8:05 a.m.

Tuesday November 13, 2001
















Conference Room

5630 Fishers Lane

Food and Drug Administration

Rockville, Maryland 20857



JOHN DOULL, M.D., PH.D., Chair

Professor Emeritus of Pharmacology and Toxicology

Department of Pharmacology and Toxicology and Therapeutics

University of Kansas Medical Center

3901 Rainbow Boulevard

Kansas City, Kansas 66160-7417

KIMBERLY TOPPER, Acting Executive Secretary

Advisors and Consultants Staff

Center for Drug Evaluation and Research

Food and Drug Administration (HFD-21)

5600 Fishers Lane

Rockville, Maryland 20857

GLORIA L. ANDERSON, PH.D., Consumer Representative

Fuller E. Callaway Professor of Chemistry

Morris Brown college

643 Martin Luther King Jr. Drive, N.W.

Atlanta, Georgia 30314-4140


P.O. Box 1362

Leesburg, Virginia 20177

JACK H. DEAN, PH.D., D.A.B.T., Industry Participant

President and Scientific Director

Sanofi-Synthelabo Research Division

International Director Preclinical Development

Sanofi-Synthelabo, Inc.

9 Great Valley Parkway

Malvern, Pennsylvania 19355

JACK H. REYNOLDS, D.V.M., Industry Participant

Vice President, Drug Safety Evaluation

Pfizer, Inc.

50 Pequot Avenue

New London, Connecticut 06340

DAVID M. ESSAYAN, M.D., Government Participant

Food and Drug Administration


Woodmont Building 1, Room 200 N

1401 Rockville Pike

Rockville, Maryland 20852

ATTENDEES (Continued)


JAMES T. MacGREGOR, PH.D., D.A.B.T., Government Participant

Deputy Director for Washington, NCTR

Parklawn, 16-53 HFT-10

5600 Fishers Lane

Rockville, Maryland 20857

HELEN N. WINKLE, Government Participant

Acting Director, Office of Pharmaceutical Science

Food and Drug Administration


Woodmont Building 2, Room 6008

1451 Rockville Pike

Rockville, Maryland 20853



KENDALL B. WALLACE, PH.D., D.A.B.T., NCTR Committee Representative

Professor, Department of Biochemistry & Molecular Biology

University of Minnesota

School of Medicine

Duluth, Minnesota 55812-2487

GORDON HOLT, PH.D., Invited Industry Participant

Principal Scientist

Oxford GlycoSciences

4 Sparrow Valley Court

Montgomery Village, Maryland 20886-1265

THOMAS PAPOIAN, PH.D., Government Participant


Parklawn Building 9B-45

5600 Fishers Lane

Rockville, Maryland 20857

ELIZABETH A. HAUSNER, D.V.M., D.A.V.T., Government



WOC 2 Building, Room 5060

1451 Rockville Pike

Rockville, Maryland 20852

ATTENDEES (Continued)



DR. FRANK SISTARE, Government Participant

Director, Division of Polypharmacology Research






Chair, Pharm Tox Coordinating Committee




by Dr. John Doull 6


by Ms. Kimberly Topper 6



by Dr. Kendall Wallace 9




by Dr. Thomas Papoian 50





(8:05 a.m.)

DR. DOULL: I think we can go ahead and start. As I'm sure all of you know, this is the meeting of the Nonclinical Subcommittee of the Pharmaceutical Science Advisory Committee.

The purpose of our meeting this morning is twofold really. First, we need to get a progress report from the working groups which we have established, and second, we need to facilitate the arrangements to support these groups and to keep track of what they are doing.

Why don't we go ahead, Kimberly, and do the security thing first, and then we can go around the room.

MS. TOPPER: The following announcement addresses the issue of conflict of interest with respect to this meeting and is made a part of the record to preclude even the appearance of such at the meeting.

Based on the submitted agenda and information provided by the participants, the agency has determined that all reported interests in firms regulated by the Center for Drug Evaluation and Research present no potential for a conflict of interest at this meeting with the following exceptions.

In accordance with 18 U.S.C., section 208(b)(3), waivers have been granted to Dr. John Doull, Dr. Gloria Anderson, Dr. Jay Goodman, Dr. Joy Cavagnaro, and Dr. Kenneth Tindall that permit them to participate fully in today's discussions.

A copy of these waiver statements may be obtained by submitting a written request to the agency's Freedom of Information Office, room 12A-30 of the Parklawn Building.

We would also like to note for the record that Dr. Gordon Holt, Oxford GlycoSciences; Lester Schwartz, D.V.M., GlaxoSmithKline; Williams Kerns, D.V.M., Pharma Consulting; Jack Dean, Ph.D., Sanofi-Synthelabo; and Jack Reynolds, D.V.M., Pfizer Global Research and Development are participating at this meeting as industry representatives acting on behalf of the regulated industry. As such, they have not been screened for any conflicts of interest.

In the event that the discussions involve any other products or firms not already on the agenda for which FDA participants have a financial interest, the participants are aware of the need to exclude themselves from such involvement, and their exclusion will be noted for the record.

With respect to all other participants, we ask in the interest of fairness that they address any current or previous financial involvement with any firm whose product they may wish to comment upon.

Thank you.

DR. DOULL: Any comments on the conflict of interest?

(No response.)

DR. DOULL: Why don't we go around the room then just so everyone knows everyone. Ken, why don't we start over there with you, who you are and where you are from.

DR. WALLACE: I'm Ken Wallace, Professor of Biochemistry and Molecular Biology at the University of Minnesota School of Medicine in Duluth.

DR. HOLT: I'm Gordon Holt. I'm Principal Scientist at Oxford GlycoSciences.

DR. REYNOLDS: I'm Jack Reynolds, Senior Vice President of R&D from Pfizer.

DR. DEAN: I'm Jack Dean. I'm responsible for preclinical development for Sanofi-Synthelabo.

DR. DOULL: John Doull, University of Kansas Medical Center.

DR. CAVAGNARO: Joy Cavagnaro with Access Bio in Virginia.

DR. ANDERSON: Gloria Anderson, Callaway Professor of Chemistry, Morris Brown College in Atlanta.

DR. MacGREGOR: I'm Jim MacGregor. I'm the Deputy Director of the FDA National Center for Toxicological Research.

DR. ESSAYAN: David Essayan, Center for Biologics, Food and Drug Administration.

DR. HAUSNER: Elizabeth Hausner, Division of Cardiorenal Drug Products, Food and Drug Administration.

DR. PAPOIAN: Tom Papoian, a pharmacologist, Center for Drugs.

DR. SISTARE: Frank Sistare, Director of the Division of Polypharmacology Research in the Center for Drugs, FDA.

DR. DOULL: Thank you.

Why don't we go ahead then. Our plan for this morning is we are not going to follow this agenda tightly since it's fairly flexible. As I indicated, we want to hear from the working groups, where they are and what they have been doing, and then we want to close by spending a little time figuring out how best we can help the working groups in accomplishing the goals that have been set out for them by this committee.

So, why don't we start with the cardiology.

DR. WALLACE: I'm real pleased to report back to the NCSS that the Working Group on Biomarkers for Cardiac Injury has, in my opinion, been very productive. There is certainly a lot of enthusiasm and commitment shared by the members, and it is a real pleasure working with this group.

Since the last time that we reported back to the NCSS, we have really focused on a meeting that occurred last week with the American College of Toxicology. So, what the working group did is that we met once here at the agency on October 12th, and it was basically a pre-meeting planning session where two of the speakers presented their science and then we spent the rest of the day just preparing for exactly what was going to transpire during this meeting where our group was invited by the American College of Toxicology to conduct a symposium on the topic of the troponins as biomarkers of drug-induced cardiac injury. So, we spent that day doing that.

We then met in downtown D.C. at the Renaissance Hotel the day of our symposium, which was last week, November 6th. It was an afternoon symposium. We had a morning session, and again we really wanted to maximize the amount of information that we gained from that venue where we had an opportunity to get feedback from the audience on the topic at hand. So, we met in the morning to strategize exactly how we were going to conduct the afternoon session with that in mind.

We then had the afternoon session where we had a series of, I believe it was, four or five presentations. The program from the American College of Toxicology meeting is included in your handout. There was one substitute speaker. I will point that out to you. So, we had that symposium.

Then we ended the symposium with, I thought, a very engaging dialogue, a question and answer period afterwards where the panel responded to a lot of questions from the audience on the topic, and I found that to be very helpful.

We then met the following evening and had a working dinner meeting where we then planned what we were going to do with that information. We planned our path forward. We concluded where we were and we decided what the next steps were going to be.

So, at this moment, what is currently happening within the expert working group for the biomarkers of drug-induced cardiac injury is that we are currently preparing a document that is suggesting that the troponins I and T are useful bioindicators, biomarkers, of drug-induced cardiac myocardial cell injury. We are just at the beginning stages of writing this document.

Now, it is very important that when we write this document, it is going to contain all of the elements that were outlined and presented last time, as far as what are the characteristics of an ideal biomarker. So, it is very important to understand that what these biomarkers, the troponins, mark is cell injury. They don't mark anything other than that. Active cell injury. They don't mark an infiltrative type of cardiac injury. They don't mark a congestive cardiac injury or an electrical malfunctioning of the myocardium. It's a cell membrane injury type of phenomenon. So, we are currently writing this document, have just begun the initial stages of writing this document as of last week.

In the process of writing this document, we are going to make sure we not only understand what type of cardiac cell injury or cardiac injury that we are marking, but also the kinetics of that marking, how rapidly this bioindicator is released to the serum, and then how rapidly it returns to normal values following the act of injury event.

But we also are addressing an issue that we think is important; that is, is there a relationship, a quantitative relationship, between the amount of troponin that is released to the serum and the amount of damage that has occurred to the myocardium? So, the data that we've looked at gives us an indication of what that relationship may look like, but the evidence is not as strong as we would like it to be in order to suggest it as a working document. So, another subcommittee of our working group is designing an ideal experiment that would address the issue of whether there's a quantitative relationship between serum troponin levels and the degree of cardiac injury so we can extrapolate on a dose or time course kind of a basis.

Then as we are doing that, we are also keeping on our radar screen the fact that we want to mark other types of cardiac injury. So, we are looking at other small molecules and biomolecules as biomarkers of other types of cell damage, whatever they may be. So, we have another group of us who are looking at the horizon and looking at what are we going to do in the next stages. The next stages I think are going to be very exciting.

First of all, let me just say that troponins offer a great opportunity in that they are ideal in the sense that they do cross platforms. The homology of the troponin sequence is such that we can use the same technology to mark myocardial cell injury in a variety of different experimental species as well as humans, and so this biomarker will have the characteristics of being able to be used to bridge across the nonclinical/preclinical studies to clinical and surveillance types of studies as well. So, we are real pleased with that.

Along the same line, it is our opinion -- and we will work this out in the document -- that the troponins may be more effective at marking cardiac cell injury than any other biomarker that is currently being used to mark the same type of event. But we are in the process of developing that argument.

We're real excited about getting into the next stages, the more challenging stages of describing other types of cardiac injury and then discussing what kind of markers that may be most useful in marking those events, whether it be the infiltrating, inflammatory pericarditis kind of a thing or what have you, but also looking at the emerging technologies of the biotechnologies, gene arrays and proteomics and such, and looking at the possibilities of whether those offer opportunities to employ those technologies as well in marking cardiac injury. But we are really looking several months down the road before we'll be able to really engage or invest ourselves in those discussions. But we're excited about that.

So, that concludes my presentation. I would be very happy to answer any questions or expand on any of the information that I presented.

DR. DEAN: Ken, will this document being prepared by your group compare the sensitivity against the classic CK and LDH, transaminase, et cetera and time course of the conventional markers versus troponins?

DR. WALLACE: The intent is to do that. It is my objective that in the process of recommending that the troponins be used, we do it on a comparison basis, that it be used instead of or in lieu of other classic biomarkers, not in addition to, unless in the process we can define how one of the other biomarkers pick up something of importance in the whole process that the troponins miss. But everything is going to be done on a comparative basis.

DR. DEAN: But both in specificity and sensitivity, the troponins look superior to the more conventional biomarkers?

DR. WALLACE: Yes. From the evidence that we've seen so far, it looks very promising.

DR. CAVAGNARO: Have you talked about how one would integrate this into a conventional development program, for example, as part of safety pharmacology or single-dose/repeat-dose studies or screening versus mechanistic? How would one then try to introduce this into a development program?

DR. WALLACE: Well, the committee has not discussed that extensively, Joy. It is all going to depend upon the application, the intentional use of the drug. What the committee has suggested is that the troponins will not necessarily be promoted for use as a standard drug screen, but in those cases where there's reason to suspect that there might be a cause for concern for drug-induced cardiac injury, that the troponins might be pursued at that point in time, but only if there's a basis for a concern.

DR. DOULL: Let me ask Ken. One of the goals of this subcommittee is to look at preclinical tests and then evaluate their ability to be carried on over into the clinic. I gather you guys are also focusing on that potential?

DR. WALLACE: Yes, we are and we are very encouraged that the troponins are being used clinically now. They are approved clinically now as biomarkers of ischemia reperfusion injury. So, we already have clinical evidence for this.

DR. DOULL: So, this paper you are going to do then would talk both about the preclinical and the clinical and efficacy and so on.

DR. WALLACE: If that is what you wish it to talk about. I'm not certain in that we've just begun to put our pen to paper, but if that's certainly something that you would like to have in the document, we'll make sure that it's there.

We weren't going to do an exhaustive review of the clinical literature, I don't believe. But we're going to talk about the promise that the troponins have and that there is already a lot of clinical evidence out there that suggest that they're equally effective in marking clinical myocardial cell damage. So, if you would like us to emphasize that in the paper, we'll add more emphasis to it.

DR. DOULL: I think Jack has already touched on the business about efficacy. You need to say something about how the troponins compare with other myocardial markers, and then hopefully you would say something about this is a preclinical test that also has clinical application.


DR. DOULL: Both of those are charges that we need to carry back to the advisory committee.

Are you guys thinking of publishing it? What's the future of your document?

DR. WALLACE: Well, I'm a big advocate of publishing, but I am trying to learn what restrictions and what parameters are put around the working group by the NCSS. At some point, I would like to see it published.

DR. DOULL: That's certainly something I think we need to talk about because that's our charge as to how we can help you, and that is certainly something, Jim, we need to give some thought to.

DR. WALLACE: My plan would be to charge ahead with the intent of publishing until I was told to hold up.

DR. CAVAGNARO: Certainly the data. The challenge becomes then in terms of recommendations. So, clearly all the data that you accumulated is very important in support of that, and whether a separate piece can be done in terms of making recommendations now, because this has, of course, huge global ramifications in terms of suggesting new parameters into development programs in the theme of harmonization. But I think clearly the data that you accumulate is very important that it be published to get out there.

DR. WALLACE: Right, and at this point in time, I believe all the data that we are reviewing is publicly available, and so we are not using any proprietary information that we are drawing our conclusions from. So, absolutely.

DR. CAVAGNARO: But that won't preclude it because I thought that there was some -- because it was important to get original proprietary data, you could somehow blind it. Wasn't that the charge, that people amongst the groups felt comfortable in presenting?

DR. DOULL: Yes. We discussed that at the last meeting actually, and the concern that some had was that in addition to the literature data, for example, you might very well have data from some of the companies, for example, that would be confidential, and how would that be handled in terms of writing a report or making that data available. We didn't really resolve that. We recognized that that was a problem. As I recall, we had no up-front solution to that, except to the point where you can get it into a review kind of document that you can publish. Then that makes it, of course, available as Joy says.

