1 DEPARTMENT OF HEALTH AND HUMAN SERVICES SECRETARY'S ADVISORY COMMITTEE ON HUMAN RESEARCH PROTECTIONS MEETING TUESDAY FEBRUARY 1, 2005 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 MARK BARNES, J.D., L.L.M. Member CELIA B. FISHER, Ph.D. Member ROBERT G. HAUSER, M.D.,F.A.C.C. Member NANCY L. JONES, Ph.D. Member FELIX A. KHIN-MUANG-GYI,Pharm.D. Member SUSAN KORNETSKY, M.P.H. Member MARY L. POLAN,M.D.,Ph.D.,M.P.H. Member ADA SUE SELWITZ, M.A. Member SUSAN L. WEINER, Ph.D. Member Ex Officio Members PEG BARRATT, Ph.D. National Science Foundation HOWARD L. BRADLEY Social Security Administration SARAH CARR (for Amy Patterson) National Institutes of Health KATHRYN LYNN CATES, M.D. U.S. Department of Veterans Affairs ROGER CORTESI U.S. Environmental Protection Agency 2 Ex Officio Members (continued) PATTY DECOT U.S. Department of Defense HELENE DERAMOND U.S. Department of Education SALLY FLANZER, Ph.D. Agency for Healthcare Research and Quality SUZANNE GAYNOR, DrPH,RN,MBA U.S. Department of Housing and Urban Development ALSO PRESENT KELLEY BOOHER, Office of Human Research Protections MICHAEL CAROME, Office of Human Research Protections JULIE KANESHIRO, Office of Human Research Protections LAURA ODWAZNY, Office of General Counsel IVOR PRITCHARD, Office of Human Research Protections IRENE STITH-COLEMAN, Office of Human Research Protections 3 I-N-D-E-X OVERVIEW OF AGENDA: Ernest D. Prentice, Ph.D. . . . . . . . . . .4 ADDITIONAL BUSINESS: Ernest D. Prentice, Ph.D. . . . . . . . . . .6 ADVERSE EVENT ISSUES UPDATES AND DISCUSSION: Amy Patterson, M.D. . . . . . . . . . . . . .7 Jean-Louis Saillot. . . . . . . . . . . . . 21 Question and Answer Session . . . . . . . . 39 SUBPART A PRESENTATION: Dan Nelson. . . . . . . . . . . . . . . . . 83 Question and Answer Session . . . . . . . . 95 PUBLIC COMMENT: DISCUSSION OF COMPLIANCE OVERSIGHT ACTIVITIES: Kristina Borror . . . . . . . . . . . . . .144 Joanne Rhoads . . . . . . . . . . . . . . .165 Michael Klag. . . . . . . . . . . . . . . .187 Tom Puglisi . . . . . . . . . . . . . . . .201 Chris Pascal. . . . . . . . . . . . . . . .223 Question and Answer Session . . . . . . . .233 DISCUSSION OF COMMITTEE BUSINESS: Tom Adams . . . . . . . . . . . . . . . . .268 CONCLUSION: . . . . . . . . . . . . . . . . . . .277 4 1 P-R-O-C-E-E-D-I-N-G-S 2 8:39 a.m. 3 DR. PRENTICE: Good morning, everybody. 4 Welcome to the second day of the 2005 SACHRP meeting, 5 first of three. What I would like to do, as I 6 normally do, is provide an overview of what we're 7 going to be talking about today. 8 We'll devote some time to additional 9 business, if there is any. And then we're going to 10 move into a panel, which will address reporting of 11 adverse events. As you know, SACHRP has had an 12 interest in IRB review of adverse events for quite 13 some time. We had one panel presentation at, I 14 believe, the second meeting of SACHRP, and we also 15 discussed it at a subsequent meeting. So we're 16 delighted to have Dr. Patterson and Dr. Jean-Louis 17 Saillot to talk about some adverse event initiatives. 18 And I'll introduce the speakers in a moment. 19 We'll take a break after that, and then 20 we'll move into subpart A presentation and future 21 plans. As you know, this is a newly-appointed 22 subcommittee of SACHRP to look at the basic 5 1 protections afforded by 45 CFR 46, so the co-chairs, 2 Dr. Felix Gyi and Dan Nelson will be talking about 3 what they intend to do in terms of pursuing that 4 charge. 5 We'll have a period for public comment 6 after -þ something is wrong here. Lunch must be, 7 lunch should be there, I think. It's not at 2:00, 8 okay? So that's a typo in this particular slide. 9 Does it say 12? Oh, is that what's happening? I see. 10 Okay, all right. There you go. 12 to 1 lunch, and 11 then discussion of compliance oversight activities 12 will be between 1 and 3. And the reason why we've 13 chosen this particular topic is, as you know, the 14 first significant, I think, OPRR shutdown of an 15 institution's research program occurred in October of 16 1998, when Russ Presbyterian Saint Luke Medical Center 17 in Chicago had their assurance suspended. And after 18 that, there was a cascade of shutdowns by OPRR and 19 then, subsequently, OHRP. Things slowed down in 2001, 20 and I don't think we've had anything of significance 21 since Hopkins, but we think it's important to talk 22 about the current state of affairs of compliance 6 1 oversight activities by OHRP and FDA. 2 We've also asked some representatives 3 outside of the government to talk about their view of 4 compliance oversight: Dr. Michael Klag from Hopkins 5 and Tom Puglisi from PriceWaterhouse. We'll then take 6 a break, and we'll have a discussion of committee 7 business. And then we're going to try to adjourn at 8 4:00. 9 So is there any additional business that 10 you would like to raise as a consequence of 11 yesterday's discussion or anything else? Okay. Then 12 I'm going to move over there. I'm going to ask Amy 13 Patterson to come up first to present, and I'll 14 introduce you when I get back to the table. 15 MS. PATTERSON: Can you hear me? 16 DR. PRENTICE: Yes. Now, in the interest 17 of time, I am not going to read all of Dr. Patterson's 18 distinguished bio. Amy is the Director of the Office 19 of Biotechnology Activities, which is known by the 20 acronym OBA, in the Office of Science Policy at the 21 NIH. And as most of you, I think, know, OBA is the 22 focal point within the U.S. Department of Health and 7 1 Human Services for addressing scientific, medical, 2 ethical, legal, and social implications raised by 3 pioneering medical research, such as gene transfer 4 research, genetic testing, and xenotransplantation. 5 So welcome, Dr. Patterson. We're looking 6 forward to your presentation. 7 MS. PATTERSON: Thank you very much. It's 8 a pleasure to be here. In speaking to you this 9 morning, I'm actually not here wearing my NIH hat. 10 I'm here on behalf of my colleagues from several other 11 agencies to share with you an update on a fairly new 12 initiative, a federal adverse event taskforce, that 13 came into being, in large part, driven from concerns 14 expressed by this committee regarding how adverse 15 events are reported, analyzed, and ultimately 16 communicated. 17 I wanted to first take a step back and 18 talk about adverse events, why they're important. It 19 seems like a fundamental question to ask, but it's, I 20 think, particularly important as we examine how we 21 collect information, how we analyze it, and what we do 22 with it. Ideally, adverse events can be viewed as 8 1 lessons from nature. They not only give a signal to 2 us to attend to the safety of one or more particular 3 research participants, they, not infrequently, also 4 call for a change in the protocol, be it the 5 inclusion/exclusion criteria, the dosing, the 6 monitoring plan, and the informed consent. 7 Sometimes, if we are very lucky, they 8 offer insight into the mechanism of drug action or 9 even perhaps an avenue to a new therapeutic or 10 application. But I would posit that our current system 11 for the collection analysis and communication of 12 safety information presents some substantial hurdles 13 to realizing the knowledge that might be garnered from 14 adverse events. 15 First of all, we have variation in the 16 regulatory terms, medical vocabularies that are used 17 to report adverse events, and this really frustrates 18 comparison of data within a trial, across trials, and 19 certainly sharing of data and comparison of data 20 across agencies. The agencies have diverse policies 21 for reporting adverse events, timeframes, and the 22 content of these reports also complicates and slows 9 1 investigator communication of key information. And 2 then, finally, and I think of keen interest to this 3 committee, at the local level the volume of safety 4 reports received by IRBs, particularly in multi-center 5 trials, hampers the detection of important signals 6 against a very large background noise. 7 From our PIs' perspective, this is a portrait of the 8 AE problem. Our PI has to report adverse events to 9 their IRB; perhaps to other IRBs; their DSMB, if 10 there's one in place for that particular trial; to a 11 sponsor, if a commercial sponsor is collaborating or 12 engaged in the research. If they're using an 13 investigational drug, biologic, or device, they have 14 a reporting relationship with FDA. Depending on the 15 nature of the adverse event, they'll also be 16 reporting, perhaps, or their institution will to OHRP. 17 And depending upon their funding streams, they may 18 have one or more reporting relationships with NIH, VA, 19 DoD, etcetera. And it's not so much the fact that the 20 PI has to report to multiple entities as it is the 21 diversity of the timeframe, the scope, and the content 22 of these reports that creates a challenge for even the 10 1 most well-intentioned and seasoned of investigators 2 and institutions. 3 From the perspective of the IRB, this 4 might be viewed as the AE problem. IRBs are 5 confronted with a huge volume of information that's 6 particularly exacerbated in multi-center trials, where 7 an IRB may be getting reports from other sites about 8 AEs happening at those other sites, but they're not 9 provided with sufficient contextual clinical 10 information to make heads or tails of what's actually 11 happened. What is the import of that adverse event 12 from the other site for what's going on at their 13 institution? What is the incident of the event? Is 14 it one death out of five people enrolled in that dose 15 cohort, or is it one death out of a thousand people 16 enrolled in that dose cohort? 17 And it's very difficult for IRBs to make 18 heads or tails about the information and really 19 extract what the knowledge that they need to garner 20 out of it to make changes, if need be, in the protocol 21 at their institution. 22 There's considerable confusion out there, 11 1 in addition, among the IRB community about what do the 2 regulators require us to do? What, under regulation, 3 are we required to do with the information that's 4 emanating from other sites? 5 As Ernie already alluded, this committee 6 has visited these issues on prior occasions, and the 7 committee, in 2004, last summer, called on the federal 8 agencies to streamline and coordinate diverse 9 requirements for AE reporting. And, in particular, in 10 a letter addressed to the Secretary of Health and 11 Human Services, your committee recommended that OHRP 12 and FDA should promptly issue clear and consistent 13 joint guidance on IRB review of both internal and 14 external adverse event reports, which will best serve 15 to protect human subjects and effectively reduce 16 regulatory burden. 17 In large part, in response to these 18 concerns, as well as in response to the agency's 19 recognition that we have an obligation to respond to 20 the requests for clarity from the research and IRB 21 communities, a federal adverse event taskforce was 22 formed. It has a number of member agencies that are 12 1 listed here. It consists of senior leadership from 2 FDA, OHRP, AHRQ, DoD, VA, NIH, and CDC. 3 And our charge is to propose some very 4 specific means for promoting harmonized and 5 streamlined federal requirements for reporting, 6 analyzing, communicating safety information from 7 clinical research. And what I'd like to do over the 8 next couple of slides is give you an update on the 9 objectives that have been agreed upon by the agencies 10 for this taskforce and give you a status report on 11 where we are and invite your input. 12 Now, one of the principle barriers to 13 streamlining or harmonizing adverse event requirements 14 across the agencies and, indeed, clarifying what those 15 requirements are for the IRB communities and the 16 research community is that the agencies speak a 17 different language. We use different terminology. 18 And so the first and probably the most 19 fundamental objective agreed upon by the taskforce 20 members is that we aim to have the agencies speaking 21 the same language. For instance, all of you, I think, 22 are fairly aware that the term "adverse event" does 13 1 not appear in the OHRP regulation. It's not in the 2 lexicon. The term "unanticipated problem" is used. 3 And unanticipated problems can encompass many things, 4 including adverse events, but the term doesn't 5 encompass all adverse events. And what does 6 unanticipated mean anyway? And how does it differ 7 from the FDA terminology "unexpected?" Well, these 8 are all fairly basic and fundamental questions that 9 not infrequently come up. So one of the aims of this 10 objective is to reduce the variability in the terms 11 and definitions and, at bare minimum, provide a 12 translation between the terminology used by the 13 various agencies. 