DISCUSSION OF TISSUE SAMPLES REPORT KATHI E. HANNA, Ph.D., AND SUBCOMMITTEE MEMBERS DR. MURRAY: We are going to talk about the Tissue Sample Report, and we have-- Commissioners should have a draft of sections of the report. They should have had it for a week or two now. No? One week or so now. Kathi Hanna has been working hard on it, including over the holidays, and I want to thank Kathi on behalf of all of us for what you have done. But we should jump into the report. Now, Kathi, we want to go over each of the sections, I think, both the ones of which we have a draft, and the ones that we just have still an outline. I am going to ask Kathi in a moment if she has any specific needs that she would like us to address. I have two things I want to mention. The first is I have a series of specific questions that I think we probably haven't talked enough about, even as a subcommittee, to know precisely what we want to say in the report, and I want us to get to them. You may have other candidates. The second thing is, and I will try to resist the temptation myself even as I remind my fellow commissioners to resist it, this is not the time for copy editing. If you have large comments about organizational scope, yes; about sections that need to be in there, yes; but this is not the time to correct spelling or the precise words. Write it down, give it to Kathi, and she and I will make sure it is taken care of. Is everyone in agreement with that? Now, the temptation is very great because that is something we can fix on and do, but it is-- I think we are better off using our time to talk about the larger issues. Kathi, what would you like to see us do? DR. HANNA: Well, I think it is pretty obvious which sections need to be discussed. I think that the overview is obviously just my first attempt to try and forecast what issues are going to be covered in the report, so anything that is missing from that section I would appreciate your input on. I think that the second chapter, which is really just a slightly condensed version of Elisa Eisman's report, there I think you just need to make decisions about how comprehensive you want to be. I think it is still too long, but I left a lot of the examples in there so that there was at least information for people who were looking at this for the first time. For the third chapter, the moral and religious perspectives, I think we have a lot of information to work with from Courtney Campbell. We still are waiting to hear whether we have someone lined up to write a commissioned paper on some of the other issues that are not necessarily, that don't necessarily have a religious orientation. So that really leaves-- I think where we need to do the most work in is Chapters IV and actually VI, since Chapter V will be mostly descriptive and that is going to be fairly straightforward. Other than that I think I just, you know, this is very patchy draft at this point and I think we need to focus. I can work from the transcripts and previous discussions on Chapter IV, but I think Chapter VI, where you really have to operationalize your recommendations, is where we need to do the most work. DR. MURRAY: Okay. Trish? MS. BACKLAR: I just would like to say that, in Chapter I, that we shouldn't forget that we really need to look in some way at the issue of minimal risk as we are also looking at the Human Subjects Subcommittee. And you may not want to do that now because-- (Technical difficulties.) DR. MURRAY: I know that has come up in some of the discussions about tissue samples, including the National Action Plan and previous discussions. MS. BACKLAR: And we are--many of us--really considering the detail-- (Technical difficulties.) DR. MURRAY: Yes. MS. BACKLAR: So we might share some of that together. DR. MURRAY: Yes. Yes. Yes. DR. GREIDER: I have a couple of comments, and the first is on the outline. And I apologize that I wasn't there for all of the discussion when we discussed the orders and what the actual tactics were going to be. But it seemed to me, looking over this proposed outline, that the public knowledge and beliefs is relegated to an appendix rather than a chapter, and I am just wondering if we would consider actually having that be part and parcel of the whole thing rather than putting that in an appendix at the back. And I am not sure what kind of discussion occurred since I wasn't here at the end of the meeting when we discussed the outline why it is an appendix. And my recommendation would be to have it be a chapter following the science Chapter II. So something along the lines of, you know, what are the public views on this, earlier rather than as an appendix. DR. MURRAY: This is the time to talk about the overall organization of the report. I think that is a good way to jump in. We had a discussion about that organization at the end of the last meeting, but that was not having a draft in front of us and, you know, your ideas get more concrete the more you have to look at. Larry? DR. MIIKE: A couple of comments. The framework Chapter IV, if we are going to be discussing the issues around which we then reach our conclusions in Chapter VI, then that is okay, but the way I, the way I glanced at Chapter IV, it seemed to mix both. So you are either going to have to combine IV and VI and make it follow V, or you are going to have to have a discussion of the framework and then the policy recommendations coming later. But in either event, I think just in sequential things, the currently-proposed policy should come before our framework because that sort of bores out there right now, and then we impose our own framework on top of that. The other problem is that I don't what to do with the religious chapter because I don't see the moral piece. And if I don't see the moral piece I don't know how useful the religious chapter is going to be in balancing that off. If we keep the religious chapter there, then I would take out the conclusion section and more or less say that the religious discussion leads to the same kinds of conclusions that we reach in the--for lack of a better word--the lay approach that the rest of us think, which is that it gives the same kinds of conclusions that we would have reached regardless of a religious perspective. Do you understand? When somebody talks about confidentiality, community, et cetera, et cetera, and those are not things that necessarily-- Particularly from the religious standpoint, I think it is something that we have all discussed. So I think that perhaps that we should say that, even when you come from this religious perspective, we sort of arrive at the same point, regardless of whether we are coming from a religious perspective, or from a scientific perspective, or a social perspective, or whatever. DR. MURRAY: Kathi, you had something to say before. Did you want to say it now? DR. HANNA: Yes. I just wanted to respond to Carol's comment about moving the section on public knowledge to an appendix. I think we are still not quite sure what to do with that section, and part of it is because I think there is some discomfort about the reliability and the validity of the mini-hearings' approach as a good gauge, other than just one indicator. And so I think what Eric and I have talked about doing is trying to find some other opinion polls, surveys, systematic types of measures, that can then be viewed as complementary to the piece that is being done, the mini-hearings. So I think the-- The other thing is that there are some interesting things that have come out of the mini-hearings that I think we should try and incorporate throughout the report as they arise, and not just segregate public opinion to its own section, but really try and, if it is relevant, refer to it in the chapter where it is appropriate. DR. MURRAY: Bette? MS. KRAMER: Excuse me. I understand the concern about the-- (Technical difficulties.) MS. KRAMER: --and I know it is important to put it into the body of the report as opposed to an appendix because if you think back to our concerns that led to the mini-hearings it was the fact that, even though all of the-- (Technical difficulties.) MS. KRAMER: So I think that if you put it in the context of recognizing that this is not a full-blown scientific poll, such as-- (Technical difficulties.) MS. KRAMER: --but put in a context of our attempt to get some feedback from the public. And then I don't know legally-- (Technical difficulties.) MS. KRAMER: And I hate to see it regulated to an appendix because I think it indicates a lack of concern of the public-- (Technical difficulties.) DR. MURRAY: Carol? DR. GREIDER: I agree with you, Bette. That was sort of why I initially brought this up. I also wanted to respond to Kathi that I think it is also true to incorporate as much of it as we can into other chapters. I very much agree with that. I was just responding to the fact that it seemed to me to be a relatively important thing to many of the commissioners that we get this information rather than sort of operating in a vacuum, and I didn't want that issue to be an afterthought the way the report came out. DR. MURRAY: Zeke? DR. EMANUEL: Two things. (Technical difficulties.) DR. EMANUEL: --and I suggested, one of the reasons I asked Janet Wells(?) to come up with those questions was really incorporate it-- (Technical difficulties.) DR. EMANUEL: --sometime in the next few months, but not-- (Technical difficulties.) DR. EMANUEL: Having those questions and maybe even including them during-- (Technical difficulties.) DR. EMANUEL: My own suggestion is that we still have a lot of boxology-- (Technical difficulties.) DR. EMANUEL: And my only suggestion is that, once we resolve that, at least we can more or less decide what the framework is. And at least on eye level there was some disagreement and uncertainty about the boxes, and I apologize for the boxes. (Technical difficulties.) DR. EMANUEL: But we had talked about whether, on previous samples, we were going to combine research and clinical care and we had talked about how we-- (Technical difficulties.) DR. EMANUEL: I mean, I think that those are the most important issues for us-- (Technical difficulties.) DR. MURRAY: Yes. I think, particularly with Zeke here, we ought to take what time we have with him to try to look at the boxes, but I want to recognize Bernie and David. DR. LO: David and I are-- (Technical difficulties.) DR. LO: --little bit while I was swimming this morning, so this may be all wet. (Laughter.) (Technical difficulties.) DR. LO: But it struck us, as we sort of stepped back from where we are around the-- We were concerned we may have lost sight of-- (Technical difficulties.) DR. LO: I didn't really have a clear picture until I spoke with David about-- (Technical difficulties.) DR. LO: One has to do with, as I said, just sort of what makes genetics DNA research-- (Technical difficulties.) DR. LO: And, on the one hand, I think I would ask that we stress this firewall that we started to talk about and the way that