So letís begin by going to Chapter 5 again. There is a section in Chapter 5 which we have much discussed in one way and not at all discussed in another way. That has to do with howóit begins on page 211 one way or another and goes to 213, and then a special case of that over on 215ódealing with the interests of groups you might say, people other than the research subjects themselves, and what responsibilities, if any, we wish IRBs and/or others to take on in this respect.

We have discussed it in many different ways but we havenít really focused on what we believe we really want to introduce into the various considerations that we are accumulating here, the interests of those who are not directly a party to the investigation as the investigators and the research subjects themselves. Alta mentioned something about this yesterday in one of the comments that you made, and, of course, it has come up many times. Kathi has produced some language here, including some proposed things for us to think about, and I think we should start addressing that issue.

One possibility, and Iím not recommending it actually but itís a possibility, is just to say that we ought not to worry about this, this is somebody elseís concern. Another possibility, obviously, the other end of that spectrum, that it is a deep concern which needs to be taken into account just like the concerns with respect to individual research subjects. So, I could ask Commissioners to address themselves either to the particular suggestions that are made here, or to the issue in general.

Iím sorry weíre going to quit this discussion, completed or not, at 8:45. At that time, we will go on to something else, but the next step in the work of the Commission in this area will be to produce a brand new draft of this report. I would, myself, hope within about six weeks to have a report that may not be finished, we may not have resolved all of the issues, but would be in good enough shape for us to send out for public comments, reactions, and so on, even though there might be some unresolved issues still at that time.

So let me open the floor for discussion of these issues.


DR. LO: I would urge us to try and say something about this, if only to call attention to it as a problem. I think that this is one of the areas where DNA-based genetic research on samples is different, may be different from other types of research in stored samples. I think to say that we recommend that investigators think about it but not want to yet try and write binding regulations, I think, is a good approach.

On the one hand, I think we need to just find out more information. What are the kinds of problems that we not be able to think about now but weíll only learn about if investigators and IRBs focus their attention and consider it?

And secondly, I think we want to try and get people to do more than just the bare minimum. To me, the regulations are a floor that you have to do it, but itís kind of the minimal decent things that you should do. And what we really say here, it seems to me, is think hard about things that you donít have to, weíre not making you do, but it would be good to do because thatís most likely where weíre going to find out about potential serious problems that now we canít even envisage or contemplate.

So Iíd like the language of recommending making it not binding on every case, but whatís binding is you have to think about it and not that you have to respond to a certain new set of regulations.

DR. SHAPIRO: Alex, then Eric.

Professor Capron: I think I agree, basically, with Bernieís thrust. If we do that, we have to be careful then in the boldfaced language on page 212 not to describe these concerns as requirements, which is what we do now.

And I also want some help from whoever on point C, which says "Consider the implications of disseminating research results where such results may identify individuals at risk of harm who are not the subjects per se of the research." Thatís, to me, very okay. Consider the implications of disseminating the research results, does that mean having gotten the results, you ought not to publish them? Does it mean risk of harm to those who are not the subjects per se of research if they are an identifiable group but not otherwise? If one of the implications is that not any particular group but the population at-large, a certain percentage of people carry a particular marker for a type of canceróbreast cancer, but not among Ashkenazi Jews, but just among women generally, or women who had uncles who had prostate cancer, or some such thing, those people are now potentially at risk. Itís just very unclear. Itís mushy what this means.

So I think if I were a researcher reading this, I just wouldnít know what Iíd been told to do.

DR. SHAPIRO: Iím glad you focused on C. Iím also sympathetic to the kind of position Bernie laid out. If I look at the boldfaced material here at the bottom, I donít like C at all. If I understand it, I donít like it. And in some sense, I donít understand it. So I donít like C at all. I think thatís rather dangerous territory actually.

The material that follows C, on the next page, begins to look more like requirements, because not only are you asking the investigator to do something, youíre asking the IRB to assess whether the investigator has considered this. And it begins to feel a little on the oppressive side to me. Thatís just my own reaction.

