Archive

IRB Policy Regarding Genetic Research: C. Christopher Hook, M.D., Mayo Clinic

DR. MURRAY: Why donít we go right to Dr. Christopher Hook and get his statement about IRB policy regarding genetic research? As I understand, Dr. Hook, you are the chair of the IRB?

DR. HOOK: I was vice chair of the IRB at the time the policy was formed.

DR. MURRAY: Okay. Thanks very much. And itís a very highly regarded IRB with ample experience thinking about these questions that Dr. Hook is going to share with us.

Just to give you a sense of whatís likely to happen, my guess is, if itís okay with Harold, weíll have Dr. Hookís presentation, some questions and answers, take our break, come back, get into the report. Weíre going to conduct a discussion in much the way that, or at least structure it much the way that the discussion about the prior report was structured, that is, according to the memo that was distributed ahead of time, although I have some additional points I want to add to the conversation. The memo is a bit different from the memo for the competency report. And weíll get as far as we can. My hope is that by 3:00 today weíll have come far enough to do the same thing that weíre going to be doing with the competency report; that is, do a second draft, submit it to commissioners for feedback, and if we feel good enough about it, to post it as a draft on the web. But we will have to see where we are at 3:00.

Dr. Hook?

DR. SHAPIRO: Tom, would you permit me just two amendments. One, weíre going to try to finish todayís meeting at 2:30, even though it means squeezing the lunch hour in some way. There are quite a few, including myself, who have to leave at 2:30.

And second, I just want to indicate for everyone elseís benefit here that Dr. Hook is also Director of Ethics Education at the Mayo Graduate School.

DR. MURRAY: Thank you.

DR. HOOK: Iíd like to thank the Commission for the kind invitation to be with you this morning. Iíve been asked to share with you our experience over the last few years of trying to craft a policy to assist our IRB in dealing specifically with questions that arise in research involving genetic information and genetic testing.

As not unknown to all the members of this Commission, there are a variety of concerns and risks that arise in genetic testing that we were confronted with quite sharply a few years ago by our investigators, including the risk of stigmatization, potential for insurance and employment discrimination, concerns about breeches of confidentiality, potential for discovering undesired or uncertain information in regards to paternity, risk for disease development, and so on. There was also significant concern about the potential impact in the lives of patients who receive genetic information, psychological and emotional harms particularly, and we also recognized that there was a need to protect a right not to know.

And yet, at the time, there was scarce little guidance provided to us in how to address these different concerns. And as our poor gentleman here, we were trying to cry out for some wisdom in how to proceed. Consequently, the chair of IRB asked me to convene a task force to draft a policy. We took the next six months in order to do this task. Then, seeking input from members of our department of medical genetics, individual investigators who performed a lot of trials and laboratory studies concerning genetic information, received input from these individuals, and then revised our draft. This revised form was submitted to the full IRB, and with some additional concerns our policy was approved in March 1996. Soon thereafter, our DNA Results Committee, or DNAR, was formed, and I will explain what the purpose of that is momentarily.

The fundamental principle guiding the activity of our IRB and, in general, our approach to patient care is the thought expressed by Dr. Will Mayo back in 1910óthat the best interest of the patient is the only interest to be considered. And in addition to protecting the best interests of our patients, the goal as we tried to put together our policy was to readdress our understanding of minimal risk and see if it was appropriate in the context of genetic studies. We wanted to provide some safeguards against premature, potentially over-optimistic enthusiasm of investigators. We wanted to ensure that proper and sufficient counseling and informed consent take place. We wanted to protect the right not to know. We also wanted to provide guidance and education for our investigators who needed to be able to address the wide variety of specimens that would be available for this type of research. And, of course, we wanted to see if we could in performing all of these other areas of concern minimize the intrusion into the work of investigators.

Addressing this question of minimal risk, one of the concerns that we received as we began to submit our first draft of the policy was some protest from investigators saying well, now, genetic testing, genetic research really involves keeping or reviewing medical records, taking blood sampling and so on, all of which would fall under the category of minimal risk by strict physical harm definitions. And yet we believed it was important to broaden the concept to encompass psychological, emotional, or social harm, using as a clinical example our approach to HIV serology and requiring informed consent and so on prior to performing that testing.

Therefore, we made a statement in part of our policy that we thought that genetic research, particularly if there is going to be communication with the patient of the results of that research or inclusion of any of that information in the medical record, will involve more than minimal risk to the participants in almost every circumstance. Now, if there will be no identifiers at all maintained with the samples drawn or with the information abstracted from the medical record so that there can be no link to the source, then we believe that the research could proceed without obtaining prior written consent. However, if there was going to be any form of identifier maintained even in coded form, in that context, it would be the opportunity to break the code and recontact, that we would require full informed consent.

If a previous consent for how that specimen was obtained did not allow for the participation or investigation of other things other than that specifically outlined in that consent, then recontact and consent was required. If the specimen had been obtained with some knowledge that additional studies may be performed of a genetic nature but that confidentiality would be maintained or contact not recur, then a waiver could possibly be obtained in accordance with standard regulations.