we address-- (Technical difficulties.) DR. LO: One direction is, as the researcher discovers things that are going to have potential clinical significance-- (Technical difficulties.) DR. LO: So I think as long as we have a possibility of-- (Technical difficulties.) DR. LO: So I think how we handle that is important when we are thinking about it. I think it also may fall under pressure in the opposite direction, and this I-- (Technical difficulties.) DR. LO: And David came up with a model of using a large-- (Technical difficulties.) DR. LO: The problem I see come up, when I signed up for that study, I said, you can only use my tissue for the disease of interest and-- (Technical difficulties.) DR. LO: And one of the things I would like to see is this notion that along-- (Technical difficulties.) DR. LO: I think we should sort of try to think about all those and anticipate those sorts of problems. (Technical difficulties.) DR. COX: Yes. Well, I run a risk of saying anything-- DR. MURRAY: David, would you bring your microphone closer? Thank you. DR. COX: Because in the past, when I have tried to articulate these things, I have been totally incomprehensible, so I will try yet one more time, but if I am incomprehensible yet again, please tell me. The-- Bernie has helped me with some of the words and the concepts and I guess one of them has to do with that there is several different processes of doing research, different study designs, and to focus on what that study design is--cross-cutting what the boxes are-- gives a more whole picture to me. And so it really has to do with not sort of what most of the samples are that are in existence now, but what are going to be the use of the samples in the future? It is an important issue to deal with the retrospective studies--don't get me wrong--but I think that, in large part, our job as a commission is to think of where we stand now but, more importantly, where we are going in the future. So in the second chapter, which I guess is the science chapter, because Carol and I are involved with that, and I think it is good to document what samples are there, but then, just like happened in cloning report, go through what the scientific process is, what some of the study design would be that could lay out what the structure for the future is going to be. That is not sort of what the ethical boxology is, but it is saying practically how the research is going to be carried out. Then that makes for a more whole discussion of-- It really boils down, in my mind, to this relationship of the subjects to the researchers, and that relationship is different in different settings, depending on how you do the design. My personal belief is that, in genetics, the relationship between the researchers and the subjects is like tight. It always will be. All right? And that is very different from what is going on in epidemiology in general. So why is it tight in genetics? And it is tight because of this stratification; taking big groups of people and winnowing them down to a narrower stratified subset of which you collect more and more information. And how you can have that kind of a process, where you divorce--in the long run, to come up with a treatment--where you divorce the researcher from the patient, I don't understand. So my specific suggestion is, is that Carol and I--and others who want to--but in specifics, Carol and I work on the second chapter to include that kind of a process of the kinds of research that may be going on, different types of research, and what is involved with that in terms of these relationships, because I think it complements the structure of the boxology. And then--I quite agree with what Larry said--that then putting those two together we come up with the recommendations on it. DR. EMANUEL: I am a little confused. I need some context-- (Technical difficulties.) DR. EMANUEL: And I think this is a situation where we are talking about it. You want to go back to the Physician's Health Study or to the Framingham Health Study and do genetics on samples that were not initially collected for genetic tests. (Technical difficulties.) DR. EMANUEL: There is no relationship between the researcher there and the-- (Technical difficulties.) DR. EMANUEL: No relationship. Even if you did accomplish-- (Technical difficulties.) DR. COX: Okay. I understand, Zeke. DR. EMANUEL: So the issue that Bernie suggested, you know, now once-- (Technical difficulties.) DR. EMANUEL: You don't know who these 10,000 are. They are just numbers. DR. COX: That is correct. So let us back up for a second and say, ultimately, what is the goal of genetics research? It is not to find a gene that predicts something. This is my personal view. It is not to find a gene that can predict that you are going to die when you are, you know, 43. Okay? It is to come up with treatments to improve people's health. So that what is the process by which, in my view of the future, that this is going to happen? It is going to start with very large populations like the Framingham or NHANES. The initial part of it is going to be finding associations of genes that do make predictions. All right? Not all DNA information is going to be highly predictive, but a subset of it will be very predictive. But that is just the beginning because, when you get those predictions, then you have individuals who you can make predictions about but there is no therapies or any options. How then is science going to proceed to come up with any therapies or options? It is going to be to enlist exactly that subset of people to do clinical trials to figure out what works and what doesn't work. That is where the relationship comes in. DR. EMANUEL: (Inaudible.) DR. COX: So it is not stopping when you find the association. (Simultaneous discussion.) DR. EMANUEL: Right. It is also a completely different type of research protocol that is separate from going back to the stored tissues. Right? I mean, that is-- (Technical difficulties.) DR. EMANUEL: Right? I agree with you. Then what you do is you go back, find the people who got that genetic alteration and-- (Technical difficulties.) DR. COX: But see, now we are going to-- Follow me one more step now. So now we are with those individuals and that one set of researchers is getting more and more information about then and are involved with clinical trials. Now, another group of researchers, because these happen to be heart doctors because they are working on ApoE, but now another group of researchers says, "Guess what? Your ApoE is important for Alzheimer's disease, not just heart disease, and so we want to enlist you in this." All right? So that what it means is that if, when patients get involved with us to begin with, they are part of an overall process that they may or may not want to be part of, but you can't inform them about it from the very get-go, so-- (Technical difficulties.) DR. EMANUEL: --this kind of example because the question is, is it covered in a way we feel comfortable with or is it not covered in a way that you-- (Technical difficulties.) DR. MURRAY: Yes. DR. EMANUEL: Is that what we are here for? DR. MURRAY: Yes. You are here until 2:30 p.m.? DR. EMANUEL: Yes. DR. MURRAY: Okay. DR. EMANUEL: I apologize. DR. MURRAY: Well, I think we should use you while you are here as extensively as we can, so why don't you do that? DR. EMANUEL: So let us say we have a stored sample, like the Framingham Health Study, so it is a previously collected sample. Right? And we are not going to-- (Technical difficulties.) DR. EMANUEL: Right? So I am not sure why that isn't on the boxology 1a. Is that 1a? DR. COX: Well, you tell me. (Technical difficulties.) DR. EMANUEL: Okay? Now say you have found that and in your first study, you know, you notice, in the Framingham, it is associated with the Ashkenazi Jewish population, and you go to another Ashkenazi Jewish cohort that has-- (Technical difficulties.) DR. COX: I don't want to go to a different cohort though Zeke. I want to take the people that I have begun to find in Framingham because there is not so many of them there, and it costs me a lot of money to find them, and I want to do more stuff with them. DR. EMANUEL: Okay. So I think the issue is what does "more stuff" mean? DR. COX: More stuff means collecting-- (Technical difficulties.) DR. EMANUEL: Collecting more clinical information, if it is done in an anonymous manner with the firewall, is perfectly fine--as I understood our agreement last time--is perfectly fine in box 1a. We want continuous information--b--as long as it is in an anonymous manner, and you can't identify that particular person. Now-- DR. MURRAY: Instrumentally, as I understand it, to see if I am following, what this would entail would be the researcher, you, now wanting additional information and more samples, going back to the stewart of the samples. DR. EMANUEL: And who is the stewart of the sample? DR. MURRAY: Presumably it might be a pathology department. And saying, you know, these were very fruitful samples. I would appreciate additional sample material and additional clinical information, but it can also be-- It could still remain anonymous. Is this what you are contemplating? DR. COX: Well-- DR. EMANUEL: But imagine two different circumstances. Imagine you have a research study, again like the Physician's Health Study or NHANES. You are-- That data is being dumped into this anonymous pool. That is the way you have created the situation. So you have got the data up until that point. Say two years later they do another survey of these people--right?