Event ID: 440661
Event Started: 12/13/2005 8:57:01 AM ET
Good morning. This is a test.

I have few announcements. Dr. Amy patterson is out and Dr. Owe rilely serving with Dr. Patterson instead. We will begin this morning with a couple of announcements. First I want to remind all of the RAC members to sign in at the desk outside this room. Second, we've reserved a table for lunch today. I will remind you of this again in alleys in the back room. Third, for any of you who are driving today, please get your complimenttary parking passes outside of the rhyme. Dr. Riley will read us conflict of interest statement.

being a member of this economy makes--committee makes you--standards of ethical conduct of employees of executive branch. You received a copy of this document when you were appointed. In every meeting in addition to reminding you--we like to review the steps that we take and ask you to take in ensuring any conflicts of interest we address. Personal profession financial interests. We use this information for the base of--that could compromise your ability to be objective in giving advice. While we waive conflicts of interest for general matters we have relied on a great degree for you to attentive during the meetings for a possibility to arise that could affect your interest. If you reuse yourself in the discussion and leave the room. You are also required to reuse yourself from the preliminary protocol or review process. If you have any questions about the rule of conflict of interest or committee managers or officers will be happy to address them.

Thank you very much Dr. O rilely. We will procedure from the constitution of September 21st meeting. Dr. NaomiRosenberg.

I reviewed the minutes along with nick and I found no errors in the minutes that were presented.

Thank you both very much. The gene transfer safety assessment board rr report is next and Dr. He willbellda will leadoff.

I am begin with a brief review.

I forgot to take a vote on approving the minutes. I'm moving too fast. All those--no I have to take a roll call for approving the minutes.

Is that really a motion?

Yeah, just a second I'm getting myself organized. We're taking a brief pause. I need a motion to approve the minutes.

so moved.

I second.

Dr. Magisca moved and Naomi Rosenberg second. Take on motion.

Aye.

Wara aye,.

Aye,.

Wara--i'm sorry weber and grave and grave Weber. Thank you. Now that we have done that he will now move ton to the gene transfer safety assessment board report.

In the last quarter this were 18 submissions to the RAC. 12 of those were not selected for in-depth review and mustic discussion. These included nine for cancer, one for peripheral artery December, one for museum diseases and enfection December. Three plasma vectors, one a retro viral vector, one a her peas viral vector. Protocol is a plasma expressing mach mow globin A and aqua pore ra one. Procall 732 replicated adeno-virus with individuals with advanced tumors. 734 with individuals of Cleo blas Tacoma. 739AAV ser row type 6 for the first time expressing human placenta with individuals for the cystic fryrose sis.

Thank you very much doctor. Dr..

RAC 700147 amendments in the last quarter 37 were PI changes and two protocol design modification. There were 14 procall stage changes, 46 annual reports and seven responses to N1C1 and 41 other amendments. The RAC received notification from Dr. Enknit--ken knit wine berg in Children's Hospital in Las Angeles with infants of children with a patient previously not have a bone marrow transplant. Procall allows for enrollment of patients and high risk of more bility or mortality. As for the potential subject add seminated the investigators fell into this category and planned to offer his parents the option of enrolling him on other parol protocol 494 as well as transplant. A second notification to the RAC was that single subject exemption will not be included with the 609 which a study of children with disease.

Thank you, doctor. We are very privileged this morning to ra brief discussion before our first protocol presentation in risk-benefit assessment in early stage.

I think there was another one that Dr. Peter offs going to read.

I'm sorry Helen. Are you doing the next?

No doctor federoff of going to do it.

I poll gize.

This is protocol 699 a study of for the treatment of high Greg low ma using blunt cells genetically modified placing themselves on hold due to very preliminary animal studies that show add normal hue mat pleas sis amild proliffertive disease transgene in the animals and full pathology report and testing for replication confident retro virus will be provided.

And that is full amendment discussion. Dr. Owebellda will included this.

The gene tran fierce safety assessment board reviewed 11 seven were classified as A1 and possibly associated and unexpected and four classified as possibility associated and expected. After reviewing we thousand there was one adverse event in protocol 619 that's protocol by Dr. Crystal to teach patients with ser roid lipo know sis with an AP vector. There was a 4 and a half-year-old boy with a moderate disease who had surgery on September 20th. Eight days later two seizures which were controlled by reduce the anti-seize sure medication the patient had seizures nine months ago and not sense that time.

thank you. And now we'll go onto our presentation this morning which involves the risk-benefit assessment in early phase pediatric research. This is dear to my heart because I'm a pediatrician. Dr. GilmanGrave will discuss this and a brief comment for his specific expertise for being here this morning from since 1985 Dr. Gilman Grave served as of NIBT ofen fans and children. He works with the NIHnd programs on adult disease and especially interested it ceases disease prevent and ascertain takening the early and os pea yo porosis. He currently chairs the NICHD institutional review board. He also serves on the subcommittee on research involving children for the secretary's advisory committee on research. In this capacity federal regulation toss protect children involved in research projects and the children's health act of with 2000. He also searched as task leader entitled ethical conduct of clinical research involving children which was mandated in 2002. For these accomplishments he was been awarded outstanding and mare torous service medals. He is particularly well-versused and-suited to discuss the issues of protection of children in the conduct of clinical research. Dr. Grave.

Where did you ever get all that material?

I was handed it this morning and I summarize it.

I don't remember being in involving in all of that. First of all I'm not a pediatrician Dr. Wara imi'mn internist I've been provide here since 1972 and emmerged in pediatric issues and six month he had a the Burnham general hospital. I'm about to subject you to abuse by PowerPoint. I was telling Dr. Nelson this morning that PowerPoint is totally eliminated public discourse in this coun and people end up looking at the slide instead of the person presenting. But we'll have to put up with that because unfortunately when you're dealing with very fine nuances of distension between--discontinuingtion between the levels of risk of the words that were wren and they are very important and written by professors of law and lawyers. And they have stood the test of time. I think the three remarkable productions of the national commission between 1974 and 1979 first research on fetus that came out in 1975 followed by research involving children that we'll pay attention to this morning and finally Belmont report in 1979 these three documents are absolutely stunning. They're sort of like the constitution. We should be so proud of these document. The national commission did an incredible job if an atmosphere that was fear struck by tuskeyee revelations and in order to save federal research in biomedicine and pave the way for the establishment of biomedicines is as stonishing--I don't have copies of these three documents but they are available on the web but I do have seven or eight books to give out ethical conduct in clinical research involving children. Sarah car was sinned enough to give you one chapter which is 4. One of panelists Dr. Nelson in the back room I was a the humble project officers. That was any involvement with that introduction let me get on with the meat of the presentation. This is to remend you of the fact that we're dealing with very 71able children in this kind of research and obviously demand special protection. They are the future of our republican and we--republic and we need to take good care of them and remember any intervention we put on them at this tender age stays with them lifetime. One of characteristics of the children is they are growing all the time. Any intervention you may come up with in a child may have a long, lasting linger effect that is the heart and essence of the pediatrics which is the field of development biology. This is the third slide I would show you about the development and care and concern about drugs in that country back in the 1906 when they passed the pure food and drug act in sponges to the children when he were getting heroin and morphine and color ra form in their medicines to stop them from couching. Thisas followed the food and drug act in 1978 to the children who were given those drugs that lead to the food and drug cosmetic act in 1938. Finally a great tragedy of pha lid mid which some of you may remember in early 19 60s in yes, ma'amThere is a pattern here followed by a legislative Fiat. One of the important things about this particular piece of legislation which was remarkably important at the time is the third point the law of unintended consequences. They mandated it be tested in adult and instead of children. It led to therapeutic or fins which are children given drugs. Why some the current legislation is so child-oriented and child-focus for very good reason. Now about 80% of drugs given to children have never actually been tested specifically in children for that point. This slide is about the national commission. It was mandated in the national research act of 1974 which was introduced by senators Ted Kennedy, Jacob Jeffadvice and Walter Mondale and this was a absolutely a very important piece of legislation. It imposed a moratorium on fetal research lifted a year later after the commission's wonderful report. Research involving children is of interest. This quote actually comes from a letter that the Chairman of the the commission Dr. Kenneth Ryan a gynecologist from harvard wrote to president Carter he said it raise ethical concerns because of in come pentancy to give informed consent. It's hard to remember when you look at the two infants sitting there the such bliss. The send part is these concerns should not be answered by restricting participation and research to people that have come tent to consent. Namely adults. This is of the hard of in your rencode it was only people who could give consent. That's adults. One of key break through of the commission and job they set themselves to do was to find a way that children could participate in research and this was a letter to the president as I said. And the letter went on and Ken Ryan pointed out there are two key questions that have to be addressed. The first was under what conditions is participation in children and research ethically acceptable? The answer as it turns out depends on the thoughtful analysis of risk and benefit including non-beneficial or non-therapeutic research which might be the nub of the argument. Under second question what participation bet subject of the parents. How can we get around the fact that children cannot consent for themselves if they are to get permission or actively accept children's assent. There were other key vic riess that they had. Each one is pretty much autumn mouse and loosely overseen by the office of human research and protection how can they review the results of all 3,000RIBs. There's a lot of local authtomy here and some may agree and others disagree. The other is component analysis some may have heard some may not. It's deacon instructing a protocol to look at the parts to see if each has a certain level of refreshing or possibly the risks differ in which case you could find for one part of the protocol but not the other. Finally is secretarial review peen the department of health and human services review by the secretary and important research for some major serious problem that the RIB felt at sea with that they couldn't approve it. Maybe we'll kick it upstairs to the secretary and more about that later. The task the commission set before then as I indicated is it ethical to do research involving children. The nurenburg code would say no. 3 low again that--that thought only medical research for the interest of the child therapeutic research and dealt with this at great length and thought the term itself therapeutic research as an inherent contradiction. If it's research, how can you possibly know the outcome and therefor it couldn't completely be therapeutic because the so-called therapy that you're trying you think will be beneficial might have detrimental effects like they did. The minute you introduce the term research you are no longer talking about pure therapy. I think this was brilliant punctureture ofl Ramsey's balloon. He was big on good works and thought that fashioning of the concept of parental permission in children's asset was key here. And in a way some consider a benefit more about that later. The Belmont report. This is their final magnumowepus and it is short like the constitution. They had three key principles ba benefitsis carries the word benefit with it. They address children in it. They said we need to find better ways to treat children and it it's a benefit to justice in serving children from they are constantly figuring outweigh we could inject those tiny helpless creatures into research for their own good especially having experienced nur iturg and Tuscany and all thetion they said protection to harm first do no harm, second minimize harm and quoted Bernard who said injury no one regardless of benefits to others. And this is where Paul went to terribly awry in the willowburg experiments and he started injecting them with the Hepatitis vi withecause they may get it anyway. He should have paid attention to Claude injury known andd an important key principle we must keep many mind. It has entailed a discussion with the research of children without the immediate prospect of direct benefit and greater than minimal refreshing is inadmissible. That's the word they used. In admissible. They said you can't use such a strong word you mittt rule out a great benefit this is the attention of the concept and attention RIBs wrestle with everyday. Respect for persons is a little more straightforward and can be handled quickly. They simply indicated that individuals with capacity to consent 18 years old and above are treated auto mousely and any by definition has legally auto me. It implies and ethical and imperative to obtain consent of children and inform them of risks. That's my own deduction from that that could with argued but I think it's acceptable. The third principle is justice and we can quote historically the burden serving as research subjects fell largely on the full ward patients and fell to the private patients. They mention the tus key gee studied this ethical principle is children should be proposed the benefits of research as well as adults instead of being treated as or fins as you've heard and extrap lated down to. Federal policy as we know it is incorporated in so-called title 45 of the code of federal regulations part 46. Title 45 deals with public hmm health and part 46 deals with protection of human subjects. Subpart B is basically derived almost word for word from the national commission's research on the fetus that I showed you in the subpart D is derived almost word for word for the national commissions record research dating on children back to 1977. Spitee many lookees in the last years not a single word has been changed in the last 22 years and it definitely stood the test of time despite repeated challenges. This is the category of per miserable research in children. There are four of them. We'll go through these quickly now and focus a little bit more in detail later. But there's minimal risk. Greater than men mal risk but with the prospect of direct benefit which means you could go to extremely high levels of risks if you have a extremely high levels of benefits. As a minor increase over minutenal risk ind no prospect of direct benefit this is by far the most controversial four part of subpart. Two commissioners out of the 11 decrended from this and thought there should be no kind of research and subpart B for fetal research there is no such comparable phraseology. This is very important and very controversial aspect of subpart D that bears a lot of discussion. And then finally not within the scope of the above categories but presents a reasonableable opportunity to prevent or treat a serious problem that is the secretarial review that you heard me mention. There's a footnote here that the FDA has comeable and regulatory proinvestigatings 21CRF that is the code of federal regulations 21 deals with food and drugs. And they're subpart is 50 which deals with human row searchrotection. The definition in minimal risk generally is product for two vectors. But the discussion actually in 45C of our 46102 mens the men mal risk or probability or magnitude of harm or discomfort about paited in--anticipate in research are not greater than those daily life orfulysical or psychological testing. What I want to focus on is the vector business because you can have an enormous amount of harm airplane crash or car crash leading to death with an extremely low probability. For instance the risk of dieing for any one of news airplane crash is about 1 in 2 pone 3ion so we would get on the plane in a minute. I think there's few of us that wouldn't. Oddly enough death in the car crash is 1 in 6400 which is several three orders of magnitude greater yet we consider going to car completely minimal risk because we figure the probability is so low even know the magnitude may be enormous. I bring this up because many of the dire consequences in children as giving them an experimental vaccine as a risk of dieing 3 orders of magnitude less than driving in a car but it's hard to convince people of RIB of that. This is actual code fication and it doesn't make difference if it's their put particular or not. That's an--their pewic or not. Greater than men mal risk. This is key focus that you probably ought to be looking at today in view of some of the protocols that you are all involved with. And the risk is justified by the anticipated benefit. I think this actually implies heroic procedures that might be entailed in life-threatening illnesses for instance total body irradiation for total transplant for Leukemia. The second bullet is about available alternatives and third is about permission of the children and their parents and guardian. Only one parent has to sign for they aretic research. Now remember the IOM the institute book it was made for the benefit of children. In considering how to ascertain the prospect of direct benefit the institute of medical empaneled a group of experts pediatrics cling Dr. Nelson. They helped to guide our thoughts about benefit and risks had this book. I brought seven copies of it. You can scalp you one want of these I can get you one. It's easy for me to obtain more copies. This panel dealt with the prospect of the direct benefit and thought there should be a tanable positive outcome cure disease, relief of pain and increase mowility and we have to think about whether it leads to transient or permanent benefit. We talked about compensating levels of risk and benefit and one break through point they had was the business about collateral or indirect benefit he's in a study he must be learning a lot about biomedicine he's experiencing hospital care from the inside. He's getting great medical care and monitored. Nop of these things should be considered direct benefits they are definitely indirect benefits. You would be surprised how often committees wrangle with how there's a benefit from being in the hospital and being monitored and from getting a little bit of remuner ration. Interestingly enough the commission 30 years before this had gone to the heart and soul of the direct benefit by saying the expert of success should be scientific sound to justice whatever risk is involved. You would be surprised how many protocols we had to Wade through that have a fatal flaw. That simply aren't scientifically sound enough in their fast draft to go forward and even offer benefit. The institute of medical report recommended providing guidelines for RIBs. That's one of key point of this. They really thought that the subpart D as rine is really fine and shouldn't be tampered with. In fact it's written very interesting ways that encourage dialogue. You could say that given the same protocol two or three RIBs might come up with two or three different anes and you can't say any one of those answers is right. The process is right. The process is right just like our constitution in ways bedevilling arrest cane and ob truce. Some of the beauty is not defined but encourages dialogue. That's very important and key heart--from my point of view the key heart of protection is the dialogue that we all enter into when we deal with these kinds of protocols. They came up with an example of a study that is not approvalable under direct benefit. And let me just read this phase 1 pediatric cancer protocol will give a doesage with no probability of amillating the subjects' disease or management. The subcommittee of research involving children in human research protection that's we just love though acronyms. And Dr. Nelson sits on this panel too and he might have come up with this example. You give this dose just it so what the connectics maybe there's to possibility this small micro dose is going to alter the cancer of this child. You can't claim a direct benefit. But it might be approvalable under the next category which a mean for increase above minimal risk with no prospect of direct benefit that would be an argue that RIB would have to make with itself. Council for international organizations and medical sciences weightd in a few years ago about thought of face 1 trails. Because children are unique and affected by childhood illnesses it maybe impossible to extrapolate from children as adults we have heard. They thought we have to deal with this dilemma and sometimes go directly to phase 1 research in children not waiting for studies in adults. And this would be especially appropriate for diseases that don't occur in adults or pediatric respiratory deficiency or some of the congenital heart disease that children have. I think we'll see more and more of when you can start research in children without sort of short-circuiting prior researches in adults. We have resent FDA draft guidance on this too. Just this year the FDA reminded people that--investigatingors--that they already had in the way the guidelines were already written under the INDs they really could use early phase 1 clinical trails without a therapeutic intent that would involve screening or micro dose studies this is generally of limited duration and conducted before tradition and safety studies. This was a remarkable guidance out of the FDA and has an important foot not that Sarah car pointed out to me. This is a quote these types of studies would not be carried out in pediatric patients but I would say in some cases they could be. Generally, the word generally permits that I think. The second guidance the FDA come out this year is how to comply with the pediatric research equity act. This act of passed unanimously by both houses of congress a few years ago in order to give the FDA authority to require pediatric studies of drug companies if they had been used in children but not tested in children from 1999 to the present. This is remarkable action of couldess to a number of things that happened recently in in the pediatric community. In 1998 the FDA put out a pediatric rule ta said the same thing we have the right to demand a drug company to study drug in children if we think they should and support is necessary. This was taken to court by a number of groups would were very conservative and felt that this way overstepped the authority of the FDA and in fact justice Kennedy in 2002 struck this down and said the FDA did not have such authority but could ask congress for such authority. Well congress was so anxious to give this authority to the FDA because it felt it was so important that drugs be studied in pediatric age groups that he passed the pediatric research accunom mousely in both houses in congress even war against the Japanese was not unanimous. This shows how important in a town in Washington how important it is. If happened in 1990 when the institute of medical sen add another conference on pediatric oncology. An unsung hero is Dr. Summer gasee who some of you may know who thumbped the tub about this leading to the network in 1994 which is just had a wonderful partnership with the FDA. The FDA has paid much more attention to pediatric pharmacology over the past ten years. A number of offices and feed--pediatricians have make--we had a pediatric research equity act. The other thing that fed into this even though they weren't children one was April-year-old one was 22-year-old they had promise and potential and also the Erica gripes suit at onhoskins--john hop kins. It led to congressional angst for life threatening disease for which adequate treatment is not available may been earlier that happen my occur in less disease. The urgecy for the needs for product may justice the trails despite the relative lack of safety and ineffectiveness of the information. I think for life threatening illness that the pediatric research equity act paves the way for important studies early on. This is 46406 the minor increase overminimal risk and no prospect. There was to con 10 shes that two out of--contentious that two auth of 11--out of 11 descending. The in stute addressed this and said this is a slight increase in minimal risk. Duration of possibility and magnitude of the risk and also I'd say yes, risk, duration and possibility of magnitude. We come back to the two vectors we saw before they added duration. There is a idea of comensable ability you dothe treatment and not related to their experience if the past life. You have to argue otherwise it cannot qualify for 406. The bedevilling word and the panel thought it referred to a set of physical, psychological or effect health or well being. Some examples might be a genetic predisposition to obesity, living if poverty, live if a high level of lead in the environment like the kids in the Baltimore or being Annie yo Nate or adolescent. He was on the panel as staff. Arely in data? In fact recently about unpublish data. They didn't know she signed up for this but it's important. Finally the situation of the proposed subject. I don't know what that actually means. 44 or 6 a--446 and ofl important and overlooked by the way. Some kicked back and found not be be 407 by secretarytary panels. You wonder why there are four possibilities in subpart D why is one hardly used in the top one.

There's a group here in NIH who are very concerned about the fac that the rules and regulations are undefined and mean so many things to so many people. Interesting enough. When they went back to look at data and risk and children how RIBs look at this a paper was written by nars canand hop Kens in 1971 they asked several directors of and chairs of pediatric programs what they thought about a risk one I recall is where you particular a needle in the eardrum and let the fluid drain out. These are pediatric professors now one-third said a minutenal risk, one-third thought it was a minor increase above minimal risk and other third found unacceptable to be approved by the RIB. Here you have premeditation trick leaders unable to determine among themselves or come close to an agreement about a procedure. Here sexual activity survey. These are RIB chairs 22% said minimal. This is stunning to me that you can have so many leaps in immediate tricks and--feedatrics. 15% thought it was perfectly minimal. This was so studied so many R I B chairs felt so strongly about protocol changeby for change of captionners] The risk of playing football is 1 in 250 games. Risk of playing football or dieing in a car crash. This brings me to an absolutely stunning quotation from Alex that I never forgotten since I learned it in college. In America sooner or later everything ends up in court this includes REB members and there is a great fear in the land now and strong demand for in demmething should go horribly awry in the a studied end up in court. It has happened and end up again. I like to say no deed goes unpunished. Ethical principle and as a result of that. Differs from adults. Quite a number of drugs Ike gappapentin and total max--one was found to have twice the mortality of other comeable drugs in use in feedatrics. Ribbowvirine increased suicidal ideation. There are other things that come out in children not seen in adult. This brings us to the fact that our ethical issues continue to need an assessment from time to time. [please tannedy for change of captioners] -- or what bob would call the fallacy of the package deal is -- is part of what goes to avoiding the term therapeutic research but it's not meant, I think, to address that there can't be research that offers the prospect of direct benefit. But this prospect has to be put into context. It's not a threshold that you have to reach for anything to be prospect. So that would be I guess the key issue there.

Can I ask just one followup question? The hypothetical, on this whole prospect of benefit for you and your IRB. Let's say same kinds of animal studies and only animal studies have been done and you're talking about doing something on kids. One -- hopefully to emiruate deafness and the other blindness. One could think that those would differ in terms of harm. Would you look for the same quantum of evidence about prospect of benefit with both?

Well, neither is life-threatening.

Right.

