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

Drugs

  • Print
  • Share
  • E-mail

Informal Discussion of PET Drugs: Approval Procedures - March 22, 2000

DEPARTMENT OF HEALTH AND HUMAN SERVICES
FOOD AND DRUG ADMINISTRATION
CENTER FOR DRUG EVALUATION AND RESEARCH

PUBLIC MEETING ON THE REGULATION
OF POSITRON-EMISSION-TOMOGRAPHY PRODUCTS

Wednesday, March 22, 2000
9:05 a.m.
Goshen Room
Holiday Inn Gaithersburg
2 Montgomery Village Avenue"
Gaithersburg, Maryland

PARTICIPANTS

PET PARTICIPANTS

Jorge Barrio, Ph.D.

Peter S. Conti, M.D., Ph.D.

Robert Ferris

Jennifer Keppler

Jerry Kuhs, R.Ph.

Dennis Swanson, R.Ph.

 

FDA PARTICIPANTS

Jane Axelrad, J.D.

John Friel

Florence Houn, M.D.

Ravi Kasliwal, Ph.D.

R.K. Leedham, R.Ph.

Eldon Leutzinger, M.D.

Patricia Love, M.D.

Cecelia Parise

P R O C E E D I N G S

MS. AXELRAD: I would like to welcome everybody to our latest public meeting on the regulation of positron emission tomography products. I have lost count of the number of meetings that we have had. I usually say "the second," or, "the third," or whatever, but I have now lost track of how many we have held.

Before we get into this, I would like to just talk about some logistical things. First of all, I apologize that we are set up in this sort of formal table arrangement. I would like to have been able to have it set up so that we could have more of a real conversation, but it is awfully difficult to be able to converse with the people at the table and, at the same time, have our faces to the audience.

So that is sort of the way this is set up.

Since we met the last time, I think that we have made a lot of progress, a giant step forward, in fact, because the two documents that we have been talking about for over two years now have finally emerged from the process.

We published a Federal Register notice announcing our findings with regard to the more commonly used PET drugs and indications and invited the submission of applications. We also issued a draft guidance document that describes, in some detail, how to complete either a 505(b)(2) or a 505(j) application for those PET products that we addressed in the Federal Register notice.

The purpose of the meeting here today is to get comments on the draft guidance document. We felt that it was important to get a draft out on the street so the people could see it and to, then, hold the public meeting so that we could get comments on the guidance document.

We are very interested in hearing what we can do to make it better, what problems we may not have addressed in the guidance document and we intend to revisit it and take into account what we hear, both at this meeting and in any comments that are submitted in writing.

We will certainly review the transcript and take notes from what we are told at the meeting, but we also welcome either people here or anyone else to submit comments in writing for us as well. I know there are a lot of people who could not be here and who have comments on it. So we will be reviewing both the transcript and the written record.

Dr. Barrio, did you want to say anything?

DR. BARRIO: Good morning, everyone. I would like to start with a light note, if I could. When I was a kid, I used to go to soccer games with my father. I did that for fifteen years and I really had a lot of fun at the time and I learned a number of things.

I learned that soccer is a great sport, probably not well appreciated here in America but it is a great sport, that the spectators and players don't like the referees to produce bad calls. At the time, anything we had at hand, tomatoes, oranges, eggs, whatever, would end up in the field.

I think that is true for any sport, realistically. Those who like basketball or football or whatever realize that we like to see the game. We don't like to see the referee. The referees are there to help us, help the players, help the spectators to have a better game and to enjoy the whole process.

Regulatory agencies are like referees. When they penetrate too deep into the game, we are ready for tomatoes, eggs, oranges and whatever. The FDA has received quite a few of them in the last fifteen years.

I think in the last two years, we have made a significant amount of progress because things have changed, I think, not only because of FDAMA '97, but also because there is a perception that PET right now is a well-demonstrated technique. It is very safe, has reached the public and it is time to give it life at that level.

This document, by the way, as Jane has indicated--this document and others--is the result of a serious effort by the community to address very, very important issues that needed to be addressed to get this somehow off the ground. I think the technology issues have been addressed quite well. I think everybody is very happy with each of the clinical indications. Finally, there is good recognition of the value of PET there in medicine.

I think recognition for the process that needs to be applied, to some extent, has been accelerated and facilitated for industry but there are some items that yet need to be addressed and some of the them are of significant concern, that we have collectively and we would like to discuss here.

For example, this document is called "Guidance for Industry." We have received phone calls from physicians who say, "Does it apply to me?" And we have to say, "Yes; it does apply to you," and whatever. "But we are not industry." And this is true.

Academic institutions and hospitals are not industry and I very distinctively remember a conversation I had with Chancellor Young at UCLA many years ago. He was the chancellor for twenty-five years. Five years ago I had a discussion with him about the issues of regulation for PET, et cetera.

He said, "Do you know of anything that UCLA manufactures?" I said, "No. The university doesn't manufacture anything." "Do you know, in your experience of any university that manufacturers anything?" I said, "No; I don't know of any university that manufactures anything."

Certainly, and this is not because it is the opinion of Chancellor Young, I think it is an opinion well respected by a group of people, we are not industry. Academic institutions, I repeat, are not industry. Then this document should have been Guidance for Industry, Academic Institutions and/or Hospitals.

Also, FDAMA mandates that the FDA recognizes the difference between industry and academic institutions and hospitals. We have briefly addressed this issue at different opportunities. I guess the last time was when we discussed CGMPs. Certainly, that would be the time to really go deep into that issue when we discuss CGMPs and we discuss the guidance.

But I think it was a great opportunity here to introduce the topic and to introduce the concept and, even though, perhaps, the elements for discussion of that difference were not there, the introduction of the topic, I thought, and we thought, was absolutely necessary for everybody to understand where we are going with all this.

There are people who see this as a pervasive tendency of the agency to be the whole FDA. Somebody perceives this as being an issue that needs to be addressed. We would like to believe that this is an issue that needs to be addressed.

I think this is to us, today, one of the most important concepts that we would like to bring forward. There are some items of discussion here that I guess my colleagues are going to bring, too, but that concept--honestly, when I saw this title, it bothered me and probably it bothered a lot of people, too.

It is more than anything else irritating. It is not so much whether academic institutions or hospitals will be doing this or that, but, rather, why are we still called industry when we shouldn't be called industry.

I think that brings--this is deja vu for many because we are coming back to the same kinds of things that we were discussing years ago.

But I am confident, and we are confident, that this is no more than something that can be corrected easily and something that can be addressed easily and this is our hope and we are intending to address these issues this morning.

Thank you.

MS. AXELRAD: Thank you. Why don't I address it first off. First of all, it can be addressed easily. The reason that this title is this way is that the Center for Drugs has been issuing guidance for many years.

We issue two types of guidance, guidance for industry which is generally a generic term that we use to apply to anybody who is regulated by the FDA. It applies to the compounding industry. We call people who are compounders the compounding industry. It is the pharmaceutical manufacture industry. It is a loose term that we use to distinguish guidance for industry from the other kind of guidance we issue which is guidance for reviewers, which is internally what the agency reviewers do.

These are the terms that we have applied to every guidance we have issued in the center. We were not sensitive to the connotations, I think, that are attached to that title and I think it can be easily corrected when we issue the final document.

I don't really want to get hung up, I think, on the title of the document. What I really need to understand and what we really want to hear from you today is what real differences in the area of the receipt of applications, the approval of applications, for these particular drugs and indications do you want us to put into our guidance that distinguishes the academic or the not-for-profit medical centers from the other people who might be submitting applications.

