By CRG staff - interview with Laura Rodriguez

Laura Lyman Rodriguez, PhD, is the Director of the Office of Policy, Communications, and Education at the National Human Genome Research Institute (NHGRI).


GeneWatch: NHGRI has a new strategic plan this year which emphasizes translating science into clinical practice. Does the plan include more focus on the return of research results and incidental findings?

Laura Rodriguez: It's not part of the strategic plan explicitly, but it is clearly an integral part of providing clinical care informed by genomics. This is an emerging area with a lot of complexities. The science is moving very quickly, and law, as so often happens, is not up to date with current science. In this particular area, even the ethics considerations are still very fluid. It's a very hotly contested domain with a lot of passionate feelings on all sides, because it is at the intersection of our genomic capacity and our genomic understanding. At the moment, we can produce a lot more information than we can understand.

To address this, we launched a new consortium earlier this year focused on the issues surrounding return of results. There were several new studies funded, along with a few studies we had already funded, which were all brought together in a consortium so that the investigators working on this can get together and try to chart recommended paths through the issues.

Are there firm NHGRI guidelines right now about returning results, or is it up to the researchers how to handle it?

There are existing guidelines in the form of CLIA (Clinical Laboratory Improvement Amendments), and as part of the federal government we abide by that law-that laboratories returning information for clinical purposes or with clinical context should be CLIA approved. We have labs that are working in this intersection, and they have developed different plans on how to work with these guidelines-for example, by partnering with a CLIA approved lab so that when there is a research finding appropriate for consideration to offer to return it can be confirmed in a CLIA lab. All of the projects are also operating under IRB (Institutional Review Board) review.

What we are trying to do is to build a foundation of data through our return of research results consortium from which we can make informed policy choices about considerations regarding the decision to offer to return results or not. At the moment there is an accumulation of data, but there is not a consensus, which makes it premature to design a national policy.

Our consortium also addresses how to return the data. It isn't simply a question of the wet lab having a finding and giving it back; there are a lot of questions about the appropriate way to offer to return this data to individuals so that it can be meaningful to them. Some of that has to do with ethical considerations of when the data is returned, and some of it also has to do with support and resources to make available to individuals, both before they decide whether they want the information, and afterward if they do choose to receive it. 

Does it seem to you that more studies are tending to make it an option for participants to receive individual research results?

As you said before, our goal is to have research translated into the clinic and to have more applications in the realm of genomic medicine, and that's going to involve providing more results to clinicians and patients in order to inform their decision making. So I would say that yes, we're going to see things move in that direction. What we're trying to do now is foster the data on participant perspectives about the scientific knowledge and the ethics conversation, so that we can be prepared to form the policies that will guide options to return results in practice.

Do you think the same is true of returning incidental findings-that is, if the variant to be returned is something that the participant wouldn't have thought about in advance?

Again, I would say that the norms about when it is appropriate to return an incidental finding-when is it even appropriate to look for incidental findings-are still developing. Right now, it's important for laboratories to work with their Institutional Review Boards on this question, because it's going to be very case specific. From a national standpoint, it is difficult at all times to make policy that will fit every situation. Right now, when there are so many moving parts and so much changing knowledge in this field, it's incredibly difficult to come up with a policy that could be applied across the board appropriately.

Part of what our researchers are looking at is what the obligations should be for offering to return findings, and if there is some obligation, how do you define which conditions you have an obligation to look for? In some regards, returning findings that aren't related to the purpose of the study might actually be very similar to returning the research findings to participants; but in some ways it might be more difficult, because the investigators and physicians (if there are physicians involved) aren't as likely to have expertise in diseases outside of their research area. That makes it even harder to think about the right way to provide the appropriate resources to the individual, let alone provide any clinical care for the individual.

At the same time, this is part of the dilemma when an investigator does run into something and feels that it's important because there is some intervention or utility the individual might have, medically or on a personal level, from knowing this result.

That's something I wondered about: What actually happens when a lab technician or a researcher realizes that they have information that could help someone, even save their life-what do you do next? What if there isn't a clear protocol for what to do, or what if you've told participants you aren't going to be returning any results? ... Or is this just a hypothetical situation?

I don't think it is just a hypothetical situation. I know there are investigators who have had a very difficult time with this. They have found things that could be important, and their IRBs have advised them that they should not return the findings because in a consent form five months ago or five years ago, a participant said "I don't want this information." That is very difficult for the investigators, because they believe the information is very relevant.

There would also seem to be a logistical issue of returning results to participants because if you agreed not to return results, you may have de-identified the samples and can no longer match them to the participants. But if you have an individual's whole genome sequence, couldn't you potentially use that to identify them?

It depends on what you're testing. If you're doing whole genome sequencing, you have enough information to find a unique pattern. But if you've made a commitment to de-identify it, there would be questions about changing the plan after the consent process. There's also the problem of what other information you have that you can match back to the sample. When you de-identify a sample, you're doing it for participant privacy and confidentiality, so you're not keeping the other information you'd need in order to make a match to the sample. In order to re-identify the sample, you'd have to break the policies you put in place. It is very dependent on study design-and that has been a mechanism for solving this issue, to say "we don't have sufficient information," or "we don't think the information we get back will be sufficiently informative to return to participants." Then you can tell participants upfront that you won't be returning results, that this is only for research purposes, and you can de-identify the data.

It seems that in a situation where you give the option to return the results, there needs to be a way to de-identify samples, but also to re-identify them. How is that navigated?

If you do plan to offer to return results, you would keep a link. You're not working in the laboratory with individuals' names on the test tube; you have an anonymous link so that the people working with the samples don't know who it is, but if there is information determined to be of sufficient relevance to return, the primary investigator or someone else can make that link and contact the participant.

Do you have a sense of whether having the option to receive results impacts how likely people are to participate in a study?

There have been studies that show this issue is a factor in participant perspectives about research participation, and there are anecdotes that suggest this is why some people choose to participate in a study-because of the possibility of having results returned. That's not universal, and we need to understand the issues better. This issue is something we are struggling with in genetics and genomics, and we're struggling with it both in cases where we might know what a given result might mean and, even more vexing, when we don't know what a result might mean. It's been a conversation that participants and investigators have long had in clinical trials: Their participation in the research is not a guarantee that they are going to benefit in any way, and in fact, a fundamental principle to convey during consent is that the research is not for the participant's personal benefit. I think it's something that physicians and their patients have struggled with for a long time. Many still make the decision that even though it might not help them to participate in a study, they want to help others. I think many make that same decision when they are participating in a genomic study but may not be able to receive individual results. But we need to understand the associated issues better before we make assumptions about what motivates participants, and before we start making policies based on those assumptions.

Search: GeneWatch
The use of forensic DNA databases by law enforcement around the globe is expanding at a rate that should be of great concern to civil libertarians.
View Project
Women and Biotechnology
View Project