And I think the assessment of the impact of blindness and deafness will vary depending upon the community you speak with. My own bias, I wouldn't limit theing withness *pness to apply any data, any adult data, any kind of preclinical data or on the nature of the condition. You know, the devil's in the details and the aply kablt of the models and the question of what can you get when you do it in adults. Is it applicable, is it not? It really has to be looked at in detail. It's a difficult question to answer hypothetically. I might add, I don't know what the IRB would do and I'm no longer the chair, so --

Okay.

I have a question. I'm also be willederred by the need as an IRB member myself to define prospect of direct benefit when our IRB, as well as nationally we've been instructed to try to separate science from our job of assessing risk benefit. I find it almost -- no, impossible, not almost, it's impossible to do.

Yeah.

Without assessing science and I'm wondering what your comment on that would be.

Well, you remember what the commission said. They said that you shouldn't even enter into this unless there's a possibilities that -- that this is going to work scientific cli. I mean, remember that -- here it is. The expectation of success should be scientific cli sound to justify whatever risk is involved. I think that is such a key quotation and implies so much about just what you're saying and our IRB wrestles with this all the time and I think that it seems to me some of the early part of subpart D actually mentions things that have to be scientific cli as well as ethically sound. So we regard this as an inherent part of our charge and I don't think it -- I don't think you can separate them. I completely agree with you. I don't think the national commission ever thought so either.

So let me then urge everyone to refocus IRBs on the science of the studies they're reviewing both locally at UCSF and nationally through a presentation by OHRP about six months ago to those of us involved in Pete yat Rick aids, I have specifically heard that -- and through the RAC actually, through comments by those being reviewed that, you know, science is reviewed over and over again and your task is to look at risk benefit. And my personal sense, my gut sense is, I can't do one without the other.

You're absolutely right.

Ms. Kwan?

I have two kind of unrelated questions. For the first one, I apologize if it's incorporated in your slides and I may have missed it. But the idea of minimal risk or slightly above minimal risk in allowing the research to go forward, was there any comment or interaction with the possibility that the risk could be eliminated by using another method or, you know, conduct -- obtaining the same piece of information without even minimal risk. Is that a factor? So, for instance, if -- if the trial were to be delayed six months to get additional data from an adult study, for example, but that the risk is considered minimal or slightly above minimal, is there any interaction in saying well, then you should wait?

I think that would be very wise counsel. And in fact, one of the key principals of the Belmont report was the -- all efforts to minimize risk at whatever level of risk you're at. So I would say absolutely.

Okay. The -- the second one, and I may be -- I'm going to take it out of context, but I think before this committee in previous sessions we have been faced with some investigators telling us that in a dose escalation study, let's say the dose was begun at a higher level than would otherwise be recommended because that's the only way that a pediatric research subject might possibly derive therapeutic benefit, how does that fit in with the reports and the directives and the advice?

Well, I guess my first question is how do they know what the dose is and why do they have to escalate it? Do you remember why? They thought that -- in other words, this is working adults at a lower dose but wouldn't work in a pediatric patient? Is that the idea?

I -- I don't -- can't cite a specific study, but I think the comment was, well, your animal studies would suggest that you start at a lower dose. Why did you pick such a high dose to begin this? And that was in response to that kind of question.

Well, it sounds like they're begging the question, if you ask me. That doesn't sound like a good answer, frankly. How do they know? But of course you could point out some of the drugs that I just mentioned that do need higher doses in pediatrics. I mean, this is not unheard of. I don't know if we're talking about actual drugs or vectors, but this is a very important problem and that's why I brought up the new guidance from the FDA that actually permits the studies of phase 1 in so called microdoses just to actually get PK without hoping for a therapeutic benefit. I think this is an important break through. But the FDA says that you always had this capability of -- of presenting a study like that but it's not used very often. That's why they came out with the recent guidance in April.

Consent is a critical component of ethical studies in adults. Can you just briefly discuss the role of assent particularly related to age and children?

Well, that's an accident children and the commissioners felt that it was up to local IRBs to determine and sometimes in cases each individual subject whether they were capable of assent or not. At the NIH we generally use and it's been promulgated a few places age 7, but that is very arbitrary and certainly there are 7-year-olds that are terribly precocious and others that aren't and couldn't understand what you're talking about. So you have to tailor it to the individual. And in some cases there have been guidances that the IRBs should actually look at individual children or have a -- a proxy or a liaison to follow the -- the consent or assent process in individual children depending on the severity or possible harms that could come in such a study.

How severe does the disease have to be before you would say the child -- that the parents consent would override the child's assent?

That's very interesting. That's a wonderful question. And people have wrestled with that and it's written down in -- in a number of documents and guidances and it is that the -- the parents -- if there's absolutely no other way this child could get similar therapy, this is the only avenue to improving the child's health, the child squirms and squawks and says no, the parents can override that, absolutely, if that's the only hope for therapy in that child in the research venue and it's clear and it's written down in a number of places.

Could you comment on the role of the states in imposing more restrictive limitations on pediatric research than those that are offered in the federal guidelines? For example, I'm told by our legal department at the Hopkins IRB that the attorney general of this state takes a pretty dim view of most of those categories of pediatric research. Could you comment on that? Not Maryland in particular, but the principle.

The Erika grimes suit brought up this very ancillary decision by the court, which was that parents or guardians could not consent to enter a child into nontherapeutic research that carried any quote -- any, underlying risk and Dr. Nelson and I were talking about this earlier in the corridor and that was changed because they realized that they -- the court realized it was out of consonance with federal guidelines that did permit research, minimal but not any risk. And the federal guidelines are written such that at least as far as determining the age of assent or consent in children and emancipated minors and minors living apart from their parents or minors who are parents themselves, state regulations hold -- are the most important or paramount. The state regulations could, at least in some cases abrogate any overriding federal regulations.

I was curious about that nice example you showed about the IRBs indecision or lack of consensus on what constituted risk. Has a similar study been done for adult? Is there a similar lack of consensus for adults?

That's a wonderful question. I don't know. This has been a very poorly studied area. Partly because it's -- it hasn't been an appealing area of study for journals to publish. I mean -- or even for people to go into. It's pretty remarkable. I mean even if people of pediatric pharmacology. When I was in medical school I heard I know he's a pediatrician but he's a pharmacologist too? I mean, there were certain areas people didn't go into in certain periods but now we have a lot of people interested in pediatric pharmacology. But now we're desperately needing quantify kags as the biomedical establishment becomes much more finely nuanced and many more risky procedures are on the horizon. We need these data but we need them desperately in children.

It would be interesting to go back and redo the perception of risk question to physicians -- kinds of risks that you had up there. Frequent blood draws, skin testing for allergies and ask those who are responding whether they've ever been involved or have been involved within the last five years. For instance, in obtaining blood from children. I suspect that the -- the differential in per spepgs is at least in part based on direct experience.

Absolutely. I thought that -- well, you're right on target it's almost like you're reading my mind because I've been thinking some of these pediatricians or chairs of departments probably haven't seen anybody except on rounds or done a procedure for years where as the interns and residents, we do these things all the time. Are there other questions? Yes.

Initially you talked about the two vectors attributing to the assessment and in the IOM introduced ration. Can you further expound on the duration and are the there any pragmatic points you'd like to convey to us?

I think the duration refers to -- I think this came up from a peripheral point of view in the guise of insulin clamps and how long would the insulin in the fusion be running. And if a child should become dangerously high bow glycemic, was there an instant remedy? And so in that case we're talking about a duration of a couple of hours, so using that as a -- as an example in your mind, you could extrapolate other kinds of procedures. Who knows how long is a child exposed to risk and sit reversible during that period of time and should something untoward happen would it be reversible there after.

Other questions? Thank you very much. This was a wonderfully informative and provocative presentation and will help us as we move forward to look at gene transfer in children and youth. We're going to take a break. We -- we had originally been scheduled to break until 10:15. It is now 10:15 so we'll break until 10:30. Thank you all.

The first protocol for discussion during this RAC meeting is protocol 740 entitled the phase I safety study using adenoassociated viral vector to deliver the gene for human RPE65 into the retinal pigment epithelium. The principal investigator is Albert Maguire from the University of Pennsylvania and the sponsor for this protocol is Katherine High from the same institution. And Dr. Maguire is going to present.

Good morning. I'm not Dr. Maguire. Dr. Jeanne Bennett and we're going to give a tag team presentation this morning.

So Dr. Bennett who will be presenting and in what order so we'll be prepared.

It will be first me, then fratzer Wright, then Albert Maguire and then Chris rocky. But first, on behalf of the team, I'd like to express our -- our sincere appreciation for the opportunity to discuss our propoetzed study here at the RAC today. I'd like to extend special thanks to the RAC committee and the reviewers for their comments and questions and I'd also like to thank Dr. Graves for providing such a helpful frame of reference this morning for studies in pediatric populations. So this is our phase I study. As you can read. The phase I study with LCA using adenoassociated viral vector to deliver the human gene remember 65 into the retinal pigment epithelium. it is involving investigators for the appointments at chop and at the University of Pennsylvania. As you just heard the PI of the study is Dr. Albert Maguire who will be presenting this morning. He's also my husband, and I -- I am the scientific director. I I'll lead in -- I'm obviously leading in now and Fraser Wright who will also be speaking this morning is the director of the CCMT vector core. So the oraleder of presentation is thus. First I will give you a little bit of background about the human disease and then I'll describe the preclinical and proof of concept studies, and the preclinical safety studies. Then Fraser Wright will briefly describe vector manufacture for the proposed human clinical trial and then Dr. Albert Maguire will discuss the proposed human clinical trial involving pediatric subjects including topics such as subject selection, trial design and future plans, and finally, Mr. Chris rocky who is a patient advocate will say a few words. So the original proof of concept studies took place over five and a half years ago starting in about July of 2000 and involved the investigators shown in this slide here. There are three centers involved. Wonderful scientists from Colonel, from University of Pennsylvania, and University of Florida and there were three different dogs involved in this -- this -- beginning study, one of whom is named Lancelot who you'll see in a second. We've also benefitted tremendously from funding for this study and the succeeding studies from the national eye institute and foundation citing blindness and other institutions including research to prevent blindness. This study has evolved because of the exciting result to include an expanding set of collaborators that you can see here. Clearly there are multiple institutions involved now in addition to the additional ones at the University of Pennsylvania and University of Florida. There are also investigators from other institutions and they're all listed here and I apologize if I inadvertently left any one person out. So what is labor congenital amaurosis or LCA? This is an early on set retinal degeneration with severe vision loss present from birth. It's also associated with abnormal eye movements where the eyes have abnormal movements. The symptoms are apparent in infancy and early childhood and mutations in any one of at least nine different genes can give rise to the disease. Importantly, there's no treatment for this disease and there's no cure for this disease. Mutations in the following genes have been found to cause LCA. And they're all listed here. And this listing is from a wonderful web site called RETNET which keeps progress of retinal genetics in retinal disease and I'd like to introduce you to the cells that are involved in -- in the primary cause of this disease. This is a cartoon of a cross section of a human eye going from front to back. And here if we focus on the back on the retina, which is shown here, this is a cartoon structure of the cells and the connections of these cells in the back of the retina. We're going to rotate this tissue 90 degrees and you see it here, and the cells that -- that are initially involved in this form of LCA caused by mutations in RPE65 are called the retinal pigment epipeel yum cells. These cells are the nurse cells to the photo cells which are the sensory cells in the retina and they provide important ingredients including vitamin A precursors to these cells which is critical for vision. Now, normally these cells contain a protein called RPE65, stands for rep specific 65 protein. As you can see here in a normal dog retina when it's subjected to immu know histic chemical for RPE65 protein, the protein is spent specifically in these cells where as it's absent in an affected dog. RPE65, the gene encodes an enzyme which is a retinoid isomerase and this is what is the critical blockade -- the critical block aid in the disease is caused by a lack of production of the important vitamin A derivative which is necessary for vision. Now, the proof of concept studies that I'm going to describe to you take advantage of two naturally -- one naturally occurring animal model and one genetically engineered animal model and these are shown here. The nch rally occurring animal model is a dog. The Swedish dog shown here is Lancelot who is one of the first three dogs who benefitted from treatment with a vector containing the normal version of RPE65. This is at his third trip to Congress where he's been an important advocate for the importance of biomedical research. And he is the dog who is now five and a half years post treatment and still seeing. We've also taken advantage of a mouse, which is knocked out of RPE65. This mouse was generated by Michael redman at the NEI and generously provided for these studies and we performed studies in adult mice and in fetal mice that I'm going to describe. Importantly, both of these models have similarity in phenotype to humans with LCA caused by RPE65 mutations. Now, the vector. You've heard the word vector early this morning by dra Graves and here it has a different sense. The vector here is what we use to deliver the corrective gene to the diseased cells. And in this case it's an adenoassociated virus. A recombinant virus which is not capable of replication and even in wild type virus has never been associated with any disease in humans or in animals. This virus that we're going to use contains serotype II ITRs. Those are inverted repeats and capsules which you see in this diagram and within it the DNA that is delivered to the cells includes a con stitchtive promotor. The Dr. Maguire will tell you a little bit more about this and production of this in a few minutes. But here are the scientific results and this shows the results that we first noted in Lancelot, the dog who you just saw who has been to Congress. So what you're seeing here are results from electroretinograms. This is a noninvasive global testing of retinal function, the electrical -- power of the retina after it's exposed to light. And here on the study was performed by Sam Jacobson and colleagues and what you're seeing here is a normal sighted dog, the wave form increasing accordingly to be very reproducible and quantitative and gives a good quantitative figure of the amount of vision that the animal has. As you can see, the wave form is very reproducible in terms of timing and amplitudes. Before treatment Lancelot had a severely abnormal ERG and you can see even at the highest intensity of light exposure there is barely a blip. These months after treatment you can see that he has an essentially normal appearing retinogram especially when you compare this what you see in a normal sighted dog and this effect has persisted. This shows 13 months after treatment and again it's now five and a half years after treatment and the results have recently been published in fact this month in molecular therapy. As I mentioned, this therapeutic effect has persisted and shown here it is evidence again published in the same article just going over three years and again it's now five and a half years where you can see year one, year two, year three. There's per distance of the effect. Now, what I'm going to show you here is a movie demonstrating the visual behavior which results from this treatment. You just saw the electroretinograms and -- and I think this is really going to show why this is so important. So what you're going to see here, there are three dogs. This dog here is just a friend. He's a large dog who is normally sighted and he's -- basically they're playing with two dogs, twins, who are both affected with the disease but one was treated and the other was not. And so what you see here, the treated dog is playing with the other dog. He's walking around picking up objects that he sees on the floor, moving without a problem. The untreated dog doesn't see the object that's thrown to him. He walks over here, bumps into this dog. Turns around, trips over the other dog. The treated dog is ready to catch objects that are thrown to him, walks around with them, eager to play. The untreated dog knows there's something going on but he doesn't see what's happening. In fact, he didn't see the handkerchief which was thrown to him. He wants to play but he just bumped into this other dog. He walks around very tentatively. He doesn't see this object thrown here and he just bumped into the other dog again. Again the other one is walking without a problem and the untreated dog walks around very cautiously lifting up his paws very gingerly because he's so used to bumping into objects and doesn't want to bump into them. Now, we have continued over the past five and a half -- five years to look at the results in additional series of dogs and this shows a series of 26 dogs which received subretinal injection of AARV 65. What this graph shows is again, electrofizz logical responses, ERGs, in these dogs and it's monitored in terms of rod photo ceptor function. The similar results are found for cone function. So what you see here is on the vertical axis, these are amplitudes of the ERGs and the normal sighted dogs who are untreated have amplitudes that fall in this range. Tun treated dogs or untreated eyes have amplitudes in this range, and this dashed horizontal line you see shows the maximal amplitude in an untreated dog. Each of these triangles, by the way, underKates one eye that was -- was subject to study. Now, when we injected this particular intervit rally, there is no improvement in rod -- however, when the material is injected subretinally so that it can -- 23 out of 26 of these eyes showed success in terms of recovery of this electrophysiological response. When we went along and further analyzed these mice in terms of histology, what we found was with these particular eyes which fell below this threshold, the injection actually went subRPE. So if we look at this data from another viewpoint, that's 23 out of 23 successes when the material went into the subretinal space. I mentioned that Lancelot is presenty five and a half years post treatment. He's certainly healthy from a clinical perspective and enjoying his trips to Congress and dinner dances and continues to enjoy vision in his treated eye. To date there have been rescue in these affected animals and I'll show you in terms of dogs and mice. So out of 35 injected eyes from affected dogs 97% of those have sho*e shown success in terms of rescue vision after injection subretinal injection of the greater than a.1X doles of AARP 65 and I'll show you that shortly. These are in ages approximately 3 months to 14 months of age when they're treated. In contrast, soo*e oh out of 16 dog eyes showed a therapeutic effect when they're injected intervit rally or subpigment epipeel yum. For mice we found that when we injected mice in utero, nine oud of 13 eyes showed success. That's 69% and I'd like to remind you what a technical feat that is. This is really submicroscopic surgery. Then with adult mice, mice at post natal ages of 2 to 47 months we found 24 out of 30 or 80% of the mice showed response to the treatment. And -- and importantly, this 80% is our success rate in accurate subretinal injection in adult mice. This is -- so this is -- if you scale it to that success rate, this is 100%. Then when weed at affected mice at post natal ages greater than 157 months, IE, very old mice, we found that only 16% of those eyes showed a therapeutic effect. preclinical toxicity studies of all the animals showed that they all remained clinically healthy for the duration of the study. No adverse effects in terms of food consumption, hematology, clinical values, there was a mild and reversible ocular inflammatory response after surgery except for one eye and by the way, ocular inflammatory responses occur naturally after surgery in any species. And in response to one of the RAC reviews I'd like to describe what we think the cause of that extra inflammation was. And that -- the -- that is as follows. In the fall of 2001, lab grade vector preparations were combined and these are four different preparations to inject a set of 11 dogs or 20 eyes. And we found that a number of these eyes had inflammation after surgery which is, what I just said, we typically see this after surgery, and all except one responded to medical treatment and that medical treatment involved eye drops and in some cases some con gynotitle injection of steroids. We were very concerned about this inflammation particularly in this one dog and we did a study to evaluate the possibility that our vector stock was contaminated with bacteria or there was bacteria present within the eyes. We tapped the eyes, grew out the sample, found no evidence of bacteria. We thought perhaps the vector was replicating evening though that's very unlikely and we looked at that by looking for presence of rep and cap by PCR and by immoo no histochemistry and so no evidence of that. And importantly, we've not seen any sorts of inflammatory reactions of any of the dogs initial Liz treated or those that have subsequently been treated with 65. And we think we've identified the cause of this inflammation which is shown as follows and this is published in this month's journal molecular therapy. When we ran out each of the vectors which were combined, I said the -- remember I said there were four different vectors which were combined into one solution. When we ran them on a protein gel and did silver stemming of the gel we found that one of the vector preps gave rise to a series of multiple bands on this gel where as a -- another vector, which was coinjected with it had what we expected to see, which are the these predominant bands associated with AAV. And so just to conclude, we feel that the most likely explanation for the inflammatory response in this set of dogs was due to the coinjection of this dirty vector prep in these eyes. Now, importantly, all of the dogs that we followed long-term after coinjection of these batches of vector showed rescue of visual function. Now, what about vector spread beyond the retina? The RAC has traditionally had an interest in what happens in terms of transfer to the testes and the ovaries and what is possibly the transmission of parent to child. We've seen no evidence of transmission of the DNA from treated dogs or mice, and in -- in our series of dogs, Lancelot, who was one of the three dogs, was -- is the proud father and grandfather of more than 50 puppies, children and grandchildren. He was mated with an affected dog and none of those dogs have first of all shown rescue of the phenotype. They were all born blind. And second of all, in biodistribution size which we'll get to we found no evidence by PCR or presence of the vector in the gonads. Mice we breed from the initial stock provided by Dr. Redman and we don't on purpose cross treated mice with other mice, but in certain situations these mice do breed actively when we're maintaining them long-term so they have reproduced and we've seen no evidence of transmission of the DNA to any of those offspring. Now, I think it's important to bring up results from a hemothelia B clinical trial with which Dr. High has been involved. In one trial, 140 trillion, that's a lot of zeros, vectors, that's more than three log units higher than what we intend to inject in the eye. This many were injected into skeletal muscle in humans and there was absolutely no evidence of PCR of vector in the semen samples of these affected males. In the second study the same amount of virus was injected systemically into the hepatic artery and in those individuals there was transient detection of vector sequence in some -- in semen of some of the individuals, but this all cleared by 16 weeks after treatment. So we think that this risk is minimal. And now I'm going to tell you about the possibility of vector spread to the brain and elsewhere. We're particularly interested in the brain, obviously because the eye is connected to the brain via the optic nerve. Let me just stop this for one second. And so one question is where is the vector localized, where's the protein localized. And so what you see here is a fundus view of one dog immediately after injection. The eye extends in this extent of the screen. What you're seeing here is actually the refleckive to pedoum of the half of the eye and the optic disk is right here. This is where it receives the injection just a few minutes before we took this picture so the virus is contacting the epipeel yum cells in this portion of the retina only. We let this dog go, measured its vision, it was rescued and then we looked at the tissue by immu know fluorescence to locate the 65 protein. And what you're going to see is a cross section going all the way across the retina and this movie was very nicely done by Kelling and it's again in this paper which was just published. So here is the entire section of retina made by overlapping these sections. And what you're seeing is immunofluorescence. We're going to get to shortly, the region where the material was injected which is right here and here you can see this apple green color which reflects positive RPE cells localized only to the injected portion of the retina. Notice also that there were no gangliocells. This is important because those cells send their ax son out through the optic nerve. They make up the optic nerve into the brain. So the question is do we see this protein in the optic chi yachl of the brain, which if we don't, that's a good indication that it's not making it to the brain. This is representative of an immunofluorescence of the optic chi yachl of the affected dog. And you see this on kick chiasm section. Now, it's also important to note that we have been able to successfully repeat the administration in contralateral eyes treated with AV 2 despite small increases in AB 2 specific antibodies after the injection of the first eye and this particular experiment involves the transfer -- we injected one eye of a mouse with AVE GSP and waited three weeks until there were antibodies present in the serum and injected the contralateral eye and this is what we see at the six week time point. Similarly in monkeys we injected one eye subretinally with GFP and waited for -- in this case, seven months for each of these animals. There was evidence that there were antibodies, mild antibodies, levels of antibody in the serum of these animals and we injected the second contralateral eye and these are the differences you see in those two eyes. There's exquisite -- Now, one other question is how do we determine the dose levels of -- for the therapeutic effect? And this has been done using the cumulative experience with 29 eyes of affected dogs after subretinal treatment with AA VRP 65 in this particular dose range, plus a dosing study in 15 affected dogs over a four and a half log unit range of doses of this virus going from this level to this level. And the data is shown here and was presented at the RAC last June. This is performed by Dr. Jacobson and colleagues, and what it shows is, again, the ERG amplitude threshold at -- before -- in untreated animals, control animals and then after treatment. And this is, again, in dogs. Each one of these circles represents an individual dog, because one eye of these dogs was treated. And the vector dose is increasing as shown in this grated series down at the bottom, a portion of the graph. What you can see is as the dose increases, there is an increasing number of eyes that respond -- show response in terms of visual function. And I'd like you to note that even at the lowest dose, which we plan to use in our study in -- in a human population, there was evidence of a treatment effect, and Dr. Maguire will go more into these details. So one more question about dose and inflammation. We have looked clinically at affected dogs and unaffected monkeys after subretinal delivery of AVRP 65. This is clinical grade vector and these animals were scored by a team of veterinary ophthalmologists or human ophthalmologists for the degree of inflammation going from the front to the back of the door. Cornea, interior chamber, lens and vitreous. And the inflammatory changes were apparent in both monkeys and the dogs after injection of the vector and after injection of the vehicle. There was a minor trend of increasing ocular inflammation with dose familiarly at the 3X dose, however, most of these -- or essentially all of these inflammatory changes resolved after one month. There were slightly higher inflammatory changes noted in the canine study than in the nonhuman primate study and this is not unexpected. In fact, the canine eye is known to be -- to respond with much more inflammation to surgical procedures than the eye of any other species including humans. So finally, we've looked at preclinical toxicity data in terms of the injections and the tissue effects. This slide shows the photograph of a retina that -- three months after its subretinal injection of a 1X dose. This is the to pedodal half and this is the nonto pedodal half. The cannula entered the space at this location and this leaves predictably a scar in that location and what I've drawn here is the borders of the retina as identified from the original injection photos. This bleb extended out in a dumbbell shape. If we look at the needle entry site, predictably you see trauma at the injection site and the retina is folded up and there's sometimes rosettes formed in this location. In some eyes, in some regions of the retina we did see iminflammatory cells and these are shown here. Please note that there are no inflammatory cells identified in the subretinal space which is where the vector was delivered and in addition, there is evidence in some of the animalss of thinning of the outer nuclear levels shown here, the photo ceptors and this is likely a result of the injection procedure itself, but please note that there are definitely distinct photo ceptors remaining here. So finally, we feel that's very important to conduct a GLP preclinical toxicity study to evaluate the potential toxicity and biodistribution of the clinical grade vector that will be made at the CCMT, and we'll -- we have designed these studies and have been discussing the design of these studies with the FDA and these -- this study will involve both early and late time points for analyses. So now I'd like to hand over the podium to Dr. Fratzer who will tell you a little bit about the vector corps in the clinical grade vector.