DR. WALLACE: The committee has talked briefly. Again, we are real early in the phase of writing this document. So, we don't know where we're going to come up against a block where we have this issue arise, at which point we'll have to address it.

The committee has talked about whether we want to go to other organizations to help generate and share information that is not yet public. But at this moment, as of last week, the committee felt that there was sufficient evidence available that we don't have to engage other organizations at this point.

DR. DOULL: Let me ask Jim. Are you aware of any constraints that we need to consider here in making this data public and so on?

DR. MacGREGOR: I don't think I'm aware of any. I think that the scientific aspects of the information pulled together could be published, and I don't see any limitations to that except for the normal issues of any proprietary data that might have been considered in getting appropriate consent or protecting that if that becomes an issue.

Since I have the microphone here, I guess one thing I might put out on the table is I think that now is kind of a critical point for this committee, which has now met several times and themselves gone over in fair detail in their committee the data on the troponins and, to a more limited extent, the comparison with other markers. But I think the committee, now that it has made a decision that it is itself convinced of the value of troponin as a flag for injury and that they're going to go ahead with the paper, one of the questions might be how this group sees their interaction with the expert group during the development of that. So, the committee is going to go off and write a document.

And I guess the question in my mind would be would you like to see a presentation of the scientific data here in this subcommittee and at what stage? I'm sure at some stage you would. Would you like to have the draft all done, read it, and then see the data, or would you like some interaction as it's developed? Or how would you like to see that development go?

DR. REYNOLDS: Well, I think some dialogue and interaction, as the paper evolves, would be very important. Many of us I think can have some insight into what should be done or where you should go and how it can be leveraged into practice.

DR. SISTARE: I think we should give the expert working group a question to answer. Some questions have been raised. Ken made the statement and it was a conservative one, but I think maybe we should ask and challenge the expert working group. Ken made the statement that when cardiac injury is suspected, to then look and measure for cardiac troponin T or one of the cardiac troponins as a measure. The question that has been bandied around is, is it better than some of the traditional markers? So, that's a good question to have them focus in on.

But another question that has come up in the deliberations is, why not make it a part of the routine clinical chemistry that's measured if we feel it can pick up things that are difficult to see? Unless you happen to make a slice through the heart and do some histopath and just happen to catch a focal lesion, there's a chance you might miss something like that. That's come up. So, I think that might be a charge to challenge the expert working group.

I guess we have to be careful. We have to not ask for recommendations but ask for data, data that would ask the question are there times when a routine measure of one of the cardiac troponins would give a superior insight into the cardiotoxic potential of a compound as if you didn't do it. Or would it be a waste of time and money and resources to do it routinely? If we can develop some data lines along that.

The other point that has come up is clearly Malcolm York presented some data with anonymous compound names where they looked at cardiac troponin and showed the real benefit in a lot of those cases where cardiac troponin measurements really helped them decipher what was going on with their particular compounds, whereas CKMB or the LDHs didn't do it.

My guess is, from talking to some of the other researchers in some of the other companies, they have incorporated troponins into some of their studies. And I'm wondering if the expert working group put a call out to sponsors doing these kinds of studies if they wouldn't be able to share data similar to what Malcolm shared that could be incorporated into a very extensive analysis of this question.

DR. ESSAYAN: Pursuing some of the questions that Frank just raised begs the question of what the stability of the assay would be over time and the reproducibility in normal, unmanipulated animals of various species. If we're going to go in the direction that Frank just outlined, that would be an important data set as a comparator to know the distribution characteristics of the normal values and then that would play into the quantitation question that you raised earlier.

DR. WALLACE: The committee has seen one data set that addresses both of those questions, the variability of the troponins in an otherwise control animal, as well as the stability of the troponins on storage.

DR. HAUSNER: Both David and Frank raised very important points, and I'm not going to reiterate what Ken just said about the kinetics of the troponins. But what Frank suggested I think is very important, and it was briefly touched upon in some of the discussions, that there seem to be two possible scenarios for use of the troponins. I guess the question is which one or do both increase the safety for participants in clinical trials the most. One could be where you have a suspicion of cardiotoxicity maybe because of other members of the class of drugs. The other scenario is a drug of unknown history, perhaps new in a class, and would you increase the safety more for clinical trial participants to have troponins in that case as a routine part of the screening? And I agree with you, Frank, that that would be very important to address.

DR. CAVAGNARO: I want to say about five years ago, but it could be longer, there was an initiative by the Association of Veterinary Clinical Pathologists. It was championed by Kurt Wyngard and Jack Bloom who is now in clinical. I know presented for CBER and there were presentations from CDER and maybe other FDA centers. The purpose of that meeting was to explore different analytes in terms of standard clinical pathology measurements.

I am wondering if you've had discussion with them or during this discussion -- because this is where you're going in terms of incorporating -- whether or not touching back with them or maybe there is somebody in your group that liaises with some of those folks. But I think to get the veterinary clinical pathologists involved at some point I think would be quite useful.

DR. SISTARE: Yes, I agree with that totally. Malcolm York, who is on the expert working group, was on that committee as well, and I challenged him with that question. You know, you guys five years ago published a paper, and for routine animal studies, you suggested these 10 or 15, whatever markers. You didn't include a real specific cardiac injury marker in that set. Let's talk about that. And would you revisit that now in light of all the information you've gathered? What do you think? And he really hemmed and hawed. He wasn't completely sure. He was still leaning on routine testing, probably not incorporate it, still save it for when you suspect it. But there was a challenge and there was a dialogue, and I still think it's something that needs to be resolved and more carefully thought through.

But I agree we may need to broaden it perhaps. I don't know if we have to broaden it or how we can dictate this or what. I'm not sure how the whole process works, but I do think that we do need to get a broad perspective on this and get real dialogue if we feel that everything is telling us this should be routine.

DR. DOULL: Yes, that's clearly one of the questions. Where do you position this in terms of it being the most useful? Early on as a preliminary screen? It doesn't sound like that. It sounds like later on when you have some clues to bring it in.

Looking through this paper, Ken, it seems to me that that did, in fact, cover several areas. I guess in terms of publication, I think it's better to have this out in the real world, what the working group is doing, rather than just present it here. We're going to have to take your information and present it to the advisory committee to show what we're doing and why it's important and all. But this needs to be out widely distributed in order that everyone is aware that we're concerned about cardiotoxic assays that are predictors of all kinds of things and how they might be used and so on. So, this is a good step I think in terms of getting the information more widely distributed and making a point about the fact that we're interested in this area.

One of the goals of our committee, at least, is that we need to facilitate this working relation between industry and academia and Food and Drug and so on. Clearly, this kind of thing I think does that. We're making that attempt to facilitate this kind of cooperation.

Yes, Dr. Holt?

DR. HOLT: We've used the word proprietary data here several times, and the difficulty is that our suspicion has been for some time that the really good data and particularly the kind of data that might answer the far-reaching questions that Frank has raised is probably in a domain in a company in many industrial settings that they don't readily want to share. The difficulty is, as I understand it at least, that when data comes to us it's by definition no longer proprietary. It might be new data, but because ours is a public forum, we have no way of participating in any kind of a confidential conversation. And that's fine.

But I do suspect -- and we've talked about it before -- that if there was a mechanism by which we could sector a certain amount of our time, predescribed of what we intend to do and post-described what has and has not been accomplished, but still keeping the actual data set in confidence, that would help us.

Similarly, for these new things, the markers that might come up after troponins have had their day in the light, there's almost certainly a fair bit of that data that the people who are generating this data would want to keep the information in confidence at least for a time. So, if there's a possibility of us getting that ability, that capacity, that would I think greatly facilitate our process. And my understanding is we can't do that right now. Is that correct?

DR. DOULL: Well, that's what Helen said at our last meeting. She implied that at least.

But we have expertise from Food and Drug, lots of it. We have expertise from industry, lots of it. And surely, this is a problem that has been wrestled with in the past, and I guess what we need to figure out is -- we don't need to rediscover the wheel -- if there's some mechanism whereby we can address this problem and find some kind of compromise that works, then I think we ought to look into that. I don't know who would do that, Jim. Who would we ask to help us get the expertise out there and how to deal with proprietary information?

DR. MacGREGOR: Well, we have people in FDA that deal with that that we would have to bring into it.

DR. DOULL: I think the point is if we recognize it as a concern, then at least we can begin to explore who we can talk to maybe and find out if we have any approaches that would work.


DR. REYNOLDS: It seems to me like this is an issue that may be overplayed a little bit. I think most anyone who would participate in this activity would certainly be aware there may be things of a proprietary nature that you would not want to divulge, but in my experience, they encompass mainly two areas. One is in the area of chemical structure, so you can keep that protected, and the other is not so much proprietary as that which would give you competitive advantage or you wouldn't want your competitors to know. And all of those things in my experience can be dealt with fairly easily. If you don't want to divulge competitive information, then you shouldn't participate in the experiments or this activity. And the other thing is, in terms of chemical matter, that's usually just a transient thing anyway until one has patent applications and things in place.

So, I'm trying to think of an example where we have been bound by, if you will, this kind of activity based on the proprietary nature of the activity. It doesn't happen very often.

Just one other thing, the main thing that companies want to do is when you divulge the information, the lawyers need to be aware of that so you can put in place when you should file patents. But most of us are not opposed to releasing a lot of this information.

The last thing is in terms of data and data sets, when it comes to clinical information, most of this is protected by patient confidentiality and those kinds of things. But one can still anonymize or randomize the data and have access to this.

So, my suggestion is we're maybe overplaying this a little bit, maybe trying to artificially constrain the activities of the group, because I haven't seen an example of where we've run into this yet.

DR. DOULL: I think what we're saying, Ken, is that the chair has to, in a sense, be involved in this and aware of it and alert for potential problems, if they do exist. We'll explore the possibility of helping you by some kind of -- if there have been arrangements in the past that have been helpful in this or we can figure out some ways that would facilitate this.

DR. WALLACE: I appreciate that. I especially am pleased that the NCSS is encouraging us to publish this data because that certainly was my desire.

The way that I see that we'll develop this is we'll start with an outline, which is basically the slide that appears on page 3 of your bound copy. It's the characteristics of an ideal biomarker. That will serve as the outline for our document.

Then as we go through each of those characteristics of an ideal biomarker, we will try to be proving or disproving one of those arguments, and in the process, we'll weigh the data that we have available to us in the public literature. And if we come up shorthanded, then at that point we'll have to stop and consider our alternatives, and those alternatives would be to go to the sponsor companies and ask for a sharing of data or to design de novo new experiments to address those issues. But as we make each of those individual arguments, we'll reach that stage and see whether it's necessary or not to approach the companies to share data.

Now, what we may want to include in this whole thing is, once the whole document is ready, distribute it to many of the sponsors, and as we make statements that the data that we looked at, let's say, proved specificity, give them the opportunity to react, if they want to share data that's against whatever claims we make in that paper. We could incorporate that into one of the near final stages of our publication to make sure that we've looked at everything.

DR. CAVAGNARO: Yes. I think that's a great idea.

DR. DOULL: We talked at one of our meetings about how you define a biomarker. Maybe that's going to help us to get some good, acceptable definition of a biomarker. It needs to be broad because this committee is thinking biomarkers really broad.

DR. WALLACE: It's been helpful for the working group to direct our discussions on that and limit them to what are characteristics of a biomarker. Ironically, there are several different subgroups not only within the agency, but in other organizations as well, that have come up with their definitions of what are the characteristics of an ideal biomarker. And it would be nice to get us all together because it doesn't matter what tissue, which drug, whatever you're working on, the characteristics are the same I would think.

DR. DOULL: Yes, that would be helpful.


DR. DEAN: I want to support what Jack Reynolds said, that I think that most of the companies, if you anonymized the source of the information, the chemical structure, and probably the indication, would have no problem in providing some kind of bridging data, if they have it. If it's in a clinical program, the data is probably already with the FDA. So, as long as it's been revealed to the agency in a good way, then I don't think the companies would have a problem sharing the information.

The second point, though, Ken -- and I know you're considering this -- is if this paper comes out being sanctioned by some subcommittee of the FDA, that it has to be of very high quality relative to the points that Frank made and that I made earlier as well around sensitivity, specificity, stability, transferability, et cetera because everyone will follow whatever recommendations are made. I would be concerned that we not just have one person's experience, that we have multiple people's experience, and the more you can pull information together from multiple sources to validate whatever people believe, it would be helpful.

DR. WALLACE: I would welcome the participation of the NCSS in the review of this document. I agree with you on that.

And, Dr. Reynolds, I would suggest that we engage the NCSS early in the process, beginning with an outline of the document to give you a chance to have some feedback on that, and then as we develop the items in the outline, at periodic stages again submit it to the NCSS for review.

DR. DEAN: Really, I would also like to encourage a presentation here by the working group of the information because people may be willing to join in early in the process of helping provide additional information if it's needed. It may not be needed.

DR. REYNOLDS: I missed the presentation at ACT and some of the background to that, but what I hear you telling us today is that you have found the biomarker for cardiotoxicity and that your work now, as I understand it, is being wound up in terms of defining the troponins. That's how I understood that. What activities do you have to look at other measures of cardiotoxicity or to ensure that troponins are in all cases the biomarker that we would want for cardiotoxicity?

I guess what I'm saying is we're here to try to determine how we can help your group move things forward, and it sounds to me like what you're saying is that what you needed us to do is to review your paper and not to stimulate further research or further activities. I know I'm probably missing something.

DR. WALLACE: No. This is the first in a long process. I want to be very careful that the committee believes that the troponins mark myocardial cell injury, but that's not the only type of drug-induced cardiac injury that occurs. So, we're suggesting that the troponins will be used to flag that type of injury and perhaps even to measure it in terms of a dose response.

However, we have not yet addressed issues of what kind of cardiac injury occurs that troponins do not mark and what are other markers then that we can use to cover those types of injuries as well. So, this is just the first stage of a much longer process we believe.

DR. DOULL: Back some time ago when this subcommittee was looking at this, we looked at a bunch of biomarkers, and it appeared to us that the troponins were kind of out ahead, that they had more data behind them and so on. So, that was a logical first group to focus on. And I heard you say, Ken, that part of the group at least will be looking at other biomarkers and evaluating those, and that's part of the evaluation of troponin as, in fact, a biomarker, is that comparison with other biomarkers. Down the road, I would imagine you all or another committee would be looking at other biomarkers of cardiac toxicity someplace.

DR. WALLACE: And we'll start that process at a much lower level where we'll need a lot more data generation in the early stages of those discussions than we had to have for troponin. Much of that data was already out there.

DR. CAVAGNARO: I think John had discussed it and others. The whole focus is to identify these markers to support clinical decision making, and so when you validate the marker, it's to a functional endpoint, to a histopathology. You had talked about some microscopic changes, whether or not they were there. Or really to a clinical finding.

So, the database really needs to have a significant number of cases were you see cardiac toxicity -- I mean, to me cardiac toxicity in humans -- and have missed it in the animal studies by the standard methods because again, we're not trying to assess cardiac toxicity in the animals. We are trying to introduce a new biomarker to help us predict clinical toxicity. But I'm sure that that's all part of the --

DR. DOULL: I think Jack raises another point, and that is in the sense that you guys are a working group of a subcommittee of a Food and Drug advisory committee, your reports and statements and what have you carry some weight which have to do with later on when a company comes to get cardiovascular drugs approved and so on, it may have some influence. I guess we need to think about that because that raises a potential problem I guess in the sense that because you are a working group of a Food and Drug subcommittee and all, what does that imply. I guess we're probably going to need some help, Jim, to be sure we do things correctly in developing all this and moving it along.