--to find out, you know, what diseases have happened in the intervening two years, the way they do in the Physician's Health Study. That data, as it is entered, gets dumped in and there is an update. DR. COX: Yes. But what if it is not the data that I want, Zeke? DR. EMANUEL: What if it is-- DR. COX: Not the information that I want. DR. EMANUEL: So you-- So now the question is, you want to walk backwards, identify those particular 12 people-- DR. COX: This is how genetics is done. So that you go and you stratify populations and you intensively investigate a smaller subset. DR. EMANUEL: I think if you then want to be able to use it in an identifiable manner, you are going to have to get their consent. I mean, that is what this says. Then you move to 1b. And then you would have to get their consent. DR. LO: Okay. But that is-- I think that is important that you can't sort of say the trustee goes back to the patient and negotiates and then anonymizes it so that, when I get it as a researcher, it is anonymous. DR. EMANUEL: Well, I think we are going-- I think we are-- I think by not having a good example, we are sloshing a little back and forth. In my mind--and I am only one member here--the trustee situation is a-- That operates really in the clinical where you have got the sample from the clinical context. Remember, in the Physician's Health Study, you have got freezers full, you have got it computerized, you have got a database. All right? There you have got an organization already. In the sort of clinical setting where you have gone and you want, you know, like the Angiogenesis Factor Study from the Brigham, you want samples of breast cancer with, you know, five to 10 years of follow up, and they are the trustee who pulls them out, who knows which ones they have pulled, and has gone to the clinical record and added that information to you. That-- There is a trustee. In a research setting, there is not a single person like that pathologist. Right? I mean, there is a whole infrastructure to dealing with 30,000 or 60,000 people. And that is a completely different setting I think. DR. LO: Right. I think the point I would want to try to make is that we need to think through what are the characteristics of either the trustee or the decision-making entity within the larger study that are such that we would feel comfortable saying they can make these decisions and, in particular, whether there should be some input in these research studies from the community as to when you cross the line from 1a to 1b because I think there are going to be a lot of judgement calls. And I think that, in the way this is interpreted in practice, I wouldn't want people to not be aware of the nuances and the controversies. DR. MURRAY: Carol had something to say. DR. GREIDER: I am going to agree with a lot of what you were saying, Zeke. And I think that one of the things that would help us all out is just to define specifically how we are going to deal with each of the situations, what the scenarios are in box 1a and 1b and 1c and 1d, and what kinds of protocols we would like to see put in place for each one of those different cases. I mean, I think you are all bringing up specific cases, and I think that they all are covered by this framework. We just have to be very specific about how we define what goes into what box. If we are having trouble here, that doesn't bode well for other groups in the future. So if we can be more specific and lay out the details. And I agree with you; we have got to do the boxology to do that. DR. MURRAY: Bette? MS. KRAMER: I am wondering if we don't have to go back further; thinking about what Bernie said about people who have opted out who might, for the benefit of further knowledge, have changed their mind. And if I recall correctly in the mini-hearings, just about every group, there were a preponderance of the members, the people there, who said that they would want to find out. At least that is my memory from the presentations and looking it over. So I wonder if what we really need to take a look at is the opt-out process? DR. GREIDER: But that fits within one of the boxes, right? That is one of the-- MS. KRAMER: Yes. Right. But it is-- DR. GREIDER: --criteria. MS. KRAMER: It is more procedural than-- DR. GREIDER: It is filling in with what you are going to do in the specific instances; it is not changing the framework. I mean, I think it would be-- MS. KRAMER: So that doesn't change the framework. DR. EMANUEL: I think again, Bette, we have to distinguish. If you are using a sample in an anonymous manner, you don't know. I mean, what that means is you can't link result A with person B. That is what it means. Okay? So you can't actually go back to that. If you are using it in a potentially identifiable manner, which would make it 1b, or any of the b's, you can go back to them, but that would have required consent in the first place. Now, I think we need to-- I am comfortable with that and I would-- You know, we can argue about that, and we did argue about it in the previous-- But I think that does cover the cases and how you can go back. If it is truly in an anonymous manner. I mean, if that is what that firewall is about, you can't walk backwards. We get that information. We can't link A to B. I mean, that-- DR. MURRAY: Harold and Larry have been waiting. DR. SHAPIRO: Thank you. I want to ask a series of questions just to test my own understanding of this because I haven't been part of all of these hearings and I just want to make sure that I understand what is laid out here. And I will try to do it within the context of these boxes. I understand what is meant by use in an anonymous manner to have nothing to do with how it is collected, but it has to do with the nature of the researcher's--in this case--knowledge or capacity to go back to the original sources. In an anonymous manner means there is some kind of wall there so that a researcher cannot go back. Is that-- That is correct? DR. EMANUEL: Right. We think that is one of our breakthroughs. DR. SHAPIRO: Okay. No. I just want to understand. That is what I thought and I have no objection to that. And I think Bernie's point is that that wall is going to achieve a certain kind of dynamic over time and we may want to address that. I understand that we will come back to that a little bit after. I take it that it is true, in all these boxes, that an opt out is always possible; there is no reason why anyone should-- Right? You have opt out in some cases and not in other cases? DR. EMANUEL: No. That is not true. DR. SHAPIRO: Okay. DR. EMANUEL: You have to distinguish previously collected samples. DR. SHAPIRO: Yes. DR. EMANUEL: The 238 million existent samples-- DR. SHAPIRO: Right. DR. EMANUEL: --from the prospectively collected samples. DR. SHAPIRO: Yes. DR. EMANUEL: Okay? In the previously collected samples, when you went in for your breast biopsy, no, they didn't ask you about anything, right? You actually probably signed away your rights in a way that you had no idea. DR. SHAPIRO: Right. DR. EMANUEL: In the prospectively collected sample, we want to make that an explicit part of the process. DR. SHAPIRO: Okay. So the answer is that opt out-- The privilege of opting out of a study, should you know about it, depends on where you fall in one of these boxes? DR. EMANUEL: No. I wouldn't have put it that way. The privilege of opting out of any study is reserved for two categories. One is if your sample is going to be used in an identifiable manner, and, in general way, in prospectively collected samples. DR. SHAPIRO: Okay. So it is only the right-hand side? DR. EMANUEL: Well, and also all the b's. DR. SHAPIRO: Along with some columns. (Simultaneous discussion.) DR. EMANUEL: All the b's. 1b, 2b and 3b. DR. SHAPIRO: Okay. I will come back to that later. Again, just for clarification, I don't know if we are all looking at the same table here, since there is quite a number of them at the end, but the one that is most fully filled out is the one that I am looking at. If I look at 2a and 3a, or 2c and 3c, or 2e and 3e, or the bottom two squares in the anonymous manner, in each case, is it the case that they should read exactly the same as 1a, c and e? Because I can't understand why 2a, 3a, et cetera--the bottom part of those columns where they are in an individually anonymous manner--should be any different from the bottom two squares and the top square. What have I lost here? DR. LO: Well, we do-- DR. GREIDER: I don't think we are looking at the correct version of filling in. DR. EMANUEL: Right. Right. DR. GREIDER: Right. What I have is the filled in thing and what I think Zeke mentioned is that this was a number we agreed on and-- DR. EMANUEL: Right. I have revised that. DR. SHAPIRO: I am just asking a question. I am not challenging anything. I am just asking whether square 2a and 3a, properly filled in, are any different than 1a? DR. EMANUEL: Yes. DR. SHAPIRO: Okay. Well, that I don't-- I don't want to take up time now, since I don't understand that, so one of the members will explain that to me later. I am just-- DR. EMANUEL: Actually, that is-- No. But that is the heart of part of this boxology. Okay. DR. SHAPIRO: Well, okay. DR. EMANUEL: And now I think I understand. The question is why isn't everything just 1? DR. SHAPIRO: Not everything. No. I am just talking about a. Let us take column a. DR. EMANUEL: Oh. DR. SHAPIRO: Column a. DR. LO: It is the community. DR. EMANUEL: It is the community issue. DR. SHAPIRO: But it is anonymous, so who knows anything about community? DR. EMANUEL: Because sometimes you go to particular samples that identify a community. For example, a colon cancer gene where you were trying to identify Ashkenazi Jews, you just didn't go to any sample, you went to a particular sample that you could associate with-- DR. SHAPIRO: So anonymous in this case is only partly anonymous; that is, you can't identify the individual. DR. MIIKE: That is what we mean by-- DR. EMANUEL: By anonymous. DR. MURRAY: You can't identify the-- DR. SHAPIRO: All right. I didn't understand that. Okay. DR. MURRAY: But it may well-- I mean, very little of these tissue samples are useful without some demographic information or illness history. DR. EMANUEL: If you actually look at the very last table--this is I think not updated since our last meeting, but updated from our previous meeting--it now says to be used in an individually anonymous manner and the word-- Maybe we should have stressed or highlighted or bolded "individually" there, because what it means is that you can't identify an individual, but you might be able, through demographics, like Tom says, identify from which community they might come. You might have gone to a particular community to get the sample. DR. SHAPIRO: Okay. So it just means I don't know their name; I may know something about their-- DR. EMANUEL: A lot about them. DR. SHAPIRO: --religion, or about their race, or about something else? DR. EMANUEL: Right. DR. MURRAY: Yes, as an individual. Right. DR. EMANUEL: To emphasize, again the current regs, the common rule only recognizes that box, 1a. It doesn't recognize-- DR. SHAPIRO: Well, I assume-- DR. EMANUEL: --the other boxes. DR. SHAPIRO: Yes. Okay. DR. MURRAY: Larry? DR. MIIKE: Yes. I want to get back to what Dave was having a discussion with Zeke about, which was he wants more information and so he wants to get back to the individuals who provided the original sample. The variable in here is this trustee concept because it seems to me that that is an issue we have got to address. When is there a trustee who acts in place and just sort of a shepherd of the existing information, and when does the individual have to be brought in? At a simplified level, one could say the trustee is there for information that is already in a record somewhere and does not need to be continually collected off an individual. If information is being continually collected off an individual, that person is actively involved and should be--should be--sought after and said do you want to participate in these future research topics? So it seems to me that the question that if you want more information and you are concerned that your model doesn't--your particular instance--doesn't fit in here, it depends on what is the relationship between the test subject and the researcher or the clinician. DR. EMANUEL: Bingo. And that is exactly what-- DR. MIIKE: But it would still fit in these because that would be-- DR. EMANUEL: I thought you articulated it right. DR. MIIKE: Yes. Because it would fit in one of your boxes. DR. EMANUEL: If you are getting information from an existing pool-- DR. MIIKE: Right. DR. EMANUEL: Right? Existant data that you are going to use in an anonymous manner, then it is in the to be used in an anonymous manner. If you need to go back to the subject and get additional information from that subject-- Right? You have got-- You are using it in an identifiable manner and you are wishing to collect special information that isn't extant. First of all, that-- You know, 45 CFR 46 doesn't apply to that. I mean, that is new information you are collecting as part of a research protocol. That is a new protocol and that means you need to get their consent. DR. COX: No. Zeke, listen. It can fit into the boxes. What my difficulty is, is that do the boxes-- What we, as NBAC, want to do with layout, you know, what the discussion, what the issues are-- And if what we do is we say--okay--that, because of privacy issues and because of the difficulties of doing research that, what we are going to do is say the paradigm by how the stuff should be done is that there is a firewall between the people doing research and the patients, I have a problem with that. And the reason I have a problem with it is I think that the majority of the future research is going to require closer relationship, ongoing relationship, between the researchers and the patients because most of it is going to require more and more information and it is going to require continued consent. The trustees are going to be the community--not the individual patients--that you work with and that the researcher is very close with. And so that my concern is not whether things can fit in the boxes or not, but whether we are sending the message that isn't sort of consistent with reality. DR. MIIKE: But you are not suggesting-- DR. EMANUEL: I don't see a-- DR. MIIKE: --that, in a continual relationship between a researcher and a subject, that once a subject gives consent he can't back out? DR. COX: Absolutely not. I am not suggesting that at all. But I am saying that there-- I think that the vast majority of genetic research, at least at the level where it is going to count--not at the level of finding associations--is going to require a very close relationship between the researchers and the individuals in the communities involved. DR. MIIKE: Okay. But can I just-- DR. COX: Yes. DR. MIIKE: Then I think that we are in a different scenario. We are not talking about a piece of tissue lying there; we are talking about the real patient now. The tissue was the entry into that patient. DR. COX: Correct. DR. MIIKE: But you are into a different relationship. DR. COX: Correct. DR. EMANUEL: And that is a different kind of research effort. That is not stored tissue research which-- So I think we are-- I don't think we are in disagreement at all. I think-- I mean, that would require a regular, every-day old protocol that you can-- You already have to do now. It would require the patient's consent because they would be giving you additional information. So I don't think it is-- I don't think it is the issue that we were addressing or, if it is the issue we are addressing, it requires informed consent. And I now apologize for running out unfortunately at a very important discussion. DR. COX: I agree with what you just said, Zeke, but I think that I would hope this is an issue that is on the table with respect to this because it is certainly broader than just dealing with the tissues that are sitting in somebody's freezer. But when we are talking about the use of genetic information in stored tissues, I think that a lot of the action is in these broader issues and not just in what is in the freezer. That is my point. DR. MURRAY: Thank you, Zeke. Sorry that you have to leave, but I understand. We will see you tomorrow. DR. EMANUEL: Yes. DR. MURRAY: I had the bad form to cut Harold Shapiro off in mid-question, so let me give him a chance to finish. DR. SHAPIRO: Well, let me-- Someone tell me if I am not speaking articulately into this microphone. As I look at this table, which does seem to cover most of the cases I can think through one way or the other, although the problem of the wall remains and its dynamic nature, would it be true that if you look at the segment of the matrix that deals with existing samples, that is the left two columns, if I understand this correctly, that the nature of the original consent might somehow matter? And I don't have any suggestion to make. I just have an observation that we deal with those two columns. You may or may not have an original consent of some kind--I don't know the vast variety of things, I expect--and I am just suggesting that that might matter. DR. GREIDER: I think that we discussed that the last time and that we were sort of going for the least common denominator approach, assuming that it was the thinnest possible, if any, consent and giving protections to that-- DR. MURRAY: When we were-- DR. GREIDER: --scenario. DR. MURRAY: Yes. Go ahead, Carol. I am sorry. DR. GREIDER: That is all. DR. MURRAY: When we were being careful in spelling some of this out, we made the point that if there was, with the consent for a tissue sample, some reasonable indication that it would not have been, the person would not have wanted it to be used for X, then it ought not to be used for X. So if there is any indication with the sample that there was, you know, that someone checked the do not use my tissue for research, then you don't use it for research, or if there was some other question that was asked which would indicate someone wouldn't want their tissue to be used-- DR. SHAPIRO: Okay. So the assumption is, as we go ahead here, that is a minimal consent, whatever minimal is. Not being asked I guess is the minimal. Okay. If I can make a few other comments. There is just related questions because I don't-- And that is the issue of community consent becomes very large in these new rows here, the second and third row of the matrix, and I certainly understand why. The question I have, Tom, is whether the committee has given any consideration or talk about just what that would mean? We tend to talk about it as community involvement, which I understand much better than community consent actually. And one suggestion I have, which came out of really a conversation I had with Eric this morning, is that that might actually be a better word to use. But I leave that-- DR. MURRAY: Which? DR. SHAPIRO: Involvement. DR. MURRAY: Well, where actually? I was disappointed to see it described as community consent here because I think we had moved to the notion of community consultation at our last meeting. DR. SHAPIRO: Okay. All right. So that is just-- I am sorry it hadn't caught up with me. DR. MURRAY: Okay. It wasn't in the document. DR. SHAPIRO: Yes. Another question that I have is--let me also just put it as a question--in your own thinking about this, on distinguishing from those that have potential harms, and those that don't; that is, distinguishing in the second row and the third row. In your own discussions, how does that happen? Where does-- How do you decide whether to throw something in one box, the second row or the third row? DR. MIIKE: I think we punted on that and decided we could not be the body that could tease that out to the degree that would be satisfactory. Isn't that right? DR. MURRAY: Bette and Bernie. MS. KRAMER: I am having a little problem because I don't recall that we ever came to a final determination as to how we felt about community involvement, period. We seemed-- The last two meetings, as best I can recall, ended with those issues in the process of discussion but no decisions having been made. So if I remember correctly, on the basis of what--excuse me--what I remember, what is in these boxes is what Zeke had prepared for us around which to have a discussion, but we have never come to a final decision. Now, please correct me if I am wrong. DR. MURRAY: No. I think you are right, although I sensed--probably by mere projection--I sensed a growing consensus that some notion of community consultation was sensible, even though we had to define it better, flesh it out, and defend it. But we still have some of that work to do. But Harold, I think, just asked a different question, which is how will we know, either in terms of substantive principles or some procedural arrangements, when to say that there is potential harm or no potential harm to the community? Bernie and Carol. DR. LO: Yes. If I can try and generalize from the last couple of comments, I think again I am concerned not with the boxes, per se, but who decides what fits into which box and who does that interpretation? And I think it is fine if you have Zeke on call to say, "Well, wait a minute, let me think this through and let me explain to you why it is really in the box here rather than the box there." But you can have zillions of IRBs and zillions of investigators doing this on their own and-- DR. SHAPIRO: How many zeros are there in zillions? DR. LO: Yes. But my concern is that, unless we give this some guidance on these issues, the grid itself won't be as useful. I mean, it is like any other sort of federal regulation on research. It is how the individual IRB struggles with it, and I think they need help on that. One of the considerations they should take into account, deciding whether it is this column or that column, and I think just the issues Harold was raising; who decides? And, again, I would push to say that it doesn't make-- I would urge that we have some community input into whether there are potential harms or not, not just researchers or trustees, or whatever, or steering committees deciding it. DR. MURRAY: Carol? DR. GREIDER: I mean, just to directly address that issue, my understanding from our conversations was that it is the IRBs. That which box to assign it to is subject to IRB-- The researcher first says, "I think it fits blah," and the IRB says, "Yes, we agree with you," or, "No we don't agree with you; it belongs in the other box." And then there is a series of recommendations about what would be done if it were in one box or the other. But to get back to the specific question that Harold asked, I don't think that we even agreed that there was a difference between 2 and 3. We were in the middle of discussing that; whether you could even make a difference between 2 and 3, or whether we should have one column or row for all of the community issues. And we were in the process of discussing that at the end of the meeting. And from my recollection, we never decided whether that should be a 2 and a 3, or simply a 2. So we couldn't have addressed how to decide which goes in it if we hadn't really come to that conclusion. And I certainly wasn't convinced whether it was 1 or 2. DR. MURRAY: My recollection of our conversations