1 U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES + + + + + SECRETARY'S ADVISORY COMMITTEE ON HUMAN RESEARCH PROTECTION + + + + + MEETING + + + + + MONDAY MARCH 13, 2006 + + + + + The Advisory Committee met in the Jefferson Ballroom in the Radisson Hotel Old Town Alexandria, 901 North Fairfax Street, Alexandria, Virginia, at 8:30 a.m., Ernest Prentice, Ph.D., Chairman, presiding. PRESENT ERNEST D. PRENTICE, Ph.D., Chair BERNARD A SCHWETZ, D.V.M, Ph.D., Executive Secretary CATHERINE SLATINSHEK, M.A., Executive Director CELIA B. FISHER, Ph.D., Member NANCY L. JONES, Ph.D., Member FELIX A. KHIN-MUANG-GYI, Pharm. D., Member SUSAN KORNETSKY, M.P.H, Member NEIL R. POWE, M.D., M.P.H., M.B.A., Member JAMES H. POWELL, M.D., Member ADA SUE SELWITZ, M.A., Member 2 EX OFFICIO MEMBERS PEG BARRATT, Ph.D., National Science Foundation KATHRYN LYNN CATES, M.D., U.S. Department of Veterans Affairs PATTY DECOT, U.S. Department of Defense SALLY FLANZER, Ph.D., Agency for Healthcare Research and Quality DAVID A. LEPAY, M.D., Food and Drug Administration AMY PATTERSON, National Institutes of Health LAWRENCE UHTEG, NIST ALSO PRESENT FROM THE OFFICE OF HUMAN RESEARCH PROTECTIONS KELLEY BOOHER KRISTINA BOOROR MICHAEL CAROME JULIA GOREY SHIRLEY HICKS IVOR PRITCHARD IRENE STITH-COLEMAN 3 C-O-N-T-E-N-T-S AGENDA ITEM PAGE Opening Statement . . . . . . . . . . . . . . . . .4 Report On Issues. . . . . . . . . . . . . . . . . .5 Overview of Charges to the Subcommittee . . . . . 14 Approval of Minutes from November Meeting . . . . 15 Overview of Meeting Agenda. . . . . . . . . . . . 15 Report of Subcommittee on Research Involving. . . 19 Children Recognition of Departing Members/Swearing . . . .189 In of New Member Further Issues of Subcommittee. . . . . . . . . .228 IO Initiative . . . . . . . . . . . . . . . . . .293 Public Comment. . . . . . . . . . . . . . . . . .302 Adjourn 4 1 P-R-O-C-E-E-D-I-N-G-S 2 8:37 a.m. 3 CHAIR PRENTICE: Good morning, everybody. 4 I think we'll get started. Thank you for coming. 5 What I would like to do is what I normally do, provide 6 a brief overview of SACHRP in terms of its charter and 7 charges to the subcommittees, review the agenda for 8 today. As you know, we are charged with advising the 9 Secretary on the protection of subjects who are 10 vulnerable, and that's reflected by our subcommittee 11 work. 12 It gives me great pleasure to introduce 13 our new member, our newest member, Dr. Neil Powe. 14 Would you please stand up and be recognized? Thank 15 you. And we are going to have a swearing in of Dr. 16 Powe this afternoon by the Secretary, so at that 17 particular time, I will go through his biography. 18 This is the list of the members of SACHRP. I want to 19 once again thank Cathy Slatinshek, who is the 20 Executive Director of SACHRP, and Kelly Booher, for 21 all of their work that they've done to make life easy 22 for all of us. It's a pleasure to be able to just fly 5 1 into Washington, show up at this meeting about 15 2 minutes ahead of time, and not have to worry about a 3 thing. The only thing I have to worry about is 4 whether or not I can get here on United Airlines, 5 which was not easy yesterday, by the way. 6 I'd also like to, once again, recognize 7 the work of the ex officio members of SACHRP who are 8 listed on this slide. They interact very closely with 9 Bern Schwetz and OHRP staff. And as you know, we have 10 a special relationship as an advisory committee with 11 OHRP. We work very, very closely with Dr. Schwetz and 12 all of his staff. And I think that's what makes this 13 whole committee work as well as it does, so Bern, 14 thank you very much for all of your leadership and 15 everything that you do. 16 And that leads us to the next item on the 17 agenda, where Dr. Schwetz will give us a report on the 18 issues. 19 SECRETARY SCHWETZ: Thank you, Ernie, and 20 good morning to all of you. I decided to do this 21 short presentation a little bit differently than I had 22 other times. I usually do it from the table, but I 6 1 wanted to take the opportunity this time to summarize 2 for all of you where we are relative to the 3 recommendations that have been made to the 4 Secretaries, and where we are on other activities with 5 SACHRP, so instead of telling you exactly what 6 happened since the last SACHRP meeting, I'm going to 7 review the status of where we are with some of the 8 things that you have been doing. 9 So three sets of recommendations have gone 10 into Secretary Thompson and now Secretary Levitt, and 11 I just want to update you on where these are because 12 I have to check from time to time myself, and I 13 thought if I'm wondering where some of these things 14 are, maybe it would be of interest to you. So the 15 first set of recommendations went to Secretary 16 Thompson in July of 2004, and by the time we heard 17 back, we had already followed up on your 18 recommendation to put into place the new 407 process, 19 and since then, the guidelines have come out and we 20 have now what we consider to be a pretty good working 21 system between FDA and us, and for OHRP independent 22 when it's not a joint process. So that has been 7 1 followed up on, and that particular activity is in 2 place. 3 With the accreditation, one of the 4 recommendations that SACHRP made when it was finished 5 with the accreditation review was that we should, 6 perhaps, have a conference and look at how the 7 entities that would be under accreditation, which is 8 not just the IRB but the whole program, how do these 9 entities self-regulate, and you are well aware that 10 we've never had that conference. And accreditation 11 has been building and adapting, going from two to one 12 accreditation organizations. So this has been a 13 dynamic process. And I'm not blaming you, Ernie, but 14 in consultation with Ernie, we have kind of agreed 15 that the time wasn't right yet for us to have a 16 conference on accreditation if we were going to have 17 one, so the possibility that we would have that is 18 still there. It's not thrown out, but at this point, 19 it's on hold. 20 We had the Subpart C subcommittee that 21 reviewed the issues that needed fixing right then, to 22 allow us to continue to work on research involving 8 1 prisoners, but you remember one of the things that we 2 decided was that Subpart C was, perhaps the word is 3 "broken," to the extent that it shouldn't just be band 4 aided and fixed that way. So OHRP funded a committee 5 through the Institute of Medicine to review the ethics 6 of conducting research in prisoners. And that 7 committee was formed. It's been meeting now for many 8 months, and we are expecting a report from that IOM 9 committee late this spring, this summer kind of a time 10 frame, at which time, presumably, we would be reading 11 it but they -- we would expect that SACHRP would be 12 looking at that report, and we'll decide what to do 13 about it, reconstituting a subcommittee or what the 14 action might be. 15 So that was the first set. The second 16 set, the primary component had to do with HIPAA, and 17 the discussions that were going on that were a 18 directive to us that we need to harmonize the common 19 rule with HIPAA as it relates to research. And in 20 fact, that recommendation was sent to the Office of 21 Civil Rights and they're still working on it. And I 22 don't know exactly what the time is when we're going 9 1 to see a product of that, but I know they're working 2 on it. 3 As of last Friday, the third set that was 4 put together in July of `05 was still in the Office of 5 the Secretary, but by the end of the day last Friday, 6 it was signed off on and is now delivered back to us 7 within OHRP and there were a number of issues there; 8 the wider range of jurisdiction within Subpart C being 9 one of the questions, and the role of the standard of 10 care, medical care in the review of protocols 11 involving prisoners were two of the issues. 12 Regarding research involving children, 13 this was the letter that had all kinds of important 14 recommendations in it that we need now to figure out 15 how are we going to deal with those recommendations 16 that that subcommittee worked on so hard. And there 17 was a recommendation in there to form a Subpart A 18 Subcommittee which has been in place now for some 19 time. So we're making -- we're going to hear more of 20 the progress on that today. 21 There were other topics. You recommended 22 that we would have an IRB workshop, which we did last 10 1 November, and that will be a topic of discussion later 2 today. In addition, the FDA has been working on draft 3 guidance relative to central IRBs, on the use of 4 central IRBs relative to FDA regulator products. 5 They're at the stage now where they are about to -- 6 they have finalized their document and it's about to 7 be printed in the Federal Register. 8 On international issues, we've had two 9 sessions. We're doing one more session tomorrow, and 10 I think that reflects the complexity of international 11 research as an issue, but also the fact that it's real 12 easy to have a lot of discussion about it and walk 13 away feeling, what did we get our hands around. It's 14 just that kind of a topic, so in the session tomorrow, 15 we're going to get our hands around a piece of this 16 whole issue that has to do with international 17 research. 18 Adverse events, you recommended that FDA 19 and OHRP would come out with guidance. We have our 20 draft guidance out. We've got quite a few comments on 21 the guidance, and we will be working hard to get that 22 into a final form. FDA is working on theirs. They 11 1 held a Part 15 hearing to get advice from the 2 community on adverse event reporting, and they are 3 working internally yet to develop what they need to be 4 able to manage adverse event reporting from the 5 standpoint of the FDA. In addition, Amy Patterson and 6 others have been meeting under the heading of a 7 Federal Adverse Events Task Force to try to coordinate 8 between agencies the handling of adverse event reports 9 and the surrounding information and activities around 10 that. 11 They are in the process of final 12 discussions on a document that they're putting 13 together that would then be beta tested in the 14 community. So if you have more specific questions of 15 that, we can ask them of Amy, but this activity 16 continues to move forward. 17 We did have one session in August of last 18 year that had to do with patient advocacy to keep us 19 connected to that part of the community that we need 20 to listen to, and we did hear some good things in that 21 particular session. We talked about the impact of the 22 fear of litigation on IRBs, on investigators, on 12 1 institutions, and there was no action that came out of 2 that discussion. It doesn't mean it's not important. 3 It's just that there wasn't a defined action that came 4 out of it. 5 Okay, status of nominees; five new members 6 are going to be invited to replace people who are 7 leaving SACHRP now, and they will be invited after 8 this meeting and we expect them to be in the mid-year 9 meeting of SACHRP in August. We will need four 10 additional new members in 2007, so that's a lot of 11 turnover of this committee that has been working so 12 well together, and we will have a Federal Register 13 going out yet this month inviting either people to be 14 nominated to self-nominate, or to nominate other 15 people to serve on SACHRP. We would always welcome 16 recommendations from all SACHRP members, and all of 17 you in the audience, if you want to make 18 recommendations for who should be on SACHRP, the 19 Federal Register notice will be out shortly, but we 20 will take recommendations even independent of the 21 Federal Register process. 22 The status of the charter; the charter for 13 1 SACHRP expires on October 2nd of this coming fall, and 2 we already have been working on that, so we're waiting 3 for the Department to receive our recommendation for 4 revisions to the charter and to get that moving to 5 make sure that there is a seamless operation here. I 6 have no -- I can make no absolute certain statement 7 about SACHRP after this expires on October 2nd, 2006, 8 but I would assure you that I have heard absolutely 9 nothing that would suggest that this will be anything 10 but rechartered. So that's about as confident as I 11 can be, but I've certainly heard no indications in any 12 other way. So that's my update, Ernie. I'll take 13 questions, if there are any, either now or any time 14 during the meeting. 