DR. WALLACE: That's why I was slow to answer your question about publishing because I didn't know if we had -- but I agree.

DR. SISTARE: I want to come back to something that Joy mentioned. I think we have to be clear that what we're asking the expert working group to do is to provide data that indicates that troponins are excellent reporters of myocardial damage. Now, they can do that in the animal study because you can take the heart and you can look at it, and you can ascertain that the damage is actually there.

Joy is asking a more complex question, and that is whether the animal model is predictive of the clinical manifestation of cardiac toxicity. That's a more complex question that's going to take a lot more work. The troponins could probably be the bridging tool to answer that question if they were measured in each and every study. But the amount of experience I believe with troponins in the clinic not to measure myocardial infarction but to measure drug-induced myocardial injury is pretty small. There's a good history for certain clinics using it to monitor the onset of doxorubicin myocardial damage.

I forget the substitute speaker that spoke, but anyway, we challenged him with that question. He mentioned a few cases. I think tributylene infusions were given in the clinic and they looked at troponins and they didn't see increases. Another case. I forget the drug that was given for infants for patent ductus arteriosus, and they did not see again that there was an increase in troponins.

But it hasn't been rigorously assessed. Studies like asthmatic patients taking puffs of bronchodilators and what happens to troponins in those cases haven't been done. And those are going to be important I think to get it incorporated into the clinic. But I think we have to do it one step at a time here and focus on the nonclinical benefits we can get out of this with the hope that this biomarker can serve as a bridge to move into clinicals. But I don't know that we're there just yet. Let's take it one step at a time.

DR. DOULL: Yes, Jim?

DR. MacGREGOR: My understanding of the focus of the topics chosen by this group is that particular attention should be paid to the accessibility of the markers that were considered by the groups for the express purpose that they could serve this purpose as a bridging biomarker. My understanding is not in the sense that the animal model would predict the human, but rather in the sense that the marker chosen could measure the same class of damage in the animal and the human so that you could then relate the animal data to the clinical situation and have a bridging marker that let you study similar processes in both species. That's my understanding of the charge. And I would encourage the working group to strongly consider that aspect, the extent to which the markers under consideration can serve this bridging function.

DR. CAVAGNARO: Can I ask a question? So, if the FDA were to somehow down the line recommend to sponsors to include troponin as a marker in preclinical, I guess to me I'm a bit confused, that then the clinicians wouldn't ask for that same marker in the design of the clinical program if there was a signal in the preclinical. So, I guess now I'm totally confused about the bridging biomarker because I thought what we were looking at is measurements that we could look at in preclinical and clinical.

DR. MacGREGOR: Absolutely, yes. I guess what I was commenting on was I thought Frank indicated that the animal model should be predictive, which it always has to be predictive, of course, but it kind of mixes two concepts. Do you have the right animal model for the human is one concept, but do you have a marker with which you can measure particular type of damage in the animal or in the human is a slightly different concept. And I think it's the second concept that we're --

DR. CAVAGNARO: Right. So, we're clear. So, if we were to recommend troponin in preclinical studies, then the clinicians on the team would recommend it in their clinical program. Is that true?

DR. PAPOIAN: Not necessarily. One way to think about troponins is like we would do for liver toxicity testing. If there was evidence of liver toxicity in the animals and say you had elevations of the transaminases, coupled with histopathology changes, we would make the same recommendations to the medical officers for their clinical trials and that they would add those markers to the clinical trials for testing not as a routine, but just as a screen, as a biomarker in those cases where there are signals in animals to monitor patients appropriately for those same sort of toxicities.

DR. CAVAGNARO: I think we're saying the same thing.

DR. PAPOIAN: I wasn't sure if that was the case. It sounded like you were maybe asking if we're going to routinely ask for troponins, are we going to routinely ask for troponins in clinical trials. That's what I thought you said.

DR. DOULL: Yes. But the reason for asking for asking for them in the animal studies is because we hope they will have some predictive significance in the clinical trial.

DR. PAPOIAN: Correct, yes.

DR. DOULL: There are studies that I guess that we do in animals for which there is no clinical significance like the NTP studies on carcinogenesis. Who knows? But sometimes they do, sometimes they don't. That's one of the questions that the working group has to look at, is this just a test that's just valuable in animals to detect cardiac damage or is it in fact predictive for something further on. That's something that you guys have to deal with in a sense.

DR. WALLACE: If I could comment on Frank's statement about the paucity of that data that's out there, that might be the type of data where we could actually mine it from some of the sponsor companies. They may have that data from their clinical trials that isn't necessarily available in the public literature that we might be able to get at. I'm an optimist.

DR. DOULL: Jack.

DR. DEAN: Just to clarify what Joy was saying, the way I interpreted what you were saying, Joy, is you thought if this was a great marker, we ought to routinely include it in preclinical. And I would caution us about wanting to do that yet.

We just had a research meeting last week where we looked at what's happening in discovery and in pharmacology, and within our company and I suspect most companies, this marker is being used to study reperfusion injury and protecting against ischemic injury. And it's a very impressive marker in the animal models in that sense.

I think that if we then went on to look at this particular compound in development in the clinic, we would want to carry that marker into the clinic to look at protection against injury. But I would not think that we are at a point of wanting to routinely include it. Maybe I misheard what you said.

DR. CAVAGNARO: Yes, you did. But the latter half was right. If you said it was worthwhile in the animal studies, then I think there's no other choice but to continue it on in the clinic. So, the last half is right.

But, no, I'm not making a recommendation for routine use in screening. That was one of my first questions. Are we talking about screening versus mechanistic here? And it sounds like more the latter if you have a signal that you're concerned about. Although I appreciate your concern of the unknown with a new class of agent where you present it, but again, I think I'm, at least at this stage, more toward the latter, more the mechanistic part of it.

DR. DOULL: Our goal is not to prevent heart damage in rats. We're concerned about more than that.

Other questions? Yes, Dr. Holt.

DR. HOLT: To emphasize something that came up in the conference, the assays themselves have some challenges. We heard about some differences that have been observed between different assay platforms. I just want to caution a little bit of slowdown here, that this is part of what will need to be documented inasmuch as we can document it. There are some challenges with the assays, but we feel pretty bullish about troponin itself. I'm sure everybody will be relieved to hear this, that it will be a while before I think anybody can put their hand on their heart and recommend all animals and all humans --

DR. CAVAGNARO: So, is it you have to take a separate blood draw? Is there something special about it? You could just include it as one of the standard SMAC analyses?

DR. WALLACE: It's fairly noninvasive. Just a small, like 200 microliters of blood or whatever is all that's needed.

The problem that Dr. Holt is referring to is that there's not a good standardization of the assay itself. It's an antibody-based assay, and there's a lot of variability between the individual assays. So, a number generated in one lab may mean nothing to a different number generated in another lab. So, we have the issue of standardization of the assay itself.

DR. SISTARE: But I would hasten to add that there is one assay for troponin T and there are about -- I don't know -- 24 or something like that for troponin I. But all of those are approved by FDA's Center for Devices. So, they have approved each of those assays with the caveat that here is the limits of each of the assays. Here's what you can detect, here's the linear ranges, these kinds of things. So, each one has their own performance characteristics as Ken has mentioned, and they're all over the map.

DR. DOULL: Well, I'm taking a couple of things away, Ken, that you have brought to us to look into, and one has to do with what it means when a working group, for example, makes recommendations and presents data and so on. We'll look into that. Also, the business about -- it's similar to the publication issue, whether we have any concerns about how we do that and what all that means.

I think the other thing that I would say is that this subcommittee is very anxious to facilitate the work of your group. So, we're going to try to help you in any way that we can, and I think that means, as you've said, we need to talk to each other frequently and see to it that we have good communication between this group and your group.

DR. WALLACE: I appreciate that. I think the whole process of developing the publication is going to challenge us to further define the relationship between the two groups because it's very likely we'll be coming back to the NCSS in the publication process saying that we need access to data. Will you help us gain that access to the proprietary data? So, we'll engage at that point.

We also may come back to the NCSS and say we need to conduct additional experiments. One of our activities is designing this experiment. Once it's designed, we'll then have to say, well, who is going to do it and who is going to pay for it and how is it going to be done. So, we'll be coming back to the NCSS for guidance on that as well.

DR. DOULL: I'm pleased that money is the last thing to come up. We've gone through all the science and now we're willing to talk about budgets.

DR. CAVAGNARO: Just back to the publication because I think that that is key for the data to be out there. This advisory committee is a bit different because, in general, advisory committees can make recommendations and the agency can decide to accept them or reject them or some variation. So, that would be easy. But this advisory committee has FDA representation on it. Oh, it doesn't? Are you members?

DR. MacGREGOR: Not the advisory committee. A formal recommendation to FDA eventually would come from the full advisory committee.

DR. DOULL: From the advisory to the pharmaceutical. We are a subcommittee.

DR. CAVAGNARO: So, then that committee doesn't have FDA representation on it.

DR. DOULL: We would make a recommendation to that committee, and then they would make the recommendation to the agency.

DR. CAVAGNARO: And that committee doesn't have any --

DR. DOULL: Helen, you got here just in time.

DR. CAVAGNARO: Okay, so that makes it a little bit easier.

Can you just comment about this process and the ICCVAM process? The rationale behind ICCVAM is to introduce alternative methods across interagency type. So, how do you see this initiative and the ICCVAM initiative?

MS. WINKLE: Well, certainly the main reason for setting up this subcommittee to the advisory committee was to get input on a variety of different projects that we might work on as far as ensuring that we could meet the scientific need and go through the advisory committee to get credentialing basically of those projects and acceptability of those projects throughout all FDA.

I think that differs a little bit from the purpose of ICCVAM. Now, Jim may have more comments on this and understand a little bit more the difference between ICCVAM and how the advisory committee works. But I see this as on a whole different level.

Some of what we had hoped to be able to do was to help with the funding of the projects through this process with the support of the agency. So, that was basically our thoughts in setting up this subcommittee.


DR. MacGREGOR: Actually I was thinking I might bring this issue up later in the more general discussion that was scheduled for the subcommittee because it's really kind of the next stage. If this expert group, for example, finds that it would be valuable to use troponin in certain circumstances and defines that, then it will come to the committee to make a recommendation of some sort, and that could take a number of forms. This is something the committee will need to think about in the context of whatever the findings are when all the data comes together.

But basically I see two mechanisms in place. One is the committee could make a recommendation to FDA that a particular assay ought to be used and maybe there ought to be an FDA guidance on this topic. So, that could be a typical kind of recommendation that could go from an advisory committee to FDA to consider.

But another thing that I just thought I might put on the table for some thought is that actually ICCVAM is undergoing a redefinition of itself at the moment, for those of you who are following what's going on in ICCVAM. It was reauthorized last year, and as part of that reauthorization, it's actually rewriting its own operating rules and structure and so on. A major part of the ICCVAM function for FDA -- it's 15 different agencies, 14 plus FDA, involved in ICCVAM are thinking about this, how is it operated, how should it operate.

So, if this committee were to feel that if ICCVAM were to restructure in a certain way to consider broader recommendations for biomarkers or whatever, this would be an opportune time to think about that. It's a good time to begin to think about it, but what is recommended probably will depend on the findings of the expert groups.

DR. DOULL: We should put that on our agenda, Jim, and probably have somebody from ICCVAM come talk to us.

Helen Winkle, for any of you who don't know her, is here.

A couple of things that came out of the discussion so far had to do with proprietary data, that the committee may want to look at proprietary data. The publication. Does the committee publish on their own or does it have the mantra of the Food and Drug on the publication, and what does that mean in terms of responsibility of the agency toward those activities? Those are general concerns that will affect -- well, and finally, I guess the budget things. Those are things that will affect both working groups I'm sure. So, those are things you might want to address.

MS. WINKLE: Well, it would be nice if I could.


MS. WINKLE: I think definitely these are issues that we need to look at in general as far as this subcommittee is concerned, but we'll have the same question come up with other subcommittees under the advisory committee. I think we're feeling our way with this subcommittee. I think as far as proprietary data, publications, et cetera, we need to put some information together and be able to address these issues, and I'm not ready to address them right now. Some of them are just coming up for the first time here. So, I think they're definitely very good issues. What I probably need to do is go back, have some conversation with some of the people in various organizations, and maybe we need to bring someone in here to answer those questions at the next meeting. But let's try and capture all of those.

DR. DOULL: And we will, Ken. That's what we're here for.

Are there any other questions or messages that we have for our working group on cardiotoxicity? Yes, Jack.

DR. REYNOLDS: Yes. I was taken by the definition you provided for a biomarker. I don't know if that was something your committee came up with independently or not, but I think that is probably the gold standard by which other biomarkers would be compared. To me it was the most comprehensive definition I've seen. Not that all biomarkers would do that, but certainly as we look at product profiles, we certainly know what we're shooting for and we know when we haven't made it according to your definition. But I would just like to commend you. If that is in fact some original work from your committee, it in fact is a very good definition. I think it's very workable for those people in the area.

DR. DOULL: That's that characteristics of the ideal biomarker.

DR. WALLACE: Thank you. That has been a focus of a lot of the committee discussions and we've used it as a guiding principle in our discussions. I venture to say that that's not the latest version of it, but we hopefully have improved it one more edition. But thank you.

DR. DOULL: We'll massage that a little. It looks like a good definition for us. We'll have a whack at it too. Why not? Right, Joy?

Thanks, Ken.

Then we're going to hear about the vasculitis, and Tom, you're going to do the vasculitis presentation.

DR. PAPOIAN: You may or may not know the co-chairs, Bill Kerns and Lester Schwartz, were unable to attend because they're contributing to the financial difficulties of the airline industry.

So, as I go through each slide -- and this is the first time I've actually seen them. I didn't see these before this morning, but I've been part of all the discussions of the working group up until now.

And if anybody wants to make any comments during the course of this presentation, then feel free to do so.

Just a quick thing. I just wanted to make a comment about what Jack just mentioned about the gold standard for biomarkers. It fits, in my opinion, perfectly for troponins, but whether it will be appropriate for markers of vascular injury is another story. And that's where it will be a little more difficult to fit those sort of high standards.

In the first slide, we've had a couple telecons in the last several months, and just last week at the American College of Toxicology, we had several hours of presentations by the various members of the working group on their individual research projects.

One of the other things part of this working group has been trying to do is identify terminology. Originally we had gone into the working group called the Vasculitis Expert Working Group, and this became an issue mostly for the reasons that from a clinician's point of view, vasculitis is thought to be an immune-mediated disease, and here we're talking about non-immune-mediated drug-induced vascular injury in animals as essentially the lesion that prompts regulatory concern.

One of the other aspects of finding biomarkers for vascular injury or vasculitis is that there aren't any available. And this is what the real difficulty is. If findings are found in animals, how do we monitor for such effects in humans? On the slide, it says, complete absence of biomarker leads. There are probably long lists of clinical inflammatory biomarkers that are currently being utilized and examined and researched for monitoring markers of vascular inflammation as risk factors for atherosclerosis, but essentially there's no validated biomarker that can sort of tie together animal and human studies.

One of the issues that the Vascular Injury Working Group has been trying to deal with is focusing exactly what the next steps are going to be and how to obtain a certain finding for various projects, particularly for the academic members of the group who are representative from academic situations.

And then also the intellectual property rights is also something that this working group has also been trying to deal with. Are there certain drugs that still have patents or are owned by drug companies? If they do research on these drugs, how is that going to be reported and how is that going to be dealt with and be able to keep your proprietary information?