14 The taskforce has completed the first 15 draft of a comprehensive mapping of all the terms and 16 definitions used by the various member agencies for 17 reporting safety information, be those requirements 18 set forth in policy, guideline, or regulation. And I 19 want to show you a page. It's actually a portion of 20 a page from this mapping. Let's see if I can get the 21 pointer here to work. And I just want to show you 22 some attributes. I don't intend that you be able to 14 1 read every little detail on here, but I wanted to 2 point out a couple of things. 3 The mapping takes the terminology, 4 including like unexpected adverse drug experience, 5 unexpected adverse drug reaction, unexpected adverse 6 event, defines it, and then, on a portion of the table 7 you can't see over here, provides the source or the 8 agency that invokes that definition or sets forth that 9 definition. 10 And then in the center portion of the 11 table, it maps out whether the term or the requirement 12 is embedded in a particular institute within an agency 13 or unit within an agency. Is it the policy of an 14 entire agency or an entire department? Is it 15 regulatory guidance, something agreed to by multiple 16 regulatory entities, such as ICH-E2A? Or is it 17 actually rooted in regulation? 18 And what we begin to see is a graphical 19 display of where our differences are and what is the 20 least to most burdensome to affect change, ranging 21 from policy of a unit within a unit of a particular 22 agency all the way over to regulation. And one of the 15 1 initial foci of our efforts coming out of this mapping 2 is to identify those things that are embedded in 3 policy and guidance and, therefore, will be the 4 easiest to change. 5 Where we are with this particular task is 6 we've completed the table, the mapping. It's many 7 pages, as you might expect, because this is just a 8 tiny sampling of the various terms and policies that 9 apply to safety reporting to the federal agencies. 10 And the mapping is now being vetted by the various 11 agencies. What will ultimately come out of this is a 12 guidance, a translation between the terminologies that 13 the agencies use, a set of frequently asked questions 14 that answer the questions I posed at the very 15 beginning of this session of the talk, what does that 16 anticipated mean, how does it relate to the term 17 "unexpected?" 18 The second objective of the taskforce is 19 to develop a best practices blueprint for reporting 20 analysis and application of safety information. And 21 the agencies are taking a look, first at the federal 22 level and then, early this spring, we will be 16 1 convening a number of focus groups that will include 2 industry, the PIs, and the IRB community, and we'll be 3 looking at taking a fresh look at what safety 4 information do we collect at the federal level and the 5 local level, how is it analyzed, and how is that 6 information used and applied. And through this 7 method, we'll be taking a fresh look at what our 8 procedures and policies are, taking a fresh look at 9 what our reporting requirements are, and identifying 10 best practices, with an eye towards streamlining those 11 reporting requirements as appropriate. 12 One of the highest priority initiatives 13 that address these first two objectives, the alignment 14 of terminology and a reassessment and the streamlining 15 of our reporting requirements is what we hope will be 16 a transfederal guidance. And, again, it's a priority 17 initiative for 2005. OHRP has taken a major step 18 toward making this a reality by preparing a first 19 draft. This is a guidance that addresses adverse 20 event reporting for multi-center clinical trials, and 21 it aims to align the regulatory terminology used by 22 the agencies to clarify the reporting requirements for 17 1 review of adverse event reports from multi-center 2 trials and to outline criteria when adverse events 3 should be forwarded to other sites. 4 And I should just give a harbinger to the 5 upcoming talk by my colleague, Dr. Saillot, because 6 there are complimentary activities underway in 7 industry and academia that point to potential criteria 8 for determining when adverse events should be 9 forwarded from multi-sites to external AE should be 10 forwarded to an IRB. So we look forward to getting 11 your input, as this draft guidance becomes more mature 12 and ready for input. 13 The third objective is that we will have 14 one core AE report that PIs can send to multiple 15 federal agencies, and this we refer to as the Basal 16 Adverse Event Report, otherwise known as the BAER. 17 And it sets forth a baseline set of core medical 18 information that would be accepted by all agencies. 19 We recognize, however, that the agencies 20 have different purviews, and there may be some 21 additional data elements or fields that are needed for 22 them to carry out their respective missions. And so 18 1 there are additional tables of data elements that 2 allow for data collection to address those particular 3 needs. 4 We're not trying to reinvent the wheel 5 with this initiative. The BAER incorporates current 6 and evolving HHS standards for data transmission and 7 vocabularies. 8 And then the last objective is, I think, 9 the most difficult and complex of all, and it's 10 probably one of the most critically important. The 11 premise for it probably should go without stating, but 12 I'm going to say it anyway because it's one that we're 13 going to need a lot of insight and advice on. 14 The underlying premise for this objective 15 is that communicating the results of the analysis of 16 safety information is necessary, both to further 17 scientific knowledge and to enhance participant safety 18 and public trust. The rubber really hits the road 19 when we start to talk about what is going to be 20 communicated. Is it raw safety information? Is it 21 aggregated? When is it communicated? When it 22 happens? During the course of the trial? When the 19 1 trial is finished and the article is published? And 2 to whom is it communicated? The research community? 3 Individual research participants? The general public? 4 And how is it communicated? 5 And these questions have taken on 6 particular salience over the last several months, as 7 questions have arisen in the public sector about how 8 safety information is communicated during drug 9 development. So this objective, while conceptualized 10 late last summer/early fall, has really taken on 11 additional salience at the current time. And so our 12 group is exploring or will be exploring strategies for 13 promoting transparency about how we evaluate safety 14 information, communication, the knowledge of safety 15 information. 16 This is an overview of the work plan. 17 It's probably a little bit hard for you to see, but we 18 began last fall between July and October and have been 19 convening a series of focus groups with each of our 20 sibling agencies to work on these objectives: the 21 mapping out of our terminology, discerning where our 22 differences are, are they rooted in policy or 20 1 regulation, and which ones are most amenable to 2 guidance and streamlining in the short term, and which 3 will require more long-term efforts to align. 4 We've also begun to map out the data 5 requirements for the BAER, incorporating the different 6 needs of the distinct agencies. And this spring we 7 look forward to the participation of many of you here 8 in this room as we begin to put together focus groups 9 that look at the process at the local level, so we'll 10 have members from the IRB, the PI, and the industry 11 communities. 12 Our goal is admittedly an ambitious one. 13 We aim to have draft products circulating in the 14 agencies for review and vetting by the end of this 15 summer. 16 And I'll just close and say that much of 17 the emphasis to date has been on the reporting of 18 adverse event information, but the taskforce believes 19 that an equal, if not greater, effort needs to be made 20 on the analysis, synthesis, and communication of 21 safety information if we're going to do an adequate 22 job in human subjects' protections. And I'll close 21 1 there. 2 DR. PRENTICE: Thank you, Dr. Patterson. 3 I would like to hold questions until the end of Dr. 4 Saillot's presentation, and it's my pleasure to 5 introduce our next speaker. He is Jean-Louis Saillot. 6 He is currently the Vice President and head of global 7 pharmacovigilance at Schering-Plough Research 8 Institute. This position includes global 9 responsibility for all pharmacovigilance functions, 10 including global product safety, surveillance, 11 aggregate safety reporting, pharmacoepidemiology, and 12 clinical risk management. Dr. Saillot, welcome to 13 SACHRP. 14 DR. SAILLOT: Thank you very much, Ernie. 15 Let me thank SACHRP for providing me the opportunity 16 to come and present preliminary thoughts from the IRB 17 Sponsor Roundtable, and I will go over what this 18 roundtable is and the proposal that we have put 19 together. 20 Before going into the details of that 21 proposal, I would like to say that it dovetails pretty 22 nicely with the presentation that Dr. Patterson has 22 1 made. And the scope of this proposal, though, is much 2 smaller in terms of the target actions from the IRB 3 Sponsor Roundtable. 4 Going back to this forum, the IRB Sponsor 5 Roundtable, today, the thought is really to provide -- 6 the IRB Roundtable is a group of individuals from both 7 the IRB community and sponsors that have been formed 8 in 2003, forming two forums on HIPAA and clinical 9 research that draw the IRB and the sponsors together. 10 The two communities engaged in a productive dialogue 11 with a desire to continue the dialogue and extend it 12 to other issues of common interest. During this 13 interaction, there was a consensus that increased 14 communication between the two communities was valuable 15 in order to provide proposals to enhance protection of 16 the research subjects. 17 The roundtable was formed in 2004 and is 18 a small group of representatives from IRBs and 19 sponsors. It is independent from existing 20 organizations, such as PhRMA, PRIM&R, or ARENA. The 21 goal is to have equal replenishes from both 22 communities and is a small group, modest in size, to 23 1 ensure a lively discussion and practical effective 2 process and recommendations. 3 The mission of the roundtable is to take 4 the constructive communications between the sponsors 5 and IRBs on significant clinical research issues and, 6 where possible, propose practical strategies for 7 improving clinical-type processes and human subject 8 production and also to engage other effective 9 stakeholders in clinical research community to 10 facilitate a broader dialogue and consensus building. 11 Just to give a background on the 12 membership, roundtable participants include a number 13 of representatives from the IRB community sitting here 14 today. I'm not going to go over all the names. The 15 group is co-chaired by Dan Nelson of the University of 16 North Carolina and by Justin McCarthy from Pfizer. 17 I would like to go straight into the 18 definition of the problem which brought us here. A 19 lot of the issues related to adverse event reporting 20 were already covered by Dr. Patterson's presentation. 21 One of the things that became very clear during the 22 discussion of the IRB Roundtable is a consensus 24 1 position that the current reporting mechanism for 2 adverse events to IRBs in multi-center trials was 3 really including a volume of reports that made it very 4 difficult for a meaningful evaluation by IRBs of what 5 this information actually meant or actually what this 6 data actually meant. 7 In terms of the background behind this 8 issue, more and more in clinical studies are sponsored 9 by the pharmaceutical industry. The trials coming 10 from Phase II but possibly in Phase III are multi- 11 center trials, multi-center trials meaning trials that 12 are conducted throughout the United States but 13 extremely commonly throughout the world. And this 14 trend is very unlikely to be reverted because of the 15 actual medical and methodological needs for clinical 16 research that require now more and more patients to be 17 included in the studies for pharmaceutical products. 