Professor Capron: The latter part of it, whether the risks and benefits are an appropriate ratio, is not an new assignment for IRBs. But what is potentially over the line in that assignment is that information developed will be harmful. Now, thinking back, itís not just DNA studies, Bernie, thinking back to the early days of the AIDS crisis, when Haitians were a focus of concern, people were saying this is a disease that seems to occur among gays and Haitians and then maybe drug users were beginning to be brought in, and the suggestion was, well, are Haitians a distinct group because of something about Haitians, or are they all gay or drug users? But they were painted with a very broad brush, and particularly in New York City, I know there was a great deal of concern for that community being singled out. Now it turned out there was nothing unique about Haitians at all I gather, except there was a certain amount of the disease in the country and so the prevalence was higher there or something.

But I donít know what this advice in that context would have been. If someone had been using live people or stored tissue samples, would it have been that research shouldnít have been done which would have mentioned a group? What would an IRB do in terms of risk? There is a risk of stigma from those results. So I want to reinforce agreement with you.

DR. SHAPIRO: Okay. I have a lot of people who want to speak.

Eric. Alta, youíre on the list.

DR. CASSELL: I agree that we have to somehow get this to come out in the investigatorís presentation to the IRB. But asking investigators to think about something may not be the best way because it doesnít produce anything hard. But I think that it is possible to come out with the equivalent of an impact statement, like an environmental impact statement, which specifies what you believe would be the possible impact on groups other than the source of the tissue and specifying specific terms. It doesnít have to come out. It can be laid out in such a way that the IRB can review it and then if they feel further action is necessary, let them take that action. But this would require the investigator to not only think about it, but write it down in its broadest form.

DR. SHAPIRO: Bernie?

DR. LO: I just wanted to say that with C, I agree that the language here is kind of hard to understand. I thought this actually referred to the starting point, this was the issue of publishing a pedigree, which identified specific individuals, not because they were members of the group but the person who is placed in the pedigree, you could infer from the design of have or to not have a certain genotype.

DR. SHAPIRO: Is that what you had thought, Kathi? Because thatís not how I read it.

DR. HANNA: I think that was the intent.

DR. SHAPIRO: I see. I didnít read it that way at all. Iím sorry.


MR. HOLTZMAN: I think the point there, I donít have the exact language, but you can write it to say that if you can publish in a scientifically valid way, publish the results in such a way to shield the identification. And we do this all the time, right.

DR. SHAPIRO: I understand that quite well. I agree with that. I just didnít read C that way at all. I misunderstood. I apologize for that.

Okay. Alta.

MS. CHARO: Again, Iím going to rely heavily on the lunch conversation I had with Karen Rothenberg yesterday at lunch just for the sake of the record.

DR. SHAPIRO: You didnít have lunch twice, Alta?

MS. CHARO: No. Dinner was fun though. It seems to me that there are two parts to this question. The first is how is an IRB or some other body supposed to interact with an investigator on this topic, at what points in the decision process and with what degree of authority. And the second is what are they supposed to be doing.

Starting with the "what," it seems like it might be appropriate for us to make a suggested list of IRB kind of bag of tricks that we would suggest that they become good at, and we also suggest perhaps that OPRR help them get good at through further work in the guidebook based on the kinds of suggestions we make here, in which we identify for the variety of scenarios, whether itís population-based studies that divide things by ethnic group or itís family groups, both of which have an effect on third parties, as appropriate design issues. Why are you using this particular population and not another? Thatís what has generated the interest in the Jews is that it is the convenience of the collections from Tay-Sachs that is generating the perception that a lot of genes are associated with Ashkenazi Jewish heritage, not simply that there was something found in those collections. So design issues of why this population, and can you expand the population and still do your research.

