• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services

Vaccines, Blood & Biologics

  • Print
  • Share
  • E-mail

Transcript for Public Workshop on Cell and Gene Therapy Clinical Trials in Pediatric Populations - Part 2

garbage pail of the cell and when you're missing an enzyme, in this case tripeptidyl peptidase one, the CLN2 and this enzyme chops off membrane proteins primarily, the lysosome eventually enlarges and the neurons undergo apoptosis and die.One of the advantages from the gene therapy point of view is that this protein in a precursor form is secreted and then picked up by neighboring cells and the same cell by the mannose-6-phosphate pathway so that you leverage the gene transfer and certainly for in vivo gene transfer if we've learned anything over the last two decades it's that we can't get every cell. This is a real advantage in that we can treat neighboring cells by cross- correction. This is what happens to the children. The ordinate is a clinical scale that we've developed that was based originally on a scale that was developed by some colleagues in Hamburg and we've seen a lot of these children and we developed a scale that's a little more


sophisticated with a few more parameters. This is the clinical rating scale where the higher number is more normal and the lower number is deterioration. That's a function of age of the children that we've seen and the scale is based on motor function, gait, language and feeding. As you can see, these children rapidly deteriorate over about 18 months to 2 years. That also from the clinical design point of view is an advantage because we're going to be able to show efficacy, it was by using clinical scales. If we can show that we can level off the deterioration, that might give us a measure of efficacy. We carried out our initial trial several years ago using what then was the standard for adeno-associated virus vector, serotype 2, a human serotype and so a simple vector and after evaluating many, many different promoters we wound up with the CAG promoter and the human CLN2-CDNA between the two ITRs. I won't review any of the details, but just to show you the final slide which was the clinical scale for that trial. At


that time we were using this prior scale or modified scale that was developed in Hamburg. We treated 10 children, 5 with severe and 5 moderate on the scale. The total dose was 2 times 10 to the 12th particle units in 12 different sites. I'll show you that in just a minute. The outcome was that the MRI study showed there was a trend, but nothing was significant. But on the clinical scale, and you can see it on the right, is the mean change and the modification of that where zero is no change, the positive number that's going up is some improvement and below has deterioration. In the control group, the untreated group, you can see of those five children that there was deterioration, whereas at 1 year or all of our data we had most of the children out to 18 months and there appeared to be efficacy. You have to take that with a caveat and I'm going to talk about that in just a few minutes which is the control group. The control group clearly is not a placebo control and it's not double- blinded. These were not


contemporaneous. They were children who we had seen around the same time that families for whatever reason decided not to enroll them in the gene therapy trial, but we were able to evaluate them. That's one of the real challenges in terms of developing these kinds of therapies is the control group. How did this trial inform us in terms of our new trial, and I'll get to the new trial in a second, using a different vector, one derived from a nonhuman primate called rh.10? One of the things that we learned is that we're limited in the children by the number of burr holes that we can make, we use six; the amount of volume that you can put into the brain at any one time per unit time and you're limited by that; you're limited by total anesthesia time. One lesson was to use a vector with better distribution. This is a diffuse disease throughout the brain so we want the vector to distribute throughout the brain. The second lesson was to use a more sensitive clinical rating scale and to minimize variability,


and I'll tell you in a minute how we do that and improve our quantitative MRI. From a risk-benefit point of view very similar to what you heard from Don Kohn, targeting early disease is beneficial, and if possible develop a contemporaneous control group. What we did learn was that our review of the consent process worked pretty well and I'll review that with you. Over the period of time when we were doing the clinical trial with AAV2 through a grant from the Neurologic Institute we evaluated 25 different serotypes of AAV and we landed on one called rh.10, the nonhuman primate serotype that by far and away was better in terms of distribution throughout the brain. The expression cassette is identical to what we used in our prior trial with the same promoter and the same CLN2, human CDNA, but a different capsid. This capsid was from AAV rh.10 and we're not sure why it's so much more effective. Probably its charge characteristics are such that it can diffuse through the brain better than any of the other


serotypes. This is an example in CNL2 knockout mice with an assessment at 1 year after administration. On the top are our controls in cortex, striatum or hippocampus, and below the brown stain is the human TPP1, the protein, the enzyme, and you can see that there is clear staining throughout 1 year later. Again with this theme of the earlier you treat the better off you are, this is again studies in knockout mice as a function of when we administer the vector to the CNS. This is survival of the mice on the ordinate and then time and days. The first two lines over on the left is treatment at 7 weeks with saline or untreated animals and they died in about 120 to 130 days. Then up at the top is the wild type that obviously live. Then the three other colors, the green, the yellow and the orange, are when we treated: At 7 weeks with green, at 3 weeks with yellow or with orange we treated at 2 days, and clearly you can see increased survival from the controls at all of these, but the earlier you treat the better.


This is a real dilemma also. This is a rare disease. Our experience is that the children at ages of about 3 to 4 develop some symptoms. Usually it's gait problems, sometimes vision problems, sometimes seizures and it takes about 6 months, sometimes 9 months before the diagnosis is made. It's not something that pediatric physicians and neurologists see a lot of and it takes a while before somebody makes the diagnosis. There is no newborn screening for this disorder so getting at the children early is a real dilemma. In terms of this vector being so good in terms of distributing throughout the brain, this is data from nonhuman primates and we're looking at the nonhuman primate brain sliced into multiple different cubes. Of course there is no knockout nonhuman primate so that we used a criteria that if we had enzyme activity that was beyond two standard deviations above the endogenous levels that we would accept that as being positive. The graph at the far left is PBS at 7 days and the graph in the middle is the vector at 7 days. It


takes about 2 weeks for the vector to start functioning because it has to convert a single strand to a double strand of DNA. The graph on the right is at 90 days and you can see a broad distribution throughout the brain and that's two standard deviations above and you probably only need about 5 to 10 percent of the normal levels to be able to treat this disease like other recessive disorders. That led to study design and then funding by the Neurologic Institute. We had a study design of 6 burr holes, 2 doses, 9 times 10 to the eleventh and 1.8 times 10 to the 12th, 2 injections at different levels per burr hole, but through the work Patrick Auberg and his colleagues in Paris who was one of our collaborators have done, they've developed a new strategy and we're going to try that clinically as well where we use two catheters simultaneously in each burr hole which will shorten the anesthesia time. We use 2 microliters per minute to minimize damage, 300 microliters per burr hole so that there are 12


sites total that are administered. This is what it looks like in terms of the OR, and one of the real challenges of doing this kind of study is how do you pay for it because the hospital could probably make $50,000 in the same time that we're doing this clinical study which ties up an OR, 12 people, 2 neurosurgeons and a whole variety of other people to do it so that it's a real challenge for everyone involved, both the investigators, the hospitals, the regulatory groups and so on. Our outcome measures are the score that we developed that I told you about and then quantitative MRIs so that we've worked hard with our colleagues over the last several years and have developed quantitative measures of ventricular volume, gray matter volume and cortical apparent diffusion coefficient which is movement of water in the brain. Then we have some secondary measures -- scale and a quality-of-life questionnaire that I don't think are going to be as useful as the primary measures. As for the


status of our trial, we've treated our first child now, a child 7 years old, homozygote from one of the common mutations that fit into the criteria over 12 sites and no adverse events. We are now 8 weeks and we have I think two children scheduled coming up over the next couple of months. Let me deal with the issues then that we face to the relevance of this meeting. This is a fatal childhood disease for which there is no therapy. There is no prenatal or newborn screening. It requires invasive neurosurgical procedures and there are challenges in terms of disease severity, do we treat mild children where we know we'll have better success if it works? On the other hand, the ethics of that compared to treating moderate or severe children. Issues of doses. We're limited in terms of the amounts that we can administer. We're also limited in terms of nonhuman primates and other animals in terms of toxicity and in terms of volumes that we can use. Then what do we do with the controls? In terms of the choice of disease


severity and dose, this was the first in humans with a nonhuman primate with adeno-associated virus to the CNS so that that was an issue. We chose to do mild to moderate versus severe to moderate in the AV2 serotype trial so that we decided that the earlier we treated the better we had a chance for efficacy. We were limited in volume and concentration to the animals and that's why we ended up with the doses that we ended up with because we couldn't go much higher in the nonhuman primates because you can't infuse that volume into the nonhuman primate brain. This is our trial design. There are a number of issues that relate to ethics in here. First we have a screening protocol and I strongly recommend that as a strategy. We don't invite children to be part of the gene therapy trial. We invite the families to be part of a screening protocol with no guarantee that they will be part of the gene therapy trial. Then if they're eligible the decision is made by four faculty representing three departments and independent of


me. I'm the PI on it. I think that's important also to separate the PI who can be conflicted for a whole variety of reasons. Then we give the family the choice to continue in the screening protocol or to enter the treatment protocol. What we found in our first trial was that some families for a variety of reasons chose not to participate so that we decided was the best we could do in terms of developing a control group. We hope to find 16 children and assess efficacy parameters again at 18 months. Then we have 16 at the 2 doses, 8 at each dose and then we assess efficacy before 6, 12 and 18 months. Another important concept is the consent process that's done also independent of the PI. We do this partially through our CTFC, the Clinical Research Center funded by NIH and we have a CTFC patient advocate present at the consenting process as well. With the critical path in developing things I see as the biggest challenge that we have is measurement of phenotype and that of course is


an issue for all of us as well as for the agency in terms of eventually should we get to the point that for instance the SCID trials are, could this eventually be accepted as an approved therapy so that there is a critical need to develop quantitative phenotypes. In terms of clinical scales, at least our experience is that at best they're crude integrated measures of overall CNS function. What we've done to try to limit that is we videotaped the assessments and after therapy we even put a stocking cap on the children who were not treated so that the observers don't know whether they were treated or not because of the burr holes. We then have three blinded observers, pediatric neurologists, assessing the videotapes. In terms of MRI and also CSF biomarkers we're interested in, this is evolving methodology and of course requires more data and verification, but I think that's going to be very useful. In terms of ethical issues with this, clearly they are a vulnerable population. Not only are they children, but because of the disease


process they can't make any decisions and the parents are very vulnerable in terms of this so that therapeutic misconception is a major issue. Enrollment decisions are an issue. Of course, there are complex relationships among funding sources, investigators and family. Don Kohn and I were talking before and in the disease he deals with and with us, the families are all connected by internet, they all find one another and they know what's going on around the world faster than we do. It's instantaneous information. We don't advertise at all, but we have all these families contacting us from all over the world. In terms of the vulnerable population, as I said, the subjects can't make informed consent and the parents have of course conscious and unconscious complex motives. They want to do everything they can for their children of course, but they also feel guilty. It's their genes that have resulted in this and it is very, very complex in terms of dealing with them. Also the procedures and therapy can have significant risk.


The biggest challenge that we've had is the failure of parents to appreciate that it's research and that there is risk. If it wasn't research it would be approved and they're clearly at risk. There is clearly a disconnect between parental expectations and the reality of these early phase trials and understanding and appreciating the informed consent document despite the approach that we use with multiple people, the PI is not involved, we have the patient advocate and so on. We send the consent form to the families before they show up in New York. We ask them to go over that with their physician, their clergy, whoever else are their advisers, but still it's a major problem. This is a fatal disease and that's different than a chronic disease, and of course there is the issue of mild versus severe and risk versus benefit. In terms of funding and personal motives, it's an orphan disease with less than 500 to 600 cases worldwide. It's not rational for industry to be very involved with this. The


federal grants often cannot cover clinical costs. There are enormous financial pressures on the hospitals. There are issues of industries ties and of academic investigators. We don't have any, but often that's an issue. And then you have the issue of funding from parents which is a real challenge. I get e-mails all the time about somewhere in the world that there are newspaper articles about us and that it costs X-hundreds of thousands of dollars and we'll treat you in New York. We have nothing to do with that. It's something that you can't stop, but it's an ethical issue that is relevant. I have a few recommendations from our experience. One is to have two distinct protocols, a screening and therapeutic protocol, that gives the investigators a little more leeway. You're not promising anything when they come. Of course, if the children are not eligible, it's a real issue for the parents and they're very disappointed. Separating the principal investigator from clinical care. That's easy in


my case. I'm a pulmonary care doctor and so I shouldn't be taking care of these kids anyway, but I think that it's very important that the PI be separated from the clinical care and have a clinical research monitor and physician medical monitor reporting to the regulatory groups independent of the PI. This is how we do it. I'm the IND holder, sponsor and principal investigator. We have what we call a Clinical Operations and Regulatory Affairs Group in our department and we have a clinical research monitor that is a dotted line to me, but also to the head of my clinical operations group. We also have a medical safety monitor and the clinical research monitor has the line authority, a line to the medical safety monitor and to the IRB, IBC and the data safety monitoring board as well. Over on the right we have research subject advocates who are part of the clinical research center. I think these are a variety of strategies that can be used to help protect the families as well as protect the


investigators in terms of the study. My second recommendation is to have standard operating procedures to cover specific items under the consent. We have written standard operating procedures for the people doing the consenting to make sure that they cover specific items. Some of those issues are efficacy is unlikely, it's a primary safety study and death is possible and we update the consent with adverse events as the trial progresses, should there be any, and we separate the PI from consenting. We have an independent research subject advocate and we also do some monitoring after the consent process. This is our consent form. As you can see, in the third paragraph it indicates, "It's important for you to know this research study involves high-level risks to your child and this includes the risk of death." You can't be more direct than that, but having talked to these parents over and over and they look at you know what they're thinking, they have no choice because their child is going to die and there are no other


therapies. So that's a challenge for all of us ethically in knowing how to deal with that. My third recommendation as I say is to have an expert panel independent of the PI for enrollment. You have to decide in terms of particularly the new therapies to consider whether to do more severe subjects or have more mild. In our first trial, our fourth subject died. The child was severe and we never could figure out working with the agency and our regulatory groups what was it was due to. The child left the hospital with no problems. We asked the families to stay in the area for 2 weeks and at that period we reevaluate. The day before the child developed grand mal seizures. We were able to suppress that by putting the child in a coma, but never could get the child out of that and eventually about 45 days later the family decided to let the child die at their home hospital. We did not get an autopsy although we requested it. All the data, MRI, CSF and all the other things, suggested there was no inflammation


so that we don't know whether or not that it was due to the vector which it could have been, whether it was the surgery and the catheters that happened to hit a site that induced the epilepsy or whether it was the natural history of the disease which does occur. But the fact that we had done it and it was a severe child I think helped in terms of the concept of moving along after a couple of months in clinical hold. In our new trial because we have the experience of the first trial, we decided to move to moderate to mild, but it's a significant ethical issue in making that decision. Then also I think another issue in funding and personal motives is that there are issues -- in relating funding and if there are, my recommendations are that this be a gift to the academic institution and not linked to any specific subjects. They have to protect the investigators and the academic institution, this is not fee for service and it's an ethical issue that has to be dealt with. My fourth recommendation is like Don's.


My experience is the same and I said this at the RAC meeting a couple of months ago, that our experience also is that I think that the IRBs, the IBCs, the CTFC, the Data Safety Monitoring Board, all that protect us as well as patients, so it's really all of us working together to do it. On the other hand, it is a huge burden on the academic groups. If you change our investigator or somebody associated with it, you have to report it all these groups appropriately in paper and the problem is the funding and the time. If there were a regulatory god out there that really could control all of this which there is not, I would recommend to clearly define the role of each regulatory group, harmonize as much as possible, clearly define specific issues. This might be the simplest thing to work out I think would be to try to simplify post-initial approval. Once you have the approval to move ahead, there are in these clinical protocols changes that go on all the time. They mostly are minor changes that do not relate to fundamental safety or efficacy and if


there was a way of harmonizing that between the groups, that would make things much, much simpler for everybody who's involved. In summary, gene therapy for the CNS at least in terms of the studies here and whether we can develop this to be a reality in terms of therapy that's efficacious, we're just going to have to wait and see how the studies go. Thank you. DR. DALEY: Let's take a couple of questions for Dr. Crystal if there are specific ones and then we'll have the panel come forward. SPEAKER: How and when do you discuss consent to autopsy in this trial? DR. CRYSTAL: In terms of autopsy, we have in the consent form that should there be death we would ask for an autopsy. Of course you cannot sign away those rights beforehand. We knew the child was going to die and we were in constant touch with the family. They had moved back to Britain at the time. We were in touch with them and their physicians in terms of that issue, but


the family decided not to. SPEAKER: I have a question on your comment about separating the PI from the consenting. Does that relate to the fact that as the PI are also the sponsor or is that just a hospital guideline? DR. CRYSTAL: It has nothing to do with our hospital and nothing to do with the fact that I'm a pulmonary doctor and not a neurologist. We do that for all of our protocols because we believe that it protects both the patients as well as the investigators. We as investigators and particularly the PI have other motives in terms of this and we want to try to do the best. We want to do it ethically and well, but we're academics and there clearly are other motives. So we think it's best to separate the PI so that that doesn't become an ethical issue. SPEAKER: Thank you very much for your nice talk. I have two questions. One is regarding the nonhuman primates AAV vectors, is there any concern about -- infection? The second


question is you injected the gene into the CNS and the CNS is a very crucial part of the human's personal integrity. Were there any ethical questions about that? DR. CRYSTAL: The second question? SPEAKER: The second question is the verification of the CNS's integrity, some people think that -- modification of the CNS cells may be change the very crucial integrity of the person. DR. CRYSTAL: In regard to the issue in terms of nonhuman primates and your concern about -- SPEAKER: Zoonosis. DR. CRYSTAL: This is a nonhuman primate vector and one of the issues is immunity that may be preexisting as well, and then of course there are issues of possible recombination. The way these vectors are designed, the latter is essentially impossible and so it was not a problem. Originally, back in the early 1990s, when we used the adenovirus for cystic fibrosis, that was a real concern of endogenous viruses and


we and the agency struggled with that. In fact, our first two patients, I was at NIH at the time who were the first humans ever to receive a virus for that were in negative pressure rooms for 2 months while we were testing for every possible thing. In terms of enhancement, your second question, that of course is an issue. What we're modifying here is a gene that has mutated and, therefore, very low in terms of expression so that we are simply augmenting what is actually normal. It's a significant consideration particularly because this is a diffuse disease throughout the brain including gray matter and including the frontal cortex and it's certainly an issue. For this particular application, I don't think that's an issue. DR. DALEY: The last question? SPEAKER: I wonder if you could expand a little bit on your recommendation to consider treating the severely affected patients first and reconciling that with some of the ideas we've heard today that recently diagnosed early onset


patients specially in CNS disease and whether it's gene therapy or enzyme replacement therapy, those are the ones we think are most likely to benefit the greatest. DR. CRYSTAL: The best example we have out in the gene therapy field of course is the Gelsinger case. In that trial the original concept was to treat severely affected newborns who die within a week. The decision from an ethical point of view was made that the parents would be under too -much duress and they couldn't make an informed decision, therefore you should do milder cases. I think had a much more severe case been done, we probably would not have the challenge in the field of having someone who was apparently a mild case that was relatively well functioning that died. That's the dilemma that we have. At least from our point of view I think if you're starting something that's very new just like taking the cue from the way cancer therapies are developed, chemotherapy is not going to help end stage cancer, but more with more mild disease


you will and I think it's the same. My recommendation if you're doing something totally new is to start severe and then work your way down and if you have problems I think from an ethical point of view it's perhaps more acceptable if you have significant problems in a more severe population. Once you have some experience there is no question that I agree with Dr. Kohn and others that the earlier you can get at these diseases the more possibility there is for efficacious therapy. DR. DALEY: Thank you, Dr. Crystal. I can invite these speakers back up on the podium for the panel discussion. When I was President of the International Society for Stem Cell Research we put together a panel to look at the issues around clinical translation of what were essentially novel products based on pluripotent stem cells and a variety of stem cells. We established a set of guidelines, but we continued to struggle. We've seen just in the last few weeks the first introduction into a patient of the


products of a human embryonic stem cell. Part of what motivated the International Society for Stem Cell Research was this growing tendency to essentially market therapies directly to patients well ahead of any proof of efficacy. This is obviously not done primarily in well-regulated jurisdictions like the United States and Europe, but, rather, in the sort of frontiers of Third World countries, middle America, Asia and the like. What strikes me is because the way these are marketed is based on their promise and the idea that somehow we're very conservative in the United States that we're denying patients access to revolutionary therapies. You heard some of us juggle the terms gene transfer versus gene therapy because when we call it therapy that implies efficacy and obviously we're still searching for that. What I'm really very impressed with with the speakers that we've had, these are intrepid investigators. They are passionate. They are driven. I was really moved by the intensity that


they bring, the rigor, the skepticism, the transparency, the willingness to open their work to independent and very critical peer review and oversight. I think that's something that we really have to applaud. Obviously there is always that tension between trying to have appropriate oversight and not hold back the pace of innovation. That's just to give you folks a chance to put some of your ideas together because we'd like to hear from you to ask questions about the common themes and challenges of the morning. We do have a couple of additional members. Let me do the proper introductions of our additional members here unless of course you want to introduce yourselves. We have Dr. Chen first. Dr. Chen, would you like to introduce yourself? DR. CHEN: I'm in OCTGT and I review preclinical studies. DR. MAYBEE: David Maybee. I'm a clinical reviewer and also in OCTGT primarily in the oncology area.