18 So some of the issues often at each site 19 is overseen by different IRBs. For example, it's not 20 uncommon that you have the IRB overseeing the sites at 21 the institution, but there's also a central IRB for a 22 country, like in the United States. There's also many 25 1 different IRBs or ethics review committees, as they're 2 called in Europe for example, and the volume of these 3 institutions basically is leading to some of the 4 issues that Dr. Patterson mentioned in her slide in 5 terms of multiple stakeholders that need to be 6 reported to and that need to take action on 7 information coming from safety information collected 8 from clinical trials. 9 So, obviously, the IRBs, part of their 10 mission to review the ongoing progress of the trial, 11 including the safety information from trials, receive 12 adverse events, and I will go through some definitions 13 that are typically unexpected and serious, as per the 14 IND regulations for example. Most of my presentation 15 will be heavily based on IND regulations because 16 that's what the pharmaceutical sponsors are mostly 17 dealing with. However, these problems are probably 18 relevant to other type of federal regulations for 19 other types of studies. 20 So the expedited adverse events, which 21 include possible unanticipated problems involving risk 22 to human subjects and others. However, as Dr. 26 1 Patterson mentioned, this is not a 100-percent match. 2 There is some overlap, but only overlap. It is not a 3 100-percent match. 4 One of the major drivers for the current 5 challenges is that the adverse events that occur both 6 at the institution and in institutions outside in the 7 other sites. Other research sites involved in the 8 study are sent to the IRBs. So the sheer number and 9 disaggregated nature of the reports make it difficult, 10 particularly for the IRBs, to effectively evaluate the 11 significance and the implication to study subjects. 12 Let me continue on some of the challenges. 13 One of the issues that has been discussed quite at 14 length during the discussion of the IRB Sponsor 15 Roundtable is that the current function and 16 infrastructure of IRBs really does not allow the IRBs 17 to function as safety oversight committees for multi- 18 site trials. They do not have access to the type of 19 relevant information necessary to evaluate large 20 volume of disaggregated adverse event reports in order 21 to put them in proper context. 22 The conclusion of that is that the signal 27 1 to noise ratio is unfavorably diminished by noise. 2 The existing regulatory framework was deadlocked 3 before multi-site trials were commonplace. This is an 4 observation that the IRB Sponsor Roundtable identified 5 in terms of the need now, maybe with the experience 6 gained with this large multi-center trial to revisit 7 guidance provided by the government. 8 We have already mentioned, Dr. Patterson 9 has already mentioned the differences in definition 10 that make handling adverse event reporting more 11 challenging across all the different types of research 12 activities. And a very important point that the IRB 13 Sponsor Roundtable felt was that, in order to really 14 affect a change from both the IRB and pharmaceutical 15 industry sponsors' perspective, guidance from the 16 agency was really going to be needed. 17 Some definitions, I won't go into many 18 detail. There are, indeed, multiple definitions of 19 adverse event, adverse experience. Obviously, within 20 the IND regulations, the international conference on 21 the harmonization or ICH guideline has also defined 22 adverse event or adverse experience with input from 28 1 the three different ICH regions, as well as WHO. 2 In summary, an adverse experience is 3 anything that is empowered that is being recorded in 4 a patient or subject involved in a clinical trial, 5 regardless of the causality of this adverse event with 6 the action being studied in the study. And that's 7 probably the shortest definition that one can make on 8 an adverse event. Obviously, it can be expanded in a 9 very large fashion and provide more details as to 10 exactly all the important points behind the adverse 11 event definition, such as whether an adverse event is 12 expected or unexpected, whether an adverse event is 13 serious or not. And I will not go over the details of 14 this definition. 15 Just a couple of definitions relevant to 16 this discussion is external adverse events. We mean 17 during this presentation that any adverse event which 18 occurs in an institution other than the one for which 19 the IRB is directly responsible as an external adverse 20 event. And an internal adverse event is a report that 21 occurs at the local institution for which the IRB 22 obviously needs to be made aware of. 29 1 An anticipated problem involving risk to 2 human subjects or others, I will not go over the 3 details of that definition. But there is a 4 significant overlap with the definition of adverse 5 event. That's the point we wanted to highlight. 6 In terms of other definitions, drug safety 7 monitoring boards or data monitoring committees, they 8 are not as well-established definitions for this 9 group, but this definition is taken out of recent FDA 10 guidance on the data monitoring committee. It's a 11 formal committee charged with reviewing the 12 accumulating data as the trial progresses to monitor 13 safety, effectiveness, and trial conduct issues, and 14 provide a set of recommendations to the study sponsor. 15 So let me now move on to provide this 16 group with preliminary thoughts on what the IRB 17 Sponsor Roundtable discussions were. I wanted to 18 highlight that these are preliminary thoughts, that 19 they have not yet been endorsed, but we thought that 20 sharing these thoughts with this group, particularly 21 in light of the initiatives that are being pursued, as 22 described by Dr. Patterson, and was hopefully going 30 1 to be helpful for directing guidance. 2 The goal of the new adverse event 3 reporting model would be to enhance the production of 4 human subject by ensuring that medically-relevant data 5 on adverse events is communicated to IRBs in a 6 meaningful way to their central role in protecting 7 human subjects. 8 The goal of the new model is to clearly 9 highlight the events that are medically relevant and 10 that are more likely to change the benefit/risk 11 relationship from other adverse events. The goal is 12 also to improve the oversights of adverse events and 13 human subject protection by promoting responsible and 14 effective AE reporting through a multi-party process: 15 IRBs, principle investigators, and sponsors, which 16 includes appropriate checks and balances. 17 Conceptually, this is not very different from what is 18 already in place, but you will see that we have some 19 suggestions, points, solutions that may help 20 improving, incrementally, the process. I think that's 21 also an important part. 22 The proposals that we are putting forth 31 1 today are incremental improvements. These are not 2 changes, but we believe that these incremental 3 improvements will bring a lot of value. So part of 4 the goal is also to reinforce the active participation 5 by all parties and to define potential unanticipated 6 problems. 7 So let me dive into the concepts of the 8 proposal. In the context of identifying unanticipated 9 problems involving risk to human subjects, the 10 proposal would be for investigators to identify 11 relevant external adverse event reports that require 12 notification to the IRBs. As opposed to sending all 13 expedited adverse events, as is currently done, this 14 communication of external adverse events to the IRBs 15 would be focused on reports that fulfill the following 16 proposed criteria. 17 If the report is leading to a study 18 protocol modification, a study protocol amendment, 19 like additional tests, additional visits, or anything 20 that basically has impacted the conduct of this study, 21 even in a modest way, reports leading to revision to 22 the informed consent, obviously any report that is 32 1 significant enough to lead to a revision to the 2 informed consent would qualify for this. These two 3 criteria are pretty straightforward and highlight 4 adverse events of concern that, you know, are easily 5 identified. 6 The third bullet is to leave some room for 7 interpretation as to anything else that may not lead 8 to a protocol modification or a revision to the 9 informed consent but that would be significant enough 10 that would be reflective of another major concern 11 impacting the study. Such report would be submitted 12 to the local IRBs. Of importance is to note that the 13 principle investigator's obligation to submit all 14 appropriate internal adverse events would not be 15 changed. 16 Going on to other modification of the 17 process, as sponsors should clearly identify to the 18 principal investigators those external adverse event 19 reports that meet the same criteria. That is, beyond 20 the fact that the investigator would have to do this 21 evaluation, a similar evaluation would be carried out 22 by the sponsors and the sponsors would highlight, when 33 1 communicating this adverse event to the sites would 2 highlight that this is a report that, in the opinion 3 of the sponsor, meets the criteria for medically- 4 relevant report. 5 Importantly, just to highlight this. I 6 think I mentioned that, earlier on, this is seen by 7 the IRB Sponsor Roundtable as being more a best 8 practice because, in practice, whenever an adverse 9 event leads to a modification of the protocol, this is 10 usually very well highlighted or at least should be 11 very well highlighted in the communication to both the 12 investigators and the IRB. 13 Some important aspects of the proposal: 14 the principal investigators should provide the 15 external adverse events to the IRBs that they believe 16 meet the criteria, even if the sponsor does not 17 identify them as such; and if the sponsor, on the 18 other hand, believed that this adverse event should 19 automatically be sent to the IRB, this should be 20 highlighted in the communication to the principal 21 investigators. 22 Now, an important point, which is the 34 1 second bullet, is whenever an investigator does the 2 review of the external adverse events and he or she 3 believes that this report should not be sent in an 4 expedited fashion to the IRB, the rationale and the 5 justification for not sending this adverse event 6 should be documented by the investigator. And I will 7 get back to that later on in terms of implementing a 8 system of checks and balances around these proposed 9 processes. 10 So let me move into some of these proposed 11 checks and balances. Obviously, as I just mentioned, 12 the documentation of the analysis of all adverse 13 events, including the external adverse events, needs 14 to be done. This is actually in line with current 15 regulations where, at least for the sponsor, the 16 evaluation of each adverse event is to be done and 17 documented. 18 The analysis and associated documentation, 19 particularly that done by the principal investigator, 20 would be available for audit by the IRB or a 21 designated compliance arm of the institution or for 22 both sponsor and site investigator by the FDA 35 1 inspection branch. Also, an important part of 2 changing the process would be the creation of a safety 3 communication plan. That should be justified by the 4 sponsor up-front at the time the study is initiated 5 and would be submitted as part of the study protocol. 6 Let me just expand on some of the thoughts 7 around this proposed safety communication plan. 8 Elements of the safety communication plan could 9 include the following: the proposed frequency of 10 submission of aggregate safety information or listings 11 of adverse events that would include all the adverse 12 events reported within the trial. And the frequency 13 of that submission could be quarterly, semi-annual, or 14 annual. 15 The frequency for the reporting would 16 obviously be dependent on the product being studied, 17 as well as the trial itself and the patient population 18 targeted in the trial. The proposed format for the 19 submission for the periodic quality assessment reports 20 covering all safety information relevant to the trial 21 and, as I mentioned, all expedited reports and other 22 relevant safety information would be described in the 36 1 safety communication plan. 2 There will also be a description and a 3 functioning of any DSMV or data monitoring committee, 4 as well as the method and frequency of updating the 5 investigators and the IRBs of the outcome of these 6 reviews. 