15 CHAIR PRENTICE: Okay, thank you, Bern, 16 and thank you to all the members of SACHRP. You've 17 done an incredible amount of work. I think it's very 18 clear by that update of how far we've come since our 19 first meeting in July 2003. Now, the schedule that 20 you have in your notebooks has been altered because of 21 the schedule of the ASH, so we are going to recognize 22 the party members this afternoon. So therefore, I'm 14 1 not going to go through that now, but we unfortunately 2 will actually lose four of our members who will be 3 replaced. 4 Now, as customary, I go through the 5 charges of the SACHRP subcommittees. We've got two 6 subcommittees in operation. The first one is the 7 Children's Research Subcommittee. And as you know, 8 they're look at all aspects of the Subpart D, 9 Additional Protections for Children Involved in 10 Research, and they're going to present their latest 11 report to us on their work. This has been a very, 12 very active subcommittee under the co-chairs of Sylvia 13 Fisher and Susan Kornetsky. 14 The second subcommittee that we have is 15 the Subpart A subcommittee. They are charged with 16 reviewing all aspects of Subpart A, the Common Rule, 17 and they are going to be presenting their 18 recommendations to SACHRP, as well. I think that this 19 is going to be a subcommittee that has a very long, 20 long tenure. I don't see them going away any time 21 soon. Therefore, Felix, don't think you're going to 22 be able to get away from SACHRP just because your term 15 1 is ending today or tomorrow. 2 The next task we have is approval of the 3 minutes. I will entertain a motion. 4 MEMBER SELWITZ: I so move. 5 CHAIR PRENTICE: Thank you, Aida Sue. And 6 do I have a second? 7 MEMBER KORNETSKY: Second. 8 CHAIR PRENTICE: Thank you. Any 9 discussion, any corrections, any amendments? Hearing 10 none, all those in favor? 11 (Chorus of Ayes) 12 CHAIR PRENTICE: Wake up everybody, wake 13 up, wake up. Any opposed? Any abstentions? Motion 14 carries. Did anybody read the minutes? 15 All right, overview of the meeting agenda; 16 I will go over today's agenda, not tomorrow's. We've 17 already gone through the welcome, opening remarks, the 18 introduction of the new member. Again, we're going to 19 go through that in a little bit more formal sense this 20 afternoon because he needs to be sworn in, but we 21 certainly welcome him, and I'm delighted that he's 22 here. 16 1 Bern has given you the report of the 2 issues. We will move into the next part of our agenda 3 momentarily, which will be the Subpart A Subcommittee 4 report delivered by Felix and Dan. At 10:30 we'll 5 take a break. And then we're going to continue on at 6 10:45 with the Subpart A Subcommittee's report. We'll 7 have lunch between 12:00 and 1:00. We'll continue on 8 after 1:00 with the report of the Subpart A 9 Subcommittee, and then at 2:00 we're going to have the 10 recognition of the departing members of SACHRP, as 11 well as a swearing in of our new member by the 12 Assistant Secretary, Dr. Agwunobi, and he will be here 13 at 2:00, unless there is a sudden interruption in his 14 schedule by the Secretary. 15 At 3:00, we'll take a brief break. Then 16 between 3:15 and 4:15, we're going to have a 17 discussion of the IRB workshop that was sponsored by 18 ASCO AAMC, and Bern will present the results of that 19 workshop, and then we'll engage in a discussion of 20 that. 4:15 to 4:30, we have a public comment period. 21 If there's anybody that wishes to be placed on a list 22 of individuals who will comment publicly, you're 17 1 welcome to sign up or simply wait until the appointed 2 time. And then at 4:30 to 5:00, we will have a wrap- 3 up, and we will then adjourn. 4 Okay, Dan, Felix, would you assemble up 5 here, please? 6 MR. NELSON: Felix just asked me what that 7 beeping sound was behind me and some of you know that 8 I don't jump on new technology. I got my first laptop 9 this year, I don't wear a watch, I don't carry a cell 10 phone, I don't have a PBA. I do have a cell phone now 11 at the insistence of my family, and they were just 12 trying to reach me. So see how that works. 13 Well, thank you, Ernie and Bern. It's a 14 pleasure to be back. I've lost count at what number 15 presentation this is to SACHRP, but most of you know 16 that my day job is at the University of North 17 Carolina, and Bern Schwetz just reminded me that my 18 presentation last year about this time often included 19 references, or sometimes moving pictures showing our 20 University's march to the national championship in 21 basketball, and if you've been following us this year, 22 we don't have quite that much glory going on, so you 18 1 can -- you won't have to suffer through that, but it 2 is a pleasure to be here and to continue our 3 discussions, and to present the next installment of 4 our formal recommendations for your consideration, and 5 hopefully for your acceptance. 6 Just to give an overview of today's 7 presentation, a few -- just a quick recap, as Ernie 8 already laid out on our subcommittee's charge and 9 membership, and then we will be giving our 10 recommendations on continuing review, which was 11 started at the November meeting. We'll then begin our 12 recommendations on the next major topic area of 13 expedited review, and finish up in the afternoon 14 session with some updates on other issues that we've 15 been considering. 16 The charge to the subcommittee then, 17 although Ernie also just flashed this up, are to 18 review and assess all provisions of Subpart A, of 45 19 CFR 46, or the Common Rule, and to review and assess 20 relevant OHRP guidance documents, and based on this 21 review and assessment, to develop recommendations for 22 your consideration in three general categories; 19 1 interpretation of specific provisions, development of 2 new or modification of existing guidance, and possible 3 revisions to the subpart itself. And there will be 4 one or two of those that we put out for your 5 deliberation today. 6 The goals of our subcommittee are to 7 enhance protection of human subjects, to reduce 8 regulatory burdens that do not contribute meaningfully 9 to this protection, and to promote scientifically and 10 ethically valid research. 11 Our subcommittee membership, you've seen 12 before. Gary Chadwick from the University of 13 Rochester, Bruce Gordon from Nebraska, Felix at my 14 right hand, and that's both literally and 15 figuratively, as we work hard on this together. Isaac 16 Hopkins is a community representative from New Jersey, 17 Nancy Jones is at our table here as a SACHRP member, 18 Myra Keane from the University of Minnesota, Susan 19 Kornetsky, also at your table here from Children's 20 Hospital in Boston, Geegee McMillan is Director of the 21 We Can Pediatric Brain Tumor Network in California, 22 myself. Lorna Rhodes is an anthropologist and social 20 1 scientist from the University of Washington, Aida Sue, 2 again at your table from the University of Kentucky, 3 and David Strauss is an IRB Chair from New York City. 4 Our subcommittee meetings are listed here, 5 and I should draw your attention, if you haven't 6 already found them in the additional binder that was 7 handed out by Kelly, hopefully all the members at 8 least have the -- all the printed slides in front of 9 you so you can be following along, and we will reach 10 points where the written recommendations are there for 11 editing and closer reading. 12 These are the subcommittee meetings 13 reaching back to January of last year, alternating 14 teleconferences with onsite meetings. The most recent 15 meeting was held in Rockville, Maryland in the FDA 16 Conference Center at the end of this January, and much 17 of what you'll see today was refined and finalized 18 during that meeting. We also split into working 19 groups basically dividing our subcommittee in half 20 that went to work on these two major topic areas 21 you'll hear about today. 22 Just a little orientation to the format 21 1 and process for our recommendations that are soon to 2 follow; we're trying to learn from our previous 3 experience the -- given our task of reviewing and 4 assessing the entire Common Rule, we've taken various 5 tacts to bring in this work forward to you. We're 6 going to spend a fair amount of time today on some 7 general background and context not only for new 8 members, but I think even for existing members, and 9 for people who are quite steeped in the intricacies of 10 IRB work. We're finding this is necessary in order to 11 get an orientation that then supports discussion of 12 the specific issues to follow. 13 We'll lay out some of our subcommittee's 14 working assumptions in these areas, then identify one 15 specific issue of concern, the subcommittee's 16 rationale why is this a problem, why should it be 17 addressed, why does it get in the way of effective 18 human subject protections. Our proposed 19 recommendation will be on paper in front of you, and 20 then we would stop at that point and ask for 21 deliberation and vote rather than throwing an entire 22 package at you and hoping that you remember what we 22 1 discussed several hours earlier in this complex flow 2 of information. 3 And then we would be repeating the above 4 as needed -- well, and it will be needed because we'll 5 have a number of recommendations throughout the day, 6 so we'd like to, with Ernie's guidance and support, 7 complete that cycle and then move onto the next item. 8 Our written report, some of you may remember from the 9 November meeting, we did produce a written report that 10 has proved to be a draft as we continue to work with 11 your feedback and with our own thoughts, and rather 12 than throwing a series of written reports out to you 13 that end up being works in progress, we're going to 14 wait till the end and hopefully tie all this together 15 as needed and then that would be presented, but really 16 what you see in front of you then is the essence of it 17 today on these slides. 18 Before moving onto the actual 19 recommendations, I would like to wax philosophic just 20 for a bit and ask what's wrong with this picture, what 21 are we trying to fix. This is -- you've heard how 22 many angels can dance on the head of a pin, and this 23 1 is an angel. She's dancing, or at least trying to 2 balance. Granted, it may be a bowling pin. I don't 3 know if that's where the metaphor came from, but I 4 would like to suggest that this is an apt metaphor for 5 IRBs in today's environment. I would suggest that 6 IRBs are spending increasing amounts of time dancing, 7 or at least trying to balance on the head of a pin, 8 and proportionately less time protecting human 9 subjects, because of where we're at in the current 10 state of affairs, our current understanding, our 11 current interpretations. 12 We've had the opportunity, and I've had 13 the opportunity to review some of progress, our 14 subcommittee work with different groups over the last 15 few months, and I get varying responses. When I talk 16 to groups of IRB people, their eyes immediately light 17 up. They recognize the issues. They congratulate us. 18 They thank us, they add more issues to the mix, and 19 there's an instant acceptance and recognition that 20 what we're doing is important, and that it's 21 problematic. 22 When I talk to groups that may include 24 1 investigators, public members, ethicists who aren't 2 necessarily involved in the day-to-day routines of the 3 IRBs, their eyes don't necessarily light up. They may 4 tend to glaze over because some of what we're talking 5 about seems so esoteric and so slicing it fine, so 6 advancing on the head of a pin, if you will. And I'm 7 increasingly forced to step back from the daily work. 8 I think Felix and I and many of you on this committee, 9 and certainly many on our subcommittees would consider 10 ourselves reg heads. We can quote the regs. We apply 11 them daily. We can stand shoulder to shoulder with 12 people as we debate issues like, well, does the 13 approval period end on midnight of day 364 of that 14 year, or did it start at noon on day 365 and when does 15 the letter get sent out. 16 And these rules are important, standards 17 are important, process and structure is important, but 18 I think we're increasingly in danger of losing the big 19 picture by focusing on increments that really 20 represent the small picture, the dancing on the head 21 of a pin. So I think as we go through today, what 22 we're really trying to do is get IRBs out of this 25 1 scenario, and even as there is structure and process 2 and rules to be followed and applied closely to allow 3 IRBs to step back to what I think you and all of them 4 would consider as the real job, which is protecting 5 human subjects, and how to best go about that. 6 So onto continuing review. Again, we 7 approach this process by splitting into a working 8 group that was chaired by Gary Chadwick, whom you've 9 heard from at previous meetings and David Strauss and 10 three other subcommittee members, their work of the 11 working group was then brought back to the 12 subcommittee as a whole, further refined and is now 13 being presented to you. 14 Just as a reminder, we did present our 15 first report and recommendations on this topic on 16 November 1st. At that point, there were 16 formal 17 recommendations that were put forth. Nine were 18 ultimately approved. I've listed them by number here 19 but I would urge you to not get hung up on the 20 numbering scheme. That was our attempt as we brought 21 forward this large set of recommendations just to keep 22 things straight on paper but I think as we move 26 1 through on slides, the important thing will be the 2 content and we'll make numerical references just 3 because there was this report already out there 4 previously but I list them here just for your really 5 recollection that we did work through and with your 6 help and input, approved nine out of those 16 in 7 November. Seven were tabled for a variety of reasons. 