We wanted to specify the information that we would require for review for our genetic research protocols: What is the information to be generated? What are the risks of incidental finding, such as nonpaternity? How will participants be protected from disclosure to other family members, particularly important in linkage analysis? Will results be recorded in the medical record? Are there circumstances which would require disclosure to the participant or the participantís physician if not previously intended? Will any of the participants be minors, incompetents, or members of vulnerable groups? Where and how will the information generated be stored, the presence of identifiers, as weíve just talked about? What is the risk of inadvertent disclosure? And what are the implications for family members, how will these be addressed as well? These things we require all investigators to have thought about and have an articulated plan or understanding of those concerns.

If information generated in the conduct of genetic research is to be disclosed back to the patient, we required a couple of things be present first. That there be a determination that the results indeed are of sufficient significance to justify disclosure. And that the participant have the opportunity to hear the pros and cons of learning that information and provide informed consent to receive that disclosure.

One of the controversial aspects of our policy has been that if disclosure does occur, that the information be recorded in the patientís medical record. This issue has been readdressed several times within our institution. But our rationale is based on the following concerns. First of all, the alternative to that would be after disclosure the information be included in some sort of a sham history or shadow file. By law in the State of Minnesota and Florida, where we have our centers, any information recorded about the patient that can influence their clinical care, regardless if itís in a separate history or not, is formally part of the medical record and, therefore, is fully discoverable upon request. Therefore, keeping it in some other alternative research file simply does not provide any protection. But if we were to suggest that we were doing that to patients, it might make them mistakenly believe that indeed their information was still fully protected and confidential, which we did not want to mislead them.

We also felt that in many cases the risks involved would be deducible from the family history or potentially altered behavior of clinical course based upon that information. And any astute observer would be able to already assess different risks and be discriminating against the patient on that basis.

Given that case, we were concerned that there might be situations in which a family history would suggest a risk leading to discrimination against the patient, but if the results of the study were negative and we were not allowing that in the medical record, then the patient would be inappropriately discriminated against or receive bias against them. We also felt in keeping with the goals of the medical record, if this information was going to influence patient care, it should be there for all care providers to be able to understand the course undertaken.

We also wanted to reduce the risk of misinterpretation. Much in genetic testing now still is very cautious in its interpretationówhat does this result actually mean, what is the actual penetrance and so on of the trait. And therefore a careful delineation of the caveats needed in interpreting the result can be put in the medical record and decrease the risk of misinterpretation. Also, we just wanted to oppose a general trend so often now of fragmenting the care of patients versus looking at them as whole individuals.

To assist in this process of determining the appropriateness of disclosure of information, we believed that we needed a separate group that would serve as an ancillary body to our IRBóanalogous to the radiation safety group, pediatric research and so onóthat would be charged with looking specifically at protocols involving genetic research. This would be a multidisciplinary group that would try to provide impartial review of the request of justification for disclosure and also to review the method of disclosure plan to make sure that the consent was appropriate, thorough, and clear, and that proper mechanisms for counseling took place.

Membership of our DNAR would include a representative of the IRB; someone from the clinical practice committee, simply because so much of whatís going in genetic research is quickly translated into clinical care; a representative from medical genetics; and emphasis that an ad hoc member familiar with the relevant area of medicine who is not an investigator. You will notice that a question that may immediately arise in looking at this list is we do not have a lay member on this particular group. We are rethinking about this. Our reason initially was that because we were only advisory to the IRB and not a separate, independent IRB that once our review was forwarded that the full IRB with the lay members would have an opportunity to review it and then we would get that input at that time. What weíre finding though is that even among our deliberations, we think it would be helpful to have a lay member there just to ask questions of and to get some perspective on. So this may be a change that will be coming soon.

What have we learned in this process? One of the first things that became immediately obvious was that the education of investigators was crucial and having an articulated policy was an excellent instrument in doing that. Because during the formative stage there were a lot of different concerns that investigators raised. But once the policy was approved, we received positive feedback from the investigators saying we know whatís required of us, we can approach this with those concerns already in mind. And compliance has been good with minimal disagreement. Again, that underscores the value of the clearly defined policy.

We also have found that at least in our institution having this separate group, having the DNAR as an independent body has been valuable. Now this may not work in other institutions. But if you have a very busy IRB, such as ours, where the full IRB meets every other week and theyíre going to be reviewing 2,000 protocols this year, that leaves five to ten minutes for those that actually make it to a full committee. And yet, as we have experienced, the concerns and all the issues raised by genetic protocols completely derail a meeting and take an hour or so in order to thoroughly discuss and evaluate. Having DNAR as the body to engage and take that responsibility has been very helpful. It has enabled us to also have the time to respond creatively to the unforeseen challenges.

And thatís the next major lesson, there are always new surprises, there are always new twists and new complications that no policy can necessarily foresee. Let me try to give you an example of one we just encountered about two weeks ago in which the investigators were looking at a delayed onset illness of relatively catastrophic impact on the patients. Three kindreds were studied. One kindred involved some local individuals, but all the members of that kindred indicated that they did not want to receive any of the results back or ever be recontacted again in the context of this study. The results of the study are quite clinically significant and are to be published. And yet, members of the kindred are individuals who will be in contact with the pertinent medical literature. And so how are we going to deal with that situation of trying to preserve their desire not to know, to preserve their confidentiality, and so on. These are the kinds of issues that a full IRB would be completely derailed if they had to deal with that in full session.