DR. DALEY: I might ask the speakers to each take a few moments very briefly to address a question and maybe even to cross-address it based on some of the issues that were raised by the other talks to give their impressions of some of the other talks. I think one of the issues that is a theme across all of these is that these are highly innovative strategies for treating disease, in some instances diseases for which there is either early fatality or no hope of treatment by a standard strategy. What is the burden on us? What is the burden of preclinical evidence before we move forward to subjecting humans to the risks of clinical trials? What are the principles on which we base the review? This is probably relevant to those of you who are coming from the IRB perspective. What are the principles and standards that we have to apply in particularly innovative areas of cell, tissue and gene therapy around the scientific rationale and the preclinical substantiation of evidence in order to move forward? I might ask that as a general


question and we might start with Helen to make a couple of comments and we'll move down. I don't want this all to be us talking. We'd like to have your thoughts as well so please come to the microphone with your questions. DR. HESLOP: I think you have to -- on the indication because often you don't have a good animal model that would be a predictor for what's going to happen in humans. I think I gave some illustration of where preclinical models would give you the wrong impression in terms of predicting toxicity where it didn't occur clinically. Conversely there are some situations where you may be reassured by a clinical model such as the TGen experience, but then the animal model didn't predict toxicities which did occur in humans. I think you need flexibility depending on the clinical situation as to whether an animal model or in vitro studies are the most appropriate testing. DR. DALEY: We'll ask our surgeon as well because obviously surgery has a very long


tradition of innovation in real time right in the surgical suite. DR. BREUER: The first thing I'd like to point out is that I do come from a little bit of a different world. In the operating room you're not regulated to a large extent. For instance, one of the alternative vascular grafts that people use is called a pericardial roll. What the surgeon does is he creates a vascular pedicle pericardium, the sack that the heart lies in, he sews into a tube and he uses that as a graft. Only a small fraction of patients have enough pericardium left by the time they come to the operation that we're studying in order to do that. But it's amazing. A surgeon came up with that and he did it and he wrote about it and other surgeons do it and that's the process. I felt a little bit like a barbarian coming in to Rome the first time I walked into the FDA and the whole notion of quality control and quality assurance was a new concept for me. That's been very valuable and I'm trying to bring


quality control and quality assurance to surgery now which is not going to be a simple venture, but an important one. From the standpoint of quality data and what do you need to do in order to bring something to the clinic, I think it depends most importantly based on what is currently available. If you have nothing available the quality of data that you're going to base your decision to go into the clinic will be less than if you have something where you're just making a small incremental improvement. I'm a big fan of preclinical work using both in vitro work and animal work and I think you always need to use a variety of animal models in order to move things forward. DR. DALEY: Can you comment quickly on what surgeons do in the process of obtaining peer review of an innovative procedure? Is there an accepted standard in your community of how that'd done? DR. BREUER: No. I think that's all governed locally. If you look at the


morbidity/mortality process, if you look at medical morbidity/mortality versus a surgical morbidity/mortality, in a medical morbidity/mortality conference the patients die from their disease no matter what the patient died from. In a surgical morbidity/mortality conference the patient died because the surgeon screwed up and that's always the assumption. Neither is obviously true, but I think within surgery that people are very independent, but I think that also accept that responsibility and the costs that come along with it. DR. KOHN: It's a difficult question. There are no perfect models and so I think you have to use the best available animal or human cell models to get what is a reasonable proximity of what you're proposing to do in the trial. The issues with vectors, the tropisms of vectors in rodents will be different than in humans, et cetera. I think that preclinical data is always imperfect and you just try and get the best possible data you can with the available models.


DR. DALEY: Ron, you have what seems like a fairly good mouse model for Batten's. What's the level of efficacy one needs to see in that model? Have you seen it? DR. CRYSTAL: There is a mouse model and of course the best efficacy measure is survival and that's what we've used. There are a whole variety of others as well that you can do on a morphologic basis, on an MRI basis or you can do it biochemically. In terms of animal models I think for safety purposes is probably is very, very useful. Although in our first clinical trial we ever did which was in 1993 for cystic fibrosis, we were fooled in terms of the volume that we used. We had done the studies in nonhuman primates with no problems. Our fourth patient developed an inflammation of the alveoli because we were using too much volume, something we totally missed in the experimental animals. I think probably doing two different species is a wise thing for gene therapies. For efficacy I feel less strongly I think. There are


multiple examples where humans are not just big mice and where the mice may not predict. So if the model is available it's great to do it, but if it's not -- and I think also we can get stuck on somebody develops an aardvark with a knockout aardvark with the disease and all of a sudden that becomes the standard at least in the academic world that you're forced to follow and that may be useful, but it may not be. DR. DALEY: Comments from Dr. Maybee or Dr. Chen? DR. CHEN: I think everyone on the panel all say all of the points that I think are what the FDA considers true. What we usually suggest is that using the existing available animal model and it has to be biologically responsive because products are different. You have cells and you have vectors and you have to consider the trans gene whether there is species specificity among different trans gene products. So we usually ask the sponsor when they select the animal model to provide the rationale.


Also, I think in terms of the safety analysis assessment, you not only need to assess the product related issue, you also have to assess the surgical procedure related to safety. Those are the minimums. In detail usually we work with the sponsor -- and they submit the protocol we put into the discussion to see if there is any issue that's from the sponsor that we can discuss and see if there is any possible solution to resolve it. DR. MAYBEE: For those of you who have had a lot of experience in the review process, you know the answer to this already. It's worth stressing what Theresa just alluded to which is that in the preclinical setting and in pre-IND evaluation you have a perfect opportunity to review prospectively with us what we would consider to be the important evidence from a safety point of view, whether it's in vitro or an animal model. This is the time to address these questions so that you're not caught flatfooted with our usual very negative response if in fact


the information submitted is in our estimation inadequate. So I think you have to start with trying to get at these issues earlier. Dr. Crystal's proposal to use a screening protocol is very interesting because in fact that's exactly the kind of thing that might be addressed in the process of this. You get patients in and you evaluate them for eligibility certainly, it will be the time to inform the parents and to get into many of the ethical considerations that we're talking about today. But indeed it also allows the chance to react about what you have in the way of information in your arsenal to support safety and efficacy in the principal evaluation and I'd like to stress that. DR. DALEY: Thank you. Let's get some questions here, please. Speaker: One of the things that struck me after this morning's earlier presentations about benefit to the individual in the clinical trial, benefit to pediatric populations in general if there was not


necessarily any potential benefit to the individual. Then during Dr. Kohn's presentation about having a situation where you have an alternative therapy that's available for children, it struck me that we potentially put ourselves in a position where as we begin to develop therapies, but they're not optimal therapies, we put ourselves into a bind where the bar becomes higher and higher. Now we have a therapy, we can use it, it works, it works sort of, but it doesn't work well enough. But then trying to justify doing something that may be higher risk to a child becomes harder and harder to do. I wonder if we by setting aside or perhaps deprioritizing the societal benefit or the benefit to future children we don't wind up harming those future children because we limit their opportunities to just what's good enough today. DR. DALEY: That's interesting. Does anyone want to comment on that? It's a changing landscape. Don, you've faced that in your work. DR. KOHN: I think it is a problem and


you have to change with the times. I guess your question is about can we justify some of these things based on the common good rather than the good of the patient. I still think that principle probably still holds for pediatric patients that it has to be in their interest, also. It would be that the prospect of benefit is even higher potentially with this new approach to justify it. I would find it hard to rationalize that it may not be in this child's benefit, but it might be good for future children. I think parents sometimes will say that and may also believe it, but I think ethically we can't operate on that with children, that we have to be protecting that individual who we're dealing with. DR. DALEY: How about let's go in the back and then we'll come to you, Dr. Nelson. SPEAKER: I have a question in two parts that follows-up on your introduction to the panel discussion where you mentioned the concept of access. Dr. Heslop mentioned this in one of her first slides that there are approaches to promote


the idea that children should have access to these therapies. I see access as being in some way in between the idea of the ethics of what is right and the FDA legal considerations what you're allowed to do and this is what do you have the right to do. Neither the legal discussion nor the ethics discussion this morning talked about the rights that children have to access to research whether or not that right actually exists. The NIH has covered this in some ways by saying that when it's federally funded research they have a algorithm to say that you have to justify the exclusion of children in a clinical trial and not justify the inclusion of children in a clinical trial. The first question is, is there a right to access to research and if so how do we balance that within the other frameworks of ethics and regulatory management? The second is under that, a right of parents to make that decision. The AAP in the last few months has released a document suggesting that the guidelines for normal donors in transplant or in organ transplants


should have a third-party adviser and that that third-party adviser has the right to trump the decision of the parent when they feel that it's in the best interests of the child. I think that's a very slippery slope in who has the right to decide for children. If you comment again in that particular instance of right where we have rights since many of you discussed in your talks this morning having a similar third-party person where that fits again in our discussion of ethics or legality in how we include that. DR. DALEY: We have two parts. We're going to first talk about the right to access and whether that exists. Helen, you mentioned that you had a trial where you had six pediatrics, six adult and I wonder whether that's driven by this altruistic notion that a pediatric patient should have the right or whether that's serving the requirements of the scientists and the investigator to expand their patient population and maybe you could speak to both of those issues. DR. HESLOP: What I said we had


six trials that were pediatric only and six that were adult only and that was based on the target diseases. DR. DALEY: Obviously when you're talking about very, very small patient populations there is some tendency to want to expand the numbers of patients and that might include going to both pediatric and adult populations. DR. HESLOP: In all our other studies we have included both pediatric and adult patients where they are in the populations that we're targeting because of the access issue and obviously the considerations discussed earlier this morning have been addressed, that the trials have to be approved as having potential benefit -- more than minimal risk. If that's so we believe there should be access to children who are eligible for the study as well as for adults. DR. DALEY: I think that the right to access is driven by the probability that there will be some benefit directly to the individual as opposed to --


DR. HESLOP: And these are all studies that have gone through the review process and have been put in the category as providing potential benefit. DR. DALEY: What about comments from the group on this issue of the value and challenges of using a third party in the informed consent process and this interesting issue that was raised that seems that in some instances there might even be a case that the third party could trump the opinion of the adult? DR. HESLOP: I think that's very difficult. In the transplant setting that was raised, this is a recommendation that has come about despite the fact that there haven't really been instances of harm to the donor in the transplant field. Personally I don't think that a third party should be able to have that decision-making capability and take that away from the parents. On the other hand though I think in IND studies it is helpful to have an advocate there for the child, not somebody who takes over the


decision-making mechanism, but that can represent the child. And I think if there were a situation where there was a discrepancy between the child's wishes and the parents' wishes then you'd obviously have to work that through and not enroll the patient until that was resolved. DR. KOHN: I haven't read those transplant guidelines, I've heard about those and I thought that some of it was to have the medical evaluation of the donor's eligibility to be done by a third party and that certainly makes sense to take that away from the transplant physician and have a primary-care pediatrician assess the child's well-being. I think whether to be a donor for a sibling obviously belongs in the family with the parents and we know children of various ages like to say no to whatever you tell them to do. So I think a parent certainly would weigh what's best for their two children. I'm not sure exactly what the guideline says. I don't know if it's now law or it's a recommendation coming along. DR. DALEY: Does anyone else want to


comment on that? Dr. Maybee? DR. MAYBEE: I had several thoughts. First off we have as you know evolved a very specific process now for looking for you might say new entities and old entities, old pharmacologic and biologic entities that should or have consideration for pediatric access if you will. That is something that a number of us participate very heavily in who are right here in this audience. But that's for products that have already been marketed or are in the process of being marketed. For earlier access, expanded access if you will, there has been a liberalization of many of the regulations and guidance that relate to that with an intent to try to allow at least the consideration for that kind of access, but that takes them outside the standard investigative role of the protocol itself. It's a question of when you say right to access what exactly is the stance that the patient is going to be in in that setting, investigational still or is purely


because this family or advocate or kid heard that this might be a real good way to go and it seems better than getting chemotherapy or getting some other odious treatment that might not appeal to them. The issue of access from a pure practical point of view is something we face all the time. We have many, many individual requests for this for various kinds of experimental therapies and some of them have to do with the kinds of agents we're talking about today. I'd have to say that looking at those abroad there is tremendous variation in what the expectations are, what has led to the decision to contact us in the first place, how aggressive are they or who is pushing it. These are very difficult individual issues to address. I don't know that we have any regulatory or guidance-driven new methodologies that we need to allude to here in that regard, we have a number of people who could speak to that, but I'll leave at that that this is an issue that sits very high with us in terms of what we see coming in the


door. Many of the ideas voiced at this table are good ones in terms of trying to determine whether that would be an appropriate route to go. DR. DALEY: Dr. Nelson? DR. NELSON: I'd like to hear some further reflections on the part of the panel particularly the investigators around this complex question of the severity of the disease and choice of subject population, balancing which is true often that those who are less severely affected might have a greater opportunity for benefit though the risk of the product may then present a greater risk to that same population. And within that context to at least set aside the OCT trial since that was as much a decision to proceed in adults and not children and therefore set aside the risk-benefit issues in pediatrics than it was a question of mild disease versus severe disease although those were obviously co-variables. I'd be curious because clearly in Batten's disease one has a choice of going to a less-severe population, just to hear some more reflections on that kind of


calculus if you will. DR. DALEY: Don, would you like to take that? DR. KOHN: An example of another disease to talk about is adrenoleukodystrophy for which there have been both successful bone marrow transplants and now gene therapy. That's a disease where the natural history now is well known that boys will be relatively asymptomatic and then start showing symptoms and they have a relatively rapid neurocognitive decline. The experience from bone marrow transplant is that once you intervene with the transplant there is continual progression of the disease for a while and then there's stabilization so that if a severely affected patient is treated their neurologic function will not improve and there the standard of care has been to try and intervene as soon as progression begins. That's now been applied in a gene therapy trial very successfully also and so I think that's a case where the lessons learned in allogenic transplant that early


intervention is important was then applied for the first gene therapy trial that was done just like successful allotransplants except using autologous cells with gene correction. I think part of it is the message that we can build on prior experience and if we know that a parallel therapy works early then it would make sense with gene and cell therapies to apply them early. DR. DALEY: Ron? DR. CRYSTAL: I agree absolutely. The animal studies and all our clinical experience is that the earlier we get these diseases and if we could do the newborn we probably would have a much better change of success. On the other hand, we have to balance that. The reason that there are 150 people in this meeting is because this is gene and cell therapy and people die all the time from drug trials. I think the Gelsinger is a very good example of what happened to our field. Gene therapies and cell therapies are something that the public is not yet fully comfortable with and so we have an emotions and a variety of other


things that come into the equation. That's why I recommended starting with severe not because I show efficacy, but it's the adverse events and toxicity that we have to worry about and I think that we are better served by doing that as we do in cancer trials where we start with late-stage disease. DR. DALEY: I know that issue has been raised in the spinal cord injury trial. There has been some criticism in the stem cell community of choosing that nonfatal condition as a first in human for the products of an embryonic stem cell. Obviously on the counter side is the sense of how risky is that product and we really won't know I think until we some more experience. John? DR. THOMAS: This issue I think has certainly been addressed this morning and already touched on in other questions, but I still hear the idea that in some investigations people feel it's appropriate to try a therapy where it applies to both a pediatric and an adult population right away, whereas other speakers talked about perhaps


it's better to start with the adult population to get some experience before moving into the pediatric population. I'm wondering is there agreement on this from the panel that all these new cell therapies for example can be tried in pediatric and adult patients at this time or do we really need to start with adults first? DR. DALEY: Is there any reluctance to initiate a new therapy in a pediatric population if there's not a comparable adult disease spectrum? Does anyone want to comment on Dr. Thomas's question? DR. THOMAS: It sounds like we have good agreement. Dr. KOHN: I think that for many of the disorders, neuroblastoma, congenital heart disease, SCID, Batten's disease, all of these are diseases that there aren't adult patients to treat, that you either get definitive therapy as a child or you don't make it to adulthood so clearly for those disorders it will be children first. For conditions like Hodgkin's lymphoma that affect


