DISCUSSION OF PREVIOUSLY COLLECTED TISSUE SAMPLES COMMISSION MEMBERS DR. MURRAY: Let us jump right into the first item on the agenda, which is-- The agenda today is basically just in three big chunks, except for the public statements. The first chunk is previously collected samples, the second chunk is community consultation, and Bernie, I hope, is going to lead us through that, and the third is tissue samples collected after whatever the effective date is of our recommendations. And we have a sample of the work that has been done by the National Action Plan on Breast Cancer that we can look to for that. At least one member of that project--Debbie Saslow(?)--is going to be joining us for that conversation. So let us begin with previously collected tissue samples. Does anybody wish to start? (No response.) DR. MURRAY: Do we know where we are on this? Zeke, would it be helpful to put your-- DR. EMANUEL: Do you want me to put up the old-- DR. MURRAY: --plan up on this? DR. EMANUEL: --framework? DR. MURRAY: Sure. DR. EMANUEL: This is just a framework. And I think one question is whether that framework still holds or whether we want to re-think it. And I think I have the recommendations for the proposals I had. I guess one question is whether that--those boxes--still makes sense to people, having thought about them now for about a month and a half. REPORTER: Excuse me. Could you use your microphone. DR. EMANUEL: Sorry. DR. MIIKE: Zeke, the bottom two, when we get to community consent-- DR. EMANUEL: Right. DR. MIIKE: --what is the difference operationally between community consent "full" and community consent presumably with "opt out?" How does a community opt out? DR. EMANUEL: Well, they raise objections I think, as opposed to-- I mean, one is putting the onus on the researcher; one is putting the onus on people out in the community who want to object, I think. That is the way I imagine it. One is we do something to inform people what we are up to. We distribute a letter, we contact organizations relevant, and we wait for them to respond. The other is we actually, as researchers, go to them and solicit their advice, but we don't go-- We aren't permitted to go ahead unless we have some sign-off that we think is a sign-off. So I think one, you know, it is a measure of who has got responsibility and where the responsibility for raising the concerns lies. It is also a measure of how much I think leg-work, effort, for really getting the community, or community leaders to sign off on it. MS. KRAMER: But, Zeke, we haven't-- We didn't really discuss all of that. DR. EMANUEL: No. No. I was suggesting initially just are the boxes around, and then we can talk about the content inside. These are my ideas. MS. KRAMER: Right. DR. EMANUEL: And they are all tentative and they are not to be suggested for the commission. If you want me to put up the other one, with the blank boxes, I am happy to do that. MS. KRAMER: No, no, no. I just-- I just wanted to make that point because I think Larry missed the meeting at which we initially began going through the boxes. You were a voice. DR. MIIKE: No. The next question I was going to ask was that I assume we are going to--Bernie is going to--lead the discussion about community consent. DR. LO: The next section? DR. MIIKE: Yes. Right. MS. KRAMER: Right. DR. LO: Yes. DR. MIIKE: Yes. I caught it. I think I was up at the tail-end of that part. MS. KRAMER: Right. DR. LO: (Inaudible.) DR. EMANUEL: No. The bottom-- Yes. You are right. I didn't update it. Or I may have updated it, but I now have so many overheads I can't remember. DR. MURRAY: He changed his mind. DR. EMANUEL: No, no. I mean, I think the first-- One question is whether these-- We are now comfortable with these boxes, and obviously these two boxes presume something about community consent. We know that, for example, at least at one--(Inaudible.)--to the large commission, Jim Childress raised about whether we use these boxes or not. But, I mean, we have done some interesting things here. One is we talked about previously collected samples. Sorry. Another, in this box, in the previously collected samples, was to fuse the clinical and research into one category, not to separate them out. To treat them as the same. To have one set of rules for both. Then we talked about not how the samples were collected, but how they are going to be used, so that we don't talk about anonymous samples, or anonymizable samples, but samples that are going to be used in an anonymous manner. I still can't say that. And then samples that are going to be used in an identifiable manner. So in this box are samples that are collected with identifiers, but the research is going to be done such that the identifiers are expunged, or encrypted. So those are, I think--going down--those are the major decisions that, you know, we have talked about. I don't think we have finalized anything, but that is what is here. And then along this column is these three divisions, which we have had for a while, but have never, you know, sort of had to stand behind. DR. LO: Zeke, the extreme bottom left box? DR. EMANUEL: Yes. DR. LO: You know, what is that supposed to be? I mean, it looks like-- DR. EMANUEL: Oh, I am sorry. This-- I do have-- Too many overheads. Hold on one sec. I have a separate overhead that has it correct. I apologize. Yes. It got shifted over when I made this. It is supposed to be-- DR. MIIKE: Potential harm. DR. EMANUEL: Just to put this in context, I will move the recommendations off, and just put the empty boxes in while we are talking about the empty boxes-- It is supposed to be community where there are potential harms. Community-- DR. GREIDER: I recall a discussion about collapsing those two boxes into one. DR. EMANUEL: Yes. We talked about that about two months ago. DR. GREIDER: And we are talking about just the outline, and so maybe we could discuss it, in terms of the community things, whether there should be two or one. DR. EMANUEL: Yes. By going down this way, we may-- I mean, we may begin to feel comfortable that we have made these decisions, which I think in some sense are slightly easier decisions, although one shouldn't minimize it. These are pretty profound changes in the conception of how we are going to deal with things. And then get to this, which I take it is our intuitions are a little more divided, and we know that there are at least some voices in the whole commission that haven't gone through. DR. LO: Yes. I missed the last couple of meetings, so I may need to be brought up to speed here. In terms of collapsing the distinction between samples that were originally collected in the course of clinical care, so the archetypal example would be cancer removed at surgery versus samples that were originally collected in a research setting, could you review for me the reasoning behind collapsing those two distinctions? DR. EMANUEL: I think part of the reason, in the previously collected samples, followed the following. In both there wasn't any understanding previously that they would be used, stored and used, for future research where people didn't consent. And there was a sense, in this case, that the distinction on the rules we might make between these boxes just wasn't significant. There really were substantial differences on the substantive matters. Now, it may be that we should go through that again to think it through. We have, you know, retained that distinction somewhat here but, again, I think part of, or the main purpose of the meeting is to go through and see whether we still, you know, whether that is really our settled judgement. DR. LO: Well, just for my clarification, I mean, making things similar can either mean you move this one over to here, or you move this one over to here. And I guess my concern would be that I have heard arguments that, well, if you consented to have some of your tissue taken for research purposes, it kind of stands to reason that you would like your material to be used for other research, and so you are probably not going to object if some other researcher comes along with a research project that is really on a very different subject than the topic you originally consented to. As long as we don't have that presumption that, once you have consented to research we can sort of assume you are going to just consent to any other project, I would feel comfortable collapsing it. DR. GREIDER: The way that I remember this discussion going, it was more in the other direction in that the kinds of consent forms that might have been used for research and the kind of consent, or lack of consent, that would have occurred in clinical care was so thin that you should consider them all-- DR. LO: To be not consented to. DR. GREIDER: --as if they really weren't consented for, for future uses. And that is how I recall the discussion going. I don't know if other people agree with that. And that is why it collapsed more in that direction if there wasn't as much consent going. DR. LO: I definitely agree with that. DR. MURRAY: Just to put this in context, very briefly, since I see some new faces--at least among the faces--I don't recognize among the visitors. Thanks to the work of some of the people who have helped the commission on this, we know that there are over 200 million identifiable samples out there of human tissue in the United States, of over 100 million different-- Well, 100 million people. We know that the research--that the tissues--can be used in some very fine kinds of scientific research. We know that people have significant concerns about privacy and confidentiality. The evidence we have also indicates that most the people, at least who we have spoken, or who have spoken at our mini-hearings, are supportive of scientific research and would like to see research go ahead, thinking that more good will come from it than harm, so long as individuals can be protected against discrimination. So those are some of the parameters within which we are working right now. We are talking about stuff that has been collected until now, often, as Carol described, with--you know, this is not to impute the motives of the people who collected it--with a kind of minimal informed consent. We also know that most people seem to have no recollection--the people we have spoken with--that they ever consented to the use of their tissue, but in fact you can, in many cases, point to their signatures on the consent forms to indicate that they did indeed consent to those uses. So these are some of the background conditions under which we are working. MS. BACKLAR: And I am doing quite a bit of catch-up because I have missed quite a few of these meetings, too. But, as I was trying to catch-up and read through some of this last night, what I was looking for is what about the issues of people who maybe were decisionally impaired? And when you are saying that not many of these people could consent, but have we been thinking at all about people who can't consent where their tissue might have been taken? Are we making any allowance for that in these retrospective in collected tissues? DR. MIIKE: Well, I don't see how we can. You would have to go back to each of these individually, and then I still don't think