7 And the safety communication plan should 8 be implemented, I mentioned that already, in a 9 flexible manner so that the specific needs of the 10 individual clinical trial or the investigational 11 product would be taken into consideration. The 12 thought would be that this safety communication plan 13 would be submitted together with the protocol so that 14 IRBs would have the opportunity to review this plan 15 and provide any comments to both the investigators and 16 the sponsors, as needed. 17 So this is basically, in a nutshell, the 18 high level proposal that the IRB roundtable has put 19 together. As I mentioned, I think I mentioned this is 20 a scope which is much smaller than the large 21 initiative and very needed initiative that Dr. 22 Patterson presented earlier on. But, really, these 37 1 are the thoughts that the roundtable put together. 2 Many of these can be considered more best practices 3 then really major changes. Our primary thought is 4 that these changes would already fit within the 5 existing regulatory framework and could be implemented 6 via guidance. However, this wouldn't have to be 7 validated by the federal agencies responsible for 8 guidance generation. 9 The next steps that we propose is to 10 continue to discuss and further refine the thinking on 11 this proposal. As I mentioned, it is a preliminary 12 proposal. It has not yet been endorsed, but I think 13 we felt comfortable enough to share these preliminary 14 thoughts with you to get the reaction of this group to 15 this proposal. 16 Continue to conduct outreach to the 17 interested stakeholders, including PhRMA, ARENA, and 18 PRIM&R, as appropriate to obtain feedback. And, as 19 appropriate, continue dialogue or presentation to this 20 group or interested government agencies, like the FDA 21 or OHRP. And I think it was clear from the earlier 22 part of the presentation we really encourage FDA and 38 1 OHRP to clearly articulate an official guidance best 2 practices for reporting external adverse events in 3 multi-site clinical trials. 4 The thought of the roundtable, again, 5 representing both, and I'm here representing not only, 6 or trying at least to represent not only the 7 pharmaceutical industry thoughts within the roundtable 8 but also the IRB thoughts, would be that if a more 9 workable AE reporting model for multi-site trials can 10 be put in place, the IRBs will be able to more 11 effectively evaluate the benefit/risk issues. 12 A very important point or goal of this 13 Proposal, I think I mentioned that already but let me 14 re-emphasize it, is that this proposal is really 15 geared at trying to address the signal-to-noise ratio 16 of all these aggregate reports on an ongoing basis. So 17 if that issue of signal-to-noise ratio can be 18 addressed, the thought is that both the investigators 19 and sponsors, the IRBs will be able to more 20 effectively evaluate the meaningfulness of this 21 information. 22 In the process of highlighting those 39 1 adverse events that are more relevant or that are felt 2 to be relevant, it is felt that both investigator and 3 sponsor will be better equipped to fulfill their 4 regulatory and ethical responsibilities. And the 5 bottom line is that the subject protection will be 6 enhanced. 7 And that's the end of my presentation. 8 Thank you very much. 9 DR. PRENTICE: Thank you, Dr. Saillot. 10 Amy and Jean-Louis, would you assemble up at the 11 table, please? Asserting Chair's prerogative, I have 12 about 25 questions here. Since we have 45 minutes, we 13 have enough time, I think, to address all of them. 14 Amy, looking at your time schedule, to 15 develop a draft work product. You indicate in June of 16 2005 that you'll have a draft product ready. Once 17 you've done that, I assume it has to go through 18 various layers of review. How long do you think it 19 might ultimately take to actually implement whatever 20 guidance recommendations that your committee develops? 21 MS. PATTERSON: Well, I think it's always 22 reasonable, and that's what we're trying to do here. 40 1 That said, Ernie, I think that the efforts of the 2 taskforce have both short-term and long-term outcomes. 3 Some of the objectives that I described are, I think, 4 inherently longer term things that will be realized, 5 particularly objective four. The more short-term 6 outcomes and I think ones of greatest and most intense 7 interest at the moment to this committee include the 8 transfederal guidance on AE reporting in multi-center 9 trials. And as I mentioned, OHRP has prepared a 10 working draft some months ago that the agencies have 11 reviewed and provided comments on and we look forward 12 to some time this year, in 2005, and I'll invite Mike, 13 if he wants to chime in here, but that's a front- 14 burner priority for this year to at least get out in 15 draft form. Mike, did you want to comment on that, or 16 would you agree? 17 DR. CAROME: I agree. 18 MS. PATTERSON: And some of the other 19 things, I think you can appreciate the mapping of the 20 terms and definitions and where our differences are I 21 think is very realistically to be accomplished by this 22 summer and provides us then with a blueprint for 41 1 moving forward on both, again, some shorter-term 2 projects and doing away with differences where they 3 are a hindrance to compliance and clarity and 4 translating those differences when they are meaningful 5 with respect to our different agency purviews. 6 That same blueprint provides us also with 7 some longer-term objectives. If there are differences 8 rooted in regulation or regulatory guidance, it may 9 take longer to streamline and clarify. So I think 10 there are multiple streams, some of them very short 11 term, deliverable this year. We hope that some are 12 deliverable this summer and then others that I think 13 that are going to require patience and considerable 14 input and dialogue with the research and IRB community 15 to make sure we end up in the right place. 16 DR. PRENTICE: Okay. I know that you and 17 Jean-Louis have had some discussions and you were 18 aware prior to today of the proposed AE model, which 19 is more focused on IRB review than your more broader 20 presentation. How does that model presented by Jean- 21 Louis fit with the discussions that your committee is 22 having about IRB review? 42 1 MS. PATTERSON: Well, again, I will turn 2 to my colleagues in the room, Mike, Lynn, Sally, and 3 others, to jump in here, but I'll just open up the 4 dialogue by saying that we view on the efforts of the 5 roundtable as being highly complimentary in informing 6 our efforts. It's of particular relevance to the 7 draft guidance for AE reporting for multi-center 8 trials that's under development. As Jean-Louis and I 9 have discussed and as his presentation reflects, it's 10 going to be very important, as any proposed criteria 11 for triaging or determining what safety information is 12 referred from an external site to the responsible IRB 13 for a particular institution, that this really be 14 framed appropriately, that we are enhancing the 15 knowledge that can be garnered from the considerable 16 effort that goes into reporting AE information. This 17 is not an effort to keep information from IRBs. This 18 is not an effort to strip away human subjects' 19 protections. 20 So I think that we view the effort as 21 needed, one that we will learn from, and we're 22 certainly, as we look forward to our focus groups in 43 1 the spring with the IRB and industry and PI community, 2 will be in dialogue with the members of the roundtable 3 and others, as we craft the guidance that we hope to 4 put out this year on this issue. 5 Mike and Lynn or Sally, did you want to 6 pipe up? 7 DR. CATES: Yes. I think Amy has done a 8 really good job of summarizing both what we've done 9 and what we've planned to do. I think that she's 10 right, that this is a very important part of our 11 objective, too, which is to look very hard at workflow 12 issues with regard to how adverse events are reported 13 and analyzed. So I think the two initiatives are 14 going to dovetail beautifully, and we're delighted to 15 hear what you're doing. Thanks. 16 DR. PRENTICE: Jean-Louis, you indicated 17 that you don't think that regulations would have to be 18 revised to accommodate the proposed AE model, and I 19 would like Amy to address whether or not she agrees. 20 And the reason why I ask that question is the last 21 time I had a conversation with David LePay about this, 22 he indicated to me, and this was perhaps six months 44 1 ago, he indicated to me that he thought perhaps there 2 would have to be a revision of regulations, as opposed 3 to simply issuing guidance within the boundaries of 4 the current regulation. So I wondered if you could 5 comment on that. 6 MS. PATTERSON: Well, I wish David were 7 here to speak on this, and I will invite Michael, as 8 a member of the taskforce and representing the other 9 regulatory agency member of the taskforce, to also 10 comment on this. But I think insofar as the extant 11 regulations do empower the IRBs to request aggregate 12 information, analysis of information, the proposal is 13 consonant with extant regulations in that respect. 14 And, certainly, the proposal of best 15 practices and criteria and the development of the 16 safety communication plan are also consonant with the 17 existing regulations. Mike, do you want to -þ I don't 18 want to speak for FDA because we understand that they 19 have a rule on suspected adverse drug reactions that 20 is currently in play within the agency, and it's hard 21 for them to be definitive exactly what the structure 22 of that regulation will ultimately be. And so I feel 45 1 at a bit at a loss to speak definitively to the FDA 2 part of this. But Mike? 3 DR. CAROME: I think the proposal that the 4 IRB Sponsor Roundtable sort of drafted is very 5 consistent with what we heard over a year ago at 6 SACHRP when I think the City of Hope and Wash U 7 presented how they handled adverse events, in 8 particular external adverse events, and it closely 9 mimics the thought process of the IRB Sponsor 10 Roundtable. And at that meeting, OHRP said that that 11 is consistent with our view of how those adverse 12 events should be handled under our regulations. So 13 what we're hearing is consonant with our expectations, 14 and I think it's reflected in the draft guidance that 15 we're working on. 16 On the issue of having plans for 17 monitoring safety, I mean that's completely consistent 18 with the regulations. Section 111 lists what the IRB 19 must determine in order to prove research, and 20 requiring, when appropriate, an appropriate monitoring 21 plan is consistent with the existing regulations. So 22 we see no need for revision of regulations to 46 1 implement that type of plan. 2 DR. PRENTICE: Don't sit down yet. 3 MS. PATTERSON: Ernie, I also wanted to 4 interject this in a follow-up on something that Mike 5 said. The component of what Dr. Saillot has presented 6 in terms of the enhanced analysis of AEs that would be 7 provided by both PI and the sponsor is a very 8 important component of the proposal. The next step 9 that he presented in terms of outreach to PhRMA, to 10 ARENA, PRIM&R are going to be critical because that 11 component really isn't set forth in a regulation. 12 It's going to be an extra task or workload on the 13 research community and the industry community, and it 14 sounds like they're willing to step up to the plate to 15 be part of the solution here. I think that's a very 16 important part of the proposal that isn't restricted 17 in any way by the regulations. And if that alone were 18 done, that would be a major step forward. 19 DR. PRENTICE: The reason why I asked you 20 to remain at the microphone is that, as you know, any 21 unanticipated problem involving risk to the subject or 22 others must be promptly reported to OHRP. There's no 47 1 threshold level of risk in that definition in the 2 regulations. I'm curious to know, considering the 3 hundreds and hundreds and hundreds of clinical trials 4 we have ongoing in this country, how many adverse 5 events actually get reported to OHRP on a, I don't 6 know, whatever basis you want: annual basis, monthly 7 basis? 