Yeah, so, okay. So I'm Fraser Wright. I'm the director of the clinical vector corps at the center for molecular therapeutics at children's hospital Pennsylvania. I have five slides and I'd like to briefly go over our new facility at the children's hospital of Philadelphia, describe the process for which we will make AV 2 gene transfer vectors and give some description of some o of the characterization that we will be forming on the vectors. In this first slide, we have a -- this is a picture of the Abe ram's research center. This is part of the children's hospital of Philadelphia. In the building is located our new clinical vector core. It's up on the 12th floor right here. This is a 10-year old building and the area we've designed this in is actually only two years old. On the right side in panel B is our clinical vector core. It's a 5 room facility. We designed this in collaboration with the FDA at a type C meeting back in April and we have implemented recommendations from that meeting. The area involves a staging room, gowning room, subculture room and a degowning room. Unidirectional flow and other standard come poets of GMP manufacture. We also have a QC lab closely associated and an R&D lab in support of these activities. In total we have approximately 2,000 square feet of space supporting these activities. Can you hear? Okay. Now I have two slides that will describe our process for making the AAV vectors. In the first slide I will talk about our biosynthesis. we start with HEK 29 human embryo cells from a well character rietzed master cell bank. We see these cells in roller bottles and once we've reached the appropriate level of approximately 70% we perform a triple transient transvection of the cells. Foe lowing this transection process we perform a media exchange into serum free media to reduce serum components. The purification method that we have developed and optimized is outlined in this slide. Essentially we harvest the material and remove some of the culture components. Media components by filtration and -- the concentrated material is liced, the cells are liced by microfluidization ho mojization and the vector, which can pass through a point to a micron filter is recovered from this and separated from the cell debris which is contained in a filtration step. Subsequently we apply this material to an exchange column. This is a 50 H S column which we have previously published details of for purification of AV 2 vectors. Concurrently we perform a digestion step on the vector. The material that is recovered from the columns and essentially purified AAV particles however it is composed of everybody advertise capsids and genome particles. Following the gradient -- we appoint two micron filtration and we have made arrangements for final labeling and Vile fill and finish at a qualified contract manufacturing organization. I won't go through in detail the -- the list of assays that we perform for our quality control characterization of the vector however I will inindicate this list includes qualified assets for the -- these assays are performed on each and every lot of the vector. I've outlined the methods that are used here and these are generally standardized and in most cases published methods. As the last slide I just want to give an indication of the quality and purity of the vector that is produced using this manufacturing or G and P manufacturing process. I've indicated two sodium -- electrophoresis gels here. In the left panel is a gel in which we've loaded two -- and stained with silver staining. It's a highly sensitive method to pick up proteins. We'll pick up the product as well as the impurities. Essentially all we see at these loadings which give substantial levels of the VP 12 proteins, these are the expected proteins, these are observed and e season usually no significant levels of impurity proteins are observed. Similarly when we perform blue staining on much higher loadings of gels so this is another way of assessing the purity using a different type of a stain. We see essentially no significant bands other than the VP 1, 2,nd 3 corresponding to the expected AA V 2 vector proteins. So I'll finish off with that and pass the podium to Dr. Ma Dr. -- Maguire for further presentation.

Thank you. I'm Albert Maguire, the PI for this proposal. I am a retina specialist and surgeon at the University of Pennsylvania. In order to eliminate any potential conflict of interest related to my participation in this and other trials, I forfeited any financial benefit related to a pending patent based on this therapy. And likewise, my spouse and collaborator, Dr. Bennett has wa*ifzed any financial interest as well. To begin my presentation, I'd like to go over some of the clinical features of this condition which are important to consider in context of this proposal Leber Congenital Amaurosis is an incurable disease which begins with near blindness and ends with complete blindness. Affected individuals are born with poor vision and this poor vision is slowly extinguished over time, ultimately resulting in total blindness in adulthood. In the year 2005, there is no treatment for this disease. Care is entirely supportive and is based solely on adapting to a life without vision. This is, therefore, a disease of morbidity, not moralty. LCA results in loss of function, namely vision. It is organ specific, actually tissue specific in that it only affects the retina. Leber Congenital Amaurosis is a rare disease. In the form of LCA formed by mutations in the RPE65 gene is rarer still. There are very few individuals that have been identifyed with this gene defect and it is estimated that in the entire United States the number of people with this disease may be on the order of 2,000. This is truly an orphan disease. Leber's amaurosis is a pediatric condition. Patients with LCA do not succumb to the disease but their eyes do. Their retinas do eventually die. It is not relevant to speak about an adult form of this disease. There is no such thing. Most investigators accept the notion that the possibility of deriving a benefit from treatment diminishes with age. And this is what underlies our rational for proposing a pediatric trial. These are photographs from two LCA patients. One age 13 years, the other 27 years old. This is the optic nerve. The macula, which is the center of the retina where we get our central vision. The retina -- note the extensive pigmentary changes in this older eye, which occurs in areas of the generated retina. The retina is fused in these areas and is surgically inaccessible. We are most interested in treating the macula, which is here, a 3 X 3 millimeter area of nerve tissue which is responsible for our ability to read, recognize faces, drive, and so on. You can see just looking at these pictures the potential challenge of doing surgery in this older patient with this surgically inaccessible area of retina. Arguably, our animal -- our animal studies are most properly considered as pediatric models for LCA. The vast majority of our gene transfer experiments in animal models were done in animals under one year of age. The dog studies were done exclusively in animals under two years of age. While delivering the vector, it's possible in older animals, the surgery is technically challenging due to difficulty of injecting the older degenerated fused retina. What is clear is that our greatest success in terms of recovery of visual function occurs in younger animals. As previously stated, success in the dog model was documented in animals less than a year and a half years of age, and investigators at other institution -- other institutions report similar results in young RPE65 mutant dogs. Studies we performed in aged mice with RPE65 disease, demonstrated a substantial reduction in treatment affect in the older animals. This was true both in terms of the percentage of animals that showed a treatment response and also in terms of the absolute degree of physiologic rescue of the retinogram. This is data taken from the studies on aged mice. First, note the decrease in the treatment effect in terms of the recovery of the electric response of the retina. In the -- notice the difference in scale. This is 25 microvolts in the older mice. 50 microvolts in the younger mice, indicating a much larger electrical response in the younger mice. Further more, the raw percentage of treated animals, which demonstrate any evidence of rescue is substantially decreased in the older population. You saw this slide before. We have over 80% success rate in terms of any recovery in ERG in the younger group and much less, less than 20% in the older group. Parenthetically, again, this difference between 100% and 80% is attributable to the fact that our success rate in achieving subretinal injections in mice is on the order of 80%. I'm pretty good at doing these injections, but I'm not that good to get every single mouse injection. So the main rationalE, so this is the main rationALE for a pediatric trial. We feel this is a pediatric disease and therefore appropriate to do a clinical trial in children. Like many other inherited pediatric diseases the opportunity for benefit is lost over time and from the point of view of a clinical trial, the amount of useful information is diminished as well. If we are evaluating respiratory distress in a premature infant, how much do we learn by evaluating pulmonary surfactant in a normal adult and should we put healthy adults in a study for which they can derive no benefit? Another factor which we should recognize relates to the difficulty of doing surgical delivery in older degenerated retinas. Not only is there less viable tissue, there's less tissue. As I stated before, the degenerated retina is fused together and cannot be successfully injected. The older the subject, the less retina available for surgery and the greater the difficulty of doing that surgery. We recognize the important ethical considerations inhernts in a phase I clinical trial in a pediatric population. We have sought guidance from our -- from representatives from the children's hospital IRB to both better understand and to respond to the issues involved. The level of protection for child participants in clinical trials should be set at the highest possible level. We have implemented several layers of protection in our protocol with special emphasis regarding informed consent and willing participation. By far and away, the most serious risks in this study is the use of general anesthesia. The risk of morbidity would be limited to the potential complications involving the one treated eye. Based on the preclinical data, we believe the above risks are justified by the anticipated direct benefit to child subjects and the relation of the anticipated benefit to the risk is at least as favorable as available with alternate approaches of which there are none. Although the primary purpose of this phase I study is the collection of safety data, based on results from our animal experiments there's also the prospect of direct benefit to partis -- participants in the study. This is not a pharmacogenetics in healthy individuals. We will be evaluating primary and secondary outcomes of visual function for evidence of a therapeutic beneficial effect. Second, while order of preference places adults before children, we feel that a study of adults may yield diminishing returns both in terms of potential benefit and in terms of knowledge to be gained from the study. As stated before, this is a pediatric condition. All agree that the ultimate target population will be infants and young children. Performing a study in older adults with devastated retinas may not provide adequate information to proceed with further -- further studies in a pediatric population. Now, on to the study. This will be a phase I study evaluating the AAV vector in individuals with Leber Congenital Amaurosis. Again, know that this is a safety study. We anticipate a potential for benefit and vision will be one of our outcome measures. This will be a dose escalation study with three groups of three subjects, each, and a total of nine subjects, so that's 3 X 3 groups, total of 9 subjects. It should be noted that the lowest dose of vector to be used did show evidence of efficacy in our animal model studies. We have chosen the sample size based on the principle that only the minimum number of subjects should be exposed to the potential risk in order to obtain valid information needed to proceed with further investigation. Three subjects per dose cohort has been typical for early gene transfer protocols. It is critical to keep sample size to a minimum, particularly for rare disorders such as Leber Congenital Amaurosis. The proposed sample size, should, however, provide sufficient safety and dose response data to establish a likely best to initiate a plan in younger children than this. The time interval between subjects and between dose escalation is based on acute safety data and the biologic considerations of transgene expression. Based on the known delay of up to 6 weeks between the delivery of AAV 2 and the onset of transunit expression, this interval will be between the first and second subject -- the interval between the first and second subject will be six weeks. There will then be a minimum two-week delay between the treatment of testimony second and third subjects, and between the third subject and the first subject in the next dose cohort. To be eligible for inclusion in this study, subjects must be legally blind due to RPE 65 form of Leber Congenital Amaurosis. Both eyes must be sighted and we will treat only one eye. The other eye will serve as a nontreated control. Subjects will include individuals between the ages of 8 to 18 years of age, and I will elaborate further on the rational between this specific age selection later in the presentation. Standard precautions will be implemented for females of childbearing age. Subjects and families must be willing to adhere to the study protocol as discussed in the informed consent document. Individuals participating in the study must likewise be available for long-term follow up. We have given careful consideration to the consent assent procedure for this study. We have met with representatives of the children's hospital of Philadelphia IRB to discuss the issues and to -- and to develop what we feel is a reasonable and appropriate approach. Our goal is to provide the highest standard of protection possible in terms of informed willing participation. With this in mind, for subjects under the age of 18 years, we are requiring both parental consent and patient assent in order to participate in the study. At any time if the subject decents or if a parent with draws consent the subject will be excluded from further participation but will be provided study related care. In our written response to this committee we have elaborated on other aspects of this process including which individuals will participate in the process, the availability of additional reference materials such as audio tapes, the use of talking points and so forth. One aspect which I'd like to mention here is the presence of an anesthesiology representative in the consent assent process. The major risk in this study will be the use of general anesthesia and therefore we would feel that it is most appropriate to have an anesthesiologist as part of this informed consent process. We have responded to the comments of the committee and have expanded our informed consent to comprehensibly detail the possible risks related to the surgical procedure itself and to the administration of either too little or too much transgene. We will implement exclusion criteria relating to ocular systemic and the psychological condition of the subject and family. These are detailed in our study protocol, but there are some aspects I'd like to specifically mention here. We will be screening all subjects for the presence of neutralizing antibodies to AAV 2 and exclude these with those of 1 to 1,000 although we found that the retina is a favorable environment in terms of immune response, we will nonetheless take these precautions. Finally, a subject may be excluded from the study if in the investigator's opinion, they are some -- they are somehow unsuitable for participation. In my experience, I have excluded subjects from other studies because I have felt their participation would be unduly stressful for themselves or their caretakers and I feel especially strongly about this in the context of this present proposal. Both -- both health related -- for our assessments, we will be looking at on -- and the physiologic in terms of effect of toxicity with regard to structures of the eye. As we have developed a extensive table, which is in your handout in the protocol for systemic evaluation based on the world -- world health organization toxicity scale and we will use this in addition to -- we have stocking rules based on the dose limiting toxicity and we will be incorporating a data safety monitoring committee which will have access to all this data and they will help decide about the need for additional s*up subjects at a dose -- dose limiting toxicity is identified and they will also determine -- determination of further participation in the trial. Both health related and vision specific quality of life assessment will be incorporated into this study. A trained technician will deliver the quality assessment instrument either in telephone conversation or in face to face interview. An assessment instrument for health related quality of life has been validated for children -- yeah?

Move your slides in the other direction.

Thanks. An assessment instrument for quality of life has been validated for children adolescents to age 2 years of age. The vision quality of life assessment is problematic. The national eye institute has developed a visual function questionnaire which has been validated to measure vision targeted quality of life in adults, but this has not been done for children with low vision. We would clearly need to modify the existing instruments to be both age appropriate and vision appropriate and I've been in contact with experts in low vision and in pediatric ophthalmology for input on this issue. before closing, I'd like to revisit the issue of age selection in this proposal. I've already discussed the rational for doing a pediatric trial in general. I'd like to now share our reasoning behind the specific age selection in this study, that is 8 to 18 years of age. There are two additional risks that come into play when we consider children below the age of 8 years. First, children under 8 are susceptible to loss of vision due to amblyopia. If there is something that transiently impairs vision, the visual system can be impaired on a central basis. The older the child, the easier this problem amblyopia is -- the easier it is to treat and by the age of eight, children are no longer at risk for this problem. Second, there is the issue of eye growth. After birth, the eye grows rapidly so that by the age of 3 years, it approaches adult dimensions and by the age of 8, the structures around the eye, the eyelids, the orbit are of adult proportions. After the age of 8, size is not a factor in terms of surgical risk. Looking forward, we foresee using the information from the 8 to 18-year-old group to develop a clinical trial for the younger age groups. Based on risk benefit considerations, we would like -- we would move to the 3 to 7-year-old population where the risk of amblyopia may be increased, but the potential benefit of rescue would likewise improve. We would proceed lastly to the youngest subjects which pose the greatest risks in terms of both amblyopia and surgical challenges, but likewise, may derive the greatest benefit. We have tried to weigh these multiple considerations bearing on risks and benefit in coming up with our age criteria. In closing, we hope the RAC members will take into consideration the balance of risk benefits we have presented in their deliberation and recommendations. Dr. Bennett and I have described the scientific and medical considerations which have gone into the design of this proposed stud diz and for a perspective from a parent of a child with LCA, I would like to introduce Mr. Chris rocky.

A year ago I never would have imagined -- a year ago I never would have imagined that I'd be standing here. Our beautiful son Ty Christopher was born on December 18th, 2004. Excuse me. He's our first baby. It was the happiest day of our lives. When doctors pronounced him healthy, you were ecstatic. He was the best Christmas gift we could have ever gotten. Two months later when Ty wasn't making eye contact -- I'm sorry. We knew there was a problem. After many doctor visits and testing, we had our diagnosis. Ty has Leber Congenital Amaurosis we were told to take our baby home and adjust our expectations. Sorry. Ty is blind.

If you want to continue, that's fine. You don't have to. We understand.

Yeah, I want to continue.

Okay.

Ty is blind. Our hearts broke when we were told that there's no treatment. I can't describe how we felt at this time. It was very dark time in our lives. The beautifully illustrated children's books we bought for him were replaced with plain white braille books. The nursery we so lovingly decorated for him he sadly can't see. The bicycle I couldn't wait to buy him will be instead a white Cain to help him get around. My dreams of teaching him to play baseball in the backyard were gone. My heart sank the day I realized my wife would never see his face light up when he walked into the room. All our dreams for his future were taken the day we found he has LCA. I'm sorry. What makes this even harder for us is that this is a genetic condition. We planned on filling our house with children, unfortunately any future children we want to have will have a 25% chance of also being born with LCA. When the time comes, we will have to make that difficult decision whether or not to have more children. LCA has not only affected us, but also our families. Ty is is a long awaited first grandchildren on both sides of our families. Seizing the pain in our parents' eyes when they come to see their grandchild is devastating. I can't tell you the amount of pain this has caused not only us but our families as well. This is why I urge you to let those wonderful doctors to perform their trial for gene therapy of LCA in children not only for Ty but for many other families affected by this. It is so important for us to do all we can to help these children see. They deserve their sight. We understand the doctors are trying to develop a treatment and that there are some risks but no guarantees, but unless they're allowed to try, nothing will get done. This Christmas, Ty can't see the festive decorations in our house. He can't see the Christmas tree. Or the pile of presents that will be under it for him. But with your help, maybe in a few years, he will. You see, our son Ty,, thank you.

Thank you very much. And thank you Ty, as well, for joining us this morning and I want to thank the entire group who presented for clearly answering at least, I believe clearly answering the majority of the RAC's questions and for directing your presentations towards what we consider to be the major issue for discussion this morning. That is the conduct of this particular gene transfer trial in children and youth. And because the thrust of our discussion this morning is going to center around the risk benefit analysis of this particular study in children and youth, we've chosen to reverse our usual discussion sequence. Generally we begin with questions, comments regarding the science of a study and conclude with questions and comments regarding the ethics, the risk benefit analysis, the consent document. Today we're going to reverse that and we're going to begin with Dr. Powers questions and comments. Dr. Po*uzers?