both age groups, there it depends how much this is really a Phase I safety study and how much is a Phase I/II where you really have evidence that there's going to be efficacy. If there is the potential for efficacy I think then it becomes almost the issue that you shouldn't deny children that potential efficacy. I think it really depends on what the disease is and whether there are adult patients and then what the evidence is that you may have from prospective benefit. DR. DALEY: I presume part of the exclusion of pediatric populations from the spinal cord injury trial was this issue of unknown risk for the new product, but it also in that setting raises some very interesting questions about informed consent. I should think one of the messages to come out of some of the discussions this morning from the perspective of you who are in the IRB community include how fluid and evolving the literature around informed consent is and how much our assumptions have to be put to the test as to whether or not the acute situation in the context


let's say of spinal cord injury allows you to even make an informed decision. Let's take a question in the back and then we'll come to the front. We've got about 5 or 10 more minutes only. SPEAKER: Coming back to the discussion raised by Dr. Kohn, when we're facing to balance existing treatment with this new treatment, the existing treatment such as -- new treatment and it's the issue in ADA deficiency, but maybe it's also in -- having a chance to have -- I would like to know what we have learned from the experience in the mid-1990s when the -- moved for example -- from palliative surgery to -- knowing well that at least for a couple of years the mortality risk has been -- I think it's a good example that at that time we moved from already existing treatment palliative surgery and then we moved to -- surgery and in between there have been many years, a couple of years during which there were uncertainties about effect -- mortality risk so what did we learn from that? DR. BREUER: I think that's a great


question. I think one thing it highlights is what an individual can do. When you look at the fathers of congenital heart surgery, I don't know what these men were made of. How do you operate on a kid and see a kid die and then wake up next morning and do it again? It's incredible, but I think it highlights the fact that they truly believed in what they were doing and that an individual can make great progress. But I think we're in a different phase now. I think improvement now is going to be incremental so that we need to be much more circumspect as we move forward. DR. DALEY: The question is often raised whether or not things like congenital heart surgery, bone marrow transplant and other very, very high-mortality-associated procedures which are now accepted as standard of care could have evolved under the current regulatory structure. I think they could, but it may not work quite in the same way. SPEAKER: I had a specific question


about escalating dose in clinical studies specifically for cell and gene therapy because in this instance usually a follow- up treatment really isn't an option so if your preclinical studies support the safety of a given dose then are you doing the patients a disservice to start out at a lower dose that would be subefficacious? Maybe you can comment on the ethics and the logistics of that. DR. DALEY: The ethics of dose-finding studies. Dr. Maybee? DR. MAYBEE: I would have been interested in hearing what the rest of the panel had to say about that. I'd probably make a distinction between cellular therapies and gene therapies in this particular one. With cellular therapies we have some fairly definitive guidelines that we usually offer to sponsors to consider in their trial design and those have to do with the range of dose that has proven over time to be adjustable, we'll say a log increase or less for instance. I'm just giving you an


example. In terms of dose escalation, there are other bells and whistles that we can put in place of course in evaluating any trial that might allow some deviation from a relatively controlled-dose escalation plan which would have to do with staggering patients or possibly creating a window of observation in between dosing cohorts. These have all been alluded to today. We have various ways of trying to take a given trial submitted to us and react to it in relation to cell dose. It's a lot trickier with gene doses. It was pointed that cellular dosing can be repeated, but with gene therapy it's usually just one application. Again the initial preclinical work that comes in that would frame that and the extent of the information available in total particularly regarding safety is the key I think in that for gene therapy. It's hard to answer that question in a more specific way for gene therapy. DR. DALEY: Helen? DR. HESLOP: I'll add that cell therapy can be quite complicated in that sometimes more is


not necessarily better and in some of our studies we've actually deescalated because it looks like we have better results with lower doses where the cells have had -- vivo culture. It's not like pharmacology products where more is considered to be more potent. DR. DALEY: I think if we think about the experience with dose escalation in highly targeted kinase inhibitor therapy for cancer for instance, it's very well informed by very strong preclinical evidence of a dose response relationship and we should strive to get that for various cell therapies. We're going to probably only take one or two more. SPEAKER: I wonder if the panel can address the ethical issues surrounding exclusion criteria for a clinical trial that's ultra orphan like in Batten's Disease for example where you have 200 people worldwide and you have a handful in the U.S. I know that Dr. Crystal had mentioned that and gave advice about separating the PI from the consent process in clinical care, but in all


actuality when you're inviting families to come for assessments aren't you then picking your age-matched control group? There is a great ethical issue surrounding that. The natural disease history study for a disease like Batten's could be used as you barometer to show efficacy against. I wanted everyone's feeling regarding setting your own control group by excluding people from the trial. DR. DALEY: That's challenging. Dr. Crystal? DR. CRYSTAL: There are a whole variety of issues that relate to that because the tension is between trying to develop a trial that will be informative in terms of moving forward versus the issue of you don't know and therefore why not just take all comers. There are certainly exclusion criteria that you have to build in from a safety point of view. Then the other exclusion criteria come primarily to try to develop as pure a study as possible to inform decision making in the future and it's a real challenge and I don't think


there are any simple answers for it. But when there are set up at least in our experience of exclusion criteria it's to try to make the trial as informative as possible to making future decisions in terms of direction. SPEAKER: This is a commentary from the parent population, not a question, but a plea. A plea to our government agencies, a plea to our institutions, advocacy groups, medical professionals, et cetera. Please continue to work on communication and education in working with parents and families. I can tell you from my own personal experience that finding clinical trials has been difficult, trying to understand the clinical trials when you go on to NIH sites, et cetera, they're very difficult to understand. They're not written in layman's terms. You don't understand if it's appropriate for your child or not. People are finding out about trials from parent to parent or one to one. There are not good forums. There is not good communication. People need to understand if the trials are


appropriate where they can find them and how they can access. It's a very frustrating thing and I know that people are working in these areas so I'd make a plea to continue to work in those areas extra hard to make it as accessible as possible. Thank you. DR. DALEY: Thank you very much for that comment. I want to thank the panel for all of their comments. I think it's highlighted the challenges. I think we don't have answers, we have a process and I think that we have great confidence that the regulatory structure is going to provide us with the kind of review, oversight and transparency that's absolutely essential. The one issue that I want to stress which I think is relevant to these very arcane, very scientifically sophisticated new approaches to treatment is the requirement for relevant area-specific expertise in all of our review. I think there is tremendous disagreement in the community of stem cell biologists in particular that I can speak to around the kind of


rigor of preclinical efficacy for many of these new types of treatments and those standards have to be respected before thinking about moving into clinical trials. I want to put in a big plug for area-specific expertise in the context of your reviews and that has to be balanced against the fact that sometimes there are very few people in the world who even have the relevant expertise. This has been a great discussion. (Recess) DR. WITTEN: I hope everyone had time at least for a quick lunch anyway. It's my pleasure this afternoon to introduce Dr. Jeffrey Botkin who's going to be moderating the afternoon session. Dr. Botkin is a professor of pediatrics at the University of Utah and an adjunct professor of human genetics. He's chief of the Division of Medical Ethics and Humanities in the Department of Internal Medicine. He obtained his undergraduate degree from Princeton University, his MD from University of Pittsburgh, and MPH from Johns Hopkins. He's the associate vice president for


research integrity at the University of Utah with oversight responsibilities for IRB, conflict of interest, responsible conduct of research, biosafety and research ethics education. His research and publications are focused on the ethical, legal, and social implications of genetic technology with a particular interest in research ethics, genetic testing for kids, susceptibility, newborn screening, and prenatal diagnosis. He's had a number of appointments in organizations related to bioethics and stem cell therapy, and currently he's a member of FDA's Pediatric Effects Advisory Committee, so Dr. Botkin, I'm going to turn it over to you. DR. BOTKIN: Dr. Witten, thank you very much, and my compliments to the FDA on a terrific meeting so far. This afternoon's session is IRB oversight and of course IRBs are the mechanism by which we provide peer review to these sorts of studies and others, that being a foundational


notion in our research ethics system. The IRB has come up a number of times through the discussions today. We've heard quite a bit about science, heard quite a bit about the regulations and the ethics, and it's really the IRB where the rubber hits the road. It is to a large extent the gate keeper that gives the final thumbs up or thumbs down as these protocols move towards recruitment of patients and implementation of the research protocol. So, very important to think about how this system works and we're going to encourage our speakers through the afternoon and in the panel session to be as innovative as possible, thinking outside the box. Frequently we think quite specifically in terms of the regulations and appropriately so, but we have an opportunity, I think, to think about those regulations in the context of this very interesting, innovative, and important domain of research. So, we have several speakers who will present to us over the next hour or so, and then


we'll have a panel session. Our first speaker is Angelica Walden who has an MBA. Angelica's daughter is relevant to her opportunity with us here today, Asia, who I understand now is 11-1/2 years old. She was 530 grams and about 22 weeks gestation, so quite an extraordinary young lady, had endured several years of difficulties with respiratory illness, developmental delays, vision complications and other issues related to her prematurity. This has been an extraordinary experience, I'm sure, for Ms. Walden and she's now working within the healthcare field. She is in quality management for the MCG Health as a service quality coordinator and patient representative, voting member of the Human Assurance Committee, which is the IRB at the Medical College of Georgia. She's also served as a patient representative there since January of this year for pediatric respiratory illnesses and this rich personal background and professional background gives us an opportunity to hear her thoughts about


advocating for pediatric subjects on the IRB. Ms. Walden. MS. WALDEN: Hi. You heard a little bit about my background. I am a parent of a formerly chronically ill child, I'm a voting member of an IRB, and I do work for the organization that helped my daughter to exist and be here today. Some of what you'll hear me say today, I'm primarily talking to you as a mom, but I will admit that some of my perspective is influenced by my participation on an IRB and kind of knowing some of the background from that. And that smiling face is my inspiration. Asia did, for six months, present quite a challenge for a bunch of people in the neonatal intensive care unit, but I'm proud to say today that she is a perfect 11-1/2-year-old and she continues to be inspiration for me professionally and personally. My concerns when considering high risk clinical trials such as cell and gene therapy, those six months that Asia was in the neonatal


intensive care unit greatly influenced my thoughts about these topics -- about this topic, for two reasons. Number one, I recollect the feelings that I had as a mom making a decision not only for my baby, but for the potential person that would exist for years down the line, today, as a matter of fact. I also consider the fact that my daughter may not have been here if some parent had not made the decisions, not unlike those that parents that will be making decisions in these types of circumstances will make. One of my -- my primary focus as a parent would be about my child's safety, the safety mechanisms built into these protocols that involve these children. As a parent I'd want to know how much I would be informed of what was going on with other subjects participating in these trials. I want to know that I'll be told about what's happening with other children and I'd want to know how soon I'd be told. There's been a lot of talk today about data safety monitoring boards and data safety


monitoring plans, I'd have to say as a parent I'm especially, when you consider protocols that are 50.53, 50.54, I'd want to -- I'd be much more comfortable as a parent knowing that there were data safety monitoring boards comprised primarily of clinicians and not of researchers, so that I can be sure and be confident that my child's safety was being monitored. I'd want to know that I was getting adequate background information so that I could make a good, informed decision about whether or not I would want my child to be involved in these types of protocols. I will tell you that as a parent, I'm going to Google it. I mean, even if I don't understand a word of what you're saying or a word of what I find on the internet, I at least want to hear some similarities in what I'm reading and what I'm hearing from you, so that's something to consider, what's out there on the net. That might be a great foundation to begin conversations with parents in the informed consent process. It


builds trust, helps to build a foundation for them to draw upon when they're making these decisions. Am I adequately informed and am I given an adequate opportunity to ask questions? Sometimes it's the hardest thing to do to sit when you're in front of a clinician hearing very difficult information to understand, it's hard to formulate questions in your mind. I also work as a patient advocate and I'll tell you that a lot of physicians call me to come in and sit down and help them facilitate conversations with patients and families about very difficult informed consent and also about treatment options that don't necessarily require consent, but just to help facilitate the conversations. Sometimes it's hard to understand when parents and when patients really understand what you're saying to them. Sometimes it takes an outside person who is an expert in doing that to sit and just observe that. You need to consider your setting. Sometimes it's not the best time at the bedside to sit and have these conversations, or in the treatment room. I


want to know that you as a doctor or a PI or whomever is doing the informed consent, I want to know that you're making a good, genuine effort to connect with me as a parent or as a subject in a clinical trial, especially in these types of trials, to help me understand as much as possible. Consider, you know, sitting down at a table, you know, with another facilitator there who, like I said, is an expert in looking at and making a decision as to whether or not -- not really scrutinizing the information that you're giving, but making a good assessment of whether or not your audience is actually getting what you're saying. Give them an opportunity to ask questions. Sometimes it takes that outside person to look and be able to tell, you know, Mr. and Mrs. Smith, do you have another question? Sometimes when I'm looking at parents and they're hearing information from doctors and the doctor is asking, do you understand? They're saying, yeah. I understand. And then they go right on talking


because they don't recognize the body language that people display when they're not really understanding. In clinical trials like this, this information is very difficult to understand. It's very unlikely that they're not going to understand what you're saying and sometimes they just need the support of that other person there to help guide them through that process. These children are separate beings. As a parent, when I was asked to make difficult decisions about Asia's treatment or about other -- about medication or diagnostic testing, or as these parents will be asked to make a decision about whether or not they what their child to be a part of something that's very risky, the constant thing that's on your mind is that this person is not only your child, but they are a person, and they have their own potential for life, and they have their own purpose, and not only are they your child, but 10 years from now this person will be a different person and you want to understand, and you want your child to understand, that whatever


decisions you made, whether they had good outcomes or bad outcomes, you want them to know that your intention was to give them the best opportunity for life. She is a person, too. Child assent is very, very important to me as a parent and as well as should be for everyone. I want to know -- first of all, let me say this. Children nowadays, with technology, are more informed than you can ever imagine. One of the worst mistakes we make is to make the assumption that children have -- really assessed what their capacity is for understanding. Nowadays when Asia comes to me and asks me something, she's went and done all the research, she knows more than I do. Mom, can I be on Facebook? Did you know that there are 50,000 kids in Georgia that are on Facebook? You know, they come to you with information, they know how to Google things and look at things and they have information in their capacity with technology and everything, you know, that's available to them, their capacity to understand is different than it


was 10 years ago. So you have to take that into consideration, and I as a parent, knowing that, it's important to me what she thinks. It's important to me that she understands and that the clinician or whomever is participating in these consent processes are actually making an effort to connect with her, and engaging her in the assent process. Does she understand that there are other options? We talk about trials and we talk about standards of care. You know, I would be more comfortable as a parent if my child understood -- to be sure that my child understands that you don't have to do this, you don't have to say yes to this. These are the other options. And if there are not other options, well, I want her to know that, too. So, in conclusion, as a parent I want to know that all measures that are possible to ensure my child's safety are being considered due to the nature of gene and cell therapy. It is important that parents have confidence in the information


that's given to them, be really conscious in how you're delivering all of this information to them, and also understand that it is a great burden for parents to make these decisions. They're going to go to bed with them every night, they're going to stay with them for the rest of their lives, and the way you deliver that information to them is important. Thank you. (Applause) DR. BOTKIN: We have time for several questions. Our next speaker is Dr. Jeff Kahn. He's professor and director at the Center of Bioethics, University of Minnesota, where he holds the Maas Family Endowed Chair in Bioethics. Faculty appointments are the medical school Department of Medicine, and he also has appointments in public health law, the Humphrey Institute, and the Department of Philosophy. Jeff has a long history of outstanding work in the field of bioethics. Early in his career he was the associate director of the White House Advisory Committee on the human radiation experiments and has served on a number of


committees at the national level in the meantime and currently, relevant to his presentation today, he's a member of the Recombinant DNA Advisory Committee. Jeff's talk is "Ethical Considerations in Pediatric Gene Therapy Trials: Perspectives from the RAC." Jeff? DR. KAHN: So while this is pulling up, let me say thank you for the invitation. I've enjoyed the day so far. I should say, by way of disclaimer, it probably doesn't need to be said but I'll say it anyway, these are my views, these are not the views of the RAC. I have served on the RAC, I think, now for four years. I'm about to finish my term so I have some experience in that regard, but I'm certainly not here to speak on behalf of that committee or OBA or anybody else. So you heard my background. I come to this as someone who does ethics of research primarily. When we were talking about the presentations for today one of the requests that


came to me was that I spend a little bit of time talking about the RAC and how it does its work. I should say too just in sitting here through the course of the morning I had some thoughts about modifying the slides, so what you're going to see here are a little bit different than what you have in the books, so it'll keep you on your toes. It's more elaboration than it is really any change, but that's why they look a little different. Alright, so you've heard the RAC invoked a number of times in the course of people's comments today. I'm going to not, I think, disagree with anything that you've heard said but maybe give a little more context and background. The RAC has a particular role in the oversight of gene transfer research. We try not to talk about it as gene therapy, as you heard Don Kohn say. We talk about gene transfer, not gene therapy. I'll say a little bit more about that later. The RAC is the Recombinant DNA Advisory


Committee. It's advisory to the director of the NIH. It does not have regulatory authority, so that's an important feature of the RAC. The guidelines by which the RAC does its business are noted there, easily found on the OBA website, their so- called NIH guidelines and you can see them in all of their many page glory on the web. The RAC, of course, has its work not in isolation, but in conjunction with really, and in relationship, with the FDA, which does have regulatory authority, so there's a collaborative relationship between the two entities. That's, I think, an important part of the feature of the way the RAC does its work, and you saw a couple of the speakers earlier today talk about the many levels of review and oversight and that's, of course, part of the trick, is how to manage that relationship in ways that doesn't overly burden investigators and institutions, but does the work that the RAC is there to do. I'll talk a little bit more about that momentarily.