Second, advice on how to do community consultations, in which how to identify community leaders that you think might be worth talking to to identify sensitivities that you may not have thought of to see if thereís a way around the design. Again, nothing that has to be done. And a bag of tricks that listed things like that.

Then on the "how," there are two possible places where this could come up. One is for IRBs that are already reviewing a protocol because it was supposed to come to them. This can simply be a non-required part of the discussion. We make the following suggestions to you, we do that all the time, to go along with the things that are mandatory.

More radical is the notion that institutions should be encouraged to change their practices and perhaps even embody that in an altered MPA to say that people who are doing exempt research will nonetheless have to have a conversation with their IRB if these issues are present. Very complicated in terms of knowing that you need to self-refer, dealing with the administrator, getting the conversation going. Thatís more radical, hard to operationalize. But thatís the suggestion that was on the table and I promised I would pass it along for Commission discussion.

DR. SHAPIRO: Thank you.


DR. LO: I want to try and tie this section into the discussion Professor Gunsalus had with us yesterday. I guess I would like to suggest that as part of our scut work reduction act for IRBs, we ought to think about what would we really like them to be doing. It seems to me one role that weíve not really talked about is they may be the best site for a thoughtful investigator to come and say "I have an issue here that Iím concerned about. Can we talk and letís learn about how best to deal with this?"

To the extent that weíre looking for some place for people to think more about these issues, weíre not going to settle a lot of things in this report, but if we can highlight certain issues that we think more thought needs to go into and investigators and IRBs and others need to think through, we should highlight these. But also somehow, if thatís what we want to do and think in the long run thatís going to be useful for IRBs to do, we need to point them to that. But then I think also take away some other things that we otherwise might ask them to do and say we really want you not to be looking at paperwork requirements, but to take the lead at your institution in having substantive, thoughtful discussion. At least we should present that to some IRBs as something they may want to take on as a mission.

Iím thinking again that what the Mayo Clinic did by spinning off a satellite IRB and having them really develop thoughtful policies and educate their researchers may, in the long run, be something that will do more to protect human subjects than sort of tightening up on some regulations.

DR. SHAPIRO: Bernie, can I ask you a question about that? I think the notion that, and weíve talked a lot although not said a lot about it, that the responsibilities that are at play here are not that the investigator has responsibility, the IRB has responsibility, the institution has responsibility, itís a kind of shared responsibility; and the idea that an institution, public or private, might have a place where investigators could go to talk about how do you design experiments, what things you should be sensitive to, sounds to me like a very useful idea.

Ideally, that, in my own sense of it, ought to be separated from the IRB so the IRB maintains a function of saying, yes, this is either satisfactory or not. Thatís just an alternative. I think the idea is an interesting one, although Iím not sure it belongs in the IRB. We havenít had a chance to discuss. But itís an interesting idea.

DR. GREIDER: It sounds like there is some support for keeping some kind of language relative to this group issue in here. Eric had mentioned that if we have it not be a requirement at all, then itís just not going to occur probably. And to have it be something formal, as minimal as having to write something down, that the IRB reviews I think is a good way to get investigators to actually think about the issues.

DR. SHAPIRO: How is an investigator supposed to know or supposed to imagine that heís onto something, sheís onto something, which a large group out there, undefined or not easy to define, is going to be impacted? How is one supposed to reach those decisions? Now there are some obvious cases, we can all generate some, but there are a lot of cases which arenít obvious and the concerns the groups have change over time and the way the groups comprise themselves change over time. How is that supposed to be resolved? How does an investigator know, if weíre going to insist, which may not be a bad idea, they write an impact statement like Eric said, how do we know when to do that?

DR. CASSELL: Itís the same way they began to discover that there were ethical implications of their work. We all remember in the early days of IRBs, even before the regulations, there were no ethical implications of lots of peopleís work. I mean, this is just the way it is. But they began gradually, even the oncologists began to realize that they might as well tell the truth. And they did because there was a requirement that they do this. And this is the same thing. We make the requirement and theyíll come to know because, after all, there are competitive forces going on in that university and within the IRB; there are forces acting on them and countervailing forces so that things happen. But youíve got to have some requirement, otherwise they wonít.