8 DR. CAROME: First of all, I clarify not 9 all adverse events in all trials need to be reported 10 to us. It's only the research we have purview over, 11 which is research conducted or supported by HHS or 12 conducted under an applicable assurance, and there is 13 a significant amount of research outside that 14 umbrella. But for the research we do have oversight 15 over, we probably get a few hundred reports per year. 16 It fluctuates from month to month, but that's probably 17 on average. 18 DR. PRENTICE: Does that suggest under- 19 reporting? 20 DR. CAROME: It may, although, as I said 21 previously to this committee, we believe that most 22 adverse events that occur in clinical trials are not 48 1 unanticipated problems because most things that go 2 wrong are expected. You know, in chemotherapy trials, 3 bone marrow suppression, infections, and even death 4 from those infections can fall within the realm of 5 anticipated problems, and no such events need to be 6 reported then, if that's the judgment of the 7 investigator, the sponsor, and/or the IRB. 8 DR. PRENTICE: Okay. And I've got two 9 final questions for Jean-Louis in particular. You 10 mentioned variable definitions in the various 11 regulations. The ICH has a definition of adverse 12 events and reporting requirements, which is different 13 than the IND definition, to some extent. There's no 14 causality that's included. 15 You know, obviously, research is conducted 16 on a global scale. Whatever we do in this country, 17 how will that impact the global reporting of adverse 18 events? 19 DR. SAILLOT: Thank you. Well, you know, 20 first let me clarify that the definitions between the 21 ICH definition and the IND definition are very, very 22 similar. Conceptually, they're actually the same. 49 1 The differences are in some of the terms that are 2 being used. The causality may be defined slightly 3 differently. In ICH, it talks about the association 4 in the IND regulation is defined slightly differently 5 in ICH, but the meaning actually is the same. So 6 slightly different words are being used, but the 7 meaning is the same. 8 I think that there are more differences, 9 and I don't want to speak out of turn because my 10 experience is mostly in the pharmaceutical product, 11 drug product, but there may be additional differences 12 with the devices that are a little bit more apart than 13 the differences between IND and ICH definitions. 14 But to go to your question, which is, if 15 we go down this path, as I understand it, as we go 16 down this path, will that impact multinational 17 research? I would venture to say no. And some, also, 18 food for thought, in terms of some of the proposal 19 behind the enhanced communication of safety is also 20 leveraging some new initiatives from other regions 21 within the world. You may know that in Europe there 22 was recently the implementation of a clinical trial 50 1 directive to implement into law good clinical 2 practice, very similar to the IND regulations. 3 These new European regulations ask for or 4 mandate the creation of an annual safety report, which 5 is, in concept, very similar to the IND annual report 6 that has additional requirements in term of the 7 interpretation consistent with European experience and 8 laws with regard to adverse events reporting. So 9 these additional activities that are being conducted 10 by the industry could be potentially leveraged to also 11 help with the issue that is at hand. 12 So some of the thoughts is in terms of a 13 lot of the ideas, we're really trying to focus only on 14 the best practices to try to see how we can best 15 address the problem at hand, again, with the 16 assumption that this was already within the current 17 regulatory framework. I don't know if I answered your 18 question. 19 DR. PRENTICE: Yes, yes, thank you very 20 much. Okay. Bob? 21 DR. HAUSER: Thank you for your 22 presentations. They were very clear. You know, this 51 1 AER morass is an ongoing problem, and it's affecting 2 the safety of human subjects. I appreciate all of the 3 problems involved in harmonizing and so forth. But 4 when we sent our recommendation to the Secretary, we 5 underlined the word "promptly," and that was done very 6 specifically because we felt this needed to be 7 expedited in the process. I would like to suggest 8 that perhaps this draft report could be ready for us 9 to see at our next meeting, which is in April. 10 The other comments I'd like to make is 11 that -- one question: are there representatives from 12 working IRBs on your taskforce? 13 MS. PATTERSON: The members of the taskforce are 14 members of staff and federal agencies that I mentioned 15 in the presentation that part of the work plan is 16 outreach and involvement of focus groups that involve 17 members of IRBs, PIs, and industry, and that will 18 happen this spring. 19 DR. HAUSER: And the other question I have 20 is these adverse events involve not just drugs but 21 also medical devices. How are you incorporating the 22 medical device community into your taskforce? 52 1 MS. PATTERSON: Well, certainly, from FDA, 2 their Center for Devices has representation on all the 3 focus groups and the taskforce, as well. And when we 4 look to the industry and the PI and the IRB focus 5 groups, it will be very important to include a broad 6 spectrum and not only intervention studies but also 7 look at the broad range of clinical research, 8 epidemiologic, natural history studies, the social 9 sciences as well. So we are trying to be very 10 cognizant of the broad scope of research with human 11 subjects, but that's a very good point. 12 DR. HAUSER: Thank you. 13 DR. PRENTICE: Felix? 14 DR. GYI: Dr. Patterson, I want to applaud 15 you and Dr. Carome and your colleagues across the 16 agencies for taking the time to address an issue that 17 SACHRP had identified initially. I'm in agreement 18 with Dr. Hauser in that we would like to see this as 19 quickly as possible, but I'm also cognizant of the 20 fact that interagency discussions take a long time, 21 especially when it comes to harmonizing definitions 22 that are previously perhaps inconsistent and 53 1 incongruent in terms of how the interpretations have 2 occurred. 3 The point that I am somewhat troubled by 4 and still remain concerned about is the illustration 5 that Dr. Carome raised. You asked a question, how 6 many of the multiple hundreds of clinical trials they 7 have adverse reports for, and he correctly indicated 8 that they only have jurisdiction over the studies that 9 they have oversight. 10 The part that we, as IRBs, wrestle with 11 have to do with FDA-regulated products. What I'm 12 concerned about is the fact that FDA folks have not 13 stepped up to the plate, perhaps, as candidly as the 14 other folks have to say where the acceptance levels 15 would be because what I've heard, as well, before is 16 that unless there is a regulatory change that the 17 acceptance of some of these issues may not be as 18 smooth as possible. 19 So I would encourage you and your groups 20 to at least push the appropriate buttons that you can 21 to facilitate those discussions because, if your 22 bottom line is correct, then by the middle of this 54 1 year we should have some discussion between industry, 2 IRBs, and PIs to be a part of your process and come up 3 with a documentation that we need to move the guidance 4 in the direction that would be very facilitatory, 5 which is to help IRBs ultimately be more efficient in 6 protecting human subjects. 7 DR. PRENTICE: Let me respond to both Bob 8 and you, Felix. I've learned that the wheels of 9 bureaucracy in Washington don't turn as fast as we 10 would like them to turn. We have a SACHRP meeting in 11 April, and we have a SACHRP meeting in August. I 12 would suggest a more appropriate time for Amy's group 13 to produce a report would be in August. I think 14 they'll have a much better opportunity to interact 15 with their sister federal agencies and develop this 16 harmonized guidance. I think April is perhaps a bit 17 premature. That's only about two and a half months 18 from now. Are you prepared to accept a later date, 19 Bob? 20 DR. HAUSER: Well, I appreciate that. I 21 appreciate the problems that government has and the 22 complexities involved. But as a member of this 55 1 committee, I have to sit here and just express a 2 concern, and my concern is that this signal-to-noise 3 ratio is really out of whack, and it's creating a 4 situation where it's impossible to analyze data. And 5 if I look at my patients and I say to myself what can 6 I do for them that might ensure more protection, I 7 think getting this taskforce report done quickly is 8 something that I can do and we can do. 9 So, yes, I will accept that, Ernie. But 10 I would like to see some progress in April. Maybe we 11 can have an interim reporter, an update, so we don't 12 lose touch with this very important activity. 13 DR. PRENTICE: Amy, how would you feel 14 about giving us a brief update in April? 15 MS. PATTERSON: That's fine. I'm not sure 16 I can just move the location along the timelines, so 17 you can see where the meter is. 18 DR. PRENTICE: Okay. Also, another 19 comment relative to Mike's response to my questions. 20 I think we need to recognize that institutions who 21 have a Federal-Wide Assurance that voluntarily agree 22 to comply with 45 CFR 46 must report all unanticipated 56 1 problems involving risk promptly to OHRP. So it goes 2 beyond simply HHS-funded studies. So I don't know 3 what the percentage is of institutions who have, 4 quote/unquote, voluntarily agreed and their assurance 5 to comply with the Common Rule, but I suspect it's 6 pretty high. And how many assurances do we have out 7 there? 8 DR. CAROME: There are about 8,000-plus, 9 and then there's about 60 percent voluntarily -- 10 DR. PRENTICE: Okay. So that's þ- 11 DR. CAROME: Some prefer to modify the 12 terms for the reporting requirements, and they're 13 going to limit that to the federal -- 14 DR. PRENTICE: Okay. So that's a lot of 15 sites, right? A lot of sites. Susan? 16 MS. KORNETSKY: Thank you both, and it's 17 refreshing to hear that there's some news, but I also 18 think, you know, what you can do to move it along 19 would be helpful. One of the questions that I had, 20 and you both have sort of very full plates, but one of 21 the things that I think will also, that I think about 22 in having a successful, is that, you know, basically, 57 1 there's going to be a sifting of the events and some 2 determinations made so that the IRBs are getting 3 what's important, and that's going to require very 4 good data monitoring committees, data safety 5 monitoring boards. And even after working in the 6 field for 20 years, there's still a lot of discussion 7 about what is appropriate for what type of study, 8 independence on data safety monitoring boards, 9 especially in today's environment. 10 So if we are relying on a certain level of 11 sifting before it comes to the IRB, there's got to be 12 a real level of trust in that process, and I'm just 13 wondering what you both, you know, can say about that. 14 The other comment that was raised is, you 15 know, I was almost knocked off my chair when I see the 16 part of the slide that sponsors and investigators 17 shall document their analysis, and this would be 18 available to investigators and to the IRB. I mean, I 19 think that's wonderful, and I think we've talked about 20 that, and maybe David has some problems. But I don't 21 know about other people, but there have been times 22 when IRB has asked for that and get responses, "It's 58 1 proprietary information; you can't get that." 2 And I realize, I mean, it's not to say 3 that we shouldn't strive for that. It would be 4 wonderful to get, but I'm more concerned about the 5 first part and what we can do to really have trust in 6 that. 7 MS. PATTERSON: Well, certainly for DSMBs 8 and DMCs, one of the issues that, at least I'm going 9 to speak for my agency at the moment, we are looking 10 at is -- and I know FDA has a similar effort underway 11 that has been circulated amongst the agencies and WHO 12 does as well, and there's a dovetailing of concerns 13 and emerging guidance on best practices and standard 14 operating policies and procedures for DSMBs. Under 15 that fairly large umbrella are a number of very 16 important issues. It sounds a bit administrative and 17 bureaucratic, but there's some very telling issues 18 about how data in a trial is really handled, looked 19 at, and communicated. And these include best 20 practices in data review. Are we looking at blinded 21 or unblinded data? When are we looking at the data? 22 When is the data monitoring committee or, if it's a 59 1 data monitoring plan, how is that being communicated 2 to the IRB? What are the respective roles and 3 responsibilities of the data safety monitoring board 4 and the IRB? What is the relationship between the 5 data monitoring committee or board and the trial, the 6 institution, the agency, or the sponsor? What are the 7 roles and responsibilities of the data monitoring 8 committee? Are they making decisions, or are they 9 making recommendations? And who's actually enacting 10 those decisions? 