We continue to encounter the concerns about disclosure and including that information in the medical record. Clearly, this impacts the choice a number of patients make in terms of whether they want to participate in research or not once they learn that information will be discoverable if they learn the results. And yet we, as we reason through it, see no other option but to include that information accurately in the medical record because it will be discoverable in some capacity once shared with the patient.

So we as a group have concluded in the face of these difficulties that there is going to be some need for national rules or legislation dealing with the question of genetic discrimination if weíre going to provide access to all the patients that can potentially benefit from this type of research as well as clinical testing.

At this point, Iíll pause and see if you have any questions.

PROF. CHARO: Dr. Hook, first, thank you very much. And on a very personal note, Iím relieved because Wisconsin also moved toward having a genetic subcommittee and itís good to know that weíre going down the same path youíve already tread.

Your approach and the additional details that you provided in the handout strike me as a fairly conservative implementation of the Federal regulations. For example, your committeeís decision to presume that little of this research could realistically be characterized as minimal risk due to psychosocial risk strikes me as a conservative interpretation of the Federal regs. Now, weíve been hearing from certain parts of the research committee expressions of fear that this is going to be a tremendous obstacle to efficient work in areas that require that you be able to continue to match medical records with tissue samples over the course of time.

I wonder since your implementation, and you didnít mention exactly when it is that you put this into place, since your implementation, whatís your sense within the investigator community at Mayo about their happiness with this level of constraint on their work as well as with their ability to collaborate outside since your policy significantly restricts their collaborations to those institutions that follow substantially similar policies.

DR. HOOK: Weíve been at it for two years. And I can say that after the initial protests expressing concerns you just raised, once we went through the process of explaining to them the necessity of why we had to broaden our understanding of minimal risk, I think most of them understood. They had to have a little paradigm shift in their thinking.

It has been more cumbersome for them, thereís no question about that. But yet they are complying. And thatís the report that I can give you is that they are trying to anticipate those needsóanticipate the need for the counseling, anticipate the need for the consentóand theyíre doing it so far.

In terms of external collaboration, that has impacted particularly on one study. And we basically had to tell the investigator that these were the rules that we would live by the patients at our institution and we would ask them to pass that onto the other institutions participating, but that the other collaborating institutions had to understand that perhaps they may not have as many patients enrolled overall because of these restrictions. But there hasnít been any massive revolution yet to have me beheaded or have the policy changed.

PROF. CHARO: Investigators arenít threatening to leave to go to a different institution?

DR. HOOK: No. No, weíve not had that happen.

DR. SHAPIRO: Well, you canít recruit, what can you do?

DR. HOOK: In fact, weíve had other investigators come in since weíve implemented the policy and at first it was very different than what they had encountered elsewhere, but, again, the feedback has been positive. Once they understood the reasons why we were approaching things this way, they said, well, weíll do that.

DR. MURRAY: I have David on the list. Does anyone else wish to be recognized?

DR. COX: I was really also sort of impressed by the idea of having a subcommittee to deal with these issues and was even willing to try and consider taking it further because given the realities of the situation where you have five minutes to do each one. There would have been another alternative, which is that this wasnít a subcommittee of the IRB but it was like a separate thing in and unto itself. And that there are good reasons to keep things together.

But my question to you is just in terms of the nitty-gritty operation of this. As more and more things come into the subcommittee, how effective is the IRB in considering this? Is it really a rubber stamp or should it be, is this going to get so complicated that the structure should be two different IRBs? You see what Iím getting to?

DR. HOOK: And itís a very valid concern. Because when you are that busy the tendency may be that DNAR has looked at it, therefore does anyone have any objections; itís then approved, rather than someone at least presenting the issues again and so on. There is that risk and weíve been concerned about that ourselves. Fortunately, thus far the volume has been that after we submit our report, it is reviewed again at executive, and then submitted to the full IRB. So there still is that review taking place. When volumes come up it may squeeze that a bit. I think certainly bringing a lay member over to us is going to be important to make sure that part of the review take place.

DR. COX: And I had one other sort of follow up, if I may, Tom, which is there is an area that you didnít explicitly discuss but certainly is something weíve discussed a lot, and Iím sure youíre actually dealing with it, which is the issue of the samples themselves. Is this something thatís going to be one of the increased complications as more and more volume comes through this?

DR. HOOK: Actually, this has been a specific issue recently taken up by our IRB, not necessarily the DNAR per se, but with the tremendous volume of specimens generated and requests for access to previously stored or banked materials, specimens and so on. We have just completed the work of another task force, setting up a central, organized group to look at those requests, to look at the resource we have, to make sure that other potential requesters for those specimens in the institution have an opportunity to get involved and how these resources are used and so on, so weíve actually set up yet another policy to try to deal with the movement and access to the specimens.

DR. COX: And whatís the relationship between DNAR and that group then? So, thatís another subcommittee of the IRB?

DR. HOOK: Actually, not so much a subcommittee of the IRB, but set up within the Department of Pathology and Clinical Medicine because they are the reservoir of this material. It was to help guide them to respond to requests of the investigators. And, they will basically submit an approval as part of what the IRB will be looking at--a checklist of things that have to be approved.

DR. MURRAY: Okay. Just to let everyone know the status of the list, I had myself next--Harold, Alex, and Steve. If anyone else wishes to speak, just let me know.