we get it. So I think we are trying to take a broad hit about what areas where we are not seeking consent-- MS. BACKLAR: Right. I understand. I just don't want it left out that there is a-- That you are forgetting about a group of people who maybe there was never any consent either, because they-- DR. EMANUEL: Well, someone consented. I mean, generally what happens is someone consented to their surgery, whether they did or a proxy did. And that consent to surgery sometimes contains a sentence and sometimes doesn't that permits people to do it. So, to the extent that proxy consent for surgery, or whatever else is taken, is acceptable, at least on the clinical side, that is usually--and also on the research side--that usually is-- MS. BACKLAR: How are you going to counter it? DR. EMANUEL: Well, that is how you encounter it. MS. BACKLAR: Right. DR. EMANUEL: I think the general question is, you know, the sort of conceptual framework, and then we can talk about the details of protections. MS. BACKLAR: All right. DR. EMANUEL: When we-- Because the framework, I mean, the framework, you know, I think we should be clear. The framework is not value-free. It has got a lot of normative claims in it and, you know, that is why I think just talking about the outside of the boxes before we even get to the inside is reasonably important. DR. MURRAY: Yes. We agree on that. Steve? MR. HOLTZMAN: You know, just to follow on Carol's point, I think--and it touches yours--effectively what we are doing is saying that operationally there was no consent for the extant samples. Just assume that. Against that backdrop, what ought to be done with them? And effectively the recommendation is to say that, to the extent that research is conducted in an anonymized manner, no consent is necessary, no additional consent is necessary. DR. MURRAY: I would actually-- I am not quite comfortable-- MR. HOLTZMAN: Putting aside community. DR. MURRAY: Yes. --with your characterization of it, Steve. There is research on what informed consent, the process, and what people remember, et cetera. And, you know, sometimes it has been used to claim that informed consent is completely meaningless because people can't recall what was revealed in the consent form often times relatively soon after they signed the form. I am not sure that is a valid inference from that, even from that data. To me it may be that someone looks at a question, makes a call, has no objection to it, signs, and that is it. You know, just doesn't feel any need to retain the information. So I would say perhaps a characterization would be that people, when they were given a choice, had no objection and so signified their consent. MR. HOLTZMAN: I guess when I said operationally--Tom, I don't disagree with what you just said--but that, insofar as we are not going to require of people, or request of people, an inquiry as to what was the consent, whether there was decisional impairment, whether in fact there was a general consent, whether there was no consent, which is the case with much operationally, we are saying treat them all as if there was none. Now what? DR. MURRAY: Fine. DR. GREIDER: I think, again, because we are talking about lumping them all, sort of to reiterate what Steve was saying, if we are going to lump them all, and say any existing samples, then we have to decide what level of protection are we going to have on those, and I think that we are going in the direction of the level of protection that had the least kinds of consent. DR. MURRAY: Right. DR. GREIDER: And so, in lumping it, we then go toward the most-- DR. MURRAY: Right. DR. GREIDER: Sort of the least common denominator in that group that we are lumping together, if we are going to lump them. DR. MURRAY: Yes. MR. HOLTZMAN: Or you could go the other way, right? DR. GREIDER: You could go the other way, but I-- MR. HOLTZMAN: Yes. Well-- DR. GREIDER: --understood all of our discussions in the past going in that direction. DR. MURRAY: Right. DR. LO: If I could ask another question about sort of the outline of the grid. The relationship between the research question and the condition for which the sample is originally connected isn't a parameter in this framework. And I guess I want to raise the question of might it not be the case that, under these conditions of not having very, you know, sort of thick consent, that it might make a difference as to whether the research question the researcher proposes to address with these stored samples is pertinent to the condition for which the sample was originally collected. So, for example, if I come in and have a biopsy done for colon cancer, I think it stands to reason that I would probably be interested in having scientists investigate something that pertains to the diagnosis, pathophysiology or treatment of that condition. But if someone just said, "Gee, there is this amazing tissue archive and I have a totally different research question that has nothing to do with that condition," are we going to treat those two sort of protocols the same in terms of the level of review? DR. EMANUEL: I think-- DR. LO: I ask that just because I think the paradigm we have in mind, when we talk about this, I think is the good science, so it is someone saying, "Gee, wouldn't it be interesting if we could find a genetic marker for predisposition of this condition which would lead to early diagnosis of the treatment?" But I think there also is-- There are a lot of proposals made that I think are either of questionable scientific merit or, frankly, you know, come out of political or social agendas. And I think, again with genetics, that whole background is-- And I think this isn't just a historical thing. I reviewed a report Eric Meslin is working on, on the genetic basis of behavior and, you know, there are a lot of studies being done on the genetic basis of antisocial behavior, by which they mean everything from school truancy to arrests for violence. I could-- One could imagine someone with enough sort of, you know, open-minded scientific agenda pursuing these questions which, you know, are interesting and important questions of ethnic differences and, you know, propensity to antisocial behavior. And I guess, from the point of view of someone whose samples were originally collected for, you know, the diagnosis or, you know, clinical study of--whatever--heart disease or cancer, it may make a difference as to whether the researcher is proposing to study those questions as opposed to something totally different and, in addition, where the nature of the study of the hypothesis may be objectionable to some people. DR. EMANUEL: I think we have talked about that and--I will speak for myself--one of the problems is you immediately get into what actually was it collected for? And this is particularly true on the clinical side where the objectives are-- So here is the example. Here is an example. You went in for a breast biopsy but, like most women, the breast biopsy actually is negative. So was it originally collected for cancer or for breast biopsy? I mean, the category you put it into turns out to be a little vague, or not so much vague as malleable. Or only in the ones that really were cancer can you test for, you know, cancer. If it turns out to be a benign biopsy, you can't do any cancer research on it? Or, similarly, you know, you go in for, you know-- Is it Tay-Sachs disease, or is it Jewish genetic diseases that you are looking at with those samples? So the categorization you put it under I think turns out to be something that could be changed or recharacterized and it is, you know, one has to do a little bit of a leap of faith into what was in people's minds supposedly when they consented to whatever the procedure is on the clinical side. On the research side, we now know, for example, you know-- Again, the examples I like to use are the Physicians Health Study or the Nurses Health Study. They are collected for a very narrow type of research issue but now a whole slew of questions, that certainly weren't in the minds even of the investigators--you know, genetic test for propensity to thrombosis, et cetera--now become relevant on those samples. And depending on how you want to characterize what the objective when it was originally collected was, you know, you can either include or exclude those. And I think one of the problems one gets to-- And I confronted this and I reported it to the subcommittee. When I tried to write a sort of generic informed consent--not for the previously collected samples but for the future collected samples--it becomes very difficult to imagine how you are going to phrase those kinds of-- "I will permit it for this but not for that." And maybe we have been slightly--I suggested last time--maybe we have been slightly skewed because of the example we have, which I think is a good one, from the National Action Plan on Breast Cancer. I mean, they start out with a small purview, right? Women who are coming in worrying about breast cancer. So you have got cancer, breast cancer, and then you add onto it genetics or no genetics. But if you are trying to get a generic sample, that anyone who comes in for clinical care is going to fit into, I think suddenly that paradigm quickly breaks down. DR. MURRAY: We are going to talk about the future stuff this afternoon so I am just going to-- DR. EMANUEL: I was just using that as an example. DR. MURRAY: A very good example, and very appropriate. DR. GREIDER: So, in addition to what Zeke said, sort of how you categorize things, I think that the kinds of concerns that you were raising about the research--that people might have concerns about certain research that is done--that those things I would hope would be addressed at the level of the review of the research or the IRB, or some other panel that is looking at the actual research itself. Rather than attaching it to the tissue samples you attach it to the research. DR. LO: Yes. I think that is a terrific point. And so my question then is what mechanism for review of the protocol is there so, for instance, the IRB either gives no or minimal review, or just administrative review. Is that sufficient review of the research protocol? I mean, if the study is not funded by an agency that has strict peer review, and many of these studies may not be, and if the IRB is only giving administrative or less review, then I think the question is where is that review going to take place? But I do agree with you that it is not a function of the consent because we are sort of speculating at that point whether people would want it. But it seems to me we would want to have some mechanism for review. And there I think we are faced with a dilemma of how much are we willing to trade-off for review which may have some delay of research versus no review or minimal review, which may let some things slip by that, in retrospect, after the study is published, people would say, "Well, wait a minute. How come they were permitted to do the study?" DR. GREIDER: But I think that that is where there are some-- That is why this is structured the way it is. There is, you know-- Are there individual concerns or are there community concerns? And the kinds of things that you are raising