In addition to relating to the FDA, there are, of course, relations to the institutions and the oversight that happens locally, so that includes the IBC and the IRB -- Institutional Bio-safety Committee and Institutional Review Board, for those of you who need explanations about the acronyms. I will say, having served on IRB at multiple institutions including the University of Minnesota where I now am on the faculty, there's a sense of burden on the part of the investigators and when we talk to people about the RAC they see it, as we've heard today, additional burden, not only about local oversight, but federal level oversight, and that, I think, has some reason for background and history. So, why special treatment for gene transfer? One of the things that happens when you join the RAC, and I hope many of you will have that opportunity in the courses of your careers -- how many meetings now, Jacqueline? A hundred and -- I didn't hear the answer but -- it's 123


meetings of the RAC in the course of its history. When you join the RAC you're given, among other things, a book about the history of recombinant DNA regulatory oversight and research by Don Fredrickson who was, at the time, the director of the NIH. I was actually reading that book on my way home from a RAC meeting early on in my appointment and I thought it was quite interesting, in fact. It outlines the history of the Asilomar Conference and the effort by the recombinant DNA scientific community to regulate itself, more or less. The reason I'm telling you this story is actually -- believe it or not, there's a punch line at the end -- as I was sitting on the airplane flying back to Minnesota, the person next to me was reading what was, obviously, to me at least, a very dry looking law review article, I think about taxation, and I leaned over to her at the end of our flight and said, boy, that looked like really boring reading you were doing on the course of this flight. And she said, you should


talk. I had no clue that what I was reading was boring. It was actually quite interesting to me. I think that the answer to the question on this slide -- so to be serious for a moment -- is that gene transfer has gotten somewhat special treatment and there are interesting historical and policy reasons for that. So, one, going back into the history, there was great uncertainty about the risk posed by recombinant DNA research in humans. So, risk about -- risk to individuals themselves, so what will happen when we put new genes into people. There were concerns about risk that were more public health in their orientation, risk to the environment, risk to people around those who were involved in recombinant DNA research. And because of that there was a sense that additional oversight was necessary and important. Now, we could argue that many other areas of research are like that too at the early stages of their development, and one question we


might ask ourselves is, at what point does that not become necessary -- is it no longer necessary -- I didn't say that very well. But I think there are strong and good reasons for creating special oversight regimes at the early phase of these leading edge, somewhat controversial areas of research, and at some point we might ask, do we need them anymore? So, to be provocative as Jeff instructed. Let me spend a little time with you now talking about how the RAC does what it does. So, you've heard it invoked a few times. I think we haven't heard much about this part so I spent a little -- I added some things to my slides for this part of the talk. So, the first question is, what gets reviewed, and you heard that referred to a little bit. That's all spelled out in Appendix M of the NIH guidelines, and it's effectively a selection of the protocols that are submitted to OBA that are human gene transfer experiments. That's the language actually from the appendix, human gene


transfer experiments. So not every submitted protocol gets in depth public review by the RAC. In fact, there is a fair amount of work that goes on before there's a decision even about public review that are requests for additional information from the investigators on the part of members of the RAC to help them better understand what's being proposed, and then once there's sufficient information in people's minds to actually render a decision about whether it deserves in depth review or not, there are three votes in the affirmative necessary before a proposal is selected for so-called in depth, which means public, review at a RAC meeting. The RAC meets four times a year for two or three days each time. When people say that they're on the RAC it has double meaning that way. We spend a lot of time inside of meeting rooms in Bethesda and Rockville. You get to know each other very well because of the amount of time you spend together. The RAC is a FACA committee -- that's the Federal Advisory Committee Act -- which


means certain very specific things about the public nature of the work that it does. You must have public announcement of meetings and you must do those meetings in the public light, so they can't be done behind closed doors, there has to be an opportunity for the public to come, and our meetings are actually webcast so they can be watched in real time from any place in the world. So, who serves on the RAC? It's smart people, maybe present company excluded. There are about 20 of us. The folks who are on the RAC have a variety of expertise, as you would expect. Some of them are researchers who do gene transfer work, some are microbiologists, there are people like me who have expertise and training and do work in ethics research, and so there's a lot of brain power that's brought to bear on the proposals that are brought to the RAC for review. So, what happens when those reviews occur? So again this is lifted from the guidelines. We look at scientific rationale, the scientific content, whether there's sufficient


preclinical data, and then we spend a lot of time -- at least from my perspective, I spend a lot of time thinking about the social and ethical issues. So, someone asks -- one of the questions, I think, from the audience about request for autopsy, and in fact that's a part of Appendix M that needs to be in the consent process. You must ask -- must talk about the fact that there will be a request for autopsy if somebody is a participant in a recombinant DNA gene transfer trial. So, there's a lot of attention paid to the kinds of issues that we've been hearing about through the course of today. I added this slide because I thought it was important. So, what are the effects of RAC oversight? We heard a lot about the additional burdens that RAC oversight causes and I think that would not be -- it wouldn't be fair to stop at that. It's true, there's an additional burden that comes with needing to submit to another entity and then having a lot of back and forth about whether you will then get public, in depth


review, and then once that happens, coming to Washington to present -- there's a lot of work involved, and that is for sure the case. There's a somewhat unique history of recombinant DNA research, as I've already mentioned a little bit, but what you get from the RAC is something very important -- some things that are very important, among them, that there is a public and transparent discussion of the science and of the ethics, and that is unlike other areas of research. There's something to be said for that. There's a way of getting on the record what the issues and concerns are. It actually helps investigators. I think the vast majority -- I certainly wouldn't say all, but the vast majority say after having gone through RAC review that it actually gives them important and helpful feedback to improve what they're proposing to do -- improving the science, improving the way that they will carry out a clinical trial -- there are lots of smart people with expertise in the same room focus on


your protocol for a chunk of time and that is a helpful thing. In addition, there is a protection offered for researchers which I think no one has really mentioned yet today which I think is quite important, which is, an additional level of approval, yes, is burdensome, but it also provides an additional level of protection for the investigator. You've gotten a federal level committee to say, yes, you can go forward under these recommended conditions, and that's a helpful thing especially if things don't go the way that you might have hoped. Lastly, and this is just a short list, this is not meant to be exhaustive, there's a help to the field by doing this kind of oversight and discussion and giving a feedback in a very public way. It helps the field advance. Now, you might say it slows the field down to have to go through this level of oversight and burden, but I would argue that there's also some benefit to having the science move forward in such a public and


transparent way. It's very hard to do an experiment about, you know, are we slowing down or speeding up or not causing any change -- it's an impossible question to answer -- but I think that there are some benefits as well as the burdens that we've already heard about. Okay. Happy to talk about any of this, of course, during the Q&A. So, let me spend a little time talking about the ethical issues that come up in the course of RAC review and in particular related to pediatric trials because that's the point of today after all. We have an ongoing problem that you've already heard about a number of times and that is therapeutic misconception. It's not surprising when we're talking about what are very early stage trials usually for diseases for which there aren't other great options, in particular when we're talking about pediatric populations when there are families who really are in somewhat unfortunate circumstances and looking for opportunities to


help their children get better. So it's not surprising that there would be therapeutic misconception. People want to get better. We have had many, many concerns over the course of the time I've been on the RAC -- I'm going to say more about each of these in succession on a separate slide in a second as well -- about how we think about how we think about inclusion of children in very early stage or early phase gene transfer research. When is it appropriate to enroll children? Of late we've started to see what, for lack of a better way to describe it, is an issue where basic science research, that is research without the potential for direct benefit to the subject, is bundled together with clinical research, and I'll say more about that as well. And so like we've heard already today from numerous others, there's an ongoing challenge and need to balance what are important new approaches to treat diseases in children with a commitment that we've already heard discussed,


that adults should go first. Those who can consent for themselves should go first. Okay, let me spend a little time about each of these. Don Kohn already said we're not supposed to talk about gene therapy, and in fact that's one of the areas that always sort of sticks in my craw, when we hear somebody talk about gene therapy and see a consent form that says "gene therapy" -- it's gene transfer research. We don't know about it as being therapy. It's easy to understand that people have a misconception about therapeutic potential when you call something therapy. So, it's an easy term, for sure, gene therapy, but I think it's somewhat unfortunate because people get the wrong impression or could get the wrong impression. Of course it depends what else you say around that term. We know, as I said a moment ago, that these are trials for diseases with often few or limited other therapeutic options and so the subjects and certainly their therapies are invested in potential treatment. They hope it


will be therapeutic, so it's something we want to encourage but not leverage in ways that leads to misunderstanding. As we've heard also throughout the day and just from our previous speaker as well, there is very difficult and complicated information to communicate. It's always the case in clinical research that that is true. In gene transfer it's true in spades. These are very complicated trials with lots of very difficult to understand science, hard to communicate that at the seventh or eight grade level, especially when people are very sick or their children are very sick and they hope that they will get better. That said, these are promising research approaches and the investigators are very excited about what they're trying to do, and so that excitement gets communicated to the families and the subjects as well, and so that's why we get this potential for misunderstanding. One of the other issues, and I think I heard this mentioned at least once earlier today,


this is what we sometimes in bioethics call the "white coat problem," the people who are caring for individuals as patients may also be recruiting them as research subjects and that's very difficult for people to get straight in their heads. You've changed your role but you look the same, you are the same person, you're talking to me the same way. How do I know that this is not about health care and therapeutic benefit, but this is now about something different? And I think that turns out to be often the case, not in all clinical research, but often the case in gene transfer research because the people who are involved in these trials are the -- you know, one of the handful of experts in treating that particular illness or disease. The second issue -- I think it's actually the third I had on that previous slide, but second here -- is this bundling together of what are basic research components with so-called clinical trials. So, we've had a number of these come before the RAC in the last few meetings. So


studies in children where the research is without potential for direct benefit is being proposed in conjunction with the clinical research, that is the research that does have the potential for direct benefit. So, we now have the gene transfer component which is greater than minimal risk, but has the potential for offsetting direct benefit to the subjects, and then appended to that is a basic science component. And the question is, is that greater than minimal risk, and if so, what do we say about it as the RAC? It's actually not clear. It's certainly an IRB issue and it relates to Subpart D which we'll hear more about on the following speakers, but it's a real question about whether the RAC ought to say anything about it at all, whether the RAC ought to even see it as part of its review, and I think we're still wrestling with that. We've been talking about it and offering recommendations to the local IRBs, but it's not obvious, at least to me, what our role is in relation to these particular kinds of cases.


We've heard now a lot of talk in the course of today about this notion that adults should go first, that we ought to do research in those who can consent for themselves, get first-person consent before doing research on children. I thought I would just offer you a little bit of historical context. I think that Skip Nelson may have had some recommendations from the National Commission as well. This is the commission that was appointed in the mid-1970s in the aftermath of the so-called Tuskegee syphilis study becoming public, so this goes kind of back to the beginning of the creation of regulations -- creation and promulgation of regulations related to protection of human subjects. So, these are just a few of the recommendations from the National Commission's work, but they relate to how we think about research on children. So, they said, "Research is valuable and necessary for the health and well being of children and can be performed ethically." So, we


can do research on children ethically and we should because there's important information to be gained. To do that it must be scientifically sound and significant. "Studies must first be performed on animals and adults," interesting lumping those together, "and older children before infants if possible." So, the idea being we're going to try to do this in people who can give consent first -- obviously the animals don't live in that category -- those who can give consent first, and if we can't do it in adults we're going to do it in older children before we do it in younger children. So the idea being, the oldest we can get into research first, the better. "The risks of research must be minimized. The selection of the subjects enrolled must be equitable, and increased risk requires potential for offsetting therapeutic benefit to individual subjects." So this, you can see, communicated directly into the rules that we have lived with ever since. I will say just from our discussions on


the RAC, and now this is just my purely editorial comments, having spent now many meetings together with folks on the RAC, we struggle with what we know to be the case that we want people to be able to give consent to participate in gene transfer trials if we can at all get there, but children deserve the benefit from research and there are many diseases in which there aren't adult parallels or analogues, and so it's important to be able to offer the potential benefits of research to children as a group, and the question is how we balance that commitment or that desire with wanting very much to learn from the experience in adults first. We had a study -- these are all public records so I'm not telling any tales here at all -- we had a study where there was a proposal to recruit both adults and children who suffered from the same particular cancer in a gene transfer trial, and so the discussion ensued about whether it was appropriate to recruit children at all in a study like this where we could learn first from


adults, and there was a long and quite heated discussion around that issue which led to a split vote on the part of the RAC about whether it was appropriate to recruit children at all in the particular study that we were reviewing, literally a split vote, eight to eight, I think, or nine to nine in that particular case. So, this is an issue that I think we are still struggling with as a committee. So, to kind of wrap up here with some questions and challenges and then a conclusion or two, I want to step back for second and not talk just about the RAC. So, one of the things that people who do work on the ethics and policy of protection of research subjects, and I would count myself among the folks who do that as part of our scholarship and research, there's a kind of a growing discussion about whether the policies in place have overprotected children, so by excluding children from research that does not offer the potential for direct medical benefit to them if there's greater than minimal risk, has meant that


we haven't included children in early phase trials. We all know that. And that's meant that we don't have information about children's diseases from very early phase trials, and as my colleagues in pediatric medicine will tell me, they rely sometimes on trickle down information from adult medicine to try to figure out how to treat children, and that doesn't seem, from their perspective, to be the best way at the treatment of children and their illnesses and diseases. So, I think that's a kind of 30,000-foot question. Can we figure out ways to obviously protect children but also make sure that they are able to access the benefits that research has to offer? That's the balance that needs to be struck. So, how do we balance that protection with the advancement of research on children's health? It's much easier to defend, much easier to justify when adults can go first, but that may not always be possible or even desirable, and I think that's, maybe, the more provocative statement. Is it always desirable that adults go


first? Let me just finish then. I think it is the case in the policies related to protection of subjects, and I know Skip talked a little bit about that this morning, there's a tension now between protection and access to research benefits -- protection from harm or risks in research, and assuring access to the benefits that research has to offer, not only an issue in childhood disease, of course. People have written that women were over protected during the course of the last couple or three decades. That's changed dramatically since the early 1990s with requirements for inclusion of women in research. That's just by way of an example. I think we've struggled a lot on the RAC with the ability to assess whether the proposed approach is at least as favorable as existing alternatives for the diseases that are being studied in children in very early phase trials. If you can't say that, then it's very hard to defend, probably illegal to defend research on


children when there's greater than minimal risk. So, how do we get to that satisfaction of Subpart D and the FDA counterpart? As I said, there's a question in, I think, our collective minds, certainly in mine, about what role the RAC has in relation to what are non-gene transfer components of the research bundled with the gene transfer protocol, especially when there's not a potential for direct medical benefit for the children themselves. Whose role is that to adjudicate? Recommendations to IRBs may be all we can and should do, but that, I think, is an open point for us to consider and discuss. I think with that, I'm done. Thanks very much. (Applause) DR. BOTKIN: We've got time for a question or two, and not seeing anyone at the microphone I'll take the moderator's prerogative here. So, Jeff, I wonder if you can comment just briefly on the layers of protection that are


currently built into the system with the institutional bio-safety committees and then the IRB. The RAC is looking at both ethical issues and scientific issues. Is it appropriate for IRBs to defer to RAC review in their consideration of protocols? DR. KAHN: Do you mean refer or -- did you say defer? Defer. I don't think it's appropriate for IRBs to defer to the RAC, no. They certainly ought to refer to the RAC in terms of both directions, and there's important information that can come from IRBs to the RAC and we presume visa versa although the RAC's job is not to take the place of the IRB and in fact we work very hard not to do that. Often what we see at the RAC is prior to the IRB review so we sometimes don't even get consent documents that are in their final form. We often will give feedback that leads to revised, and we hope, improved consent documents which then make their way to the IRB, but that's not the end of the


story, and ought not be. So, I think that the answer is, so, de facto, no. It ought not be the case that IRBs defer to the RAC in terms of their oversight responsibility. They may accept the recommendations from the RAC as something that they ought to consider and maybe act upon. DR. BOTKIN: And just to clarify some of your comments from the presentation. When RAC reviews a pediatric protocol, do you look specifically to Subpart D, in compliance with Subpart D in your determination? Or is that not part of your mandate? DR. KAHN: No, we certainly do. And we look very hard at Subpart D, and we talk to the investigators about whether what they're proposing satisfies the requirements of Subpart D. Our job is not to tell them whether they can or can't go forward, but to make a recommendation about whether what they've said seems to satisfy Subpart D or not in this particular instance, and then that may make its way into the final


recommendations that go to the director of the NIH and eventually back to the investigator and the institution. So, at the end of the day, it's up to the IRB to approve or disprove whether research ought to go forward and that's not the RAC's role or purview. SPEAKER: I thought it was interesting that you alluded to the traditional exclusion of women because that was the parallel that I was also thinking of in the back of the room, and I was wondering, you know, now that we've included women more and more in clinical trials we find that women are not men, and women's disease is not men's disease, and so probably we can anticipate that the same thing would happen even in parallel indications in adults and children. So, in the interest of potentially parallel advancement, pediatric trials, or at least understanding pediatric disease, would it be possible that we might consider doing screening protocols in children while we are doing actual


trials in adults, and at least that way you would identify the potential patients and you would have some additional information to compare the disease process in the pediatric population and adults? DR. KAHN: I'm not going to speak directly to your question because I don't know that -- I mean, I think that's a fair suggestion and a good point. But I tell you what I will say, I think part of the issue is, are there sufficient incentives to perform earlier research on children, and that's been one of the arguments about why we haven't learned as much about research in children as we might, that the incentives are not there to perform them, although we've now got new incentives -- or newer incentives from the FDA to increase licensure time in return for doing earlier research on pediatric populations. So, that's the kind of policy thinking, I think, that is required. You know, just to say in response to you, women are not small men, children are not small adults, and that's exactly


the right way to think about it and that's what people have argued for a long time. It seems -- it was an easier fix in the case of women than it is for children. DR. JOFFE: Jeff, I wonder if you could follow up on the issue you raised at the beginning of whether the RAC has sort of outlived its usefulness, whether there was a time in history when it was necessary but maybe no longer is? And conversely, the alternative is to say that it's actually a good model for other high-risk very novel kinds of interventions that are like gene transfer in some ways but are not gene transfer. DR. KAHN: Steve, I appreciate that comment, and I actually do think the latter, which is that I think it's a very effective model for leading edge, cutting edge science. What's interesting is we don't really have a model by which we sunset that approach, so I'm part of a project back at Minnesota that's NIH funded around how we should think of oversight of so-called nano medical research, whatever that means, and that's


part of what's difficult to define what counts as nano medicine, but one of the directions that we're heading is to look at RAC-like models and say, for other very early stage kind of novel approaches in humans, we ought to think about some kind of federal level oversight for the positive reasons that I tried to articulate in my talk. So, I'm not one of those who actually thinks that there's so much burden involved with RAC-like oversight that it ought to be done away with. I actually see it as quite valuable and helpful. Whether it ought to always be the case that the RAC continues to oversee recombinant DNA in human, that's, I think, a really important question and I don't know that I am ready to say yes, no, or even maybe. I think we ought to ask that question at particular times over the course of the evolution of the technology, but I do think I would endorse the idea of the RAC-like oversight for other areas of research as they come on line. xenotransplantation might be another, for instance, and I'm just talking off the top of my head here,


so if I get quoted saying that I guess it's my own fault. SPEAKER: I was fascinated by your description of the literally -- the split vote on the RAC and I was wondering if you would go into a little bit more detail about that? What was the split point? And a follow up question, how often does that happen? I sit on -- I've sat on IRBs and, frankly, controverted issues don't normally happen all that often on institutional review boards. So, I'm kind of fascinated to hear about how and what happens at the RAC. DR. KAHN: So, this is all public too, so I have not at all, outside of -- you know, it's not outside of a closed room that I'm having a discussion I ought not. In my years on the RAC it never has happened. I'm looking at Jackie, has that ever -- one other time? Okay, so she says one other time in her -- maybe -- ever? Ever? In your history. So, it's rare. In the case I mentioned it was about whether it was appropriate to recruit children in a trial where they were


also recruiting adults in that it was the same disease and why would we recruit them together when we could do adults first, sort of following the kind of ethical argument that I made mention of. I didn't really argue it, I just sort of said, adults first. And we have gone that way for the reasons of consent. I think some of the people who voted against adults and children being recruited at the same time together thought that it didn't satisfy Subpart D to do it the way they were proposing, so I think there were a couple different reasons why people voted against it but others thought that there was such important information to be gained from research on children for children's health, that it was appropriate to recruit children at the same time as recruiting adults. Is that a fair -- Jackie, why don't you come to the microphone? So, this is Jacqueline Corrigan-Curay, if you don't know her, who's the executive secretary of the RAC who is the fountain of all knowledge related to recombinant DNA in humans. I


say that truly. Dr. CORRIGAN-CURAY: The other thing that came up in this protocol, because it was immunotherapy and it was using a specific type of viral T cells, was whether it was comparable -- first of all, whether the diseases were comparable, which is always -- whether the effect of the therapy would have been comparable in adults and children and whether safety data from the adults would necessarily have been informative for the children, or if it didn't prove to be efficacious in the adults, would that be a signal that you shouldn't move forward. So there was concern amongst those members who were -- that if you went first in adults, and you got that information and you acted upon it in your decision to go to children you might deny a potentially efficacious -- so that was the other part of the argument. DR. KAHN: Maybe you should say one more thing is what happened -- when we deliver a split vote to the director, what happens?