I agree that we will also be getting into this in more depth when we do the GMPs, but I would be interested--we have discussed, in a brief way at other meetings, but what we really need to hear from you today is how the procedures that we have for submitting applications and getting these drugs and indications approved need to be made different depending on whether you are an academic medical center or you are a for-profit commercial entity.

Then we will need to go back and explore the flexibility that we have under the statute to make any kind of differentiations that you suggest to us because, in fact, we are limited by the statute.

Congress, when it passed FDAMA, indicated that we were still supposed to be regulating the PET industry under Section 505. Now, we have some flexibility, once you get beyond the statute, to make some distinctions. In fact, FDAMA taught us to do that but it didn't exempt anyone from the requirement of Section 505 of the statute and our application requirements are derived from that.

When we talk about 505(b)(2), that is a section of the Federal Drug and Cosmetic Act that defines a certain kind of an application where the studies were not done by the application. When we talk about a 505(j) application, that is another section of the Federal Food Drug and Cosmetic Act that defines how applications are submitted, abbreviated new drug applications are submitted.

Those statutory provisions were not changed by FDAMA so we have to start with those. After we get below that into specifics, we need to see what you would like to have be done differently to make those distinctions that the title, obviously, didn't recognize but that we may be able to look at when we revise the guidance.

DR. BARRIO: Do you guys want to comment on that?

MR. SWANSON: As an advisory committee, I think one of the concerns and problems as advice to the FDA is you read through the background material in the guidance document, both in the Federal Register and the additional guidance document, the background material takes several sections of FDAMA and uses those sections and presents those sections to support the position of the FDA.

Yet, the guidance document, the background section, does not address it at all. The requirement of FDAMA, "The Secretary of Health and Human Services shall take due account of any relevant differences between not-for-profit institutions that compound the drugs for their patients and commercial manufacturers of drugs."

As advice to the FDA, it is imperative in your background document that you address the considerations that have taken place, or that are taking place internally, related to that mandated requirement of Congress on the FDA.

I think, to your own benefit, you have got to discuss it. You can't just ignore it. It is conspicuously absent in your discussion of the background information. You have got to address it. Congress has mandated that you address it.

MS. AXELRAD: We can fix that by addressing it, but the question is what are the relevant differences between for-profit and not-for-profit entities as it applies to the procedures for submitting applications to the agency for these three PET drugs and for these three indications because that is really what we are focussing on at this meeting.

You need to tell me what you see as the difference and how those procedures should apply differently and what are the relevant differences and how should these procedures be applied differently for those two different types of entities. Otherwise, we can put it in and we don't see any relevant differences.

I need to hear what you think the relevant differences are and how we ought to change what we are doing to take that into account.

MS. KEPPLER: One way to approach this might be to go through the guidance as well as the Federal Register announcement and talk about concerns that we had and, as well, bring up, as we see them, any areas that we feel might be differentiated.

DR. CONTI: I think we have to have an open discussion about this, Jane. It is very clear. We have been avoiding it for the last several meetings and I think it is reasonable to include it. We should just have a good hour discussion about examples and how we see it as individuals and get some input from the audience as well, here, and then maybe condense that into something that you can work with for your document because I think that process has just been lacking.

I have a top-level question in terms of 505 and I need that to be clarified. Does 505 provide a mechanism to either A, discriminate between the entities and does it allow modification of the actual application process? Are you locked into a very specific application process?

MS. AXELRAD: I am turning to our attorney from the General Counsel's Office, David Horowitz.

MR. HOROWITZ: Section 505 is pretty specific and it sets forth particular standards for safety and efficacy. There are regulations and guidances that have developed over the years setting the standards for safety and for efficacy. There is not a great deal of flexibility on the 505 standard.

The are different types of applications within 505 that are discussed in the guidance document. To a large degree, the discretion and flexibility has been used in the guidance document and the approach that the agency is taking.

DR. CONTI: But this is specifically according to what is available within the document. In other words, my question is 505(k); can that be created as another application form or can the 505(j) be modified to the extent that we can do that?

MR. HOROWITZ: The 505(j) application has to be a 505(j) application. I will look at Section (k) and get back to you.

DR. CONTI: That is just an arbitrary--I don't know what 505(k) is. I am just saying (j), (k) in a sequence. But the point is, can you take 505(j) and add a new box to the application form?

MR. HOROWITZ: No; I don't think so. There are a finite number of options within 505 unless Congress changes the statute. There is the ordinary 505(b) application and there is the 505(b)(2) which is discussed in the guidance document in the context of literature. And then there is the common generic application and those are essentially the three boxes that Congress gave us.

DR. CONTI: That begs the next question. In the application process, what I am hearing is that we will not be able to discriminate entities because there are no boxes to do such on those forms.

MS. AXELRAD: I think we should spend our time talking about--I don't understand. Do you want to distinguish between entities by saying that some of them don't submit any applications; is that what you are trying to do?

DR. CONTI: No.

MS. AXELRAD: If we could explore how you want to differentiate it, then we could give you a better idea whether the statute would allow it.

DR. CONTI: For example, between an academic institution and a commercial institution; for example, between a person that could be applying for a non-profit versus a profit type of organization; for example, a drug manufacturer versus a PET facility.

DR. HOUN: With respect to what? A user fee? Amount of paperwork to fill out? What is the difference between the two?

DR. CONTI: If it is a name, first of all, then it has certain implications in terms of what the community perceives as a drug manufacturer, for example, versus a PET facility. That could be defined within the guidance. Do you have the flexibility to do things like that within 505 is really my question.

MS. AXELRAD: We can talk about any different entity we want. We can say that there are five different types of entities that might apply for approval of a PET drug. There can be a commercial manufacturer. There could be an academic institution. There can be, I don't know, anybody. You could define however many there were.

But the question is, are you asking us to say that one of those, the commercial manufacturer, would submit an application and the others would submit something different, or nothing at all?

DR. CONTI: No.

MS. AXELRAD: We can say that there are a lot of people out there who would apply for this, but the question is do you want the requirements to be different among those groups?

DR. CONTI: No. I am just asking can the application form be modified to discriminate, at the front end of the document, what the entity is that is applying for the application.

MS. KUHS: Maybe I can clarify that a little bit. I think several meetings ago we had suggested to the agency that it begin with the registration process, itself. It was kind of neatly tucked away in the back of all of the guidance, but we didn't really address the registration process.

We are still required to register as drug establishments. I think that we had suggested that if FDAMA was going to really create PET drugs as a separate class of drugs that maybe that recognition ought to start with the registration in that, rather than being a drug establishment, we are a PET establishment, and that we begin with the PET establishment and then subdefine that between the profit and not-for-profit and then see who fits into those categories.

I think it is unfortunate in how the ultimate legislation was written, profit versus not-for-profit, when, clearly, the spirit of that was the academic research-based institutions versus a commercial distribution establishment. I think we are struggling because we can't really fit what the spirit and the intention of the legislation was into the actual wording.

MS. AXELRAD: That is a real problem because we have the statute as it was written and we are bound by what it does and what it does not do. We can't change other sections of the statute that were not changed. I think we have discussed a bit at the last meeting the issue of registration and the question of whether FDAMA and what it said about PET changed the registration listing requirements of Section 510.

David may want to comment on that. I know that we talked about that at past meetings. But the bottom line is FDAMA did what it did, and there are certain things that it did not change. We cannot change that. We can't change the statute, ourselves.

We can try and work within the statute. I think we have gone a long way. One of the things that I was sort of a little confused about here was that we have been talking about this and we have been talking about this approach for over two years, now.