Future plans is what I guess is going to happen after this group in subsequent meetings. Some of those issues I just mentioned will be part of the future plans.

Terminology. As I mentioned before, the issue of vasculitis versus vascular injury, rather than having it discussed at the whole group, essentially Robert Johnson and myself have had several discussions regarding which would be the more appropriate terminology. When it comes to identifying a biomarker, one of the issues of focusing on the inflammatory component is that you don't necessarily incorporate the initial lesions of the endothelial cells, smoothe muscle cells, subsequent possible repair processes that occur afterward, even though vascular injury would entail and would encompass the inflammatory component as well. So, whether we're talking about vasculitis or whether we're talking about vascular injury as an initiating event with all subsequent processes that occur after that injury as a potential source of finding one of more biomarkers was the reason for trying to essentially redirect the terminology into a more encompassing term such as vascular injury.

The clinical and preclinical observations have to do with the fact that vascular injury or drug-induced vascular inflammation is well described in many drugs that are submitted to the agency. One of the difficulties is finding evidence of drug-induced vascular inflammation in humans, and the only time it has been looked at is in the case of minoxidil. In those cases, they looked at autopsy specimens from minoxidil-treated patients and found no obvious drug-induced lesions.

One of the other problems with trying to find sequelae of vascular injury is how would one monitor for subsequent events. And one of the possible clinical subsequent events would be rupture of vulnerable plaques and that would occur because we have vascular inflammation and trying to pick up such a signal such as myocardial infarction may not be necessarily an easy thing to do in small-scale clinical trials. It may require some sort of large-scale monitoring after a drug is on the market or some sort of clinical trials that are done after a drug is outside.

As far as how to influence the selection of potential biomarkers, that again is simply what I just described. If we're looking at the injuring event, as well as the repair and all the inflammatory processes that occur in between, that's the range that the subcommittee is trying to focus -- is really not focusing but leaving the arena wide open as far as what potential biomarker will eventually be correlated with these drug-induced changes.

As I mentioned, the subgroup was established between myself and Bob Johnson to look at the terminology. One of the things that we will try to do is develop a glossary of terminology, something that we discuss all the various terms to make sure that we all understand and have the same understanding and definitions of the terminology that we're trying to encompass, including the biomarkers themselves, as well as the lesions and the cells that are involved.

These will be put into a draft, it looks like, by next March, and so we'll get that together.

The working group also wanted to get a copy, when it's available, of an internal draft guidance on drug-induced vasculitis that the Center for Drugs and Center for Biologics are currently working on. Once this gets published in the Federal Register, then the working group would like to have some comments and review that as well to make sure that we're all on the same page on that.

Just real briefly, the draft guidance takes a risk assessment approach to findings of drug-induced vasculitis in animals.

The research budget. Now, I can't address this too much myself, and if there are any comments on here, I would be happy to hear any comments.

The ILSI update. I believe that we're discussing aspects of how to obtain funding. ILSI apparently may be a potential source for that, as well as standardization of protocols and getting sort of a consortium of methodologies together.

There were apparently some discussions with NCTR to see if there's any way that research programs can be funded. I'll defer these points maybe to the discussion.

There was a plan to look to NTP, as well as contacting NIEHS, for potential ways that a particular project can be funded.

Also, Pharma apparently is another source that I believe was contacted following our last meeting last week to see if there's any way that they can help collaborate in this endeavor, and also Pharma and NCTR may be good sources where this working group feels that that might be promising.

Now, as far as the targets themselves, the working group, as I mentioned, is really leaving nothing out. It's looking from the very initial phases of injury to the first changes that we see at the endothelial cell layer. There's evidence of possible apoptosis occurring. Some of the endothelial cells apparently detach and circulate. The following inflammatory component, of course, has its cytokines and acute phase proteins that are possibly elevated. There may be a repair phase. Markers there are worth looking into.

And regarding whether it's a fishing expedition or not, it is focused in the sense that it is looking at vascular injury, but since there are so many potential candidates in cells in up-regulations of surface membrane proteins, as well as released cytoplasmic constituents, as well as changes in gene expression in the involved cells themselves, not just the target vascular cells, but also the inflammatory cells themselves, are all potential targets. In a way, it may be a fishing expedition, but we have a pretty good idea of what cells and what methodology is available to us.

A recent interesting approach was using NMR spectra to look at urinary proteins or urinary constituents. In fact, they're not proteins, but everything except proteins, as well as possibly in plasma. This is very sensitive methodology that needs to be really validated to make sure how well it correlates with actual injury.

And then proteomics is also being currently examined as well.

As I mentioned, the urinary NMR markers seem to be very interesting. They're very sensitive. Specificity is an issue that still remains to be established. It does require a lot of validation, and that's something that I think both subcommittees are working on.

There were some early results with PDE, phosphodiesterase, IV inhibitors as a potential research tool to see how to correlate the actual lesions themselves with the actual changes that one can monitor either in blood or in urine.

The acute phase proteins were shown to be changed with some of these compounds with the PDE III inhibitors particularly in the various species, and how this correlates with changes in acute phase proteins in humans still needs to be established.

And the concept of circulating endothelial cells as a marker of vascular damage is something also that's being looked at. Here again, too, whether these sort of assays and markers hold the criteria of a gold standard for a biomarker is probably a little more problematic, and I'm speaking here my own opinion that because these sort of changes can occur in all sorts of conditions either naturally induced, infectious, or drug-induced, and so these sorts of things have to be correlated with drug exposure specifically. That's where these will have to be key.

And the next steps. As far as the committee members go, several of the people are still involved in their own independent research either with their own pharmaceutical companies or in their own independent academic labs.

We also need to work in parallel to see if we can get a standard model for testing. In other words, we want to examine all the species, but a lot of the data, particularly when it comes to assays for these particular constituents, is usually human-specific or mouse-specific, and we need to get more data in the standard species that are used in studies for support of drug submissions. Those are rats and dogs, mini-pigs and primates.

We need to select a few compounds, and I think the list can be longer than shorter. I think there's no lack of potential compounds that are known to induce these lesions in animals. Also identify who is going to supply these compounds. Some you can buy off the shelf. Some are still proprietary. Where is that money going to come from if it needs to be purchased?

We need to develop a standard protocol for, I guess, looking at different species to be able to correlate specifically whether rats and dogs and humans share the same endpoints and effects to the same drug, and also agree on those endpoints.

We need to establish some sort of centralized way of doing these animal studies, and I'm not sure about this one so much as far as understanding that other than if there's going to be a central source for the animals, somebody is going to actually do the dosing and the analysis for the group or for several people involved.

Also, consider asking other people to participate in these studies, that we haven't done so already, and there are probably lots and lots of people out there that might have interest in this. Particularly, I think there's possible other academic labs that, if there was a source of funding, either through NIH or other sources, they might be willing to do some studies in their own laboratories.

And the goals. Again, to look at biomarkers that give you an idea of the mechanism, because one of the things that was discussed in the committee is rather than identifying a specific protein or marker that's elevated in an animal and looking for that specific marker in humans, also another way to think about it is to look at the mechanism that was affected and look for a similar mechanism marker in the humans.

And also look at late-stage biomarkers. That way you'd be able to find biomarkers -- you don't have to monitor people like right way after they take a drug, but hopefully the next time they come in for their clinical visit, a week or so later. That certainly would have an advantage.

And also, use all the technology that's available to us today of genomics, proteomics, and metabonomics.

And finally, the time line as far as when this is all going to be accomplished. I guess we're looking at February of next year to having an agreement on the study protocols and the appropriate endpoints to proceed on. What are the good model compounds that we need to use and just start using selected models so we can short of share data on using the same methodology in the same compounds, and somehow agree on the collaborative responsibilities, who's going to do what and how is this all going to fit together.

Finally, as I mentioned, the terminology position document would essentially just make sure that whatever we call vasculitis, we all agree on what we mean by that or any other terms that we throw around.

Define a budget for purchasing particular compounds that we need to test and be able to review all the data that comes in from the committee. And finally, finalize the plans between NCTR, the pharmaceutical industry, and Pharma and any other support that we can possibly generate.

Further on the time line, from next April to the following March, actually start possibly some of these protocols, actually start the studies themselves, after we have consensus on the protocols and the drugs that we're going to use, and then somehow have a central place to compile all the data.

Then in March of 2003, review the first data sets from these experiments and finalize plans I guess for a time line for guidance recommendations if that's appropriate as an end result being a guidance.

Then November of 2003, have a target date for an initial biomarker. That would certainly be optimistic to have such a thing by that time, given the complexity of the various issues involved, but that would certainly be a reasonable thing to begin with.

And finally, the committee members: Bill Kerns, Les Schwartz, David Essayan, Don Robertson, Fred Miller, Kerry Blanchard, Jim MacGregor, Frank Sistare, Paul Snyder, Prakash Nagarkatti, Robert Johnson, Scott Burchiel, and myself.

Thank you. I guess we will open it up for discussion.


DR. MacGREGOR: Just a point to keep the record straight. The way we've set up these committees, we actually had an FDA process with representatives across centers to select the experts in these areas, and then in addition, we assigned center liaisons to the committees. So, some of the people on these slides listed as committee members, in fact, are the FDA center liaisons to the committees. We have actually only one FDA member who is a formal committee member, and that's Gene Herman on the cardiac injury group.

DR. DOULL: It is clear that this report and the previous report are quite different. In the previous report, you guys have kind of focused in on troponins, and in this one you're really surveying the field to find out which one is really going to be the best.

I think this is a very exciting, a very fascinating kind of program and one that would, in fact, move us ahead to do exactly what it is you're intending to do.

You talk about focusing on injury and on repair, and I guess I'm thinking about the time line for that. Frequently the time line for injury is rapid and the time line for repair is very slow, and I'm wondering if looking at the time of that will help in designing those studies.

The other thing, ILSI has just finished a big cooperative study on cancer. I don't know how they did that, Jack. They went to industry and got support to do all those tests of all the various carcinogenesis testing procedures, and then the companies tested each of those against selected --

DR. DEAN: You're referring to the genomics program?

DR. DOULL: Right, and Ray Tennant's animals and so on.

What you're talking about sounds kind of similar to that, that there would be a list of drugs or agents that would be useful in this procedure. You would look at it against a variety of biomarkers, and out of that hopefully would come some clues as to direction. I gather that is sort of what your committee is thinking about.

DR. PAPOIAN: Yes, that's right. The list of compounds I think is one of the first things. Actually they sort of working independently to first identify the list of potential biomarkers that one can use, that assays are available. If no assays are available, how do we go about developing certain assays to be able to detect them in very small amounts most likely. And then to validate the actual markers themselves, we need to have a standard or small list of compounds to use.

Now, there's no shortage of experience as far as what compounds cause lesions in animals. I think one of the issues was which ones are proprietary. How do we obtain the ones that are not proprietary? Where is the funding going to come from? Which ones are we going to buy? But we can certainly select anywhere from a half a dozen or less. That's not a problem. And to be able to agree on these are the ones we're all going to test and use as the validation compounds. That's the thought currently right now.

DR. DOULL: Let me ask one other thing. You're mentioning the Food and Drug paper. That will help, you're saying, because it will talk about some of the same areas and so on?

DR. PAPOIAN: I'm not sure how much it will help because it's a guidance to be used for reviewers, as well as for industry, for everyone to be on the same page as far as when findings of vasculitis are shown to occur in animals, how appropriate risk assessment is done to determine what the human risk is in those cases. The guidance right now is in a draft stage. The subcommittee members have had their input, and it's just a matter of putting it into the correct format for inclusion in the Federal Register at the stage it is now.

But the process essentially -- there are a couple members here at this table -- is using standard assessments of safety margins, therapeutic indexes, mechanisms, known mechanisms, whether these are species-specific effects, whether they occur only in dogs, whether they can possibly extrapolate to humans. But those are all considerations as far as how we determine whether a compound has potential human risk.

DR. DOULL: That paper focuses on vasculitis.

DR. PAPOIAN: That's correct.

DR. DOULL: Whereas, you guys are looking broader at vascular injury.

DR. PAPOIAN: We're looking for ways to monitor these effects in humans. We're trying to find a biomarker for if these effects occur in humans. The only tools we have now are just standard risk assessment tools. Having an appropriate biomarker for monitoring patients is the task apparently of the working group, to be able to have some tool to allow these drugs to continue in clinical development, the safeguard being that you can detect these lesions if they were to occur in humans.

DR. DOULL: Questions for the Vascular Injury Group? Yes, Jack.

DR. REYNOLDS: Yes. I think Tom has focused on several points that I would want to make.

One, I think we talk about biomarkers, and I think when you look at cardiac toxicity, you can begin to imagine a biomarker like a troponin. I think with vasculitis, because there are multiple components of the organ, if you will, that are involved, there are multiple pathogenesis to the injury and the repair and the like, I think to target for a biomarker at this point in your research activities to me is kind of narrowing the field. I guess the term I use and would propose is helpful is if there were a diagnostic that would help us identify vascular injury from that diagnostic, where we identify that, we might be able to go back and identify a biomarker.

I think that certainly in my experience with my own company, when we begin talking about some of these possible diagnostics, folks always come back and say, well, you haven't identified a biomarker or you don't have any activity. Where are you going to find a biomarker? So, I think with vasculitis in particular, I would advise us to be a little careful and perhaps refer to a diagnostic for vascular injury with the hopes of finding a biomarker or biomarkers.

The other thing that's problematic, I think, in the area of vasculitis -- and you've touched on it as well -- is that since there is not a biomarker or a diagnostic for vasculitis in humans, most of us who study vasculitis-inducing drugs cannot go so far as to do experiments in humans where we can assess are the animal models a good predictor of what's occurring in humans or vice versa. There is an occurrence of vasculitis in humans that we probably never would see in animals either.

So, I think one of the things this committee could do is to maybe help bridge the gap of how do we get from our knowledge preclinically of vasculitis-inducing drugs, how do we do those exploratory kinds of activities in humans, obviously without endangering humans. And I think that's where we can have significant input into maybe building those bridges or those collaborations.

DR. DOULL: But the methodology you're talking about, a lot of that should be applicable clinically, shouldn't it? The NMR, plasma, urinary things, proteomics and so on?

DR. REYNOLDS: Yes. I would think of the methods they referred to here or others, there's probably not a method I've seen that isn't applicable in humans under some circumstances.

DR. DOULL: Yes, Frank.

DR. SISTARE: There are definitely a lot of issues to discuss here. I'm just wondering, with respect to timing, do we want to take a short break now and come back and discuss, or do we want to discuss for about a half hour?

DR. DOULL: Okay, 10 minutes.


DR. DOULL: I guess we're back on track. We're now at the point to discuss the vascular injury.

Since the name is formally Vascular Injury, it is no longer the Vasculitis Working Group. It is the Vascular Injury Working Group. We need to have all our records show that since we have made that change formally.

Okay, questions for Tom. Joy.

DR. CAVAGNARO: So, commenting again to Jack's point, this is a huge and ambitious initiative and very comprehensive, and if you do it by 2003, that will totally amazing. But this is a real issue now. There are drugs in development that are on clinical hold or various stages of clinical partial hold or whatever. So, in efforts to move forward in that regard, I'd be interested in some comments on that.

The other point is the diagnosis now is based much on histopathology data in animals, and whether or not on your committee you have representation of veterinary immunopathologists and human immunopathologists just because I think there are some species differences in animals that sometimes we call it vascular injury or vasculitis. I don't know what we're calling it now. I think that there are some differences, and to try to get a handle on the significance of perhaps some species differences.