MS. CORRIGAN-CURAY: Sure. So, it was split and we have to make a recommendation on behalf of the NIH. OBA does that for the director. And in that case we actually laid out all the arguments and we made a recommendation to try and enroll some adults first. The real issue was how many adults? Do you have to do a whole trial? Do you do three adults? Do you do four adults? When do you get that data? So, the recommendation was we had to go one way but actually the letter is quite long and it really laid out all of those arguments and tried to give the institution the ability to go and take it anew and look at it. It's important to say that the RAC makes recommendations, sometimes they're followed, sometimes they're not. One of the things that happens because it's public is the investigator responds to those recommendations and lets us and the director and the public know, because it's public, what recommendations were accepted, why and why not. And because maybe there's


information that comes down the line, that changes that assessment. DR. BOTKIN: Jeff, thanks very much. Our next speaker is Susan Kornetsky. Susan's a widely acknowledged leader in the field of pediatric research ethics. She's currently serving as the director of clinical research compliance at Children's Hospital in Boston, a past member of SACHRP, Secretary's Advisory Committee on Human Research Protection, and was co-chair of the Subcommittee on Children, currently serves on the Subpart A Committee of SACHRP as well and was a former member of the Institute of Medicine's Committee on Clinical Research Involving Children. So, Susan can talk about IRBs and Subpart D, a practical view from the field. DR. KORNETSKY: Thank you. I want to thank FDA for including me on the panel. I think it's -- I feel very honored to represent the IRB community but also feel very humbled because there are lots of different IRBs out there making these


determinations and my discussion will be from my own work and also from talking with others. Just a show of hands, how many of you sit on a pediatric IRB? And how many of you sit on a pediatric IRB that has involved a first-in-human gene therapy or cellular therapy trial? Right. So this -- the point really here -- I mean, this is a new area for many of us and so there really aren't a lot of answers and the people who are here are people who are probably more attuned to doing those types of trials. But this is a new area for us as well and what I'm -- my perspective is really based -- I've been working with pediatric IRBs for 27 years, actually at Children's Hospital in Boston, and I have participated in review of both gene therapy and cellular therapy and tissue engineering trials, all that have been first-in-human pediatric trials, not a lot of them, but some of them, and so my comments are going to be based on issues that are more pronounced in first-in-human gene


and cellular therapy trials, although as I listened to the discussion this morning, a lot of first-in-human trials in pediatrics brings up some of the same issues. The topics that I'm going to talk about that come to my mind what really happens at the IRB when we review these things is to talk a little bit about some issues and challenges in the areas of expertise, the multiple levels of review, which has certainly been talked about this morning, the risk-benefit assessment, also continuing review, that hasn't been brought up as much -- there's been a lot of concentration this morning on the discussion about getting that trial out there and approved, but to really think about how to think about continuing review because then you really begin to get some information on how it's going -- and then challenges regarding consent and assent, and then lastly, although not an IRB issue, I think it's a very important issue since a lot of this research is being done in academic medical centers including our own and is


a human subject protection issue for an institution is that many of the investigators are also holding their INDs and serving as sponsors and the responsibilities of understanding what that means, and I think that's a big, big challenge. So, expertise. Many IRBs do not have members with pediatric first-in-human gene and cellular therapy experience. I think just by the show of hands in the room we see that. And so when these trials come through -- and I think about expertise -- I think about expertise in two different areas, one, in the technology, whether it's tissue engineering or cellular therapy or gene therapy, there aren't a lot of people who have experience in that area, and in the other areas, the actual discipline that it's being applied to. I think IRBs do a better job at that because we have different disciplines but we need to have that expertise to understand does this make sense in a technical sense and does this make sense to introduce this given the other therapies


for that particular discipline. So you really need both expertise. Fewer trials means less frequent review and fewer experts, so that's the challenge. So, what IRBs do is they will look for consultation and that's often what we've had to do, who can we get to help the IRB either give the IRB an education lecture about a new type of therapy or help them with an individual review of an individual protocol, and this is primarily in the area of the technical expertise that's needing to understand this. So, IRBs often need to identify consultants, sometimes they may need to be outside the institution. PI consultant -- often we will ask the investigator themselves, who can you recommend, someone who's not involved in the trial, who's outside of the field who may be able to help us understand or think about the implications of the trial. And some of the recommendations are, it may be appropriate for IRBs to rely on RAC for scientific expertise, and


I was listening to the discussion and in some ways in my experience, I'm relieved, and our IRB members are relieved when we know that 20 people who are experts, have spent a lot of time thinking about this, I mean, that really helps us. We have the documentation, we have the data, we have the back and forth, and in some respects I find that a lot easier than some of the other trials where we haven't had that. So, I think that from my point of view, the RAC review has served a very useful mechanism. It does take time and the order of reviews and all that I think is an issue, but I've always, and I think our IRB has also appreciated that. And we asked PIs to identify several names of individuals who could serve as a consultant and the IRB can select those individuals, and we consider conducting research -- the other issue is, you know, when possible, to consider conducting research in sites where there is appropriate, ongoing experience, and hopefully there will be more institutions that have this


experience. The issue of multiple levels of review, this certainly does present logistical review challenges, the different levels of review, the RAC, the Institutional Biosafety Committee, the IRB. Other committees, also, if a research protocol involves extra scans or different things with radiation exposure. So there are lots of different levels. You may need radiation safety review at your own institution. And so, you know, you can think about, do you do this simultaneously or sequentially? The results of the RAC, I said, are beneficial for the IRB reviews, as are the biosafety committees and I have always found, and I think many IRBs have found it most helpful to have a review by the RAC first, then perhaps the biosafety and then come to the IRB. But there may be reasons why -- timing reasons why that's not possible or not desirable, but that is -- it's a lot of balls up in the air and what the problem really is on who's going first, because there may be requested for-


changes and amendments and then the investigator needs to let others know, until everyone comes to some type of agreement that this is what the ultimate final protocol and informed consent can be. When you add multi-institutional studies to that, I think you just duplicate that because you have many institutions doing the same things, and I'll have a comment about some of the other issues with multi-institutional studies as well. So, I think the recommendations are, I do think IRBs benefit by the review of the RAC and the biosafety committees and if that can come first when possible, I think that that has been very helpful. A couple things about risk-benefit assessment. We heard a lot this morning, I'm not going to spend a lot of time, I mean, IRBs are no different. When there can be good preliminary adult data, that's great. In the experiences of the protocols we've reviewed, the diseases and disorders just did not exist in adults and were


also in some of them difficult to replicate in animal models as well. So, it's not always possible in a lot of these first human trials there isn't going to be adult data. We do rely on the background provided, animal studies, justification for moving forward, and I think it's really important when our investigators are submitting their protocol, is to provide some of that thought process to the IRB as well, not just, this is the trial, this is what we're going to do, but to explain what else has been explored, why there aren't other models, why it may or may not make sense to do this in adults at the same time. That background will also help bring the IRB up to speed on the particular protocol. IRBs, you know, are generally not mandated. Most IRBs don't look at their first responsibility as the scientific review, although there is no way really to avoid looking at the science when it comes to the ethical judgments and in many institutions there is a scientific review


process in addition to the reviews that, you know, have occurred here. So, I think that, you know, we need to understand something about the science, but I don't think most IRBs are going to get really heavily involved in looking in depth at the science and are really hoping that the investigator gives them the appropriate information in looking, certainly when we have the RAC, at those determinations. So, the recommendations as far as IRB really are no different. The protocol needs to demonstrate well- designed lab studies, appropriate animal model, and promising data on disease mitigation or correction. IRBs do prefer to start in older children moving to younger adolescents or children, and I think this has been brought before because mostly because of the issue of assent and involving the child in the process, but again this isn't always possible because even in pediatrics and in a larger age group you have diseases that


happen in neonates that don't occur, you know, in older children, and also the issue that the condition -- the nature of the condition -- may be -- the treatment may be most promising the results if it's often done the younger the child. Earlier treatment in younger children often does offer the best chance of benefit if successful versus treating older children who are more severely affected. So, again, a strong justification requiring for starting an age group and how severely affected and if applicable, it's often helpful to specify to the IRB why other alternatives were considered but rejected and why the trial is better than the alternatives. The other issue about risk-benefit assessment when we look at clinical trials, first in human gene therapy and in treatment trials, is that the nature of some of the potential risks are very different than other trials and much more difficult to consider. Gene therapy and cellular therapy are permanent. It's not like drug trials where you can begin and end and have transient


results and adverse events. There are some things that can be more long term, but this is something that is permanent, it's integrated into the human -- the body and the system. So, the risks may not be transient and reversible, there's also the increased potential for long- term risks where we need to follow the 15 years and longer to find out what happens long-term. And the request for withdrawal raises significant follow up safety concerns. I mean, IRBs will spend time thinking about, is this a reasonable population to require the follow up, and how are you going to be able to do that? That was alluded to the resources that are required for that because you need to do that not only for the science, but you also need to do that for the safety of the research subjects as well. And the other issue with risk-benefit, if the therapy does not work, how does this impact future treatments? Is someone still eligible if their disease has progressed, does that put them


at a different state? So, IRBs will look at those things and ask questions about the implications of, if this doesn't work, now what? And that's an important part of the risk-benefit assessment as well. So, the protocol has to really demonstrate the ability to maintain long-term follow up as far as what the investigator can do and also what the subjects are going to be expected to do. Satisfying the regulatory criteria, there's been a lot of discussion about that this morning. And I think we would all have consensus that this type of research does present greater than minimal risk. I will say that there probably still is some inconsistency about IRBs in the interpretation and risk and benefit categories and I don't think there are many IRBs that would say this is not greater than minimal risk, but I still think by the nature of some of the questions this morning, the issue of the potential for direct benefit. I think there's been a lot of food for


thought out there about why we can consider issues of gene therapy and treatment and the cellular therapy, to present a potential for direct benefit, but I am sure that there are lots of pediatric IRBs that may not share that particular view, and there's been inconsistency in those types of discussions. So, as far as the benefit, is there potential for direct benefit? Considering the magnitude and the probability, and certainly the benefits may depend on the specific aim of a study. In general, I think, when we can IRBs will try to find a potential for direct benefit and the reason, if you look at the regulatory criteria, you certainly can't justify category 50.52 because I don't think many people think that it's a minimal increase and then the experience presenting commensurate, actual, or expected situations. Those just, you know, won't work for these types of trials. And the 50.54 is a long and involved process and I think more IRBs, when


they can, will try and work with the investigator, try and understand if there is a potential for direct benefit even though the study may not be aimed at that design, but if that is a potential result. I think to help in this, we've mentioned several advisory committees, the IOM Report. They've spent years and many members -- there are a couple members of those committees are in this room -- came up with recommendations to help there be a little bit more consistency among the pediatric regulations about how to interpret minimal risk, greater than minimal risk, disorder, condition, those types of things. Unfortunately, even after two federal advisory committees and an Institute of Medicine report, that guidance has not been issued and I would just make a plea that I think that that's helpful for -- would be helpful for the pediatric IRB community for this topic as well as others. Parental permission and assent, we talked about, sort of, we have a motivated subject


population, certainly in the issue of therapeutic misconception -- I'm not going to go into that. It's a complex science and topic to explain. Emphasis really needs to be placed on the parental permission and assent process. The form is important, but again the process -- and we've heard discussion about how that needs to be considered, how that has to be ongoing, how that needs to be a longer process, who approaches the families -- I think all of those things are very, very important that IRBs will look at. And then also a description to the families about the alternatives and how enrollment can impact the ability to receive alternative if the research fails. Long-term follow up may require getting permission when individuals turn 18. Don't know if IRBs are spending a lot of attention, they should be thinking about and making that known to parents even when you get permission initially, that there will be a requirement when a child turns 18 for them to give their own permission and


continue to be followed, so that needs to be thought of right up front. And for rare disorders, subjects -- another sort of very practical issue, certainly come from other U.S. and foreign sites, and this can present, also, logistical issues as far as obtaining informed consent. A family isn't going to travel from another country to come for a particular therapy and then say no to the trial. So, how do we -- you know, does -- the informed consent process usually has to begin before an individual shows up at the site. All of those details are important for an IRB, you know, to understand, and that involves even travel within the United States, and then also for follow up, what are the plans for follow up. Also, non-English speaking, language considerations, we've run into this discussion that we have individuals coming from other countries who don't speak the language and what is being done with the translation not only of the consent documents, but for communication on an


ongoing basis. So, recommendation, this was mentioned many times, that consent monitors are advocates. Another model that is used at our own institution is that the consent process takes place and then an uninvolved individual speaks with the family after the fact, they're not part of the actual discussion, and it's decided beforehand, what are the key issues that we really want this individual to understand before they agree to really start the research, and that feedback is then given back to the investigator from an independent person about whether these individuals understand these things or thinks that they may perhaps want to go back and re-explain. And also to think about things like funding for translation of consent documents and other important documents. If any of you are familiar with the process of short forms, I don't think this is a good area to use a short form. I think that you need to take the time to translate the informed consent, that families have a full


document that explains what they participated in. Data and safety monitor is another thing that the IRBs will look at, certainly essential for the IRB to review and consider up front the recommendations. Really, all pediatric first-in-human gene therapy and cellular therapy trials require some type of robust data and safety monitoring plan. I think that's very, very important and I would be surprised a protocol were to get through without having some type of plan. I'm not saying it has to be a fully independent constituted board, but there needs to be some type of plan. And the plan and process really must be independent from any conflicts of interest. There hasn't been too much discussion in this area, but I think this is another area that is ripe for discussion about how to avoid conflicts of interest in some of these technologies that are also being developed. Continuing review is another essential human subject protection. IRBs should consider requiring continuing review, potentially at


shorter periods of time if IRBs are concerned about how things are going, depending again if it's a very -- if it's a low-recruiting trial and you get someone once a year, you're not going to get more information, but IRBs have the ability to say, I want to hear what happened after that first subject and then we will give approval. So, I think when IRBs are uncertain or uneasy, this is something that they can do and we have done this in these situations. Require and copy the data and safety monitoring board reports, or evidence or other forms that data and safety monitoring have been performed. So, if the protocol says this is what's going to happen, then somehow the IRB at a continuing review should be assured that that did happen. And requiring detailed report of how many eligible subjects were approached and how many enrolled. Another amount of information that could help the IRB think about how this whole process of informed consent and enrollment is


going. Obtain information on continued follow up of subjects also to ask about individual subjects and ask for a report if you're using a consent advocate or monitor on an ongoing basis. And these are all different things that we have asked for or have made sure at the time of continuing review or thought about up front on approving these first-in-pediatric trials. And lastly I just want to, you know, bring up this issue if the PI holds an IND. While this may not be an IRB specific issue, it is certainly a human research protection program concern. The investigator needs to follow a sponsor responsibilities. Independent monitoring is required. How is the PI educated and who monitors fulfilling the sponsor role within an institution? Some locations are required that the institution itself hold the IND and that assures that the institution is making sure that all the sponsor responsibilities -- but in many institutions it is the investigator themselves and


certainly from my experience, not so much in gene therapy, but in tissue engineering where a lot of the work is coming out of the surgical fields, these are individuals who are not accustomed to typical types of clinical research protocols and as was alluded to, need help with the methodology, developing the protocols, the biostatistics, and institutions really need to have those resources available when you really have, right from bench-to-bedside work occurring at your institution. So, in conclusion, IRBs need to balance the need to protection with those limited -- with autonomy with the need to conduct first-in-human trials. There is definitely a higher bar with regards to justification as to proceed with first-in-human pediatric trials and there needs to be strong and convincing preclinical information. IRBs will rely on the PI, outside expertise, the RAC, and Biosafety Committee, for information and justification and I think that's reasonable, I don't think there's anything wrong


with doing that. Conflicts of interest need to be disclosed, carefully monitored, and in some cases, eliminated in order to ensure the safety and integrity of the trial. Expertise at all levels is required -- the investigator, the IRB, external reviewers, and the application of the pediatric regulations is essential, guidance based on NHRPAC, CACHRP, and IOM reports would also be helpful. Consider the use of consent monitors and advocates, and the need to maintain robust data and safety monitoring and continuing review processes is also very important. I will stop there. I hopefully gave you some food for thought, and also invite other people, other IRB folks here, don't want to -- I feel very humbled speaking for all IRBs, pediatric IRBs, so other thoughts that people have, that would be very helpful. (Applause) DR. BOTKIN: Susan, thank you very much. We will, on the schedule here, have a very short


break. Let's take 10 minutes or so and come back for additional talk and then our panel discussion. (Recess) DR. BOTKIN: Let me invite everybody back to their seats. Okay, we're going to go ahead and get started again this afternoon. Our next speaker is Dr. Scott Denne. Dr. Denne has extensive experience as a clinical investigator. He's the associate director for the Indiana Clinical CTSI where he's director of the Clinical Research Center, and coordinator of all the pediatric activities within the CTSI, also chair of the Pediatric Project team. Vice-chair for the Indiana University IRB, and he also serves as the chair of the American Academy of Pediatrics Committee on Pediatric Research. So, Dr. Denne's going to be speaking with us today about Subpart D, view from one IRB. Dr. Denne, welcome. DR. DENNE: Well, thanks. I know it's late in the day and we're just after a break so