8 Some just lack of time to really understand the issue, 9 some a failing on our part to provide adequate 10 background, some sending back to the drawing board to 11 rethink our stance on a particular issue. These 12 remaining issues then, are what will be laid out in 13 the next hour or so. 14 Only items then not approved or tabled 15 last time are being presented today for your further 16 consideration. Just a bit of historical context, the 17 requirement for continued review is one of many 18 legacies from the U.S. Public Health Service Syphilis 19 Study. I think many would recognize remembering this 20 historical landmark that even if you could have 21 reached an ethical construct that allowed this study 22 to be conducted back in the early `30s and addressed 27 1 issues like informed consent and other things, the 2 study failed on many levels that even if that study 3 could have been approved in some state, the lack of 4 continuing review that allowed the study to go for a 5 period of decades long after the advent, the 6 introduction of penicillin as an effective treatment 7 for Syphilis is one of the reasons we have continual 8 review requirements today. It's at least got people 9 thinking about the need for this as they set up the 10 regulatory structure. So continuing review of 11 research was intended to prevent continuing research 12 activities in the face of unacceptable harm, futility 13 or scientific or ethical obsolescence. 14 The Belmont Report does not mention 15 continuing review as an application of its fundamental 16 ethical principles; however, we view it as important 17 and indeed, it was embodied in the regulations as are 18 documented on this slide at a number of levels. So 19 Section 46.103 instructs that IRBs must have written 20 procedures to conduct both initial and continuing 21 review to determine which projects require more than 22 annual review, to verify that no changes have occurred 28 1 from sources other than investigators. Again, IRBs 2 are given considerable discretion as to how they 3 establish these written procedures. 4 46.108 talks about IRB functions and 5 operations and although not naming continuing review 6 directly, it certainly applies there because the 7 deliberations, the actions of the IRB must occur at 8 convened meetings unless they qualify for expedited 9 review and we'll spend a lot of time talking about 10 expedited review in the next segment. 11 46.109 is highlighted because that's the 12 only section that really gets into any depth at all 13 and I'll share that with you on the next slide, on 14 continuing review, and it talks or instructs us that 15 continuing review must be conducted at intervals 16 appropriate to risk but not less than once per year. 17 Section 110 deals with expedited review which to the 18 extent it's used to conduct continuing review, 19 applies. 46.111 are the criteria for IRB approval and 20 I'll review those with you later but they apply to 21 both initial and continuing review and one thing I 22 think the subcommittee would throw out for your 29 1 consideration is to what extent must they apply. Do 2 they have to apply with the same rigor with the same 3 intensity in all circumstances? 4 46.116 are the regulations that guide 5 informed consent and obviously all of that applies to 6 continuing review but there is a specific element that 7 requires that significant new findings be provided to 8 subjects that might effect their willingness to 9 continue to participate and that's obviously relevant 10 to continuing review. 11 So this is the primary regulatory 12 citation, Section 109 and this is really the only 13 section that addresses continuing review process with 14 any amount of detail and in its entirety states, "An 15 IRB shall conduct continuing review of research 16 covered by this policy at internals appropriate to the 17 degree of risk but not less than once per year and 18 shall have authority to observe or have a third party 19 observe the consent process and the research. 20 Well, given the relative sparsity of those 21 references, it's predictable and understandable that 22 there's considerable uncertainty and variability in 30 1 how IRBs have historically interpreted and applied 2 those regulations. So we've asked for guidance and 3 have received it from OHRP and others. HHS guidance 4 has become increasingly detailed and I think we would 5 say increasingly restrictive over the years and some 6 of the specific elements that we will be addressing 7 shortly are those areas where we feel that the 8 guidance has become, perhaps, too restrictive in ways 9 that impedes the appropriate implementation, the 10 appropriate conduct of continuing review. 11 The current, most latest guidance is dated 12 July 11th, 2002 and that document notes that it 13 replaced former guidance on continuing review from 14 1995, guidance on IRB approval periods from January of 15 2000 and discussion of continuing review of DSMB- 16 monitored clinical trials from May of 2000, so those 17 are subsumed into the most recent document. 18 By comparison, the FDA guidance has 19 remained comparatively permissive to the HHS guidance 20 and that's one of the things, of course, that IRBs 21 grapple with are the differences at times in 22 interpretation and detail between the FDA and HHS 31 1 regulations and their interpretation. But the FDA 2 guidance from the FDA information sheet says that 3 continuing review. After study approval the 4 regulations for that outline minimum requirements but 5 do not provide specific instructions to IRBs on how to 6 set up their own rules for continuing review within 7 the framework of the regulations. Therefore, the 8 regulations allow institutions or IRBs to impose 9 greater and more detailed standards of protection for 10 human subjects than those specified by the regulations 11 and permit each IRB to develop procedures appropriate 12 to its needs. 13 And, of course, the HHS regs and 14 associated guidance follow the same concept that they 15 provide a floor, of course, and many IRBs can and do 16 go above that floor. So our working assumptions as e 17 entered this task are that continuing review plays a 18 central and often under-valued role in the IRB 19 process. It's critical. We don't want it to go away 20 but it must be applied appropriately and we feel that 21 practices that do not demonstrably enhance the safe 22 and ethical conduct of research actually diminish 32 1 overall human subject protections because at some 2 level it's a matter of resource allocation. The more 3 time we spend dancing on the head of that pin, if you 4 will, the less time we can focus on some of the more 5 critical tasks. 6 So I'd like to, with that general 7 background and overview of continuing review, identify 8 specific issues of concern. Our subcommittee, if you 9 will remember, was given a list of some 12 to 14 10 questions and as they worked through them, then they 11 came up with the following recommendations. And here 12 again I've maintained the numbering schema that we put 13 in place last time around just so it can be inter- 14 digitated with the work already approved but don't get 15 overly focused on the number. I think the content of 16 the question is what will be intact throughout this 17 little segment here. When can continuing review stop? 18 Must review continue as long as identifying data 19 exists or is there a point at which the IRB and the 20 investigator can close the study file? 21 Why is this an issue? HHS regs did not 22 address when research ends for oversight purposes. 33 1 The full application of 46.111 and those are the eight 2 criteria for IRB approval to minimal risk research or 3 I think we would say to any research past a certain 4 point in it's life cycle and that's an important point 5 that we'll return to, does not add meaningfully to 6 protections and is, therefore, not a reasonable 7 requirements. 8 Another complicating factor is that HHS 9 and the FDA have different operational definitions for 10 what constitutes human subjects research and 11 obviously, that complicates not only continuing review 12 but the oversight of research in general. How do we 13 even know what research requires IRB review versus 14 what doesn't. When do we apply expedited review, et 15 cetera, et cetera. FDA frames human subject research 16 within the context of using a test article with 17 patients or controls for example, a clinical trial. 18 HHS on the other hand, rolls in the concept of private 19 identifiable information in addition to a more active 20 intervention and we're not saying that's inappropriate 21 but nevertheless there are differences that must be 22 accommodated especially by IRBs that need to and 34 1 intend to adhere to both sets of regulations. 2 There are additional scientific factors 3 that factored into our analysis. One is that data 4 analysis may go on for years, of course, and there may 5 be differences in single site studies where all the 6 data is in the hands of one investigator within one 7 institution versus multi-center studies, clinical 8 trials, for example, where different locale, different 9 sites conducting the study may be starting and 10 stopping at different points. When does the overall 11 study stop as opposed to the study activities at t 12 single site. 13 An additional factor, sponsor requirements 14 and simply good clinical practice good research 15 practice may dictate keeping identifiable data for 16 years or forever. Many of us have files full of or 17 boxes back in the archives full of identifiable 18 private information that exists from studies long 19 since closed out, nowhere close to being conducted, 20 nowhere close to having any meaningful role for IRB 21 oversight, yet investigators remain in possession of 22 that identifiable data if that's one of our defining 35 1 factors in continuing review starting or stopping. 2 So these are a few examples where 3 interminable and here, I guess interminable can be 4 taken in a number of ways. We really mean it without 5 end or without a concrete end, but some investigators 6 and IRBs may use it in a more pejorative sense. When 7 is it really not value added? I think many in this 8 room could come up with a much longer list of 9 examples, but a few for your consideration; social 10 science surveys or minimal risk surveys of any kind, 11 individual sites in a mut-center trial after activity 12 ceased at that site but continues at other sites 13 conducting the same study, cooperative group studies, 14 cooperative group oncology studies, AIDS clinical 15 trial group studies, that remain open solely to 16 collect survival data. Many institutions have managed 17 this by rolling all of their studies when they reach 18 that point in their own life cycle in the study's life 19 cycle to roll things into an umbrella study that 20 allows collection of survival data but long after the 21 really active phase of the study is done. 22 Is this a meaningful process even at that 36 1 point? And I would suggest that every study at some 2 point we have to be asking this about, at some point 3 every study reaches a point beyond which the repeat 4 application of all the elements of 111, all the 5 elements of the regulations are no longer meaningful 6 and are no longer value added, and so we ask ourselves 7 who are we protecting and from what? 8 So a number of possible staffing points; 9 when does or when should the IRB's oversight 10 responsibility end? Does it end after the last 11 subject is enrolled, after all the interventions are 12 complete, after data collection is complete, after 13 data analysis is finished or all the papers are 14 published or do we go to infinity and beyond which 15 worked for Buzz Lightyear but doesn't necessarily work 16 well for IRBs or for investigators. 