11 So these are some very important issues. 12 I'm glad you raised it. We're taking a very hard look 13 at this, and, again, I wish FDA were here because I 14 know they're preparing to go out with some guidance on 15 this topic very shortly, as is WHO. From my own 16 agency, we look this year to be holding a conference 17 on best practices on data safety monitoring boards and 18 DMCs. And so that may be another topic that we can 19 come back and share with you and get input on. Jean- 20 Louis? 21 DR. SAILLOT: Well, you know, the whole 22 issue of trust and transparency, when we put forth the 60 1 concept of a filtering or a sifting, basically led to 2 a lot of discussion and debate within the IRB Sponsor 3 Roundtable. I don't know that we have really nailed 4 this down to the extent that it should be nailed down, 5 but the thought really is to address the transparency 6 issue and the trust issue. The safety communication 7 plan, where all this information would be available 8 regardless, it is more of a question in timing. And, 9 actually, Dr. Patterson's slide on what is submitted 10 to whom, when, is a very relevant question. 11 So some of the checks and balances 12 proposed, that are part of the proposal are to address 13 some of the transparency and trust issues. One of the 14 challenges, though, is that the people that really 15 have access to all the information to make meaningful 16 decisions are really the sponsors, and the 17 investigators, they have to relay all this information 18 in a meaningful way to the investigators. Some of us 19 do it well, some of us do it better. I think it goes 20 back to the best practices. But you cannot take the 21 sponsors out of the equation if you want to get to the 22 level of protection of the subjects. 61 1 So all of this discussion, I mean the 2 safety communication plan, the fact that this 3 information would be available but on an aggregate 4 fashion at a defined time point that the IRB can see 5 this information is one part, you know, one of the 6 different checks and balances that we have discussed. 7 Obviously, the four certain studies, the need for 8 independent safety oversight like a DSMB, is critical. 9 And FDA has issued guidance as to what they consider 10 independent from the sponsors and what they consider 11 not independent. That is also important for us to 12 take into consideration. 13 I personally don't believe that a DSMB 14 would be required for each and every study. The large 15 studies, obviously, with specific end points, outcome, 16 you know, death as an outcome, the large outcome 17 studies come as real good candidates for this. I 18 believe that the, you know, the experience in getting 19 input from a larger community on this proposal is what 20 we are looking for in terms of making sure that the 21 proposal is acceptable to all parties. 22 But it is an issue that we have been 62 1 struggling with in terms of trying to balance the 2 efficiency or the issue of the signal-to-noise as 3 opposed to meaningful information. The question is is 4 data really, is providing all the data on an ongoing 5 basis in an aggregated fashion being more transparent, 6 as opposed to taking the best practice of looking at 7 this data and trying to highlight what is really 8 relevant. 9 MR. ADAMS: Question. 10 DR. PRENTICE: I want to respond to Susan 11 first, if you don't mind, Tom. Susan, you mentioned 12 concern about the IRB auditing the PI's analysis of an 13 external adverse event, an IND safety report that is 14 not sent to the IRB, so they have some kind of an 15 analysis that they perform. For example, they decide 16 the protocol does not need to be modified to minimize 17 risk further, the risk/benefit relationship is still 18 acceptable and the consent form is still acceptable, 19 and they document that analysis, they put it into 20 their file. That's subject to review or audit by the 21 IRB, and you indicated that was a problem at your 22 institution for proprietary reasons, and I'm trying to 63 1 understand why that would be. 2 MS. KORNETSKY: It's not the PI part. The 3 slide that I was referring to, it said sponsors shall 4 document their analysis of all external AEs and that 5 this analysis would then be available to the IRB. 6 DR. PRENTICE: No, no, I don't think that 7 was the intent of this. 8 DR. WEINER: My question is the same as 9 yours, about the investigator having a discretion to 10 report to the IRB. Now, I mean, in terms of the 11 criteria that are in the slide, Dr. Saillot's slide, 12 describing about the reporting model, the protocol 13 modification and changing form consent are external 14 criteria that is external to that particular PI. So 15 one doesn't really even need a PI to make that report 16 to the IRB. It's a sort of automatic review by the 17 local IRB. 18 But then there's this third item that is 19 discretionary, and that's a situation in which it 20 strikes me that the investigator would either over- 21 report or would either tend to over-report or under- 22 report under those external AEs under those 64 1 circumstances. I mean, they're either going to 2 protect against suits or not want to violate the 3 relationship with the sponsor. 4 DR. PRENTICE: Well, that's a good point. 5 First of all, they're over-reporting now. And the 6 idea is to cut back on reports that really are 7 basically meaningless. And the first gatekeeper is 8 the sponsor and their DSMB. So, hopefully, that's 9 going to be a filtering system whereby investigators 10 are not going to get these reports that they have to 11 send to the IRB. They're going to get the reports 12 under the IND regs right now, but they are reports 13 that do not have to go to the IRB. That's the idea of 14 establishing criteria. 15 But, nonetheless, the PI still has the 16 option, if he or she feels that a report should go to 17 the IRB, even though it doesn't meet one of those 18 three criteria, they can send it to the IRB for any, 19 any reason. Now, will PIs adopt a protective stance 20 and send everything to the IRB? Well, I don't know. 21 I suspect, if they're educated, they probably won't 22 because it's a lot of work for PIs to send these 65 1 reports to the IRB, particularly if you're requiring 2 them to fill out some kind of an analysis, as we do in 3 my institution. 4 DR. WEINER: I wonder just, from Dr. 5 Saillot, you know, whether that was a point of 6 discussion and, if so, what the content was. 7 DR. SAILLOT: Yes, it was, indeed, 8 discussed. You know, there's always, this was 9 something that was brought on earlier on in terms of 10 the conservative nature, you know, that could impact 11 both the sponsors and the investigators, by the way, 12 with sponsors just saying, "Well, just in case, we're 13 going to send anything anyway, the way that we do it 14 right now." You know, the thought really is to try to 15 move away from that, you know, "Let's send everything 16 out; that's it, I'm done," kind of mindset and really 17 focus on what is, you know, the surveillance of the 18 safety during clinical trials. It's really to look at 19 this signal issue, and this is not easy. I mean, you 20 know, signaling in pharmacovigilance is not an easy 21 task. But by reinforcing the best practice and really 22 the thought process, the intellectual thought process 66 1 of that, is this report a really relevant report, as 2 opposed to something that, administratively, fulfills 3 the administrative definition of expedited reports. 4 By just providing guidance to refocus both 5 the sponsor and investigators on why is it that 6 they're expected to review the data, it is to gauge 7 whether there's an impact on the subjects. By even 8 reinforcing this, this may have already a training, if 9 I could say so, aspect to it. 10 Now, additional training for both sponsors 11 and investigators may be needed, too. You know, 12 that's something that may be discussed as part of the 13 taskforce. 14 Always, the IRB Sponsor Roundtable came 15 back as that the current system is not better or has 16 a significant number of flaws. So we have two 17 options: do nothing or to try to incrementally improve 18 the process. That's where we're coming from. 19 DR. HAUSER: Could I ask a question? 20 DR. PRENTICE: Related to what he just 21 said? 22 DR. HAUSER: Well, go ahead. 67 1 DR. PRENTICE: Okay. Then you're after 2 Tom. Wait, hold on, hold on, I got to start writing 3 down names. Tom, you're next. 4 MR. ADAMS: First, I'd just like to join 5 with what everyone else said. I think these are two 6 really great initiatives, understanding what you go 7 through with intergovernmental taskforces. There's a 8 lot of work that I know is going to be occurring 9 between now and the time we do get the final report. 10 I have two comments. The first is that I 11 do share somewhat the concern that we've don't 12 actually wind up here encouraging the reporting of 13 everything. And I go back to the experience that we 14 had with the implementation of the HIPAA regulations, 15 where every hospital staff and attorney in America, 16 you know, further the intent, if you will, of that 17 legislation and the regulations. And so I'm pleased 18 that you've got a period of time for the roundtable 19 and for the taskforce to seek public opinion here. In 20 particular, I think we need to hear from attorney 21 groups to see what their reaction of this is, as well. 22 I don't think we want to do something which could be 68 1 perceived or painted as providing less protection or 2 less information for the public. So that, to me, 3 would be important. 4 DR. HAUSER: Susan made a comment about 5 the IRB requesting information from a sponsor, and 6 there was a claim that it was proprietary. Under the 7 regs, can a sponsor withhold adverse event reports 8 because they are "proprietary" from the IRB? 9 DR. SAILLOT: Not that I know of. 10 DR. PRENTICE: I don't think so. 11 DR. SAILLOT: Let me just add something. 12 I mean, the proprietary concerns, which are real 13 concerns within the industry in terms of protecting 14 innovation and all the issues related to patents 15 really have to be looked at, you know, at the proper 16 level within a company. There may be folks that just 17 react and say, "Well, this is proprietary," without 18 really understanding all the consequences. And I 19 personally can't foresee that providing safety 20 information to an IRB in the context of clinical 21 research would be preempted by proprietary reasons. I 22 mean, that's clearly within the regulation, the 69 1 obligation of the sponsor to make all relevant safety 2 information to the investigators and, you know, 3 directly or indirectly to the IRBs. 4 DR. HAUSER: Just let me follow-up. So if 5 I understand this right, if an IRB requests all of the 6 adverse event reports that a sponsor received, the 7 sponsor uniformly discloses those events? 8 DR. SAILLOT: I'm not sure I got your 9 question, sir. Could you -- 10 DR. HAUSER: Well, if I understand you 11 correctly, if a IRB requests all adverse event reports 12 from a sponsor, it is the sponsor's reaction to 13 provide all those adverse event reports to the IRB. 14 DR. SAILLOT: Yes, that would be the 15 action. It obviously comes with a burden on the 16 sponsor, and people may try to discuss with you as to 17 the rationale for the request and so forth, but this 18 information I don't, you know, I don't see that that 19 information could be filtered. 20 DR. PRENTICE: I think you can get it. 21 It's not always easy to get it. We've had a situation 22 where we had a suicidal ideation where a young 70 1 adolescent almost killed himself, and we were unclear 2 as to whether or not it was related to the 3 psychotropic drug. We literally asked the sponsor for 4 all reports with regard to that particular adverse 5 event, and they initially said no. And then we said 6 we're stopping the trial, and then they, you know, 7 came forth. 8 So you can get it. I would like to see 9 more communication between IRBs and sponsors and IRBs 10 and DSMBs than we've had in the past. Nancy? 