Two questions, Iíll try to be quick. First of all, thanks very much. That was very interesting, very clear, and commendably brief. On all counts, thanks.

As clear as it was, I have one--I had some questions, just try to puzzle some things through. Did I read between the lines that samples that are acquired in the course of research--for the purpose of research--receive a higher level of scrutiny and protection than samples acquired through clinical care?

DR. HOOK: I hope not. I mean, that was not our intent because, again, the issue is what is going to be done with that specimen regardless of whether itís obtained for research or whether itís obtained for clinical care. What research enables us to do is to prospectively obtain consent more readily than a clinical specimen would. But if we had stored clinical specimens and that was their sole means of acquisition--in other words, patient wasnít asked, "Oh, would you mind if we should also use a bit of this for research?" That same degree of the need for recontact and consent and so on take place if identifiers are going to be maintained. There shouldnít be discrimination between those types of tissues.

DR. MURRAY: There was just some ambiguity, I think, in some of the slides.

PROF. CHARO: Yes, itís clearer in the written policy.

DR. MURRAY: Okay. Good. The second question is: I think your very last point on your last slide was what your group determined was a need for legislation to prohibit medical discrimination. And, Iím just wondering what experience--data you have that says this is an important issue. I mean, are subjects refusing to participate for this?

DR. HOOK: Particularly coming to our familial cancer project. There are a lot of individuals who could benefit from participating in the research protocols, or even in the clinical sphere, as testing has become more clinical rather than research-oriented, who are refusing to do so simply because they know that once they do that, that information could be used.

DR. MURRAY: Thank you very much. Harold....

DR. SHAPIRO: Yes, I was very interested in a statement made in one of the earlier slides regarding the barrier you had for the release of information. You said particularly you want to decide if information is valid and significant. I believe thatís, or phrases--I really would just want you to expand on that a little bit. I think thatís a very intriguing idea. Iím wondering how one decides, who decides, what issues come up in trying to decide whether something is significant enough and to whom.

DR. HOOK: Our concern was that as investigators were learning the results of their experiments and so on, that they would make conclusions and want to rush to recontact and rush to inform the patients of that. We thought--try not to be obstructionistic, but at the same time just to encourage people to take a deep breath, to pause, review the situation again, possibly have some nonparticipants subjectively look at that information and say yes, you know, we believe you have something here that is of clinical importance that fits the data...itís hard enough toward going back and sharing that information. And so that was our concern--was to just make sure that sort of impartial--.

DR. SHAPIRO: Well, let me just ask a specific question to see if--help me understand that. Supposing you discover some information which is real about someone, but about which there is nothing anyone can do. Is that significant or not? In the way you think about--?

DR. HOOK: Well, it depends on particularly the action of what the real thing is; in other words, is it something that people are going to be needing to make decisions about? I know in the conduct of their lives and family planning and other things of that nature--these are the things that we would want to be asking and considering. But of course, part of the reason for putting in this step is to say, you know--yes, you found something but if thereís nothing anyone can do itís not going to influence their life any way youíd notice, other than potentially take an emotional hit, then we would try to counsel against that or think of some other way that we can approach the issue.

DR. SHAPIRO: Thank you.

DR. MURRAY: Alex is next.

PROF. CAPRON: Two questions, Dr. Hook. One relates to the guidance we can gather from your experience on already stored samples; the other, samples that might be collected in the future. On the already stored samples, the document that was provided to us in advance that has more details than the slide seem to place a very great restriction on sharing samples outside the Mayo Clinic. If there were to be any identifiers--and in fact it says, on page 3, "...The samples are sent to an outside organization or investigator without any identifiers." We had talked in our deliberation for--about this distinction between the repository and whatís in the repository by way of identifiers and what in the sample had to be used for a particular research project. Your comment orally just now is that you are that you are exhibiting that issue to another committee, did I understand you correctly?

DR. HOOK : The other committee is facing a technical issue of triaging the requests and helping to distribute a specimen instead of a repository.

PROF. CAPRON: Not the policy.

DR. HOOK: No, sir. Now, let me give you some contacts. The Mayo Clinic has always had a very strict requirement, even before this policy regarding genetic research, in how much information or the presence of identifiers we will release outside of the institution--to outside investigators. We have always tried very hard to protect patients from being recontacted by someone outside the institution that they did not, you know, give consent to, that they donít know who this person is. And so, this is just in keeping with our already existent policy of eliminating or removing identifiers before specimens leave our institution.

PROF. CAPRON: Because having made that broad statement in section--in that same subsection on sharing the data, or sharing the sample--you then go on to talk about consent and suggest that with consent you would allow identifiable data. Is that--?

DR. HOOK: Yes. If it is explicitly discussed with the patients that a specimen will leave our institution, we allow that to take place. Under consent we specify a number of things: If it leaves the institution but there are no identifiers, we let them know that specimen will leave our hands; afterwards we have no control over its use. And if there are identifiers, then we have to be very careful that they understand to what extent and whatís safe for us to put in the place to preserve their confidentiality.