17 So as a result of this uncertainty and the 18 lack of a concrete commonly accepted stopping point 19 for continuing review, IRBs have come up with a number 20 of ways to handle this. Some err on the side of 21 keeping things open forever, which again, we would 22 argue is not meaning full and it's not value added or 37 1 we establish local compromise solutions. At our 2 institution just as a single example, we advise 3 investigators when they come and say, "How long do I 4 have to keep this darn thing open, how long do I have 5 to keep coming back to you", we advise them to keep it 6 open at least through the active phase of data 7 analysis and publishing the results because journal 8 editors may send them back to the data for further 9 analysis. They may even send them back to the 10 subjects themselves to collect more data and at that 11 point, certainly none of us are arguing that we 12 shouldn't have -- that the IRBs shouldn't have an 13 active oversight role. 14 That's sort of a local finesse, if you 15 will, and that's not necessarily commonly practiced Or 16 I'm not saying that's a gold standard that should be 17 practiced around the horn. What is that gold 18 standard, what is -- what should the conventional 19 wisdom be, because that's what we really like people 20 to be following and not making it up on our own 21 somewhere. 22 So as we grappled with this at our last 38 1 subcommittee hearing, I stepped up to the white board 2 and tried to sketch out the -- sort of frame the 3 direction with this diagram and some members of our 4 subcommittee found it helpful so I've tried to 5 translate that into a slide. One of the problems with 6 the communication that the SACHRP received in 7 November, I think, was that it was filled with 8 references to, "When does the study end", and I think 9 we really need to separate in our thinking and 10 conceptualize that we have an entity called the study 11 by which we mean all study related activities and 12 that's quite different and should be different and 13 should be thought of separately and analyzed 14 separately than the parts of the study that require 15 IRB oversight. 16 IRB oversight doesn't necessarily end with 17 the study, which both starts before and ends after, we 18 would argue, IRB oversight as required. So I think if 19 we needed to, although that isn't the point of today's 20 discussions, we could identify a discrete starting 21 point for the study as whole, study related 22 activities. We could also identify and have, although 39 1 that's not at issue today, when is IRB oversight 2 required. I think we would all agree before human 3 subject are enrolled, before human subjects are put at 4 risk in a study the IRB needs to be involved. 5 But there is a gap there. There's a time 6 lag. And activities that may include and do include 7 developing the protocol, going after funding, all of 8 these things occur before we say IRBs really need to 9 have an active role before the study needs to be 10 approved in the eyes or in the files of an IRB. 11 At the back end of things, although this 12 also isn't up for debate today and doesn't need to be, 13 I think we could identify a point at which "the 14 study", all study related activities are done but 15 here's where we're having problems and here's the 16 point of today's discussion; when does IRB oversight 17 end, and that's a much more blurry squiggly line and 18 we're having trouble defining that and that's what 19 we're going to take a shot at on the next slide and 20 ask for either your acceptance or modification. 21 So and actually, I think for ease of 22 reading, this goes onto two slides and after I lay 40 1 these out then, Ernie, we would ask for you guys to 2 step in and ask questions and discuss and give us some 3 feedback on this. So our recommendation is that OHRP 4 should clarify its guidance on the required duration 5 of continual review. Continual review may end when 6 all research interventions and interactions with 7 subjects are over and when data collection for 8 research purposes is complete as described in the 9 approved study plan protocol at the research site for 10 which the IRB has oversight. 11 And it continues, the IRB must have 12 reviewed and approved the investigator's plan for data 13 analysis and the safeguards in place for 14 confidentiality protections. The investigator still 15 retains the responsibility to notify former subjects 16 and the IRB of subsequent analysis and/or new 17 information, raises concerns about rights, safety and 18 welfare of human subjects. 19 So those two slides encompass our 20 recommendation as a whole. The second page is really 21 added as we debated internally how this should be 22 handled. That first -- the first page is very similar 41 1 to what you saw before, that we consider the data 2 collection phase to be the part that really needs 3 oversight, if you will, the part where subjects need 4 most protection. We recognize certainly and there was 5 a fair amount of internal debate and discussion over 6 what happens after that active data collection phase 7 is complete. We recognize that there are still risks 8 when people are retaining data, when data are in hand 9 and that things may come to light months, years later. 10 We don't want the IRB to be completely removed from 11 that process but neither do we feel that ongoing 12 active approval, that requires an IRB file to be kept 13 open by the IRB and the investigator is value added. 14 So the subcommittee members were 15 comfortable with noting that the investigator's plans 16 for data analysis and safeguards, for confidentiality 17 protections which is really the primary risk that 18 we're talking about once active interventions, once 19 active data collection is over. Now we're into the 20 area of confidentiality risk as the primary concern. 21 If those are being adequately addressed and they 22 should have been as part of the up front approval, and 42 1 through the life of a study, if those are in place, 2 and if we remind investigators that they still have a 3 responsibility to come back and, of course we would 4 hope that's in place right now, long after we close a 5 study, if something comes to light, we'd still like to 6 know about it, we'd like subjects to know about it. 7 If those caveats and qualifiers are added 8 to this, this was out subcommittee's feeling as an 9 appropriate recommendation. And with that, I'll turn 10 it over to you, Ernie. 11 CHAIR PRENTICE: Thank you, Dan. What we'd 12 like to do is consider each recommendation as they are 13 presented and, if possible, actually vote on 14 recommendations as appropriate. So let's enter the 15 discussion phase. We'll follow our standard 16 procedure. You raise your hands if you want to ask a 17 question or discuss something in particular and I'll 18 recognize you and we'll try to proceed in an orderly 19 fashion. So does anybody have any questions? 20 Okay, hearing none, let me ask a question 21 with regard to the CR Recommendation 1.1 (continued), 22 I understand the rationale for requiring the 43 1 investigator to still assume the responsibility for 2 notifying former subjects of any new risk information 3 or any other information that may have clinical 4 utility for them or impact them in any way. But you 5 also have a statement to notify the IRB. 6 When studies close, IRBs have various 7 mechanisms to either maintain records for the three 8 years or they actually eliminate records of studies. 9 So I'm trying to understand the rationale for keeping 10 the IRB in the loop, particularly if we're talking 11 about a study that's been closed for let's say a 12 decade. Could you comment on that? 13 MEMBER GYI: Subcommittee did not consider 14 the logistics issues regarding the IRB's decision to 15 destroy the records beyond following the regulatory 16 requirement of three years. I suppose just to hazard 17 a guess, if it were to come before an IRB and the 18 records had either been archived or destroyed, the IRB 19 would then have to sort of rethink what that process 20 is. Just by way of example, let me share with you the 21 story of the Salk Vaccine. 22 I had shared this experience with a couple 44 1 of investigators awhile back and we had talked about 2 the fact that the Salk Vaccine came to the marketplace 3 very quickly following research and without the 4 appropriate oversight and protections in place today 5 back in 1954/1955. And I was reminded that the 6 original vaccines were harvested from the monkey 7 kidney cells and subsequent exposures to subjects 8 resulted in kidney damage. And the person who was 9 talking to me reminded me that perhaps, this is the 10 role where the IRB might have had an impact. If they 11 had known about it, even after the fact, the IRB could 12 have made some determination as to what the research 13 related implications were. 14 And I think that that has value here as 15 well. Even if the research records had been 16 destroyed, I think it would be incumbent upon the IRB 17 to make some determinations of how they wanted to 18 address that particular issue. Our initial impulse, 19 though, was stimulated by the discussions around this 20 committee of oncology related studies that Susan 21 Weiner had reminded us about and I think we had a 22 great deal of discussion surrounding that, and you 45 1 know, the IRB, I think there was a strong feeling that 2 the IRB ought to know, even if the study had been 3 closed and archived the record is destroyed. 4 So that was the reason behind the 5 subcommittee's continued discussion about making sure 6 that the IRB was informed of that, especially relating 7 the handling for the rights and welfare of the study 8 participants. 9 CHAIR PRENTICE: Okay, thank you. Other 10 questions? Celia? 11 MEMBER FISHER: I was trying to figure out 12 the relationship among 102(d)-(e) and (f)(2) in terms 13 of the continuing review because I think one of the 14 points that Dan was making was that HHS includes 15 identifiable information which then would mean, as 16 long as you're analyzing data and you have the records 17 of the -- the researcher has the records of the 18 identities of the individual, it would continue to be 19 under IRB review but I was just wondering if they 20 could help out why -- how they reached that assumption 21 because -- and whether the regulations actually say 22 that. Do you understand what I'm asking you? 46 1 MEMBER NELSON: Not entirely. 2 MEMBER FISHER: Okay, I'm sorry. You 3 point out that there's a difference -- you point out 4 there's a difference between the FDA and HHS, right, 5 on page 11, I don't know what slide it is, and you say 6 that HHS includes the concept of private information 7 in addition to intervention. 8 So I'm assuming -- what is the 9 relationship between when continuing review can stop, 10 the definition of research, the definition of human 11 subjects and why you would assume that as long as a 12 researcher can identify the subjects well after data 13 collection is over, the HHS regulations in and of 14 themselves, are creating an obstacle. I guess that's 15 my question. 16 MEMBER GYI: Just as a point of 17 clarification, I wonder if there's an FDA 18 representative. I don't see Dr. LePay here, so -- 19 because this is an issue that sort of touches on both 20 sides. There is one, thank you. 21 MEMBER NELSON: Okay, a fair question, but 22 if I can paraphrase slightly but not to lose the gist 47 1 of your question, Celia, if you go back to the 2 beginning of a study, what requires IRB review in the 3 first place, IRB review and approval, well, it doesn't 4 have to be an active intervention at all. You create 5 a human subject. You make someone into a human 6 subject by obtaining data about them even if they 7 don't even know you exist. Going to do a medical 8 records chart review study, retrospectively, creates 9 human subjects. So I guess taking that to its logical 10 extension through the life of a study, as long as you 11 have those data that person remains a human subject. 12 I think what we're saying is that at some point, we're 13 accepting that even though those data remain under the 14 control of an investigator, if appropriate 15 confidentiality measures are in place, it's not value 16 added to continue, I guess to continue to regard that 17 person as a human subject in need of IRB protections 18 for the rest of -- well, for the rest of their life or 19 for the life of those data, accepting that the data 20 may be out there forever. 