I think would fall into the community concerns area. And so then the question is what do we fill in that box as to what the kind of review is? DR. LO: Right. DR. GREIDER: But I guess we are discussing the framework. First, we will go through the framework and then we will go through-- Hopefully we will get to filling in the boxes today. DR. MURRAY: Steve, then Bette. MR. HOLTZMAN: You suggested, Tom, we will take up future samples tomorrow; I mean, later today. For those of us who got in at 3:00 a.m. (Laughter.) DR. LO: It is tomorrow. MR. HOLTZMAN: Yes. But I can't help but wonder, in listening to Bernie's question, if a lot of how we think about the existing samples is not really shaped by how we think about the samples that we will be collected tomorrow. In other words, if you believe that, with respect to a sample taken tomorrow, there ought to be some consent, that the individual has some say in how it is to be deployed, that, with that background assumption, where your intentionality, your decision-making, all of that can be brought into play, that when we then think about the sample where none of that was in play, you then say, "Well, how can I use it? All of that was absent." And you start to try to come up with the cases about, "Well, how would I conform with that individual's intents and desires if I had known that?" Which leads you down the kind of path Bernie started down. So I think it comes from this notion that, with respect to the sample where I did have some control, all right, if you go back counter-factually and say if I had had it, you end up concluding you can't possibly ascertain what the intentions were. And I think that is to Zeke's point. So you, I think, then have to look at how would we feel-- How do we feel about the samples to be collected tomorrow and how thick is the consent in that instance, which drives you back to questions about what is your relationship to your tissue, and to what extent you should have control over the destiny of it. Because if you think that that is really, really thick and that there is this ownership relationship, for example, with complete dispositional authority under all cases, you are going to be more troubled about thinner consent in the past. If, on the other hand, you feel, as reflected in, say, Zeke's example, that we can make distinctions and use mechanisms like opt out, in the case of the clinically collected sample, you may be less troubled about the thinness or the absence in the consent of the past. It is just a thought. DR. MURRAY: Bette? MS. KRAMER: I think that my concern is that whatever assumptions that we make; that they are very, very clearly stated. And that if we are going to assume that, in collapsing this in terms of the stored samples, that we are going to assume that any consent that was given in the past was so thin that we would not regard it as adequate going forward but, you know, in an attempt to allow the scientific research to continue to go forward, that we are going to, you know, we are going to make these rules. We are going to go work on that assumption. But state it very, very clearly up-front and probably include language that calls for a high standard of review of any protocols that will be using this research, fully understanding that a consent--an adequate consent--may not have been given. So just call for a very high standard of review, I believe. DR. MURRAY: I want to add something to the portrait of the way things were. I mean, if I were convinced that researchers in the past knew that they were going to be using tissues regularly for research purposes and that they could get enormous amounts of personal information out of such tissues, then I would probably insist on very rigorous standards, even for consent for past use of tissues, even anonymous use. I am not convinced that that is the case. And my sense is that most people-- You know, most of the people know most of the tissue was collected for other than research purposes. Overwhelmingly that is true. They were collected in the course of clinical care. Collections were built up partly out of legal requirements, to keep for quality control, to keep tissue samples for quality control purposes, and the like. It is relatively recently that--people--the techniques for pulling information out of these tissues, genetic information, have become, that the techniques have sort of reached a kind of initial maturity. They still have a way to go before this is achieved and is an easy thing to do. And it is relatively recently that the concerns about genetic discrimination and privacy have been raised to new heights. And, you know, we have people around this table who have been a part of the course that insists on the importance of protections against incursions on privacy and protections against genetic discrimination. So I think the past is-- I mean, it is not just that research is valuable, but that I think it was not unreasonable for the people gathering the tissue in the past to think that we are not doing anything all that exceptional, or all that of likely personal significance to the individuals whose tissues are being collected. I just wanted to say that that is an important part of the background for me. But that won't be true. And it is really at the cusp of not being true any longer, and so we will have to have different rules I think for the past and for the future. Larry? DR. MIIKE: First, I want to say hello to Dave who is-- David Cox is sitting at the middle table here. DR. MURRAY: This isn't even a virtual David Cox today. DR. MIIKE: A couple of things. One is that, on the issue about past informed consent, I mean, another way to look at it is that it is simply stale. I mean, it was given awhile ago and, you know, just the fact that time has passed. I just want to react to some of the things Bernie is raising. Maybe I am wrong, but what I hear, and the kinds of concerns Bernie is raising, is sort of outside the box that we are looking at. I mean, it is sort of like what kinds of research should fall under the purview of IRB review? What kinds of things should fall under the federal agency type of review? Because you are raising concerns about shady research, or research outside of these areas. And it seems to me that we really can't address it in what we are dealing with right now. And now we are sort of very, very focused on the consent issue around the whole issue about research on samples. So am I wrong or, Bernie, are you raising some of the issues that I think are really outside the discussion--the framework of the discussion--that we are trying to have? DR. LO: Well, I guess what I am saying is that, if we only focus on the consent issue without attention to the larger question of whether the research is appropriate and should proceed, I mean, consent is only one of the mechanisms by which we judge whether or not we think a research protocol is appropriate and needs to proceed. And I think it is hard. I mean, as some people here were saying, I think that, to the extent that we have questions or reservations or just don't know about the quality of consent, one may want to look at other mechanisms, such as either IRB review or community review, to satisfy us that we feel comfortable that the research ought to proceed and, you know, meets some ethical standards so that-- Although I think, you know, it is important to look at the consent issue, I am not sure we can view it in isolation with the other tools we have to review research. DR. MIIKE: Well, I agree with that. But the way that I have been looking at it is that you sort of-- We can't move and look at all those things all together and try to change the IRB--we are going to talk about the community consent issue later on-- but the issue about whether it is a legitimate research or not isn't just applicable to tissue samples; it is across the whole research spectrum obviously. MR. HOLTZMAN: Unless Bernie is saying something along the following lines; that there is a set of research activities which, if individuals consented to them, they are okay. All right? But in the absence of consent, one would question whether or not they are okay. So there is research beyond the pale, there is research which is scientifically wonderful and, in between, there is a sense in which that, in that kind of research, one ought only be part of it if one has specifically consented to it. DR. MIIKE: Yes. But, Steve, I mean-- MR. HOLTZMAN: But I-- DR. MIIKE: But, Steve, who would say, "I consent to a bad research on my tissue?" I mean, you know-- MR. HOLTZMAN: No, no. It is not-- I don't think it is an issue of bad research. I think it is an issue of, if one-- I am not saying that this is easy to do, or one ought to do this. I am just trying to frame for myself what Bernie is saying. That there could be a scope of research activity where reasonable people could say, "Well, maybe Zeke would be interested in participating in that, but I am not." Okay? It is not bad; it is not good. Just a difference about whether one wanted to participate in that kind of research. And if one could say that falls within that--all right?--then that would be the sample, the kind of thing that Bernie would be pointing to. I am not sure it is doable. DR. MIIKE: But I don't think that is what he was raising. MR. HOLTZMAN: Well, I will ask him. DR. GREIDER: Let us ask Bernie. (Laughter.) DR. GREIDER: We need to-- (Simultaneous discussion.) DR. LO: I mean, I think the harder questions are more in the gray zone as opposed to beyond the pale. I think that there are some types of research that some individuals may not want to participate in, but since we don't know--we couldn't ask them way back then and we couldn't have anticipated what the questions are--I think we may want to find some way of looking at the question as to whether a significant number of people who were included in this tissue bank might have had objections. I think another way to look at it is when you consent, particularly in the kinds of, you know-- Even the good consent where you go to a hospital, you trust your surgeon, the surgeon actually does talk to you about, you know, "When I take out the tissue we are going to use some of it for your diagnosis but we would like to use-- The standard routine is to archive part of it for all kinds of research." And then explain to you the kinds of studies that are usually done. It seems to me part, to the extent the consent had any meaning, you had some idea that good scientists would be doing it, they would be working on important research questions in a rigorous way. And in a sense I think you consent, to the extent that you consented at all, to that quality of the study. But then the question is, when the protocol comes down the road 10, 15, 50 years later, what mechanism is there in place for assuring that all those criteria are met? And I just think that, if you look at the current federal regulations, there is a lot of sort of ways you can have research get through with minimal approval that, frankly, I don't think any of us would want any of our staff or people under our supervision to be doing. So I think that is my level of concern. That if we can't, because of the way the