So, I guess there were two questions. One, what are we doing now for those products that we are posting animals and seeing some findings and calling it that? And then, two, making sure that we're understanding potential species differences in how this syndrome is presented.

DR. PAPOIAN: I could begin and then the other panel members can take over.

As far as the current drugs go, until recently findings of vasculitis in animals were considered to be related to exaggerated pharmacologic activity. A new class of drugs, the endothelial receptor antagonists, sort of changed that in the sense that you can have vascular injury in the absence of systemic hemodynamic alterations. This has subsequently been shown. It was presented also last week and was very interesting. It can occur either in the absence of marked regional flow changes as well, which sort of further shows the complexity and difficulties of using any sort of analytical biomarker or any sort of test to be able to monitor for such effects in humans. The idea was that these sort of changes are unlikely to occur in humans at therapeutic doses.

So, I don't know if that's the reason why other divisions and other divisions throughout the center have put drugs on hold for findings of vasculitis. I think they were always doing that before, but previously if it could be shown that these changes occurred with marked hypotension and marked tachycardia, that these sort of effects were unlikely to occur in humans and therefore posed minimal risk. That's the current state of affairs with drugs in the center for now.

I don't know if anybody has any comments to that effect.

That's one of the things that holds up drugs and why they are put on partial holds or in holds is that people feel uncomfortable as far as how to be able to know, to be able to monitor whether such effects are occurring in humans or not, particularly when the drugs are not always cardiovascular drugs. They're not all vasoactive drugs. You can have other classes of drugs now that produce these lesions, and so it's not really clear what the mechanism is. The idea is they don't know whether these effects occur in humans or not.

DR. SISTARE: You had asked about the working group makeup in terms of pathologists, and we do have a few on the committee. A number of industry representatives on the committee are bringing slides that their pathologists have diagnosed. David is on the committee and Fred Miller. They're clinicians clearly, but not human pathologists, clinical pathologists.

Dr. Jun Zhang is actually trained. He's not formally on the committee but again, these people are sort of behind the scenes, but not formally on the committee.

A couple other points.

DR. PAPOIAN: Could I just make a quick point? Identifying lesions in humans is what currently does not exist. In other words, there are no lesions that have been ascribed in humans. So, it's not like one of those pressing issues that we have to have a clinical pathologist involved to be able to look at these sort of effects in humans and correlate whether these are the same things that occur in animals. I think maybe Dr. Essayan might be able to discuss as far as what drug-induced vasculitic lesions occur in humans, but the mechanisms behind those are obviously markedly different from what we see in animals.

DR. CAVAGNARO: I thought you said minoxidil was.

DR. PAPOIAN: That's the drug where human tissue was examined in people that were treated with minoxidil for up to a year, and after they died, their hearts were looked at to see if there was evidence of vascular inflammation in the hearts. Even though there was some vascular inflammation, some damage, it was thought to be not related to drug. So, that's the only case where human specimens have been looked at to correlate with what's known to occur in animals.

Minoxidil is sort of a different story, because the effects in animals occur -- it's a combination of vascular inflammation, as well as myocardial necrosis. The necrosis is thought to occur from the work overload from the tachycardia that occurs as a result of the drop in blood pressure. So, those sorts of effects are unlikely to occur in humans. If that were to occur, they would be given beta blockers, beta antagonists to prevent the increases in heart rates. So, it's probably not unreasonable to think that -- you know, it wasn't surprising that lesions were not found, but it gave a certain amount of reassurance that it's just not a toxic effect, but it possibly could have been related to its pharmacologic activity or the marked pharmacologic activity that was seen in animals.

DR. CAVAGNARO: I guess I'm referring to some of the cases where the preclinical data suggests some type of a vasculitis-like syndrome, but it doesn't totally fit to a human syndrome. So, then the question becomes, so do you err on the conservative, which generally happens? You err on the conservative I guess.

DR. ESSAYAN: Yes. I think to a large extent, these are the issues that we're trying to deal with prospectively, and it's difficult to comment on the question that you're asking because that's the question that we intend to ask through the research we're going to do. We understand your cause for concern.

As far as a direct effect on the current applications, I think we have to bear in mind where we are with this committee. We've generated a series of short-term goals where we're going to take the candidates that we've already identified and try to set up collaborative research efforts that would look into those. Our probability of finding the magic marker or the magic diagnostic from that group is relatively low, but these are leads that we already have which we wish to pursue.

The emphasis of this group is really much more on the long-term goals where we're going to embrace a more far-reaching discovery mode using these different other techniques to look for a pattern which may more appropriately be called a diagnostic, and from that pattern, either by NMR or proteomics/genomics type technologies, potentially identify a specific biomarker wherein we would be able to narrow down.

To what extent it's going to address mechanism is unclear. The kinetics, also unclear. Its applicability across the different types of injury that may be seen, also unclear at this time. In a way, that's what makes this particular committee exciting to me because we really don't have a lot of answers and leads right now and the field is very much open.

DR. SISTARE: It's tough to give one label. As Tom pointed out, there's a lot of debate about what to call this: vasculitis, vascular injury, vasculopathy. There are sponsors who have drugs which cause an inflammation on just the venous side. There are sponsors who have agents which cause vasculitis on just the arterial side. There are agents which kind of bridge the very small vessels, the arterioles, the capillaries, the venules. There are agents which cause vasculitis in just the mesenteric bed of the rat. There are agents which cause vasculitis in the rat, the dog, and the monkey. It's a huge spectrum.

And then there's IL-2 induced vascular leak syndrome. Where does that fit? That's a sort of a vascular injury. There's certainly an inflammatory component, but there isn't blood coming out.

So, there's a very diverse spectrum of things happening, and that does occur in the clinic, IL-2 induced vascular leak syndrome. That does occur in the clinic. So, there are a lot of questions here.

One of the first focal points is we see a manifestation of this diverse class of injuries in different species, different beds, and sponsors want to develop this. It's otherwise a very good compound, no other problems with it. This is the dose-limiting toxicity. Well, if this dose is 100 times where they want to go in the clinic, could it happen in the clinic? It's still there in the back of your mind, but you're not so concerned. But what if it's tenfold, what if it's fivefold, what if it's fourfold, the blood levels that you want to get to?

So, are we being too conservative in preventing the introduction of compounds into the clinic, into development that could be very beneficial on the one hand because of this concern that we have no way of monitoring for? Clearly if benefit outweighs risk and if you don't know that there's a real risk, you make those kinds of decisions. Those kinds of difficult decisions are made by regulators and sponsors all the time.

But what we need is we need something to shed some light on this very complex subject right now. We need to open up a new can of tools that we can use to measure something to tell us what's going on. Whether it's going to be mechanistic or whether it's going to be very proximate to the injury, both of those have value. Right now we just have nothing. So, we're not talking about an ideal biomarker. We're talking about something, some biomarker.

So, the first data that I've seen, data that came out of our lab, data that came out of Boehringer's lab recently, agents were dosed and we found some inflammatory biomarkers. Now, some inflammatory biomarkers are good in some species, but the same ones are not good in other species. C-reactive protein, for example, is very good in humans and dogs, but it's not good in rats. Alpha-2 macroglobulin is better in rats, but it's not good in humans. So, it's not a perfect biomarker, but at least it's something.

Now, if you're going to use that drug to dose someone with an inflammatory disease, arthritis, it's not going to be the best biomarker because they're going to come up with high levels already. Or if they have some underlying viremia when they come into clinical trials, they catch a cold, you might see some inflammatory signals going on there. You don't know if it's drug-related or what. So, it's not the perfect biomarker, but it's a start.

So, that's sort of the state of where we are with this. And what we're trying to do right now is keep the momentum going that's in the laboratories, as Tom pointed out, that there are laboratories that are doing things here. We're trying to keep our momentum going, but let's optimize what we do. Let's decide on one of the most important unanswered questions and which laboratories have technologies that they can tap into and the same samples generated from one study can be looked at in six different labs. If someone has the strength of flow cytometry to look at endothelial cell markers in the circulation, they can do that rather than five of the other labs trying to develop the methodology and get it in their lab. So, we're trying to optimize. So, that's one strategy, and that's an ongoing thing.

But as Tom pointed out, the expert working group has a number of options we can consider that are going to take longer. Should we try to get NIH to get grants out there? Should we try to get ILSI involved? These are all questions that we have. How can we best optimize this? Can we get other sponsors who are wrestling with this issue to the table? Can we bring them to the table as well in some sort of a collaborative mode without ILSI involvement? Maybe that would be the quickest way to do that. And how can we best do that? Set up a workshop, for example. Publish in the Federal Register.

Because right now, I don't know. We have three or four companies that are at the table with problems, but I know there are other companies that are developing compounds where they see this problem that are not at the table. So, maybe we could somehow open it up and get them involved in it. Again, if they're not willing to share their compounds, maybe they can dose animals and then they can share samples to other labs that can do specific measurements.

Some people have monkey studies. We can't do monkey studies. It would bust my budget to do one monkey. But there are other companies that could do monkey studies. Those are invaluable. The specimens you could generate from that would be invaluable to test some of the hypotheses.

DR. CAVAGNARO: Yes. I think a workshop is a good idea actually.

DR. DOULL: History is on your side. I'm thinking the problem you have really is sorting effects from toxicity. There are a lot of effects that we see in animals, for example, that are not predictive of toxicity either in animals or in humans. They are simply effects. But eventually, as one keeps massaging those effects, like P450, for example -- we knew about P450 years before we had any toxicity associated with P450. But eventually as we learn more and more about what those effects really mean and what they have to do with mechanism, somehow they get woven into the fabric of what really defines toxicity.

And you're kind of that way with vascular injury. There are a lot of effects, and as you point out, they vary around species and they vary around agents and so on. But until all that comes together and you have a handle on it, I think it's hard for that to be particularly predictive.

On the other hand, what else can you do? You have to start in that way, amass the information, and somehow try and make sense out of all that.

DR. SISTARE: I'm not sure I'm interpreting what you're saying accurately. There's no one, I think, that would doubt that what we're seeing in these studies is toxicity. These are not effects. These are toxicities. We're seeing injury. We're seeing hemorrhage. We're seeing inflammation. So, there was no one who would doubt that this is toxicity.

When we measure changes in proteins in the serum or when we measure changes in things in the urine, those are effects. And you're right. To try to sort out what are effects and what of those parameters that have changed are toxicities, that's a real challenge. And when we talk about a global approach that taps into genomics and proteomics and metabonomics, those are very important things that have to be sorted out.

In the working group, we have examples of laboratories that have strain differences. So, you can give the same dose and you can see pharmacology, but not get vasculitis. Give the same dose to a different strain of rat, get vasculitis. So, that's a nice tool to try to sort that out.

Then there's also specificity questions, things in similar classes that are not bad actors, these kinds of things that we're using to sort out. And then there's always dose ranges, those classical things.

You're right. It's going to take a lot of work to get us to the point. I'm a little more optimistic that we may have something prior to 2003 that won't be a validated biomarker, i.e., along the lines of a troponin, but it could probably be something that could be incorporated in a developmental strategy that could be used in the developmental strategy to get something from the nonclinical/preclinical into phase I and to learn at that point, to learn about the value of that biomarker, to incorporate that into the clinical study. If you have confidence, if the logic is there, you've done monkey studies, you have dog studies, you've done rat studies, everything fits into place, and now we'll say, okay, measure this in your clinical trial. I think there's a real learning opportunity, and it's something that won't have like an ICCVAM blessing. You may not have a diagnostic that's been approved by CDRH, but I think you might have something that you can incorporate into a clinical trial.

DR. DOULL: I think it will be a little slower with proteomics and metabonomics than it will be for some of the other things you're talking about.


DR. PAPOIAN: Frank mentioned something that was discussed with the working group about when would be an appropriate time to monitor patients if an appropriate biomarker or diagnostic was found to be predictive in animals, and that would be to also minimize the background, just not specific inflammation. And that would be possibly with the phase I clinical trials as part of the normal safety and tolerability studies where you can sort of minimize background disease incidents. And these would be healthy people free of disease, free of infections, and to be able to see whether the drug would produce the same changes that were found in animals, at least by the biomarker status. So, as an initial screen to be able to at least get into a clinic, they'd be able to find out at least in healthy humans and find out whether the drug has potential to produce the same sort of effects that were found in animals.

DR. DOULL: It's cheaper to do it in rats than in phase I.


DR. SISTARE: I want to make one other point, and I have to be really careful how I make this point. But a paper came out recently in JAMA which talked about trying to understand why certain people who were not otherwise at risk for a myocardial infarction -- they didn't have high lipids, these kinds of things -- yet had heart attacks, had cardiovascular events. And they found that there was a subset of patients that had elevated levels of myeloperoxidase and IL-6, and these are markers associated with inflammatory vessels. So, there's a logical link there. So, the thinking of the article is that maybe there's some inflammation that might be precipitating and predisposing certain patients to cardiovascular risk, risk of heart attack.

So, as we start and use animal models to uncover and unravel what might be some good markers of arterial inflammation, venous inflammation, vascular injury, we may be able to ultimately work up a set of endpoints that would be able to pick up patients or people that might be very sensitive to these kinds of things and ought to avoid certain things. We're focusing a lot on genomics, SNP, these kinds of things to identify patients that would be at risk, but I think as we develop the biomarkers for looking at some of these insidious effects, I think we're also going to be able to pick up responders and nonresponders or cohorts of patients that might be very sensitive to certain things and ultimately really improve safety profiles of medications. There may be some people that are very resistant to these things that can take a lot and maybe open up markets.

So, these are things that we have to think about I think in terms of expanding potential applications of biomarkers and seeing where this thing leads. But I think here and now the focus of this is we are in situations where we have drugs in development and we don't know how monitor, and we're being somewhat conservative here. So, I think that has to be the focus. But there's always potential leads.

DR. DOULL: Let me back up for a minute and go back to what we talked about a little earlier and that's ICCVAM. As I understand it, the whole focus of ICCVAM is to find alternative testing, protocols, whatever that are in fact predictive. One of the problems they have in ICCVAM is what's the gold standard. If you develop a new procedure, what are you really going to compare that against? If we have a new test for cancer, for example, you compare that against the NTP bioassay or whatever. And how good are the gold standards in a sense?

It seems to me we're going to have this kind of problem here also because Joy was asking what are you really comparing the biomarkers or the effects you're describing against in order to validate them, if I understood. I guess what that says to me is that somehow we need to be thinking about what ICCVAM is doing and keeping track of how they're making progress in defining gold standard biomarkers, if you will, or whatever they're going to do that will help us.

That's always the response. When you talk to those guys, you say, well, how does it compare with the NTP bioassay, and everyone says, well, that's not so hot. The rat doesn't predict for the mouse and neither predicts for man. That's always a problem of what is the final really test. Well, the final really test is does it predict the right answer for man and do you have clinical data that, in fact, validates the biomarkers somehow.

But ICCVAM must be dealing with these same issues, I would think, Jim.

DR. MacGREGOR: The reason I mentioned ICCVAM before -- well, let me just back I guess before ICCVAM.

I think when you come to the point where you have a new biomarker and become convinced that it has applications, then what you might recommend as a committee or what the parent committee might recommend really depends on the arena of applicability. So, if it's something that's specific to drug development, obviously there could be a recommendation to CDER for guidance. If it's something that cross cuts FDA, there could be across FDA FDA guidance. And then if it cross cuts other agencies, then you need to start thinking about ICCVAM.

And then when you begin to think about ICCVAM and biomarkers, I guess the way ICCVAM has operated in the past in my mind has not really been focused on specific biomarker application. It's more a whole new assay approach. And the way they've gone about it has been really a rather time consuming and slow process such that over their history, they've really only addressed a very few assay systems.