21 And that's the dilemma here. I don't 22 think we have any problem up front saying, you know, 48 1 back to that right here, the beginning of IRB 2 oversight, we're all in agreement, I think that an 3 investigator obtaining data, whether or not it's 4 reactive intervention, creates a human subject, 5 creates the need for IRB -- triggers the need for IRB 6 oversight and protection, and it's on the back end 7 where extending that rationale, that solid logic 8 creates problem. 9 And I think what the subcommittee is 10 trying to say is that at some point we have to be 11 comfortable that those data are out there. They may 12 have gone off to sponsors. They may be sitting in 13 locked files. They may be in computers forever, but 14 the act of oversight by an IRB is no longer value 15 added. 16 I might, if Mike Carome or others are 17 willing, invite OHRP's response to that same question 18 because some of this gets back to -- well, a lot of it 19 gets back to obviously, OHRP's definition of the -- 20 not the end of a study but the end of need for IRB 21 oversight. 22 CHAIR PRENTICE: We can certainly do that, 49 1 Dan, but I do want to offer what I think might be a 2 clarification for Celia, which is not self-evident 3 necessarily in the slide you're referring to. The HHS 4 definition of human subject, of course, includes the 5 attainment of identifiable private information but 6 that's in the context of conducting research. So if 7 the research is complete and you can debate when that 8 is, but if you look at the recommendation of the 9 subcommittee, basically what they're saying is when 10 the analysis of the data as described in the protocol 11 is complete then the research is complete. 12 So even though you would still retain 13 identifiable private information, it does not 14 constitute research any longer, and therefore, not 15 subject to IRB review assuming that we accept this 16 particular recommendation. Does that help? 17 MEMBER FISHER: Well, I absolutely agree 18 and I guess I agree with the recommendation. The 19 question is, how it should be phrased and I think in 20 the earlier slide, it suggested that OHRP or whoever 21 would make the -- with this guidance that in some 22 sense one is saying something different from what is 50 1 in the regulations and so that's what I was trying to 2 articulate. I think the recommendation is excellent. 3 Whether or not it's a shift from how the regulations 4 could have been interpreted for a very long time, I 5 think is the essence of how it will be framed when 6 it's a recommendation. So that's why I was asking is 7 this now a new direction or have IRBs been overly 8 conservative in tying (f) to (d), when, in fact, 9 they're two different letters, which is, I think, what 10 you were saying. So it's really how it would be 11 phrased. 12 CHAIR PRENTICE: Okay, I think it's new 13 direction but I would certainly invite OHRP to comment 14 on their view as to whether or not this is a new 15 direction. Dr. Carome, would you care to comment? 16 DR. CAROME: What OHRP has long-stated is 17 that the use and analysis of identifiable private 18 information from living human individuals represents 19 human subject research. And so with respect to when 20 does continuing review need to continue at least 21 annually, we have not said that as long as 22 investigators hold identifiable private information, 51 1 they must continue to subject the research in which 2 that data was collected to continuing review. 3 We have said that the analysis and use of 4 the collected identifiable private information as 5 described in the protocol as proposed usually in the 6 statistical or analysis section of the protocol, that 7 represents ongoing human subject research. And so 8 this recommendation would, if adopted by the 9 Department and our office would reflect a change in 10 our current position. 11 CHAIR PRENTICE: I don't understand what 12 you just said because you said that as long as an 13 investigator is using or analyzing data for the 14 purposes of research, it would necessitate continuing 15 review and you also mentioned the -- when the analysis 16 as described in the protocol is complete, then there 17 would no long be a requirement for continuing review 18 which seems to reflect this recommendation. How is 19 that different then than OHRP's position? 20 DR. CAROME: I think the recommendation 21 would allow for the investigators to continue to 22 conduct the analysis as proposed in the research of 52 1 the identifiable private information collected but no 2 longer required for continuing review. Dan, that's 3 how I read it, so I think Ernie is reading the 4 recommendation differently that we are reading it. 5 CHAIR PRENTICE: Yeah, I'm reading it 6 differently. Dan? 7 MEMBER NELSON: Well, let's read it 8 together. So "Continuing review may end when all 9 research interventions", and I should have probably 10 had a follow-up slide with some of the operative words 11 here highlighted, but I think speaking for the 12 subcommittee, research and interventions, the subjects 13 are over, that's nearly a no-brainer, I think. We 14 all want that to be done before we would close out a 15 file. And data collection for research purposes and 16 that was added at the last subcommittee hearing, 17 perhaps, Ernie, at your suggestion but certainly 18 agreed to by the subcommittee, that there may be other 19 data collection that goes on, for example, for 20 clinical purposes, if it's a patient in a clinical 21 trial, but data collection for research purposes is 22 complete as described in the approved study plan and 53 1 then the -- after the comma, at the research site for 2 which the IRB has oversight recognizing that if you're 3 part of a multi-center study, other sites may continue 4 with these activities. 5 So I guess the point of discussion right 6 now is data collection for research purposes is 7 complete. We are recognizing that data analysis may 8 go on. Again, if you go back to those scientific 9 factors as I term them in that slide, that impact on 10 our discussion, data analysis may go on for years and 11 we're saying that IRB oversight during those years is 12 not value added. That you know, if something came up 13 today that would send an investigator back to a study 14 that had been closed out two years ago, and they 15 realized that by going back and pulling up the file 16 already in their possession, that the sponsor has told 17 them should -- in good practice, has told them should 18 stay in their possession for what seven years if it's 19 NIH funded, forever if you were brought up the way I 20 was, which is to not throw anything away, if those 21 remain in your possession going back to them, 22 analyzing unless you're going to need to go back 54 1 through medical records, back to recontact subjects 2 directly and re-engage them in the process, that 3 they're okay without our IRB protection at that point. 4 CHAIR PRENTICE: Okay, Mike, I agree with 5 you. It is different. I was not looking at the term 6 -- I was looking at collection and probably we need to 7 refer to analysis as well. It doesn't -- so this is 8 a difference between your current position and the 9 recommendation. Celia? 10 MEMBER FISHER: I was jut wondering and I 11 don't know enough about how IRBs operate. I do 12 understand that we want to reduce burden and at the 13 same time protect human subjects. So I'm just 14 wondering as some kind of a midway point between 15 OHRP's concern about the continued protection of 16 subjects when there is identifiable data, and the 17 concern about not having full IRB review, for years 18 when data is being analyzed, whether or not there's a 19 way to do something akin to an expedited. 20 Once data is -- data collection is 21 completed, confidentiality protections have been 22 approved, then there's some kind of yearly update to 55 1 insure the IRB as the data is being analyzed or kept, 2 that the confidentiality continues to -- protection is 3 continued to be maintained. Is that possible within - 4 - is there a way to streamline the continued review of 5 confidential information storage? 6 MEMBER NELSON: Go ahead, Ada Sue. I bet 7 you have the same answer I do. Go ahead. 8 MEMBER SELWITZ: In fact, that's what we 9 do. So that mechanism is in effect. It is 10 streamlined. It is expedited but that doesn't negate 11 the fact that it does require a substantive review, 12 even if it's expedited. I mean, just because it's one 13 person versus an entire committee and from an 14 administrative standpoint and the burden on managing 15 the entire system, it's -- it is a problematic. 16 I mean, from a staffing standpoint, you 17 spend as much effort on expedited as you do full 18 except that it isn't in the minutes. So that, in 19 fact, is currently being done. I think the reason it 20 is so important in this particular recommendation that 21 we include that PI's have an obligation to report any 22 problems that should occur with the data analysis, I 56 1 think that's an important part of this recommendation 2 because that -- otherwise, I would be uncomfortable 3 with saying, "We never hear from you", but on the 4 other hand, I think if you tell PI's that if there is 5 a problem and they have an obligation to report. And 6 I'm not worried about the logistics on that, Ernie, 7 now that so many of us are moving to not even 8 electronic submission. 9 I mean, most of us have an electronic 10 record that we do not destroy over a long period of 11 time. So I think the logistics of receiving a report 12 we could manage it quite effectively and would want to 13 receive it. But I think ethically, that does a great 14 deal to strengthen the merit of this particular 15 proposal. 16 MEMBER FISHER: Can I just say this? I 17 don't see the oversight on a sloppy investigator here. 18 I only see in the recommendation that if the 19 investigator decides that the data analysis itself 20 raised a safety issue, that then they should be 21 encouraged to report it to the IRB but I guess the 22 question I have is, where is the oversight which is 57 1 the purpose of the IRBs? 2 I guess the other question I was asking is 3 I understand that regular expedited review is the same 4 as a review. You have to look at lists of benefits, 5 et cetera, et cetera. I was asking whether or not 6 there's even something more streamlined for 7 termination of data collection, data analysis and then 8 the only thing that needs to be reviewed is something 9 that says confidentiality, procedures are continued to 10 be in existence. That -- I was wondering if that's a 11 possibility. 12 MEMBER SELWITZ: You know, it's not a 13 possibility in terms of the current existing guidance. 14 It's not. 15 MEMBER GYI: I think that perhaps we ought 16 to put on the table as well, the differentiation that 17 Celia is making and that is something along the lines 18 of an investigator initiated study where perhaps an 19 oversight may be a little bit different than a multi- 20 center FDA regulated sponsored research where the 21 investigator may not have any access to the analysis 22 information but may have access to data as we're 58 1 talking about right now. 2 And so part of the dilemma that the 3 subcommittee had was in trying to balance those 4 differences between what constitutes an investigator 5 initiated study, perhaps even spanning into behavioral 6 science where data analysis as Dan was talking about, 7 and data access may go on for years and the very large 8 volume of sponsored research where once it leaves the 9 investigator's shop, that study essentially ends at 10 the investigator's site and the analysis and access to 11 further data resides primarily in the sponsor's shop. 12 MEMBER NELSON: Let me just add to that, 13 that the members of our subcommittee who are most 14 experienced and spend their time in the social 15 sciences felt quite strongly stemming from the point 16 Felix just made, this is not all about clinical trials 17 where you have concrete bundles of -- maybe concrete 18 is the wrong word, they're heavy but there's discrete 19 bundles of data that are analyzed numerically and 20 statistically. That's quite a bit different than a 21 qualitative data analysis. 