I'll try and be brief. Today I just wanted to give you a view from one IRB, probably more accurately, one IRB member from one IRB. I won't make any statements about whether our IRB or myself is either typical or representative. I'm not sure there is such a thing, actually, but what I wanted to do is give you, sort of, briefly an idea of how our IRB works so you can kind of put this into context, talk a little bit about Subpart D and how we address that issue, and then really spend the bulk of the time giving you an example about a study that we had to review, an NIH study, really relating to cellular therapy and how we tried to work through the issues related to that study. So, I work at this sort of weird place called Indiana University, Purdue University at Indianapolis, a real mouthful, and this is actually the place where the Indiana University medical school is located along with a dental school and the nursing school and health-related sciences school and undergraduate -- a fairly


large undergraduate population as well, and the IRB serves all of that and also is the IRB of record for seven hospitals including the Children's Hospital, the VA, and the county hospital among those seven. The workload is fairly large so we actually do have 5 IRBs that are about 20 members each. One is strictly a behavioral/social sciences focused IRB and then there are four biomedical IRBs. Three of these biomedical IRBs review pediatric studies. There isn't a specific pediatric IRB. We actually serve on these multiple IRBs as pediatric representatives and there are two each on each of these -- three of the four boards. It's a fairly large place, as I said, we have about 2,500 open studies that have required full board review. We get 250 to 300 new full board studies a year and we have over 700 open studies with children and about 150 FDA studies with children, just to kind of give you a context of the volume, and in terms of the studies, the


FDA studies with children, almost every one of them have been approved under that potential for the prospect of benefit provision in Subpart D, which isn't a surprise. So, this is a form that we use, and I know you can't really see it, and it was just an idea -- just a point to give you the lay of what form we actually ask investigators to fill out if they're going to include children in research, to deal with the Subpart D kinds of issues, and so we actually ask the investigator explicitly to justify the necessity for including children in research, and then ask them specifically how this actually fits into the -- into what category in the Subpart D. For simplicity, we call them category 1, 2, 3, and 4: One being minimal risk; two being prospect of benefit, three, minor increase over minimal risk; and four, otherwise not approvable. And so, again, we ask them to actually explicitly address these issues. So, this is sort of a blow-up of the relevant category. Again, this prospect of


benefit category, and again we ask them to really fill out this form, necessity for involving children in research, and then explicitly address, in fact, what really is the regulatory language in this particular category, the potential prospect of benefit. So, with that very brief background, I wanted to talk a little bit about a particular study that we had to review in IRB, and this is a study from NIH, the National Health, Lung, and Blood Institute, and it was evaluating the safety and effectiveness of stem cell transplants from unrelated donors in children with sickle cell disease, in other words, bone marrow transplant for children with sickle cell disease, again, from unrelated donors. This is an active study. There are multiple sites, we're one of them, and this is just actually information straight out of -- straight off of clinicaltrials.gov. So, this is a description of this study, it's an interventional study, non-randomized,


uncontrolled. The end points are safety and efficacy. It's an open label study. The primary purpose it says on the clinicaltrials.gov is treatment, which is kind of interesting since it's really evaluation of treatment, it's not really treatment -- this is a research trial, and the primary outcome measure is event-free survival measured, basically, at two years out from the procedure. This is the overall inclusion criteria. The ages eligible for study is three years to 16 years, so pretty broad pediatric population here in terms of age range. And the inclusion criteria included sickle cell disease with one or more of the following: Clinically significant neurologic event, meaning basically a stroke, a minimum of two episodes of acute chest syndrome, a very severe event in sickle cell disease, and a history of three or more severe pain events per year in two years before the study, and this was really the investigator's attempt to identify a severely affected subgroup of children with sickle cell


disease. So, we were presented with this study for review and in terms of expertise, and there's been some comments about the appropriate expertise, we, again, this is a board of 20 people and with really broad expertise -- pediatrics, oncology, cardiology, medicine, neurology, pharmacology, genetics, dentistry, biostatistics, law, surgery -- so, a pretty good accumulation of expertise in a wide range of experience and also community representation. There wasn't anybody on the IRB who specifically did bone marrow transplant, but I think at least our sense was that we had enough expertise to ask the right questions. As I said, this was an NIH study and as an NIH study it comes to us with a certain degree of credibility. We know that it has been extensively peer reviewed, it has some degree of sort of ethics review as well, so although that's not dispositive, it certainly does help and really increases the credibility of this particular


study. I think I would say the same thing for FDA's -- for studies that are blessed by the FDA as well. So, what was the major issue? Well, there are lots of issues that I think you can probably sort out in your head before I start, but this is the one that came to fore right away and that was, we have sickle cell disease which clearly has a high morbidity, but has a very low risk of death within sort of the childhood range. And we're proposing a treatment that may ultimately, you know, affect something close to a cure -- that's probably too much -- that's probably a bridge too far, would substantially improve this disease but also carries a very high risk, or a substantially higher risk, of death and morbidity, at least in the short run. So, to what extent is it reasonable to take children and subject them to a treatment that actually might kill them instead of sort of a standard approach that's, although they may suffer more -- longer -- they're probably not going to


die. So, this came to us sort of straight up like that. This issue wasn't very well addressed, I didn't think, in the protocol, so the IRB had substantial questions, and so the study was tabled and sent back to the investigator with a whole variety of questions and we also asked the investigator to come to the next IRB meeting so that we could ask additional questions. So, these are the things that we sort of sorted out as we were working through this study, these are the questions we wanted to address, and I'll give you just a brief part of the answers here, I'm not going to go into a lot of detail, but just get you an idea of the nature of the conversation. So, what is the experience of using bone marrow transplant for sickle cell disease? Well, there is some. It's not overwhelmingly extensive, but there's a reasonable amount, I mean, in the 50 to 100 patient range, both children and adults. The risk of mortality for bone marrow transplant


in this context is probably about in the range of 10 percent, so not trivial. What's the risk of mortality for sickle cell disease in childhood? Substantially less than that, probably 1 percent or so -- probably less than that actually. What are the current therapies for sickle cell disease? Well, there are a variety of therapies including sort of multiple transfusions, hydroxyurea, a variety of other things, that are sort of leading on to the second question of how effective they are, they're modestly effective. They have some significant downsides, they don't really prevent all of the complications associated with sickle cell disease. We also asked, under what circumstances is bone marrow transplant standard of care for sickle cell disease, and it turns out it really isn't under almost any circumstance is it standard of care. So, this is -- this really is a research protocol. This was an area that we spent a lot of time on, do the criteria select the most severely


affected children with sickle cell disease? And we've talked a lot about that throughout the day, how if we're going to subject this risky therapy, this new risky therapy, we might want to start with the most severely affected group and, again, I told you that this was the attempt at selecting the most severely affected children were these kids with stroke -- there wasn't much argument about that because those kids, once they have one stroke, are at risk for multiple other strokes, but things like multiple pain episodes, is that enough of a severely affected group, and there was a lot of discussion about that, is this the right group. How certain is the prognosis for these children? We were reassured to say that when you meet these kinds of criteria you're likely to experience additional complications of sickle cell disease. We also asked why the study should be conducted in children rather than adults, and there was a little bit of adult experience but the


concept here was that this was the group that could most benefit by this treatment, that by the time these children were adults, they would have experienced these significant complications, so that was the discussion. We asked how consent and assent would be obtained and it turns out that will be over, you know, weeks and multiple discussions. We asked who would do this, actually the research team was going to do this including the PI and the research coordinators. There has been some talk today about other alternatives -- child life representatives, other sort of research advocates, which certainly would have been an alternative in this particular study. We also asked why should children who are too young to provide assent be included. It's one thing if it's a 14-year-old who has certainly some sense of what this might mean, it's probably a different thing when it's a three-year-old, and again, this goes to that these young children with these significant complications may be the ones


that would best benefit. We also asked how well was this study vetted beyond just the scientific community. Has this study been discussed with children with sickle cell disease and their parents? And as it turns out, they had actually done this through a variety -- how was it done? Through a variety of interactions with advocacy groups. And what the investigators reassured us about was that they were very positive, they wanted this option as a potential treatment, at least as a research study. Some additional questions: are all the procedures, risks, and alternatives clearly presented? Obviously, that means going through the consent form, but it also meant to us really having a conversation with the investigators and coordinators of this study to really ask how this was going to be presented, I mean, beyond just what was on the consent form. We asked how safety would be monitored during the study. There was a DSMB, there were stopping rules, and we also asked, is this treatment possible outside of a


research protocol, because we wanted to know, if somebody wanted to have this done, could they actually have it done, and the answer was, yes, they could. Now, you may or may not think that's relevant. I think our IRB thought it was relevant, that somebody could be doing this outside of research protocol which seemed to us probably the worst of all circumstances. So, having gone through those issues, those and some others, we ultimately decided that children were appropriate for this study and that this -- the prospect of benefit and the other provisions were met, and so the study was approved by our IRB, and as far as I know, by every other IRB where it was considered. I have to say that it would be not unreasonable to have a different view on this particular study, but, again, from my IRB, having gone through a pretty extensive process with the investigator, I think we felt like it was a reasonable thing to pursue.


Okay, we've talked a lot about this, cell and gene therapy in children and what are the considerations in IRB, and I don't think I'm going to say too much different than what lots of other people have said, but again this is a view from at least our IRB -- the severity and natural history of the disease, the availability and effectiveness of current therapies, and again, effectiveness is another important aspect, it's not just available, but how good they are. All relevant information needs to be provided by the risks and benefits, basic animal human data, and I think this may be a slightly different twist than what others have talked about, but I think from our IRB's standpoint, the strategies for obtaining information in adults need to be explicitly considered and discussed. I mean, Susan talked a little bit about this. Give us your thinking. It's possible that adults can't be studied, we can't really get there from here, but we want to know why and we would like to know that you've really thought about all of the potential other


options -- similar conditions, heterozygote carriers, a whole variety of other potential approaches that we would like to know that you've thought about and said, won't work for all of these reasons. So, with all that, what would actually be helpful to us as IRB members? Well, I think that, again, advice to the FDA, development of an ancillary document with relevant questions -- and many of the questions that I outlined here are really relevant to all cell and gene therapy -- really should be addressed specifically, and this document can provide these questions, these questions can be given to the investigators and sponsors and actually affirmatively addressed, and then submitted at the point of IRB. I think with a more standardized approach, and at least a very clear outlining of these questions, I think there will be a little bit more consistency across IRBs, for one. And secondly, I've looked at lots and lots of protocols and virtually none of them address very explicitly these kinds of questions.


A few things tend to be addressed, none of them very comprehensively, so I do think this would help certainly IRBs and I suspect, actually, investigators and sponsors into thinking about what really needs to be provided to IRBs. Thanks. DR. BOTKIN: Any questions for Dr. Denne? All right, we're going to go ahead and convene our panel now and I'll invite the panelists to come up and join us. One individual who is on our panel who has not been introduced as yet is Dr. Jerry Menikoff. Jerry has an M.D. and a JD, currently director of the Office of Human Research Protection. Prior to that job in just the last year and a half or two years or so he was the director of the NIH's Office of Human Subjects Research for the intramural process, so extensive experience, obviously, with IRBs. So, we're going to welcome questions up here fairly shortly. I think the thrust of the conversation has really been twofold about IRBs, one is, how is it that IRBs can do a better job in this domain, particularly complicated set of


protocols that are new for many IRBs? So, how is it that IRBs can be better supported to do a better job with human subject protection issues? Now, the second vein is clearly one of trying to reduce the burden of regulatory oversight and sometimes those two are in stark conflict, but not always. So I think we want to think of methods with -- by which we might achieve the goals we're looking for in terms of human subject protection while reducing the extraordinary burdens that this entire process oftentimes has on investigators. So, I'll start with one of my own questions and it really relates to a question that an audience member brought up a little bit earlier, and it has to do with, sort of, the communication process. I'm interested in the panelists' ideas or thoughts on what the obligations are of IRBs to go above and beyond the review of the consent document and process alone in terms of helping kids and families to understand and make more fully informed choices in


this particular context, again the context of complicated, innovative protocols. Are there ways in which the IRBs should go above and beyond what is typically their normal role of looking fairly concretely at the form and the process? Ms. Walden? MS. WALDEN: I would greatly support the idea that the informed consent should be a process as previously stated by other speakers and that perhaps the IRB could consider prescribing intervals at which there are situations where clinicians or PIs or whomever is responsible for having those conversations with families, actually sit down and review information with them again, give them opportunities to ask questions that may have arisen since the initial consent and to ensure that it is, indeed, a process and ongoing. DR. KORNETSKY: I think some of the ideas that have been talked about using consent advocates or monitors are getting at some of that, that there be some independent party that is helping to make sure that the families understand


or need help with the decision-making. An interesting question that I think arises with the informed consent as a process is what happens after the trial begins and what is the obligation to keep informing families about the nature of the research and what's happening with it. And is there, especially in a field where someone has agreed to be in a first-in-human trial, you certainly don't have conclusions about the trial, but is there any type of information sharing outside of just the involvement of that individual in the study, about the study in general? DR. BOTKIN: Skip? DR. NELSON: Susan, let me ask you a question. One of my questions about the whole area of monitoring is, what do you tell people who are put in that role to look for? It's sort of, I'll see good consent when I see it. I was struck by your comment that you have people go afterwards and then ask key questions about certain items. And so my question to you is, how do you identify those? And I assume those don't include sort of


all of the minutiae that's included in the consent documents. So, how do you go about identifying those key issues? DR. KORNETSKY: So, the process that we use that has worked very well is we set up a meeting with the investigator and anyone they want to include -- the consent monitor, the IRB, the IRB chair -- and we have a discussion about the trial and really take from the investigator what they would really want to be assured that someone understood before consenting to participate. By also having myself or the IRB chair there, we also know the issues that were on the mind of the IRB as well. So between both sides there, we come up with, before the actual study begins, these are the key issues and we actually prepare a document and say, you know, this is what the consent monitor is going to explore with the individual, and that monitor will talk after the process. They're instructed not to answer questions, if this brings up questions to say that that will be, you know, up to the investigator, that they'll


bring them back, and then they really -- it's almost like a report card, in a way, back to the investigator saying, we went through these questions, they really understood this, or they really didn't understand that, you may want to go back and do it. So, it's really a joint process up front that we use for that and it's not the minutiae, we think about the real key issues that we want people to understand. DR. BOTKIN: I wonder if anybody -- oh, Jerry. DR. MENIKOFF: I guess I'd sort of suggest stepping back a little bit, not even so much concentrating on their process, there might actually be benefit to looking at what IRBs do in terms of the consent form itself. It was interesting during this meeting seeing a lot of information about consent forms and the groups that have studied this and looking at some of the language and I've separately advocated making all of this more transparent in terms of getting greater public buy-in in terms of


a lot of this stuff. So, for example, in one of the presentations it was noted that their consent form on some of the studies -- I don't remember which group -- they would note it was unlikely to benefit people. And it would be interesting to see how many of the other consent forms in this scenario or in similar scenarios, are routinely saying that your likelihood of benefit -- that that's a particular unlikelihood of there being benefit, because the bottom line is, a lot of this stuff isn't that open right now. The last time, I guess it was a few years ago, I tried to sort of find some of the RAC-approved, or at least the consent forms that went through to RAC, and it wasn't that easy to do. I think see this stuff and air it and see if that could all lead to greater ability on the part of IRBs in all of us in terms of getting better consent forms, because the bottom line, partly, is that regardless of the process, the consent form starts out spelling the basic conditions under which what we think are the


relevant facts to look at. DR. BOTKIN: Very good. Steve? DR. JOFFE: Jeff, I'd like to suggest that as a rule, IRBs don't have any idea what actually goes on between investigators and potential subjects and families unless there's some -- and I think this is still uncommon -- some mechanism put in place to observe that process either in real time or one could imagine a system where some sample of consent conversations were audio taped and then somebody from the IRB could listen to those audio tapes later as a kind of quality assessment, quality improvement kind of exercise. What Susan has suggested in terms of somebody from the IRB speaking with families later and doing a check on their understanding, on what was communicated to them, et cetera, as an alternative mechanism, so there's many mechanisms by which one could do it, but the bottom line is, I think right now for the most part the informed consent process is a black box to IRBs.


DR. BOTKIN: Yeah, I think that's certainly been my experience at our institution and I think many investigators are resistant to that as are the IRBs themselves. They see that as an additional burden on the process rather than as an assist to the communication that they're engaged in. Jeff? DR. KAHN: I want to maybe suggest something we haven't heard much about today which is that many of the CTSA recipient institutions have clinical research ethics consultation services built in and that IRBs need not think about this as solely their responsibility and domain, that there now is an increasing capacity to deal with these issues in other places within at least some of our institutions and that we ought to think more broadly about who might help in improving the process by which we do, not only consent, but protection of subjects, think about early phase trials and the ethical issues that come with them.


DR. BOTKIN: Good. I might pick up on Ms. Walden's comment earlier, see if anybody has experience with this notion of a longitudinal engagement with research participants. We had a little bit of discussion about this at our institutions and one of the investigators said, sort of concierge research -- I don't think that was meant to be complimentary, but I think there's this potential notion of being able to engage participants over time rather than the front-loaded process that we have now. Anybody have any experience with particularly these complicated studies, ways in which the IRB can maintain that relationship? DR. NELSON: Well, I don't have personal experience from that perspective but there are certainly longitudinal studies that have been known to be successful and, for example, the Nurses Study is one that has been able to keep nurses in for years and through a process of birthday cards, postcards -- I mean, the kinds of things that keep people engaged. I think one of


the tensions is that there might be a certain skepticism among certain individuals that the kinds of things that you do to be nice, basically, that would keep people interested in your project and keep people engaged and coming back, are sometimes seen by some -- I would not include myself as part of this group -- as potentially manipulative. In other words, you should treat them neutrally, treat subjects neutrally, and that way they cannot be unduly influenced one way or the other. Well, if you do that, then they will obviously not want to come back to see you in follow up. So, I'm not sure that's the approach I would take. I think engagement, community engagement, involvement of the community, feedback of research results to the community, all of the kinds of things that I think would be expected, I think would be an improvement in that kind of longitudinal experience. DR. JOFFE: In my field of pediatric oncology, as many people know, there are many


protocols that -- where the intervention "treatment" part of the protocol can extend for a year or longer and often there are milestones or research interventions that happen late in the protocol where discussing those at the very beginning is not going to be salient to somebody who's having to think about something that might happen a year down the line. So, we're beginning to build in processes of re-consent or refresher consent or at least documentation of renewed conversations at milestones that happen down the line. This is in part driven by the fact that, particularly for a family whereby their child was just recently diagnosed with a malignancy, they are not necessarily at their decision-making best and the situation is overwhelming enough that the passage of time may afford greater clarity and so the more things that we can decompress out of that first conversation and move to a later time, I think the better the decision goes, the better the process goes.