consent was obtained or not obtained in the past, you know, really say the people wanted to do this and they kind of understood what was involved, and there might have been some concerns and objections, but they nevertheless went ahead, if you can't say that, then I think I want to look even more closely at the other mechanisms we have, as are these, you know, other lines that Zeke had in there with regard to the IRB review or, as we are going to talk about next time, the community consultation. DR. EMANUEL: I think it is helpful here to stick to the divide between the tissue to be used in an anonymous manner and an identifiable manner. Let us remember, on the identifiable side, I think our general intuition for everyone in the room has always been that full, informed consent, you know, even if you are using a previously-stored sample, has to be obtained from the people. So this issue of consent doesn't-- Our debate doesn't apply there because we agree you have to get consent for that. So we are really now into the tissue to be used in an anonymous manner. Now, within those boxes, it is important to think through that traditionally there have been only two types of protections. One protection is lumped into the IRB. You know, is this a valid study? Are they going to get useful information? Is the question not harmful? And then there is the second level of protection, even if all of that is true, "I have the right to consent or not consent to participate in that study." In the already-collected samples, the problem arises because that second level--the consent level--doesn't exist. And I should remind us--again, jumping ahead to this afternoon because they are not all that separable--even in the samples to be collected in the future, we recognize that you are not going to be able to have detailed informed consent. You are just not. Full informed consent is not a possibility because today we have samples that are, you know, 75 or 100 years old and, in the future, they are likely, you know, if I get surgery today, you know, the day after our sample that--who knows?--it may be 100 years before someone decides that Ezekiel Emanuel's, you know, pancreas or lungs or heart--or whatever it was--might be useful. DR. GREIDER: Brain. DR. EMANUEL: Brain. Yes, right. That they know is not useful. There is not a cell left. MR. HOLTZMAN: (Inaudible.) (Laughter.) DR. MURRAY: I want you to notice it is not even 8:30 in the morning, and they are insulting each other. DR. EMANUEL: You need a little more caffeine in that coffee. (Laughter.) DR. EMANUEL: So I think we have to operate under the assumption that consent just isn't a good--or isn't sufficient here--protection. Now, some of that is for practical reasons; some of it may actually turn out to be, you know, more comfortable for philosophical reason, separate from the practical concerns. And then I think you are right. Then we have to be reflective and look back on the other kinds of protections. The traditional protection of the IRB is one. And I think what we have been discussing for the last few months has been do we add a level, another level, of protection that doesn't even exist in the common rule now, which is this community? You know, granted it is vague, it is mushy, it falls between the fingers, but we think somehow it is very important. And I think in part we think it is important because it addresses really, at least to some extent, the concerns you have. If this research really has a potential to be stigmatizing and down-right harmful, then we are not even satisfied with the IRB. You know, they have-- We want to think about adding another level. But I think-- I do think, following up on a number of comments here, we really do need to get out of the consent box because we just can't satisfy it here, even if we wanted to. It is just not going to be possible in those tissues where it is to be used in an anonymous manner, without I think really grinding the whole system to a halt and harming things in a way we wouldn't want to. So I don't think consent is a-- We shouldn't rely on it as a safeguard at all. And I think, in some ways, it may be more an informational process than a real safeguard. DR. MURRAY: Bette? MS. KRAMER: I think Zeke has probably already said it-- DR. EMANUEL: Oh, I am sorry. MS. KRAMER: No. No, no. Not at all. Well done. DR. LO: Can I raise another question in terms of concerns one might have of doing research in stored samples? I need to defer to the scientists here. Is there any sense that you may run out of the sample? That if you only have, you know, a couple of tubes and people in 1997 are doing all these studies, by the time someone has the definitive DNA probe, in 2005, there may not be enough samples? And how do we factor that into this kind of analysis in terms of appropriateness of research as sort of the big question? So I know Carol would be the best person to ask. DR. GREIDER: With any given sample, of course, you would worry about running out of the sample. But I think the kinds of things we are talking about here when we have, you know, 100 million samples out there, most studies that are done, you know, you are going to use several thousand to, you know-- You are not even going to get to the 10,000 level. And so I don't think we are concerned about depleting all-- DR. LO: How about a more specialized sample like pedigrees of-- DR. MURRAY: Well, I think Jack has a good example. It may be useful to hear from him. DR. KILLEN: Yes. Jack Killen. I am from NIH, the Division of AIDS. I think there is a few that just spring immediately to my mind. I am speaking from the perspective of prospectively assembled collections done for research. Certainly in the case of HIV, there are millions-- We have millions of samples already stored away in a repository. But about 90 percent of our requests focus on about 1 percent of the samples. And, yes, running out of them is a very big issue. One of the things that we have started doing is generating immortalized B cell lines as a way of getting a reproducible supply of DNA from the individual into the future. The process of going through the decision-making about whether or not that was good maybe we can talk about later on in the community consultation part. Is that, Zeke, what you are referring to? DR. EMANUEL: Yes. An exact example, which I think is a helpful example. Running out of cells and then the decision to make them immortal so that you can, in the future-- And, I mean, I think it is inevitable that there are going to be some tissue samples, for whatever reason, that turn out to be very, very valuable, either because of the combination of clinical information or because there is something unique about this set of people separate from even doing it in an identifiable pedigree manner; just, you know, you have collected 10,000, and it happens that all the relevant diseases you are interested in, you know, turn out to fall into a small group. But fortunately there are techniques at least that may, you know--immortalization--raise certain questions. You know? Does anyone have the consent for that? In what manner might they consent? But it does somewhat obviate the issue of are we running out of tissue permanently? DR. GREIDER: But that is only for certain kinds of tissues though. DR. EMANUEL: Right. DR. GREIDER: We can't immortalize-- DR. EMANUEL: Cell blocks. DR. GREIDER: Right. Or certain tissue types, too. DR. EMANUEL: Right. DR. KILLEN: Or certain research questions. DR. MURRAY: Well, how are we doing on the overall structure up there? Do we agree? What are the key choice points? Let us see. Are we agreed that, for existing samples, we are not going to make-- We are not going to claim the distinction is important between those collected for the expressed purpose of a research proposal versus other purposes, if that were-- This is a very indirect way of saying it--I apologize--but we are going to say we are not going to observe that as a significant distinction? Is that right? (No response.) DR. MURRAY: We are going to treat them as effectively the same for our purposes. Is everybody comfortable with that? Do we have a good argument for that? Do you feel we have a good argument for that? Okay. We are also not going to pay attention to the specific terms of the-- Well, that is-- Let me put that a little differently. We will pay certain attention to the specific consent that may have been given or withheld if-- I mean, I think we need to state that, right? If someone has said, "I don't want my tissue used for research," in my view that should be a veto. That tissue is not used for research. Do we all agree on that? (Whereupon, there were several affirmations.) DR. MURRAY: If someone has said, "My tissue should not be used for research of this kind," do we all agree that that should hold? People should have the right to veto it. If there is a record of any kind of objection of that sort, that must be observed is my-- I think that is a very clear view I have. But, in the absence of such opposition to research, we are going to--in the past, again--we are going to assume that, you know, barring evidence of some malicious motive on the part of the gatherer of the samples, that the samples ought to be at least possible to be used for research in an anonymous manner. Do we also agree that any research that would include identifiers must--even samples collected in the past--must include prospective consent? Are we saying that? You must go back and get the individual's expressed consent to do it if we are going to use their samples in an identified manner. I can think of one case, one set of cases, where that might be a problem. DR. GREIDER: We really haven't discussed the identified stuff. I thought we were pretty much in the anonymous box so far. DR. MURRAY: Yes. DR. GREIDER: I mean, to back up a level, as Zeke had pointed out, it is somewhat different to say research, how the tissues are to be used as opposed to defining the tissues. So at the second level up there, to be used in anonymous manner and to be used in an identifiable manner. Right? Maybe we should agree-- DR. MURRAY: On the anonymous. DR. GREIDER: Maybe we should agree whether those categories--that that framework--is the one that we want to adopt and say it is how the research is done, not the actual tissues themselves. DR. MURRAY: Oh, we have accepted that. That is-- If I got that wrong, I apologize. It is how the tissues are to be used because we are going to assume that most of these tissues are-- They exist in some state in an identifiable way. That there are identifiers linked and we are going to have to create a recommended structure of a kind of stewart of the tissues who will then forward the tissues and other information, as appropriate, but stripped of identifiers so that you can't walk back and find out who the tissue came from. DR. GREIDER: Okay. So you are just assuming that whole discussion from the past? DR. MURRAY: Well-- DR. GREIDER: I just am trying to go through and-- You know, since we are sitting around the table here agreeing, do we agree to lump, yes or no? And we just said yes. DR. MURRAY: Right. DR. GREIDER: Then the next level is do we agree that this is how we