biology."
This project also got me and a couple of my colleagues really thinking for the first time
about how were we going to manage what are going to be some of the ethical, legal and
social implications as we start to integrate this technology, this whole genome sequencing,
into clinical care? And so we thought about those issues and wrote about them, kind of
trying to tease out what would be the health system challenges and implications of routine
integration of full genome sequencing. And that's work that I continue to think about
and continue to work on. So that's an example of how we one example
of how LC research and the genome science has kind of integrated itself in a way that
I think was fairly successful. And again, there are other models that are out there
as well.
The second theme that I want talk about is collaboration. This, from the LC research
perspective I think we are and should be interested in ethics and legal and social issues associated
with this research. And to me, I really feel strongly that if you don't combine sort of
a multi disciplinary approach and consider all of these various facets of a given topic
then you won't have a clear picture of what you're thinking about. So let me give you
an example of this. I've been in some research around direct to consumer genetic testing
so when some of these companies like 23 and Me, which you can see the 23 and Me spit kit
here, which was named Invention of the Year a couple years ago. When these technologies
first came out and these companies first started advertising their services I think it was
in 2007 I became very interested from a variety of different perspectives. And early
on in sort of the life of these companies, there was some push back by a couple states,
New York and California. They sent letters to a couple of the companies. They raised
questions about whether providing these services was somehow in violation of their state laws.
And one of the laws that they were particularly interested in was the law that most states
have that prohibits the unlicensed practice of medicine. And so from a legal perspective
I was very interested in whether this type of service constitutes the unlicensed practice
of medicine according to state laws. And so I partnered with a legal colleague of mine,
Cindy Marietta, and we did an analysis of state laws on the unlicensed practice of medicine.
And what we found was that there is really not very good clear guidance, which sometimes
you do find, and that most states do have laws that prohibit the unlicensed practice
of medicine but whether this is the practice of medicine really turns on whether you consider
giving somebody predictive genetic information to be a diagnosis. And there's not very clear
guidance. There's a lot of opinions but not very clear guidance on whether that is diagnostic.
So it occurred to me that how consumers perceive this information may become very relevant
from a legal perspective of the whether it's diagnostic information. So if the consumers
are perceiving it to be diagnostic information that could become relevant in a court of law
if this ever gets challenged. And so we decided to look at this from an empirical perspective,
and we did a very, very pilot study. It was a cheap, unfunded quick study to kind of get
our finger on the pulse of what people thought about this and a couple of the other issues
associated with this emerging industry. We posted in the "American Journal of Bioethics"
and what we did was an online survey and we used facebook.com users and this was done
in 2008. And the reason we chose facebook.com users is because some of these companies engage
in social networking and we thought that perhaps users of facebook.com may be some of the early
adapters of these companies.
So we surveyed just over about a thousand individuals and you can see their demographics
up here, and we first asked them, have you ever use these services and if not, would
you consider using the services? And about six percent of our respondents reported that
they had used the services and about 64 percent said they would use the services even though
they hadn't done so. Then we asked them, why would you use the services? And over 80 percent
they could choose more that one answer here over 80 percent said they just had general
curiosity about these genetic testing services. More interesting to me was that over 70 percent,
about 73 74 percent said that they would use the services specifically to see if a
disease ran in their family or was genetic, so they were using it, were interested in
it for health care purposes. And also, very interestingly, about 40 percent said they
would use the services to learn about if their genetic makeup specifically without having
to go through a doctor. We didn't get to kind of follow up on that and probe into why that
was so important to such a large minority of our respondents but I think that's an important
issue that that deserves further exploration.
So we asked them a bunch of questions and relevant to what we're talking about right
now we asked them whether they considered the information you would get back from these
companies to be a medical diagnosis. And of that small six percent of our respondents
who said that they had used these services, 60 percent considered the information they
got back to be a medical diagnosis. Of the 64 percent of individuals who were interested
in using these services, just under 40 percent said they would consider the information that
they got back to be a medical diagnosis.
Finally, we were interested in sort of how is this information from a policy perspective
going to interact with our health care system and the practice of medicine? And so we asked
individuals that had used the services, did you talk to your physician about your results?
And about 55 percent said yes, they had spoken to their physicians about their results and
another 10 percent said they hadn't but they planned to. Remember, this is only six percent
of our respondents so it's a very small number people.
Those who said they were interested in the services, just under 80 percent said they
would talk to their physician about their results. And then we asked everybody do you
think that physicians have a professional obligation to help individuals understand
results from these direct to consumer companies? And about 60 percent agreed or strongly agreed
with that statement that yes, physicians have a professional obligation to help patients
understand this information.