We have indicated at public meetings what direction we were going on this. So I am a little sort of surprised that now, instead of focussing on the specifics of the document, we are raising all these issues. I really don't understand what you would want us to do differently in the way something would apply to an academic medical center as opposed to the way it would apply to a commercial manufacturer, other than the registration issue which we have talked about before.

DR. BARRIO: Jane, we have discussed many times that one thing the PET community wants, and is actually adamant, is that the quality of the PET pharmaceutical to be given to the patients. It doesn't matter what the source is, maybe from academic centers, industry, the hospital or whatever, it has to be of the best possible quality and the standard has to be the highest.

Then it is not a matter of lowering the standard or quality of the pharmaceutical that we are discussing here but rather the legalities of not calling industry what shouldn't be called industry. That is essentially what the fact is.

As I said before, it is, more than anything else, irritating, more than the fact of looking for different ways of doing things. I think the process, in general, of an NDA or an ANDA application that you developed except for some questions that may exist--I thought was well done. It was carefully considered.

We worked for two years with you and we have done this and I don't think--we are note talking about the good cause here. We are talking about the bad cause. And I think it shouldn't be confusing to the agency. I think that is an important point, a very important point, because that differentiation, as Dennis has indicated, is central to the law.

And I understand. We understand what you are saying that you have difficulties defining within the statute that you have to make these differentiations. That is why, I guess, Peter, Dennis, myself, whatever, the whole audience here, is trying to figure out how to simplify this. It has to be addressed somehow in this document and we feel, collectively, as a group of people, the society that we somehow represent, when physicians come to us and say, "You know, are we industry?"

This is a stupid question, but we feel terrible to say, "No; you are not. But then this applies to you anyway." They don't understand anything about law and we don't either, probably.

But I think it has to be addressed. Even if there is an issue that cannot be addressed, then it should be said here it cannot be addressed within the statute, but some indication, something, some process in which everybody will understand what we are talking about.

Otherwise, we are back to explaining to everyone what it is and different people explaining the thing a different way and different people understanding the thing in a different way, too. I think we really need to have everybody have a very clear description of this.

I don't think we are talking about different ANDAs. The mechanism, the different way to synthesize it, to synthesize the radiopharmaceutical, different quality controls; we are not talking about this or we have addressed these issues.

DR. LOVE: In listening to what is being said, it sounds like part of the concern is semantics, if you will. What do we mean when we say "industry." We do have specific language that we are responsible for using. When Jane talks about the guidance for industry, the Center is only allowed, at this moment in time, to issue something with this title regardless of the persons to whom it may be addressed in terms of the different entities and what is being recognized.

I understand what you are saying, I believe, in terms of how that term might be perceived. It may not be understood that we had some constraints. When Jane is saying we weren't sensitive to the fact that that term might be perceived in a different way, then that is part of something that we would need to explain.

On the other hand, as I listened to your comments about some of the issues, the registration, CGMPs that we still have yet to finish, that is certainly a place where we recognize that there may be some issues that need to be clarified between profit and non-profit entities. We may need, in this document, to give some history of how we arrived at where we are, why this is being issued at this point, why it applies to all to make sure that all the products are comparably prepared and used regardless of whether you are a non-profit or a profit entity that is making the drug.

I think we are just saying we all agree that is important. So if what you are asking for, at this moment in time, the fact that we take a bit more time in the guidance to try to clarify the deliberations that we went through to get to where we are, why we are using certain terminology, how it would apply or not apply to different groups or entities and what this document is to be addressing and the fact that, perhaps, other differences, if identified, between commercial and non-commercial would be addressed, perhaps, in the CGMP document. Is that essentially what is being said?

MR. SWANSON: I think you have to address it; okay? You can't just ignore it. It is conspicuously absent now. It stands out that you are not addressing it. That is a major concern. Getting back to registration; we have heard your lawyer tell us about registering as a drug establishment. This is no big deal.

This is a big deal to the nuclear-medicine community. This is a big deal. Understand that. You can explain it away all you want, but listen to us. This is a big deal.

I work at an academic medical institution. It is a hospital. We are not a drug manufacturer. That is not our mission, nor will it be our mission. There is a perception that registering as a drug establishment makes us a drug manufacturer. It is an issue. You can argue it away. You can say anything. But this is a real issue to the PET community.

MS. AXELRAD: Dennis, could I stop you right there and ask you to please elaborate on what your concerns are about why it bothers you or what you think flows from the fact that you are being designated, if you were to be designated, as a drug manufacturer.

What do you think flows from that?

MR. SWANSON: Let me give you a dollars-and-cents example. We are an academic medical institution that has a PET facility, $15 million investment, in 1994 dollars. Our PET facility was established specifically with the primary emphasis on doing research, involving positron emission tomography.

It is not the mission of our institution to be a commercial supplier, a commercial distributor, of PET agents. That is not the mission. Let me finish. But we register as a drug manufacturer and now we move to having to comply with CGMPs.

You can say what is the difference between USPs and CGMPs. The difference lies in the degree of documentation--you can go that route. Let me finish. Assuming you go that route, it requires an increase in documentation and record keeping, validation requirements, that are being imposed on us through this process.

So now let me give you a dollars-and-cents example. To comply with those requirements, versus what I am doing today, let's say we have to hire one additional person to make sure the standard procedures are written, standard testing procedures, t's are crossed, the i's are dotted, make sure that we can sustain a prescriptive-based inspection of our facility rather than a performance-based inspection.

Let's say it is costing me $100,000 to hire this additional person. Now, my PET center is not going to distribute. I have got a multi-million-dollar investment. I would like to be able to use that investment to at least provide dosages of FDG for the care of my own patients.

Let's say I do a thousand doses a year for my own patients. So I take this $100,000 increase in investment, a thousand doses per year--that is about twenty per week which is where we are at--that means that my cost, my increased cost of compliance because I am now in this realm, comes out to be about $100 a dose.

I have got a commercial supplier of FDG down the street. That commercial supplier, operating under the same standards I do, has the same requirements, they invest $100,000 to address the compliance issues, they are doing 5,000 doses, because they are a commercial supplier.

I have chosen not to do that. Their increased cost comes out to be $20 a dose. So now I have an $80 differential between my operation and their operation per dose in my cost because I have chosen not to be a commercial distributor but am subject to the same requirements.

Yet, when I go to bill this radiopharmaceutical to the insurance company for providing patient care, they are going to pay me at a rate of what the commercial manufacturers are charging. So now you have put me at a major disadvantage, a cost disadvantage, simply because the mission of my facility is not the commercial distribution of radiopharmaceuticals.

That is a real issue, folks, and you have got to pay attention to that. You have got to address that. I know you find this hard to believe. There are people out there that are doing things not for entrepreneurial purposes. We are not commercial distributors.

You deal with people that are doing this on a for-profit basis every day. What you have got here is you are trying to regulate people with a different mission and you don't seem to understand that.

We keep bringing it up and we keep bringing it up and we keep bringing it up. But that is fundamentally what we are talking about here.

DR. CONTI: Let me just add to that in another form. The universities across the country are very concerned about registering as drug manufacturers. There are liability issues that are not being addressed. As Dr. Barrio mentioned a few minutes ago, his chancellor doesn't know of any universities that are registered drug users. We had the same problem at our university.

There is a connotation, stigmata, whatever you want to call it as well as some real dollar terms that Dennis had delineated and some liability issues that academic centers and hospitals are going to have to deal with that are not traditionally dealt with outside of the industry.

How are we going to face that? So these are the types of things that have just been, to date, ignored in this process.