are going to deal with the tissues? And then it is how the research is going to be done. I personally do agree with that category, but I don't know that we have sat around and had that agreement at the table. DR. MURRAY: That is a good point. Let us find out if we agree with it. DR. EMANUEL: I might say that, I think, is a very important reconceptualization to the way the debate has been held. If we remember back to the arguments between--I hate to be so crude--but Cord and Clayton and, you know, the American Society for, or the College of American Pathologists, the ELSI Working Group, et cetera, they focused on the sample nature, and we are re-doing it to say it is really not the sample we are interested in; it is how the research is going to be conducted. And I think that is an important break. We have discussed it a number of times, but I think it is, you know-- Because it is important, we should really be very self-conscience about that change. DR. GREIDER: And we should highlight the reasons in the report as to why we are considering reconceptualizing that. DR. MURRAY: Right. DR. GREIDER: Or why we did reconceptualize it. DR. MURRAY: Yes. And this is sufficiently important. It is worth making sure that everyone on the commission is fully comfortable that they understand the distinction and believes it is the right one to make. DR. LO: Let me-- DR. MURRAY: If this is a time when you have any uncertainties, you should speak up please. Bernie? DR. LO: I actually don't personally, but I guess since we are saying this is an important reframing of the issues, maybe we should just sort of think back and how would-- What would the objections be to this? How would someone like some of the people in the ELSI Working Group respond to this proposal? So if we can anticipate what some of the rebuttals and objections might be, that might be helpful. Because I agree. I think this is very important. I actually think it is a very useful step. I mean, maybe we could float this by Clayton, or some of the people in that group. DR. EMANUEL: I mean, I think part of what is going on here is the idea that we feel somewhat comfortable that you can, even though the tissue itself is labeled, you can-- The researcher will get it in an anonymized manner; that they can't walk backwards to identify the people; that the potential harms that are present therefore are obviated by that kind of protection; that the concerns one might have about identifying a community are taken up in a different way, not by focusing in on whether the sample has got a label or not. I mean, I think those are some of the rationales. At this hour, I am not sure I can reproduce all of that. MR. HOLTZMAN: Can I-- DR. MURRAY: Yes. Trish and Steve. MS. BACKLAR: I just want to say again, as you went through this list of people who would object, consent, and so on and so forth, I still think you have a tricky area in there at the group of people who may not be able to consent. And even though Zeke said there would have been a proxy, and so on and so forth, it could still be a little sticky. I think that you have to identify that group as you are going through; people who say no absolutely. You understand that you are not going to do it and-- You made a list of people who would consent or not consent and how you would deal with that. And I just want to make sure you keep the decisionally impaired in there in some way. DR. MURRAY: Trish, I think I will count on you, when you think that quite a different policy or set of rules ought to be enforced for people who are not decisionally competent at the time that the tissue was taken--which would include all children and would include all adults who are unable to give full informed consent-- to signal that; when you think it actually makes a difference in how we ought to treat those. I haven't heard that yet. I have heard you say be conscious of it. But if you see a point where you think it makes a substantive difference in the rules we ought to propose, please say so. And have I missed you? Have you indicated that already? MS. BACKLAR: I wanted to say that I was concerned that they were left out in your list because it may alter the rules-- DR. GREIDER: What I understood-- MS. BACKLAR: --in some way. DR. GREIDER: --Tom to say was, if there is something contrary already written down in paper that is collected, we are not going to ignore that. Is that not what you were saying? That there was already somebody going on the record in paper saying, "I don't want my tissue used for that." DR. MURRAY: Yes. And that is it. End of story. DR. GREIDER: And that would include decisionally impaired as well as-- MS. BACKLAR: That you wouldn't use their tissue. DR. GREIDER: That is right. That is whatever is already written down will be followed. But we are just making the assumption that a lot of things weren't written down, and so you might want to have additional protections. MS. BACKLAR: Right. And that-- DR. GREIDER: So I didn't hear Tom excluding the decisionally impaired in any way. I heard him say-- MS. BACKLAR: But they might fall into the group where something was not written down. DR. GREIDER: And so that is the best-- DR. MURRAY: Which is going to be overwhelmingly the case for-- DR. GREIDER: That is the-- DR. MIIKE: Wait-- DR. MURRAY: Larry? DR. MIIKE: I don't want the exceptions to the rule, our general rule. I think that, for previously collected samples for which there is no indication about not wanting to use the tissue for research, I don't know how we can go back and ask on an individual basis. On the looking forward side, clearly we are assuming that someone can give informed consent. If there are issues raised about their ability to give that, then that falls within that purview, so when we say that there will be informed consent, it doesn't mean it is an automatic process; it means that they can really give informed consent. So when we get into the discussion this afternoon, those are the kinds of areas that we will be looking toward your group to tell us whether the, you know-- I mean, you can sort of overlay your structure on top of ours in terms of the prospective type studies. DR. MURRAY: So let us-- Trish, keep raising the issue of decisionally impaired persons because we want to ask at each point does it make-- Will we want to sort of make a special provision or special rule or other treatment of such persons? At this point, as I understand it, we would treat all samples which have been legitimately obtained--and I am being purposefully vague about what it means to be legitimately obtained--in the past, previously collected samples, we are going to treat them all the same way. We are going to look to see if there is any opposition to research, in which case there will be no research on that sample. We will look to see if there is a specific opposition to specific kinds of research, in which case there will be none of that research on the sample. But otherwise we will treat them pretty much the same, I think. If you think that is inappropriate, we need to hear that and we need to hear why. It will be different I suspect as we go to the samples to be collected hereinafter, where consent will be, you know, we will be able to hopefully have a more informed, more robust kind of consent. Not perfect, but-- DR. MIIKE: Even in the second column, where we determine that previously collected, you know, but still identifiable, then this same issue comes up. DR. MURRAY: Yes. That is--thank you, Larry--that is absolutely right. Steve had his hand up, and then Rachel. MR. HOLTZMAN: I think in moving to the distinction, where we have moved of not the condition of the sample but rather the kind of research, as we are writing that, I think it is worth reflecting on whether, in fact, we have moved very far from where historically the reg was, because one thing that struck me when I read the Clayton paper is I thought they had taken the reg and changed it in their mind, intentionally moved from a standard which had been research conducted anonymously to the sample itself. So I don't think we need-- We shouldn't immediately assume that we are, in fact, changing the way this has traditionally been thought about, which is one comment. The second thing is, in focusing on research conducted in an anonymous manner, we should be clear on what we mean by that. For example, do we mean that the individual conducting the research can't hook up the research to the individual or that, say, the publication is not one from which one could discern the individual? And I think we should just be clear on what we mean by that. And I think we mean the former and therefore, by definition, the latter as well. DR. MURRAY: That is my understanding. Does everybody share that understanding; that, once the tissue and whatever information goes forward with it, that it would be practically impossible to walk back and find out who it came from, so the researcher would not have the information. MR. HOLTZMAN: So is that practically-- DR. GREIDER: That is my understanding, yes. MR. HOLTZMAN: Right. Because then that practically raises the sort of thing the theology group talked to us last time about which is that, if the pathologist is to be the researcher, then they need to go get someone else to be the stewart effectively. DR. MURRAY: Yes. DR. GREIDER: Yes. DR. LO: Is it what you intended also, Zeke? DR. EMANUEL: Yes. I think that is quite good. The researcher has to be blind. DR. MURRAY: Right. Now Rachel and then Bernie. DR. LEVINSON: Just very quickly, because we were talking about special groups, I just want to raise the question of the dead when you are talking about retrospective studies, not because those should necessarily be treated differently under the boxes that you are defining, but because they are currently treated I believe under the common rule very differently, and so you would be making a change to that that you should keep in mind as you think about the recommendations. DR. EMANUEL: I think actually that is important especially when we get to the identifiable because, if they are dead and we are going to use them in an identifiable manner and we require full informed consent, we have a problem. But that is a very-- I mean, that is an important category especially since probably the vast majority of our current samples actually do come from people who are currently deceased. DR. GREIDER: Can I just ask how are they currently treated differently if they are dead or not, under the common rule. DR. LEVINSON: Exempt. DR. MURRAY: Exempt from IRB review. Is that right? DR. LEVINSON: If they are dead? DR. GREIDER: Yes. DR. MURRAY: Anonymous or-- DR. LEVINSON: Not specified. DR. GREIDER: Really? DR. MURRAY: I think-- Yes. I think in all cases. DR. LO: If I could just follow on one of the points Steve made. I think that what I understand will emerge from our discussions is a rather sort of high tech foolproof way of making samples anonymous to the researcher. And I think we want to distinguish that between much more informal ways of "unlinking" samples which, in fact, are of varying protection. I mean, you know, if my close colleague in the office next door is the person who does the coding, I don't think