So, we in fact in the cardiotox group had some discussion and invited Len Schechtman from ICCVAM in to talk about ICCVAM and where it's going because it was felt that it would be useful to have an organized multi-agency group that could address biomarkers. But in my mind, probably ICCVAM would have to change the way it goes about its business a little bit in order to be able to do that in an effective manner. That is, they might have to think about establishing focused expert groups to just look at the evidence on a particular biomarker where it might be used, and they could do that perhaps in a fairly efficient way, if they were to organize in that way, because they do have the multi-agency structure. And the criteria are all laid out, so all that background is done.

And I bring it up in part because our office is heavily involved and because Len Schechtman, who is in our office, is chairing the committee that is rewriting the guidelines for how ICCVAM should be operating. So, I don't know if the timing will come together or not, but if the timing were to come together that this group could see a way that ICCVAM could help develop broad recommendations more efficiently, then some recommendations could even go out in that regard.

But I wouldn't want to see this group get diverted into that. It would only be if it made sense that that would be the arena where it might go.

DR. SISTARE: My understanding of ICCVAM also is that ICCVAM is an evaluation committee. All the work is done. All the I's are dotted. All the T's are crossed. All the data is there. It's submitted and it's either blessed or it's not blessed.

So, what we're talking about here is we're talking about developing the data. That's what's needed here. Even with the troponin group, it's like where is the data. At what point are we with the data? Is it at the point where we're close to a final thing? And maybe if we are, then we can go to ICCVAM with it as well in parallel.

But with this effort, we're talking about calling the cavalry and saying we've got a problem. Who can help us solve this? Who can help combat the enemy here, and let's solve this problem. So, we're not ready to talk about ICCVAM with vasculitis I don't think. We're not close to it. Maybe at some point we will, but right now I don't think we're there.

DR. DOULL: Yes. I'm not even sure we're close with cardiotoxicity to talking to ICCVAM.

Well, ICCVAM has only approved one procedure, haven't they, in all that effort they're doing?

DR. DEAN: There are actually three or four.

John, about ICCVAM and cardiotoxicity or the troponins, if the test is a test that's been approved by the FDA for clinical use, and you then apply it to the animal, I'm not sure that would become part of the ICCVAM process because ICCVAM is really, as Frank said, about evaluating new test methods in animals to see if they're applicable and have any predictive value and are they reproducible, sensitive, et cetera.

The local lymph node assay was probably the most classic example where you had conventional test methodology in the guinea pig testing and now you had a newer, more rapid test system in the mouse, and did they give you equivalent data in terms of risk assessment or hazard identification.

But if you take this area of vascular injury, I mean, you're so far removed from anything you could validate, that it probably doesn't fit.

But let's go back to your first point about ILSI a minute. Both Ken and I are on this Emerging Issues Committee, and in the conference call a month ago or less, this was presented as a topic that people were very enthusiastic about, at least on the committee, as something that companies might be interested in helping with. And I know in the January annual meeting, it will be proposed in the Emerging Issues Committee as something for the membership to see if they would be interested in collaborating.

As you pointed out, I think ILSI is a nice platform where you need to do work to get the industry and academia and government people together to try to do work and where you can put in sweat equity and the companies can put in money, et cetera. So, ILSI could be a very interesting platform for this kind of a fishing expedition. Maybe fishing is not fair, but where you need some industry people to do some work in some animals and share data.

DR. SISTARE: Yes. The issue came up at our last expert working group meeting. Clearly the feeling was that we have some momentum going now, we have some progress going on. Let's build off of that, and maybe we can find funding mechanisms with the academic representatives that were there through individual partnerships with some of the pharmaceutical companies. And they said, we see a very focused question. You've got something you can do for us. We'll find a way to make that happen. That was sort of the dialogue. So, clearly that was said.

But then the issue came up -- because I believe it was Bill who submitted the recommendation to ILSI, and then Bill found out that, yes, this will be presented in January. And there was some discussion about is that the best thing or isn't that the best thing. I'm not exactly sure how to say everything that was conveyed there, but I'll try.

There is clear need to get very focused things done and try to do them as quickly and as expeditiously as possible. There was some concern that going through the ILSI -- I don't want to say bureaucracy, but the ILSI committee structures and everything and then formalizing the proposal and getting it all to big Pharma and then big Pharma coming back and saying yes or no, might take a lot of time. And it might dilute potentially some of the funding that could go into direct, focused efforts. Not to take away the real benefit that definitely come out of these ILSI efforts. They've had tremendous success, and I applaud all of that and I encourage that.

So, I think it's something that could go on in parallel. We should probably proceed with that, but maybe something for the committee here to consider is what are the best mechanisms to make this effort succeed, to ensure success here.


DR. MacGREGOR: Just maybe to reiterate the charges to the various groups, I think the charge to these two expert groups is very clear, and the charge is to, within the assigned area of damage, to identify what we need in terms of biomarkers, what we have, and if we're lacking information, to define what information we need and also to define for the parent committee mechanisms by which that information might be gained. So, I think it's really a clear part of the charge to the expert groups that if you find that a certain set of research is needed to get a key bit of information, we'd like to see the expert groups come back here with alternative ways of getting that done for discussion and implementation.

The charge to the NCSS subcommittee is to serve as a steering committee to any such projects that might evolve, so that after the expert groups have really identified what needs to be done and alternatives for how, then it would, I think, be the charge of this subcommittee to make some recommendations and to facilitate and serve as a steering committee to the work to bring that information back in in a form that then could become the basis of a recommendation. And then recommendations would have to go through a parent committee to FDA for implementation. That's the way it should work.

DR. DOULL: Actually the Vascular Injury Working Group has, in fact, done that. You are saying to us you are envisioning some kind of a study for which you would develop protocols and drugs and procedures and so on that would provide you with some information about sorting out biomarkers for vascular injury. And you talked in there about ILSI and you talked about NPR, other sorts of funding and so on. So, I guess in a sense we're at that stage.

And the cardiotox. You have talked about some activities that also would need funds.

So, I guess we're at a point where we need to talk about how this subcommittee can help the working groups take the next step, and the next step you're talking about is really I think probably the budget. There are some formalities about how you do things within the agency and all, but budget is clearly an issue that we need to get around to. I guess, Helen, I wonder is this the time to move into that area because we're now talking about how this committee is going to function to help the working groups.

MS. WINKLE: I think Ken had a question first. Didn't you have a question before we get into this topic?

DR. WALLACE: I was only going to make the remark, as we were talking about formulating relationships with ILSI, that the vasculitis has already ventured into -- I was just going to make the comment that the cardiac injury group had also had the same discussions and is considering those opportunities, whether we can set up a synergistic relationship with ILSI and other organizations. But we are really looking to the NCSS for guidance on how to do that.

DR. DOULL: I might just mention that when we were talking about looking at various biomarkers and looking at imaging and proteomics and all those things, we didn't go into the liver specifically, and one of the reasons was that ILSI already had a working group that was working on the liver and was developing all those topics. So, we thought that would be redundant. So, there is, in fact, in ILSI an established interest in that whole area of biomarkers, and they've talked about it at the meetings every now and then. So, clearly there's some history for ILSI there.

Helen, why don't we go ahead and talk about --

DR. MacGREGOR: Can I make one more comment about these funding mechanisms and organizations before we go into that?

DR. DOULL: Sure.

DR. MacGREGOR: We've had some other discussions also. I guess personally I would say I think we're very close, but I'm not sure absolutely ready to go out and try to bring in specific support because I think both groups are moving toward a set of recommendations for an approach, but I think in both cases there are not yet specific research proposals or even -- and I don't mean in the protocol sense, but I mean specifically what studies need to be done still remain to be defined.

I guess there are a couple ways then you could go for resources and when. I guess my vision going into this was that the expert groups would identify very specific things that might be done and funding alternatives.

As the discussions have gone on, I think a second line of thinking has now surfaced that maybe some of these organizations could be brought in and might be interested enough to put up money and help refine and define the questions. So, that's I guess something to think about.

But the other thing I wanted to mention is we've had a lot of focus and talk about ILSI too, but I think FDA mechanisms are something that may also be available if things can be defined tightly enough. I am not sure if everyone on the subcommittee is aware, but the National Toxicology Program actually has a block of funding that's administered through the NCTR for the specific purpose of addressing major regulatory scientific needs of priority to the FDA. So, there's a mechanism there to bring science priorities that FDA considers its priority to NTP, which has got a funding block that's set up to conduct the work at NCTR. So, there's another mechanism to think about.

Now, the way that system works, it would require I think a fairly specific plan because there's a priority committee that's set up for the utilization of those funds that essentially compares competing priorities and recommendations from the FDA centers and then decides what would make sense to fund. So, that's another thing to think about and talk about when we have a plan at that stage.

DR. DOULL: I would make one point. One problem that we want to avoid is that we now have momentum in our working groups. They have had meetings. They have presented at the outside meetings and so on. Clearly that effort is recognized. It is moving forward, and we do not want this to bog down in the middle because of something that we have not considered, and funding certainly is one.

One option would be to get all the funding from FDA, if that were possible. That would be one option.

The thing that's wrong with that for me is that our charge says that our goal is to initiate joint efforts from industry, academia, and FDA. So, I think we need to think about joint efforts in terms of funding, joint efforts in terms of protocols, joint efforts in terms of what we do, and finally joint efforts in recommendations. And that means then that we would look around for money, not just from FDA.


MS. WINKLE: I certainly agree with you, John. I think the emphasis has always been for collaboration to have joint efforts in being able to get this research done. You and I had a conversation during the break. I think one of the things that we do need to take into consideration that has always been an issue and a problem is that CDER itself does not have the money to fund this research. So, we have the laboratory support that we could do some of the projects, but we definitely need to be thinking more broadly in how we get these projects done. This was one of the reasons for even setting up the committee, to be able to do more collaboration, to get people involved in some of these projects.

This brings me to the discussion we wanted to have. I think last time that I met with the subcommittee, we talked a little bit about moving this subcommittee under the auspices of NCTR. And I wanted to talk a little bit more about that. Jim is going to talk about it some. Because I think it's important for you to know where we are, and I think this is part of the budget issue as well.

I think it's very important to take into consideration the fact that NCTR is budgeted to do toxicological research and is very focused on this type of research. As I said to John, the center is focused on review. Obviously, in doing that review, there are regulatory questions that come up that we need to be able to answer, and we want to be able to do that through research so that we have data to support our regulatory decision making.

But let me go over again some of the issues that I brought up last time as to why it would be beneficial for this committee to be under the auspices of NCTR, and then Jim can sort of catch you up with where we are. I actually expected by this meeting that we would have made some decisions, but we're still focused on coming up with the appropriate way to handle this.

So, anyway, just to reiterate some of the advantages of moving under NCTR. And one of the things I've already mentioned, which is a real important advantage, is the resources. The NCTR currently has the resources to support the working groups, but also to support the research under these programs and the projects. I think what I've heard here today and I think what I've heard in the past from this subcommittee is there are some very significant projects that could be very beneficial in moving head in the regulatory realm. And I think it's very important that we make sure, as John said, we don't lose the momentum. I think the expert groups have really moved forward coming up with projects, and we want to be able to support that. And I feel that NCTR, because of its focus on toxicology, is in a better situation to do that than CDER.

Also, I think, as we've already mentioned, the ICCVAM process for the agency resides in NCTR, and if there is some need for collaboration or some activities that ICCVAM can help support in some of the projects we have, I think that's important to be able to have that collaboration.

I think also that the Science Advisory Board at NCTR really has the scientific expertise to help support these projects and help direct these projects. Although on the Advisory Committee for Pharmaceutical Science, we definitely have a toxicologist, John, that advisory committee is not focused as much in toxicological problems in general. Their focus is a little bit different, and I think moving to NCTR would help benefit this group in being better focused on those projects and issues and be able to provide expertise and direction. They also have the networking and more of the connections with the community than obviously CDER is going to have or our advisory committee.

Last of all, I think at least the last time I talked to Dr. Casciano, there was a real desire to bring this subcommittee together with NCTR, a science advisory board, to better coordinate some of the science. I think it's important, though, that CDER stay involved, and after Jim sort of brings us up to date with where we are, I will talk a little bit more too about how I think CDER can keep that collaboration with this subcommittee, if it is, in fact, moved to NCTR, and how we can ensure that interactions are continued.

So, I'm going to hand it over to Jim for a few minutes.

DR. MacGREGOR: Well, thanks, Helen.

I think Helen summarized very well kind of the FDA thinking, and we had quite a number of discussions within FDA about advantages and disadvantages of how we really might administer this group as it moves forward to really coming to the reality of developing collaborations and expanding and overseeing research projects and so on, which I think now is coming to fruition.

So, the issues that Helen brought up actually have been brought to the NCTR Science Advisory Board for discussion. Basically their feeling, if I can summarize it -- and I'll say it's unfortunate that Ken Tindall couldn't be here because the plan today was to have Ken Tindall be here for this discussion to represent the NCTR Advisory Committee and to discuss the issues that they raised. So, he was scheduled to fly yesterday when the news was coming in and decided not to do that and volunteered to call in, but unfortunately, technically we were unable to hook him into the public record. So, therefore it wasn't possible to even bring him in by phone. So, I'll just have to try to summarize, as best I can, the input from that committee.

So, their basic reaction was, yes, this makes a lot of sense. Scientifically this is where we are focused at NCTR. So, scientifically that makes a lot of sense.

The major question that they had and the major concern about moving the committee between the advisory committees was they saw the importance of maintaining a connection with the important constituencies which are, namely, the pharmaceutical industry collaborators who historically have been tied closely to CDER and also CDER itself, which is a critical link.

So, I think we have the situation that was plain to them that the regulatory science motivation kind of came out of the drug development arena, and then as we picked our focus areas and decided where we were going to go, those areas led us down a scientific path, namely biomarkers of safety, that happens to be a major scientific focus of the NCTR.

So, I think the consensus is that NCTR has made the decision that it's focusing a major part of its resources in its research program in the biomarkers area and that it does have a science board that has in-depth scientific knowledge of the toxicology issues, which would make a good sounding board body for the recommendations that would be coming up from this subcommittee. But the concern was that we not lose those critical links.

So, I'm going to now kick it back to Helen because Helen and I have been talking about how we might move to the center that has the resources, use those scientific resources for the discussion/recommendation development, and still not lose the critical links between CDER, the regulatory center, and also to make sure that all parties are comfortable that the information channels and flows would be maintained. So, I think we have talked about solutions that would make them work well.


MS. WINKLE: Jim, I just have a question before I talk. When this was discussed at NCTR, did you talk any about actually supporting these projects and whether the resources were going to be available? I'm just curious. As we talked about budget and how these projects would be supported, I was curious if this is one of the issues that came up at your discussion.

DR. MacGREGOR: No. The funding of specific projects really was not discussed. I think there's a recognition by all of us in FDA that if this effort is successful and we move into further expert groups in biomarkers, this is not an insignificant investment just to run the necessary meetings and oversee the travel and then have the necessary scientific dialogue where the expert groups are coming back to the appropriate FDA bodies to talk about the regulatory science and where it should go and data will start coming in, and so on. So, I think those are the kinds of issues we discussed, not the specific funding of projects.

But there was a recognition, just kind of in passing in the discussion, that yes, the center already is focused in this area and certainly would hope to be very heavily involved scientifically in projects that would flow out of these recommendations.