PROF. CAPRON: Okay. Now, the question that relates to a future collection of--I was interested in the category of samples--itís on page 2, point 3--of samples with identifiers previously obtained for use in another research protocol, and there youíre very firm about informed consentó"If the sample use will go beyond the terms of the original consent, "óand one of the issues that we talked about is precisely what that would mean. Is it enough to say a DNA study or an molecular genetic study or a genetic study or an examination of family history--I mean, in other words, if right now I would collect a sample for use in a study having to do with a gene that has reproductive effects where the concern is prenatal screening, and then five years from now--and I said, and this is being collected for this genetic analysis and we may want to do further genetic analysis and by signing, weíd be allowed to do that. And then a colleague from oncology comes along five years later and says, "Iíd like to screen these samples for the patientís subsequent experience with breast cancer and do a genetic screening to see if thereís a marker for breast cancer. How do you define in beyond the terms of the original consent?

DR. HOOK: If in the context of the example youíve given, the patient has agreed that some other form of investigation can take place on the specimen other than the explicit one at the time, we would allow that oncologist to have access to the specimen, but we would not allow them to recontact the patient or share any of the results of that until there had been an explicit plan for recontacting counseling to cover the new context, which has just arisen. But if they want to have access to the specimen to do some generic study or whatever, we could allow that.

PROF. CAPRON: But these are specimens with identifiers and so I would assume that if this were an in-house investigator that person has access to the Mayo patient database; and with the identifier, could then see if this patient who was in for reproductive issues five years later developed one of the cancers that Iím looking at.

DR. HOOK: Right. If the initial consent did not specify that any other form of genetic testing may be done in the future, recontact for reconsent would be necessary before anything could be done to that specimen.

PROF. CAPRON: I tried to get to how broadly you feel comfortable. Itís a matter of genetic research or--.

DR. HOOK: General research. Genetic research. Weíve tried to be broad in that regard in terms of we want patients to be able to say "Yes, you can look at this but this only" or "Iím willing to let you have access to this for other types of research beyond the scope of your individual trial."

DR. MURRAY: Thanks. Steve . . .

MR. HOLTZMAN: A couple of quick questions just to follow on that last point. If itís collected in the clinical context, do you also go into that kind of specification about what kinds of future research, or do you just go with a "may be used in future research?"

DR. HOOK: If itís a clinical specimen but yet additional material is drawn at that time for research purposes, the patients are offered a consent to allow additional forms of research that is going to be provided.

MR. HOLTZMAN: So, it can be just a very general kind of--any kind of research.

DR. HOOK: Right. What weíre trying to do as much as possible, particularly in the genetic ground, if we know thatís going to be the questions asked, weíd like that consent to explicitly talk about genetic testing.

MR. HOLTZMAN: So, two more quick questions. The second question is: If you could comment about the second question, how the Minnesota law could be affecting this. The first question is: Imagine I come to you either in Mayo or outside of Mayo and I want 50 prostate cancer samples; I want the complete medical histories. I donít care about the identifiers of the sample and in one instance what I want to do is look at germ line polymorphisms and so Iím looking at vertically transmitted genetic traits. And in the other case, Iím going to look at transcripts from the expression-level differences in certain oncogenes, so Iím not looking at vertically transferred genetic traits but all of the same issues are at stake with the exception of transmission. Would you handle it any differently?

DR. HOOK: If weíre able to strip the identifiers--in other words, you have abstracted clinical information and a specimen that have no means of contacting that patient or knowing who they are--no, I would see that there would not necessarily be a reason to discriminate between the two. In regard to your first, which is your second question, you made reference to a Minnesota law which has a bearing upon all research--not just genetic research. The legislature has passed a ruling in the State of Minnesota that no medical record can be accessed for any purposes of a research nature, even for just an epidemiologic study, without there being explicit consent from the patients. So, the old practice of being able to pull up a list of the last 3000 patients that we have seen with this disorder to try to obtain some epidemiologic understanding about the disease is now going to impact it, in that we have to go through and make sure that all of those histories obtained have given us consent to approach their medical record from a research perspective. Now, 95+ percent of patients have indeed given us their consent and itís now requested as any new patient comes to the institution, but thereís still those patients that say "I donít want my record approached for any purposes other than my direct patient care" and this will have a negative impact on epidemiological--.

DR. SHAPIRO: Could I just ask a follow up question to Steveís, namely, heís made the request of a certain amount of material and youíve taken the identifiers off of that, and youíve sent them onto him providing you have something appropriate. When you say "take identifiers off," does that mean that even you can never trace back where those samples come from?

DR. HOOK: Thatís correct.

MR. HOLTZMAN: But in the sense--in the sense of which he still has John Jones and John Jonesís medical record, and I get No. 1 with the medical record but he gets rid of No. 1 equals John Jones. It hasnít been anonymized in the sense that if you just have a medical record back in Mayo without the individualís name attached. Itís just thereís no code. The link has been broken.

DR. SHAPIRO: Right. You could not get additional information because nobody knows where that record came from.

DR. HOOK: Correct.

DR. COX: Well, then, an extension on that. I donít mean to cut into the line--.

DR. SHAPIRO: Do you want to use your slot, David?

DR. COX: Yes, this is my slot.

DR. SHAPIRO: Okay, use your slot.

DR. COX: This is my question actually, and Iím sorry to--so now, that researcher has those prostate cancer samples that have been stripped and he says, "My God, you know, I found, like, the cure to cancer. But I just need to know one more thing about those people." Canít he help me? Can you help them?