11 DR. JONES: Dr. Patterson, just thank you 12 for the harmonization work that you're doing. But, 13 along that, as you're looking at this analysis, how 14 did we get to this point, and so are there things in 15 place that we wouldn't get to the point that regs are 16 discord between the different agencies as we move 17 forward and new guidance and new regulations? Is the 18 climate different, or are there things that you've 19 learned so that you don't have to do this process 20 again for something else? 21 MS. PATTERSON: Well, it's an interesting 22 question and an important one and one that I've 71 1 actually thought a lot about. I do think we're at a 2 different point. I think that there's the different 3 culture that pervades now than it did even five years 4 ago. I'll again be a bit parochial and speak about my 5 own agency. We have 27 different institutes and 6 centers, 17 of which have clinical research as a large 7 component of their portfolio. And five or six years 8 ago, the general philosophy was let a thousand flowers 9 bloom. And the etiology of the differences among our 10 ICs and one can extrapolate this to, you know, to 11 other agencies even more so because of different 12 purviews, different scope of applicability of their 13 regulatory or fiduciary authorities. 14 But many of those differences between the 15 ICs and the agencies have come up. They're 16 historical, they're idiosyncratic. People developed 17 approaches, policies, requirements to meet a 18 particular situation they were dealing with. And it 19 worked, it worked for them. 20 But as trials become increasingly 21 collaborative, multi-centers, Jean-Louis has said 22 international in scope, we really become confronted 72 1 with the diversity which in and of itself makes it 2 more complex a task to comply. But when the diversity 3 is actually true differences in content and timeframes 4 and what's to be reported to whom, it really becomes 5 a barrier. And I think the day has come for 6 streamlining and harmonizing. 7 Now, that said, there are considerable 8 challenges, as some of you have already alluded to, 9 that government can be as slow as molasses in January. 10 We're in January, and I hope we're going to have a 11 thaw here. 12 But the challenges to harmonization are 13 several fold. One is that the status quo is 14 comfortable, and we have to fight that. We have to be 15 self critical, and that's the large part about what 16 this taskforce is about is taking a very hard 17 introspective look at ourselves and how we do 18 business. 19 The second thing is that, like you, most 20 federal workers have a stack of work to get through, 21 and it's hard enough to get through your own tasks, 22 much less collaborate with the person down the hall, 73 1 far less to collaborate with somebody in a different 2 institution. Coordination and collaboration takes 3 time. 4 But I think, again, we have reached a 5 point where it is almost impossible to do our jobs 6 correctly and well and serve the public without 7 actually coordinating and harmonizing. So I think the 8 day has come. 9 DR. PRENTICE: Mary? 10 DR. POLAN: I really hadn't realized how 11 complicated and unknown this whole area is in terms of 12 numbers of adverse events and reported adverse events 13 and who gets reported what to whom. So it occurred to 14 me that as you do the interagency harmonization and as 15 the roundtable proceeds, how have both of you set up 16 metrics for measurement? Whether or not you're going 17 to be successful in what you do, who do you measure 18 it? The number of reports the IRB gets or the number 19 of reports the sponsor gets? Because it seems to me 20 it is so confusing that, unless you have very clear 21 measurements of outcome, you're never going to know 22 whether what you've done is worth anything. 74 1 DR. SAILLOT: Well, this is a very valid 2 question. I don't know that we have given enough 3 thought, at least in the IRB Sponsor Roundtable, on 4 metrics or key performance indicators of the proposal. 5 This is something that we'll have to take back and 6 start giving some thoughts. 7 There's some basic metrics that we can 8 get. For example, the problem we're trying to tackle 9 is a qualitative problem, not a quantitative. It is 10 what is the quality information that needs to be 11 provided to the IRBs. Putting metrics on qualitative 12 issues is always more difficult than on quantitative. 13 We can probably nail down quantities pretty well, but 14 what does that mean to the actual bottom line: 15 increased subject protection? That requires some 16 thought, and I don't have the answer to that yet. 17 DR. POLAN: Well, perhaps, we could ask 18 you when you come back again to elaborate on how you 19 plan to measure these things? 20 DR. SAILLOT: We'll try. 21 DR. PRENTICE: Mark, you've been very 22 patient. 75 1 MR. BARNES: I'm exhausted from yesterday. 2 A few little things. One is, aside from thanking you, 3 obviously, for the work that you've done, is that one 4 thing that has occurred to me in the past when I've 5 done investigations of protocols that have gone wrong 6 in one way or another is the fundamental kind of 7 conceptual problem of thinking that the adverse event 8 calculus of describing it as unanticipated adverse 9 events because it's actually that I think an IRB would 10 want to know really all significant adverse events, 11 whether they're anticipated or unanticipated. And the 12 reason for that is that even within anticipated 13 adverse events there are degrees of adversity. There 14 are degrees of seriousness, and that may, in fact, you 15 know, weight into the risk/benefit calculus. 16 The other thing, just as an anecdote, is 17 that once when I did a, a few years ago I did an 18 investigation of an oncology protocol where there had 19 been several unexplained and unanticipated, just from 20 our perspective, patient deaths. The response of the 21 oncologist, who is the PI, said, "Well, I said in the 22 protocol they might die, so I didn't report the 76 1 deaths, so what's the problem?" They had anticipated 2 they were going to die; they had cancer. 3 So you need to be careful about just the 4 terminology. Unanticipated may work if what you're 5 talking about is the mandate to report significant 6 departures from regulatory guidelines, but it may not 7 work when you're talking about adverse events. So 8 just put that on the table. 9 Another thing about what Susan was talking 10 about, the problem has not been, I think, that we 11 can't get the actual adverse event reports from 12 sponsors, speaking on behalf of medical centers and 13 IRBs. It's been that what we can't get from DSMBs, 14 even perhaps from NIH DSMBs and also from the 15 sponsored DSMBs, is the actual content of the 16 analysis. That's what we can't get, and that's what 17 some IRBs, at least the ones that have enough 18 expertise to care about this. 19 In some cases, we've wanted the analysis, 20 and that has been what has been shielded either by 21 government agencies, like DoD or NIH, by saying that, 22 like, it's none of your business or by the sponsors 77 1 saying it's proprietary. 2 And the final thing I wanted to say is, 3 and if you want to react to that, I'd like to hear it, 4 but the final thing is sort of a slightly separate 5 point. It is just a question. Has the FDA actually 6 issued guidance about the independence of DSMBs yet 7 and clinical events committees? 8 MS. PATTERSON: I'll answer that last 9 question and then also speak a little bit to the 10 unanticipated or the unexpected. With regard to the 11 independence of DSMBs, DMCs, and other matters related 12 to their operations, FDA published in 2002 a draft 13 guidance that many of you probably saw and reviewed 14 and will be coming out, as we understand it, shortly 15 with another iteration of that. So I think that's 16 imminent, and it speaks to the importance of the 17 independence of the data monitoring committee or 18 board. 19 I just want to echo and agree with your 20 observations, and I think the taskforce would, too, 21 about the term unanticipated. One looks at the FDA 22 regs, and David isn't here so I'll just note that, 78 1 with respect to the terminology unexpected, as you 2 pointed out, there are other dimensions of that that 3 need to be taken into consideration. In the FDA regs, 4 the IND regs, those dimensions are actually spelled 5 out in terms of the specificity and the severity of an 6 event. 7 So I would echo your concerns about this. 8 I think this is an education and training issue, that 9 people understand that, you know, a headache can take 10 many forms, as we all know. It can be a tension 11 headache. You're exhausted, maybe you have a headache 12 from yesterday, but it's not a neuropthalmologic 13 migraine, we hope. And, likewise, other events may 14 assume different grades of severity and the 15 specificity of their clinical manifestation may make 16 them fulfill that initial criteria set forth in the 17 IB, the investigator brochure, or the monitoring plan, 18 or not. And one has to look very critically at that. 19 Furthermore, when an adverse event does 20 occur and there's a determination that, well, yes, it 21 was a headache, and we said that in the informed 22 consent, but this was a severe headache that required 79 1 hospitalization and pharmacologic intervention. Then 2 the informed consent needs to be revisited and note 3 that this is a particular type of headache that may 4 occur. So I just agree and feel strongly about this, 5 as well. 6 DR. SAILLOT: Well, Amy said very nicely 7 for our components, well, the relevant components 8 around unexpected. I just want to add one more thing, 9 that even in the current regulatory framework, if the 10 adverse event in question is already known, the 11 frequency of reporting changes, and then that changes 12 it to be unexpected. And that's probably a training 13 issue, as well, because people just, you know, go 14 directly to the serious unexpected definition without 15 thinking about, you know, the degree of severity. 16 Again, serious is a non-administrative definition. 17 Severity is a medical one. 18 The other point was on the DSMB, and 19 that's a very valid þ- something that I didn't expand 20 on my presentation. But in the draft proposal that we 21 have, we have this caveat, and I'd like to hear 22 reaction from this group on it. It is not so much the 80 1 proprietary nature that shields DSMBs from disclosure. 2 It is a methodological one. You have to protect your 3 statistical analysis and your safety analysis from the 4 people that are involved in conducting the trial, so 5 that the knowledge of the DSMB outcome or review of 6 the data, which is protected very closely, is 7 critical. 8 So there's a caveat in the proposal that 9 says as long as it does not impact the methodology of 10 the trial and the validity of the trial, then this 11 information should be shared. But in my past 12 interactions with IRBs, folks were saying, "The fact 13 that the DSMB met and did not close the trial, we 14 would like to know about this. We are not 15 automatically made aware of this." So one of the best 16 practice would be part of the safety communication 17 plan to lay out when the DSMBs are to meet, what are 18 they going to review, what the decision criteria that 19 they may be faced with, and just communicating that, 20 even though the content may still be protected and, 21 again, more for methodological reasons than 22 proprietary reasons. 81 1 DR. PRENTICE: Okay. I'd like to close 2 this up before our break. First of all, on behalf of 3 SACHRP, I would like to thank you both for some very 4 interesting and important presentations. Amy, I would 5 like to also extend our gratitude to your committee 6 for all the work they're doing. We recognize the fact 7 that it involves multiple agencies and an awful lot of 8 work, and we really appreciate that. 9 And, Jean-Louis, the IRB Roundtable 10 continues to pursue this particular problem, and I 11 want to recognize Marie Hartwick as the young lady 12 sitting right there, second row, who volunteers her 13 time. She's a lawyer and volunteers her time to 14 organize the IRB Roundtable to pursue these kinds of 15 issues. 16 I also want to recognize OHRP and, 17 particularly, Mike Carome. I always give Mike a hard 18 time. He knows that, he expects it, but he has 19 tackled this adverse event problem in terms of 20 developing draft guidance, which I had the privilege 21 to look at. And that has been shared with Amy's 22 group. 