Professor Capron: I havenít noticed particularly if people are competing to be the most ethical.

DR. CASSELL: No, no, but they are competing to get the things through.

Professor Capron: But that suggestsóI mean, if you take Altaís point from Karen Rothenberg which suggests a major alteration of the regulations and the requirements imposed by the IRB to take research which is otherwise exempt because it is using publicly available, nonidentifiable samples, for example, doesnít qualify, goes right around the process, right now the investigators send it in, and now the IRB administrator is going to look it over and say, well, it would have been exempt, but we think you may have some implications for groups here, and so at that point you have to file an "impact on others" statement. Is that wható

DR. CASSELL: Weíll have to figure out how we make the specification. But certain kinds of work require you to state whether the work you are having might have an adverse impact on ethnic, etc. groups and what might that be? It doesnít have to be very complicated.

Professor Capron: But your work itself may create a group that didnít even exist before.

DR. CASSELL: Yes, it may. But weíre not primarily concerned with the little subtleties on the other side of the curve. Weíre concerned primarily with the center of the curve where people donít pay attention now.

Professor Capron: But Iím not quite sure what this impact statementóI like the idea of people having it in their minds somehow that there are perhaps slightly better or worse ways of going about this. But suppose you are doing a study in which you propose to use stored tissue samples from Tay-Sachs screening fifteen years ago, and the point that itís a study of convenience, you donít have comparable samples from Episcopalians, so what in the end? If you do find that there are risk factors or particular risk factors for a disease in that group, is that wrong to have done it when you know thereís going to be an impact? What do you do with that?

DR. CASSELLL: Letís go back a step. If Bernie is having us think about it, what are we thinking about? Our work sits here isolated? We were only interested in Tay-Sachs. The fact that a whole population suddenly felt themselves threatened by a disease when, in fact, they may not be so threatened, thatís none of our concern. After all, all we did is identify that a certain population has a prevalence of something. What do we care if it got people upset or not?

In other words, the thinking about it is that process. What is the larger impact? And you can take the genes of the prevalence of breast cancer or ovarian cancer as the perfect example of the creation in a group of a bunch of fears that wereóexaggerated might be not strong enough.

Professor Capron: Eric, I wasnít kidding when I said that the work itself, even without any ethnic identification, may do that. People like Barbara Lockwood have written eloquently about the way in which research looking at breast cancer has now made women generally more apprehensive about that disease, although compared to other risks in their lives it is not a huge risk. But the fact that there are genetic factors and now the ability to identify two of them, BRCA-1 and 2, has generally elevated concern; and you might say that has caused discrimination, stigma of women generally.

DR. CASSELL: And you donít think that any investigator knew anything about that?

Professor Capron: Well, what do you do once you have that knowledge? I donít understand. If the statement that youíre going to file is simply, "If we find a risk factor exists and are able to identify it, all the people who might use this screening test who are potential candidates for the test will realize that they are at greater risk." Period. Iím not trying to be obtuse about this. Iím really trying to think what we expect to come out of this in terms of changes in behavior.

When you talk about investigators not being sensitive thirty years ago to telling people that they had cancer and the reason they wanted to give them experimental radiation was that they had cancer and they had to disclose the diagnosis, I can see both the change in behavior that was expected and a clear ethical mandate for it. With the problems that new knowledge brings, itís harder to understand what change in behavior other than not developing that knowledge one would have.

DR. SHAPIRO: First of all, let me remind you we have ten minutes left for whatever part of this discussion weíre going to have today. Alta, Bernie, and Rhetaugh are next. Letís see what they have to say and see how far we get.