MS. KUHS: I think there is one other thing that we could add from a commercial standpoint on this and I think that Bob will probably agree that if the commercial drug industry, traditional drug industry, were to look at the changes that have been made in the application process, the simplification of the application process and the really, step-by-step procedures in the outline, they would all be delighted.

I think, from my standpoint, I would say I am not real disappointed in that. But what I am disappointed in is that PET is an emerging industry. What we, as a committee, and you, the FDA, decide today is going to decide what happens to this industry from here on out. If we stifle the academic institutions where the research is coming out for these products--it isn't coming out of a Dupont or Merck or anyone like that. It is coming out of the small academic institutions.

If we stifle that today, we stifle the entire industry from here on out. I think that it is incumbent upon all of us here to find a solution to that problem.

MR. FERRIS: I think the issue, with respect to the check box "drug establishment" is primarily a categorical one. In our conversations amongst ourselves, there are many times that we ask what latitude do we have under this law, what latitude does FDA have.

As you are bound by 505, so are we. What we sometimes don't know is what are the options that we may have to take an alternative pathway that reflects our business, the PET business, the radiopharmaceutical business in a more accurate manner.

It is not to avoid regulation. It is to be more consistent with what our science is about. So I think, with regard to the registration issue, it primarily sets off a category and if we had a check box that said, "PET tracer, processor" or something to that effect, that sets a distinction. That distinguishes this whole industry, if you will, with respect to the FDA process, if Section 505 can provide for that.

MR. HOROWITZ: The registration provision is actually a separate provision from 505. I do think we have some flexibility to distinguish for registration purposes the title that applies to the establishments that are registering. We will look into that.

It sounds like there is an enormous amount of sensitivity to the symbolism of being called a manufacturer, of being called an industry. We are going to go back and do everything we can to be sensitive to those subtle distinctions.

Registration and the title that is given to the party registering might be an area where we have some flexibility. It is not going to have substantive consequences for the Food Drug and Cosmetic Act, but I understand there are other issues that you are talking about.

We will look into that. Registration, as I said, is a separate provision of the Food Drug and Cosmetic Act. It is separate from GMPs. It is separate from approval. It is separate from just about everything else.

Earlier, it was said that if you register, then CGMPs apply. The two provisions are really independent. Whether you register or not does not determine the applicability of GMPs. The statute is written in a way that, if it is a drug, essentially GMPs apply to it.

The FDAMA provision didn't distinguish registered, non-registered, establishments. It didn't even create an exemption for compounded PET drugs. It did require FDA to establish GMPs for PET drugs without distinguishing how the PET drugs were produced or where they were produced.

But there is the statement in there that the agency should take into account any relevant differences. We have grappled with what those relevant differences would be and we have had an incredibly difficult time.

We understand that the industry--well, that would be a bad choice of words--but the PET community grappled with this before the legislation was enacted. I understand there was a lot of back-and-forth about trying to draw the line between what would be a commercial or even an industry establishment and a not-for-profit and academic establishment.

Because the PET community, itself, was not able to come to terms with that, as I understand it, the legislation has this somewhat ambiguous statement leaving it up to HHS and FDA to determine what, if any relevant differences there are.

We are seeking guidance from you, still, to help us figure out what we can do within the limited framework we have of the Food Drug and Cosmetic Act as amended by FDAMA, and we appreciate any help we can get on that.

DR. CONTI: If you listen, we have helped you. We have given you some of our concerns. Please take them into account beginning with the registration process because it is a first step in going to the unique nature of this business. That is what we are trying to do here. We are trying to give you a top-end way of discriminating PET from other drugs and then, from there, you can begin to derive a series of consequences of that renaming of business that may be more applicable to meet the requirements of the law.

It may require the complete discarding of CGMPs from other drug processes in rewriting CGMPs that are specific to PET, as we have been trying to do, but you might have that top-end to work from because we have now classified the business or the drugs as different than the traditional mechanisms.

So listen. We are giving you the way to do it, if you would pay attention to that.

MR. HOROWITZ: I assure you I am paying attention and we have been paying attention for quite a while. We have certain restrictions that are built into the statute. We are trying our best to deal with those restrictions and take into account your concerns.

Your last statement, I think what you are saying is that, by starting with the registration, that we can use the distinction we develop in registration to develop exemptions from the other parts of the law. What I am trying to explain is that the registration distinctions don't give us the authority to create exemptions of modify the other legal requirements. FDAMA did not provide us with that authority.

MR. FERRIS: Not necessarily exemptions. By establishing a categorical difference for the drugs, that does not necessarily categorize it as not a drug. It still remains a drug. It is to seek alternative pathways of dealing with the regulation of these particular classes of drugs.

It is not to necessarily avoid or to go for exemptions. I will reserve that with respect to waivers, or with respect to the normal provisions that exist for exemption.

But that is not the intent here. Many of our discussions are what latitude--and we don't have a lawyer in the group--but what latitude does FDA have to be able to seek alternative means. But it sets up a category a little different but it does not make it not a drug.

DR. LOVE: The difference is that you want to be able to say that the different types of entities that might be producing PET drugs, when you register, if you identify yourself as one or another type, commercial, non-profit, whatever category might be appropriate to put there, then the steps that would follow from that might then lead to CGMPs or something. Is that what your thought is?

MR. FERRIS: I think that setting up the category--this is the PET group, this is the PET community, whatever, it is--we still are left with, and I really believe it is a very difficult distinction, that the law requires FDA to make, and I certainly don't want to necessarily attempt that, today, anyway. But the point is that I think that distinction still could have, underneath that category, to some extent, and that extent would really have to be tested in the context of 505.

DR. LOVE: But, then, after that distinction is made, then the assumption is that, based upon that distinction, then if we are either looking at CGMPs or something else in the process, whatever--at this moment, it would remain unnamed--but that that would be some way of deciding what types of relevant approaches might be taken for one institution versus another?

Is that what is behind this?

MR. FERRIS: I would say that would be fair; yes. That's correct.

MR. SWANSON: Or distinguishes these entities as, perhaps, being subject to a different set of CGMPs from the beginning.

MS. AXELRAD: Isn't that what we have been doing? How many meetings have we had where we have been writing--I sort of don't understand. I am having a hard time even following the discussion. That is what we have been doing.

MR. SWANSON: But we don't know what your CGMPs are.

MS. AXELRAD: We are going to have another public meeting to discuss those as soon as we get along--

MR. SWANSON: But that relates to what we are talking about here and the concerns of the PET community, getting back to your initial issue. I am a little curious about your interpretation of FDAMA. Again, I am not a lawyer--it does seem to me, though, it says, "The Secretary of Health and Human Services shall establish appropriate procedures for the approval of PET drugs pursuant to 505 of the Food Drug and Cosmetic Act."

Basically, you are interpreting that as saying that the submission of ANDAs and NDAs, the actual submission process, need to follow 505, basically.

MS. AXELRAD: Yes.

MR. SWANSON: But we do still have the flexibility under FDAMA to develop a different set, or to codify a different set of CGMPs; am I not correct?

MS. AXELRAD: FDAMA specifically says, "Develop special CGMPs for PET." It specifically says do them. And that is why we have been doing them and are doing them.

MR. SWANSON: I think that is all we are really saying is, perhaps, a different registration category, then, is viewed by everybody as meaning that, from a CGMP standpoint, we are going to be looking at those differently from traditional manufacturers.

It is sort of, as was pointed out, a distinction as a special drug category.

DR. HOUN: I think we really welcome those comments. In the discussions previously, we had not heard where you would think, in personnel or in the quality-control parts, those differences should be. I think we have been playing to a common denominator of academic institutions, but maybe there are different things you would like to bring up in future discussions where they could be separated more.