Thank you. Just as a preparatory remark, I'll say that just reiterate what was said this morning by Dr. Grave. There are very few kinds of protocols that would come before the RAC, probably of greater public interest than one that would involve pediatric research subjects. I think Dr. Grave pointed out quite nicely that it's -- it's a very controversial area. It's been controversial from the beginning, even though the language has remained as Dr. Grave pointed out the same, so too have the debates. And so trying to understand these very difficult slippery concepts of minimal risk and ordinary life and things of that sort, it's really tough and it would vary among members of the RAC, if I had to predict if we were to be looking at this or any other protocol, our own gut feeling, I think at this one moment might vary in part because, you know, we talk about these decisions as being decisions under risk, but they're probably decisions under uncertainty and that is to say that we really truly can't quantify some of these risks or some of the benefits either. We -- we rather impressionist cli gesture at those benefits soy want to thank the investigators for spending a lot of time in the written remarks they gave us which were only briefly summarized here in a public arena. They tried to do an awful lot to address a task. I think it's really a tall order. Let me start with I guess my most relatively minor points first in inverse order as I've put them in my questions for the investigators. My first question for the record was an open ended question asking the investigators to discuss the quality of a life assessment, how it would be administered, by whom, what kind of information would be gathered, what its overall focus was. They provide add considerable amount of information in the public discussion today about that. It was very helpful to see, not only in terms of just what it would be like from the experience of the family and the subject, but from what kind of information scientifically would be gathered. So that was very helpful and my question has been rather fully annalsed, I think, in that regard. Other scientific comments -- colleagues, I'm sure will have comments in that area, nonetheless. Secondly, I asked to describe the -- the process of assent. And they did provide a great deal of information about that, some of which we've heard in a public presentation today. I want to commend the investigators for their detailed and thoughtful discussion they provided us of how that process would work and I want to commend them also, I think, for taking probably the most risk aversive posture that might be done here. We're requiring the -- the con sent of both parents and the consent -- the assent of the child, and to make an individualized judgment about whether a participation would be too stressful or inappropriate otherwise for particular subject and subject families. And the process that was laid out in some detail struck me as a -- as a very thoughtful and nuanced and cautious one so, again, I appreciate the level of reflection and responsiveness to that question. Thirdly, I asked about the selection process, and within the -- the discussion today and then the document, much information was given -- was very helpful to see why they thought about not going below 8 why they thought about staying as low as 8 as compared to maybe reaching up to the slightly higher years where the process of comprehension and understanding might be considerably advanced in the subject themselves. They seem to provide a great deal of information in that area and again, I -- I would dez fer further comment on that to my scientific colleagues as to where they have more fine grained accounts. But I appreciate the efforts to provide some considerable information about subject selection and the age choices. Finally, in essentially the biggest central question, is the overall process of risk benefit. Dr. Grave offered this morning a -- a cat gorization and typically we would think of -- of a phase I study in this arena or any other arena as one in which it's nonbeneficial research. Where the prospect of benefit is -- is one in which we need to be sure and informed consent process is not overstated and that we're really doing safety studies, and so you start at least with a baseline assumption at least for the purposes of further discussion of whether this is a nonbeneficial study such that you need to have a more cautious view about how much more than minimal risk is likely to be there and how much generalizable knowledge. I gather from the presentation, this is my first really question at this juncture for the investigators is that their ultimate discussions with their IRB and their own internal discussions would lead them to characterize this as the study in which there is some prospect of benefits, and that -- and it's in that category among those shown by Dr. Gave this morning that they would characterize this study and so my first question is a compound one, if you will. Is that, in fact, your assessment of which category for analysis this falls within, and a second part of that is what would you say to one of your own IRB members if they were to say, I see this as nonbeneficial research, greater than moral levels of risk, considerably more and perhaps maybe a good candidate for the 17th of these studies to go to the secretary.

Well, to answer your first question, I -- I would reiterate, we do consider this a phase I study with the prospect of efficacy, prospect of benefit and that's how we have -- we will be presenting it to our IRB. And in terms of the science -- in terms of going to the IRB, in fact, we've had informal discussions and very helpful with our IRB. We would address that question by -- once again presenting the science aspect. I think what you've seen with the animal experiments rarely do, I think, have two animal models for a condition like this, rarely do you see such compelling evidence, both in terms of the functionality of the animals, their behavior, and in terms of the objective physiologic response. And those are the -- those are the arguments I would present to the IRB in that -- in that context.

The -- I guess another component is -- is to think out loud about the kinds of things that might change the risk benefit equation even if only marginally. Could you comment I guess that your conclusion is that some proceeding ahead with some adult studies would -- would not even marginally change the risk benefit calculus in this instance? Is that a fair judgment

Well, I'd like to point out that this is a continuum, this disease. And actually, we will have -- in looking at it by definitions from different groups, we have adult, addless sent and children in that 8 to 18-year-old age group. The concern I have is when you look at, again, from a scientific standpoint, a retina, a tissue which is completely atrophic and nonviable, there is much less possibility of deriving useful information that can be used to project studies in the younger population. In addition, I think somebody had mentioned the aspect of doing a -- the pharmacology of a smaller eye. And, yes, these eyes aren't completely -- these aren't totally adult size, so we will be deriving information that I don't think you can get in normal full-sized adult eyes by shifting into this smaller group. The dosing may turn out to be different.

Yes. Thank you. I appreciate the one part of the equation, which is the uniqueness of information to be gained from this population. My -- my question actually went to the other side of the coin slightly, and that is whether proceeding first in a kind of staged way with some further adult studies would provide you with information that would change the risks that you profile as you move to this population. Would there be no gain or some gain and what kind of gain if there is some gain would be attainable?

I guess my answer to that is we have a lot of excellent animal safety data already, which we feel more than adequately addresses the safety issue. And in fact, we've had discussions with the FDA about the safety issue, and they're in agreement, they -- they feel that we would only need a small additional animal study to satisfy their cry tier iteria which I think are pretty stringent and they had not recommended a adult trial.

Okay. I think at this juncture I would like to hear more from the scientific colleagues but I think that certainly sets the predicate for discussing whether this in fact fits that particular category or one of the other kinds of categories. These are not easy calls. I will simply say that. And I -- I do commend you in particular for having this extended discussions at the IRB at the front end and I appreciate the time you spent trying to help us get a handle on that. What you're likely to hear in here and beyond is -- will be messy because reasonable people will deeply disagree about how to characterize that, even among those who very much appreciate the terribleness of this condition and the importance of moving forward with a lot of clinical -- preclinical animal model material. So I will stop for the moment with that.

Ms. Shapiro?

I too want to thank the investigators and the other speakers both for their presentations today and for the very complete response to -- to our questions. Many of my comments and concerns and questions overlap with Dr. Powers so I'll try to be brief. But just a bit more on the risk benefit ratio which of course is key here. I, myself, found compelling the discussion of the prospect of benefit for children in this protocol, which leads me to ask the investigators if they care to answer this as an aside whether given that IRBs always are charged with a risk benefit evaluation, whether they were wrong to have approved the adult study. You can answer that or not, I guess. U I'm still concerned and --

I'd actually like to hear that answer.

Good. Me too.

And perhaps not as -- as -- as directed as wrong, but perhaps -- well, let's see what you answer. Wrong is a strong word.

Yes. Sorry.

I'm going to take the advice of my colleagues and just say that that is a study that's -- has another sponsor and although I am actually the surgeon for both studies and I think there is a rational for both, I don't think I should comment in detail on somebody else's --

Pleading the 5th. That's wise.

Perhaps inserting different study. They're two different studies with different goals.

Yes.

However, I, too, like Dr. Powers am going to be anxious to hear about the cat gorization of risk for these children and I myself was not focusing when I head this as a nonscientist and the risks of the anesthesia as nearly as much as the risks of the unknown, I suppose. I appreciate the investigators' responses about cleaning up the informed consent document a bit so that the references to the child and the child's parent is not going to be confusing. There -- there was an issue that -- that I asked the investigators and they kind of threw back to us about whether this particular informed consent form should talk about the Gail singer case and to me it's just really out of place -- and you think it is too. They responded that in the past, RAC reviewers had asked for that to be included so I don't know if we're going to talk about it or what. And finally, the assent process which I do appreciate the expansion in terms of description about that, a statement is made that if a prospective child participant decents the child won't participate in the trial which sounds good although I don't know what would count as decent from an 8-year-old so I hope some additional thought will be given to those tough questions about the process.

Thank you. I'm going to go next and I've already commented that the reason for asking that this study be further discussed publicly discussed with the RAC is the nonfatal nature of LCA, the likelihood that these children will live for decades, we hope they will, and the issue of risk benefit in children and youth when a similar study in adults has been proposed and apparently is moving forward but is not yet enrolled. Terms of the preclinical studies, I asked whether any of the data from the nonhuman primate studies that -- were performed in conjunction with the earlier adult study, which we've reviewed would be relevant to this study. And the answer simply is and it's perfectly acceptable, different sponsor, different study, truly we should be relying on the preclinical data that's directly relevant to this study and I certainly accept that. I asked that the group discuss the evidence for the absence of vector spread beyond the retina. -- in particular, any evidence of spread toward the central nervous system, because, again, in the earlier adult study, we heard evidence that there was spread toward the central nervous system. I really appreciate the careful attention to this sort of data that was presented to us today, and am very comfortable with it. That is, there's no evidence of spread towards or to the CNS from either the dog or the mouse model. I asked for some evidence or some description of whether vector transmission occurred from mother or father to infant, that is vertical transmission, and the information which was presented to us was compelling, especially -- I don't remember the exact number of dogs that Lancelot has fathered, but many, many, many, and although, unfortunately they were subject to the recessive nature of this disease, none -- no vector -- no demonstration of vector was seen. I asked for the range of antibody both to AAV and to RPE65 following gene transfer in the animal models. It does occur. The protocol has been modified to take into consideration preexisting antibody to AAV and to add that as an exclusionary criteria. I -- I'm not sure. I'll just -- this is my personal thought, not to go into the RAC statement. I'm not sure I would exclude because in dogs, who have been -- who've received A AV to one eye and who have had the presence of antibody, successful surgery has occurred. Successful gene transfer has occurred to the contralateral eye. I'm not certain that in the presence of that data I would have modified the exclusion criteria, but I'll leave that up to others. That's truly not for the RAC's discussion. That's your decision. In terms of clinical trial design, I asked whether there was any other experience with subretinal injections and the proposed age group and asked whether that specific risk could be discussed with us and I -- and the answer I received was perfectly satisfactory and really is directed towards my own ignorance, which is, we don't inject subretinally in -- in children, so there is not, for many indications, so there is not a large body of evidence oris k specific data to present to you, but we believe, we the investigators believe that this is virtually risk free in this patient population and at this particular age. And of course the -- the group has invoked the risk of anesthesia which is perfectly appropriate in this age group so I think that's fully satisfactory as well. I asked how the dosing was determined and we heard a perfectly satisfactory discussion of dosing based primarily on Lancelot and the other -- other dogs. I asked that the investigators include both in the protocol and in -- and in appendix M some discussion of how dose escalation would occur taking into consideration the acute safety data. This was not in the original protocol that we received and it now is, including a TOX table for DSM V and it's fully explained in the material that we received from the investigators is absolutely more than adequate and I thank femme there are that -- for including all that for us. And I asked about the time interval before dose escalation in order to fully assess DLT and I suppose potential efficacy as well in -- in the children and youth. We heard that there will be six weeks between the first and second subject in each dose cohort and then two weeks between the next two and that that is justified on the time that it takes for expression of the vector in the preclinical data. That's a very -- it's a satisfactory explanation. Terms of the consent, I thought the consent was quite well written and easily comprehended. I, of course, appreciated the provision of an audio tape for the youth who by the time they will assent will be blind and unable to read. I appreciated a copy of the consent in braille for exactly the same reasons. I asked that the investigators consider surgical risks which we the RAC have been made aware of during our earlier public review of the adult protocol in which I recalled as including such issues as corneal decompensation potential cataracts and the consent has been modified although I also understand that these are remote risks for children and young youth in comparison to older adults so the consent has been modified. And I asked whether there was a risk of receiving either too much or too little of the transgene for obvious reasons and the consent again has been modified. So all of my concerns have been more than adequately addressed and I thank the investigators for their careful attention to details which are perhaps more difficult to address in children and youth than they are in adults and Dr. Vial, you are next.

Okay. Thank you. So yes, I would also again like to reiterate my thanks to the investigators. They have provided extensive responses to my questions in great detail and I will just for the record sort of read my questions into the -- into the record. So my opinion is scientifically I think there's a very strong rational for this trial that's well grounded in the literature and in the prez clinical studies, certainly Lancelot and the other dog studies I think are very compelling in terms of the effects of this gene therapy. My predominant issue was related to the toxicity, and although toxicity is relatively low, it is not absent, so I was particularly interested in the severe toxicity in a dog if with the rapid ocular inflammation was not controllable and I just asked if there were f*urltder details of that and the investigators presented a lot of clear data regarding their interpretation that this -- this adverse event was due to impure vector preparation and I think the -- the evidence for that is compelling. Just had one question, just -- just to sort of push it to the limits as to whether there was any expression data obtained from that dog just in case this toxicity could have been caused by massive overexpression of the transgene indicating some toxicity that if -- if in vivo gene transfer would be very efficient there would be some sort of toxicity but in general terms the data provided regarding that toxic event was very compelling, I think, and thank you for providing that. I asked whether the investigators had any evidence of integration for the vector and they replied, again, with a lot of detail that there's -- there's preliminary evidence. There is none -- there is evidence for persistence and with -- with Lancelot 5 and a half years post treatment there don't appear to be any toxicities with integration which were addressed by the investigators. One of the one of the areas that I was interested in is long-term, if -- if there is expression of the transgene and this transgene is being expressed in children who have never seen it immunologically due to the mutation. One of the chronic chances of developing immunity, the cells which are expressoing this gene and this protein and again, there was extensive discussion from the investigators which I am grateful for. I'm not sure if my questions can be answered readily in terms of chronic levels of T-cell reactivity against protein over long periods of time but certainly there's good evidence for the more acute levels of auto immunity being very low in terms of antibodies and so on. I asked also that in protocol 677 the studies were undertaken with monkeys and are these data relevant and available to the PI in this protocol, that's an issue which has come up earlier, and the -- the response to that is essentially it's a different study and I might come back to that just right at the end. I asked whether there would be any selection as to the RPE65 mutations that will be included or excluded related again to the chances of raising auto immune responses to the transgene if the mutation is one that prevents the immune system ever having seen this product, and the canine mutation in indeed as I understand it is a known mutation so again Lancelot is ongoing evidence of that and may be not an issue. So I also asked about the systemic distribution and the sensitivity of the studies and they were very well addressed. So essentially all of my questions have been very well addressed. Any outstanding issues would be the chances of auto immune reactivity against the transgene in a chronic situation down the line, and I guess I would just -- my final point is -- which has been ratzed earlier, one of the links between this current group of investigators and those investigators in the previous adult study, I understand that these are different -- different studies, different sponsors, but I would perhaps like to raise it just one more time in the sense that there's a -- there's a compelling argument here that this is a pediatric disease. We've heard that today so it does raise the question of whether there's any value in gene transfer to adults as proposed in an earlier trial and if that was a completely separate study, then I think it would be sufficient to say this is a separate study, separate sponsor, but there are data from -- from the previous group and there is clearly a connection with the current study in terms of shared surgeon and so on which just leaves me with a feeling that maybe there is some sort of conflict -- I don't mean a conflict of interest in any way, but it would be northeast nice to see a greater level of open communication between the two studies if there were data from the previous -- from the adult study that was relevant, will that be readily accessible by this group? If there were shared members of the group. So that's -- that's really all I have to say, but I do thank the investigators and think they have some excellent preclinical studies here. And thank them for their responses.

So we are left, really, with three issues that I -- that I see. Yes? Ms. Kwan?

No, we're not -- we're going to have a discussion. I wanted to summarize first what I think is left and then we're going to have a discussion. I hope with some participation from the public who's here as well. Issue number 1 is Dr. Powers, what is the risk benefit as we see it for children in youth? Is this slightly greater than minimal risk with prospect of direct benefit to all those who participate? We need to discuss that. Second, is Dr. Vile's remaining concern that there may be chronic autoimmune activity in these children as they grow older with express transgene?

Yes.

That's a remaining concern, I believe.

That is something that is in my mind. You know, I don't want to be hyper critical and I'm not sure how I would test that but that is a concern.

That is a concern and third, I believe the group, three of us mentioned it, remain concerned that the ongoing studies in adults, study in adults with shared members of the current group, that those results should be accessible and freely exchanged between groups in order to enhance both studies. I think that's as well as I can state that. So I'd like to open this up for discussion. In particular, I think discussion of the first of the risk benefitish shuz where -- where are we? Dr. Powers, where do you see us? I -- I'm inclined to see things in the same way that the investigators in the discussions are seeing the same things. I think this is the kind of case where the quality of the animal model data give us at least a reasonable basis for thinking this potential benefit. I mean, it's something incongruent with the way we categorize phase I studies but this is something the RAC has learned to realize that this is not like cancer studies and other things, that classification scheme would seem to recommends one cat gorization, but I don't think that's a poz it here. So I think that scheme ma, I think that's a reasonable cat gorization. As to the second half of that given that kind of cat gorization, I think it would fit quite reasonably within the category of slightly more than minimal risk. And I was initially, like Dr. Wara, particularly curious about the interretinal, it has the kind of layperson's yuck factor, like my God, that's got to be scary. But I suppose we submit ourselves to all sorts of things in routine medical environments and this doesn't sound that much more disturbing. Given the -- the information about the vector, which seemed to be another big problem, it doesn't seem to be looming as large in my mind given the information about the -- the anesthesia. I think that's important and I think the investigators, I mean counter intuitively to me, I think they have the right to further elevate this up in their discussions. Given the prospects of the -- of bondness more fully, slight loss of function in one eye given what they can expect otherwise, that particular adverse outcome doesn't look beyond the pale to me either. So increasingly, absent any further kind of conversations and my scientific colleagues the main things I see in the tradeoff list look within the reasonable range of judgment that an IRB might make and it wouldn't occur to me that the RAC would be necessarily in a position at this juncture to advise differently from what I think an IRB might reasonably reach as our own conclusion in this matter. That's my preliminary assumption of that absent hearing anything else more compelling that I've missed on that list.

Thank you. Many Shapiro?

I have a question for the group. The risk benefit ratio necessity quantifying somehow kind of both risk and benefit, I myself as I said before am convinced that there is a prospect of benefit. So my scientific colleagues and those with experience on this RAC which I don't have much of, do we feel comfortable getting our arms around to a sufficient extent, the nature of risk with animal data only? And the quantum of animal data that we have on this? Short term, long-term, acute, chronic, the whole ball of wax, or do we need something else?

Well, I'll take a crack at this. For me, the problem is that as the investigators presented to us today in the protocol, this is a pediatric disease, so the risks that we see in adults may not inform at all what's going to occur in children. For instance, postop rative cataracts. Those kind of risks that may occur in adults may never be seen in children or vice versa. There may be risks in children not seen in adults and that, for me, is at the crux of -- partly at the crux of why we need to do research in children. Diseases are different, risks are different, benefits are different. So it's really, in my view, the balance of risk benefit in children and youth that needs to be looked at. And sometimes one can learn from adults, but you know, as we do more and more research in children, it's becoming clearer and clearer that we overexpect, we overanticipate being informed by ongoing work in adults unless -- unless there's something very clear that happens. And that's just not the way research usually is. Both risks and benefits are usually more subtle although I have to say in this disorder, disease, it may be that the benefit hopefully is not subtle in this particular age group. So I think, you know, when you have really good preclinical large animal data to go further to ask more may not even be respecting physiology or growth and development, but I'm a pediatrician. That's my bias. So Dr. Weber? And we're going to open this up to everyone.

I think from my standpoint when they put in risks there the local risks cataracts, bleeding and infection and then the more systemic risks but they're choosing the worse eye in a disease that degenerates even if they have one of the bad outcomes they still have the vision and the quote unquote better eye and the worse eye will become the better eye later. So the function is still remaining for the individuals even in the worst of side effects in many ways and that changes for me at least the benefit analysis.

Dr. Federoff?

I was just going to try to address the -- the extrapolation of data that might accrue from understanding the outcome in adults and whether it's relevant to understanding the particular application here in these younger individuals given the progressive nature and the degenerative nature of the disease, it's very likely that what might be learned may not allow you to assess the extent to which there might be benefit and I would say perhaps even the risks associated. So if one relied or one thought one could rely on that as a way to guide this study, it may effect *ifzly preclude the effectiveness of this calculus that Dr. Powers was referring to and this is true for many degenerative diseases where the end stage may not inform you about the earlier stages where there might be an opportunity for remediation. So my sense having listened and read and I also commend the investigators in thoughtfully responding that there seems to be an overwhelming amount of preclinical data that support the application and I think the question is -- really is the age range, the correct one to assess this calculus and my estimation is from the discussion that this is something that seems quite reasonable.

I was going to encourage you to add a fourth point of discussion which has to do with scientific validity of the study which seems to me is particularly highlighted in the circumstance that we've all discussed so thoroughly this morning, the additional burden on pediatric research, and I'm not sure if this is the time to bring it up, but if you'll come back later, I'd be happy to express some concerns. I've been able to resolve my personal debate about risk benefit in favor of this being an appropriate intervention in an appropriate age group with regard to risk benefit. I'm persuaded by my own attempts to make extrapolations from older individuals and from other species which of course we do all the time, age and other species are just surrogates for what we think we know about biology and if we think that the biology extrapolates, then we're comfortable making those leaps. If we're uncomfortable with the biology being different, then we wouldn't extrapolate even from other pediatric experiences in -- into this circumstance. So I -- I find those reasonably persuasive. What I'm not totally satisfied with, though, is the scientific validity of the research design that's actually proposed. And let me just make a couple of points and then maybe those can be picked up by my colleagues if they seem to rez Nate. The stated purpose of the trial is to determine safety and toll rablt and the second is to look at clinical efficacy. If you look at the protocol, there is no discussion or definition of either the safety or tollerability in the outcome of this study. Instead what we have is a dose escalation. That's the basic design and then if you also look at the dose escalation as being somehow integral to the study, there's no discussion of how the dose escalation information will be used. So I think that the disconnect between the basic design of the study and the stated purposes raises a concern about whether this is really the -- the proper design. The denominator that's chosen for these cohort sizes of three individuals, I think, is although familiar to us and by now almost rote may not be appropriate given the goal of safety and tollerability. For example, we had a considerable appropriate debate about a rather extensive series of animal studies with regard to safety. Yet we're prepared to accept a much lower standard of evidence in terms of cohort size of 3 human subjects to believe that we have illustration of safety. And I -- I think that the fact that we are debating a large series of preclinical evidence, but would be persuaded by a minuscule experience of human evidence is -- is a disconnect. It's frob my not -- not correct. So I think at the very least, there needs to be a definition of what safety and toller ability means in the outcome section of the protocol and how the acquired data will be used to assess it with some properties that are reasonably reliable. For example, just to make a point about safety, if in a cohort of three individuals, no serious side effect or no serious adverse event were observed, there's still a reasonable probability that such an event could occur with a frequency as high as 70%. So in other words, a small denominator doesn't rule out some pretty clinically significant frequencies of events. Similarly, on the efficacy side, if you observe some dramatic evidence of efficacy in three out of three subjects on a given cohort, you could still have a lower bound on efficacy that was only 30%. So in other words, a pretty modest or even poor frequency as efficacy that might not justify pediatric patients could still give you an apparent success rate of 3 out of 3. So I think those are the kinds of issues that might be reflected in the protocol that investigators have thought about them and chosen a research design that would -- that would explain why they're doing what they're doing. Thank you.