DR. KAHN: I actually serve on a DSMB for an international longitudinal trial looking at genetic risk for diabetes and they have been extremely successful in keeping families and their children in that study over the course of, now, I think 8 or 10 years by providing information to the families, not only about the findings, so the genetic risk that they've been able to determine for those individuals, but also around sort of general health issues and sort of ongoing sort of newsletter-like information that has built a kind of cohort in a community sense of cohort, and that, is, I think is an interesting model that's likely to be something that will need to be replicated. Think about the National Children's Study, I don't know if anybody here in the room can speak to those plans, but that's one that's obviously going to have to figure out how to keep cohorts together over a long period of time and some of the bio bank research going forward is likely to have some of these issues, too. So, I think your question's a good one.


I don't know that we have a lot of experience yet, but we certainly need to build up the ability to deal with these kinds of longitudinal populations. DR. BOTKIN: Well, and that's nice when an intervention of that type meets the needs of the research and the study by maintaining satisfaction, maintaining communication and participation while also reminding folks that they're in research and we know from the radiation experiments and other work that a lot of folks forget that they're in a project before long, and so being able to maintain that communication and a reminder of what's coming down the pike could be important. A question here? SPEAKER: Yes, I wish to touch on one of the themes, the common themes, that have been running through today and that's the challenges of all of the different regulatory institutional committee reviews. And I'm entertaining the thought of if someone were to be working on a small multicenter study, and let's say, for


example, two sites in the U.S., asking the panel, is it possible to get those two IRBs to work together through an approval process? We all know that, you know, running a program like this, time is, you know, always at the door and wouldn't that be great if we could find a way to, you know, make that happen, to get to, again, the endpoint of getting that first patient on the trial as quickly as possible, and I wonder if any of you have ever tried it. Would you consider it? What would you need to have addressed in order to, you know, put that forward with your IRB to, you know, work with another body? DR. BOTKIN: All right, terrific question. I think Dr. Menikoff may want to have something to say about this. DR. MENIKOFF: Yeah, certainly it's a great issue. Can it happen? Of course. It happens all the time. There are many examples out there in terms of existing relationships in which various institutions have chosen to rely on outside IRBs and there are numerous models in


terms of how this might work. Again, it depends on the specifics of the study, the specifics of the relationship between the two IRBs, lots and lots of issues there. As a more general matter, I think both the FDA and my office, OHRP, have both gone on the record as saying it would actually be a good thing if there were fewer duplicative IRB reviews, so we're certainly supportive of this and I think we all are looking for more examples and methods for encouraging this to take place. And there are a variety of things that can take place. There are systems out there that -- currently the VA is a good example. For certain centrally funded VA studies you have to participate through their central IRB, so if your particular site -- or considering being a site, you either say you agree to accept that central IRB review or you don't. There's nothing preventing somebody creating a new study from setting up the same model saying, I am now the funding entity for this particular study and if you want to be one of the sites, here's the


IRB we've chosen and we could try to demonstrate what an excellent IRB it is, but that if you want to participate in this very prestigious and important study, you basically have to sign on to this. So, it's something definitely to be encouraged and, yeah, it is sort of a good thing. DR. BOTKIN: Susan? DR. KORNETSKY: Well, I think the idea of reducing the number of IRBs is a very good idea. I will say that from the perspective, I'm representing IRBs, I have to question, and this may be controversial, whether first-in-human trials in gene therapy and things like this, are the types of things that IRBs will feel comfortable relinquishing to other IRBs at another site. And I'm just being perfectly, you know, honest about that. I'm not so sure that that would be the first thing on my mind. DR. DENNE: Just to follow up on that a little bit, I do think that there may be some middle ground between a central IRB and individual reviews and that is that we all do this


independently and literally don't know anything of what the considerations were in the other IRBs. I think we really would benefit by knowing -- and I think the process would be a lot better, if we could actually share some information about these reviews and I think we could streamline things a lot and actually improve the reviews. DR. BOTKIN: Skip? DR. NELSON: I agree with that, but I would also put the burden then on sponsors and investigators to make it clear that IRBs sharing information like that is fine. Both of them sign confidentiality agreements, but I do know of anecdotal cases where, not academic, but industry sponsors have felt that the sharing of information between IRBs violates that confidentiality agreement, and has actually then tried to imply to investigators that their IRB is doing something that's considered non-compliant. So I think if -- I think transparency and sharing between IRBs is part of the solution, but that's something then that the community of


sponsors and investigators need to embrace and need to empower IRBs to be able to do that. DR. BOTKIN: I would also point folks to a recent publication by Dr. Menikoff that actually touches on this issue both from a burden of regulation standpoint but also the potential down sides to human subject protection of the redundancy within the system. So, that's a different new perspective. Question here? SPEAKER: Yeah, hi. I would just like to ask for the opinions of the panelists going back to the issue of 50.54 federal panel review of some of these studies. Dr. Denne, I'd just be curious whether or not that's something that your IRB had considered during your review and under what circumstances the panelists might feel like that would be helpful. Thank you. DR. DENNE: It did come up. It came up briefly. I have to say that there is some inertia, as I suspect people recognize, about pushing that process. I've sat on the IRB for 15


years, we've never referred anything. I mean, I've pushed to refer a few things. We've not, and I do think that there is the sense that that's a drastic measure and people will go to all sorts of lengths to not move in that direction. DR. BOTKIN: Let me ask Skip to comment on that. I think the reputation for this process is that it's years and extremely onerous. I don't think that's quite accurate any longer but a more informed opinion. MR. NELSON: Well, it's a complicated question. I mean, if you look at the recent statistics I think I would argue that one could get through the process in six to nine months. Now, because of the requirements that we have largely around the public nature of the review, and so those reviews take place under the Federal Advisory Committee Act and since we don't have routinely scheduled meetings, as opposed to the RAC where you've got four meetings per year and it's sort of set up, there's a certain delay that it takes in just setting up those meetings.


So, from that standpoint we have tried to routinize and make transparent the process. My own view is for the same reasons that Jeff argued for the potential benefits over time of the RAC in terms of transparency and the discussion of the science, that for many of these controversial issues, that that sort of public review would have some advantages. Now, I'm not advocating that certain protocols have to be through a 50.54 or 46.407 process, but my own view is there is a certain virtue to the transparency in public discussion of the science in what might be considered controversial areas that over time, even if the -- would be, if you will, advantageous to the field. But I don't have a protocol in mind that I would say needs to be a 50.54 review, so don't over interpret my remarks. DR. BOTKIN: Yes? SPEAKER: Hi. My question is about primarily first-in-man studies with complex cell and gene therapies. When IRBs are reviewing that first protocol, are you also looking at the


long-term follow-up plan? And second question, in addition to looking at the follow-up plan, are you looking for whether the resources are there in the commitment to execute on that plan? DR. KORNETSKY: I think that you definitely should, if IRBs aren't, they definitely should. I think that's part of assuring the safety and welfare of the research subjects to ask those questions about the plans for follow-up, how they're going to do that, how that's going to be resourced, so absolutely. DR. DENNE: The other part of that, and this is the tension between sort of central review and local review, understanding who the investigators are and their capabilities of doing that, I think, is really important at the local level. I'm not sure it's really possible at the central level, so, I mean, these are the tensions between sort of central and local reviews. DR. JOFFE: This is a really important issue and seems to me like a great opportunity to create an infrastructure for long-term follow up


of participants in certain kinds of trials -- gene transfer trials, cell therapy trials, potentially other kinds of trials -- that an infrastructure that crosses studies, that investigators on individual studies can tap into, and there are some existing registries for other areas that could be potentially tapped to create arms for this or perhaps a new registry, but I think it would be advantageous to be able to say that, our -- in addition to being on the protocol, somebody who enrolls in our trial will also join this registry that will assist us and have an infrastructure for the long-term follow up and in the context of five-year grant cycles, it's very difficult to make commitments to long-term follow up beyond the time of the ending of that grant and so if there was some infrastructure that could sort of take that off investigator's shoulders, I think that would be very advantageous. DR. BOTKIN: Yes, sir? SPEAKER: I'd like to get back to the process of getting consent a little more and let


me throw out a scenario here a bit. For instance, if we have an ongoing therapy such as allotransplant for FAA or SCID or Wiskott- Aldrich or whatever, and we develop a new gene therapy or cellular therapy, and we're going to potentially be comparing these, we may actually have two different protocols open within the same institution that potential patient may go on, and so the issue for some of us as investigators is how do we really attain consent in that sort of circumstance. I mean, I may be the PI of both protocols and, on one hand, I might be -- you know, have more insight into how these things were developed and others, but, on the other hand, I don't know that I'd want to put myself in that position of potentially being thought of as steering one patient or a family from one protocol to another. So, you know, in regards to this, kind of, how does one be impartial about this and how does one potentially have another level of obtaining consent, because I don't think a social


worker is going to help with that, I don't think a child life person is going to help with that, and if there's either somebody from IRB or the ethics group or whatever who can be in that sort of circumstance up front, we have never done that, and I don't know, does the panel have any precedent for doing any of that? Because that would seem to make sense to me but that's not part of what currently happens. DR. DENNE: I will say that we do pay some attention to competing protocols. If you have competing protocols with the same patient population, that's something we really highly discourage. And if, in fact, that's true, then we specifically ask investigators, how are you going to deal with that situation? Because it is difficult, I mean, to ask a patient to try and choose between two competing studies with overlapping inclusion/exclusion criteria, I think is really sort of unfair and that's, at least, the bias of IRBs is to not actually have that situation. And again, if you do, you have to have


a very clear plan and convince us, anyway, that you have some mechanism to deal with that. SPEAKER: And what would that look like? DR. DENNE: Well, I mean, right. So, I mean, sometimes it's simply, we're going to do this one first and then we're going to do that one. You know, sometimes it's, you know, these inclusion criteria -- we're going to do this one at this site and this one at that site. I mean, I'm not very happy with those particular solutions, but the competing protocols in the same institution, the same investigators, I think, is just problematic. I think it introduces, just from a research standpoint, you know, potential bias that's hard to address. SPEAKER: Well, the problem, of course, is that you may have therapy where you've got -- you know, again from a transplant standpoint, you may have 80 percent survival for Wiskott-Aldrich and you want to introduce gene therapy, which you think might be better but it may not be better, and can you ethically not offer allotransplant in


that circumstance if you want to explore gene therapy? So, that's kind of the thrust of the question. DR. BOTKIN: I think it's an excellent question. From our institutional perspective in the cancer arena we have the Clinical Cancer Investigation Committee that is supposed to look at this issue to support the IRB or support the institution and our neonatology department that does a fair volume of research also has a pre-review process to try to prevent competing studies for the same patient population, but seems to me the IRB is not in a very good position to know when this situation arises. I mean, you've got two studies coming forward at the same time, you might pick that up, but if they're six months apart, it's not clear to me the IRB's going to recognize that issue. Skip? DR. NELSON: I heard a slightly different theme in your last comment which went to the question of whether you as an


investigator/clinician indeed were in equipoise about the virtues of enrollment in the clinical trial compared to what other treatment might be available, whether that's another clinical trial or whether that's standard of care, it's considered evidence-based, and the bottom line is if you're not, then you have a dilemma and you shouldn't be recommending people to do things that you believe in fact are not equivalent. And, you know, the complexity of that comes down to a discussion within your -- I mean, I used to work in a critical care unit, pediatric -- I mean, as a group we would have to come to some agreement and then whether I as an individual would be involved in those conversations then becomes a matter of personal choice. So, that's a much different question than competing protocols and the need to allocate a priority within a given unit, for example. I mean, I heard that theme coming out through your question if I heard you correctly. SPEAKER: Yeah, I think you did. I mean, essentially if we have, you know, a decade's


worth of experience, even though it's a rare disease, we have a fairly good idea of what the current, you know, state of the art is as far as interventions go, but now you're adding something that you think might be better as a potential competing therapy and of course you don't know that it's any better because it may not have existed for, you know, until now, and so, you know, that's the conundrum, I think. DR. JOFFE: I want to take a slightly different tack in addressing issues raised by your question. A theme that's been implicit in a lot of discussion today, although I'm not sure it's been named explicitly is, what should be the appropriate scope for parental decision-making when there are choices to be made, either what came up earlier between, for example, enrolling their child in a research study versus doing whatever else is available to them as part of standard care. This was an important part of the case that Dr. Denne presented of unrelated donor bone marrow transplant for sickle cell or you're


alluding to the possibility that there might actually be two protocols with very different kinds of flavors and people might be asked to make a choice between those and how reasonable is that. And I'd like to argue that we ought -- so, the IRB process or the human subjects protection process is unavoidably paternalistic and that's true for review of adult research and it's even more true for a review of pediatric research. That said, there's a question of just how paternalistic it should be and an extremely paternalistic process would then narrow the scope of acceptable parental decision-making down to a very narrow window. An alternative view would be to say, let's make that space for parents to make decisions, that scope as wide as possible. And I'd like to argue that we should actually try to open up as much space as possible for parents to consider the decision they want to make for their child, be it to decide between two competing research studies or between joining the protocol for unrelated donor bone marrow transplant for


sickle cell versus doing supportive care for sickle cell which would be "standard of care." And so we want to, as human subjects protection -- as investigators and as people involved with the human subjects protection process, take things off the table that are truly unreasonable. But having done that, try to leave as much space as possible for parents to make the range of decisions that they think is in the best interest of their kid that is consistent with their values and where they're coming from. And I think, you know, I think parents and families, particularly when there's time and a lot of support from institutions and investigators, are capable of doing this. DR. DENNE: Just a brief response. It's not just about patient -- or parent autonomy, which certainly is an issue here. It's also about getting research studies done, and if you have a rare disease and you're allocating among four different things, the likelihood that anything's going to come out of it is a major issue.


SPEAKER: I just wanted to make a brief comment about what was said because I think this is going to be a problem that will be very, very frequently seen once gene therapy starts in a lot of these rare diseases because many of them -- of the therapies, are combined with bone marrow transplantation, so the situation where you would have in the same unit maybe two investigators offering different treatment options actually could be quite common, and I think it's very difficult for the families to decide. So I just wanted to bring it up because I think it will be a frequent problem that investigators will have in the near future. DR. BOTKIN: Thank you for that comment. Yes, sir? SPEAKER: So, the system requires that for a protocol to move forward that there's an FDA IND authorization and the other side of that is the IRB approval. My question is -- can you hear me? DR. BOTKIN: A little louder, please.


SPEAKER: The system requires that the FDA authorize an IND for the research to move forward, and the system requires IRB approval, and then there are other committees involved given the particular protocol. My question is, how does an IRB -- how is it impacted by an FDA IND authorization? Are there perspectives that are changed that influences that? Someone brought up the fact that NIH gives a credibility to a certain trial. I'm just wondering about when the IND is authorized by the FDA, what impact that has on IRB. DR. KORNETSKY: It certainly does have an impact. I mean, IRBs, you really -- the trial can't get going until you have both, and, you know, there's the question of, you know, which comes first, the FDA or the IRB, or do you do it simultaneously? Ideally, I have found, although it's not always possible for timing reasons, that the IRB should be looking at the same protocol that the FDA is looking at and if one changes, the other one has to look at it, too.


So, it does help to receive a protocol from an IRB perspective that has been reviewed and approved by the FDA and all those changes that may or may not have been requested are incorporated in that. It's a little bit more difficult when an IRB reviews it first. They could certainly review it and then it goes to the FDA and you're all set if the FDA says, okay, if not, you're going to come back and change it. And it gets complicated when you review them in parallel as well, but I do think IRBs do depend on the level of review and what's being required because they are also looking at safety in human subjects issues and it's another set of eyes, so I think it is considered. DR. DENNE: I would agree with that. I just also want to say that at times when the IRB -- certainly it does provide some credibility that it's gone through the FDA process, but when -- sometimes when the IRB starts asking questions of sponsors, sponsors come back and say the FDA approved it. And so they do sort of push back on


-- and sort of assume, in some ways, that it's through there, why are you asking us these questions? So, that doesn't always work, but it's not an infrequent response. DR. NELSON: So, I think it's important to repeat the view that the IRB and FDA assessments ought to be independent. You heard earlier about the IND criteria in terms of clinical holds. Personally I would argue that failure to be in compliance with Subpart D ought to be considered an unreasonable and significant risk and therefore be considered a criteria for not allowing for an IND to go forward. Having said that, the judgments about whether that's true or false are complex and there may be differences of opinion about whether that holds or doesn't hold. So, from that standpoint I would hope that IRBs would not cave in to the pressure on the part of a sponsor to say that just because the IND has gone forward, you shouldn't make an independent assessment. I would like to turn around a question


to the audience and ask this. I mean, we heard a couple different suggestions about ways for IRBs to get information about what the FDA review is. I think generally there's a reluctance for the FDA to communicate directly with IRBs, in fact, I'd say generally there's a reluctance. We don't do that. I mean, we inspect them afterwards, but we don't communicate directly with IRBs. Why not? Well, we're communicating directly with sponsors and with investigators about how to design that trial and if we were then communicating directly with IRBs about how to review that trial, I call that the Enron Problem. So, you know, the general approach I recommend is that the sponsor, or investigator, sponsor investigator, has the benefit of the FDA review and hopefully could incorporate it into either their protocol or into an ancillary document, as you've recommended, answers to the kinds of questions that the IRB would like to know, many of which, in fact all of which, the FDA has, in fact, considered and discussed, but may


not be communicated transparently from the FDA through the sponsor to the IRB. So, I guess my question is, I mean, that's generally how we conceive of that communication flow rather than our assessment going independently to the IRB which would raise some interesting questions about areas of jurisdiction, conflict, et cetera. DR. BOTKIN: Yes, sir. SPEAKER: So, first a comment about the prior discussion of conflicting protocols. So, our institutional IRB actually requires with submission of a protocol a prioritization list that's developed by the treating group of physicians. For instance, if you are within the pediatric neuro-oncology group, your group must maintain a prioritization list that already answers those questions of competing trials in terms of priority, or if there are two listed priorities, the submitting investigator must submit along with that, some answer to how those conflicts will be resolved and I think that's just probably a good thing for IRBs to include if it's


not already routinely done. A question, though. In an earlier -- I don't remember if it was the end of the morning session or the first part of the afternoon session -- one of the speakers made a comment about -- as a suggestion, that the informed consent process be streamlined to the actual informed consent document, only including the elements that are actually the research portion. I just wondered how this panel considers that, particularly within cell therapies that are done in the context of transplant. It's getting more and more difficult -- the transplant itself is complicated, there are multiple chemotherapeutic agents, there's the stem cell product, there's graph versus host disease management and all those elements end up being included as part of the whole research project when really the research is all about adding one cell product, you know, to the whole. And we're trying to pull those out into separate protocols, but it is getting more difficult and our consent documents are getting now to be, you know, 25, 30,


35 pages long listing every possible side effect of everything that's included in the study, including the Tylenol that's given for -- and how do you see that? Is there room to swing the pendulum back? DR. BOTKIN: This is a great question and it's been a longstanding problem with or issue with oncology trials. DR. NELSON: I made the comment so let me just elaborate a little bit and then get other people to comment. I think all of the other aspects of what's going on with that research subject are important for the risk- benefit evaluation from the perspective of the IRB, so I wouldn't want to imply that you're sort of taking all of that off the table from the standpoint of at least evaluating what's going on. But strictly speaking the research risk is that incremental risk from whatever is different in that research one, and if it's just the column you've got -- and I don't even know if it's done this way anymore, you've


probably got fancy monoclonal -- but if you're putting your stem cell product through a column and the T cells go left or right, depending upon what they've got on them, and that's the total research question. Everything is basically standard. I would advocate the research consent ought to simply be this one -- you know, however many pages you need to explain the difference between that column process versus what would otherwise be done, you still need a consent process, a consent document, something going on for everything else. But the research consent should say something like, everything else that's going on here is going to be exactly the same as if you weren't in this research if you chose to have a bone marrow stem cell transplant with this exception. This is the difference. Here it is. And then you've got that document, and then you've got this other 40-page something that goes through the Tylenol and whatever. So, that would be the proposal. It's not often done that way.