So just as a point here, there was a follow up, much more rigorous study that was recently
conducted out of Eric Topol's lab, [unitelligible] published in the "New England Journal of Medicine"
I think just this month. And they looked they did a much more intensive survey and study
with over 2,000 navigenics consumers. And what they report in their findings was that
26.5 percent of their respondents reported that they had shared the results with physicians.
So it may be that people think they're going to share the results with their physicians
and they're not actually doing it. I think we need a little bit more data to understand
that and tease that out.
Regardless, however, this research and the results of our survey raise for me a lot of
questions about sort of the health care ethics and implications of this. So I was really
interested in, well what are physicians supposed to do when patients come to them with this
information? And how is it going to interfere with or enhance the physician patient relationship
and what are the obligations of health care providers? And so I partnered with a colleague
of mine, Wylie Burke, and we started thinking through these issues from an ethical perspective
and wrote an article where we were kind of teasing out the appropriateness of acting
on some of this information. And I think one of our main points there was, when you have
genetic information that you get back that has uncertain or unproven clinical utility,
there may be a tendency to want to try to follow up on it, because that's what, as health
care providers, we typically do is try to follow up on stuff and find out the answer.
And if it's truly the science is not there, then we make the argument that to follow up
on it might not be a justified use of health care resources and it could also pose additional
risks and create iatrogenic [spelled phonetically] harm for patients.
So coming off of sort of that ethical analysis and the discussion that we had had, it brought
us back to sort of the legal issues and we became interested in the regulatory issues
of, should these companies be regulated more than they're currently being regulated? And
this issue really kind of came to a head last summer when FDA announced its intention to
start regulating direct to consumer drug testing companies and genetic testing labs more generally.
And so in the last couple of months we partnered, Wiley and I partnered with some lawyer colleagues
of ours, Barbara Evans and Tim Caulfield, again, and really did sort of a more in depth
legal analysis of how the FDA possibly could regulate this burgeoning industry with all
the complexities of the genome Y type of testing that they're doing.
So hopefully that gives you a sense of sort of how the legal analysis that we initially
did informed kind of the empirical questions that were raised and then our empirical study
informed our ethical analysis which raised additional legal questions. I think now we're
at the point where there are additional empirical questions that we'd like to study and I've
started collaborating with Robert Green and Scott Roberts who have done some really, really
fascinating work outcomes measures and how patients react to certain genetic risk information
to look at it in the context of direct to consumer testing. They have a big grant and
I'm consulting on that grant and they're doing some really neat work in that area.
OK, the last theme that I kind of wanted to talk about is the expansion theme and here
I think I've already sort of brought this up in some ways. The fact that LC research
in general has really outgrown its place solely within the context of NHGRI; it's a much broader
area of research now. It's also in some ways outgrown its exclusive emphasis on genomics.
I think there's a lot of other diseases and conditions that are touched by this. I would
argue that all areas of medicine and research have ethical, legal and social implications
that deserve serious consideration and study. And so I think that in some areas of research
and medicine, there are unique ethical, legal and social implications and it's also the
case that some of the issues that we study in the context of genomics have direct impact
on researchers, investigators in other areas of health and disease and that are funded
by other institutes at the NIH.
So one example of this has to do with data sharing and I've spent quite a bit of my time
working on the issue of data sharing. And as you all probably know, NHGRI has had a
very liberal culture and policy around data sharing since the beginning of the human genome
project. They have called for the rapid public release of all generated sequence data since
the beginning of the genome project and there have been several data sharing policies that
have been published in that spirit. In the last couple of years at think there's been
increasing concern about the ethical implications of public data release and concerns coming
out of some of the informatics research that's been done showing just how easy it is to identify
an individual on the basis of their genomic information. There have been concerns about
whether these types of policies appropriately strike the right balance between advancing
research and protecting individual research participants and their privacy. And so the
latest NIH policy on data sharing which was for genome Y association studies calls for
the release of all GWAS data to dbGaP which is a controlled access database and you have
to apply for access to dbGaP through a data access committee. So this creates an extra
layer of protection.
So you'll notice that this is an NIH wide policy; it's not restricted to NHGRI. There
are a lot of investigators NIH wide who are conducting research that would make them fall
under this data sharing policy. And at some institutes I think there are special issues
that come up that complicate the issue of data sharing and just one example of probably
many is the National Institutes on Drug Abuse. And they held a workshop a couple of years
ago that I attended and the sole purpose of this workshop was to think through some of
the ethical challenges associated with data sharing when you're dealing with sensitive
information like social environment and psycho social information and you're merging that
with genetic data.