I guess we thought we didn't hear that difference. But I think now, saying that we understand that CGMPs are a particular concern, then when we review them with you, we can pay special attention, in terms of when you are recommending for what category of institution. We can take a look at all those sections and look at that if there should be varying things.

I think what we are interested in is before you are some proposals for application for getting approval of FTG ammonia and sodium fluoride, and should there be different application procedures? Should user fees be treated differently? Should the forms be treated differently?

I think we want to hear your views on this before--you have our ideas. We were not sure there were application differences, relevant differences, related to application. But, perhaps, there are and we would like to hear them.

MR. SWANSON: My gut reaction, my personal feeling, is I don't have a problem with submissions of NDAs or ANDAs, especially under the mechanism that you have developed which makes it relatively simple. Please understand, I think we all appreciate that you have simplified the clinical indications part of it. You have simplified the application. I think those are wonderful efforts.

I think the issue for us in academic medical centers is probably not the submission of NDAs or ANDAs, but it is ultimately going to be the CGMP issue. I realize that you don't have that option, but let me bring up an example of where our problems are with what you are proposing.

If you look at your formats for the submission of CMGs, if I look at those formats, they look a lot like CGMPs, to me. You are requiring references to standard testing procedure, you are requiring references to validation procedures, et cetera, so your current standard formats that you are asking people to submit as part of their NDAs today, are those standard formats your interpretation of USP standards or are those standard formats basically "your proposed CGMPs" for PET agents.

This is very important that you answer that question because if you are saying that what you have proposed here in your sample submission is your interpretation of USP, then when you go out to inspect facilities over the next two years, we have got a big problem because our interpretation of USP and what we are seeing in the submission format are very different.

So if you go out and inspect people looking for standard testing procedures, looking for the extensive validation that you have outlined in here, you are going to have people in violation.

So, right there, we are talking about, in your guidance documents today, your sample submission formats look a lot like CGMPs.

MS. AXELRAD: We put those out there at least once and maybe twice. And we got back feedback from you all that you thought that the formats were basically okay.

MR. SWANSON: No. Time out.

MS. AXELRAD: You did not raise the issue before, at least I have not heard the issue raised before, that you felt that these formats were too complicated or that they were prejudging what we were going to do for GMPs. This is the first time that I have heard that raised. In fact, we got back that you thought the formats were very well done. You had very minor comments on it.

MR. SWANSON: I think if you go back to the last meeting and review the minutes, you will find out that there were several concerns expressed that these formats looked very much like CGMPs. I think that might have been an opening comment in our letter back to you, "They look a lot like CGMPs."

Go back and look at it.

DR. CONTI: Let me just comment on this. I think we are heading to a log jam on CGMPs. It is very predictable. We are going to get into a very detailed discussion as these come out and we deal with them. We have had a couple of rounds with them already, but I think, again, the issue boils down to trying to create a mechanism up front that will allow us to recreate these CGMPs in the context that would be most beneficial to the broad PET community.

We do not want to change the quality of the product for any given entity, the quality of the products we want to be the same. We want to move to outcome-based as opposed to prescriptive-based approaches in CGMPs. I don't know how else to say that because it is very specific.

The PET community will not accept a prescriptive- based CGMP plan. So what we are trying to do is identify, at the top level, some mechanisms to discriminate these entities, to create a basis for which people can maybe, then, look at the issue of profit versus non-profit based on the input and based on the discriminations that we have made in these entities.

And then, at the end game, when it comes down to the nitty-gritty, which is going to be CGMPs, we will be able to know that this business is different than a traditional drug manufacturer, this business is based on outcome as opposed to prescription and, therefore, those CGMPs need to follow suit with what has predated the discussions of the CGMPs.

DR. LOVE: I think we understand the proposal, or at least the concept behind what you are talking about. Maybe just a little bit of perspective on the goals of today's meeting. The reason I say that is a lot of what you are addressing is the CGMP issue which will be a separate meeting.

There will be separate documents. There have already been. We really see that as a different venue. Also, we don't have all the persons here today to go into detail about the CGMP aspect of things.

Nevertheless, I do understand some of your perspective in the sense that maybe there is something that is contained in the check box or in this form that might lead to that. Certainly, we can consider all those things and, as David said, we can look further into whether there is a way to make some distinctions.

Having said that, is there something else, then, that is part of these particular documents that are published that also needs to be considered, changed, adjusted? Our goal today is really to seek your input on these particular documents so we can determine what may need to change in these documents.

I think we have heard you on the issue on the registration check box. Is there something else that we need to address in some way or another that leads to-- you were talking a bit about the CGMP potential implication from the CMC documents.

Could we go on to some specifics about these documents? Maybe we can, as someone said earlier, just go from the beginning through these documents to find out exactly where the issues are that need to be addressed. I think that would be productive.

DR. HOUN: Did you want to see CMC sections for FDG and kind of go through where you think the imposition of requirement for filling out would then impose more rigor than you had wanted or where there should be a difference? There should be more rigor in certain facilities but not in others?

MR. FERRIS: I would like to go back for one second with regard to the application process and that is the distinction that comes at the front end as to whether or not one goes with an ANDA or an NDA. Clearly, I think the way it is put forth is if there is an NDA out there with the indications that you want, then you can go in with an ANDA, if there is a legend drug.

What is confusing to me is the first thing that FDA does is they approve several indications. We thank you for that. That is excellent. By the way, I think the documents coming from industry, the application documents, are fundamentally outstanding in terms of being able to fill out these applications.

But the confusion that I have is you have--FDA has, on its table, one approved NDA for FDG with one indication. Then FDA goes and studies the literature, for whatever reasons, and says, "These indications are appropriate based on the scientific literature to apply to FDG." What FDG?

My interpretation is that the only FDG you have is the legend drug in front of you. Why, then, do you ask for an NDA if a center wants to produce FDG for these three indications--why do you ask for an NDA, and until one is approved, then we can't get to an ANDA?

Why doesn't FDA simply apply those to the drug that it has before it and then cause everybody to go in with an ANDA.

DR. LOVE: Some of those are the issues that we talked about before in earlier meetings. They are just mechanics of the process. Certainly, the current NDA holder can simply submit an efficacy or labeling supplement that references the Federal Register notice and complete their application.

But we are trying not to hold the rest of the community to what one entity does. So the other entities in the PET community can also submit their own application. All they have to do is cite the Federal Register notice. They don't have to do anything different.

The distinction between 505(b)(2) and the 505(j) is based upon whether your chemistry, the manufacturer process that you are using, is similar to the existing NDA or future NDA, or not. And that is what helps make that distinction.

DR. CONTI: I am not sure I understand whether you have answered his question or not because the question on the table, I think, is was it not the existing NDA holder, in effect, the PET community, the one that requested the review of the literature for the expansion and, having done that, shouldn't it automatically be applied to that NDA?

Why would the holder, then, have to request that it be added?

DR. LOVE: That NDA holder is the legal entity. In order for us to actually change the label, the person who is legally responsible for that NDA has to submit an application. Then we would turn around and say, "Yes; it is approved," because it would be submitted.

But we don't have a legal entity that has come forward yet to ask for that. So this is simply a process in order to legally get this information out there.

DR. CONTI: But the request, then, does not have any bearing on this.

DR. LOVE: The legal entity who holds the NDA, that legal entity has to do it.

MR. FERRIS: Specifically methods.

DR. LOVE: Yes; the official holder of-record of the NDA.

MS. KEPPLER: The ICP is not the sponsor of the NDA. We sponsor the DMFs, but the ICP is not the sponsor of the NDA.