MS. CHARO: Somehow I feel like around the table among the Commissioners our understanding of the range of situations and the kinds of solutions has not deepened as much as we might like over the last year, which may be why weíre struggling to come up with a set of operationalized suggestions what IRBs ought to do. The situations are enormously varied, as this conversation indicates, and there are other situations that arise in IRBs. People want to duplicate old studies and the old studies themselves suffer from these problems. Do you recreate the problems in the name of duplication, or do you now expand your study? Weíve got studies that are on Caucasians only, for example. So itís not about stigmatizing the group, itís about not generating generalizable knowledge today.

Perhaps the more appropriate recommendation, which goes to the heart of investigator education, which is really the bigger thing here, is to ask to convene several meetings with people who actually sit on IRBs and people who actually do this research to really talk about the list of situations where this might arise and what some possible solutions might be to ameliorate unnecessary harms to third parties in each of these solutions and create that bag of tricks. And then, with that bag of tricks in hand, figure out where, if at all, it belongs in the IRB process as opposed to a more generalized investigator education. This may sound like a bit of a slide-by, but maybe that goes to the center of the problem is that we donít really know enough to know what we want to accomplish.

DR. SHAPIRO: Bernie.

DR. LO: I agree with Alta that we donít really know what the range of situations and what the range of solutions is. But I think the conversation that Eric and Alex had is exactly the kind of conversation I would like somebody at each research institution to try to stimulate and to have it with, not just in front of, a bunch of investigators who otherwise wouldnít be thinking about it.

And to respond to Alex, I think we should try and be clear in the text that we donít mean donít do the research necessarily for most of the time. But we do, I think, mean things like when you write it up you ought to have a couple of sentences in there saying whether or not there are any implications for Jewish people in your study as opposed to all other people. Sort of explicitly say both in the paper and when you talk to the press, "this doesnít mean that this group is more burdened than other ethnic groups by the risk of the disease.

Also, I think you may thenófollowing with Ericóonce as a naive investigator I started thinking about this, I might later on decide, Iíd better go talk to some womenís advocacy groups and explain my findings to them in lay language, and not just say itís not my problem, I just do the research, but to really do more than that. So I think that the list of solutions, I would like to see us start to develop a list, but I agree with Alta, that list is really not going to come out of this Commission. It is going to come out of investigators, IRB members, and other institutions focusing more attention on it. But maybe we can at least jump-start or catalyze that process, which is not now taking place.

DR. SHAPIRO: Rhetaugh.

DR. DUMAS: I think it is very important that we have something that addresses this issue. I think we need to be careful how we state it for fear it might be seen as a statement that certain kinds of research should be avoided or constrained, and I donít think thatís what we intend to do.

I believe that it is possible for investigators to consider these issues and develop greater sensitivity to the possible impact that their research will have on a broader group. I think that there are ways that the results can be reported to minimize a harm, as has been mentioned before, short of withholding information or not reporting honestly, but rather in the way that the data are presented and also the way that the report is written. It can have a certain sensitivity to the impact and make sure that there are statements that would warn people not to go beyond what the data presents.

Iím very much in favor of addressing this issue. I realize there are difficulties in doing so, but I think thatís a part of the reason why we need to continue to struggle with this.


DR. MIIKE: Having just woken up, I may repeat what other people said. I guess our discussions in the past around this issue never really involved the IRB review process. It talked mainly about involving group issues in the research design and directed investigators before they even came to the IRB. Maybe we should return back to that focus. The language here tries to leap and make it as part of the IRB process, and I donít recall any discussions along that line when we were discussing this issue.

DR. SHAPIRO: Tom, Carol, and that will be the last two comments.