that should count as anonymous for the purposes of research. And, as some of you were saying before the break, given that the technology now exists to really make it anonymous to the researcher, we should insist on a sort of fairly rigorous standard for doing that. DR. MURRAY: I can't help this. It is an irrelevancy, but I am going to take the prerogative here. When Rachel talked about, you know, working with tissue samples, et cetera, from the dead being exempt from IRB review, it reminded me of a colleague--a friend of mine--who years ago was the editor of the medical section of the Encyclopedia of Texas History. And he and a group of research assistants went out and interviewed eminent but very aged physicians in the State of Texas. They had a rule though on this encyclopedia. When it went to press, they would only publish biographical essays of people who were dead. So at that point I said, "Ah, perish, then publish." (Laughter.) DR. MURRAY: So sorry. Who is next? DR. MIIKE: What I was curious about was what Rachel had raised. Because clearly there still can be harm to family members, and there is the interest of the family members, so are we just going to assume that informed consent must be obtained and not get into the difficulties? I mean, you know, I am sure there is some protocol that says who you go to first and et cetera, et cetera. Because if we apply to people who are dead and their tissues are stored some people are going to be asking, "Well, how do we address that?" But do we want to explicitly address it in the report or we just-- DR. GREIDER: Which box are you talking about? DR. LEVINSON: Yes. What are you-- DR. MIIKE: Well, I am talking about identifiable dead person with explicit informed consent. And I assume there is a proxy. If we are going to insist on that, the assumption is there is potential harm or interest by family members or relatives. DR. GREIDER: I think we have to go through all the boxes. MS. BACKLAR: Right. DR. GREIDER: I mean, we have to start filling the upper right-- We haven't filled in really any of the boxes. It is hard for me to-- DR. MIIKE: Well, I went under the assumption that-- DR. GREIDER: The bottom right-- DR. MIIKE: Yes. I know, but-- DR. GREIDER: --box without having gone through all the-- DR. MIIKE: I know, but what Zeke proposed was-- I didn't see any objection to that, even in past stored tissue samples. If it is identifiable, we are going to try to get individual informed consent from it. I haven't heard anybody say no to that. DR. GREIDER: I don't think we have agreed to anything in any of those boxes. I certainly haven't-- DR. MURRAY: We are still-- DR. GREIDER: --agreed to what is actually in the boxes. MR. HOLTZMAN: But we could take Larry's generic suggestion. If, for any box in which we say an individual's consent is necessary, then, if we are dealing with the sample from a dead person, then that consent should be obtained from whoever is the relevant guardian, et cetera. DR. MIIKE: My second follow up to that is that you clearly have to make a reasonable attempt for-- If a reasonable-- What is the end product of trying to do a reasonable attempt and you are negative in it? Okay. Then do we have to-- They cannot do identifiable research? Do they have to do it anonymously? Or is it, you know, is it an absolute? DR. MURRAY: Right. Those are important questions, and I think we will need to come back to them pretty quickly. But I still want to get to if we have the-- I want to ascertain whether in fact we have full agreement on the elements of the framework. And one of the key elements we were just talking about was that we will view-- Whether or not a tissue is anonymous is, in our view, with respect to its use in research, not with respect to what may lie in some tissue bank somewhere, but in terms of what the researcher might see. And I take it that there is full agreement? It is not just agreement; that we actually have very good reasons which we will state for why this is the appropriate way to think about it. And I think Zeke is right. This is a change from the way it has typically been conceived, but I think it is actually-- It is the way it ought to be done. I think we have actually made an advance in coming to think of it this way. Now, there are cautions to be born in mind. There are ways of using fragments of information, particularly information that can be then sort of linked information, electronic databases, to do a certain amount of walking back, so we need to be very conscious about taking the technical issues seriously. We are not technical experts and I don't propose that we are going to-- I think it would be unwise for us to recommend a particular encryption scheme, or something, but to signal what we think the right principle is, to remind everyone, you know, that this can be a difficult thing to do properly, and that it ought to be done properly, and then to suggest perhaps some procedural mechanisms for how that might be looked to. MR. HOLTZMAN: So is this the place where we should flesh out a little more about anonymous such that, for example, we intended that there could be continuing epidemiological information flowing in one direction; that that does not compromise, in the relevant sense, anonymity, and that we left open whether, under any conditions, one ought to be able to go back in order to reveal to the subject results. DR. EMANUEL: I propose we go down the left side of the column first. MR. HOLTZMAN: Okay. I didn't know if that is built into the question of what is it to be conducted anonymously. That is why I am asking that question here. DR. EMANUEL: I don't think so yet. MR. HOLTZMAN: Okay. DR. EMANUEL: Not yet. DR. MURRAY: Questions we need to address. I agree with Zeke that-- DR. GREIDER: Why isn't that the upper left box? What Steve just raised. DR. EMANUEL: Wait. You mean this box? DR. GREIDER: Yes. DR. EMANUEL: It is this box, but I think-- DR. GREIDER: Oh. DR. EMANUEL: I suggest if we do these three things-- DR. GREIDER: Yes. DR. EMANUEL: --first because they actually turn out to be also controversial and a new addition certainly to the common rule. And I think if we have all the outside, while it won't be easy to go through the inside, at least we will be very focused. That would be my only suggestion. And, in part, because we already have had controversy from the full commission, at least on their additional gut reaction without, you know, our explaining the framework in any detail, or the rationale, to a three- level divide as opposed to just a two-level divide. DR. LO: Do you want to turn to whether we want the three levels; three rows rather than two? Okay. MR. HOLTZMAN: Let me just get clear why I asked that question there again. Do we mean by conducted anonymously; something can be conducted anonymously even if there is additional information about the sample over time flowing through? DR. GREIDER: Yes. MR. HOLTZMAN: We agree with that. So that does not compromise the concept we are trying to articulate here. DR. LO: We talked about this last night. DR. EMANUEL: I think this is the question. MR. HOLTZMAN: I am trying to stay in the outside boxes. DR. EMANUEL: Right, right. I understand. But I think the question is, again, one shouldn't-- We shouldn't focus in on somewhere does that information exist with an identifiable label. The question is, when it gets to the researcher and the researcher is doing it, is it in an anonymous manner such that it can't be walked backwards? If you can guarantee that, despite a constant flow of updated clinical information--the researcher doesn't know who it is, can't walk backwards except maybe with some safeguards which we can talk about, and the research is going to be done in an anonymous manner--that is what qualifies it as being done in an anonymous manner, not how the sample is, where the clinical record is, et cetera. MS. KRAMER: But at some point we are going to discuss the criteria that we want included for this encryption, without describing the exact method, right? We are going to address the question? DR. EMANUEL: I think inevitably, you know, and as I have heard it--and this is just my synthesis of our conversation--we are a divided subcommittee on it. We haven't really debated whether, you know, if you find some, serendipitously find some important information that is relevant to the health of the people, you should be able somehow to break that code. MS. KRAMER: Well, I don't know so much that we are divided as I think we haven't fully discussed it yet. DR. EMANUEL: Right. MS. KRAMER: Right. DR. EMANUEL: But I think I hear people's intuitions being on different sides. That is all I mean by divided. I agree. We just haven't had a thorough thrashing of that issue which would tell you whether that is going to be, you know, potentially permeable in the other direction or not. DR. LO: Well, I guess one procedural question is do we want to enter into that discussion now, or go back to the framework and try to see if the grid for the framework is-- Because I think it is something we are going to have to address. MS. KRAMER: Right. DR. LO: It is really do we do it now or later? DR. GREIDER: Well, I mean, it is coming up now. We are going to have to do it today, right? DR. LO: Do you want to do it now? DR. EMANUEL: I prefer to do it later. DR. GREIDER: You prefer to do it now? DR. EMANUEL: Later. DR. LO: Later. DR. EMANUEL: I think we need agreement either to collapse or to retain the three levels of-- DR. LO: Procedurally, how many people want to defer this until later and move on to the sort of grid as it stands? DR. GREIDER: Yes. DR. LO: Yes. DR. GREIDER: Move on. DR. LO: Move on. Okay. So we will come back to this later today. Steve, we will count on you to raise it because I think it is a terribly important question. With regard to this grid, I think the question that we need to look at is are we happy with the three rows, or do we want to collapse the bottom two into one? Zeke, do you want to-- DR. EMANUEL: Well, I think the other thing we need to be very, very careful about is that, in the current standard, the bottom two rows just don't exist. DR. GREIDER: Don't exist at all. DR. EMANUEL: So the first-level question is are we all comfortable with raising, or adding, or elaborating a row that recognizes community harms, potential community harms, or potential community implication? And then, if we recognize that there is potential community implication, and we are not just dealing with isolated individuals here but connected somehow with relevant characteristics, do we then feel that this divide, where some of the research even though it identifies a community, may not have any potential harm that we can think of or that it has a harm? DR. LEVINSON: How can ever say that there are no potential harms? DR. EMANUEL: Well, I mean, we have tried to think of some examples, and I will just give you the examples I have heard from the research community. You know, the ear lobe. That is yours, right? Carol's ear lobe example. You know, you are interested in ear lobe design, or structure, or eye color, or things that-- Or baldness. Things that might not have real harms. DR. LEVINSON: How do you know that the gene coding