Can I ask you to discuss how you would approach a very rare disease such as the one proposed in the context of what I agree with you on is clearer, more robust clinical trials design. In other words, if you have a disease where, you know, if there are 2,000 subjects in the country, which is I believe what I heard with this disorder and the investigators propose for scientific reasons and risk benefit reasons to limit this study to those between age 8 and 18, let's just -- let's just suppose that there are only 60 such subjects in the country of that age group and they're scattered all over the country and those that are on the east coast may be willing to engage in a clinical study. Let's say a third of them agree. We're down right almost at the number of subjects that the group is proposing to enroll. So how do we deal with this tension? The desire to have more scientifically robust phase 1 studies but in disorders that are fortunately limited to very few subjects? I don't know how to do it.

Well I'm not going to be able to give you a satisfactory answer to that here today. I mean, I'm very sensitive to the fact that there are some conditions that are extremely orphaned and where the logistics of doing the trial is quite difficult. But I'm not willing to make the leap from the fact that it's difficult to do a trial to say that we should be doing studies that may not provide us with reliable answers to the study questions. I and this commit tease and no one told the investigators what they had to investigate. They said it. They said they want safety and tollerability. And all I'm asking is that they produce a plan in the protocol that would convince friendly skeptics, friendly reviewers that in fact, that goal is achievable.

Got it. Thank you for helping me. Ms. Kwan?

I have some concerns considering the participant selection that I think could affect risks and benefits there. I -- I was -- I noticed that there was considerable weight put to the hazards of associated with general anesthesia to the extent that the investigators intend to include an anesthesiologist in with the interview. There was also mentioned by the investigator that he also was concerned about the psychological evaluation of potential participants. Yet I think that that dissolved to his personal insp*inth chul response to the potential participation as opposed to having, let's say, psychologists who would specialize in the family dynamics and my particular concern is of within the 8 to 18 age group, the group that is preadolescent and early teens, say from the ages of about 12 to 16, I've had quite a bit of experience working with that particular population, and if anything can be said about that, it's a highly diverse population psychologically. A lot of things are happening. It is also a time where family dynamics get to be blown out of proportion because of hormonal issues, issues of separation from parental support and guidance, depressions, just those early adolescent years would make, I think, some participants much more vulnerable to a catastrophic psychological result from a failure in outcome. And I would just suggest that some greater thought needs to be butt in play in terms of actually evaluating potential participant in the preteen and early teen years as well as that -- that participant's family support system and the dynamics within the family in the selection of a participant.

Any response from the group?

First, I'd like to speak about the safety issue. I guess as I stated in some respects we're looking at safety as a lack of toxicity and we've defined our dose limiting toxicities based on the world health organization toxicity scale for the systemic aspects of where we define abnormalities and we've adapted a ocular toxicity scale from a gene transfer study that is currently in phase II and actually have expanded on that to specifically look at dose limiting toxicities with regard to the eye, and the absence of those toxicities being, I guess, our unstated definition of safety and we have stopping rules when we have decided that we have exceeded that as well. But this is not just cosmetic.

And clearly, we would agree with that, though I would add that in a -- this phase I study which typically the number of individuals would be limited than expanded in further efficacy studies,. The second -- the -- from the invest sga torrs' standpoint, the estimate of 2,000 people in this country with Leber Congenital Amaurosis is again, from the investigators' standpoint optimistic. As of now, there are only 200 or so identified patients with this condition and this is just a projection of what may be derived by extensive screening. So we are really working with a very, very limited population to respond to the general anesthesia issue, I did have a slide up that specifies that we do consider indeed the major risk based on -- that is a risk of moralty for a ASA I or II individual in this population between 8 and 18 years old to presumably would not have cardiovascular disease, the risk is -- published is 1 in 50,000 but in our institution I believe there have been no reported deaths that have occurred.

My name is Dr. Katherine High and I'm the director of the center for molecular therapeutic which is acting as a sponsor for this study so I wanted to respond to one of Dr. Vile's questions. About immune response to the transgene product. And our -- my particular research interest has been for hemophilia B and as is the case with liver congenital amaurosis, there are both dog mode models for hemophilia B. We have done studies for dogs mice and humans and we can say that we have a complete data set and that we have found is that the mice and the dogs actually are a good guide for looking at immune response to the transgene product and that is a data set where I think that we can -- or that's a data point on which we can say with some confidence that the animal data does allow one to extrapolate to the humans. So in their case they have carefully sought antibodies to RPE65 and have not found them in mice or dogs and so I would -- I would feel some level of confidence that that is an issue that will not impact on safety in a human study and so to echoDr. Wara's comment, for me it's always difficult to judge risk benefit without talking about what we know in terms of the science, so I think your issue is one that we can probably place considerable weight on the animal data in terms of the response to the transgene comment. Oh, sorry. Go ahead.

I would -- I'm sorry. I would agree with that and I'm familiar with a lot of your work and so I -- I'm raising the issue mainly, I think, because this is just a different anatomical site which I personally don't understand what those possibilities are and I -- I certainly the ongoing dogs there appears to be no issue arising and I'm raising it just as a -- the data that seems to be lacking here and that's not a criticism. It's just that it hasn't gone out long enough is what happens several years down the line if you have got some transgene which is a nonself-protein which may be the case in some of these patients so I take absolutely your point and I think the investigators have done a great job in looking at nearly everything they can look at and I'm sort of -- as I say, at the risk of being hyper critical as sort of raising this as a possibility, that you know, may appear down the line.

Well, of course, as they mentioned they do have data that extends out five and a half year from the time of injection without -- without developing what -- what appear to be harmful immune responses to the transgene product in the dog, so that five years isn't 30 years, but --

And if that -- that's Lancelot, isn't it? And I assume and I hope that there's no plans to look his logically at Lancelot, if you see what I mean. But you know, these things could be developing. But I absolutely take your point.

Right. And then I wanted to also clarify one point about the two different sponsors, so -- so as you know, the sponsors are the only individuals who can really authorize the trading of data and you know, we -- we have certainly offered to share whatever information we develop in support of this trial with the sponsor of the adult trial, but it's really his prerogative to -- to give permission for us to do the same. And so we've requested that and would be interested in it, but you know, it's really not our prerogative to answer.

And I again appreciate that purely as a member of the RAC. I would strongly -- I would like somehow that the RAC could strongly recommend that in two such closely related trials there was intense communication and again, I take your point that probably from Q4 point of view there is and there would be and so on. But I -- I do find a conflict certainly in reading this that there are manuscripts which have offers from both sets of trials on them so there is some degree of data sharing and so on. There's a limited patient population which we've already been talking about and just -- just to try and ease that tension, it -- maybe that could be a recommendation. I don't know whether I'm out of place there, but, you know, but from your point of view I appreciate you making that clear.

Could I add one comment to that? One of the things you might want to consider is having a shared DSM-V at the very least. And that would seem to be a very simple thing to do.

Okay. Brilliant.

Although it's not clear to me -- I mean, it's not unclear to have a DSMV to have a -- it's not clear to me what -- I'd be interested in what the involvement of DSMB would be in this type of trial.

I don't know this context extremely well, buttive eve been involved in any number of true phase I oncology studies that have had safety monitoring committees.

And I -- I wanted to make one other point to the many excellent points that he raised and you -- you noted that you know, we have a great weight of dog data backing up this trial, but I think, you know, sometimes investigators continue to do studies in large animal models because they're possible and they can be done. And you know, so the total number of eyes -- of canine eyes injected may not be what was projected was needed to ensure safety but rather an accessible model where one could continue to gain information. I think we all know what the -- what the difficulties can be in assembling the resources needed for clinical investigation and so sometimes one continues to accumulate a great weight of preclinical data because it can be difficult to -- to move to the species where we all have the greatest ultimate interest. So I -- I think -- I hope you won't extrapolate too much from the weight of animal data as to what number we believe is required to assure safety.

Okay. Dr. High, thank you. Other comments from the group? Dr. Vile?

Just to belabor the point a bit about the risk benefit issues which were being discussed earlier, I certainly see from the -- the wealth of prez clinical data that you have the, the efficacy that you have, I think there is a potential benefit component here. In terms of the risks, I would -- the risk of moving from the over 18 to the younger 18 age group, is there any data on if you inject mice -- I'm sorry, dogs at different ages whether the inflammation associated with that injection is -- is more aggressive in the younger dog or the younger nonhuman primate because it seems to me the risk -- if there is a potential benefit which I think there is the the risk is likely to be associated with the local reactivity which you could argue you might expect to be higher in a younger immune system. Is there any way of testing that that would perhaps satisfy some of the concerns that are being voiced there?

It's an excellent idea. Due to the precious nature of these canine animals and the difficulty in breeding them, IE, they are only fertile twice a year and gestation takes several months et cetera, they need to be weaned, it's not possible to set up cohorts of different ages that cleanly with a large end in each cohort

Right.

In looking over the data from animals treated at different ages I have not seen any clear cut evidence of -- of increased inflammation with increased age of treatment, but I don't know that we can answer that at present.

Thank you.

I'm sorry to -- one more nagging little question about risk benefit since we're -- we're on it. I -- and I hope this has a simple answer. Why is the risk benefit appropriate for one eye, but not two eyes in a given subject?

There is a lot of redundancy in the visible function. In my responses I stated that if you, worst scenario, completely lost function in one eye, indeed, you do not lose 50% of visual function because central visual field -- visual acuity overlaps between the two eyes so by eliminating one, you don't eliminate an entirely separate individual organ function. There is some overlap so actually just by injecting one eye, it's less risk than in both. I don't know if I answered the question, but --

You probably answered it. I probably just don't understand its.

Okay. Could I -- could I just follow up on Dr. Viles question on the immunology? And so it seems to me that the real, one of the risks is a long-term immune reaction against expression of RPE65 and the data that have been provided in -- from the dog studies, so one of the questions I have is how inbred is that dog line? For the -- for the assessment of the immunology?

Well, the field of study of dog models disease is getting more and more like mice in that different strains are -- different breeds with different mutations are being crossed and they've done amazing jobs doing that sort of thing, but these dogs are still outbred and so typically what happens is they identify a model for study, they will obtain F 1s from what study and they will obtain those from breeding with small dogs that are easy to maintain in a colony and they have short hair. And so the ultimate idea is to get small short-haired dogs that are friendly and easy to deal with and so they're totally outbred.

Can you just expand then Lancelot is inequal to one five years out and his genetics may not -- you know, we don't know what his genetic makeup would be. So in total, how many dogs -- what's the end number there that you haven't seen any reaction in? And the time that you observe, five years observed the most?

We have not carried out more than one dog for five and a half years. Lancelot has gone the longest. And the other dogs have been considerably shorter periods of time than that. And the longest one is about 7 to 10 months. The longest set of dogs.

Thank you.

Other comments or questions from members of the RAC? Yes, Dr. Due Hurst?

I just had one thing that I -- I guess I wanted some information on and that's relating to the fact that you've very accurately described the disease as being a disease of children and not of adults and that the scarring and fusion of the adult retinas precludes performing a trial, but I guess what I don't have a feel for is among your older children and this sort of, you know, the 16 to 18 age range, how manying of those children, if any, might have a scarred or damaged retina that may also preclude potential benefit?

First I would say -- I wouldn't say preclude, because Dr. Jacobson and other investigators with some very sophisticated techniques ocular coherence comog ra fi have been able to identify in older individuals patches of retina which appear to have some viable cells which might potentially be amenable to a surgical treatment. And again, this is a continuum, so you start to see areas of degenerated retina times soo*e oh and it just continues in essentially a constant rate over time, so yes, you would expect that subjects between 16 and 18 would have considerable areas of degenerated fused retina which may not be amenable to treatment but again, less than a 35-year-old.

I guess where I'm going with it simply is whether or not that ought to, based on your judgment, I suppose, but ought to form some sort of potential exclusion for subjects in whom you felt in your best clinical judgment that the potential for benefit was so low that it didn't really offset the risk.

And point of fact, it is, because we also intend to use the coherence tomogram to decide if there is treatable retina and if the retina will be able to -- we will be able to derive information from that. We have a retinal degenerations expert working with us to determine those met tricks in fact.

But just to extend that a little bit, that's -- I don't recall that being a formal exclusionary component, formal cause for reason for exclusion.

I would put it under preexisting conditions that would prevent -- would prevent treatment. I think that --

Okay. Thank you.

I think that's in that.

Let's move to public comment. Comments from the public? I have to get ready to type.

I have two comments. One is a common comment about the informed consent document. I thought that in places some of the language was somewhat complex. As I said, that's a common comment that I have. Regarding the discussion of the risk benefit ratio that has been going on, I just wanted to point out that the regulations have -- there's a different balancing that goes on when you're looking at research involving children than when you're looking at research involving adults and particularly children -- when the research is approvable under 405 which you have provided some support for you know, prospect direct benefit and in this child subjects. And the difference is that when you're looking at adults, that we would look at subpart A and 111 says that the risk to subjects must be reasonable in relation to anticipated benefits, if any to the subjects and the importance of the knowledge that may be reasonably expected. On the other hand, the calculus is different under 405 which says that the risks are reasonable in relation to benefits to the subjects. So this would allow for research in adults that is greater than minimal risk but has absolutely no prospect of direct benefit to those adults and where as research that's approved under 405 would not allow that.

And that's why we chose with Dr. Paterson's leadership to spend the entire morning really here today talking about the risk benefit pension, if you will, when clinical research is done in children and youth. And I guess I -- the other thing I would want to point out is that I believe that because of exactly what you've just said, the standards for preclinical data are higher, substantially higher when research is going to be performed in children and youth. That's really what we've been, I think, struggling so hard with. Likely, if this were a study only in adifficults, Lancelot would not -- and the incredible results in the dog model would not be required. I mean, we wouldn't be talking about a dog having lived to older than five years as a -- as the primary example of why we expect benefit. I really do think the guide -- the guidance is -- is different for children. And should be, by the way. It's -- other comments from -- yes?

Well, it's been a real privilege to be here this morning and I love the level of conversation. It's so in-depth and I've certainly thought the principal investigators' presentations were stunning. I would change one thing in Dr. Maguire's slide where he says parental consent. It really should be parental permission. Nobody can give consent for somebody else. That's one little tiny point. The issue of deafness versus blindness has come up and actually there's a study or two, I can't quote you act exactly what they are but they found sighted people became blind. They found deaf people that became blind and various combinations and they've far preferred being blind than being deaf and I was very surprised to hear that. Even having had a blind uncle who I became very close to who became a multimillionaire and found his way around New York City on the subway system with no cane whatsoever so nobody would ever know he was blind and he left a huge amound of money to Ober land college and played an accordian and a sax phone. So there is hope for blind people. It's not completely hopeless and you wonder if it's because he was blind from birth if that drove him to wanting to accomplish in many other spheres of life. My comment now is about amblyopia and I'm looking at the 3 to 8-year-old group. So I'd be interested to hear from Dr. Maguire just what is it about amblyopia that's so threatening?

Again amblyopia is a loss of vision on the visual development laying down the development of the eye and the central nervous system. Amblyopia in the youngest age group tends to be the most profound and most difficult tree. There are permanent functions -- structural changes which occur in the optic nerve as a result. The under 8-year-old group, amblyopia is considerably less damaging and easier to treat than in the younger age group. There is a type of amblyopia specifically exnops ya where vision is deprived very early in life, there is no ambulance for any restoration, so in the -- in the older age group, 3 to under 8, we feel that while it's a factor, it's treatable and -- and not nearly as significant a risk as the younger group.

I think it's a really good point. I read this protocol as stating that the results from the age 8 to 18 cohort would rapidly inform conducting this study in the younger group. So I -- I may have overread an intent. Dr. High?

Well, I just wanted to add that another reason that we thought to do the initial study in 8 to 18-year-olds is because these are people who can provide assent where as the children in the 3 to 8-year-old group are not really candidates for that mostly.

Any other comments or discussion? Yes?

I know all you want to hear now after four hours is from another group of parents but bear with me and I'd also like to tell you as a citizen how much we appreciate --

Are you the Brents?

We're the Brents.

We will get there. It's right in the second sentence. Anyway, we do appreciate how much time you spend outside of your normal work doing -- doing work like this. I had no idea this went on, so thank you. Anyway, in spite of what Dr. Graves' study has shown, other studies have shown that people fear blindness second only to cancer. Eight years ago I knew little about blindness except the idea of living in darkness scared me. Then when our son Allen was born we jumped into the blindness community with both feet. My name's David Brent. This is my wife Betsy and we're parents of three children, the youngest of whom, Allen, is 8 years old and blind with only light perception. Betsy and I founded the foundation for retinal research in 1994, 7 months after Allen was diagnosed with LCA. It supports research in education for people with LCA uft our web site conferences and networks touches hundreds of families from around the world. I'm also an officer for the foundation fighting blindness. Cochairman for the science committee foundation. A board member of the Chicago lighthouse for people that are blind and visually impaired. Betsy is a member of the national association for parents of people with visual impairments. The associate board of the Chicago lighthouse and has certained on several committees on special needs education. When I say we jumped in, we really jumped. As parents our job is to protect and nourish our children to the best of our ability, to educate and prepare our children so they can meet the potential and live happy and enriching and independent lives. It's hard enough to do this with our two sighted children but measurably more difficult when it comes to Allen. Not only is it emotionally draining at times, in fact about 50% more marriages end in divorce when it involves aIt takes a tremendous amount of energy on the family and others. The reality is that the unemployment rate is that it's more than 70% for blind people. We don't want to draw tremendously sad picture. The truth is Allen is an absolute joy. He couldn't be here today because he insisted on going to school, just so you know. He has enriched our lives and the lives of all who meet him. He's a nationally recognized braille whiz. He plays the piano with expertise and passion. He makes us laugh with silly jokes but there are so in everyday things going on around him that he's not aware of and needs to be taught. The idea of providing Allen with functional sight approaching normal would be worthwhile on several fronts. Only 10% of people who are blind today were blind since birthed. The 10% who are congenitally blind face secondary challenges every day. To highlight why, let us walk you through some everyday challenges that have faced and continue to face Allen because of his lack of sight. We'vebone broken these down into into areas of physical, social and educational development.

This is where a lot of those benefits of vision and gaining vision come in. The first area is physical development. Physical development is often delayed and sometimes hindered because of congenital blindness. For example people learn to lift their heads and stand erect because of vision. Sighted babies first lift their heads while laying on their tummies so they can see an object in front of them. Without a visual cue, there's little or no reason to lift one's head. Even this must be taught behavior for a blind baby. If not taught, the person may goo through life with low muscle tone in his neck, severe back and pos chur rale issues as an adult and social amallties because of his head drooped all the time. This is why you see blind people looking down and interpret that as low self-esteem. we gain 80 to 90% of our needed stimuli through vision. As a result, many people who are blind have quote unquote blindisms. You may recall seeing people who are blind rocking, bobbing their heads, swaying back and forth. This is done to help give vestibular stimulation that our bodies so need. These behaviors are not socially acceptable and they must be compensated for through activity, motion or self-control. Again, this must all be taught for young children who are blind. And artificially recreated several times during the day. The secondary ya is social. Social interaction with other children in an age and an appropriate way is often very challenging. Most of the boys Allen's age want to play basketball and baseball. They want to play video games and e-mail their friends and go bike riding. These are not Allen's forte. Allen likes to play with his little talking electronic toys. He can tell you everything you want to know and a lot of things you don't want to know about the presidents, the vice presidents and their histories. He can tell you everything about the states, where they're located, what shapes they are, but for him, this is fun. For other people, kids his age especially, it's a little weird. During recess at school, most of the third graders run out to play ball, play four square. Allen runs directly to the swing, usually with the kindergarteners. He does that to get that vestibular stimulation that he needs. For some reason people with Leber Congenital Amaurosis like to rub and poke their eyes. It becomes almost an involuntary action, there's many speculations as to why they do it, and they do it so often they become unaware. Many times the result is sunken dark eyes cataracts and very social awkwardness. The third area is education. Educating a blind child is a whole other set of issues. Allen is in our neighborhood school thankfully. And he's in the age appropriate grade and doing very well, I might add. That said, Allen's educational team includes his third grade teacher an aids, a physical therapist, a mobility and orientation instructor. An occupational instructor, a social worker adaptive physical education teacher activities of daily living consultant. The school principal, us as part of the team and a service coordinator. He's the only student who is blind that our school district has ever had. You can only imagine the hours that go into transcribing his materials into braille, adapting each lesson so Allen can relate to what's being said just as his peers do. The other children have all been learning about our world through what they see in everyday life, in the movies, by what they see on television and in the museum exhibits but all those little details, the little tiny things have to all be described to Allen. Allen's world is very concrete. Abstractions are very difficult for him. For example, what image do you see in your mind when I say the sun is setting on the horizon or the dust is swirling around the legs of the stampeding herd of zebra? We easily form pictures in our minds and comprehend these words but for Allen, he has very little, if no point of reference for horizon, dust swirling or even a zebra. These barriers make reading comprehension very difficult and almost laborious. Another educational barrier for people who are blind is a lack of access and the bulkiness of braille materials. When we ordered the first Harry Potter book for Allen it arrived in two giant cartons. This one book is almost ten volumes long and each volume would stuff a regular sized briefcase. This is one of the reasons that the literacy rate for people who are blind is much lower than the national average.

Having said all this, we accept Allen and love him exactly as he is. My love for my son is unwaivering. But I'm his dad and will never stop trying to make his life as rich as possible. I want to show him the world literally. Yes, I would weigh the risks and the benefits just like you have been doing here, but that is something Betsy and I do every day as do every parent. Should we allow our older son to ride a motorized scooter or play ice hockey or walk to town with his friends and should we allow our daughter allowed to be able to attend coed parties, ride her bike across an intersection? Parenting decisions affect the parents of sighted and unsighted children every day. Eight years ago when we were told our son was blind the prospect of being a cure seemed to us like a man walking on the moon in the 1960s. This study is the first step on the way to the moon in curing blindsness. Then people with other forms of labors and eventually people with other forms of blindness. The risks that I've been described today are really very limited. The litmus text as Dr. Graves refer to it would be would we enroll our child in the study? And the answer for Betsy and me is yes. This human trial is the light at the end of a tunlt. The benefits to Allen and to society are enormous. Let's move Allen and ourselves and our society towards that light. I want the tell you two other things, please. First off, I know there's an issue regarding the sharing of information between the two studies. To the extent that we possibly can at the foundation fighting blindness we'll try to service a bridge between that since we fund both investigators. I can't promise anything, but on behalf of the foundation we'll try to facilitate that. And finally, I know this is a -- a phase I clinical safety trial and I nose that's the risk reward you're weighing, but understanding the problems of blindness for young children like we've just demonstrated is far more than just they can't see. They really have to be gotten to early so that we can have interventions for lots of things so they can be normal functioning adults. Thank you very much for your time.