22 We lump it all together under the rubric 59 1 of data analysis but the qualitative assessment that 2 an ethnographer may do that also goes on for years is 3 quite a bit different, done in different ways, 4 certainly than the clinical trial data analysis. And 5 they felt quite strongly that subjecting that "data 6 analysis" in quotes to this continuing review is 7 really one of the things that make IRB oversight of 8 social and behavioral research such a mismatch in 9 their eyes. 10 A second point, I guess I'd make is that 11 I returned to this slide just a couple slides earlier, 12 this list here of possible stopping points, what was 13 drafted at midnight on Saturday was put in a -- I 14 think a logical sequence in terms of chronology. So, 15 nobody is proposing that we stop it as soon as the 16 last subject enrolled, after interventions are 17 complete, too soon we feel. After data collection is 18 complete, this is where we're comfortable drawing the 19 line recognizing that below that line so data 20 analysis, publication, again, across the scope of 21 research that we're talking about, everything, is a 22 reasonable place. 60 1 And there was a lot of discussion amongst 2 the subcommittee, and we have three of the members 3 sitting at your table today and they can add to this 4 discussion, there was a lot of question about where to 5 draw that line. It was felt that after data 6 collection is over, but not necessarily after data 7 analysis is over was our comfort zone, if you will, 8 and that's why it was brought forward in the present 9 manner. 10 CHAIR PRENTICE: Dr. Powe? 11 MEMBER POWE: Yeah, one situation that 12 commonly comes up in my experience has been the issue 13 where data finished being collected and an 14 investigator has a colleague, another investigator, a 15 trainee that would like to use that data and actually 16 might like to use identifiable data, in fact, let's 17 say, you know, medical record reviews or whatever, 18 that is related to the parent study that falls under 19 the aims of the parent study, but pursues a slightly 20 different hypothesis. The universities are 21 responsible for assuring that any individual who 22 engages in human subjects research is certifiable and 61 1 has undergone training. 2 And so when that new investigator comes 3 in, that's usually reported to the IRB to make sure 4 that, in fact, they have had training. 5 So my question here or if there is a 6 dilemma here, if continuing review ends and there are 7 new investigators who join and then want to explore 8 new issues, would they have to submit a whole IRB 9 proposal -- 10 CHAIR PRENTICE: Yes. 11 MEMBER POWE: -- to undertake that 12 investigation -- 13 CHAIR PRENTICE: It would be a new 14 proposal. 15 MEMBER POWE: -- as opposed to falling 16 under the original one and this could be for studies 17 that go on for, you know, years and years. 18 MEMBER NELSON: Excellent question and I 19 think many of us recognize that scenario. I think the 20 operative phrases in your words as you've described 21 that were new issues, new hypothesis and one of the 22 reasons we tried to incorporate that and maybe we 62 1 haven't gone quite far enough but so as described in 2 the approved study plan or protocol, the idea was that 3 the investigators -- what we're closing the books on, 4 if you will, staffing IRB oversight is as described, 5 as -- they've done what they said they were going to 6 do. Now, if somebody new comes along, a trainee and 7 they're exploring new avenues, that would -- yes, that 8 would require -- we wouldn't want that to occur 9 without IRB oversight. 10 MEMBER POWE: Right. 11 MEMBER NELSON: And I suppose it could be 12 up to the local IRBs whether they had -- now we're 13 back to logistical issues. It would not be out of the 14 question that an IRB could find itself comfortable or 15 doable to reopen the former study adding on a new sub- 16 investigator, assuring training, et cetera, and 17 allowing an amendment, but I could hear other people 18 arguing that by the time you do that, maybe that 19 person should have just brought forth a concrete study 20 on its own merits recognizing that the data came from 21 a pre-existing study. 22 MEMBER POWE: Yeah, and I think a study 63 1 that was very active and a number of investigators 2 wanted to use that data, my concern was would it then 3 trigger, you know, 20 new full IRB reviews so the 4 balance of what the intention of this would be to 5 reduce burden would actually increase burden. 6 MEMBER NELSON: I think -- do you want to 7 respond? That's a good point. That could happen. 8 MEMBER POWE: I don't know -- it's a 9 dilemma. 10 CHAIR PRENTICE: Right, it is. I think 11 Susan is next and then Ada Sue. 12 MEMBER KORNETSKY: And serving on the 13 subcommittee for this and struggling between the way 14 that OHRP has interpreted it and understanding, the 15 way that I personally, you know, feel comfortable with 16 this is that when we review our research protocol and 17 talk about data analysis, part of that data analysis 18 usually includes that identifiers are removed, they're 19 kept separately and the data analysis, you know, 20 continues. So I don't -- my own feeling is that as 21 long as the IRB in the initial protocol reviews and 22 approves that, that as long as that continues to 64 1 happen that we can stop it after the data collection. 2 Now, if an investigator were to then say - 3 - we would have to educate them and then they'd say, 4 okay, now they forgot a data point and they need to go 5 back, then that would need to come back to the IRB. 6 So I'm almost looking at this as sort of a -- that 7 we're looking in the future and approving a plan for 8 data analysis which should include that identifiers 9 shouldn't be on the primary data, and that as long as 10 that is followed, there's no reason that the IRB -- 11 there's nothing additional that the IRB is going to 12 add. 13 However, if anything changes, even to the 14 point of having to go say, "Whoops, you know, we have 15 to go back to 10 subjects because we didn't know that 16 this data" -- you know, we want this data analysis 17 point, this data point", then that would be a need to 18 come back to the IRB. So that's the way I've tried to 19 sort of pull together the concerns of OHRP, what the 20 regulations say plus sort of what's really protected 21 for the research subject. 22 MEMBER FISHER: Building upon what Susan 65 1 said, I think that's what's missing from the language 2 of the recommendation. That is that -- I don't know 3 how to put it in, but that once the removal of 4 identifiers from the data occurs, because that's true, 5 in much research you don't need the identifiers any 6 more when the data collection is truly over, even if 7 you're doing longitudinal research. And then it could 8 be that what -- so if the limit was once the 9 identifiers have been removed, then the -- for the 10 first part, then the obligation on the investigator 11 is, is that they must inform the IRB if for some 12 reason there is a need to go back to the identifiers. 13 But I also think one would then have to 14 address what happens, are you going to be keeping a 15 subject code book in terms of subject numbers and 16 identifiers and how that would be considered within 17 the mix. 18 MEMBER NELSON: That's an interesting or 19 I think a good point conceptually at least but I 20 think, again, viewing the scope of all research that 21 we're talking about, if we made it a requirement that 22 data were coded or stripped of identifiers and that 66 1 link broken, now we've moved truly to not human 2 subjects research based on OHRP's August, 2004 3 clarification. So if that was the case, if you have 4 in your possession data that are de-identified or 5 coded in such a manner that the current investigator 6 does not have access, the -- Dr. Powe's downstream 7 trainee who is going to do a secondary data analysis 8 if you will, that no longer in the eyes of OHRP 9 guidance and in the processing the procedures of many 10 IRBs is no longer human subjects research requiring 11 IRB review. 12 MEMBER KORNETSKY: I'm not suggesting that 13 the link be broken. It's that the link not be used. 14 I mean, that's the way I'm looking at it. The link 15 maintains -- the code book maintains, it's there but 16 for purposes, you know, the way I'm interpreting -- 17 this is the way I would think about it and maybe I'm 18 interpreting it wrong, is that once you have -- you've 19 collected your data and you're ready to analyze it, 20 there's really no need that someone need -- although 21 the link is there, there's really no need that anyone 22 as to go and look at it. If for some reason they have 67 1 to go and look at it, then that is where I'm thinking 2 it needs to come back to the IRB but as long as 3 there's a plan that the analysis is ongoing. 4 MEMBER NELSON: And, indeed, I think that 5 already happens in a lot of settings. Maybe the 6 question is could we impose that on all research which 7 is really what we're aiming at with a recommendation 8 of this nature. 9 CHAIR PRENTICE: You know, I like this 10 recommendation, but I want to raise a couple of issues 11 for discussion. One is the definition of research 12 which is a systematic investigation, et cetera, et 13 cetera, which obviously includes data analysis. 14 That's what a systematic investigation is. And in 15 some protocols, it is the investigator that performs 16 the data analysis and in other protocols like co-op 17 studies, or pharmaceutical sponsored studies, the data 18 is submitted to the sponsor and the investigator does 19 not perform any analysis. 20 So I'm wondering what kind of impediment 21 it would be to change this to say when data collection 22 and analysis for research purposes is complete as and 68 1 I emphasize the word "as" described in the approved 2 study plan protocol. And I know that, Dan, I think 3 you made comments about, you know, sometimes the data 4 analysis takes years. 5 Well, I would contend that in the vast 6 majority of research projects it doesn't take years to 7 analyze the data according to the plan described in 8 the protocol because investigators are not going to 9 spend five years working on a publication that results 10 from the data that they gathered in an IRB approved 11 research project. 12 So what do you think the downside would be 13 to insert the word "and analysis as described in the 14 approved protocol", because sometimes there is going 15 to be no data analysis described at all. It's not 16 there because the investigator is not doing it. And 17 sometimes they are doing it and if you -- that would 18 be consistent with OHRP's interpretation of what a 19 systematic investigation means. I mean, I think 20 that's where Mike was going when he said, it's got to 21 be not only collection of the data but also the 22 analysis of the data assuming that the investigator is 69 1 the one who's doing that. 2 So I'm just throwing that out for 3 discussion. 4 MEMBER GYI: Ernie, I think that that gets 5 us in the right direction but the concern that I have 6 and a question to you was we're now looking to help 7 bridge the OHRP 45 CFR 46 language and part of the 8 difference interpretation-wise in 45 and 21, if we say 9 completion of data analysis, then that hamstrings the 10 IRBs that are overseeing FDA regulated products 11 because then you'd have to provide that oversight 12 until the sponsors have finished with their analysis. 13 CHAIR PRENTICE: No, no, it's -- the key 14 phrase in your recommendation is "as described in the 15 protocol". You're not going to have data analysis 16 described in the sponsor's protocol, because you know, 17 the investigator's not doing it. 18 MEMBER GYI: No, but you do though. I 19 mean, many times you do have data analysis that's 20 described in sponsored -- commercially sponsored 21 research protocol and there are people like Dr. Powe 22 here who may want to address that but it may take 70 1 years for the data base to be cleaned up and locked 2 and then the data analyzed. 3 MEMBER SELWITZ: I'm getting confused. 4 Can I ask for a clarification, Ernie, because I 5 thought I agreed with Ernie but I want to make sure I 6 understand. I thought you weren't -- I thought you 7 were proposing that we do stop after the data 8 collection, just as proposed here. You were just 9 proposing that the analysis has to be the analysis 10 that was originally approved in the protocol, which I 11 would agree with that. I mean, I think that is a 12 condition, but you weren't changing the line where we 13 would -- you were leaving the line, right, Ernie, 14 after -- before data analysis. You're just saying 15 that the data analysis that's going to be done, has to 16 be that which we approved and rather than going in a 17 new direction. 