DR. HOOK: If we truly stripped and do not have a linkage--you know, a list at least--we could help, you know, in that circumstance other than if a generic list--these were the 50--the names of the 50 individuals that we sent out to such-and-such laboratory. If we have not maintained the code, then the alternative would be to say "We found something significant--we think we need to investigate this further in the clinical realm.." It would be, I think, justly so to go back to all those 50 individuals and then prospectively say "Okay, your specimen--you permitted your specimen without any identifier--we now have found some information that may be clinically relevant--we are offering all participants an opportunity to learn more about this discovery and whether or not they would like to have more formal clinical testing or involvement occur."

DR. COX: Yes, Iíve got it. Now, what do you do right now? You said that you strip things off but that the subtleties of these, this point, are really quite important because if the 50 people, even if things are taken off, are singled--you know, are kept somewhere. Okay, identifiers have been stripped but you still know who those 50 people are, and why I believe this is important is because then they are people that really are a different type, not individually but just as a group. And, so, Iím not putting value judgments one way or another, but just--is this an issue for you right now, or in general people donít ask for that very much or--?

DR. HOOK: Actually, itís important to maintain some record of every trial that a history or patientís material has been included in.

DR. COX: Thatís right.

DR. HOOK: Because, what we have learned after the advent of the Minnesota statute was there were patients that said "I donít object, but could you tell me of, you know, how many trials has my history been used in and what for and what were the questions being asked?" I think thatís very appropriate for them to have access to that information.

DR. COX: But itís a big difference if you have 10,000 who have prostate cancer, and go back and test all 10,000 as opposed to knowing that it was that group of 50, when you donít know exactly who it was but you can go back and identify who they are. And thatís really where this issue is coming. And for me itís a complicated one because identifiers may be stripped off, but if you can go back and know who that 50 are, then have they really been stripped off? Right? I just--this is a very complicated, an important point.

DR. HOOK: Anonymized only to a certain extent.

DR. COX: Thank you.

DR. MURRAY: Steve, is that an urgent immediate follow up to this?

MR. HOLTZMAN: Itís something Iíd ask people to ask about because itís something that troubles me and that is if you look at guidelines like this, if I ask for those 50 prostates and theyíre good County people whoíve been there for 30 years and youíve got the whole record of them till then stripped of identifiers, you can give me that so I can have the progressive inform--the longitudinal information. On the other hand, if Iím doing the study over the course of a year, and that person has come back in, I canít get that further information. What I struggle with intellectually is why if itís okay if you get the 30-year history up until that point--again, without the identifier--is the problem, is there something problematic about getting the next yearís worth of information or is only problematic because in order to get, have access to that additional future longitudinal information, it opens up the prospect of going back.

DR. HOOK: The answerís yes.

MR. HOLTZMAN: Well, weíve had arguments from Eric previously that after you start the research if youíre going to get new information, they are a research subject and you need consent. So--and I see Alex shaking his head, so I think we need clarification on that issue. Because, again, this goes to this issue of one-way encryption schemes. One-way encryption schemes take care of--if the nub is the recontact, they take care of it; if the nub is a research center, it doesnít. I think we need to get clear on that.

DR. MURRAY: Well, could you ask them to consent? If you anticipated this, then we will continue to--. Okay, if thereís no consent you canít ask. If there is consent, okay. Alta, youíve been very patient.

PROF. CHARO: First just to be responsive to Steve. For me, the major issue is the possibility of recontact because that poses the most opportunities for complicated interactions with information that is clinically ambiguous, has the potential to be as destructive as it is helpful with everybody youíd seeópatient, doctor, investigator. But I appreciate the kind of generic concerns about invasion of privacy that come up even with the identifier stripped. What Iíd like to ask, though, is on a somewhat different subject. Itís back to the notion of minimal risk. Now, since minimal risk is a term that is subject to interpretation by each IRB, Iíd assume that around the country there are going to be widely divergent views on this. And, in your case, I suspect itís possible that the conservative interpretation of it was influenced by the existence of a State law, since to find things as minimal risk in the context of tissue sample research and thereby potentially get yourself closer to obviating the need for consent for the tissue sample research is still not helping your PIs very much if they need to get consent for the medical record use. You know, it doesnít get you very far if you still need to go back to the patient, so there may not be as much pressure on you to have a different interpretation of minimal risk that allows your PIs to argue for a waiver of consent.

But, in addition to that thereís a lot about your evaluation of the psychosocial and legal risks. And Iíd love to hear you expand a little bit about how you came to your conclusions on this. Itís not that I necessarily donít share the instinct; itís that Iíve not seen a lot of good empirical data that in fact confirm peopleís fears about health insurance discrimination or employment discrimination. Thereís a lot of fear; there are some good anecdotes. I donít know if good empirical information that shows it as a widespread phenomenon and I wonder how it is that you all came to this judgement call about the level of risk this represents in reality to your patients, or are you really basing it on their perceptions in regards to reality?

DR. HOOK: In regards to one point that you made, actually our policy was created before the Minnesota statute came along. So that didnít influence our thinking at all. It was our fundamental concern about recognizing impact that this information can have on individual lives. And youíre correct. There are not a lot of good, hard, universally accepted data about this discrimination. There are several scattered reports. One that received a fair amount of press a couple of years ago made the claim that 50 percent of patients who had some form of genetic testing done had experienced some form of employment or insurance discrimination. These were even for things such as being a carrier of a diseaseóman lost his job--things of that nature. And the problem with that study was that it was anecdote-driven. You know, and how the information was obtained was basically through support groups where patients obviously had concerns and the few that--or hopefully few--that had been actually discriminated against were a large number of people that actually went forward.