DR. CONTI: I understand that. But the request to do the review effectively represented the community. You see where we are trying to go here, obviously, is to try to avoid an NDA application.

MS. KEPPLER: I'm sorry; we don't want to avoid an NDA application for everybody because it provides flexibility for people who can't or who don't want to meet the reference-listed drug as it is. At least, that is my understanding as to why we have that option as well. People can also do a 505(b) if their configuration is slightly different; is that not correct?

DR. CONTI: What we are trying to do is have everyone file ANDAs as opposed to filing another NDA.

MR. FERRIS: At least for FDG.

DR. KASLIWAL: If I can comment; some of the reasons are why we kept 505(b) around was, as Jennifer said, to keep flexibility in the process because to submit an NDA, your drug has to be the same as the reference listed drug which is currently, for FDG, Peoria's.

So if you make a higher-strength drug, more radioactivity in the vial, it will not be the same because the stability consequences and because some manufacturers may want to change their composition and active ingredients. It won't be the same to accommodate that kind of differences.

Really, in terms of what you have to submit, I don't think it changes too much whether it is an 505(b)(2) or an ANDA.

DR. CONTI: That flexibility exists. What we are more concerned about is the use of exclusivity and marketing and things like this that are going to affect the entire PET community. Here we have an entity that effectively holds an NDA, and we went through a review process to expand those indications at the request of that holder, essentially, and now we are told that in order to maintain flexibility, we are going to allow other NDAs into the foray here that could affect operations of all the PET community because of marketing and exclusivity issues.

MS. AXELRAD: We don't have any choice here. Unfortunately, the statute was written that way. We asked for our attorneys to look to see whether the exclusivity provisions would apply because we recognized how complicated it would be if they applied here, but the answer that we got is that they are applicable here and there is really nothing we can do.

It is in the statute. We can't change that, period. It just is required by law.

MS. KUHS: I think the sort of carrot-and-stick approach that you took with the fee and the fee waiver issue, though, as it applies to the FDG has serious complications and implications for any future drugs or new chemical entities that might be submitted as PET drugs.

This is really a very small industry. The user fees, particularly for new chemical entities, are significant, I think well beyond the means of anyone in this room that I know of. I think to apply that sort of carrot-and-stick approach for FDG when it has long-term implications for other drug entities in PET is sort of dangerous to do.

MS. AXELRAD: We tried to confine what we did with regard to user fees to the specific fact pattern that was before us because we felt that these were really unique circumstances in that we had done the literature review, we wanted it to benefit everyone equally.

We knew that it was sort of a matter of chance who happened to come in first and that the first person would have to pay whoever it was--we wanted to create an incentive for people to come in first. At the same time, once one is approved, nobody else has to pay or everybody else can come in as ANDAs and they don't have to pay at all.

So we really wanted to keep a level playing field. We didn't think it was fair to really disadvantage the first person and then everybody else would be off. But then, when we came to exclusivity, we were bound by what the statute says.

But it really is specific to these three drugs in this situation. We don't see it setting a precedent or anything that we would be bound for for any new molecular entities. Those will be examined under the factors in the Prescription Drug User Fee Act for waivers and we will judge on their own merits whether we think they will quality for a waiver or not.

DR. BARRIO: This issue of exclusivity; we understand that you are very much probably stuck with what we have from the legal point of view. But the community has been working for what, probably three years, trying to get these five PET radiopharmaceuticals approved.

For anyone to have the opportunity in some of them to claim exclusivity when this was an effort of the community and the FDA, by the way, to get this done is maybe legal but it is immoral. Of course, both things tend to be put together many times, but I find this astonishing that this may happen.

At the same time, I am worried--we are worried that the effect of this is, or may be, that if anybody could come out and say, "Okay; I claim exclusivity for FDG," for example, even though this may not be possible, then the reaction of the PET community will be, "I told you so. This regulation stink. See the consequences of this regulation. We have to pay now for using FDG, something we have been using for many, many years."

Then it will backfire, in a way, and it will be atrocious for the whole process. I hope we can avoid this and I understand, Jane, that you are very much without relief to have alternatives for that process. But this is going to be very, very bad for the whole process and I hope the community takes a position which respects what we are trying to do here and what FDG is trying to do for the benefit of everybody.

MS. AXELRAD: I guess the question is where do you want to go next? Do you want to get into--I really think it would be useful to get into the specifics. If you want to get into the specifics of the chemistry format and where there are things in there that are troubling to you or if you want to get into the specifics of the application procedures.

I think we need to get into the specific of the documents as to things that you think should apply differently.

DR. BARRIO: I would like to make a comment about the chemistry. I would like to say that the work that Ravi has done, particularly, and you all have done, too, and the staff has been outstanding in many respects.

What Dennis is referring to is the fact that there are certain little components over there that are on the validation side that we discussed several times that, perhaps, could be minimized, reduced, things like that.

We discussed the issue of whether the quality-control procedure has to be validated. Of course. Who is going to use a procedure that is not validated. It doesn't make any sense.

The question is that validation, for me, could mean to be different than validation for the FDA. We discussed that, too. A validation for us chemists and scientists is you apply a procedure, you use a certain column, you use certain chromatographic conditions, you have your rotation factor and you have your standards and you know how the process works.

What else do you need? How many times do you want to do it; 150 times? You know, it is true the column can deteriorate. You can change solvent manufacturers and things may be modified a little bit. Do we need to validate every time, everything?

The answer to this is no. There has to be some common sense in the whole process, too. I think, in general, we have opened these forms to the community and the feedback that we received was very positive for the most part.

There was one issue that I would like to bring back and it is the issue of the fluoride ion present in FDG, something we discussed many times. The last time was, according to the USP standards that we discussed with FDA for a period of a year or something like that, the understanding was that the FDG should be 90 percent pure, I guess. That was the standard of purity or something on this. That was the number; 90.

By implication, it is that the fluoride ion can be up to 10 percent. Ravi was concerned about that because he said, "Well, probably the quality of the image and the diagnostic value of FDG may be reduced." In fact, our own experience of many years has shown that this is not the case.

But the issue of fluoride ions is an important issue because if the specific activity of FDG crosses the line of 300 millicuries per ml, and this is possible with some high-energy targets, then we have a possibility of autoradiolysis, shelf autoradiolysis.

Therefore, the FDG may deteriorate as a function of time. That is something that we also have discussed. Then I understand--I don't remember the number that was put here in terms of the limit, but was it 4 percent, 5 percent. It was something like that.

DR. KASLIWAL: It is 4 percent now.

DR. BARRIO: The question was why 4 percent? Somebody asked me the question several times. I remember submitting to you a paper by Egbert, from Germany, who purposefully introduced fluoride ion in FDG and used it for cancer diagnostics. I don't know how many patients he had but I believe it was 80 with FDG, et cetera--this fluoride ion, I can't remember the percent but I believe it was 50:50 and FDG and used them and compared.

In fact, interestingly enough, they found that the FDG fluoride ion mix was better in terms of a diagnostic, in terms of sensitivity and specificity, than the FDG for the same kind of patient.

Of course, that was one study. We have our own experience at UCLA, something we have done and published--I don't think we published that part--in which we also found that it was at least--at least--we can say it is not a problem, it is not a problem to have some fluoride ion there.

Then, a question I received recently was, "Why not 10 percent fluoride ion? Why do you restrict it to 4 percent?" Isn't that an arbitrary number?

DR. KASLIWAL: No, not really. We have an approved NDA for sodium fluoride. If you consider that the FDG dose is 10 millicuries to the patient, at 5 percent, you will have 500 microcuries which is the effective dose for sodium fluoride for imaging bone.