18 CHAIR PRENTICE: Well, I don't know. I 19 know what you're saying. I guess I'm trying to get a 20 handle on -- trying to make a bridge between this 21 recommendation and what I understand OHRP's position 22 to be. And I think that they feel like systematic 71 1 investigation includes data analysis. The PI is 2 responsible for the data analysis and therefore it's 3 got to be subject to IRB review and this 4 recommendation may be contrary to OHRP, interpretation 5 of the regulation which I can understand, you know, 6 the rationale for that interpretation. 7 I don't have too much of a problem unless 8 someone wants to enlighten me, when the PI -- the PI, 9 let's take an investigator initiated study. They're 10 responsible for the analysis of the data, okay. I can 11 understand perhaps a rationale for saying, "Okay, 12 continuing review under those circumstances is 13 warranted. On the other hand, once you've collected 14 the data it's out of your hands, like in a co-op study 15 or a typical pharmaceutical sponsor stud, it's out of 16 your hands. Then you're no longer performing a 17 systematic investigation. 18 The systematic investigation is basically 19 ended with the data collection. So obviously, there's 20 no reason for an IRB to conduct continuing review 21 under those circumstances. So what I'm trying to get 22 a handle on, is there a down side in terms of 72 1 additional undue burden on IRBs to say, all right if 2 the PI is responsible for the data analysis, then it's 3 still subject to continuing review. That's what I 4 want to get on the table, because I heard your comment 5 about years and years and years, but I'm trying to 6 understand under what circumstances a particular 7 research project, as described, as described, the 8 initially approved IRB protocol would take years and 9 years and years. I can understand Neil's comment 10 about utilizing existing data and having different 11 hypotheses. You're going down different tracks, okay, 12 that's an amendment of a protocol or it's a new 13 protocol. 14 But you're talking about years and years 15 and years and I don't know any investigators who spend 16 years and years and years analyzing the originally 17 collected data for the same purpose described in the 18 project. 19 MEMBER SELWITZ: All the time, I mean, I 20 just have to say, I wish I could say that 21 investigators don't and I think -- I unfortunately was 22 not at the subcommittee meeting but I read the minutes 73 1 and I read this recommendation and I'm extremely 2 supportive of it from the perspective that I think 3 it's ethically viable, I mean, appropriate plus I 4 understand what's happening in the field. If we don't 5 find a way to take off our plate those issues that 6 really aren't, as Dan says, value added, we're going 7 to continue to drown to the point that we're 8 ineffective on those things that aren't terribly 9 important. So let me just be clear. 10 I support this. I also fully understand 11 and was hoping see, you wouldn't bring that question 12 up, the regulatory problems that this recommendation 13 will invariably, if it goes forward, be left with 14 OHRP to sort out, but I don't see -- for me this 15 recommendation, the value is for my single 16 investigators in the social science area, in the 17 education area, that simply do not get around to 18 analyzing their data and keep it for years and years 19 and years. 20 I also think your point about the new 21 protocol and the change, I think that's very 22 important, that it be very clear that new comes in, 74 1 but see, that's my current policy. So in a way, what 2 you're proposing, this is not -- would not be at least 3 in my system an increased burden because we bring it 4 in as new. Do you understand? When they switch to 5 something new and they're talking about a different 6 kind of analysis, and they've got identifiers, we 7 bring it in as a new protocol. So then at least I my 8 system, the issue you raise, sir, this wouldn't be an 9 additional burden. 10 All right, but if you cut that -- you 11 know, I mean, if you cut that line before or after 12 data analysis for those single investigators, then 13 this will not be value added. 14 CHAIR PRENTICE: All right, I have a 15 question. I want to go back to the regulatory 16 considerations that this recommendation may 17 precipitate and I guess I'll have to ask OHRP to 18 comment on that. The definition of research is the 19 definition of research. It should apply equally 20 across all situations, so to go back to the two 21 examples that I gave earlier; one, you have a co-op 22 study which involves a systematic investigation 75 1 designed to contribute or develop generalizable 2 knowledge that obviously must necessitate analysis of 3 the data which is not carried out by the PI. It's 4 not, okay? 5 Under the current OHRP guidance, as I 6 understand it, you say, well, all right if the IRB, if 7 the PI is not involved in the analysis of the data, 8 therefore, you don't -- the IRB does not have to do 9 continuing review. That's what you said and I think 10 you said that at a subcommittee meeting, Mike, right? 11 Okay. 12 On the other hand, if the PI is 13 responsible for analysis of the data, then it's the 14 position of OHRP that research is continuing; 15 therefore, you need to have IRB review. Now, in terms 16 of consistency of human subject protection, across 17 institutions and across types of studies, how does 18 OHRP justify having what is more or less two different 19 standards where the IRB is involved in one 20 circumstance but the IRB is out of the loop on the 21 other circumstance because of a technicality in that 22 they don't have the data for analysis. 76 1 To me, I understand the definition of 2 research but this doesn't make a lot of sense when we 3 apply it that way. So would you care to respond? 4 DR. CAROME: I don't think there's an 5 inconsistency. It depends upon which sites or 6 institutions continue to be or remain engaged in the 7 human subject research activity. So if you have a 8 multi-center study in which 10 sites have enrolled 9 people interacted with them, collected identifiable 10 private information and now they're sending all that 11 information to say a coordinating site at another 12 university or to a sponsor or to an oncology group 13 site. Only that site or institution that continues to 14 do the analysis of the identifiable information would 15 continue to be engaged in the research and it's that 16 site, if the research is subject to our regulations, 17 that we would say needs to continue to have continuing 18 review done at least annually. 19 The other sites that are down there, were 20 they sent the information and the data and they have 21 no other involvement in the research. Well, they're 22 no longer engaged and they don't need to have their 77 1 IRB continue to have continuing review because they're 2 not engaged in the research any more. So we don't see 3 an inconsistency there. It's just a matter of who 4 remains engaged in the human subject research and it's 5 that site or institution that continues to need 6 continuing review. 7 And just because I'm highlighting the 8 differences between the proposed recommendation and 9 what our current view is, you should not interpret 10 that as we're rejecting the recommendation a priori at 11 this point. I'm just -- you asked me to clarify what 12 our position is and how it's contrasted. It doesn't 13 mean because there's a contrast we are a priori 14 rejecting the recommendation. 15 CHAIR PRENTICE: Do you have the power to 16 reject our recommendations? 17 (Laughter) 18 MALE PARTICIPANT: He does have a uniform 19 on. 20 MEMBER SELWITZ: Ultimately, Ernie, 21 ultimately, they do. 22 (Laughter) 78 1 MEMBER GYI: Ernie, there's just, if we 2 stay with that notion of being engaged defines 3 systematic investigation, if we use that as an 4 assumption, then this definition should work, right, 5 because then if the investigator is no longer engaged 6 in systematic investigation because the data has been 7 submitted elsewhere, then this definition should work. 8 MEMBER SELWITZ: It does for biomedical 9 multi-center trials. It does not work for those of us 10 that are involved with social science, single 11 investigator trials. 12 MEMBER NELSON: That's right. 13 CHAIR PRENTICE: Dr. Powell, do you have 14 something? 15 MEMBER POWELL: Yeah, I was going to 16 comment on a couple of things. And that is with 17 industry sponsored types of clinical trials, perhaps 18 two cases where in the protocol we may state that 19 we're going to continue to conduct follow-up on people 20 that were participating in trials. For instance, we 21 often have a situation where we include women of 22 child-bearing potential that requires us to continue 79 1 to follow up on if someone, for instance, becomes 2 pregnant during the course of a study, which it says 3 in the protocol you're going to continue to follow up 4 and that may be for quite a long period of time. 5 The other place in -- that's becoming more 6 frequent in the types of trials we do for industry are 7 biomarker analysis, meaning we are including 8 statements in our protocols that say we are collecting 9 samples, some type of samples for biomarker analysis 10 that we, in fact, may not totally describe in the 11 protocol except that we're going to do it from time to 12 time and that may be genetic analysis, other kinds of 13 things that may take years and years and years to 14 really do, and I wonder if this -- how that fits upon 15 -- under your plan for discontinuing -- stop the 16 continuing resolution at the completion of analysis 17 when that analysis may, in fact, be triggered by other 18 things that are within the protocol that you can't 19 anticipate. 20 CHAIR PRENTICE: I assume that based upon 21 what you said, there would be a description in the 22 protocol about, you know, looking at biomarkers at 80 1 some point in time, so I would say that that's further 2 collection of research data, so it would be subject to 3 continuing review because it's in the protocol. Your 4 data is not completed yet. It's not collected yet. 5 And it might have to go on for years and years and 6 years. That's the way I would interpret this 7 definition applied to your scenario. 8 On the other hand, routine clinical 9 follow-up data not obtained for research purposes 10 would not factor in. There would be a difference. So 11 you could follow, for example, a cohort of patients 12 for life as long as you're not collecting data for 13 research purposes. It's considered routine clinical 14 follow-up. To be doing anything for research 15 purposes, it would be research subject to continuing 16 review. Okay, questions, discussions and see if we 17 can move on. 18 MEMBER FISHER: I was just trying to 19 figure out how to reword it to try to add what you 20 were talking about as well as the issue of 21 confidentiality. I know that the one thing we can't 22 solve is the fact that if the data analysis as planned 81 1 continues for a long time, that seems, you know, is 2 going to be a remaining issue, but if it said 3 something like OHRP should clarify its guidance on the 4 required duration of continued review, continued 5 review may end when all research interventions and 6 interactions with subjects are over, and data 7 collection and analyses for research purposes is 8 complete as described in the approved study plan 9 protocol that the research cycle which the IRB has 10 oversight and confidentiality protections remain in 11 place. 12 Now, the only thing that doesn't answer 13 is, is that if they continue to analyze the data as 14 planned, it has to continue to be reviewed. 15 MEMBER SELWITZ: And if you do that, then 16 for all practical purposes, this is no different -- 17 we're not making a recommendation. I mean, for all 18 practical purposes that's what we have now. 19 CHAIR PRENTICE: That's correct. It's no 20 different. All right, look, we've spent a lot of time 21 at the last meeting discussing this. We spent more 22 time on this one recommendation last time and it looks 82 1 like we're repeating what we did before. I think we 2 need to, you know, get off the mark and decide what we 3 want to do. I would certainly entertain a motion with 4 regard to this recommendation. I don't' see it being 5 modified. I mean, you either go one way or the other. 6 Well? Ada Sue? 7 MEMBER SELWITZ: I move that we accept it. 8 CHAIR PRENTICE: All right, you all know 9 parliamentary procedure. How long does it take to 10 get a second on something here? 11 MEMBER KORNETSKY: I'm going to second it. 12 CHAIR PRENTICE: All right, very good. 13 Moved and seconded. Further discussion? Hearing 14 none, I want a show of hands. All those in favor of 15 the motion raise your hands. We're going to need to 16 get a count here. We've got one, two, three, four, 17 five. And we have -- yeah, we do have a quorum and 18 it's a majority of the quorum, yeah. Okay, so yes, 19 the motion passes. 