and interim conclusions corresponds to the one that Carol just reported. I want to mention that if we were to decide that the distinction between no potential for harm and potential for harm, if we were to decide that was an important distinction, one we wanted to keep in the proposal, we have got three things to do. One was the procedural thing, which is namely let the investigator give the investigator's view. The IRB then makes a determination. So we did get that far. We haven't provided anything by way of substantive guidance as to what we think counts as significant harm, nor have we given examples. Now, I think it won't be that hard to come up with some examples, and it wouldn't be dishonorable to stop there; to give procedure, to talk about some examples, and to give some very general guidance, if we wanted to have that. We don't-- I am actually-- I think you can sometimes do as much harm by trying to sort of precisely specify all contingencies-- MS. BACKLAR: Right. DR. MURRAY: --which you will never do well-- MS. BACKLAR: I think so. DR. MURRAY: --or perfectly, than you will by giving some fairly flexible and vigorous procedure and some general guidance. And that is-- I confess that is my bias in these matters. Bernie? DR. LO: There is no-- There may be another option there, Tom, and that is not to try and specify all contingencies, but to lay out the considerations you ought to take into account and some of the problems that-- DR. MURRAY: Yes. That is what I meant by general guidance. Absolutely. DR. SHAPIRO: Tom, can I just make a comment? DR. MURRAY: Yes. DR. SHAPIRO: If you considered pulling together rows 2 and 3, and maybe that is what some of you want to do, and you went from consent, as you have already done--that is just a mistake in the layout--to consultation, then, if you make that move, the distinction between 2 and 3 becomes much smaller. Because if you put community consultation in both 2 and 3, then it is less and less telling, it seems to me, to make distinctions between 2 and 3, so maybe those things are not independent of each other, at least as I react. DR. MURRAY: I think that is a good point. David? DR. COX: So I-- This is again for my understanding of how the boxes are being used. So now I am over on the right-hand side of the research protocol where people get informed consent to do a specific study with respect to high blood pressure. All right? And their samples-- And they are informed about that. That is the reason for the research. It is just a standard, you know, informed consent. Now is it the case then, when those samples are collected, that if somebody wants to use that for behavioral genetics research, in an anonymous fashion, that the researchers have access to that material? DR. GREIDER: How were they collected? To be used in an anonymous manner or to be used in an identified manner? DR. COX: They were collected to be used in an identifying manner. DR. SHAPIRO: You don't make distinctions like that. You just-- You don't distinguish when you collect it; you just distinguish only how they are used. DR. COX: That is right. DR. MURRAY: Right. DR. COX: So they were used-- DR. SHAPIRO: It doesn't matter. DR. COX: --in an informed consent in an identified way. DR. MIIKE: But now you are asking to use that in a different research protocol? DR. COX: That is correct. DR. MIIKE: You would need-- DR. COX: In an anonymous fashion. DR. MIIKE: Well, in the original consent, they would consent to the research as well as a general consent for use in other research areas so, if it is not identifiable, they have given a general consent and you would be able to use it. DR. COX: Well-- But that is why I want to know what the consent is. This is what-- Because this is the practical issue of where this stands right now. There is-- Patients are being collected under specified research protocols and other people want to use those samples for other stuff and they don't want to be bothered by going back and asking if it was okay. DR. MURRAY: Right. DR. COX: So I am looking at how that fits into our boxes. DR. MURRAY: Carol? DR. GREIDER: One thing is how it fits in the boxes and another thing is what we are going to recommend; the kinds of informed consent that one should get. DR. COX: Exactly. DR. GREIDER: Right? I mean where it fits in the boxes is very easy to answer. DR. COX: Okay. So-- DR. GREIDER: And the IRB would review it because it is a new protocol and say it fits in this box. DR. COX: Well, then tell me where-- DR. GREIDER: The question then-- DR. COX: --it fits, because I don't see where it fits in the box. DR. GREIDER: It fits-- It would be f, 1 or 2 or 3f. Research studies to be used so identification is possible. DR. MIIKE: Well, he is talking about-- He is talking about anonymous. It would fit-- We had a-- DR. COX: It would fit under one box, and it wants to be used in another box. DR. MURRAY: That is fine because we are focusing on the use. DR. MIIKE: Use. DR. MURRAY: When you are focusing on the manner of how-- In terms of-- In that set of our recommendations, which will deal with tissue to be collected in the future, we will deal also with the circumstances under which it is collected in the consent. DR. MIIKE: Remember-- Yes. DR. MURRAY: But our primary interest here is also--our primary interest, not is also--is with use, and whether or not a tissue is regarded as anonymous is anonymous in use or not. DR. COX: No. But that is what I understand because that takes the participant in the research out of the picture. DR. MURRAY: Right. DR. COX: It is the user in terms of defining how they want to use it that has the control, the researcher. The subject no longer is involved unless the informed consent is appeared informed consent. DR. MURRAY: Right. DR. COX: That is the only point I am bringing up because-- DR. MURRAY: And I think-- We will look at possible ways that have been suggested about getting that consent, including getting consent say only for a particular study or a particular line of research versus a general consent to research which could not, at this point, be contemplated, the details of which couldn't be contemplated. That simply are-- Those simply are some of the choices that exist right now, and we may or may not choose to recommend that they be incorporated in the sorts of consents that we envision once this report is in effect. DR. LO: Tom? DR. MURRAY: Yes? DR. LO: If I could add one other point to this discussion, it seems to me that it is important that we keep clear the distinction between research context and clinical context; that we are pretty much in agreement that, in a clinical context, it is hard to imagine how practically speaking you can get a very detailed or thick informed consent. And the only thing-- If I am conducting a big prospective cohort study where I am going to follow people over time, I ought to have ample opportunity to explain, have them ask questions, re-explain, and get a much more detailed consent. And I think the thing that is striking about David's example isn't which box it fits in; it is the-- If I am asked to sign a general consent form and I am not really told what sorts of things I might be signing up for and, in particular, I am not told that there are certain types of genetic research that some scientists may be very eager to do that others find very controversial or down-right objectionable, if I am not told that, what I check off may not be very informed. And I think that in our discussions of the extreme right-hand columns, which I think is different than the clinical context discussions, we need to take into account that there is research and there is research. And I think if I am thinking that it is all going to be for diseases like diseases I have, or things like cancer or heart attack, but someone else is really thinking of--whatever--behavioral things, or other types of really, you know, socially controversial and stigmatizing conditions, that ought to be part of the consent discussion. DR. MURRAY: Yes. Yes. DR. MIIKE: Well, I don't agree. I don't agree because what do we have-- It is not as though this is the only time that someone is going to review the research. If it is going to be used in a controversial research topic sometime in the future, that is going to be reviewed by an IRB or other mechanism to see whether that is legitimate research, and the issues around that will come up. DR. LO: Absolutely. But that is a different issue as to whether I want my sample used in that research. The research may be perfectly okay to the IRB, but I may, as an individual, say, "I choose to opt out." But you didn't-- DR. MIIKE: True. But you didn't-- But you have that choice if you are going to be identified. If you are not going-- If you are going to be anonymous, I don't see how you can-- I just don't see how, when you are being recruited into the research now, you can ever get any kind of a notion about possible uses in the future. So, I mean, that is what our whole scheme is about, is about trying to protect that person if you use it in an anonymous manner or if you use it in an identifiable manner. I mean, you know, what we are trying to do is trying to find a balance between the two, and I don't think you can use the entry into the initial research topic as the be-all and end-all about everything that will go on in the future. DR. LO: Right. But see we draw different conclusions about it. So that depending on how much value you put on-- DR. MURRAY: Bernie, I am getting-- DR. LO: What? DR. MURRAY: --clues that you need to act like a rock star and stick this right in your face when you talk. DR. LO: So starting from that-- DR. MURRAY: Thank you. DR. LO: --observation, you can either-- I mean, you can go two different ways. One says it is so important that we not hamstring scientists that we are going to allow research to be used--material to be used--in an anonymous way even though the patient didn't really have very much idea of what they were getting into as opposed to saying some types of DNA-based genetics research may be so controversial we are going to bend the other way and make it a little harder for scientists and favorable to more of those subjects, albeit perhaps few-- DR. MIIKE: But that is a decision to be made in the future-- DR. LO: --who object. DR. MIIKE: You can't make it at the time that someone is being recruited into a research topic, into a research protocol that has nothing to do with any future. DR. LO: Well, can you at least tell me-- DR. MIIKE: I mean, sure. I mean, you know, I-- I mean, I will sign a form that says, "Don't use my tissue for unethical research." I mean, what good is it? I mean, there has got to be-- That decision has to be made sometime in the future. DR. LO: Maybe-- But my point is maybe we shouldn't-- I mean, another way to say it is that maybe we shouldn't-- It is not all clear to me that you should say that we are going to allow that research to be done because we can't go back and get