And so this gives you a sense of sort how the issue of data sharing really touches a
lot of different institutes and a lot of different researchers. I got engaged in the issues around
data sharing again through collaboration with our Genome Center. This is actually how I
got interested in the whole field of genomics and it was through a consultation for the
Genome Center where they were really concerned about releasing their individual level participant
data into publicly accessible data bases. And this was about in 2005 so two years, about,
after the genome project finished. And so Richard Gibbs and I worked on this quite a
bit and I went to all of their lab meetings and spent a lot of time over at the Genome
Center trying to understand the issues from their perspective. And then went back to my
office in the Center for Mental Medical Ethics and did a lot of research and did a lot of
ethics research and spoke to a lot of colleagues and tried to get a handle on what some of
the issues were. And we published this paper back in 2006 and made several recommendations
for managing the issue of data sharing. One of our recommendations was that since the
decision about data sharing necessarily involved a tradeoff between individual privacy and
advancing research, that people are going to differ in terms of how much they care about
their genetic privacy and so we should leave that decision up to individuals to make in
the informed consent process.
One of the biggest policy push-backs we got on this was well, if you tell people about
data sharing and you ask their permission and you give them a control over it, nobody's
going to participate in research and nobody's going to share their data publicly. Only what
Zack Cohony and Russ Altman [spelled phonetically] call information altruists, which are those
people who have very little concern about their genetic privacy are going to be participating
in public data sharing.
So we decided to test this more empirically since we thought it would seem like a legitimate
concern but wanted to see what participants felt about this. We started by doing some
focus groups with individuals who were participating in a genetic epilepsy study at Baylor. And
from our focus groups it seemed that they were very interested in the issue about data
sharing. They seemed to want to get engaged in the decisions about data sharing and we
developed a couple of consent models that we ran by them and got their feedback on and
modified in response to their concerns. And then we conducted a larger randomized trial
of three different types of consent that dealt with data sharing. And really the only thing
you need to know about this for these purposes is that each of the consents only varied in
terms of how much control they gave participants about data sharing. So the traditional consent
here just simply said, we will share your data both in publicly accessible databases
and controlled access databases. Binary consenting gave them an opt out, so you could participate
in a study and you could choose not to share your data if you liked. And tiered consent
gave them three choices: You could participate, you could share your data publicly, you could
only share your data in controlled access databases, or you could not share your data
at all. We then debriefed participants. We showed them all the consent options, allowed
them to change their minds and invited them to participate in a follow up interview.
These are the w partnered with these six studies at Baylor that were ongoing and were recruiting
participants. And you can see, about half of them are cancer studies and then there
was an epilepsy study and an autism study. And we worked very close with these investigators.
All of these studies were collecting samples and they were potentially using them for genetic
analysis and they would involve broad data sharing.
We randomized 323 individuals and surveyed about 230 and just very briefly, our main
findings were that initially this is pre published data, but initially about 84 percent
of people agreed to public data release, regardless of what consent form they had been given.
And everybody who had been randomized through the traditional consent where they didn't
have an option agreed to participate in the study. So nobody said, I'm not participating
in this because you're telling me you're going to release my data. After we showed them all
their options about 53 percent stayed with their decision to release their data into
publicly accessible databases. About a third chose only to release them into restricted
databases and about 14 percent chose not to release them at all.
Other key findings that we found through our interview process, and this is sort of just
a summary of some of our some of the findings that we found more qualitatively, people seemed
very willing to share not only their genetic information but also their clinical data but
they seemed to draw a line at personal information and this is defined as name, Social Security
number, address, things like that. There was also a tremendous amount of trust in researchers
and in the research institution and I think this contributed to people's willingness to
participate in the research and their willingness to share their data more publicly. And finally,
we asked participants, do you want to be involved in the decision about data sharing and the
vast majority said yes. When we asked them, how do you want to be involved, overwhelmingly
we heard, just ask me about it. I just want to know about it and I want to be asked about
it. And when we kind of probed that more qualitatively we found that this really wasn't about, primarily
about concerns about their privacy; it was about respect. It was about being respected
as a research participant and feeling like they were being treated respectfully.
So I am presenting all of this to you, really to make the point that this type of research
has implications, I think, for all areas of research funded by all NIH institutes not
just genome research. There's a lot we had cancer researchers funded by NCI, we had neurologic
researchers funded by other institutions and so this has implications for all of them as
they move forward in their research.
So I'll just bring us back to our three themes and just to highlight these again and I just
want to reiterate one more time that this I'm presenting you with my experience and
my sort of narrow lens at which I'm looking at these three themes. I do think they're
important themes and I hope that this can contribute to a discussion about the best
ways that we can integrate, collaborate and expand so that LC research has the greatest
impact on science practice and policy as we move forward in the next decade. And I'm a
multi disciplinary researcher so this is a multi disciplinary team of investigators I'd
like to acknowledge. Thank you.
[applause]