DR. MURRAY: One of us will be cut off in mid-sentence. I hope itís Carol. Iím just trying to get some closure for myself and for the staff, and for the people whoíll be involved in the next draft. What I think I hear is a general consensus among the Commissioners that there needs to be some way of registeringóon the consciousness of investigators, at leastóthe idea that there may be implications for someone other than the subjects of the research. It may be family, it may be some group with which theyíre identified. We donít know how to do that. In fact, we think that there isnít enough evidence either to fully characterize all the kinds of ways in which this problem arises nor all the possible solutions, and here Iím just synthesizing what a variety of people said. But we can certainly say the first thingóthat it is important. We need to think about how to say the second thing. What Iím going to propose is that you dig into your creative potential here and provide some suggestions to the staff and the rest of us about ways in which that can happen.

I think it is perfectly fair for us to say, a kind of "let a thousand flowers bloom," that we think IRBs and investigators need to be more attentive to this and weíd like to see some suggestions from them as to how that might happen. We might suggest some ways to foster that.


DR. GREIDER: I agree with what you just said. I think that we need to focus on exactly how that gets implemented. Is it at the level of the IRB, or, as Alta has pointed out, if research is exempt from IRB review, what is a mechanism by which we get investigators to consider this? So I think it has to be a little bit more concrete than letting a thousand flowers bloom.

DR. SHAPIRO: Okay. We have two minutes left. I want to go back to a particular part of the suggestion Alta made that came up a couple of times, and that is the issue of research that is otherwise exempt, and see whether you feel that on this issue, regardless of just exactly what procedure is used, that otherwise exempt research should also sort of fall under this and therefore require review of some kind, call it review for a moment, because of this particular aspect of a research protocol. How do people feel about that? Is that something we want to do? I donít know if itís quite radical, but it is more extreme than the earlier one, thatís all. I donít know if itís quite radical.

DR. MURRAY: It narrows the field at least. You wonít have pedigrees presumably if it is research that otherwise would have been exempt, or am I wrong about that? Is that correct, Kathi, do you think? But you will have information perhaps about ethnicity. Yes. So that would be the kind of case weíd be talking about.

DR. SHAPIRO: Carol, Alex, and then weíre stopping.

DR. GREIDER: So an alternative way to think about this, if we look at the flow chart for determining if something is exempt, one of the questions that you have to answer is will waiving/altering informed consent adversely affect subjectsí rights and welfare? We could suggest that at that point subjectsí rights be consideredóthat groups also be considered in them.

Professor Capron: Youíre looking at the consent chart, not the exemption chart.

DR. GREIDER: Sorry. But anyway, we could fold the group issue into one of the other decisionmaking processes rather than have it be separate. Thatís the general idea.

MS. CHARO: If I may. The institutions are permitted to exempt from review anything that falls under those Federal regs that say this can be exempt. But they are not obligated to. And many institutions choose to have things that could be exempt nonetheless come before them for one reason or another, and thatís probably the place where an institutionalized change could be made, or IRBs could simply institute something, or the deansí offices and provostsí offices, et cetera, institute something much more informal in terms of generalized education that doesnít commit them to any kind of formal interaction.

DR. SHAPIRO: Okay. Alex, and then thatís it.

Professor Capron: I think the logic of everything that has been said up until now is that we should endorse IRBs moving in this way. In other words, a good IRB will do this. Because the whole point of this discussion is that there are interests totally independent of any direct interest on the people whose samples are being used and it doesnít matter if their interests are so attenuated that the research is exempt. So the logic would be, yes, we should either urge regulatory change or urge voluntary steps in that direction.

DR. SHAPIRO: Okay. Thank you very much. I want to repeat what Tom said before. Any of you who have views, and most of you do have thoughtful views, on aspects of this report, if we could get those in writing one way or another, that is a huge help to us rather than wait until our next meeting and then hear from you about what your views are. Thatís not very helpful because that just prolongs this process. I am anxious to put some pressure on the staff and ourselves to get this report out for some public review. I certainly want to do that in the next month or two. So I really plead with you to, on issues on which you feel strongly and care about, to let us know and let us know in writing one way or another so that we can be as responsive as possible to them.

Letís move on now. Eric, you have some brief reports on some other activities.