20 All right, let's see. We have some more 21 time, would you like to try to get through your next 22 recommendation? 83 1 MEMBER NELSON: Yes, that would be great. 2 Thank you for that. Despite the time, and Ernie's 3 right, we spent time last time and we spent time again 4 today, but in our minds this is a big ticket item. 5 There's a reason, I think this was number one and we 6 appreciate the good discussion that just occurred. 7 This is just what we need to have happen. 8 So question number 2, are there 9 circumstances when continuing review can appropriately 10 begin to be conducted less often than once per year? 11 I'll spend comparatively less time painting the 12 backdrop for this but one part of this historical 13 context is to share with you the preamble to the 14 original regulations at that time Department of 15 Health, Education and Welfare, that came out in 1981 16 which stated that, "The precise procedure adopted by 17 the IRB for continuing review without unnecessarily 18 hindering research," that was part of their thinking 19 at that time, "should be left to the discretion of the 20 IRB". 21 The next sentence of the same preamble 22 stated that, "Reporting requirements may vary from a 84 1 simple annual notification in the case of research 2 involving little to no risk to more frequent reporting 3 or for clinical trials the IRB may require a special 4 mechanism to carry out data and safety monitoring 5 functions", and in a previous presentation to you, I 6 think Gary Chadwick, our task force chair, asked 7 somewhat rhetorically if we had moved quite a ways 8 beyond the notions that were embedded in that 9 preamble, the simple annual notification, are we 10 making adequate use of the special mechanisms to carry 11 out data safety monitoring functions which were 12 anticipated way back then but as we know, IRBs are now 13 trying to remove themselves from roles that might 14 better be covered by DSMBs and other bodies that are 15 in place to do these analyses. 16 Just to repeat the reference that I shared 17 before, the discrete reference in the regs to 18 continual review. "An IRB shall conduct continual 19 review of research covered by this policy at intervals 20 appropriate to the degree of risk but not less than 21 once per year". There is no regulatory basis for the 22 current content of the continual review process 85 1 itself. Both Health and Human Services and FDA have 2 pointed IRBs to Section 111, the criteria for IRB 3 approval that must be satisfied whether full board 4 review or expedited review, whether initial review or 5 continuing review. 6 The subcommittee feels that the 7 requirement for at a minimum annual review limits the 8 flexibility of IRBs to employ appropriate procedures 9 and criteria for ongoing review. For minimal risk 10 research, the requirement for annual review is neither 11 related to nor, in quotes, "appropriate to the degree 12 of risk". So this is, I think of all the changes 13 we'll put forth, the first one and the only one today 14 that would actually require a regulatory change as 15 opposed to a change in interpretation and guidance. 16 But the recommendation is that OHRP should 17 issue an advanced notice of proposed rule making to 18 seek comments regarding changing Section 46.109(e) to 19 allow IRBs latitude in setting review dates beyond one 20 year and the subcommittee would propose on further 21 discussions beyond what you saw last time, that there 22 be a cap of two years put in place, recognizing that 86 1 open-ended terms are not desirable, for minimal risk 2 studies but potentially for other studies as well". 3 This recommendation received a lot of 4 further discussion and debate and I think I would 5 stress the point that we're not rewriting the 6 regulation with this recommendation. We are 7 suggesting that a call be put out for comments from 8 the research community, from IRBs, from investigators, 9 from the public as to the desirability of the current 10 limit for one year. 11 This recommendation actually comes in 12 three parts and perhaps let me go through the next two 13 slides and then we'll return to these and you can take 14 them in order if you wish or as a set. In this 15 advanced notice of proposed rule-making, OHRP should 16 also seek comments on the regulatory application of 17 46.111 to continual review and/or adding a new section 18 that would define simplified criteria and the 19 expectations for the content of continuing review 20 being based upon current risk level and this somewhat 21 addresses Celia's earlier point whether a streamline 22 process could be put in place at continual review 87 1 beyond what's currently practiced or replacing what's 2 currently practiced. 3 In the interim, OHRP should revise its 4 interpretation and develop new guidance to permit IRBs 5 to develop within their written procedures, policies 6 and procedures for the selective application of 7 Section 46.111 to continual review and we'll return to 8 this theme. We would like, to the extent possible, 9 that FDA guidance be harmonized with this so in this 10 case, that should, likewise be updated. 11 Those are our recommendations on question 12 number 2. 13 CHAIR PRENTICE: Okay, thank you. You 14 indicated, Dan, that you're not proposing changing the 15 regulations. You're proposing that OHRP seek comments 16 with regard to changing Section 46.109(e). And you're 17 also recommending that OHRP seek comments, presumably 18 at the same time, in order to define, simplify 19 criteria and the expectations for the content of 20 continuing review being based upon current risk level. 21 Then you go onto recommendation 2.3, which 22 basically states that in the interim and I don't know 88 1 how long that would be, before not only a proposed -- 2 advanced notice of proposed rule making would be 3 published in the Federal Register and comments 4 analyzed, it's obviously not going to be within a 5 matter of months. It will be longer than that. 6 But you're saying that OHRP should revise 7 its interpretation and develop new guidance to permit 8 IRBs to develop within their written procedures 9 policies for selective application of Section 46.111 10 to continuing review, so here's the thrust of my 11 question. On the one hand, it would appear in 12 recommendations 2.2 you want the public to comment on 13 this issue but for the period of the next couple of 14 years, you're suggesting that IRBs be permitted to 15 develop their own procedures. Is this in any way a 16 bit of a contradiction and could that create problems 17 ultimately or not? 18 MEMBER NELSON: A fair question. It's 19 clear to us and clear to all of you that the change in 20 timing of review extending beyond a minimum of one 21 year would be -- or a maximum of one year, sorry, 22 would require regulatory change. It's our sense that 89 1 the application of 111 in its full-blown glory, if you 2 will, to continual review, would not require the same 3 regulatory change but is based more on current 4 understanding and interpretation. 5 Now, whether we're trying to have our cake 6 and eat it too by asking for comment at the same time 7 we're starting to implement which is really your 8 question, Ernie, it could be interpreted that way. 9 Again, I think the sense is that the current 10 application of 46.111 in all its elements to continual 11 review is unnecessarily burdensome, it's not value 12 added and should be addressed sooner rather than 13 later. 14 Felix, go ahead. 15 MEMBER GYI: Part of the subcommittee's 16 discussion revolved around the preamble and 17 understanding of what the initial intention of 18 continuing review was meant to be. OHRP has, in its 19 guidance, said that the continuing review should carry 20 the full force, if you will, of initial review and I 21 think part of the discussion for the subcommittee 22 revolved around is there something we can do in the 90 1 interim while we're waiting to see if we can extend 2 the period from one to two years to see if we might be 3 able to get some flexibility regarding what the IRB 4 has to do at the continuing review period. I don't 5 know if that helps, Ernie. 6 CHAIR PRENTICE: Okay, let me adopt a 7 slightly different tact. I fully understand the need 8 for an advance notice of proposed rule making for 9 recommendations 2.1. With regard to recommendations 10 2.2, I certainly endorse the recommendation but I'm 11 also reminded that it took OPRR/OHRP a very long time 12 to issue an guidance on continuing review, despite the 13 fact that regulations never changed and we all know 14 that. It's been an evolving process. 15 Therefore, given that fact, is there even 16 a need for OHRP to go through the advanced notice of 17 proposed rule making rather than come up with a 18 response to simplified criteria based upon current 19 risk level which is certainly something we considered 20 anyways? That's inherent in the application of the 21 regulations and thereby, try to avoid the -- you know, 22 the couple years this is going to take to have any 91 1 changes actually made. So I guess that's what I would 2 like to throw out on the floor for discussion. 3 MEMBER NELSON: So to paraphrase, Ernie, 4 you're basically suggesting striking 2.2 and indeed, 5 these are, although they're related elements, they 6 could have been presented to you as truly two separate 7 recommendations, because one deals with the timing of 8 review, the other deals with the content or structure 9 or intensity, if you will, of that review. 10 CHAIR PRENTICE: Yeah, I guess what I'm 11 suggesting is that you go with 2.1 which is obviously, 12 distinct, separate, but in terms of 2.2 and 2.3, you 13 eliminate the need for the proposed -- advanced notice 14 of proposed rule making and you simply indicate to 15 OHRP, you know, you need to issue guidance that is 16 relevant to the nature and risk level of the project 17 as opposed to what is considered current guidance 18 which is very, very comprehensive which I think is 19 what we're saying is needlessly detailed and 20 comprehensive; is that not correct? 21 I'm just throwing that out for 22 consideration. Susan first. 92 1 MEMBER KORNETSKY: Yeah, I mean in 2 thinking about this trying to think about it with the 3 subcommittee, it makes sense, and then you think about 4 it months later and it's different. I tend to agree. 5 My concern would be if IRBs are given the latitude to 6 start doing this and then after that we get comments 7 and then it gets potentially tightened up on it, then 8 it's going to be looking like we're being very 9 restrictive and over-regulating. 10 CHAIR PRENTICE: That's where I'm going. 11 MEMBER KORNETSKY: So I mean, I think I do 12 have a concern with that. I do think OHRP probably 13 has it within their power to issue guidance about what 14 elements should be applied and I totally -- I think 15 thinking about this a second time, agree with trying 16 to separate the issues out, handling what we need to 17 by regulatory change and not making the -- you know, 18 not adding additional regulation where we don't need 19 it. And I think that's maybe what we would end up 20 doing here in thinking about this a second time. 21 CHAIR PRENTICE: Ada Sue? 22 MEMBER SELWITZ: I like this idea, Ernie. 93 1 I think it's a good idea. I think it's a good idea to 2 separate this out into two totally different 3 recommendations; one deals with the period of time, 4 the other deals with whether you apply 46.111 when you 5 do a continuing review and make them separate. And if 6 I understand you correctly, Ernie, what you're really 7 saying is the issue of 46.111 could be handled with 8 guidance. And I think it could be handled with 9 guidance and I'm a big proponent of continuing review 10 for particularly certain types of protocols and don't 11 think that 46.111 really serves as the best model when 12 you do continuing review. 13 You know, that there are some issues that 14 don't -- aren't, in fact, necessary and there are 15 other issues that you need to add that aren't in 16 46.111. So I love this idea, Ernie, of separating 17 them out, two separate recommendations and one of 18 which could be done by guidance, we hope. 19 CHAIR PRENTICE: Okay, we are five minutes 20 past the break. Do you want to try to bring 21 resolution to some of this? Let's take recommendation 22 2.1 first. I'll entertain a motion. Well, we can go 94 1 on a break. 2 MEMBER KORNETSKY: Let's get this done. 3 I motion that we accept recommendation 2.1. 4 MEMBER JONES: Seco