for the ear lobe is not going to be found later on to have some behavioral implications? DR. EMANUEL: But you wouldn't know that now. DR. LEVINSON: But that is-- DR. EMANUEL: And so, therefore-- DR. LEVINSON: But that is not a-- But it is still-- DR. EMANUEL: No. But then, even if you got community consent, you couldn't even talk about it to them. I mean, it wouldn't effect you if-- I mean, of course, down the line, some information, but that is not going to effect-- You know, do you go ahead with the research now? Because no one knows about that kind of information. I mean, that wouldn't-- That wouldn't be relevant to the consent, right, Rachel? DR. LEVINSON: No. Only to the extent that the anonymity might be effected. You know, whether someone would be concerned about what the implications of that study could be later on. DR. EMANUEL: But this-- DR. GREIDER: But this is anonymous. It can be anonymous. You can have anonymous research. DR. EMANUEL: Let us say you are interested in ear lobe design in Ashkenazi Jewish women. Okay? It is hard to imagine-- Forget future attachments. So you go to the community and say we are interested in ear lobe, the genetic of ear lobe attachment. Okay? And we are going to divide your community up and look for a gene that goes. All right? It is hard to think what the harm of that could possibly be. Okay. DR. LEVINSON: But-- DR. EMANUEL: Now wait a second. Five years. You have done the study. You have shown that it tracks with some gene. Five years later you find out that that gene is related to, you know, the high risk of heart disease, or something like that, or cancer, head and neck cancer, or something. When you went to the community, I mean, do you feel more comfortable because you got their formal sign-off, as opposed to whatever else we are going to require? I just don't see how it would make a difference. I mean, of course, all sorts of genes that we think are innocuous now might be related to something important, or potentially harmful. I mean, I think-- Remember why we are distinguishing--first of all, why we brought the community in--why we are trying to distinguish these two. We are trying to recognize that there is some category of research which might not, which might have implications for a community, qua community not individually seriatim, and we want to recognize that, especially with genetic research, that therefore the community should have some input as to how the research is done and whether it even goes forward or not. MS. BACKLAR: Wait a minute. Have we decided that? DR. EMANUEL: Well-- Okay. But we want some input, period. All right? And then we will leave it to the extent of the input. But I think the question here is, is it reasonable to imagine that, even if it implicates a community, there are some things which aren't going to-- I mean, what are the harms we are worried about? We are worried about some stigmatization and some discrimination. I mean, suspect categories. Everything is a suspect category. MS. BACKLAR: I think that Rachel has a very important point. Why are you distinguishing between no potential harm and potential? Why do you have to distinguish? Because, in fact, you don't know what it might be. It simply might be that you are identifying a certain group with a certain shape of their ear and ultimately people say, "Well, that; I don't like that group and it is because of their ear." I mean, their ear shape, or whatever. DR. GREIDER: But you can't provide for things in the future that you don't know anything about. Right? You can only-- MS. BACKLAR: But my question to-- DR. GREIDER: You can only protect against what we know about. MS. BACKLAR: But my question to you is why must you distinguish between potential harm and no potential harm? DR. LEVINSON: It is too nebulous. It is too subjective. DR. GREIDER: Well, you are going to have to distinguish. The IRB will have to distinguish it at some point, right? DR. MIIKE: Yes. But, you know, you folks are making-- If you are going to argue that there is no distinction, and I would go along with that, my question would be, would come down on the opposite of where you are, where you would want to have rigorous protections. And I would argue, if you are going to combine the groups, then my problem is with dealing with communities. What the hell are we talking about when we are saying what is a community? I mean, you can talk about the Ashkenazi Jewish women as one good example, but if they don't agree in Boston but agree in New York, or in San Francisco, then what is the utility at? But if you are talking about a very localized group of Alaskan natives in a little village, to me that is a community definition that you would want to be very careful about protecting. So I would prefer to go with the separate of harm/no harm, and I assume that there has to be some group like an IRB looking at it to say whether there is a harm or not, rather than combining both, because if you combine both then I am going to go the opposite way of where you are going to go. MS. BACKLAR: And also my understanding that you are thinking of this community as different from the Canadian collectivities, which could be families, or can these communities be families? DR. MIIKE: Well, we-- DR. GREIDER: We haven't said that. DR. MIIKE: --haven't really said that. MS. BACKLAR: But that is why I am asking. DR. MIIKE: Well, no. I think if you are talking about blood relatives, families, no. No. That is more on the individual side to me. DR. EMANUEL: But wait a second. If you are getting down to families, you are probably getting down to pedigrees which means you are going to be on the identifiable side. MS. BACKLAR: Right. DR. EMANUEL: Let us try to keep the boxes clear. I mean, when you get to a small unit such as a family, and you are going to be doing research on the relationship of the family, I mean this may not be completely-- I see that Steve is puzzled. But I think that you are probably going to end up on the identifiable side. I think, again, it is important for us to try to keep some paradigm cases intact. Now, you may be right that the intuition is no matter what it is, if it implicates a group, any group, it is automatically a suspect category. I personally don't like that idea. I think that is a very bad standard to take. I mean, we do have some suspect, some groupings which, you know, where-- And harms that we are seriously worried about. Harms that could lead to, you know, some form of discrimination or, usually for historical reasons and reasons of social marginalization, stigmatization. But that doesn't include, you know, every group. And I would remind you that one of the papers I handed out last time, or two times ago--I can't remember any more--was about a study they did out of the Physician's Health Study that identified African-Americans and whites where it turned out that the whites were in a much higher risk category. And I had suggested that that ought to fall into community, no potential harms, because that-- You know, you don't usually harm all of such a big group, I mean, of the dominant group. That is just not the way it is usually thought about. Discriminating against all whites is a very difficult thing to do. DR. LO: But the other way, if the study had come up the other way, one could argue that, from the African-American perspective, and they said, well, the prospect of discriminating against the whole class is real-- DR. EMANUEL: Yes. DR. LO: --had the results gone the other way. So you are going to have to take it when the research was planned, not when the results come out. DR. EMANUEL: Right. DR. LO: So, I mean, like any ethnic division is possibly suspect because it could show increased susceptibility among the class, which is already disadvantaged socially and, therefore, adding to whatever burdens and discriminations you have, so at this state it may be a suspect category on the face of it. MS. KRAMER: The problem I am having, in trying to deal with the decision you are asking us to make now, is that I have great uncertainty as to how we ought to deal with community altogether, and I know that is on the agenda for this afternoon, but I personally am going to have trouble making this decision until we have talked about that. DR. MURRAY: I am going to ask my fellow commissioners for an act of faith here, which is difficult I know. But it is my faith in Bernie Lo actually, which I don't confuse with any deity, although I think-- (Laughter.) DR. MURRAY: That Bernie is going to offer us some constructive ideas about how it is that, at least in certain circumstances, one can think about community and get community input into decisions about the appropriateness, design, et cetera, of research. So let us just-- And if I am wrong, I will tell you. I will be honest with you. DR. LO: You are wrong. DR. MURRAY: I am wrong? Okay. Am I wrong, Bernie? I am? DR. EMANUEL: But I think we-- DR. MURRAY: Seriously, do you think community consultation is-- DR. LO: I think-- I think-- Well, I think these are very, very tough issues. And I think you are starting to raise some of the complexities. I think what we can do is sort of help begin to sort out. I think, out of whatever discussion we are going to have after the break, we are not going to reach conclusions, but I think we are going to be able to be more aware of what some of the dimensions are, both the possibilities and the pitfalls. DR. GREIDER: Well, why don't we have that discussion before we decide whether there is one category or two? It seems like we can't make that decision until we have discussed the whole community. MS. KRAMER: That is what I am suggesting. DR. MURRAY: I suspect that is-- DR. MIIKE: What I was going to say was that my problem is not with that three-line group with community harm/no harm. My problem is with getting informed consent from communities. And I think that what we have been seeing--and I think the kinds of things that Bernie has raised--is consultation with communities, wherever you define, is good because it helps to sharpen the focus and make the research project better. I have no problems with consultation. I have problems with getting an informed consent out of a group. That is my problem. DR. EMANUEL: Can I-- I want to raise two points. One is there are huge problems with community, but I want to raise these two points. One, we are not the first to tread into that pond. Okay? The FDA is already plopped a big stone into that pond and I think, as we have recognized over time, you know, it is an area which we have ignored for 15 or 20 years. That doesn't mean we should continue to ignore it just because it is hard. Second, I want to-- I think we need to be very clear about distinguishing two things here. One is whether we think that categorization is accurate, and the second level is what kind of protections that entitles you to. And I don't view-- I mean, we may want to end up saying, you know, we want to recognize this category. We are not sure of the kind of protections, or here are the kind of protections for well-defined communities. This is a concept which is undergoing debate and interpretation now. And our notions of what the correct protections may be may need