MS. WINKLE: I think Jim makes an excellent point. We want to be certain that we do have the appropriate connection with CDER if this subcommittee were to fall under NCTR. I think it's very important from the standpoint that obviously those appropriate regulatory questions, those questions that are necessary in doing the day-to-day regulatory activities of the center -- we need answers to those, and it's through this group or through the projects that come out of this group that we hope to get those answers. So, that connection has to be maintained.

We have looked at several ways and thought about several ways they can do this. First of all, I think that it's very important that our advisory committee has a member that participates with this group. It has been John in the past. I think that's a very important liaison back to the advisory committee. So, I would hope to continue to do that.

Of course, John's term won't last forever. I think it expires in about a year and a half. I would like to appoint another person with the toxicology background that can serve in the same capacity. I think this will be very helpful in continuing to bring these issues before the advisory committee and CDER. I think that will help some of the interaction.

I'd also suggest that at periodic times that maybe the Science Advisory Board out of NCTR could meet jointly with our advisory committee. We have done this with other advisory committees in CDER. We find it an excellent way of really talking about issues and talking about solving regulatory problems that may relate to two different advisory groups. So, I would like to propose that as well.

We would continue to have members on the expert working groups. We already have CDER membership on these groups. We would hope that they would continue to participate because again they're part of who understand the regulate issues that we need to address in CDER's activities. So, I think that's very important.

And we certainly are open to other suggestions for keeping this relationship viable. I think it's necessary that we participate very much in these issues because again, unless we have projects come to fruition and the results of those projects are used in the review process, the decisions that are made by the committee are really not going to be viable. So, we see the importance of this and really want to stay part of this activity.

DR. DOULL: Dr. Anderson, Gloria, is also a member, of course, of the advisory committee.

I had a couple questions. If we had working groups on imaging and other areas, I guess I'm wondering are we getting away from tox. That doesn't bother me. In the sense that NCTR is focused on tox, if you will, and genomics, there's no problem because they're already big in that area and hoping to be bigger. But I don't know about PET scanning and all that, whether that would fit or not.

DR. MacGREGOR: Well, that's a good question. I'm not sure I know the answer to that. I think in terms of technical expertise in imaging within FDA, you'll find that primarily in CDRH, and CDRH is a group that so far hasn't been too involved here. So, this would be kind of a future thing maybe to think about in terms of the molecular biomarker and how you would marry that to imaging, which is the other follow-on project that the committee had discussed.

But the one thing I did want to add to what Helen said is I think we already have a very good mechanism to ensure the cross-center participation in both the expert groups and the subcommittee. That is, at both the subcommittee level and at the expert group level, we have center liaisons from all the centers that have expressed interest in being involved in those activities based on the current definition of what they are. So, scientifically I think at that level, there's not any need for any change at all on the functioning up to the parent committee level.

The function of the parent committee is going to be when this committee and the expert groups kind of come to a scientific consensus that they think the body of evidence supports a regulatory implementation and a recommendation will come out. Then that will go through that committee. So, it would make sense to have that be the committee that's most knowledgeable in the regulatory science that's being addressed.

In fact, I think the reality is no matter how you cut many of these projects, they're going to have to go through parts of multiple centers. So, I think in the pharmaceutical area, any recommendation or guidance, however we set this up, probably ought to go through the NCTR for the scientific participation/evaluation, and eventually would have to go through one of the CDER committees and eventually into the appropriate CDER policy committee which would then be the arena that would implement the regulatory.

So, it's going to be necessary to maintain contacts with multiple centers I think. And as Helen pointed out, I think that ongoing working contact could be maintained by having representation from the appropriate CDER bodies on the working groups and on this committee, as well as on the CDER advisory committee, so there is a constant report back and feedback from that parent committee.

DR. DOULL: It may be a new ball game. What do you think about that, Jack?

DR. DEAN: Helen did a really nice job of talking about linkage back to Pharmaceutical Sciences. I guess my question is how do you propose linkage to the Scientific Advisory Board at NCTR? Would this be a subcommittee of the Scientific Advisory Board or would there be a liaison, or what's the thinking there?

DR. MacGREGOR: Well, the specific discussion is whether to make this a subcommittee of the NCTR Science Board, rather than a subcommittee of the ACPS. So, that's the specific thing being discussed. And the reason is all the reasons that Helen said. The major resources to support this and the scientific interaction and so on, as the focus is currently defined, has really merged more closely to the NCTR than to CDER in terms of the scientific operations, collaborations, et cetera.

Now, the concern of the NCTR board was that eventually recommendations are going to have to come back into CDER, and you want to be careful you don't distance yourself from that center. And I think that's an important thing, that we not distance ourselves from that center.

DR. DEAN: Excuse me, Jim, but would there person from the Scientific Advisory Board on this subcommittee then? Or would there be someone from here going and reporting to the Scientific Advisory Board?

DR. MacGREGOR: Periodically this subcommittee reports back to the parent advisory committee, which at the moment is ACPS, and obtains feedback from that committee and oversight from that committee. Now, as a routine what we're considering and what I think the FDA people involved have come to the conclusion -- we just want to make sure before we do this -- that, as John said, we don't do anything that causes a loss of momentum or that causes a feeling that we've taken an important party and distanced them from the process. We want to make sure that isn't happening.

But what it would mean simply was that the major reports and feedback would be occurring on a routine to the NCTR Science Board with one member or two members, whatever is decided by the CDER committee, also remaining a member of the subcommittee and reporting back and obtaining input. But the meetings where we bring all the people together, where we bring all of you together with all the Science Board members, that would then be with the NCTR Science Board rather than the ACPS.

DR. REYNOLDS: Well, I guess what I see, if we need to maintain this, that the NCSS I think has been a good group to tee up issues that need to have some scientific exploration through expert working groups. What I don't see and what I think may be a major charge of this type of activity, or at least that we're talking about and perhaps why it's important whether it's affiliated with NCTR or the Advisory Committee on Pharmaceutical Sciences, is as I saw this group at least, as we explore the science, we define diagnostics for biomarkers.

What I don't see happening is where drug sponsors can come and say, well, we think we have a marker for or we would like to progress this drug in development. And we already referred to several drugs that are in development that are on clinical hold, for example, because of vascular injury, that there's no clear-cut marker or way forward. What I think this committee, whether it's ACPS or others -- what I don't see happening very well is the ability to come and have this group advise the FDA on how one can move forward with these proposals or with these protocols. So, to me that's a very important part of the process, and so long as that process is maintained or available to all drug sponsors or people that develop drugs, that to me is the essential process that we need to retain.

And I think certainly in terms of spawning a lot of activities that are ongoing now under NCSS, NCTR is perfectly capable of doing that. What I want to ensure is the linkage back from the data and the science that's been generated there to its application to drug development and to advising FDA how they would let drug sponsors do clinical trials or do assessments to validate, if you will, or to further explore the data that you have generated.

DR. DOULL: Yes. I think in the early history of this group, as you know, that's really how it started out. We looked broadly for biomarkers in any area that would, in fact, be helpful to the clinical development of drugs. There were some others besides biomarkers and genomics and cardiotoxicity and vascular injury and so on. Liver, for example, which ILSI already is in, and there were some other areas mentioned. I think that option, through this subcommittee, is always a possibility once things get to the point where they need to be discussed or developed or described. That option would always be there I would think.

Yes, Joy.

DR. CAVAGNARO: So, then in terms of the funding, would it be similar to -- you mentioned NTP. So, basically the FDA gets a project through NTP, and at least in a previous life when I was involved, each center made a proposal and then it would be judged, but the FDA got something. It went through and sometimes it varied by project or centers gave each other chances to present a case that they thought was very important to them.

So, in terms of the NCSS, then we would be given discretion to identify things that were important, which wouldn't be second-guessed by the NCTR's Science Board who's dealing with all the other issues related to NCTR. So, when we get to the very important funding situation, it would be the recommendation of this subcommittee as to prioritization, and we recommend for funding. And then that wouldn't get re-prioritized in the whole mix of NCTR initiatives that go through their Science Board.

DR. MacGREGOR: Well, I think that's not correct. I think there are existing mechanisms for prioritization of NTP funding and NCTR research funding, and it's not being proposed that this committee would play a direct role in setting those priorities. What this committee could do is bring forward a recommendation to those bodies that we would like to see these funds used in that way. And then the weight of this committee would weigh in those decisions.

So, for example, there is a committee for NTP funds' use that includes representatives from all the center as well as NTP from outside FDA.

DR. CAVAGNARO: Oh, no, no.

DR. MacGREGOR: And they set those priorities, but the recommendations come through FDA.

DR. CAVAGNARO: Yes, right. That was an analogy. Right.

So, would it be similar to the way that is funded? That was the question. Would the NCTR look at this as a recommendation by NCSS that's been vetted through this subcommittee? And that's the recommendation, so that would be the recommendation.

DR. MacGREGOR: Well, as far as NCTR funding and research programs, this subcommittee would then become a part of that process; that is, it would be a part of that advisory committee that does advise NCTR on the use of its funds. There are two different processes: one for funds that are part of NCTR's FDA-funded and one that's part of NTP.


DR. MacGREGOR: And they're two separate processes for prioritizing.

DR. CAVAGNARO: I was talking about the NCTR because that's what this committee would be under.

DR. MacGREGOR: Right. It would become then a part of the advisory committee that advises on that, yes.

DR. DOULL: Frank.

DR. SISTARE: We look to NCTR for funding. So, would this compete with other initiatives at NCTR, other research initiatives that NCTR is involved in? We're not looking for extra money anywhere. We're looking to compete with their existing budget?

DR. MacGREGOR: I'm not sure I would put it that way. Let me rephrase your question I guess. Let me restate it in the way that I just said.

At the moment, there is a Science Advisory Board at NCTR that advises the NCTR on the quality of its programs, the focus it's taking, and how it should be spending its FDA appropriated research funding. If the subcommittee moves to that advisory committee, it becomes a part of that advisory committee and will then make recommendations to the NCTR management about the management of those research funds and how they're being used and how they should be used. We're not proposing moving any block of money anywhere because there is no block of money to support research from this group at the moment.

DR. DEAN: I'm not sure what the issue there would be because we actually go from a zero budget to at least having a budget that we can request funds from. Maybe I'm a little naive, Frank.

DR. SISTARE: Well, let's review the history of how these initiatives got to where they are. I was a little confused by Jim's comment that there was already some collaborative mechanisms in place at NCTR to do these things.

These efforts were brought to the forefront because of close discussions between researchers in CDER and reviewers in CDER. What are some of the vexing toxicities. What are some of the safety issues that are haunting, that keep coming up, keep recurring, and are not being solved? And I presented something here a while ago, and I listed a panoply of these, four, five, or six or seven of these things. Some of the more mature that the research lab in CDER had some ongoing momentum had some sort of like R01 data, preliminary data kind of thing, that gave maturity to the project, were the ones that surfaced.

So, that process worked very well because there was very close interdigitation between researchers who felt there's a way to do things better and reviewers who have a charge. They've got to review petitions on their table. They have to make decisions here and now, and they're going to do it the best way they can. But they're not thinking about 5 years from now, 10 years from now, what's a better way of doing things, but the researchers, working closely with them, sitting in the same committees with them, are.

The one danger that we have to be careful about with NCTR is just a geographic one and the lack of interdigitation in ongoing committees, the lack of ongoing communication in terms of the fine tuning. We have two projects very well scoped out now. There's momentum there. There's momentum there because there's ongoing laboratory research there. There are collaborators that have been brought to the table through things like CRADAs, funding mechanisms, from sponsors who also share the same issues, the same concerns. So, the sponsors are willing to put money on the table to get these problems solved.

What's going to happen five years from now when there are other issues? How are those problems going to be surfaced? How is this process going to continue? How is it going to perpetuate, to continue to evolve, to continue to ensure that we evolve to ensure that the best questions are being asked and the most important problems are being addressed? That's something I think we have to be careful about if we split off into NCTR. We don't have a CDER member on the NCSS right now.

DR. DOULL: Let me back up and ask. Our working groups have come specifically to NCSS saying we are planning something that will require funds. We are thinking about places that we could go to ask for that money. The issue then is, does that come through this subcommittee? Do they come to this subcommittee and say, help us go to NCTR or to ILSI or whoever to get money? Or are they on their own out there to go to those places and get money?

I think what the subcommittee wants to do is to help them get the money to do these jobs, and the question is, are we involved in all this money-raising thing? Do we go to ILSI with you guys to help you get the money, or do we say you're on your own to get the money? I guess I'm not clear exactly how best we can help our working groups.

DR. MacGREGOR: My understanding is that the subcommittee or any FDA committee cannot play a role in fund-raising and FDA can't fund raise. But what you can do is you can help us identify where is the science that can improve our regulatory practice and where's the common interest between people that develop products and people that regulate products to do that more efficiently and produce better products, and where are there existing resources that, if they could be brought together by appropriate mechanisms, could bring that to fruition. That's the role of the committee, not fund raising. There are mechanisms to bring the funds to bear to do things if the opportunities and the directions are identified, and if all the parties agree they want to go there, then there are other mechanisms that can make that happen and become funded.

DR. DOULL: Okay. So, what you're saying is our role is to encourage that and to support it by writing letters and what have you, but it is the working groups that will actually do the asking for money, so to speak, not us.

DR. MacGREGOR: No. Working groups won't ask for money either, I don't think. Working groups will identify what it would be useful to do, ways of doing them, how groups could be brought together, and then groups that have appropriate relationships to fund, which are doing that at the moment, can do that. And we've talked about some examples. So, ILSI consortia is one example. NTP funds is another example, and there's a list of examples that have already been identified as possible ways to fund these things.

The important thing, I think, for this group to do is to set the direction that we should be taking and pull the collaborators together, the internal and external stakeholders to the whole process of pharmaceutical development to develop a consensus on what's important to do that actually could be done.

DR. DOULL: Jack, you're on HESI over there. You know how that works. People come in and say, hey, I need some money to do this. Or I'm on RSI. That's what happens to us. People come and say, hey, we need this, we need some money to do this thing, and you say, yes, it's a good idea and say to ILSI or whoever, why don't we fund this thing? I'm not exactly sure how we can help with the mechanics of it. I hear what you're saying, Jim, and we need to figure that out.

Gloria, help us out.

DR. ANDERSON: I don't have an answer to the money question.

I think there are several issues here on the table, and let me say right up front I don't have any problem with the transfer of the responsibility, I guess it is, to your agency.

But I'd like to go back to the original goals of this committee which I thought you did an excellent job of defining.

First of all, I'm having difficulty making the leap from the goals and objectives in your original background paper, which we're reminded of each time we get a document, which makes it easier to prepare for the meeting, but I'm having difficulty making the leap from those goals to where we are now in terms of what the expert working groups are doing. But perhaps that's my problem.

The thing that I would like to point out, however, is that it was my understanding based on the problems that you defined, problems you defined initially, which this subcommittee was to address, that the expert working groups which were created were addressing only a small portion of what was a larger problem in the drug development industry. Am I correct in that?


DR. ANDERSON: Now, if that is in fact the case, if the responsibility for what is happening now is handed to an agency that does toxicology research, what happens with the ongoing discussions or the potential for ongoing discussions about other problems that are related to bridging the gap between the nonclinical studies and clinical studies and increasing the predictive value of nonclinical studies?

And I hope I said up front I'm not against what you're talking about. That's just a lingering question in my mind.