consent later. I mean, maybe the scientist who wants to do very controversial topics is going to have to put a little more effort into recruiting their subject and selling it on the merits of the research subject, not because the sample happens to be there. DR. MURRAY: Trish and Bette have indicated they would like to speak. Trish? MS. BACKLAR: What is interesting about this is, of course, that we do use advance directives; the things in the future that we really don't know exactly what is going to occur. So there is some history that we have of dealing with the future which is, of course, uncertain and often unanticipated. DR. MIIKE: Well, that is a different question from what he has raised, Trish. I can say it now; "I don't want my tissue to be used in the future." It is a different question that he raises. It is-- I agree with you there are advance directives. MS. BACKLAR: Correct. But we also have some history of advance directives that don't just say no; they also say this is what I would agree to, this is what I wouldn't agree to. The problem with this is it becomes exceedingly complex. And I absolutely agree with Tom and Bernie in trying to keep this as open as possible. I didn't mean to direct you into this. I just wanted the point that there was something there; that we have some history. And we may find it useful to suggest it in some way; that it could be employed in this. Not to close it off though. DR. MURRAY: Thank you. Bette? MS. KRAMER: I am having trouble understanding Bernie's objection because, if it is going to be used in an identified fashion, then it requires a full informed consent. So if it is going to subsequently be used for a research study that had not been anticipated at the time that the subject was initially enrolled, then there is, without going back to that person, there is no way to get that full informed consent, so I think you have to make the assumption-- You have got to I guess have to make one or two assumptions. Either it can't be used at all, or it can be used in an anonymized fashion. DR. LO: That is a very different assumption. MS. BACKLAR: Right. DR. MURRAY: Right. MS. BACKLAR: And I would like to go back to-- I would like to go back to a discussion that we had, a very brief discussion--gosh, I don't know--two or three meetings ago and that was when we raised a question of it is hard-- Before this is all over, aren't we all going to be part of a group, a community, to which somebody might feel there is stigma attached, and isn't that just a part of-- Isn't that just a part of the risk that we all accept, I mean, or that we should all accept? DR. MURRAY: I-- Yes. Let me see if I can press Bernie's point because I think I understand the point, but I may come to a different judgement about it. I am not sure. Let me see if I can just press it. Is it possible, under the kind of thing we are proposing, someone's tissue gathered under one set of circumstances, to then become part of a tissue collection and to then have the use of that tissue requested, in use in an anonymized manner so that my identity doesn't go forward but my tissue does, that it might be used in a way that I would, if I had known about it in advance, find offensive? And I think the answer has to be yes, that remains a possibility. Now, what are the alternatives? One alternative is to go back and knock on everybody's door and ask them for consent again. For a variety of reasons, that is seen as incredibly inefficient. Also at times impossible for certain people who will be untraceable or dead or whatever else. And also, in some cases, it might do more harm than good because people don't want to be re-contacted about certain things. They may not want to be reminded of so and so. I guess the issue is, yes, there is a judgement here. That could happen. Are there any safeguards against it? There are at least two kinds of safeguards. Safeguard number one is the IRB. And frankly if I were sitting on an IRB with such a proposal, to do something which I thought was explosive, my inclination would be to say to the investigator, "Look, I could easily see many people objecting to this so you better go out and get some new samples from people who are consenting specifically to this study." That is protection number one. Protection number two is if it implicates a particular group and it, you know, we keep that category in our proposal, then we have to get community consultation. And if the community says, "Hold on a minute; this is outrageous," then presumably the IRB is going to listen to that and say, "You can't go forward in the way you planned." So there are levels of protection, number one. One and two, really. DR. LO: No. I agree with what you and Larry both said about the IRB. My concern is are IRBs, as currently constituted, really fitted to play that role? And I think having community consultation is important, but I think we have all seen a lot of examples of IRBs composed predominantly of members with affiliate, institutional affiliations just overlooking things that more public and community input might have pointed out. DR. MURRAY: Right. DR. COX: So that, Tom, that you just gave, was a very clear answer to the scenario that I laid out, which is if somebody comes in for a specified research protocol to work on heart disease and somebody wants to do behavioral research anonymized, the answer is they can do it--okay?--with different projections. DR. MURRAY: Well, that you can make the request to use it. Absolutely. DR. COX: Now-- DR. MURRAY: You may not be granted the-- DR. COX: But it is anonymous. But now let us go one step further. And the one step further is that there is a really fascinating result in that behavioral research and people really want to go back and they want to look at these people. The researchers want to do it. But they can't because there is a firewall. "Oh, well, I guess I just won't go back and look at it." Give me a break. They are going to find out who these people are. I guarantee you they will. That is how it works. Okay. And those people are going to be contacted. I guarantee you they will. Now, that is how it works in the real world. DR. MURRAY: Will you give us a guarantee in writing on that, David? (Laughter.) DR. COX: Yes. So-- Because I know my colleagues that do this work. Okay? They are like bulldogs with respect to if they have an interesting scientific finding, nothing will get in their way. So it is not that they are going to violate the law, but they will go and they will identify the physicians who worked with those people. Is it so hard to find these individuals? No. So I am just saying that the firewall-- Okay? You can do encryption, you can do any kind of coding that you want so that the researcher doesn't have access, but so long as there are ties, there are ways for that to happen under the table. Now, I quite appreciate what you are saying, which is to have to go back and talk to the patients every time when you do different types of research, that is not practical. All I am doing is talking about the other side of it which is that, when the patients are taken out of the loop and it is being done anonymous, are we really like talking out of both sides of our mouths on that, and is it in fact the case that those people aren't going to be re-contacted and you won't have more information? And I am quite skeptical about it myself. DR. MURRAY: Now, you-- DR. MIIKE: Can I-- DR. MURRAY: Yes. We have two comments, and then I want to say something. Larry and Kathi. DR. MIIKE: My question to Dave is, is that what you say is a standard practice, or is that aberration? Because I always come back to the point that we cannot develop rules aimed at the aberration. We have got to develop rules aimed at the majority. And then you develop special sanctions for the aberrations. DR. COX: I am sorry to say that I think--okay--although I use it in a really extreme example, that it is more the standard rather than the exception. When researchers have an interesting finding, they pursue that finding, period. And it is not the patients' interests that are the ones that take primary concern. It is not the subjects' issues that take primary concern. DR. MURRAY: Well, you know what, David? I realize this is kind of an awkward thing to say at an ethics commission, but that is wrong. (Laughter.) DR. COX: I agree. DR. MURRAY: Okay. Then we will put that on the record. I mean, I think part of what we are designing are systems, as Larry points out, not that won't stop every possible malefactor from doing something wrong, although we would also like to have them in backup systems and come and try to nail those people and punish them for it, but we-- But that is wrong. And I don't care how enthusiast you are about your finding, you don't violate the protections of human subjects to get those findings. I thought we established that about 50 years ago. DR. COX: I agree. But the problem-- The reason-- One of the reasons why NBAC exists is because all those things are written down and they work not so well because people give lip service to it, but they don't act on it with respect to human subjects research in the way that there are laws with respect to animal research. So I have no problem with a firewall, but I would like not to see, you know, a fire go through it. And right now I am saying that I don't think our society has ways of implementing the concept of the firewall because I think that it is too leaky right now, socially and culturally. DR. MURRAY: Okay. I have more to say but I want to give my colleagues-- Carol, Steve, Bernie. DR. GREIDER: I just have a quick question for David. Thinking about it very soberly, what proportion of scientists doing research do you really think would have that sort of a bulldog attitude, knowing that there are people concerned with the kind of research that might be going on; that there are concerns with their research; that they would ignore it anyway and go ahead? A serious re-estimation of what you just said. DR. COX: All I can tell you, and this is printed, and it was the head of a very-- The president of a prominent scientific society who, at the end of his presidential remarks, made the comment that if it is the patient's consent or our right to do research, I will go for our right to do research. It was a public presidential statement that is written down. MS. KRAMER: David, I am sorry, would you repeat that. I didn't-- (Laughter.) MS. KRAMER: I couldn't hear you. DR. COX: That if it is the patient's consent, informing the patients, or being able to have the sample to do the research, I will vote for going and doing the research. DR. GREIDER: But that is one individual? DR. COX: It is one individual as the president representing the society. MS. KRAMER: And he made that remark in public? DR. COX: In public. A presidential address. DR. MURRAY: I skipped over Kathi in coming up with the list of speakers. Kathi? DR. HANNA: I just wanted to make-- I wanted to make the