Even in the absence of a lot of hard data, though, there was just a concern that could happen, and if it could happen the patient should know about that potential risk. But even more so, particularly myself, I was concerned about the literature of which there is much more that dealt with the emotional impact of learning this information, particularly in some late-onset neural degenerative disorders--the high amount of depression, suicide risk, things of this nature, even within the malignancy literature--the rising data. It talks about the need for psychological or psychiatric counseling, depending upon the result. Twenty-five percent of some women learning about their one positivity needing some additional assistance. Itís the 10 percent that test negative that have problems with survival guilt--survivor guilt--and issues of that nature. That was significant enough to us to really drive our concerns about re-looking at the question of minimal risk.

PROF. CHARO: Okay, thank you.

DR. MURRAY: Thanks. Now, I have three names left on the list: Carol, Larry, and Bette. I know that we all have our longer and shorter versions of questions, so Iím going to ask you to use your shorter version or Iíll ask Dr. Hook to give his shorter version of answers so we can break as soon as possible. Carol....

DR. GREIDER: Most of what I wanted to bring up has already been brought up in the discussion that we just had about identifiers and stripping identifiers. So, if one says when itís truly stripping identifiers, but yet you know who those 50 people actually are, the conversations that weíve had in the past--we wouldnít really consider that necessarily completely stripped.

And to get back to the issue that you were just discussing with Alta, which is the psychosocial impacts--if you know who those 50 people are, and this was supposedly anonymous research and yet you can recontact those 50 people and say "Well, thereís a study thatís ongoing using your sample and...," you know, "...maybe one of you has something interesting in there. Would you like to participate in a further study?" That would have significant psychosocial implications. So, simply by "stripping identifiers" in your definition does not necessarily protect against these kinds of recontact issues.

DR. HOOK: There are protocols that we write or encourage the consent to be written in such a way that patients allowed the opportunity to say "no contact" will occur for any reason so that they have prospectively stated "I donít want to know period. Even if you find something, I donít want to be recontacted." And I think patients should be given that opportunity, which may help at least those patients that know that this is going to be a potential--.

DR. GREIDER: But if they came in--weíve had many discussions about people who come in in the context of clinical care and the mental state that people are in when they have to sign some form they donít necessarily know what theyíve signed. And then those are supposedly stripped of identifiers and those 50 people are sent on and then theyíre recontacted, not really being aware that they were in research at all.

DR. HOOK: They should never--youíre making an assumption that thereís never a discussion or some verbal interchange in addition to giving them a form to fill out, which at our institution should not be occurring and isnít, hopefully, to my knowledge. That if a clinical--if a research specimen is obtained at the time of a clinical specimen, that consent form, that discussion, should indicate either that there will not be any recontact or any circumstances, or in the event whatever the investigators think may happen so that that individual has some advance warning that recontact may occur.

DR. GREIDER: But Iím not talking about research at all. Iím saying somebody came in for, you know, to have some of their colon removed. And then--and there was no talk about research. And then that sample is stripped of identifiers and sent on, which I think could occur in your protocol. And then they are recontacted saying, you know, of this--you know--these 50 people, you were one of this set--.

DR. HOOK: Right.

DR. GREIDER: And so what is really stripping identifiers in--to protect people that donít necessarily have knowledge of being involved in research?

DR. HOOK: If there is no prior knowledge of any chance that research occurs, I can tell you that our goal would be, as much as possible, to anonymize any recontact. In other words, to simply minimize any potential--you have been a previous patient at the Mayo Clinic with a prostate biopsy. We are contacting everyone who has had a biopsy to potentially participate or learn in a new study. It would have to be done prospectively that way. If there was no knowledge that the patient had--that they are participating in a study, then we would need to minimize the potential for them to be concerned other than the fact of what they know about what they participated in. "I was a patient who had prostate biopsy."

DR. GREIDER: I wasnít criticizing anything or asking for a specific question. Iím just trying to highlight the issue of what is stripping identifiers and, you know--and then you even brought up the question of you can sometimes strip name, Social Security number, address and still have it be identified by the person reading the journal and recognizing their own pedigree or some other--somebody else doing some other DNA tests, so--so itís not a simple issue of stripping identifiers, which is what we have been grappling with most of the time.

DR. HOOK: I agree with you. I mean, this is the creative part of what weíre all trying to do right now, to preserve confidentiality when there are so many potential leaks in the system.

DR. MURRAY: Larry....

DR. MIIKE: I pass.

DR. MURRAY: That was very brief. Bette....

MS. KRAMER: My question is about the consent procedure. You did not share with us your sample forms. Theyíd be interesting to take a look at, but could you just tell us briefly how you handle the consent procedure and how you handle it specifically in the clinical setting? And, could you give us a evaluation as to whether or not you think that these people are truly informed? Because weíve been concerned about people who come in the clinical setting who are understandably traumatized by whateverís going to be done and they sign and thatís it.