Before we -- first of all, thank you both very much for all the work that -- that you've done and for coming here and sharing your very personal thoughts with us. We appreciate it. Rob and I just have one question. Were your concerns regarding the assent process satisfied?

I'm assuming that the IRB is going to take a good look at this as well. But I think they were heard and I think that the response back was in the right direction.

Thank you. Then I'd like to -- if there are no other -- perhaps there is. Are there any other public comments or comments from members of the RAC? All right. I would like to go ahead then and read what I've got and I'm very open to any and all modifications. General, central question for the conduct of this protocol is the analysis of risk benefit in the proposed pediatric subjects. The preclinical results both in dogs and in mice with LCA support the likelihood of direct benefit and even the lowest dose proposed in this phase I study. The risk is assessed at light sli greater than minimal risk based on information on the surgery/anesthesia. The vector distribution in the animal models and the potential risk of diminished function in the eye receiving gene transfer. The risk benefit balance is judged by members of the RAC is appropriate for the conduct of the proposed study in children and youth. Okay? Concern was raised by the hypothetical, but potential risk of chronic autoimmune activity in children and youth who eventually express the transgene. Concern was raised about the scientific validity of the proposed research design. The stated purpose of the trial is safety and tollerability, yet the protocol does not address their specific evaluation. The cohort size of 3 may not be appropriate for the goals of safety intoller ability. Perhaps that should be strengthened to is not appropriate?

I'm okay with may not.

Okay. Good. In a cohort of 3, if no adverse events were observed, the probability remains as great as 70% that such an event may occur in a larger group. Similarly, eventual efficacy of 30% may be overestimated by this study design. Other, there is compelling evidence that the disease is pediatric in nature. The RAC remains concerned that the ongoing study in adults with shared research members of the current group but different sponsors is not accessible. Tw two such closely related trials one in adults and the second in children both would benefit from intense collaboration. A vehicle for collaboration would include a shared DSMV. Would you like to modty the wording there? All right. I were formed consent. It is likely that study participants will be -- and Terry I need your help -- will be diverse psychologically and subject selection would benefit from a formal psychological evaluation to include the assessment of potential vulnerability if failure occurs. And finally, the RAC suggests that the informed consent document be carefully reviewed to decrease the complexity of language. Is that what you meant? All right. Then if there are no recommended changes, I would like a motion to accept

So moved.

Dr. Weber, a second then let's vote beginning with Dr. Rosenburg.

All say yoo*i AYE.

Thank you all very much and I -- we all appreciate the depth of the discussion this morning. We're going to break for lunch. It is not 12:15. It is 1:15. And I apologize. Any lapses in time management are my responsibility. We will reconvene at 2:00. It's going to take 45 minutes to get lunch. Thank you again.

We'll reconvene in just one moment. I apologize for the delay.

Let's go ahead and reconvene the group. I apoll joy for the delay. The first protocol this afternoon and the second protocol of the day for public discussion is protocol 732, a phase I 2A dose escalation trial of intratumoral injection with oncolytic adenovirus vector. In patients with advanced solid tumors. Dr. Nemunaitis, who was to have been here as the PI to discuss the study is in china. He had a gene transfer meeting and Dr. Bill wold will discuss the preclinical aspects of the study followed by Neil who will discuss the clinical trial design. Dr. Wold?

Good afternoon. So we are proposing a phase I dose escalation trial of intratumoral injection of our oncolytic adenovirus vector INGN 007 or double 07 as I will refer to it in my talk. In patients with advanced solid tumors and the PI of the study is John Nemunaitis. I'm Bill wold and I will be given the preclinical presentation and I'll be followed by Neil and the sponsor is Introgen therapeutics. So in my presentation, I want to focus on four major points. First, that 007, which expresses a protein named ADP has greater oncolytic activity than other replicating adenovirus vectors. In particular, the conditionally replicating adenovirus oncolytic factors that have been developed in many labs and some of which have been examined in clinical trials. ADP is an adenovirus coded protein that functions in the infection to cause efficient lysis of infected cells at the culmination of infection allowing release of the virus from the cells so the virus can go on and infect other cells. The rational for incorporating ADP over expression into our vectors is that in the tumor, the -- the infection process will be speeded up, the cells will lice more efficiently than would be the case and the vie virus would be released from the tumor cells so it can infect other cells and spread through the tumor. Our second point is that we've developed a -- an animal model, the Syrian hamster to evaluate our vector and we believe that this is an appropriate permissive immune no competent safety studies with 007. The third point is that the preclinical data that we've obtained in the hamster model suggests that 007 will not be more permissible than wild at 5 and that was one of the questions posed by the RAC. And the final point is that 007 is well tolerated in our hamster model at the no observable adverse effect level at the highest dose that we propose in our clinical trial. So adenovirus is a DNA virus with the genome enclosed within a protein capsid. Our vector is based upon serotype 5 which is a ewe bigtous virus which causes a mild upper respiratory infection in infants leading to life long immunity. Add 5 does not cause significant disease in healthy adults. Also importantly is there's a long history of administration of live adenovirus vaccines to millions of military recruits to vaccinate against acute respiratory disease and I am not aware of any adverse aefkts from administration of those live vaccines so I think both of these points argue that a replication competent vector such as 007 will in fact, be safe in humans. So this slide shows genomic schematics at the top and for the 007 vector at the bottom. I won't have time to go through the slide. For 007 I'll point out two major features of the vector. First is that there's a deletion in the E 3Gs which are present in the wild type virus. We consider this to be a safety feature of the vector because the E 3 protein's function to protect infected cells against destruction from the host immune system and so we believe that with our vector, because the vector lacks these genes that it will be more susceptible to elimination from the patient by the patient's immune system. The second feature of 007 is that 007 overexpresses ADP at late stages of infection from this alternatively spliced MRNA. So this slide illustrates the consequences of ADP over expression by adenovirus. And what we're doing is comparing infection of a human lung cancer cell line named A 549 by 007 and by wild type ad 5 and we're using different multiplicities of infection ranging from 10 plaque forming unitss per cell or infectious units per cell down to the 10 minus 4. And what you can see with 007 is that at a multiplicity of 10 to the minus 2, the model layer is destroyed. In contrast with wild type ad 5 the monolayer is in tact. And in fact we don't see an amount of cell killing with ad 5 until we use 100 or even 1,000 fold more virus. Now, those of you who are familiar with oncolytic adenovirus vectors will appreciate that the conditionally replicating vectors are mostly, if not entirely, attenuated as compared to wild type ad 5. The genetic mutations in those vectors attenuate those viruses compared to wild type ad 5. So I think you can appreciate that those vectors will be more attenuated when compared to 007. We've done this kind of experiment in about 20 different human cancer cell lines of different indications. And without exception, 007 is more effective in destroying those cells than is well tap ad 5. So we think 007 will be a broadly applicable anticancer vector. So I've shown you that 007 replicates well in cancer cells. This slides shows that the vector replicates less well in normal human primary bronchial epithelia cells. This is a single step growth experiment where we compare replication of 007 is in A 549 lung cancer cells and in the primary cell line. And what you can see is that there's 100 fold greater yield per cell or birth size, this being a log scale in the cancer cell line than in the primary bronchial epiPeete yal cell line. This slide illustrates the ability of 007 to suppress human tumors, in this case, human lung cancer tumors growing at signe graphs in nude mice. And what you see is that 007 is very effective in expressing the growth of those tumors as compared to the bufferered injected control tumors. Now, we've had extensive discussions with the FDA regarding 007. And they advised us to develop an animal model to evaluate the vector and they indicated that the animal model should be both permissive for the virus, that is the normal tissues of the model should be permissive and also the animals should have an intact immune system. And so we've explored two animal models. One being the cotton rat and we published a paper earlier this year on our cotton rat studies and the second model was the Syrian hamster and we've just had a paper accepted for publication in cancer research on the hamster model. The key features of our model are that 007 replicates in hamster cancer cell lines nearly as well as in human cancer cell lines. Secondly, that the hamster lungs and liver are permissive for 007 and these are the relevant organs and thirdly, of course, that the hamster has an intact immune system. So this illustrates the ability of 007 to replicate in a hamster cancer cell line named HAK shown here in red in this single step growth experiment as compared to replication in the human lung cancer cell line A 549 and A 549 cells are the most permissive human cell line that I'm aware of. And the replication is excellent in the cancer cell line, only 7 fold less than in the human cell line. If we were to do this experiment in mouse cells the yield of virus would be somewhere between 1,000 or perhaps 100 fold less than in the human cell line. So here we're examining the ability of 007 to suppress subcutaneous tumors formed by that HAK cell line in hamsters following infratumoral injection. And we see that it's quite effective at suppressing the growth of those tumors as compared to the buffer injected tumors. Now turning to replication of this vector in the normal tissues of the hamster, here we're addressing whether the virus replicates in the lung, so 10 to the 7th plaque forming unitss were administered by -- this is ad 5 I'm sorry instead of 007. They were administered by intratracheal administration and then the virus was extracted from the lungs at different days post infection and then tightered in subculture on A 549 cells. So what we see is our four log increase in titer in the hamster lung and the virus levels remained elevated even after nine days. A similar study was conducted by a group back in the 80s and they obtained identical results to the results that we obtained. This slide demonstrates that 007 and ad 5 replicate in the hamster liver. This was part of a biodistribution study that was initiated based upon advice by the FDA. In the study, 1.9 times 10 to the 12th VP per kilogram of vector was administered. That's 1.9 times 10 to the 11th virus particles or about 10 to the 10th PFU. To virus us were injected intravain nowsly and then different tissues were harvested and then titered for virus using a TC ID 50 assay. So what we saw at the two-daytime point replication of both 007 shown in the red crosses and well type ad 5 roughly 10 to the 7th PFU per liver were observed. At the 7-daytime point it dropped by couple of logs and then by days 29 and 92 we were no longer to detect infectious virus. In contrast -- shown in the black crosses we were never able to detect infectious virus. So what the experiment indicates is that both ad 5 and 007 replicate in the hamster liver and also that they replicate to a similar extent. And I'll now turn to our toxicology study that was conducted again in consultation with the FDA. The study had five animals per sex per group. It was a single dose intraveinous injection study. A low dose 3 times 10 to the 9th, a mid dose and this is equivalent to the maximum dose that we will be proposing in our human trial. And then a high dose, 1.9 times 10 to the 12th VP per keg which is 60 fold higher than the maximum human dose in our clinical trial. The control groups were well type ad 5 and a replication defective adenovirus, both administered at the high dose that was the same high dose as with 007, and also vehicle was a control. And the animals were sacrificed at days 2, 7 and 29, and we examined typical clinical signs, clinical chemistry, hematology and histopathology. I don't have time to go into the details of the study so I'll just present the summary. What we found is that for 007, the no observed adverse effect level was 3 times 10 to the 10th VP per keg. And this is the maximum dose proposed in our clinical trial. We found that there were no adverse affects in the liver and no effects at all in the lung or in other organs. The second point is that we found that 007 and ad 5 have virtually identical effects and that the adenovirus was less toxic at the high dose at which the viruses were tested. So in overall summary of our preclinical results, we have found that 007 is effective in suppressing human signe graphs in nude mice and cancer tumors in hamsters. We believe that the hamster is an appropriate model to test the safety in pharmacokinetics of 007 and our preclinical data suggests that 007 should not be more infectious or transmissible than well type ad 5. So in this slide, what we're doing is presenting our proposed clinical trial and our starting dose will be 2.8 times -- or 2 times 10 to the 8th VP or 3 times 10 to the 6th VP per keg. This is 10,000 fold lower than the no adverse effect level that we saw in our hamster model. Then there will be a gradual step by step dose es skalation to a final dose of 2 times 10 to the 10th VP and this is equivalent to the no observed adverse effect in our hamster model. I want to emphasize that in the clinical trial the vector will be given by intratumoral injection and no doubt the tumor will absorb much of the virus. On the other hand in our hamster study the vie would you say was administered intraveinously and in those conditions it should be more effective to infect peripheralish tissues. I'll now pass the laser pointer.

Thank you very much.

Good afternoon. I'm the subinvestigator for the clinical part of this study. As Dr. Wold stated and based upon our communications with RAC, I want to emphasize that this is indeed a phase I dose escalation trial of the intratumoral injection with the adenovector 007 in patients with advanced solid tumors. The primary objectives of this trial are to evaluate the safety of the intratumoral injection into accessible lesions and to determine the maximum tolerated dose with the maximum dose given in this trial as reviewed by Dr. Wall being 2 times 10 to the 12th viral particles. Secondary objectives will include assessment of the pharmacokinetics of 007, which will encompass measuring viral particles in tumor, blood and urine and cytopathic effects of assays. We will be evaluating and documenting antitumor responses. Additional evaluations on this trial will include a overview of 00 pharmacokinetics and we will be looking at tumor on day 8 and day 28 to evaluate adenovieshl proteins as well as pathology as well as the evaluation of the immune response as well as neutralizing antibodies, cytokinds and surface of round cell infiltrates and cells in response. The entry criteria for this trial confirmed carcinoma with accessful superficially accessible lesions in patients who have failed standard therapy. The performance status will be 70% or greater in patients who are 18 years of age or older with a life expectancy of three months. We will require adequate laboratory demonstration of essentially within normal range or close to normal range organ and hematologic function. The patients will be required to be negative for HIV and hepatitis and will not be entered if there's a history of a viral syndrome within two weeks of registration. Patients cannot be on current imdown knee suppressives and no malignancies and of course, the IRB approved consent will be required. The study design includes five cohorts containing five patients per cohort. There will be a single introtumoral dose starting as Dr. Wall noted, four large lower than the preclinical no observed adverse effect level and if you'll allow me a seasonal acronym I'll just use NOEL for the rest of this with IV administration. And we will cautiously dose escalate in one log increments from 2 times 10 to the 8th viral particles to the maximum dose that will be given in this trial if we do not reach any toxicity prior to that of two times 10 to the 12th there will be a two-week interval from the last treated patient in any given cohort before we will initiate the next cohort with a higher dose. Now, in each cohort of five patients, three of the five patients will be evaluated for safety and efficacy. These patients will have one lesion injected and they will be followed clinically. For the other two patients in the cohort, in addition to safety evaluation, we will also assess pharmacodynamics. These patients will require at least two lesions, one of the lesions will be resected prior to treatment and then the second lesion in each patient injected, in one patient that lesion will be resected on day 8 and in the second patient, that lesion will be resected on day 28. The dose limiting toxicities that we have defined for this protocol will be any grade for toxicity and a grade 3 hematologic toxicity lasting greater than three days. Grade 2 neurotoxicity and a flu syndrome lasting ten days beyond treatment. If a dose limiting toxicity is observed, three additional patients will be enrolled at that dose. There will be at least a two-week interval between the enrollment of each additional patient and if a second DLT is observed then the prior dose level, the dose level lower than the DLT dose will define the maximum tolerated dose for future investigation. Now, as Dr. Wold had indicated, we feel in response to questions that were raised by the RAC that pulmonary side effects with this agent are unexpected and unlikely. Our feeling is based upon the absence of pulmonary toxicity and permissive animal model. That adenoviral respiratory infections have been well tolerated in the normal and cancer population, and again as brought up previously, the safe use of the intercoded wild type adenoviral in millions of military recruits with studies done at that time showing no evidence of collateral con tap nation or transmissability. However, should any of our patients develop respiratory symptoms, it is important to be noted that we will be following them closely. They will be maintained in the clinic for at least six hours on the day of injection. They will be seen at 2 days and 4 days following. They will then be seen weekly. They are given instructions regarding our expected reaction to a viral vector. The short term flu like reaction that we do see and specifically instructed to contact us if any symptom occurs out of keeping with the information they have been given. If indeed any of the patients develop some systematic pneumonitis and there is evidence of a chest x-ray infiltrate we will have those patients undergo bronchokos copi, whichever is appropriate. Now if we find a nonviral etiology for these symptoms they will be treated appropriately. However, if we cannot find the cause, then 00 identification will be performed and until we assess the etiology of that respiratory symptomatology no patients will be entered on trial. If we document the presence of 007 in the aspirated material or the patient clinically develops a grade 3 pneumonitis, then we will immediately initiate antiviral therapy with one of the antiviral agents primarily with sigh to prevary. So in summary, our starting intratumoral dose in this trial is four logs lower than the intraveinous NOEL dose in hamsters. And the proposed clinical trial and dose escalation schedule is supported by these -- the preclinical data as Dr. Wold presented in a per miss v animal model and we thank you for your attention.

Thank you both very much. You may either sit there or come up -- both be up at the podium, whichever is more comfortable for you. All right. Let's begin, then, with a discussion from the formal reviewers. Dr. Albeda.

I'd like to thank you both for a very clear presentation. I thought this was an important protocol discussed for a couple of reasons and you've very nicely gone into some of those. First of all, it's one of the first protocols if not the first protocol that's not using some sort of selection to make it a tumor selective virus. And I had a question. You didn't really go into this in much detail but maybe you could make a couple of comments about it. Why do you think this virus is selective? You showed some data with a 549 and epithelial cells with 007. Do you see the same thing with wild type virus?

The answer is yes. The adenovirus in general has a greater propensity to replicate in cancer cells than it does in normal cells. I've shown you the experiment with 007, we would get the same results with ad 5. When we compare the -- we have compared the replication of 007 and ad 5 in normal cells growing that are ak we es ent and we find that viruses don't affect those cells at all -- I'm sorry. They do replicate in the cells but not very well. They do replicate more efficiently in proliferating cells, but there is a -- the reason, I believe, that adenoviruses replicate better in cancer cells than in normal cells is because cancer cells have a deregulated cell cycle and normal cells don't and the virus needs to drive the -- deregulate the cell cycle, push the cells into S phase in order to replicate efficiently.

The second reason I wanted to discuss this was an issue that's I think bedeviled the field and many of us that have been interested and that is the lack of a good animal model. And frankly, it was a little surprising that the initial trials were allowed without the animal model that you really need, and so I think you're to be congratulated for really, in my mind, being the first group that's developed an animal model that I think is a realistic animal model where you can show replication in the lungs and liver so that it means something.

Thank you.

All the preclinical studies in the mice, it doesn't replicate in the normal mouse tissue I don't think it means very much. So I think it's important for the field and future investigators and I think you'll be getting plenty of phone calls for your tumor cell line in addition to opening up a lot of research that was never able to be done. I think it's very important but until a good model was developed couldn't be studied. And again, I think you're to be congratulated for that and I think the data that you showed convinced me that it does look like it's a good preclinical model to look for safety. Soy wanted that presented here for future trials and for the field and I think it's going to be a great advance. I had a couple of questions on the clinical side. The first is, in our experience with adenovirus, this is intraplural administration we usually see our side effects about 12 hours after the installation of virus and that's when the patients will get the fevers, and the hypertension. You're letting your people go out for 6 hours and I was wondering if you think that's long enough, maybe the side effect profile would be more quick but I was a little concerned that you're doing this as outpatients and only observing them for six hours.

We have based the timing in large part on our own past experience using other viral vectors. We've done a lot of work with the onyx 015 vector which I think bears repeating that is in addition of course to the E B 155 K deletion. It is an E 3 deleted virus with the mrens of ADP interestingly enough and those patients that we treated were treated not only intratumorally, we have treated patients via a direct injection of colorectal carcinoma and liver. We have treated patients at two times 10 to the 13th viral particles as a single dose given intravain nowsly and 2 times 10 to the 12th viral particles along with chemo therapy patients. And with recombinant IL human IL 2 in patients and what we noted was a very typical and reproducible pattern IL 6 levels tended to peak not out of proportion to what we had seen lower intratumoral trials at approximately 4 to 6 hours. The symptom time range was not exaggerated in these patients. We occasionally saw some certainly some fevers and perhaps some muscle rigor starting at about 6 to 8 hours and certainly as you've already indicated, we have seen additional flu-like symptoms beyond that to about 12 to 18 hours. In our experience, the patients since they are being seen 24 hours later, we have not really had a situation. We're beyond that six hours and being seen 24 hours later. So the patient is essentially in all cases staying close by to the hospital and does have very specific instructions. So based upon our experience with patients attentive to our instructions and our own experience with vectors given in much higher doses where even though one might make the argument that the onyx 015 is conditionally replicating, we were able at those high doses to demonstrate replication in our patients. Four of the patients, the titer on day 4 was actually one log higher than the titer 4 hours later. Certainly very much supporting replication in those patients. So based upon that, we felt comfortable following the guidelines we have been following with our other viral vector studies.

Sounds reasonable. It's certainly much easier doing an outpatient study than having to admit the patients. Just one last question which you may want to think about when you write your protocol. What happens if in a Gwynn cohort you don't get two patients where you can get two biopsies from? Is that going to stop you at that cohort or have you written it so that you're trying to do that but not qualified to do that.

We may have changed that and Al those the original protocol that you had the chance to review, on review for a number of reasons including what I felt was not particularly an appropriate request to make up a patient undergo three biopsies one prior to treatment rngs one on day 8 and yet another one on day 28, the fact that if indeed we followed those initial guidelines, those patients would then have had to have had a split dose with a proportionate amount, proportionate to either the size of the volume given in each of the two injections and we have data suggesting a dose response relationship at least locally, I felt that that would be one inappropriate from the patient's point of view and would not address one of the questions we have, which is what is the pharmacodynamic pattern at the dose that we indeed say we are doing given in this given Tuesday more? We have done other studies in other patients where we have used a similar approach not using the same patients but using superficial lesions we've done this with two different vectors one bacteria, and the other adenoviral vector in which as we followed the sequentially we had very reproducible amongst patients and informative data so I think this will be appropriate for the patient, enable us to get the information we want and therefore we will not be doing multiple biopsies in the same patient and not having to find a patient with 3 to 5 lesions.

I think from practical standpoint it's going to be much easier to do your patients.

I agree.

Dr. Albeda, are you comfortable with the information presented on the efficacy -- potential efficacy should they be needed? You'd written that as a concern.

Yes, there was was a detailed response -- written response and although there's not very good clinical data there is some data at the case report level that supports the use of this. And I doubt it's going to be really needed based on what your data is, but it does look like it does have some efficacy.