point, and this ties into what I think David is trying to say, and maybe there is a more diplomatic way of putting it, which is that I think that, for some, I guess-- The question I would ask is how truly useful would anonymized samples be? And I think for most geneticists they would say not too terribly useful unless they are doing molecular epidemiology. They are just trying to find the prevalence of a marker in a population. And that the data--the clinical data--that are tied to that anonymized sample are probably insufficient if they want to go further and try and find gene function, or do reverse genetics. So I think that it is not so much that they have some malicious intent to go and find these people; it is more that the system that is being proposed would render these samples virtually useless to them unless they could go back, and so they are going to have to go back and they are going to find a way to go back. DR. COX: Sure. That is what I am saying. It is not enough saying that these are bad folks at all. DR. MURRAY: Yes you are. (Laughter.) DR. COX: Okay. That is not what I am saying. What I am saying is that if there are ways to get around the system, they will do it because of what Kathi said. It is because in order to have things of utility, the system precludes what they need, and they will go and find ways of bending the rules to be able to get that. DR. MURRAY: Steve and Bernie. MR. HOLTZMAN: I am a little puzzled. Speaking as an organization that spends $50-$100 million dollars a year on genetic research, we conduct all of it in an anonymized manner as we have been talking about and have no problem doing it in an anonymized manner. So I think we have to get a little more granular in our detail in what you are talking about, David. It is one thing to say, "I want to go back and get more clinical information." It may not be sufficient for identifying the individual, hence it can still be in an anonymized manner. If I collect it in the context of a research study, prospectively, I probably address the issue of re-consent, or rather re-contact, in that precisely because I thought it might be the case with respect to a subset that I found I might wish to go get more phenotypic information. Okay? All of those things are consistent with what we have been proposing now. You are suggesting that the paradigm case is the instance in which you have phenotypic information about a research sample collected, let us say, in a clinical context with minimal consent--all right?--so the individual had no idea they were participating in this research, and that going beyond wanting to go back and get some more medical information that is non-identifying, rather the researcher has to get to that individual presumably because they have to get another sample. Okay? DR. COX: Okay. MR. HOLTZMAN: And to me that is not the paradigm case at all. I find that very infrequent. And so I am not sure what you are referring to, Kathi, when you say that what we are proposing doesn't work for the majority of genetic research. I just-- It is palpably false to me. DR. COX: But, Steve, I will tell you why I believe that is the paradigm case, or will become the paradigm case. It is because that it is not the-- My belief that what genetics does is it takes big populations and stratifies them down to smaller groups, smaller groups that are difficult to find; to go out and to recollect because those people are fairly rare. They are maybe 1 percent of the population. So you maybe have 10,000 people but that 1 percent is going to get a lot of attention, a lot of attention. Everyone is going to want to jump on and study that 1 percent of the people that have a particular genetic make-up. So those people are going to get inundated by being studied and you say, "Well, do it anonymous." Right? Or go out-- I mean, I don't understand that if it is anonymous how you, as a researcher, who do you talk to go and get the extra information that you need to design your clinical trial or to ask about relationships between different types of diseases? What is the process? I mean, I am open to the process. I guess what I am saying now is I don't see that we have a process for doing this and I don't-- (Simultaneous discussion.) MR. HOLTZMAN: You said two different things. If my desire is to go back to do a clinical trial, obviously I have to find the body, the person. DR. COX: Yes. MR. HOLTZMAN: If, on the other hand, as we have proposed, a one-way permeable membrane, so that additional phenotypic information, or clinical characterization can be available, I can go back and say, with respect to Sample 71, where you gave us the following phenotypic information, it would be really useful, given what we have discovered, to see if you have any additional phenotypic information of the following sort. And get that. DR. COX: And I am with you. Okay? So-- But the only dispute that we have, in terms of the paradigm, is that most, is most of the research going to be in the context of doing stuff that doesn't relate to sort of clinical trials that are coming up with therapies? Is that going to be the most useful research or is most of the research going to be focused more on clinical trials that involves the body of the individual? And I would argue that the future is going to be more in the direction of clinical trials involved in the body than anonymized stuff where you get a little bit more information to publish another paper. Again, that is just a personal opinion. MR. HOLTZMAN: Yes. I think it is highly improbable because, when we have looked carefully at the economics of thinking of doing family studies as clinical trial populations, it just doesn't make sense. It doesn't work. You can't get the numbers and if you can get the numbers, then the labeling you will get for your drug is so small that you couldn't economically justify it. DR. COX: If I can make one more statement in this regard, and it is a front-page article in the San Francisco Chronicle last Friday, and the headlines to it was "Big Biomedical Budget Push by Clinton." And in it were some statements from Richard Klausner, of the National Cancer Institute, saying that in the next year he has a $3 billion dollar budget planned because he believes that clinical trials are really at risk in this country and that he wants to have more access to patients that want to be involved in clinical trials and thinks that we need a new mechanism because that is the future of research. So that, if he was correctly quoted, was the front-page news article from the head of NCI. MR. HOLTZMAN: That is not inconsistent though. All right? If you are asking me whether or not there will be a pharmacogenetic basis for most selection of individuals for clinical trials, I think that is true. All right? But I don't think it will be necessarily familial or with respect to specific ethnic groups. I don't want to get into the details here about snips and common variants, but you know what I am thinking. DR. COX: All right. DR. MURRAY: Bernie? DR. LO: I just want to suggest that I think it would be really helpful, at least for me personally, if we could have some specific case scenarios. So I think David has some in mind. Steve, you clearly have some in mind saying that we can do a lot of really good research in an anonymous way. Kathi, you are concerned. I mean, to have that give flesh to the report would be helpful. I also think it would help us as we deliberate because it is one thing to look at an abstract grid and it is another thing to say, "Here are some typical research protocols that our best thinkers are saying are typical of what we are going to be facing." How well does our analysis fit? DR. MURRAY: Right. I sense that issue has burned itself out, at least temporarily. Am I right? DR. SHAPIRO: Could I just ask a question? DR. MURRAY: Yes. DR. SHAPIRO: Is the subcommittee decided on why you wanted two rows or three? DR. MURRAY: I couldn't hear you. DR. SHAPIRO: Is the subcommittee decided on whether you want two rows or three here? And I think it really makes some difference. And I don't have an-- I do have an opinion, but I would rather hear the committee's. Or, Tom, if you think that is premature to even discuss now, by all means let us come back to it. DR. MURRAY: I don't know. I don't think we have decided whether to have-- When you say two or three, I take it you mean-- We have all agreed, I believe, that where you have got an individual and there is no sort of group at risk, no community, "identifiable community," that that is Column 1, and we all agree that that has a certain set of rules. The issue is, is there-- Where there is an identifiable community, ought we to do things differently? And do we need to have separate rows for--rows, not columns--rows for when there is no likely harm or some possibility of harm if, in fact, we intend to recommend some model of community consultation, rather than a kind of community veto? I thought it was a good question. MS. KRAMER: Tom? DR. MURRAY: Carol and Bette. DR. GREIDER: I just wanted to say something because I think that Zeke was the one that was really in favor of having three rather than two, so I was trying to recreate in my mind what Zeke might say, just trying to remind myself what his arguments were for having three separate rows. And it might have been because the sort of hoops that one would have to jump through would be different for Column 2, Row 2 versus Row 3. And that is why initially it was set out as three, to not put in the extra added burden where one isn't needed. I am just trying to think through why we initially had three. DR. MURRAY: Right. DR. GREIDER: And so that might be coming back around to the issue of a consultation or community involvement might change somewhat how we would address that if it is less of a consent versus an involvement. DR. MURRAY: Maybe I should just add something. When we heard the-- I can't remember the fellow's name from the last meeting. DR. LO: Jack Killen. DR. MURRAY: Pardon? DR. LO: Jack Killen. DR. MURRAY: Jack Killen, yes. One of the things that Jack Killen helped me to understand better was that community consultation did not merely constitute an obstacle or a punishment. Quite the contrary, in fact. It often contributed in some very substantive but also sometimes subtle ways to the design of a particular study, to the ability to access subjects for study. And maybe one possibility then is to simply say not have two and three, just have two. Just so where community is involved, to then recommend community consultation be undertaken. Now, I think that is, at this point, where I would lean, but I would like to hear what others have to say. And Bette and Trish and Bernie and Harold are all in line. Bette? MS. KRAMER: I just-- I want to understand clearly. When we use the term "consultation" as opposed to "consent," that implies that the community does not have the right to veto the project. Is that correct? DR. MURRAY: That is my