DR. HOOK: I did not share with you any sample forms because we have not generated any sample forms. There are so many specific new ones that no single forms may able to encompass all the basic instructions of genetic inheritance, and we do have sort of a sample of that.

PROF. CHARO: Eric, if I may--just for your information Iím happy to share with you as well as Bette right after this. They are the Wisconsin model forms. With all the different modules.

DR. HOOK: And I want to look at what you have done, but we have tried to articulate more pieces of information that the investigators should include rather than having a generic form. Part of the process of our review is to ensure that consent is adequate is the inclusion of a genetic counselor, DNAR to review to make sure that the information concerning that specific diagnosis being investigated is sufficient to allow the individual to understand the process that is being studied, the type of information generated and so on. We have in some cases asked formally that there be a medical genetics consultation occur during the initial consenting process so that they can determine if they want to learn the information and participate in this study or not. Others have not necessarily required that. The degree of involvement depends on the disorder in question.

MS. KRAMER: So, if I were to come in to the Mayo Clinic for a stomach biopsy, I would not be asked to sign any form that would permit my tissue to be used for any kind of genetic research on admission, and my clinical sample would not be available for genetic research unless I was recontacted?

DR. HOOK: No. You would not--if there was going to be research done and we knew that that was going to be the case, then you would be contacted prior to your endoscopy and given the opportunity to participate in this or not. If we have biopsies that have been obtained for research purposes, which you would need to sign a consent for anyway because the biopsy ó gastric biopsy is not minimal risk, so you would have to have given us permission to obtain those additional specimens in some capacity anyway. But, letís say that it wasnít specified and we wanted now to do some specific genetic studies which may involve recontact of yourself and so on. Then you would be formally recontacted to participate in that way.

DR. MIIKE: But youíve left unsaid though, for those people who just come in for surgery or whatever, that you never use their tissues?

DR. HOOK: No, we do. There are discarded tissues. The issue is--.

DR. MIIKE: Oh, but, but you were just--in response you were just saying we would anticipate research, etc., etc. Iím talking about the case where youíre not anticipating research and you have the tissue. What happens then?

DR. HOOK: So, if you have a section of resected colon, which would be removed purely for clinical purposes, if there is no identifier then the research proceeds without need for recontact. If there is explicit maintenance of a contact of that information, the patient identifier, and so on, and there would be the desire potentially to recontact the patient, then recontact would have to occur.

DR. MURRAY: Alex had a brief follow up.

MS. KRAMER: Well, all right, but in terms of your concerns about psycho-social harm, Iíve come in now, youíve taken my tissue, I wonít sign any consent form for DNA research but my tissue is sitting down there in the path lab that, subsequently, you stripped of identifiers, and then I come along and I read an article in the journal about research that was done at the Mayo Clinic on patients with exactly this problem and these are the proven, you know, and this is--subsequently the DNA research and this is whatís been shown.

DR. HOOK: And your concern is--I mean, are you concerned that your specimen was used for research purposes?

MS. KRAMER: No.

DR. HOOK: Well, you would have--by our Minnesota statute, you would have signed upon entry that information may be obtained for research purposes that may not necessarily be linked specifically to your name and to your record. So, you have already signed the consent for that. If you had said, "I want none of my medical record, which includes tissue specimens, to be used in any form of research,"then your specimen would not have been part of that cohort that was sampled or studied.

MS. KRAMER: And your--and your consent procedure covers these--covers these patients coming in for clinical procedures?

DR. HOOK: Yes, because tissue specimens are by definition part of the medical record.

DR. MURRAY: You want to follow up, Alex?

PROF. CAPRON: I did have one question following up on this. Is it your sense as an informed and experienced IRB person that you were hired to do that, or did you read the Federal regulations to make that simply an option that you had in treating research with stored tissues samples?

DR. HOOK: We--Iím going to answer this--we felt a strong moral obligation to protect our patientsí interest as best we could in a number of spheres. The federal regulations implied some of these concerns were not explicit, and that is why we felt the need of generating our own policies, be more explicit, and deal with different types of specimens.

PROF. CAPRON: You did not feel required.

DR. HOOK: No.

DR. SHAPIRO: Thank you. Let me just make one comment--really two comments here. One is on the issue that has been discussed back and forth here about maintaining a list and the fact that that doesnít ensure in the sense weíve talked about that were unidentified because you do suffer some risks because you might be identified as a group that has some risks associated with it. Itís also true--I was discussing this last night with Dr. Hook, and thereís another aspect of this and that is if you take a little different view and think that you may encounter what you feel is an ethical responsibility to recontact someone and the maintenance of a list has that potential benefit as well as the potential harm thatís identified. It would be interesting for us to think through some things like that as we go ahead, but weíll come to that a little later.

Well, I have good news and bad news. The good news is weíre going to break for 15 minutes. The bad news is Iím going to recommend that we work through lunch. Weíll arrange some kind of logistics to get people a sandwich or something like that. But letís break for 15 minutes. Thank you very much.

DR. SHAPIRO: As I say, weíre going to try now to meet from now to last through 2:30, but I would expect to adjourn at 2:30 for those of us who do have to make planes. It doesnít--others may choose to sit around and pursue some of these matters further, and I certainly hope that will happen. Fortunately, I will not be one of them, so youíll be free of those opinions for the latter part of the afternoon.