82 1 And I think we're making progress in this 2 particular area, whereas I cannot say the same thing 3 over the last ten years or so when I've been pursuing 4 this particular problem to no avail, as have many of 5 my colleagues. So I think that the federal government 6 is listening to the concerns of the IRB community and 7 the concerns of investigators. So I'm optimistic, 8 Bob, that, while we won't have a solution in the next 9 six months, I would see that within a year or two, 10 perhaps two years, we're going to have this problem 11 resolved. That's my hope. 12 So thank you very much. We're going to 13 now take a 15-minute break. 14 (Whereupon, the foregoing matter went off 15 the record at 10:24 a.m. and went back on the record 16 at 10:48 a.m.) 17 DR. PRENTICE: All right. The remaining 18 agenda for the morning will consist of a presentation 19 and discussion led by the two co-chairs of our new 20 Subpart A Subcommittee, and those individuals are 21 Felix Gyi and Dan Nelson. And then after that we're 22 going to have a public comment section. 83 1 Dan has graciously agreed to present, so, 2 Dan, would you proceed, please? 3 MR. NELSON: Thank you, Ernie. As Sue 4 said, I had to stand up and entertain, so I'll try to 5 accommodate. So last week, as we were finalizing 6 these slides, my co-chair, Felix Gyi said, you know, 7 "These presentations usually work best when one person 8 does all the talking, and that should be you," and 9 I've belatedly realized he's both self-effacing and 10 clever all at the same time. 11 But I do want to say publicly that, while 12 I certainly appreciate the confidence that OHRP and 13 SACHRP have put in me in asking me to co-chair this, 14 I'll have to say that a big factor in my decision to 15 accept was the opportunity to work again with Felix. 16 Even when he's not up here talking, I know that he's 17 constantly thinking and he's thinking sound, critical, 18 thoughtful thoughts, and I know together we're going 19 to be able to do great things. So, Felix, he's with 20 me right here, I know, today, and any questions we'll 21 certainly both jump in on. 22 This should be a fairly short presentation 84 1 because we really haven't done anything yet, although 2 there's been a lot of behind-the-scenes work, and 3 we're happy to bring you up-to-date with that. So I'd 4 like to walk through just a bit of the background 5 behind the Subpart A Subcommittee, review our 6 membership with you, review the charge from SACHRP, 7 and then talk a bit about our future plans. 8 So just to remind you, certainly as the 9 group at large knows, SACHRP already had active 10 committees working on Subparts C and D. I understand 11 you discussed those yesterday. OHRP has been 12 developing guidance on Subpart B, and so we have quite 13 a focus, although broad enough as it is, given the 14 focus on Subpart A. But that is our focus of this 15 subcommittee. 16 There were Subpart A panel presentations 17 last October. At least two of the presenters are here 18 in the room. And this then led to further thoughts on 19 the issue and to the formation of this 20 subcommittee. Just to paraphrase, I think some 21 conventional wisdom on Subpart A regulations were 22 promulgated in an era where the protections of human 85 1 research subjects were perhaps nominal and oversight 2 mechanisms were either non-existent or inconsistent. 3 And out of that environment grew the current 4 regulatory structure that we follow today. I think 5 it's fair to say that the modern research environment 6 has outpaced our system for overseeing that and, 7 perhaps, is in need of some examination and some 8 updating, and that's what we're all about. 9 I think the application of Subpart A in 10 the modern research environment has resulted in 11 conflicting interpretations of what constitutes 12 appropriate oversight and, perhaps, as the regulatory 13 burden without commiserate additional meaningful 14 protection of human research subjects. So at the 15 October 2004 meeting, SACHRP did act to establish a 16 subcommittee to review Subpart A of 45 CFR 46. 17 These are the appointed members drawing on 18 guidance as to the committee membership put forth by 19 SACHRP last October. Some of the members, many of 20 them are, in fact, in the room: Gary Chadwick from the 21 University of Rochester, well known to many of you 22 already; Bruce Gordan, a pediatric oncologist from the 86 1 University of Nebraska and IRB chair; Felix you know 2 from in his day job as CEO of Chesapeake Research 3 Review; Isaac Hopkins, Mr. Hopkins is from New Jersey 4 where he serves as a community representative on an 5 IRB and as a community research advocate; Nancy Jones 6 you also know from SACHRP from just down the road from 7 me at Wake Forest. If you follow ACC basketball, you 8 know that her team beat my team recently. We still 9 look forward to working together. Moira Keane, also 10 known to many of you from the IRB and communities at 11 University of Minnesota; Susan Kornetsky from Boston 12 Childrens'; Gigi McMillan is Director and Founder of 13 the We Can Pediatric Brain Tumor Network. She's the 14 mother of a child who survived a brain tumor and went 15 on to found this network of support and advocacy and 16 education. Many of you heard her give a very moving 17 talk on her experience at the Community Research 18 Interface at last year's national IRB meetings.Me, you 19 know. Tom Puglisi, formally from OPRR and now with 20 PriceWaterhouseCoopers. Lorna Rhoads is an 21 anthropologist from the University of Washington, who 22 has great experience doing research of vulnerable 87 1 populations and in difficult circumstances. Ada Sue 2 Selwitz right here, and David Strauss is from the New 3 York State Psychiatric Institute and IRB chair and 4 increasingly engaged in this area. 5 I believe six out of the 13 members are 6 investigators, which was one of the stipulations put 7 forward by the SACHRP, in terms of at least a third 8 should have investigative experience. And we have 9 community representation, as well. So a diverse 10 group, and they're all coming ready to roll up their 11 sleeves and tackle this large task. 12 Additional members and input will come 13 from Mike Carome, our liaison to OHRP, and we will 14 have many ex-officio members and representatives of 15 agencies. We're told by Dr. Schwetz that there's 16 great interest in the work of this subcommittee from 17 a number of groups that will be contributing ex- 18 officio members, and we look forward to their 19 input. So our charge to the 20 subcommittee came from the chair with input from 21 several people. I will say it's a starting point. 22 It's a good framework to launch us. Ernie has made 88 1 clear this is not cast in stone and subject to 2 modification as our work progresses. But to review 3 that with you here today, we are charged to review and 4 assess all provisions of Subpart A and related OHRP 5 guidance documents. 6 Based then on this review and assessment 7 to develop recommendations for consideration by SACHRP 8 in three categories, which we'll go through on the 9 next few slides. The goals of this examination and 10 reassessment are to enhance protection of human 11 subjects and to reduce, where possible, regulatory 12 burdens that do not contribute to the meaningful 13 protection of human subjects. 14 So these three broad groupings and, on the 15 next slides, we'll show you some specific examples of 16 the regulatory categories that we're talking about 17 here, in some respects are grouped in degree of 18 difficulty, if you will. I think none of us are 19 looking, nor do we see a need, to totally rewrite 20 Subpart A, nor is that likely to occur in any of our 21 lifetimes, but we think there's a lot of room for 22 making recommendations and, again, working through in 89 1 this increasing pattern on these next three slides: 2 first, to make recommendations in the interpretation 3 of specific Subpart A provisions; secondly, to make 4 recommendations for the development of new or 5 modification of existing OHRP guidance on those 6 regulations; thirdly, where needed, to make 7 recommendations regarding possible revisions to the 8 regulations. 9 So the first of those again: 10 interpretations, understanding, and applicability. 11 The charge from SACHRP notes that the following terms 12 and/or provisions in 45 CFR 46 should be considered by 13 the subcommittee in the context of current problems 14 concerning interpretation, understanding, and 15 applicability to the contemporary biomedical and 16 behavioral research environment. 17 You can see the regulatory citations 18 there, and many of these problem definitions will be 19 immediately recognizable to you: the definition of 20 research, the definition of human subjects; therefore, 21 the definition of human subjects research requiring 22 IRB review; the definition of minimal risk, minor 90 1 changes, unanticipated problems. You've heard much 2 discussion on that already this morning. We're very 3 cognizant about the ongoing work by other groups that 4 will inform and, conversely, which we do not want to 5 replicate and will be keeping close tabs on these 6 other groups as they proceed in drawing on that. 7 Clarification of serious or continuing non-compliance, 8 which would then trigger reporting requirements and a 9 clarification of waiver or alteration of consent 10 requirements. 11 The second major grouping of our charge, 12 recommendations for development of new or modification 13 of existing OHRP guidance on Subpart A, the 14 subcommittee should consider both current OHRP 15 guidance, as well as areas where guidance is needed in 16 order to enhance human subject protection, achieve 17 greater consistency of protection across sites, and 18 reduce regulatory burdens. So examples in this 19 category are types of social and behavioral studies 20 that are not human subjects research under the 21 regulatory definition. Examples of these might 22 include journalistic activities, quality improvement 91 1 projects, and some oral histories. And, again, you're 2 familiar with the guidance, at least thus far, in 3 these areas. The notion of third-party consent, 4 clarification in that area, and an examination of 5 continuing review requirements relative to the work 6 intensity and detail needed at initial review.The 7 second category has a longer list, so it's continued 8 on this slide. Again, OHRP guidance relating to when 9 use of a resultant data set from a study no longer 10 requires continuing review. In other words, once a 11 study enters the data analysis phase, in what manner 12 should continuing review continue? 13 How to interpret minor changes that might 14 be eligible for expedited review, something open to a 15 lot of interpretation at present. Clarification of 16 when the full IRBs should be expected to review the 17 response of an investigator to initial IRB review. In 18 other words, the dividing line between minor 19 contingencies or substantive issues that might lead to 20 a deferral and return to the full IRB. 21 Clarification on what level of detail and 22 format is necessary in IRB minutes in relation to the 92 1 regulatory burden that this might impose on IRBs, how 2 to interpret the requirements for waiver or alteration 3 of informed consent, and, in terms of documentation of 4 informed consent, when would a signed consent form be 5 required for certain types of behavioral and social 6 science research. 7 Third grouping: the subcommittee should 8 identify any sections of Subpart A which should be 9 modified in order to resolve problems for OHRP 10 guidance based upon current regulations is not 11 sufficient. Examples: where modification or 12 clarification might be indicated include the 13 categories of exemptions, either existing exemptions 14 or the possibility of additional exemptions, where 15 required; again, the definition of research, the 16 definition of minimal risk; assurance requirements, 17 IRB membership requirements. An example here is 18 should we be considering a proportional requirement 19 for membership of non-scientists or non-affiliated 20 members, as opposed to the absolute that we currently 21 have. 22 Under continuing review, consider allowing 93 1 the IRB discretion to determine that or when 2 continuing review is not needed for certain minimal 3 risk research or could potentially be done less 4 frequently than the once per year, as currently 5 required. And, finally, to examine categories of 6 research that may be reviewed by the IRB.