DR. BOTKIN: Jerry? DR. MENIKOFF: Basically I'm supportive of the same type of thinking. Let me just give one qualification because you get people making similar arguments saying, for example, in a study in which they're asking you to get something instead of standard care, so some new treatment, they will say, well, therefore, you don't have to actually say much of anything about standard care. And I think the danger in that is the consent form, by and large, should give the core pieces of information that a person needs to make a decision about participating in the study. So, for example, if you're proposing some new antibiotic and that's all you'd get in a research study to treat your pneumonia, it's highly relevant to that decision-making process to understand whether standard care for that type of pneumonia would have given you a 99.99 percent chance of being cured and, therefore, again, I support the premise. But you would have to give some basic background information on how good


standard care is, that it's risk-free, et cetera, et cetera. So, with that sort of qualification, I do agree that there are ways of not including additional information about other types of treatment's risk-benefits, et cetera, that aren't -- where they aren't actually relevant to the decision-making process. SPEAKER: One qualification here, to take that to the practical end of what you just said you would have to have a section -- if I was going to substitute, you know, new antibiotic A for rocephin, I would have to list all of the potential side effects of rocephin and then all the potential side effects of antibiotic B so that they could actually make a choice about whether or not the first -- and that isn't what we do. That's not common practice. DR. MENIKOFF: I understand. And I'm not saying you have to do that. This partly gets back to what level of information you need in the consent form, but I will tell you right now, you're probably familiar with this, there are


plenty of times when a consent form will basically say nothing about standard care. They'll just say, well, you know, you have this pneumonia, we want to try this new antibiotic, and here's what we know about the new antibiotic, without even putting in one paragraph of basic information that the alternative treatment you would get is a highly effective standard care that is highly unlikely to give any side effect to you adding one other element. I think the system would work a lot better instead of wedging a ton of information in consent forms which is going the same direction you're going is, to make a lot of additional information available, either as appendices or something else, that, by the way, we have a lot of information about what standard care is and here it is in case you want to look at it, but we're not shoving it all into this consent form that clearly is just going to overwhelm you with information that by and large is extraneous. So, I don't think we're that far apart. DR. NELSON: And just one other comment.


Ask yourself this question: If we got to the point where we had developed outcome-based measures of the quality of informed consent -- I don't think we're there, but people are working on it. I know Steve has a tool that he's done some research on, but let's imagine we have an instrument like that. Then the question would be this, all right, we put two forms up against each other. You've got the one form over here which lists everything, all right, because that's what the lawyers at your institution say you must and because someone back in 1942 objected and sued because something wasn't on that form. All right? So, you've got that one, and then you've got a form that's just boiled down to what is considered the essential components that a "reasonable research subject" may want to know to make a decision, and you are able to demonstrate based on that outcome instrument that the people that get the shorter form know more about the study. Which consent document is better? All right? So, we're not there, but I


don't see why we couldn't begin to argue, instead of having a sort of document-driven approach, if we can design certain outcome measures and then start to test the outcome of those forms so that we can get to the point where we could begin to tease away some of the information that gets just added on and added on and so no one in their right mind knows what's important anymore because you can't tell when you read those things. I can't tell. It takes me hours to figure it out personally sometimes. SPEAKER: Can I ask the representative parent to weigh in? MS. WALDEN: Yes. DR. KAHN: I'm not that person, but I'm going to talk first while she thinks. You know, this comes up -- I'll say this comes up on the RAC, too, as you might guess, and there are a couple things that I think complicate the attempt to simplify. So, for one, when you introduce recombinant DNA -- so, doing gene transfer -- other things kick in, like you're


concerned about reproductive risks, not so much in pediatric trials, obviously, but in adults, which is different than standard of care. You ask people sometimes to use barrier contraceptives for long periods of time, for instance. There may be additional imaging or blood draws. We ask people about -- we talk about autopsy. So, there are things that sort of become part of the research consent that tend to make it look longer and longer. It's not as simple as we might otherwise hope. And the other thing I would say is, it sometimes helps people, I think, and we ask investigators to untangle the standard of care from the research-only parts, because then people can see, this is what I would otherwise be going through, and here on this other column, or on these other pages, or on a chart or something, are the things that are different because of my potential research participation. So, sometimes it's helpful, I think, to see those together. I agree, we don't like 25 -- no one


likes 25-page consent forms in 10-point font, but I think we have to be a little careful about slimming down things so much that we're not giving people all the information they need to decide. MS. WALDEN: I would have to say that I have mixed feelings about this because in certain circumstances I think that it would be okay to take the standard of care information out, but, on the other hand, I think that there are lots of circumstances where parents really don't have a really good understanding about even the standard of care information as well as the risks associated with that. So, I think that having too much information in circumstances, especially with protocols like the ones that we're discussing today, I would rather take the extra time, take the extra effort, to read and understand and be sure that everyone is on the same level of understanding or at least that they feel -- parents feel comfortable enough with the information in order to make good informed


decisions. So, in some circumstances I think that it's okay, but in cell and gene therapy clinical trials, I don't feel very comfortable saying that I would consider that, or protocols like that. DR. BOTKIN: Good discussion. Thank you. SPEAKER: To what extent is the investigator responsible for explaining other options that families have? Let's say there is no standard of care but there are two clinical trials that are open and you're trying to represent those, you don't want to misrepresent the other one, however, you want to inform this family about their options. How much of that should go into the consent process? DR. BOTKIN: Good. And these would perhaps be options elsewhere as well. SPEAKER: Yeah, elsewhere. In other centers, so this has come several times up in my practice personally where there could be two different options, one of which could be gene


therapy which, you know, once it could actually affect their eligibility for the other trial. So, to what extent do you go into the explanation of other options the family may have without, you know, misrepresenting the research of the other group? DR. BOTKIN: Good, very good. Susan? DR. KORNETSKY: I'll take an initial stab at this. I'm sure other people have things to say. I mean, as far as your own institution, I was thinking about the comments that Steve had, and I think that if there are other alternatives or options at your own institution you absolutely have an obligation to do that. Whether you need to spend ten pages in an informed consent doing that, no, but you might be able to include that in the process and say, you know, there are other alternative trials that involve these things and we can talk to you about that and to include that in the informed consent process. As far as things at other institutions,


it's -- there are lots of other things going on and I don't look at it as specifically an obligation to have to do a literature search of all the other trials and other places and share that, but to share with the families that there may be other options at other institutions, this may be a mechanism that you may want to look at, this may impact your ability to participate in trials. To make them aware of that is important. You know, I have heard of individuals, research individuals, who have been research subjects, who have been very dismayed at -- and have made comments after -- well, why didn't someone let me know that there were other things? Or why didn't someone talk to me about that? Or they didn't tell me that I could have this done because they were so interested in getting me involved in their research trial. So, I think we can do some things. I think we have obligations at our own institutions, but to make people generally aware that there may be other things, I think, is ethically important.


DR. BOTKIN: Skip? DR. NELSON: Susan, I'm wondering, how would you fold into that an assessment of the capability of a family, in fact, to travel great distances? I mean, where would you be -- I mean, you're in Boston, so obviously there's a lot going on within the metropolitan area, but trials that are in San Francisco, for example? I mean, how would you begin to fold that information into it? DR. KORNETSKY: I mean, I think that's a family decision. I mean, again, I don't think you need to say, well, you can go to LA and get this, you can go to, you know, Houston and get this, you can go there, but to let them know that there are trials available, that there are different centers that are doing different things, to maybe give them information about clincialtrials.gov, or whatever, to say if they need some help in thinking about that, but if a family wants to pick up and travel to California for something, that's their right. I wouldn't hide that from them because of that.


DR. NELSON: I guess let me reframe my question. I guess, could you use -- I mean, some families don't have the resources, some families don't have support systems, they basically would not want to travel. I mean, could you use that as a screen, in some sense, to know if they're, in fact, unwilling -- they wouldn't make a decision to go out of state to the not to describe out of state protocols? I mean, how would you approach that? DR. KORNETSKY: I think you're getting into a tricky area for me to -- for us to assume that a family wouldn't have resources, therefore, we -- you ask them. I don't think -- I'm having problems with that. DR. BOTKIN: Jerry? DR. MENIKOFF: Just briefly, one related issue is depending on the nature of the therapy, it may be something that could be given outside of a research study, sort of an off-label use, that sort of thing, and that in many instances will be an important thing to describe being available and


the existing data there is that in many instances this option is not clearly stated. DR. BOTKIN: And this obviously is an issue that spreads across research in general, but, I mean, a little bit less of an issue, perhaps, here where there might be a very small club of people who are treating a very rare condition and you sort of know what other protocols might be available elsewhere. Anybody else have a comment on this issue? Yes, Ms. Walden. MS. WALDEN: I would just say that as a parent I would appreciate just the information or the reminder that there may be other things out there, that I can pursue information out on my own, and then it's my responsibility as a parent to do the research and to find a way to ask the questions to those who might be able to answer the questions for me. But I would appreciate the opportunity or at least the reminder to seek out and to look and to make sure that the options that are presented to me are the best ones for me and


for my child. DR. BOTKIN: Very good. In front here? SPEAKER: It's basically on the same subject. In terms of other clinical trials that are available and are, in essence, in competition with one institution's trial, there's a lag between when data are released and data are published. So, for example, if there's been data presented at a major meeting about some early results at another center, until that appears in a journal that might be six months or a year later. How do you deal with that? In other words, there's some data out there, patients don't really have access to that because it was at a -- you know, an American Heart Association or the diabetes meeting, and it's not going to appear in the literature for another 6 or 12 months. DR. KORNETSKY: I think this whole discussion about alternative trials is important as far as informing individuals, but the problem that there are trials or that data isn't out there, I'm really having problems thinking that's


strictly an IRB issue. It certainly is an institutional issue, it's certainly a practice issue, it's certainly something that you need to think about with your department, and there certainly are ethical issues. But I'm not so sure that this is something that the IRB has to totally manage, to say, well, you can't offer that trial because we know that we're waiting for data on another trial. I mean, I think other people in an institution, in a practice, and colleagues, need to take responsibility. So, I don't have an answer for you, but I'm also saying I'm not so sure this is totally up to the IRB to deal with. DR. BOTKIN: Skip? DR. NELSON: Just a cautionary anecdote, I would be concerned if early data that's not been fully analyzed or vetted, whether it's in abstract form or other form, could be potentially influential in ways that are unintended. I've heard, years ago, Jesse Gelsinger's dad talk about being told about one anecdote from the trial, that apparently there was an early response that didn't


then pan out on subsequent analyses, but he was told that by the principle investigator and that influenced his decision -- the dad's decision, not necessarily Jesse's -- to say, okay, I guess this trial is fine. So, I would get worried, if we're not just providing information about trials being open, but we then start providing early, abstract information or early information that's not fully vetted through peer review, that that begins to get a bit fuzzy. I know that information's available to parents, I'm not saying don't make it available, but there's a danger in that kind of data. DR. BOTKIN: Jeff? DR. KAHN: You know, just listening to this conversation I think that we can expect that investigators will understand their trials very well and be able to describe that to potential subjects. They may understand the trials that are being offered in their institution somewhat well. But once you get outside of that relatively small


orbit, we can't expect them to know what's going on in any great detail -- maybe in their field, but as you get further and further away, it gets harder and harder to imagine that they could actually give good information except to say, this is not the only trial out there for the particular disease or diagnosis that you or your child is suffering from. I don't know, absent that, what more we can require of investigators to say as they talk to subjects about recruitment. DR. BOTKIN: Well, and I would add, too, that -- I mean, this is the role of the expert on the IRB, why IRBs need experts, because they may well be aware of what else is going on in the field that could be relevant to a particular trial, and also the role of the annual review, and if new data comes through during that first year, then the trial needs to be reconsidered in the light of the new data. SPEAKER: Yes, I attended a three-day workshop in May that the FDA hosted, and folks from CBER were there presenting and the take home


message was basically they're seeing so much tox data with AAV, they said, don't assume you need to do non-human primates as that can be cost prohibitive. Robust rodent studies are oftentimes sufficient. Of course you have to show over expression if your transgene is not toxic, but is there the same perception on the IRBs that you sit on with regard to AAV? Are you seeing a ton of great safety data? DR. KORNETSKY: Too technical for me. DR. DENNE: Yes, it is a technical question that I'm not qualified to answer, but I will say that those are the discussions that we're certainly open to. We hope those come through in the protocol with strong justification of what the models are, why they're appropriate, why alternative models don't need to be pursued, so all I will say is that the IRB is certainly open to looking at the arguments. SPEAKER: Okay, thanks. DR. BOTKIN: Thank you. SPEAKER: Although I'm currently with


the NIH, I'd like to actually speak from the perspective of my former life as a clinical investigator and address the issue that's been the source of discussion, the recent discussion, about disclosure to patients about clinical trial options. I think that I would agree with what the panel said about the IRB's responsibility being the monitoring of competing trials for all the reasons that were suggested, including the potential to complete those trials which is a human subjects protection issue. I think with respect to identifying the options, I think -- well, certainly I would have always considered that to be the option -- to be the responsibility of the treating physician. Now, this is the treating physician. Now, when the treating physician and the investigator are one in the same, and they may have a trial that may be competing with another trial, I think this is a good test as to how well the physician investigator can actually wear both hats.


But I think, you know, a lot of people need to remember that that's not always in the investigative realm and it only becomes investigative when the patient becomes a potentially eligible subject or agrees to be a subject in the trial. The presentation of options is often in the realm of the treating physician and not the investigator of one or the other's protocols' responsibility. DR. BOTKIN: Good. Thank you. All right, I think we'll have one last question. Sir? SPEAKER: Yes, I see on the IRB forum that there's a great deal of -- I'm sorry -- on the IRB forum we see a great deal of confusion about who is a lawfully authorized representative or legally authorized representative, and of course the law differs among the several states. And I wonder if IRBs would find it helpful to get some reminders from the regulatory agencies that they do have to inquire into whether this representative is truly a lawfully or legally authorized representative.


DR. BOTKIN: Thank you. Comments? DR. DENNE: I mean, it is an area that we actually do look at and pay attention to, so, I mean, I don't -- I mean, even in the sense of periodic audits, while we will actually see whether in fact the legally authorized representative was in fact legally authorized. Now, those audits aren't overwhelmingly extensive, but there are random audits and we do look at that as part of that issue. DR. KORNETSKY: I mean, the only thing I can also add is that I think it's important to sensitize investigators at your own institution that there are -- if there are any issues that are raised that someone questions whether someone's a legally authorized representative, to seek guidance. And I do get, you know, several calls every other month about situations of, you know, this one came in with papers from here or there, are they legally authorized? So, I think that's the best that we can do is to get the message out if you have questions about that for research


purposes, it's really important to find out the right answer and get the right, you know, legal advice. DR. BOTKIN: SACHRP had a subcommittee on decision-making for individuals with impaired decision- making capacity and that went forward to the secretary. I don't believe SACHRP has heard back as yet, is that correct? So, there may be additional materials out there in the not too distant future about that particular issue. They spend quite a bit of time with the problem. All right, please join me in thanking our panel for an excellent discussion. (Applause) DR. WITTEN: I'm just going to make a few brief comments to wrap up. We've had a long and informative day and I think it would be impossible for me to try to summarize all the recommendations we've heard, but I'll just say it's clear that there's a need for studies in these very small populations which are many of the pediatric studies. The challenges are considerable. It requires a great commitment on


the part of the PIs to get these studies done. There are issues relating to the populations, the IRBs, informed consent, and from the perspective of the investigators that we've heard about today. So we'll be, you know, taking these lessons back to FDA and considerig what else we might be able to do. So, in closing I'd really just like to thank the speakers, the moderators for the panel session, the planning committee, Bernadette Kawaley, and a number of people who have provided logistical support for the meeting, and I'd also like to thank the audience who's really actively participated in this meeting today. So, thank you very much and safe travels. (Applause) (Whereupon, at 4:47 p.m., the PROCEEDINGS were adjourned.)



I, Irene Gray, notary public in and for the State of Maryland, do hereby certify that the forgoing PROCEEDING was duly recorded and thereafter reduced to print under my direction; that the witnesses were sworn to tell the truth under penalty of perjury; that said transcript is a true record of the testimony given by witnesses; that I am neither counsel for, related to, nor employed by any of the parties to the action in which this proceeding was called; and, furthermore, that I am not a relative or employee of any attorney or counsel employed by the parties hereto, nor financially or otherwise interested in the outcome of this action.

Notary Public, in and for the State of Maryland
My Commission Expires: August 11, 2014