DR. MacGREGOR: I think Frank made an excellent point. Where did the issues that we identified come from? The principal place they came from were twofold, I think, discussions that happened among essentially the subcommittee and people that are in the room, number one. And number two is we went to the CDER Pharm Tox Research Coordinating Committee, of which Frank is co-chair, and we said, how do the CDER regulatory scientists see this? What do you see as the major problems in your view? And that's really where the specific two problems of the cardiotox and the vascular injury arose from, were recommendations from that committee based on current activities, largely by Frank's lab driven by really the regulatory issues and recognizing the potential to make much greater progress if various external parties might be drawn in in an effective way to solve those two problems.

So, that's really the way it happened, and I think the way it would continue to happen. That part wouldn't change at all. That is, we'd go out to the centers and we'd solicit their input from those kinds of committee into this group, which was constituted to be a knowledgeable group to evaluate that kind of information and make recommendations. So, all that would be essentially the same as it has been.

DR. ANDERSON: I guess my question, if indeed it were a question, was not clear. My understanding is that the currently existing expert working groups have a specific responsibility in a specific area. Perhaps I should rephrase the question then.

If these two groups solved the problems that they are addressing now, will that solve the problem that you identified in the original document and will it address all the goals that are listed there? Because if that does not happen, it means then that there is a need for someone to address whatever is left, and even if the responsibility for what is now taking place is sent to another group -- and I'm agreeing with that -- then who will address or how will the other parts of the problem be addressed?

DR. MacGREGOR: Well, in my view the background paper that you're referring to basically makes the argument or identifies a number of areas of scientific advance that those of us involved in putting it together felt presented opportunities to improve both drug development and drug regulation and identified some of those scientific bases of how they might be brought to fruition. Really that was the reason that this subcommittee was created. It is this subcommittee's charge to evaluate those opportunities and to make recommendations on how to bring stakeholders together to pursue those opportunities. That's really the charge of this subcommittee.

In other words, I think what you're asking is how would that change if the subcommittee is moved to a different oversight committee. In my mind, that part doesn't change at all. That is, this subcommittee was constituted of experts that ought to be able to address those issues and make those recommendations.

Shall I say this in public? I don't know, but I will I guess. My belief is that probably there could have been a specific advisory committee just charged for this if it weren't for kind of technical reasons that exist within FDA that necessitated this effort to be part of an existing advisory committee. So, in a sense this committee has been constituted in a way that it holds public meetings and it sets a direction and it is charged to do some things that advisory committees really haven't done before, which is to not just identify areas and provide advice, but try to actually facilitate the implementation of that advice. So, that's the thing we're on the verge of addressing now that we're hoping that this subcommittee will be able to do.

DR. ANDERSON: This is the last thing I'll ask. What I've heard this morning is we're transferring responsibility for the current expert working groups, that piece of it, or are we transferring responsibility for the Nonclinical Studies Subcommittee to another area? That's the question I'm raising because I think it's important. The answer may be important to achieving the goals that are set forth in the original document.

DR. MacGREGOR: If I understood you correctly, the responsibility for the expert groups resides with this subcommittee, and that part won't change. But this subcommittee recommends to the parent committee when it's time to make a scientific recommendation to the parent. That's what would change.

So, the question really is this committee is charged with the responsibility for implementing this backgrounder, and the question is, what's the best parent committee to dialogue with and to present recommendations through and what's the best mechanism for kind of overseeing and facilitating that? I think that's the question.

DR. DOULL: Yes. When the subcommittee was set up, the two groups that we have expert working groups on are the two groups that are most mature and ready for that. Imaging we looked at and thought maybe, but it wasn't quite ready. Genomics, we're not exactly sure. So, down the road potentially there's a whole batch of additional working groups that might be formed as these sciences come up to the point where they need expert working groups to be brought about.

The other issue then is once these committees develop a good methodology, solve the problem, as you say, Gloria, then the question is how does one bring that information into a recommendation and back into the agency where it can, in fact, improve clinical use. That I think is the issue, as Helen has pointed out and Jim has pointed out, that is of some concern. This subcommittee needs to have the ability to do that, to bring that information to the agency in a way that will help improve drug development and use.

DR. MacGREGOR: I notice Joe DeGeorge has arrived. Part of this discussion that we're having at the moment is to be sure that we're maintaining the connection between the regulatory implementation and also identification of the regulatory science needs. Joe, as I'm sure everybody knows, is chair of the CDER Pharm Tox Coordinating Committee, which is the regulatory body to which these kind of recommendations ultimately would go and it is also the parent of the Research Subcommittee that provided this group with the initial recommendations on these two areas of cardiotox and vascular damage that we're pursuing. So, Joe, do you want to comment on your views?

DR. DeGEORGE: Thank you.

I was listening to the discussion and I think it's interesting. Frank raised the point I think people are concerned about, and that is, is the connection going to be maintained to the regulatory center and, in particular, to CDER? I think that there is a possibility that the committee may have valuable information and gain valuable insights from other centers such as Biologics and their aspects that might be more centralized if, in fact, were part of the reporting to the NCTR committee because they can address biologics issues, not just CDER's concerns about cardiovascular toxicity or more vascular toxicity.

I think our center would continue to function with its researchers and with its evaluators, regulatory review staff, to identify issues that are of concern. We identified two that were thought to be mature enough to take on, and I think we'll identify some other ones as they come along as well and they'll surface through our committee and we'll make recommendations that we think need to be pursued in terms of developing approaches to address these problems. And I don't see why they would not flow back through a committee, which in fact is now situated in the part of the FDA whose mandate is to address these problems.

The mandate of NCTR is, in fact, to develop methodology to support regulatory decisions in relation to toxicology, and I might expand that to safety not just toxicity, if we're thinking only of the nonclinical aspects. Clearly a lot of the work they're doing down there has to do with biomarkers that are looking in humans, not necessarily looking in animals for outcomes.

So, I don't see that there's any difference in the process, with the exception that the committee will be at that point part of a committee that influences -- not directs, but influences -- the conduct and direction of research for a substantial portion of FDA's budget -- that is, the budget of NCTR -- and could respond to the various centers, not just CDER, but other centers. But clearly we'll keep our voice loud enough to be heard even if it is in Arkansas. So, I don't think there's any problem with that. So, I hope I can help.

DR. DOULL: Frank?

DR. SISTARE: I think the mission of the NCTR is more focused. Jim, do you know the exact verbiage for the mission? I think it's fundamental toxicology research. That's really their charge. They do support the other centers, but I believe their mission is focused on fundamental toxicology mechanisms, applications, these kinds of things. The mission of the Center for Drugs is to ensure that safe and effective drugs are available to the American public. A subtle, but very important philosophical difference there. The Center for Drugs is focused on making sure that drugs are safe. NCTR asks and answers very important questions regarding risk assessment with respect to toxicity questions and what might the relevance be to people. Subtle but it's distinct.

The other thing is the Center for Drugs is very closely intertwined with the drug development process. NCTR is not. They're very talented researchers focused on toxicology issues. These questions are problems because they are involved in stymieing progress in the developmental pipeline.

Ensuring a dialogue with a research view that's focused on very critical safety issues that are hindering that process is something that has to be ensured. I don't know what the best mechanism is. I don't know what the best mechanism is. I'm expressing some concern. That's all, by transferring it to a center with a different mission.

DR. REYNOLDS: I think one of the things that's important is to see the role of this Nonclinical Studies Subcommittee. To me, it's kind of an oversight body, a steering committee, for doing this basic research, and I do like the notion of safety, not just toxicology per se. So, I think that NCTR can do that well, in fact maybe better than what's going on now, not to be judgmental.

But I do echo what Frank and Joe and others have raised, that the mechanism for getting out what are the issues for CDER, that is, in the reviewing divisions, what are those issues that impede drug development and impede our ability to make decisions about safety. Identifying those issues I think we have done very well. And I think Frank said it pretty clearly, that what we need to make sure of, wherever the oversight lies, is that that information comes back to the reviewing division that individual reviewers and decision makers within those divisions can take advantage of what has the Vascular Injury Working Group decided, what is an appropriate measure of vascular injury in a clinical trial, phase I, phase II, or phase III, so we can progress drug development.

So, where this reports to to me is not so much an issue as it is we need to ensure that the processes in place for getting the ideas to the researchers and then getting the outcome, the experiments, and the results of the experiments back to the people that need to use them. If we ensure that, which I am sure we can, then I'm comfortable wherever this might be.

DR. DeGEORGE: I'll just make a comment, Jack. First of all, it's not like this is a new process to us in terms of developing new approaches. Transgenic animals were not done within CDER. They were done by a collaboration. In fact, they were done at the NIEHS's facility. We looked at that information. We made determinations that we could use that information in regulatory decisions. So, clearly if the science is brought and a case is strongly made for it, we can find a way to incorporate that into the regulatory process.

The working groups are supposed to identify how to solve the problem, what is the state of the problem, what is the state of the science, and what needs to be done to solve that. And then someone has to do it. It may be that NCTR is not interested to do that, but perhaps this committee, working through their parent committee, would make recommendations that some other group does it, NTP does it, NIH does it, FDA CDER does some of it, and that they try to make a recommendation this is important research, it should be done. And then when the outcome is available, we'll find ways to incorporate it into our regulatory process. Whether it's done under a parent committee that's part of CDER or a parent committee that's part of NCTR, that won't really influence us because the other processes were done and made recommendations that didn't come out of any parent committee related to the FDA at all. So, clearly we'll use the information.

So, the feedback loop that you seem to be concerned about I don't think is going to be any different in the future than it is now. You might actually have an opportunity to get to influence the support because you're working through a committee that actually has the resources to garner some support.


DR. CAVAGNARO: I was the ex officio member to the committee for CBER, the NCTR committee. So, it's like deja vu. There was always opportunity for centers to interact. As you know, that was a big thing -- I don't know -- five-seven years ago, this interaction with the various centers and NCTR.

I guess I see the difference here -- again, to go through the prioritization -- is if we've identified issues that are the critical issues to address, is no an acceptable answer for research? I mean, right now we're coming down to the point of funding. All of this, wherever we end up in terms of oversight, the bottom line is still going to be funding these initiatives that for the past two years we've identified as the key initiatives. Whether or not we all agree that they're the key initiatives, they ended up being the key initiatives, if nothing else, to serve as a prototype of an interaction between academia and industry and government, to answer the question.

So, I guess my question is, when we ever get to the discussions in terms of funding, which clearly has to happen for the vasculitis to meet any goals and cardiotox might be a little bit better served because there's more data, but there still needs to be resources to address some of those issues, will somebody still go forward? Will Frank's group still go forward trying do something? Will drugs still be on a clinical hold until we go forward? That's the piece that I am totally not understanding.

MS. WINKLE: Joy, I think you bring up some really good points. I don't think we're going to solve the oversight of this group sitting here at this table. I think that there have been a lot of issues that have been brought up today that are very, very important. I think the things that we need at FDA to do, when we go back to the table, is make certain that the interactions do remain here, that many of the questions that have been surfaced today, as far as continuing with these projects, how future projects will be done, how CDER will stay involved, need to be put on the table and we need to discuss them further. There have been issues brought up here today that I have really, truly not thought of. So, I think this discussion has been extremely valuable as we go forward.

I think what we're talking about here, though, mainly is oversight of this committee, and I think the important thing is that the projects we've already identified continue to move forward, whether the oversight is at NCTR or if it's at CDER. These are very important projects. I think the whole success of this subcommittee, regardless again of the oversight, is going to be based on these projects moving forward.

I have been very involved with the Product Quality Research Institute, which I think some of you all know about. We've had a very, very difficult time moving forward with our projects. You all are actually farther, in some ways, with the two projects you have in identifying what needs to be done than we've been able to get to at PQRI. And still, it's important to think through all of the issues and all of the process and stuff, but that's not the important thing. That's not where the focus needs to be. The focus needs to be on these two projects, getting them done, moving ahead.

And I think a lot of the questions and issues that have come out here today will sort of sort out as you do some of this stuff. As you look for dollars, as you identify how the dollars are going to get, I think that automatically you're going to see the interactions that need to take place, whether it's with the industry or whether it's internally within FDA.

So, again, we need to go back. Jim and I, Dr. Casciano need to talk more about the process, the issues, and make sure that they're in line so that we can ensure the success of the subcommittee regardless of oversight. Again, I really think it's necessary for you all to focus on the projects and getting them done.

So, unless Jim has any other thoughts on where we need to go from here. Again, I think a lot of things have come up that have been very important. So, I appreciate the conversation.

DR. DOULL: I think that's a very good point, Helen. We can't solve all these immediately, but at least we now know what we have to get to and how to do it.

One thing about getting involved with NCTR, I was involved with Bern Schwetz a lot when he was the director down there. Part of the thing is just talking to those people. They need to hear what the concerns of this subcommittee are, and I think that is an important issue, is to take that message to them. We need to figure out ways in which to do that.


DR. REYNOLDS: Just a quick comment.

DR. DOULL: We need to have the public discussion.

DR. REYNOLDS: Just one quick comment. I think that we can't solve the problems here of where this reports, but I think what we can do is raise concerns that we have. I've not heard any major concerns here raised by any of the committee members on where this is reporting to, but I think it does have to be an administrative decision for you all. So, I think that was to me the outcome of the dialogue. We have concerns, I guess, collectively as a group where this group lies administratively.

DR. DOULL: So, I will assure the chairs of the working groups that we're in your corner and we're going to figure this out and help you in some way as best we can.

We need to ask if there is any public discussion. Have we had any requests?

MS. TOPPER: We've had no requests.

DR. DOULL: So, we just need anybody who has a burning concern a chance to air it. I don't see any burning concerns.

Are there any other issues that we need to do, Helen, for this group?

DR. HOLT: Just a very brief point. But being from a corporate background myself, I think of these things. Wherever this committee sits and whatever we do, particularly on the vasculitis program, I've understood, if it is going to succeed, there will be new things that will be found and whatever is done, whatever mechanism is taken, I urge everybody to be cautious that what could happen -- it's happened before -- is that through the great skill of publicly funded things, that if it's exclusively publicly funded, it actually ends up making important information that is by definition not a good product.

And that's one of the things on a biomarker, what makes for an ideal biomarker. The very last thing on there is that it's commercially viable. And there are certain things that you guys you can do inadvertently that will make it so that it's not commercially viable like a full disclosure of information before there can be intellectual property taken and so on. So, in the background, just one of the things that I would encourage is if there's funding mechanisms or whatever, that there be some sensitivity to make sure that whatever is done in finding new and important things, that it doesn't sort of get in the way of eventual commercialization, which frankly is not part of FDA at all, at least my understanding.

DR. DOULL: Yes. That's a legitimate concern. I think the subcommittee needs to keep that in mind.

Any other final points? Jim, do you have anything more?


DR. DOULL: The parent committee is meeting the end of this month, but we are not scheduled to report to the parent committee at that meeting, as I understand. The agenda came out the other day, Helen. There was nothing on there.

MS. WINKLE: We will have some time set aside just for reports back from the subcommittees. I think as you all know, the advisory committee actually has three subcommittees. It's getting ready to take on two more. So, it's a very active committee. We're taking on a subcommittee on emerging technologies and manufacturing, and we're also taking on the subcommittee for risk management for the center. We're going to set up a subcommittee within our advisory committee. So, this is going to be a very active advisory committee. So, we will have a time set aside at each meeting where we report back from the subcommittees.

DR. DOULL: Well, on your behalf, Gloria and I will give them a quick report on the progress at this meeting.

MS. WINKLE: I appreciate it. I think it would be very helpful to update them just briefly on where the two working groups are and to get their buy-in for continuation of the efforts that are being put forth.

DR. DOULL: No other comment?

(No response.)

DR. DOULL: Thank you. We are adjourned.

(Whereupon, at 11:42 a.m., the subcommittee was adjourned.)