So the intention was to keep it below the effective dose. That was the rationale for that.

DR. BARRIO: I remember, now, that you mentioned that to me before the decision was made about the percentage. By the way, most centers will produce FDG well below that number. But I think the issue is mainly important when the physical activity of FDG is too high or is high at the level, again, of 300 millicuries per ml or plus, and that FDG is sent X number of miles, after a few hours, the physician receiving that somewhere, when it is 5Êpercent or 6 percent, is the physician able to use it?

In reality, he is, that it is going to be a very good product. But it will be in violation to use it, according to that restriction. There are instances in which that may be a problem.

DR. KASLIWAL: I think that is where those manufacturers who intend to make doses in that range need to demonstrate, in their stability studies, that their formulation is stable for that during the expiration dating period so if the expiration dating period is 12 hours, those limits are achieved during that period.

DR. BARRIO: You are right. There are ways to minimize autoradiolysis; addition of more alcohol, now, as a component and whatever it is. But then they will have to modify the ANDA, right--in that particular case?

DR. KASLIWAL: No; when they submit, accordingly, that was one of the considerations for having 505(b)(2) to applications because, by law, the labeling has to have a quantitative statement of all the ingredients that are in the drug, the drug substance as well as the inactive ingredients.

For example, if alcohol is used as a stabilizer, that will need to be disclosed and its concentration revealed. That will change. The labeling won't be the same as your approved drug labeling reference drug so we couldn't accept that as a reference as an ANDA. Those were some of the reasons why this drug was chosen.

MR. SWANSON: A couple of comments, Ravi. First of all, your reference to the minimum dose of sodium fluoride, I don't understand the relevance to approving a different agent and the safety and effectiveness of that different agent. I just don't understand how you are correlating those two.

It doesn't make any sense to me. FDG should stand on itself with regard to what is the appropriate strength, quality and purity for the sensitivity and specificity of that agent. You don't base decisions on issues related to a different drug.

Secondly, we discussed this for hours as the USP meeting, if you recall. USP gave you strong advice not to do a radiochemical and purity for fluoride ion, to base it on a radiochemical purity of 90 percent, period. This is another example of having input from the PET community that you have simply chosen to ignore.

I think that is something that we need to open to discussion.

DR. KASLIWAL: I agree. That is one way of doing it. Usually, the way we will do it, we will look at the drug-product quality on an individual drug basis. For safety and efficacy, if you have evaluated a radiochemical purity of 99 percent, it will be tough for us to accept the 90 percent radiochemical purity.

We don't know what radiochemical purity is usual. The literature we have seen, people usually have that in excess of 97, 98 percent. If you go up to 10 percent fluoride, is it the same?

We have to have documented evidence and the only other things that we have to go by is we have an approved application and the impurity level, that impurity is a factor in imaging bone.

MR. FERRIS: Isn't the only issue, though, whether or not the in excess of 10 percent of impurity affects the safety or efficacy?

DR. KASLIWAL: Yes; it has to be shown whether it is affects the safety or efficacy.

MS. KUHS: I think there is a problem with that argument. The clinical approval that was given to that drug was not based on any controlled clinical trials. I am not sure anywhere in there there was any reference to the radiochemical purity at that point.

DR. KASLIWAL: I think we would be glad to have any data on an individual application basis. If you want to have a 10 percent limit, what I would suggest to you is submit data to show that, at 10 percent, it doesn't really interfere or affect.

DR. BARRIO: Is the reference I sent you some time ago from Egbert appropriate for your purposes?

DR. KASLIWAL: Where they are mixing 50:50?

DR. BARRIO: It is something like that. I don't remember exactly the number. It is a huge, huge amount of fluoride ion.

DR. KASLIWAL: No.

DR. BARRIO: That was a loud "no." What part of "no" I don't understand.

DR. KASLIWAL: We don't really know what you are imaging there. Are you imaging both things? The intent--it also depends on what the intention is. Your intention is in the labeling. Your intention is to image FDG.

DR. BARRIO: But the question is whether fluoride ion is interfering with the diagnostic and value of the drug and the answer to that is most certainly, "no." But question is, what would satisfy you? More references for the purpose? More data?

DR. KASLIWAL: If we can get some more references; yes. That will help us.

DR. CONTI: I might be able to help out here in an indirect way. There are many compounds that have been studied with PET, with fluoride that defluorinate in vivo to some extent. In those circumstances, the fluoride will go to bone just like it would if you had injected it directly.

You are going to come across probably dozens of those types of drugs over the next few years. Usually, if it is less than 10 percent, we accept that. We can actually do work with those particular pharmaceuticals. So there are probably a number of research pharmaceuticals, there are probably INDs, that have listed these types of activity in vivo that are allowed--these drugs are allowed to be used or, at least in the research environment, at this stage, can provide useful clinical information despite the fact that there may be high concentrations of what we would consider a contaminant, if you will.

But fluoride ion is not an uncommon phenomenon to see in many of these pharmaceuticals. So even at 10 percent, you could still provide useful clinical data. I think putting the onus back on us to say that it is okay, when we have that experience already, I am not sure if that is the right pathway.

We might want to throw it back to you and say, "Well, tell us what is wrong with that. Since we think it is safe, there must be some sort of safety or efficacy issue."

DR. KASLIWAL: The issue was last time and I think we discussed it. We may have it on the record that some of you indicated that there may be interference with the image from the bone. That is the issue, whether the image that you are looking at is the FDG or the bone fluoride image, whether it is consequential to the whole process.

MR. SWANSON: But you have arbitrarily chosen a figure of 4 percent. You don't have the evidence to support that except based upon a different drug. That is what I am saying. That is an inappropriate basis. So the only thing that you should be basing that decision on is the input from the expert groups, which you are obligated to do again as part of FDAMA.

Again, this group is telling you today, this group has told you before, that that is an inappropriate standard. What do we need to do to convince you of that?

DR. LEUTZINGER: If I may try to answer that charge you are giving to us. Basically, what we need to have from you is data. This is a data-driven agency. We need to have some data from you that actually shows us that there is no interference with the image, with FDG, that is caused from the free fluoride that might be in the product.

MR. SWANSON: Then I demand data from you to establish a 4 percent threshold. Where is that data?

DR. HOUN: I think we also took a poll in that meeting in September of folks in the audience about what problem limit people would have because our original proposal was, I believe, 2 percent. People felt that 4 percent--I know people came up to the microphone and said that that was reasonable, 4 to 5 percent.

I think that was also some of the advice from people in the audience.

DR. CONTI: It is also an issue, not necessarily of the image. It is interpretation of the image. So let's be clear. The image might show bone uptake but if the interpretation is not affected, that is the efficacy issue of the actual drug. That is important here.

Again, keep in mind that other radiopharmaceuticals either decompose in vivo or have contaminants. If you look across a spectrum of all the radiopharmaceuticals we use for both PET and SPECT imaging, I am sure that there are many that have 10 percent or around that figure of contaminants or impurities that are administered to patients and cause no effect on efficacy of the use of the drug.

DR. BARRIO: I would like to add to this that for the most part the information we have in academic centers and hospitals, et cetera, when relatively low doses of FDG are produced, 300 to 400 to 500 millicuries total for internal use, we, at UCLA, and as I know, have done stability studies and performance of FDG, particularly, analysis--continually, we do that. And the amount of fluoride ion that we normally inject with any preparation for FDG also, in particular, rarely, rarely exceeds 1Êpercent.

I think that is probably what happens in most institutions. But I think that the most important issue here, in regards to the fluoride ion and FDG, may pertain to industry. When that FDG activity is too high, again, and then distribution may exceed to a few miles, a few hundred miles, a few thousand miles--I don't know; 100,000--and then that FDG stays for too long.