to change over time, as the debate gets clearer. We are going to get a lot more experience from the FDA rule. We are going to get a lot more experience in other areas. And so I think, you know, we need to make this a two-step process. One is does that divide match with some ethical intuitions, and the second question is what are the regulations that go with each of those boxes? Those are two separate questions in my opinion. DR. GREIDER: Can you remind me what the FDA stone is? DR. EMANUEL: Oh. In the emergency exception to informed consent. So you can do a study without the informed consent of the person participating in the study in the emergency room context, if you can't get informed consent because it would delay or harm them; you don't know who to get it from, et cetera, et cetera. Before you can go ahead with that protocol, you need to get what they call community consent. And they are very vague on what that actually is, what would quality. And it is a thing which, to some extent, they have punted to the local IRBs. But they recognize that if you are going to be treating people and you can't get their informed consent, you want another level of protection. I think, in some sense, though not articulated exactly as we have, they are coming to the same kind of conclusion from a different way than we are. MS. KRAMER: See, given that-- DR. EMANUEL: There are lots of places that are doing it now, but I will tell you what I think is going on in, you know, Boston. All of the emergency vehicles from one area coming from Brookline go to the Beth Israel Hospital, so if you are going to do a protocol in the Beth Israel Hospital, you go to the Brookline community. You tell them the kind of protocol you are going to do; that you are going to do it on everyone in the following circumstance. That is a method that you might approach. I mean-- MS. KRAMER: But that is a perfect example. How does one go to the Brookline community? DR. EMANUEL: Well, I mean, you have got mailings to all the people in the, you know, geography. Right? You might have public forums. MS. BACKLAR: Advertisements. OHSU is advertising everywhere about this. Little boxes in the newspaper describing what is going on. DR. MURRAY: We should just state OHSU-- MS. BACKLAR: Oregon Health Sciences University. DR. MURRAY: Yes. MS. BACKLAR: Sorry. MR. HOLTZMAN: So you are not seeking the consent of the community; you are rather letting them know that a certain practice will be taking place and they should be aware of it? MS. KRAMER: So it is informational? DR. EMANUEL: Right. But, I mean-- MR. HOLTZMAN: But they have the opportunity to object. DR. EMANUEL: Yes. I don't know, you know, I think this is so new people aren't quite sure what happens if the community gets up in arms. "We don't want you doing that with, you know, our people who are coming." I mean, these tend to be dynamic processes. They don't tend to be, you know, all we are doing is shoving it out there. MR. HOLTZMAN: So, if I could come back to what we mean by community without getting philosophical, just what we meant here, and explain my puzzlement. Under the current rule anonymous, or anonymized, or whatever, refers explicitly and only to the individual, so before we even get into lines 2 and 3, there is the question are we going to introduce another line or lines? And that is that we believe it is a relevant consideration to ask, with respect to a piece of research which is conducted in an anonymized manner with respect to the individual, of whether that research is nevertheless identifiable with respect to a community, and that we think that that is a relevant question that needs to be asked and answered. I think that is the fundamental first thing we are saying, which really does raise the question immediately did you mean a community as constituted by some social definition or did you mean it is a community in the sense that it is research which is not identifiable with respect to an individual but it is identifiable with respect to any others, in which case you would then get into collectivities, families, et cetera. I must admit I always thought that the primary divide we were making was along the latter lines; that is, that while not individually identifiable, nevertheless it is identifiable with respect to others or additional people, as opposed to definition of community. DR. MIIKE: Yes. But I always got the notion that, okay, if we are doing studies such as this and it-- Well, let us take the case of breast cancer. Obviously it would not apply to the male members. Right? I mean, the issue was the women altogether in that ethnic grouping. So it didn't seem to me that we were talking about-- I guess what we are talking about is that you have individual research in an anonymous manner where the individual is not identifiable, but the research is conducted such that it consciously looks at a particular grouping. DR. EMANUEL: But- DR. MIIKE: It may not be a particular family. DR. EMANUEL: But, I mean, I think it is important to-- You know, one is you could have a sort of historical traditional grouping like the Native Americans. You might have a geography, you know, the Mayo Clinic area, Olmstead County. You might have ethnic groupings. You might have racial groupings. You might have disease groupings; the AIDS community we sometimes talk about. And then you might have families. I mean, there are sort of six kinds, and this is just off the top of my head. I haven't thought it through completely. Now, I think the issue is, you know, not you do research and it shows up because you have these sociodemographics that tracks with, say, Jews, or it tracks with some racial grouping. The issue is you are going to that, to a particular grouping for a purpose. I mean, your intention is to identify it within this grouping, either ethnic, racial--it might be geography--for all sorts of reasons. You know, you are trying to highlight environmental issues possibly there. It may be a convenience sample that might have geographic implications, you know, implications for people living in that community. So I think we need to be open. I mean that is why, again, in the sample where I handed out the papers, the question of, you know, whether doing the study about breast implants in Olmstead County might not qualify here as the community. While they, you know, may not have any geographic or racial, you know, you might find out something about Olmstead County residents. They have breast implants at a much higher rate than anyone else, or a lower rate, or something. DR. MURRAY: It might be-- I am tempted to do two different things, and I guess I will do them both quickly. It might be worth asking what problem our sort of concern with community consultation was meant to address. And I will just state how I see it. Namely, that there are certain circumstances under which one can imagine that, even if my sample had been rendered anonymous for the purpose of research so no one would know it was me, but nonetheless there might be information about some group or groups to which I see myself belonging to, and which others perceive me as belonging to, that I might find either potentially harmful to that group or, in some way, offensive to that group, even if it didn't result in harm. We would just object to it. We might object to it for religious reasons or other kinds of reasons about our views about tissue, or we might object to it just because we think those are the wrong kinds of questions for scientists to ask and, in fact, most of the people that are in the group I belong to seem to feel the same way. That is the problem I took it we are solving. Do we agree at least that is the problem we thought this was addressing? DR. EMANUEL: Yes. DR. MURRAY: Okay. Now one answer I guess is to say, well, there is no good way to solve the problem so we will just shove it aside. That is one solution. That is not one that I am prepared at this point to embrace. I would rather see if there is a way where we can do honor to these concerns about offense and about harm. And that is what the community consultation idea is an effort to address. That is the one thing I want to do. And we have a whole section of the program devoted to that. Bernie? DR. LO: Go ahead. DR. MURRAY: The other thing is just I want to know if we have sort of reached the point where we have at least agreed on the structure--the framework as he calls it--where we can move on. Maybe what we should do, if we have reached sufficient agreement on that, we can move on to the question of community consent a little ahead of the schedule, and then come back to the structure and see whether or not we want to have this distinction between harm and no harm. DR. GREIDER: So agree on the structure, but don't agree on whether there is four boxes there or six basically? DR. MURRAY: Yes. DR. GREIDER: There will either be four or six. DR. MURRAY: Yes. DR. GREIDER: So we have agreed on the top part of the structure. DR. MURRAY: Yes. DR. GREIDER: But not the-- DR. MURRAY: There is one distinction there that we haven't-- We haven't-- We haven't decided whether we are ready to embrace. DR. GREIDER: Exactly. DR. MURRAY: Is that an adequate perception of where we are? Okay. Let us-- We have a break scheduled in 20 minutes. Are you-- Do people feel the need for a quick break now, and we can pick up community-- I see yeses. All right. Let us take a really-- We are going to have Carol's comment and then we are going to take a really brief break and then come back. Carol? DR. GREIDER: Can I just make a plea because we are going to be discussing this again. Can we number those boxes--can we go one, two, three, four, five, six--with my pen so that we can discuss the boxes. DR. MURRAY: Only if we do it randomly. DR. GREIDER: No. I want to-- DR. LO: Do it one, two-- DR. MURRAY: No. I am going to make a suggestion as to how to do that. Okay. We are going to take a brief break. Five minutes. See you back here. Carol? (Whereupon, at 9:21 a.m., there was a brief recess.) DR. MURRAY: Elisa Eisman(?) was good enough to distribute a reprint of an article about stored Guthrie cards, DNA banks, for the commissioners. Thank you, Elisa. We are going to talk about the idea of community consultation/consent right now. And in less than a minute I am going to turn it over to Bernie Lo who will chair this part of our meeting today. I want to mention that the issue of community consultation and consent is-- Not only is it not unique to the subcommittee and the FDA, it is not unique to the subcommittee and the commission. I mean, there is a paper on community involvement in research that--that is in draft now, I gather--that the human subjects research half of the commission is working with, and I have been assured that we can have at least a draft of that paper in advance of our next meeting in January. So it is important here to let the-- I don't want to characterize one of us as the right hand and one of us as the left hand, but let the other hand know, each hand know what the other hand is doing on the commission. So it is, as Zeke pointed out, it is not unique to our problem; the concern about community involvement in research. Thank you very much. Bernie?