understanding, yes. MS. KRAMER: Okay. DR. MURRAY: Although I think practically speaking, if your community with whom you were consulting said, "This is a God-awful thing and we would recommend that no one in our community cooperate with it," you would be foolish to go ahead with it. So I think, in effect, there is a kind of veto, but we are not going to call it that. We are going to call it consultation. DR. GREIDER: You might not be able to go ahead with it. DR. MURRAY: You might not be able to go ahead with it. MS. KRAMER: Right. And then I don't think we ever really addressed satisfactorily, at least not in my mind, what do we do when there are dissenting opinions within the community? DR. MURRAY: As I think it is not an uncommon feature of discussions with the various groups involved with HIV research, which is where a lot of our experience with community consultation comes. You deal with it. MS. KRAMER: And we have the problem-- DR. MURRAY: Negotiate. MS. KRAMER: --of the Ashkenazi Jewish women in the Boston area who didn't want to consent to a study. DR. MURRAY: Right. MS. KRAMER: That was nonetheless being done in other places. DR. MURRAY: Right. MS. BACKLAR: I think if you go back to the first section, where you have community, no potential harms, and community, potential harms, the reason you have to get those boxes into one is because who is going to make the decision about what those harms are other than the community itself? That they need to address it. I mean, it is not going to be very good if it is an outsider who is saying to you, "Oh, no harm to you in this particular case." So that is the argument that I would have for putting them in one box. DR. MURRAY: For putting them-- For not separating them? MS. BACKLAR: For not separating. DR. MURRAY: Yes. I think it is a good argument, Trish. Bernie? DR. LO: I think, to follow up on a point you made, Tom, I think we should try in the report to put forth a position that community consultation is a beneficial thing for the scientists and for the research. It is not a hurdle. It is not an extra administrative burden. That, in fact, anyone doing genetics research, where some sort of ongoing interaction with patients or cooperation of the community is needed, would be foolish to try and plan a study without involving the community from the onset, it seems to me. So that this shouldn't be a conflict; it should be a congruence of interest. DR. MURRAY: Harold, and then Larry. DR. SHAPIRO: My main point was the same as Trish's so I am not going to repeat that. I just want to say one thing. When I looked over the overview of this paper, as Kathi knows from the comments I gave her, I really objected to some of the distinctions that were made there between the researchers and clinicians. If I believe David, I may reconsider my position there. But, in any case, that we will come back to later. DR. MURRAY: Larry? DR. COX: Some of these researchers are clinicians. DR. MIIKE: Just on the issue of community involvement in a community consultation, that has been going on for several years now, even outside the genetic area, so it is not really a controversial issue and I think it is-- Anybody who is going to try to do research nowadays is not going to do it on separate individuals, and I think it is just a practical and an unavoidable process that one has to take up anyway, so-- And I agree with Trish that it is that consultation process that decides whether the harm is minimal or severe and then, even if it is severe, whether the research protocol should go ahead anyway so-- DR. MURRAY: Carol? DR. GREIDER: When we first went through this grid and tried to decide whether there were two or three, I was thinking about it in the mode that we first started discussing it, which is the community consent and how one was going to get consent from a community. But the discussion that we had with Jack Killen last time, about his experience in the AIDS community and how they really had a very integrated involvement of the community with the research, really dispelled in my mind the sort of confrontational us-them sort of paradigm that had initially been set out. And so I have moved from feeling like we really needed to have three to agreeing with what other people have now said; that two probably would fit the bill if we can articulate very clearly the kinds of things that Jack was laying out for us as to why community involvement is important as a part of our report. Because he was very convincing to me about that it is not a hoop to jump through, but rather it is an integrated process of doing the research. DR. MURRAY: Thank you. I think I hear a consensus that we are collapsing Rows 2 and 3. Are there any strong descents to that? Are there any weak descents to that? DR. GREIDER: But what about Zeke? DR. MURRAY: He is not here so we will have him defend it, defend that decision to the group tomorrow, right? That would be-- MR. HOLTZMAN: I mean, I was with Zeke as well. As persuading as I am by Pat's line of thinking, the issue was, to the extent that it was a burden for the kind of study that palpably couldn't be stigmatizing--the number of whorles on your finger or what-not--it seemed kind of onerous. Again, I think just to echo what two of you at least have said, it is conceived of as consultation. It is a very different kind of hoop and, in fact, it is positive. I think what we have to acknowledge, however, is the pressure it then puts on us to give some guidance here to whomever we are asking to make these decisions as to what is a community? What is a-- Kathi, I think in the intro, used the collectivity definition that was found in the Canadian report. Because we are asking the IRBs to say is there a community involved, number one, and, if so, to go get some consultation. So I think we will have to give some pretty specific guidance. DR. MURRAY: That is an important point. I-- We have-- I want to make two other observations. One is that we may have achieved a kind of enlightened consensus, or we think it is enlightened--we know it is a consensus--about the value, potential value of community consultation in these kinds of cases. My guess is that a lot of our scientist colleagues are going to have the same reaction that I am sure some of the scientists at this table had initially. Be aware of that. Be prepared for it. We can do our best in the report to anticipate that and to explain, you know, why we think it need not amount to that. But there is going to be a certain amount of protests and a lot of education that will need to take place. Just be prepared for that, number one. Number two is we are still going to have to, in line with Steve's suggestion, provide some substantive guidance--maybe also a little procedure--for figuring out when a "community" is involved. And maybe that becomes-- We may need to say that that is, in the end, that is an IRB decision whether there is a "identifiable community." At issue here, if so, one needs community consultation. Trish has been waiting, and then David. MS. BACKLAR: I am struck, as we discuss this, of how so much overlaps with our discussions in the Human Subject Committee. And one of the things that I noticed when Jack Killen was here last time was that we did not address the issue of therapeutic misconception, which can occur here and which we want very much to make sure that we get this into this report, and that when communities do become involved they do start to muddle up; that between treatment and research and what may be an advantage to them and what may not. And we need to make sure that the researchers and the IRBs are very aware that, just because this is genetic research, the same issues obtain. DR. MURRAY: David? DR. COX: So I am very in favor of collapsing the three into two. DR. MURRAY: Okay. DR. COX: I think that while it may scare some people doing research to think that they have to have community involvement, in fact almost every paradigm that you look at that has been successful has involved community involvement when it had specific communities, whether they be ethnically defined, or even people with specific diseases. DR. MURRAY: Could you help us by providing some examples of that-- DR. COX: So when-- DR. MURRAY: --to Kathi so we could actually put those, and name them, and describe them in our text? DR. COX: Absolutely. DR. MURRAY: Great. DR. COX: So I am very-- DR. MURRAY: That would go a long way I think towards making the point. DR. COX: Yes. DR. MURRAY: Bernie? DR. LO: I also agree with collapsing the two columns. I think that as we think about the report, I agree, there is going to be a lot of resistance among many scientists; this notion of community consultation. I think, on the one hand, we do need to acknowledge that we talked at first; that both the scientists and the community people need to learn how to talk to each other, need to understand, you know, the languages the other people are speaking. I think it is going to be acrimonious at first. I mean, the first couple of years are not going to be any easier than they were for the AIDS researchers. But you have got to get people to look at the long term, not the short term. And then, to follow up on what Steve was saying, and Tom was saying, I think we should think through how far we want to go with this. I think we should do more than just say, "Do community consultation." We should at least, it seems to me, identify key issues that need to be worked out to make that meaningful. And maybe this is just a process. I don't think it should be left up to individual IRBs. I mean, they are going to all stumble around in the dark. I think the NIH and other national organizations need to take some leadership in calling some national meetings to achieve some guidelines on how you do community consultation. I am not so sure we need to do that, but I think we can sort of say someone else needs to do this. You know, my own feeling is that we don't have to settle all the issues. We just have to point people in the right direction. If we point them in the direction of saying community consultation is a good thing, here are some issues that you need to address to make it really work, here are some procedural things that we think might help get scientists thinking about this. There is going to be a lot of re-training. I mean, a lot of geneticists just really aren't that good talking to people and that is what this is about; talking and listening. They are going to have to re-train themselves. DR. MURRAY: Thank you, Bernie. This seems a good time to take a 10-minute break. If anyone wishes to do public commentary, would you please tell Patricia Norris. Pat, would you raise your hand so that people know who you are? That is even better. Thank you, Pat. Would you please let Patricia Norris know that you would like to do so. And we will see you--I have about 3:33 p.m.--so we will see you at about 3:45 p.m. We will start then. Bye-bye. (Whereupon, at 3:33 p.m., there was a brief recess.)