And then one final thing. You'd commented on the consent form and asked that mention of liver toxicity, shock like syndrome or even death be added. You comfortable with the proposed modifications?

Well, I personally believe that it would be best to include some of these other complications as unlikely such as liver toxicity. I think you've convinced me the lung toxicity would be very unlikely but I -- you know, based on the past known problems, what few there have been with adenovirus it's been liver toxicity, it's my own suggestion and the investigators can do as they want that they include that as an unlikely complication.

Comment from the investigators?

We have elected to do so based upon your recommendation.

Thank you. It's part of our job here is to report as public record the written concerns that each of us as RAC members have had and then whether those written concerns have been satisfied, if you will, by the discussion, so that's the only reason I -- these -- your two final concerns needed to be read into the record. Thank you.

I'm satisfied.

Great. Dr. Due Hurst?

Yeah, I also wanted to underscore Dr. Albeda's comments about the model system. It's nice to see a real model system in which you can get some predictive information about replicating adenovirus vectors so I also wanted to express my feelings about that. We had actually very similar comments in many similar aspects so I'm just going to read my comments into the record. The first one was really just asking for information which the investigators have provided in terms of the relative invie troe replication and spread of wild type ver Sirrus 007. Whether there was potential shedding and this convinces me that there's no real concern in that regard. I had a second concern. Again, really relating to the relative replication of wild type versus 007 with respect to some data in response to appendix M. That is also a satisfactory response. A question about replication of 007 within lung tissue and the possibility of the pathology at the lung site and again, we've seen, I think, compelling evidence that that's an unlikely outcome. Like Dr. Albeda, I had a question about the ability to use cytoprevar or cytovie reason and there's -- in addition to Dr. Albeda that is certainly convincing that were this necessary, which is very unlikely, that it would be likely to be effective. I had a question about az tempts with respect to virus isolation from subjects who might develop respiratory symptoms or symptoms that might be suggestive of virus shedding or virus replication or pathogenesis and I think we heard pretty clearly that you have a well delineated plan that would involve identification of 007 were it present in being shed in association with symptomatic respiratory illness so that seems appropriate also. And finally like Albeda and Dr. Powers I had a comment with respect to the consent form which has also been addressed. Soiz don't have any other remaining concerns.

Thank you. Dr. Powers?

My only comment on the view of biomedical ethics is the question whether to include in the consent form the -- the risk of death from adenovirus vector. And I suggested to either include it or indicate why you think you don't need to include it. You did both. I -- I was almost convinced that maybe you were right not to include it, so I'm --

Best to ere on the side of caution.

Well, that's for sure, but I'm satisfied if you wish to include it but I'm less convinced that it's necessary.

Thank you. Thank you all. We'll open this up for further discussion by members of the RAC.

I'd like to just go back to a question that Dr. Albeda raised first for Dr. Wold. Is -- am I understanding this correctly that the reason the virus shows somewhat preferential replication is not necessarily due to the ADP protein that's being expressed but rather other factors such as cell cycle? So is it, in fact, the case that expression of the extra ADP protein from this vector is actually increasing the yield of virus per cell? And how do you measure that? Are you actually prethawing the cells or are you just collecting virus after infection?

The -- the overexpression of ADP by -- by 007 does not increase the yield of virus per cell. What it does is speed up the time at which the cells splice. If you do an infection in a -- say tissue culture cells and then you do a cell lysis assay at different days post infection, with wild type ad 5, for example the lysis assay you'll find that the cells don't begin to lice until about five days post infection. Where as with 007 the cells will begin to -- so it speeds up -- as you know normal replication cycle, the DNA starts to replicate at about 12 hours and if you titer the virus produced it normally peaks at about 2 days and I showed a growth curve in my presentation. But it will round off, and it will sit there with the virus in the ceils and that's what ADP does. It increases the speed at which those cells lice and it does that because ADP gets synthesized earlier. But if you titer the total amount of virus that you get out of the cells it's the same. We had thought when we first made the vector that we might lose virus. The cells might lice too early in the cycle but it turns out that that's not the case.

The one last question I would have would you care to comment just briefly on the fact that this vector lacks the E 3 region and so that may have some impact on the immune response in the lung? Do you have any general comments on the use of that kind of vector?

Well, we believe -- as I mentioned, that the lack of the E 3 region is a safety feature of the virus. There are a number of theoretical arguments one could make of why the region evolved in the adenovirus. And most of the animal viruses have the region too. And many of the E 3 genes that are in ad 5 are conserved in those animal models and we know quite a bit about most of those genes, as you know and they function as a group to prevent killing of infected cells by CTL primarily

Pause to change captioners. ey showed in that study if large amounts of E 3 deletion units were administered directly into the lung, under those conditions, they saw increased pathology as compared to the wild type virus. We don't believe that that scenario will ever apply in that study because of the dose of yours in the beginsburg study and because of your administration in the human trial. And also our conjecture is supported by our data in the hamster model where we administered 10th of a 10 PU in the hamster, two logs higher than in the beginsburg study. Nevertheless we didn't see any toxicology at all in the hams sister lung.

You talked about antibody responses -- you talked about monitoring antibody responses in participants, are you going to look at T cell responses to adenovirus as well?

Yes. I mentioned.

Doctor.

Could I just clarify? The ADP in response to the second to the last question, that doesn't increase the efficacy of the virus. So when you showed the picture, the culture picture where you used different MOIs compared add to ADP. If you let that go longer, would the wild type add been as good as the other?

The answer is yes. If there is -- in an experiment like that, if there is one intechs particliered a particle dish, there would be a series of infections. That will spread through the culture dish.

It's not showing that 007 is more Poteauant. It's just more Poteauant.

If we do an experiment such as the one I showed in the nude mouse model. If we compare a virus that expresses ADP with one that does not then that is much more effective at suppressing the growth of the tumors.

Did you show that?

I did not.

That's published in a paper in 2000.

Hams hamster cell line does that grow in hamsters?

Yes. And then we injected vector directly into the tumors and the vector suppressed the growth of the tumors as compared to the growth.

Thank you. I just wanted to make a couple of points that don't necessarily rise to the level of recommendations Bush I thought it would be important to clear it up. You have fairly good evidence it's a safe final dose. Is that fair to say?

Yes.

And so you -- I guess that would lead to the fairly strong expectation that you won't see any adverse effects during --

We believe so.

-- during the whole clinical trial. That's the way I interpreted your evidence. If that's the case, then the idea that somehow you're determining a maximum dose is not in effect correct. You're determining an operationally defined dose that's probably going to be the last design point or the highest dose of the he is calculation, is that fair to say?

We believe that we will have the maximum administered dose.

If you see something, then that will become. That would lead to the idea that maybe thinking about this in terms of a maximum tolerated dose is not the best way. But if you could define operationally or clinically or from a safety point of view an optimal dose. And optimal could incorporate feasibility. Sometimes it's not possible to administer another higher log worth of viral particles, but the point is if you were to define what an optimal dose is then you could reexamine your dose range to make sure you had in fact bracketed that optimal dose. And you might well wind up with exactly the same experimental design that you have here, but you could also have shifted your dose range one or two logs in either direction to that?

With the protocol designed as it is, you're correct. It's more likely to be a backlogically effective dose. Evidence of tumor effect with the injections locally. Part of looking for the ranges as you've said bracketed a range of doses how would we use this subsequent to the study assuming we proceed with evaluation. Given, of course, as we all know the problems we have currently with targeting and delivering intravenously given vectors, we would be looking at regimens, looking at this virus with other forms of treatment, chemotherapy, radiation therapy, so that having as the goal of this particular study demonstrating a dose that indeed in the clinic is biologically effective and allows us to combine with other agents in the fiewch thatre you are that could -- once we have defined additional safety with concurrent therapies as opposed to continuing to do this on a moan owe modeel level.

That makes perfect sense. I guess I'm just arguing not to disseminate the idea that this is tolerated because you don't want to effect people's thinking that they could increase this if you wanted to put this with something like radio therapy, it could be that a lower dose of radio therapy and a much higher dose of virus was the right combination.

Exactly. Those studies would be more appropriately done in the combined mow talent mode and then having to back off again, it will not give us additional information for proceeding with the way that this agent will be used.

Thank you very much.

You're very welcome.

Doctor, does your comment rise to the level of something we should mention?

I think not. I mean, I really just wanted to make the investigators aware. It sounds like they're pretty well tuned in. I'm okay with that.

Other comments?

Yes. Just out of curiousty, you both mentioned the vaccine that was given the military recruits why was that developed.

There's something about young men of that age under the stress associated with boot camp perhaps, they become susceptible to informations in the U.S. and sear fight 14.

That was not an issue with add five then?

No, not an issue with add five.

Comments from the public?

Of course. Going back to the doctor's questions, so given that you now have a -- the hamser model, have you compared in an intrafume recall study?

Yes, they are about the same.

Statistically the same.

So overexpressing KDP doesn't buy you much, would that be correct to say? In the cell culture, definitely you see the connection of it earlier.

In one experience we did the two gave the same results. As I mentioned, in the zone graph model, the 007 ADP overexpressing viruses are better than ADP minus viruses or ADP normal viruses.

Then does that point to the fact that -- does that differentiate the competency of the immune response and the cancer?

It's quite possible that we could be receive -- possibly virus elimination based upon immune killing of 007. The greater immune killing of 007 versus add five. We're proposing experiments to address those kinds of issues.

My question is also if the data for the safety aspect is this army injection that had wild type identifiers, and you would support that manipulating the virus to overexpress a protein doesn't give you a benefit in the tumors, then why not just use the add five?

Well, because we think 007 will be safer. At the be safer because it lacks the --

We think that the 007 virus will be safer than the other. It will be cleared more easily by the immune system. Animal data indicates to us that it's not more toxic.

Or the efficacy will be better because --

Experiments to -- other questions. Comments?

I just had a quick comment about the N.ed consent document, I think that a lot of the language in it is a little too complex. Any other comments? Well, it's just a good thing you made your comment because it's the only one we have left. In my time on the rack and as certainly as the chair, I've never had an occasion to read this kind of summary. The gem the development of an applicable animal model is an important contribution to the field. Then I was going to add a second sentence which would have read the protocol team has satisfactorily answered all the -- we must include, and I agree with you, that the informed consent document should be reviewed for complexity of language. I think then we should say though that other concerns set forth by members have been satisfactorily answered because this truly is a very different sort of hour and a half and protocol review than we're used to. And I thank you both, and your team, very much for providing this kind of discussion for us. Are there any comments on this terse? Yes.

After that first statement, I would just offer the discussion of continuing to exploit the tumor model in this virus to really get at the underlying mechanisms of differences. .

We certainly plan to do that. Thank you. Can you state again what you just said? I started typing. I've written that we recommend that the investigators continue to exploit the tumor models to define?

The underlying mechanisms that result in the differential tumor cell killing by wild type versus the 007.

Okay. Let me read what I've written again then for everyone the development of an applicable animal model for safety studies replicating adenovirus specter is an important contribution for the field. We recommend that the investigators continue to exploit the tumor model to define the underlying mechanisms that result in the tumor cell kill by wild type versus 007. The IC document should be reviewed for complexity of language, otherwise the protocol team has satisfactorily answered all of the concerns set forth by members of the rack. May I have a motion for approval? .

I move.

Dr. Has lip a second? Dr. Ail bedda. We will take a vote beginning with Dr. Weber.

Dr. Pant dosesy. I've almost got it. Dr. Fetter off.

Yie.

And Dr. M.. Thank you all very much. We're going to take a break. My watch says almost 3:20. Take a 15 minute break and be back at 25 till 4:00. Thank you very much. We're going to reconvene. I'd like everyone to take a big breath and get their energy level back up for the next hour and a half. Big breath. And before we -- or drink three cups of coffee. Before we begin, I want to remind everyone that we are queening tomorrow morning at 8:30. It's very important to be on time so that we can systemically work through the protocols. For our reviewers and presenters to be able to catch their flights in the afternoon. Our final discussion for today concerns protocol 734, an open label phase one dose he is calculation study after ad CDR 33 and surgery for patients with resusceptible recurrent high grade glee roamma. And Dr. Braoddus is going to present. .

Shift some pocket contents here. Thank you very much. Good afternoon. I am here, I'm Bill Braoddus as the doctor introduced. And I am accompanied today by these who drove all the way from richmond Virginia and happy to be here. We are presenting a protocol targeted at Geo blast tomorrowma which has its roots based in lab inventory and preclinical work cor carried out by the doctor and Dr. Valerie. Their work represents over 25 peer review publications that began by focussing on the role of the open determineel factor receptor including brain tumors and glee roamma's particularly. the protocol we're reviewing today was presented in the summer of 2004 leading to much of the features of its design presently. As many of you probably know, Dr. Schmidt died probably one year ago. Those of you who knew him and would be a considerable presence in the room if he were still alive.

Dr. Ted Chung was recruited. Since then Dr. Chung has launched VCU's first clinical trial using the add know viral vector -- Dr. Chung is coPI with me for the present proposal. The proposal is focussed on attacking geeio blast tomorrowma by taking advantage of the unique biology, the growth reaccept -- many of you probably know it is the most common primary brain tumor in adults. In spite of considerable advancements in surgery, radiation therapy, chemotherapy strategies, we still only manage to keep people alive for a little over a year, 12 months or so. The most comprehensive lent mow lek particular signature is application of EGFR and it is found in greater than 40%. In many study up to two thirds demonstrate that. In a considerable number there are actual mutations causing add vacation of that system. Our lab inventories have shown and many other lab inventory tores have now shown that inhi basing leads to sensitivity in these tumors inzerovo and animal studies. The installing network is represented by this cartoon that shows the receptor as a trans membrane protein with an extra cellular component. Binding of a applyingen, EGF or other applyingens to the receptor, depending on which one we're talking about allows for them to form die inners and die innerization of the normal receptor leads to the deposit fortunatelylation and that allows interactions with a cascade of other components in various important path ways and the results are increased proceed live rages as well as increased secureal in the face of radiation or chemotherapy attacks on the cell. In glee Joe blast tomorrowma particularly two mechanisms have significant as I mentioned earlier, one is called an auto inactivation where both the receptor and applyingens that can activate it are released or produced by it themselves result in activation of the normal receptor and in a significant fraction of these cancers, there is a mutated form. In fact, the extra celller domain does exist on these, does not need to -- these form and remain highly active. And the cell is overactivated through this mechanism. Now, one of the interesting phenomenonna demonstrated a long time ago in the laboratory and I think was the genesis of attacking this mechanism this these cancers was the fact that radiation of cells itself DOS causes -- it causes this it's very rapid within second and can result in proceed live rages of cells. So the very treatment that we try to attack these tumors promotes tumor growth at the same time. And this is known as accelerated repopulation, a term coined by Dr. Schmidt and really has one of the important features or was one of them that led them to think this was a key strategy or key mechanism to attack in going after these tumors. Another important finding was a paper published in 1991 by Dr. Ulrich's elaborate from Jeromeny, different reacher not related, where they noted lacking terminal 533 meanno acids leaving this one 647 meanno acids was enough to allow the extra cellular domain to bind EGF and allow them to -- rain cored it in the member brain thear

Actually allowed this moll dual to serve as a decoy or dummy and inactivate, prevent the activation of the normal growth factor receptor moll duals in the member brain itself. That interaction is shown here on the far right in this cartoon and is shown by this idea that there is an extra cellular domain, but the inextra cellular domain is missing. The purpose of this slide is to present this in relation to other strategies of attacking the EGF receptor system for the purpose of therapeutic activity, ant bodies are another technique that had been used. One of the problems with that is that this moll dual needs to be on the surface of the cell in order to be accessible to the ant body. These particularly that active mutant. Small moll dual inhibitors actually have great promise for attacking this. And there are quite a few that are in development and in some cases in trials. And it is a valid strategy. There are some problems with specificity of the inhi television that render them less successful. One of the things I'd like to point out in a minute or two, the dominant EGFR technique -- that the EGF receptor is in so that it actually has a broader target within the cell than the EGF receptor alone. Again, I've here I've scheme advertised that activation of the receptor that occurs not by applyingen binding but by some other effect of radiation on the cell to cause that to lead to both set of protective responses as well as other responses. This is one of the pieces of data from a study from the lab showing the efficacy of the CD 533 dominant negative mutant in the growth of brain tumors when implanted in the flanks of nude mice. And I guess I'll walk through this because it's somewhat interesting that the normal growth of the cells is represented here when a control virus is used to ineffect the cells prior to implantation. A more rapid growth of the tumors is seen when that active mutant EGFRV 3 is used to trans secretary a cell. This is reCPA pit lating that. Mutant form. Then you see that both the tumor cells own EGF receptor is inhibited by the CD 533 and those that have been transected and activated even greater can be overcome with this. This was very, very exciting findings at the time and was certainly a big part of our interest in proceeding with a clinical trial. This actually goes further and looks at the change in growth of these tumors, not only with the effect of the dominant/negative mutant itself but showing that those withme more effective itself. We envisioned beyond this protocol within fiewch protocols a trans techs with subsequent radiation treatments. I mentioned earlier that the CD 533 interacts with multiple other proteins of the same family. And these are the herb B family members, herb B one is another name also called her one, her two, is also known as herb B 3 and herb B 4 are the remaining members of the family. All of these have significant activity, although herb B one still remains the strongest in terms of promoting both proceed live rages as well as cell difference itself to that. There is one family member that is not interacted with CD 533 that's B 3. In cells that can represent an escape mechanism of the cells to avoid the dominating effect of the CD 533. I wanted to move on briefly to the discussion of some of the issues with respect to the protocol that we've proposed. One was regarding the actual tumor -- the volatile doses to be administered to patients. And there are two important studies that have now been published on results of add know viral therapies, gene transfer therapies for tumors. One is the add know viral P 53 study reported by Lang. They used increasing doses of virus up to three tens of particle units. Even at that highest dose, they did not find any dose fox isty. They were unable to define maximum tolerated dose in that study. They have used a virus which is actually rep occasion confident, it is capable of replicating in certain cells, primarily in the P 53 path way which would represent tumor cells and not normal cellings in a brain tumor setting and does not have any therapeutic gene other than that activity of causing the cells to produce other virus and die. They reached a maximum of 10 to the 10th platforming units. If we used a conversion of between 20 to 50 viral particle units per platforming units, that represents a dose of 2 to the 5 times 2 to -- no fox citiesty was noted in that study. For that reason, we propose to use five different overall doses of virus which we're now referring to as particle units. There was some ambiguity in your submission regarding this. The doses were the same, just the terminology was not used properly. So particle units we're going from 10 to 11th to the 13th units in half log units. The mechanism of administering the virus in this study is also unique to our protocol. And that will involve a continuous control rate direct infusion. The distinction is that in the Lang study, the virus was delivered by hand injection into the tumor bed over the course of ten minutes. And we are proposing to deliver the doses of virus into the tumors of the subjects for the study over the course of between four and 40 hours. And when we do that slow controlled-rate infusion, what happens is that the material actually passes through the tumor if a rankma actually in the extra cellular space, the extra cellular fraction of the tumor. If you think about that, that means that half of the space and it's a little more is extra cellular, pushing in ten milliliters of viral suspension will eventually fill 20 milliliters of total volume. Now, this was done -- this is a rat study which much smaller volumes of infusion. In effect, the largest volume we are going to be infusing represents an equivalence of about here on this graph. That means that we're expecting approximately twofold filling of brain volume for the infusion of virus. Some of the issues to consider when delivering virus -- and these are things that have been learned both from reports of other viral protocols, but also protocols that we are participating in where we are infusing other than virus for similar patients -- is that we want to avoid a sump effect where the possibility of material is escaping into a cafeteria that has fluid in it, either a tumor cafeteria or the sub arachnoid space around the brain. Those are things we want to avoid. One of the things that's been learned about the brain is if the catheter point is a centimeter away from one of these surfaces, there is effectively no escape of the fluid into that fluid space. I mentioned the CSF is containing space already. In fact, there is one other -- this one centimeter distance in terms of avoiding the escape into ventricle just as I mentioned a moment ago. Another is that the material can flow back along the catheter as it sits in the brain. And that's a known and predictable fennomenon really. We are using a distance of at least two centimeters from where the catheter weters the brain surface to where the infusion point sits is enough to avoid reflux of material back to the surface. That would then allow us to avoid escape of the viral infusion into the cerebrospinal fluid. I just thought there might be some interest in the technology and techniques that we use to do that. This is actually a couple of photographs from a needle biopsy of a brain tumor that we carried out. This is not actually -- it was not a case in which we were replacing a catheter for infusion. But you can still see the biopsy needle. In effect, there's a little LED meter on the tip of this needle. That's why the wire comes off to power that LED. There are also localizers on this arc which are fixed to the patient's skull after they're asleep. This little device here provides the information to the computer graphics system to actually calculate the trajectory of the needle going into the head so that that trajectory is then plotted on reconstructions of the MRI imaging data, this is a reconstruction of the patient's scalp surface. And that's actually showing that this needle is actually dead on track. These why these circles are all concentric, a little like being a fighter pilot but a lot slower and less stressful I suppose. We can advance the needle all the way into the target. In this case a biopsy was taken. In other cases we use a similar technique to advance catheters. This is a patient that's participating in a clinical trial for a protein foxen for attacking these recurrent blast tomorrowmas. This is a set of reconstructions of what are called sagittal images. This is the eye socket. You can see the catheter starting here, passing down to its target point here. We would call this length of tract about 12 to 15 centimeters much more than adequate. Said more than two centimeters is adequate to avoid back flow along the catheter. This is a view of savings accountel it will -- because of the two centimeters of distance beyond the ventricle and one centimeter next to the ventricle, this is pretty at what time to avoid the -- into the ventricles. Studies that have been done in preparation for this protocol are still in the preliminary stage. We have completed preliminary studies and definitive studies are pending. The studies that have been done included mouse muscle injections of the adenovirus CD 533 construct. And the results here were local inflammation of the highest dosage given. Three out of the seven animals fox isty had he pat I can was noted in the highest dose animals. This should be studied in our current proposal, first, the infusion was in the muscle where it's well-known that virus can easily activate the -- something that would be quite different with infusions into brain and brain tumor. And then secondly, the -- and the second point is that the amount of virus injected was many fold greater than the proposed greatest dose of virus that we are going to be using in the protocol. We've done rat brain infusions as well. Again, some instances of fox isty were noted. But there was no relationship to the dose of virus given. And we suspect this is more related to the preparation of nonlaboratory grade virus. The careful definitive studies are planned and of course they will be completed prior to beginning any accrual of patients into the trial. And negotiations are currently under way since this is a fairly costly aspect of protocol preparation. So in summary, the preclinical data supports the proposal for using this viral vector and eventually radiation therapy in an adjuvant technique against glee Joe blast tomorrowmas. But certainly more work on safety will need to be done prior to testing if the efficacy in combination with radiation treatment. Some of the changes that have been made that we've delineated since the rack submission is to prepare abstract -- we've corrected that typographical error that virtually every one of our reviewers picked up on and thank you for bringing that to your attention. We did change the dose to be a regular half log increment as substituted. We've made a variety -- made revisions in the protocol itself to address some of the deficiencies raised by those questions and then have also made multiple revisions of the consent form based on the comments of the reviewers of this committee. I think that's it. I thank you touch for your attention.

We're going to go through a process of reviewing the original questions put forth by rack members, the three assigned members reading them into the record and determining whether they've been satisfactorily addressed or not. If you'd like to -- since you're the single presenter, if you'd like to stay there, I think it's probably easier if you could do that. Thank you. Dr. Federoff, if you'd like to begin.

Sure.

Thank you.

I'd like to thank Dr. Braoddus for both responding and detail to a variety of different questions possessed by myself and others. I will make this as much to the point. I'm going to divide my remarks into those that are biological relevance and then protocol. The first is the concept of trans dominant inhi television of a receptor is intriguing and it's one of the reasons we selected the protocol for a public review. And the question I had possessed to you, it was really intended to be two parts, one was whether there was knowledge of that capacity of that moll dual to trans dominantly inhibit others that are not members of the herb B family. And second in cells that depend upon herb B installing that are not tumors whether the trans dominant inhi television is something you have any concerns about with regard to the safety issues.

Right. The answer to the first question is we don't know. That we have not studied the ability of the CD 533 mutant to interfere with other receptor activities. I would go on to say that since that moll dual, the CD 533 moll dual is identical to the extra cellular and trans membrane component of the EGF receptor, another way of asking that question is does herb B one itself react with others. There's no support for that, that I'm aware of. In my literature search for such, it certainly is a very interesting question. From a scientific viewpoint, it represents an ideal tool for asking that point.

With regard to the second part of the question in cells that are nonneoplastic that may depend upon this receptor or others in the family, is there a level of concern regarding the enter diction, the trans dominance of that path way as it might affect their survival?

I think that is a valid point. Certainly there would be situations in development where the effect could be quite Poteauant in the epithelial elements. In the setting that we are proposing, there is no known dependants on it for normal cell function.

And in fact, I would add that that's the purpose of our fox logic studies is to try and get at that question, sort of at the working end of the way that you would try to answer that is how does it effect the normal brain.

The only thing that I'm aware of is that lien age stem cells reflect -- to the extent that that might be a consideration, it's just worth while noting.

Absolutely.

Moving on. One of the things that is intriguing about this approach is that you can target cells where you believe that the either mutated or other indodge news receptors are playing a role to characterize their knee plastic potential. I asked whether it was known or you might anticipate that you need levels of this moll dual to produce an effect. I know that's a difficult answer but I think it's vital to understand the anticipated fiewch efficacy of this. The experiment that you presented both in the protocol as well as in the presentation or when you've coexpressed an adenovirus that has the truncated EGFR along with the other. I think the more relevant question is how many moll duals are present in tumors and whether there's an expectation that you can drive levels of this to sufficient levels where you can have an impact on those tumors.

That, again, is a difficult question to address. I can say that the original work with the dominant/negative mutant actually was done in cell system where they were able to control both the level of the wild type as well as the CD 533. That demonstrates a level. It was adequate to completely -- work has been done in vee throw in Dr. Valerie's lab on the amount of tran section required to interfere with this system and it has been well within the levels that we thought were possible to deliver. And then finally, this work and subsequent work shows that direct delivery of the viral vector into a growing tumor is adequate to stop it from growing and respond in this fashion. So we've got sort of -- we've got functional impure I can results to suggest that we can be successful with delivering adequate viral vector -- I'm sorry adequate trans gene to impair the growth of these cells and impair their response. The actual details of that would be very interesting to continue to pursue. It's the sort of thing that continues in laboratories as we're talking I had asked about the fox isty you clarified it. In the intramuscular studies there were a number of animals that developed had he pat I can -- that was not observed when you --

Sure.

You addressed that to my satisfaction.

Thank you. I had also raised a question about the catheter placement. You addressed it both in the response to my question as well as in the presentation about the potential for seepage, the reflux issue. It relates both to the depth the catheter tip is placed. I assume it also relates to the design of the catheter.

Yes. On the pressure induction side there are now numerous examples using con convention enhanced delivery. I was going to ask you, how is it distinct from CED or similar.

It's quite similar. And the reason that we don't use the term is we are now talking about delivering part chat agents. The assumptions on delivery have to do with the assumption that sol autos developed within the infuse ate are going to be carried by bulk flow. So bulk flow would occur and in fact the volumes would for a given volume of infusion are smaller than we would expect because of the size difference of the particles.

One of the other issues I just ask clarification of was when following admission to the GCRC --

Can I ask you to stop for a minute. Are you comfortable then with the catheter as well as plans for cath they are placement?

Yes, I am.

Thank you.

My apologies for not making that clear. When after admission the patient would undergo re -- it was to my satisfaction that this would occur on protocol day four where protocol day zero was the admission.

Correct. Coming to the end here. This is an area where I'm going to ask several additional questions which really related to the preoperative planning in the section following the installation. In that, I suppose that the volume metric analysis of these tumors is going to be partly approached in your preoperative planning and that you have a protocol where you're going to advance the volume in it's lowest dose and then bureau going to go and use that higher dose which is 10 CCs. Given the volume that you anticipate using, which will say is 10 CCs for the moment, what I was curious about was whether that approach is independent of your estimation preoperatively of the tumor volume itself, and how do you think about that with regard to individualizing this therapy to make it both safe, tolerated and maybe in the future, if it progresses, he have easyus.

The doses that will be used will be used independent of the tumor volumes of the individuals that enter the study. The alternative would require a much more extensive knowledge of the behavior of viral distribution than anyone has. And even in the studies that we are participating in that are multiclinical phase two trials, there is no means of monitoring or monitoring the distribution relative to the tumor size. And there is no change in the volume delivered based on tumor size. So that's the first sort of element of my answer. The second is that -- and I provide a bit of a discussion in my response that we provided to the committee -- what we find in the neurowrote oncology clinic is that our patients are followed very closely so recurrences are generally caught when they're not large, shall we say, and certainly not inappropriately large for this sort of approach. In fact, we can see a recurrence as mall small as one centimeter. We anticipate entering patients with tumor recurrences of the order of one and a half to three centimeters in diameter. If a tumor is perfectly spherecall equals a little over 14 milliliters total volume. As I said earlier during the presentation, we anticipate filling a volume of around 20 milliliters with your maximum infusion of 10 milliliters.

So in actual fact the range of tumor volumes that we will see in your patients will be commensurate with the volume of that maximum dose that we are proposing. We felt it was important to establish the safety of a delivery of a viral vector by this technical neck since it lass not been done in humans, in terms of the larger volume of infusion. And that's why we proposed the initial part of the phase one trial, which is to go from one milliliter to three milliliters from three to ten as progressive steps. We anticipate that the ten milliliters will be well -- will be tolerated with acceptable safety. And then we can get on that with question of what dose of virus in ten milliliters is acceptable. I can say that toxicity problems or maximum tolerated -- I'm sorry, dose limited toxicity with the protein foxen studies have found that volume is not as critical as the concentration of the therapeutic agent in that volume, so that's why we were anticipating that we would get to the ten milliliter volume and then proceed from there. We have designed it as such that if ten million liters appears to be unacceptable.

Thank you I take your points. I had asked about the location of the catheter, whether it would be localized and you provided a detailed response to that that I find quite complete. The last thing that I actually wanted to ask you about, which you highlighted in your presentation relates to the nature of the inflammatory responses that you saw in the rat studies. And if you would tell us a little bit more about the characterization of that inflammatory response and the progression of that inflammatory response with regard to the normal brain tissue in the -- whether there were any functional outcomes associated with those inflammatory responses.

Very good. I take that backwards. There was no functional toxicity of the study. The nature of the inflammatory changes were for the most part inflammatory cell infiltrates in the region localized to the catheter itself. There was -- I would say that the study was not large enough to really characterize the time course well because the toxicity was noted at both short and longer time points. And there was, I think, one instance of what appeared to be an abscess. And that would have to relate to technique as opposed to the agent and delivery process itself. .

I've gone through the small grouping of questions. I'm satisfied with the answer to this last one. I still think, as I've identified, there are some unknowns that came up in the Q and A here relating to stoic onlytry, relating to potential effects on other herb B family members and their survival function but I'll turn the microphone over to another reviewer.

Before we leave this because it is a little -- I want to be sure I capture everything, the major concern that you are left with is that if the mechanism of trans dominant inhi television is present to other moll duals that are not members of the herb family, this may present additional and unknown risk. Similarly if it is Poteauant against nonmalignant cells it would be an additional risk. In order to assess this, further data is necessary. Do you have any thoughts about how you would proceed to gather this additional preclinical data? .

Well, I can comment on the latter first. It would seem to me -- and it was the intent in the question I asked you about the brain related toxicities to carry out the study for a longer period. If there were herb B studies -- patterns of expression are delineated and published on, then you should be able to ask the question whether the delivery of this reagent effects the cells. With regard to the former, I think there are a variety of ways to do this and you've already begun to outline some of them when it gets to the question of are there other moll duals this can react with. I find that to be reasonably reassuring but is not definitively. One could approach that by a variety of stitch culture -- colocalize in some manner by cross linking or by direct precipitation of this moll dual with other receptors. That would give me a higher level of confidence that would there be any other installs for signals -- did that answer your question?

I wanted to put something into this beyond just the caveat of doing more work.

Those were the two.

Doctor.

I found it also an interesting application of a new trans gene. And I have some of the same questions that came up with C. Federoff, but also some additional ones. The first comment was that there was some confusion about the actual amount of virus that was going to be administered, particles versus plaque forming units. And the investigate fater has addressed that. But I would still like to ask, given the fact that it appear that the highest dose that will be used in this study is about three fold higher than the highest dose that had been used in patients and given some toxicity that has been seen any preclinical trials and the potential risk of getting the virus into the CSF, although low, is still a real risk. How firm are those doses that -- particularly the highest dose. Are these written in stone or are they up for further consideration at the completion of further toxicity studies that you indicated?

Well, I can assure you that they're not written in stone. In fact, we feel that one of the most important tasks ahead of us is to do the definitive toxicology studies. And that is going to provide information on the toxicity of this agent, of the variety vector, and would be a time at which we would consider altring our plans. At one time in designing this protocol, we were concerned about being criticized for not using enough virus because none of the studies have shown dose limiting toxicity even if the maximum dose is used. That is partly why we have included the highest dose. This represents the highest dose that would be feasible to administer because of the consideration with preparation of the virus and so forth. And if there were suggestions of greater toxicity with this viral vector, then the other -- the previous vectors that have been used and certainly we would be adjusting downward the proposed doses.

I guess I could skip down to my fourth one, which is related. And that is that, as you mentioned, there's been relativelily limited toxicity studies to do and you're one thing that would be helpful again is to make sure that those are reported as particles in these studies, but also it might be helpful to look at a standard E one deleted virus because you are seeing toxicity. One of the things that would be raised. -- in the two hours of driving from richmond, we covered that, so, yes, we fully agree. So going back to the -- my second point, which was the same as Dr. Federoff's really, was there any ability of this dominant/negative to any others and you've addressed that pretty quickly. One thing you mentioned that herb B 3 is not regulated by the same mechanism. Is there any thought that cells, tumor cells that might overexpress -- or if B 3 and what would be the consequence of that in terms of escape. ?

I can tell you that herb B three is not known to be a major factor. The potential of tumor cell escape for the strategy that we are using is certainly a real possibility as is does also apply to any other antitumor strategy out there. The part of this problem that makes it so malignant is that it is actually agree netically unstable so that when it comes back in the same patient it's different in terms of its biology. It's part of the problem we have to tackle. We believe this has a -- we hope to couple it with other strategies as well.

Perhaps a simple expect that could be done would be just to look at growth rates of primary cells and trans duesed with this gene to get a feel for whether there is some fundamental or consequences of that expressed dominant/negative. One of the other issues is I noticed the construct -- although it's depicted to -- actually, I guess contains 20 meanno acids?

That's correct.

What do you think the consequences of that would be in? Is there any evidence that this particular residue or series of residues would be a sub straight for anything? Have you looked for any of this?

Well, the fact that the -- 20 ameanno acid contains this could mean -- almost certainly not the auto deposit forlation of the site blast milk component of the growth factor. About over 300 mono acids distal to where the ameanno acid sits in the pep tile chain. That does not answer the question whether they could surface sub straights for others. We do not know. We have not seen overt evidence of that. There are specific ways that we can and will look to address that question. .

Okay and one of the last science questions was there was some difference in the amount of the trans gene expression in the scene graph tumor model versus the tissue culture model. And I asked for some clarification about that. I think the response was more than adequate in terms of describing differences that others have observed in tumors in veevo versus cell culture, so that's fine. The last thing I just had was that in the consent form, the first sentence in the last paragraph on page four was unclear. And I guess that's been reworded?

Yes, it's been revised. Thank you very much for that.

So I'm satisfied with the responses that the -- that have been made and revisions. .

Okay, so Dr. Them Earl, I was left with two items and I suspect there are more. The first one is to compare the toxicity of the EGFR vector with that of a standard E one deleted has been vector?

Yes.

The second one was that tumor cell escape wouldd be CB growth rates in the model.

Well, I think those are two different things that kind of got combined. The potential of the dominant/negative to promote escape by regulating herb B three would be one thing to look at. The other would be in a tumor cell background.

Cue repeat that for me?

Whether overexpression of the add encoded trans gene has the effect of upregulating herb B three in the tumor cell glee roamma, that might lead to escape. That should be -- that could be examined. And then the other issue is whether the trans sheen itself can act in a normal background by tweaking the normal cell processes. And one of those might be disregulating normal cell growth. .

And I guess I'm confused. How does that differ from what Dr. Federoff mentioned earlier?

I think it's an extension of what we talked about earlier.

Right.

The doctor has proposed to be cell culture experimentings. Part of what I had proposed was to take that insleevevo and ask whether there were cell types in a normal brain as well as the receptor, whether they will undergo some altered survival or altered function. .

Go on to Ms. Kwan then. I'm certain I don't have this correct, but you all can help me as I move forward.

My comments were largely waysed on trying to understand the protocol as a layperson. And one of my initial suggestions was that, unlike most protocols, this particular submission did not have a noun scientific abstract. And since that comment a lay abstract has been prepared which does help greatly in the understanding of this by nonscientific member of the rack. I would just take a moment here to kind of remind everyone that one of the reasons originally for the creation of racks and IBCs was to law the lay public better insight into some science that, at the time 30 years ago, was sometimes considered scary by communities and surrounding neighborhoods. I think while over time redominant DNA research has become less scary to the public, we are now at a time where neighborhoods and communities surrounding research labs against feel some degree of suspicion and concern about the research that's being done. And so anything that can allow better insights by the general public and calm people and help them understand what is going on would be beneficial to the entire research community. So, I think taking the time to try to describe things that are highly complicated, such as this particular protocol, is well worth it, not only for the researchers of a particular investigation, but for research science in general. So I would encourage that that continue to has been. There's also a commonly accepted idea that when you try to explain yourself to somebody who has no clue of what you're talking about, that really helps clarify in your own mind what's going on. So that would be another reason. As I said, the investigators have now prepared a lay abstract isat does explain the protocol in much simpler terms and helps people understand it. I also had a number of comments regarding the informed consent document in particular. This is common to a number of protocols that come before us for a phase one trial or early investigation, to try to reduce the term of therapy and treatment and medication. And in fact, a lot of that has been done in the informed consent document has been reworked substantially. I will leave it to Dr.Sory because things described as drugs, medication and therapy that may kind of induce more optimistic outcomes than might be expected. But, nevertheless, the entire informed consent document really was rewritten in many ways that I think makes it clearer and also less optimistic, which is appropriate for a phase one trial.

Is there further work to be done on the document?

I would guess that in the ordinary review process, the solution, the IRB will pick these things up. I don't think it's necessary to harp on anything further on that. There was another item that I did not understand only reading the informed consent document. And that was that it seemed to indicate that the biopsy had to confirm that it was a recurrent tumor in order for the study agent to be delivered. So it almost seemed to me, well, why would you -- does the placement of the catheter and then nondelivery because the biopsy didn't confirm the thesis, would that pose an additional risk? In fact the conformed consent document was rewritten to reflect the catheter would not be placed until the biopsy was confirmed through the section that there was indeed a recurrence of the tumor. That was satisfactory. I think that was probably in the methods and the protocols in the -- that was just not clear in the informed consent document. Then the other comments really were picked up. I was just reflecting the fact that some of the initial scientific members of the rack's comments were indicating some confusion with regard to the dose he is calculation.

I am satisfied.

You are satisfied. Thank you. Let's open this study up to a general discussion by members of the rack. .

You weren't here this morning but we had an interesting discussion about risk benefit. My questionings relate to that and I would like to hear your answers. I'm going to play the devil's advocate, if I might. I would consider this in my fairly high risk protocol in that you are subjecting patients to an additional surgery, placement of the catheter, the risk of fluids going in, there shallbeen some deaths after brain transfer and seizures. On the other hand because you're going to resect the entire area where the catheter is placed, EBITDA see -- very hard for me to see how that would have any potential benefit for the patient. Point number two, as a cancer team therapist, the number one word in my mind is bistander. I would like you to discuss what sort of by stander -- even if it's 50% of the cells with no bistander effects, as I think everyone learned I doubt you're going to get a tremendous amount of advocacy. Now, on the other hand because you're doing the resection, I think you will learn something very important about does your technique work and is it safe? So I'd like you to take these comments and put them together and justify to me whether you think the risk benefit is appropriate.

Very good. Well, -- they are facing a survival of less than six months. I think that's the important context to begin with. The other important point is that these patients are not being offered surgery for the purpose of the study. They are being offered the study, if they are eligible and need surgery for a recurrent tumor so that the risks that are specific to the study relate to placement of the catheter, infusion of the Victor infuse year-to-dated itself in this particular study. And I think those risks are hard to quantify, but I think they've been carefully discussed so I won't try to review that again. The question of benefit is also very important. I think it actually was suggested by my discussion earlier relating to the dose that we've chosen and the tumor size that we anticipate so that for the average tumor -- the average-size tumor in the patients that enter the study, we will very likely fill most of the tumor and extend some of the infusion of the agent into the brain just jaysant to the tumor. You're probably aware, I'm not sure -- they don't grow simply as a lump of tissue, but they have many cells that have left the main lump of tumor and are travelling, invading beyond the margins of the tumor that we can see on the scan. It's by virtue of the fact that we can't take out a bunch of brain around these tumors that they virtually always come back. There is some potential benefit with respect to the fact that these tumors will be, as I said, largely filled with tumor and the ajaysant brain where there are remaining cells of the invading cells will also be subject to the treatment effect. We hope to do much better than the efficiency that you alluded to because of the distribution of the viral effecter by this slow, continuous infusion. We are not in a position to prove that, but your preclinical data will argue that we do have a reasonable chance of better delivery to the tumor cells. I'm trying to remember if I -- by stander by stander. I guess my answer there is that that protocol is not designed to take advantage of a bistander effect. And in fact is designed to obtain optimal tran section by improving the delivery of the virus to the tumor cell. And therefore, in fact, the purpose of the study is to determine whether we do have -- we are able to attain the level of tran section that we want. We believe we can. It remains to be seen whether that's true or not. I guess the final comment with regard to the risk benefit is that the there's a small potential benefit but it's not enough to argue with respect to benefit to this trial. This is a phase one trial. The purpose is actually focussed on safety. We do believe that if we can establish safety, we can then go on to utilize this strategy to provide benefit as well. How could that be? Well, some patients might not require surgery if the delivery of the virus is successful. So therer potential benefits not necessarily for this particular trial.

I would note that you changed a sentence, I guess at Terry's suggestion about it's very unlikely we receive benefit from it it's possible. I think that's an appropriate thing to do. There are many patients that enter into a clinical trial with a motive of gaining knowledge for others. If you do get patients that are willing to volume fear for this -- although I think your answer about getting gene beyond the surgical margins is a good one. It's going to be mostly in that kind of feel. That needs to reflect that. These people are going to be undergoing an extra surgical procedure and an extra risk, with that, I think the gain of knowledge being the primary gain.

Other comments?

Thank you. I'd like to pick up on the doctor's perspective but take it in a different direction towards study design. There were four things that caught my eye in the protocol, not entirely in a favorable way. One is in the introduction where it says, quote, knows statistically valid conclusions can be drawn. The second is in the sample size discussion where it says up to 40 subjects could be enrolled. You might expect 36 plus some additional ones at the high dose. The other item that caught my eye is something that wasn't there, section nine, dealing with statistical considerations had a title was completely blank. And then fourth you did, on the front page of the protocol, have a statistical collaborator. If you put those things together and put them in front of a statistician, you're going to have a war on your hands. The notion you could study up to 40 subjects without any study of the reliability -- is not state of the art for this kind of research. For example, there are three NIH sponsored ones in the United States that does this type of study. There's one based in Baltimore. That bred and butter is doing early developmental studies. I think they've done the virus before, at least one of them. And then other small moll duals, buy logical therapies and so on. So if you're going to margin rise the risk benefit ratio, which we've already heard, I think that it's obligatory to reflect on what quantitative and general risible knowledge will come out of the study. That will lead me to the recommendation that I hope the committee will endorse that is to reengage your collaborator to review the design of this study and revise that is consistent with the way these trials are being done by your colleagues in the other places in the United States. That would include a very careful consideration of the objectives and what generalizable knowledge and how relibel that knowledge will be based on the objectives and the design chosen for the study. I mean that as a comment. If you want to take it as a question, that's fine. That's the end of the thought. Thank you.

I'm just writing. Just a minute. .