If the radiopharmacy is sending that FDG from this place to whatever, it may arrive three or four or five hours later and then we may have a problem when that FDG activity is very high. But there are ways to remedy this, of course.

But it may cause a problem under this particular circumstance because if local quality control is met, then the physician in Nebraska could receive this from who knows where on the East Coast, and they may say, "I can't use this because I would be in violation," or whatever.

I think it is important when, particularly, there is nothing that proves the contrary. In fact, there are data that prove that it is not detrimental to the quality of the product.

DR. KASLIWAL: That is the whole point. George, you said that most of the people are using 1 percent radiochemical impurity and then that leads me to believe that is what has been studied for efficacy. Then, to justify 10 percent, we really need to have data to extrapolate to that.

When you say that when you ship the product to the end user whether the product will still be good or not, that you would have demonstrated in your stability study for twelve hours it is good. So I don't see that as a problem. The end user doesn't necessarily have to do a quality control on that.

You do a release test and you have established that your product stays within those limits through your stability studies which you submitted initially. You don't have to do RCP on a batch-to-batch basis, certainly not by the end user.

DR. LOVE: From a perspective of the different approaches that are in the document with the 505(b)(2) or the 505(j), if there are some differences, would it be possible to just simply document those differences? This approach, it would seem that Ravi is discussing, would allow you to look at different things and validate the fact that your particular product is still comparable as an end result.

But if you needed to make changes based upon whether you are distributing or if you don't need to make a change because you are not distributing, then you still have the flexibility in this approach.

DR. BARRIO: I agree. I think, as Ravi has indicated, that is the way to do it. It can be done that way, too. I was only referring to the fact that this is an area where concerns may exist and, therefore, changes may be necessary in order to assure--but I still think the issue of the fluoride ion is not necessarily the perception that people are injecting continuously things with a very high level of fluoride ion.

This is not the case. But the point is that, even if this were the case, it is no indication whatsoever that this is a problem. In fact, you should read that paper again because that paper is very good, the one from Egbert in Freiburg, a couple of years ago.

DR. KASLIWAL: We will reevaluate that.

MR. SWANSON: Are we still on CMC?

MS. AXELRAD: I was going to say, if we are done with the fluoride-ion issue, maybe we could take a ten-minute break and then get into the other CMC issues.

[Break.]

MS. AXELRAD: Shall we continue, then, with comments on the CMC section of the guidance document. Dennis?

MR. SWANSON: Getting back to the model application, it is hard to discuss what I see as the problem with the model application without discussing CGMPs and the community's interpretation or problems with CGMPs.

I think I brought this up before. Basically, the difference between USP and CGMPs is a basic different between a performance-based approach to regulating this area versus a prescriptive approach to regulating this area.

It is a difference between an emphasis on process controls versus an emphasis on the purity of the final product. We have discussed this at considerable length before. So now I see a format coming out that, as I said earlier, when I look at this, it looks a lot like CGMPs to me.

It is not just me. This is comments from the PET community. It looks like CGMPs. You want us to reference our standard testing procedures. Let me give you a problem with that. I am a nuclear pharmacist working in a PET center and every day I do chromotography two or three times a day on FDG.

I know how to do chromotography. Yet you want me to write a standard test procedure for how I do chromotography and reference that in the application. Or you want me to write a standard test procedure for how I visually observe a product.

To us, that is just busy work. From your perspective, you have us write those procedures so when your inspectors can come out, they inspect us on those procedures from a prescriptive approach. That is exactly what we are trying to avoid.

If you think that I am going to agree to a prescriptive approach to regulating this area without arguing about it, you are wrong. I have argued for the last seven years in the NRC for them to go from a prescriptive-based approach to a performance-based approach and they are finally doing that.

I am telling you the same thing. That is the problem that we have with this in the PET community. It is a basic concept in going with CGMPs and your sample format looks like CGMPs. Do you kind of understand where I am coming from there?

MS. AXELRAD: I understand where you are coming from sort of in the broad perspective. I certainly understand the big point you are making, but could you give us the specifics of where in here, what specific sections you think need to be fixed in order to address this?

MR. SWANSON: The question I pose to you is this model application, which you are asking them to submit today with their application, to me, this looks like CGMPs. This ought to be somewhere out here. This does not resemble what people are doing under USP today.

I think that you are going to find that now people are going to have to take a considerable amount of time to write standard test procedures and they are not going to like doing that because it is busy work, things like validation of pH paper with two reference standards.

We buy commercially available pH paper. This doesn't seem to be a major issue to us. Your model application looks like it is moving people very rapidly to CGMPs when, in fact, you are going to go out and inspect on USP and--

DR. KASLIWAL: I think, then, if we look--I am looking at the USP compounding chapter which has just come out, each of those require, for example, to have established in writing the quality-control tests and then it also requires that you will follow the routine procedure.

So I don't see whether we are being prescriptive or not. You are going to have to have written procedures. Those are the procedures that we are asking--the application is how I am going to make my drug, the drug that I intend to put in humans. This is how I am going to make it. These are the materials I am going to use. This is how I will make it. This is how I will test it.

So all those procedures have to be in a written format. The USP requires that, as well as CGMPs are required. The way we work at the Center, is the application, you tell us, "This is how we make and test the drug." And then CGMPs are basically whether you are doing that or not.

MR. SWANSON: I am not arguing that you need procedures. You have seen what we do and I have sent it to you. You have looked at what we do. You know that we have procedures. We certainly need standard operating procedures to a point.

But what I am concerned about is, and maybe this waits to get into the detail of it, but how much detail do you want? How much detail are you talking about? If I say I have, as a quality control procedure, I am visually observing the product, do I have to describe in detail the procedures associated with visually observing that product?

To me, it suffices that one of my quality-control procedures is I am visually observing the product, which I do and I document the results of that. But when you reference a standard testing procedure, do you want me to describe under what conditions of light, et cetera, and so on, that I am observing that product?

DR. HOUN: I think what will help the community is what we are doing now, developing an actual filled-out application to put on the web as an example. Those examples could be used, copied, we hope, by the community in terms of this is one way to answer the question, what is the level of detail.

So I think those examples on the web that we did not have--we don't have it now, at this point, but we will this year, will answer those questions. They can be used. They can be printed and copied as we do this exactly, too.

So some of that, I think, the PET community just doesn't know, what do you mean when you have a blank here, how detailed. I think the answers to those questions can be answered by the example that we will be having on the web. The person we are working with to develop them is going to be using a variety of data to fill this out so there will be some reassurance, especially to new people and people who are not accustomed to writing SOPs, what is an SOP.

MR. SWANSON: Then I would encourage you to get that out before you actually--or I actually encourage the PET community not to submit their applications until that is out there. The last thing you would want to do is overly commit yourself. Advice to the PET community.

This is a major question. This is a major difference between what you are doing today and what you are going to be required to do under CGMPs. Maybe this discussion needs to wait until we talk about CGMPs, but it is a major concern.

If you look at your application, which is part of what we are talking about today, you have a model format. We are asking for standard testing procedures and so forth. We don't know what you want.

DR. HOUN: It does sound like--we have members here from the Office of Training and Communication that, in terms of our outreach activities and communications we have with the PET community, some of those questions about what level of detail, what do they mean, those are guidances or examples that we need to develop.

I know for the mammography program, there were a lot of questions as well on what do you mean, develop these quality-control procedures? What do you mean document my CME levels? Do I have to have the actual certificates? Could I just type out?

For each section of requirements, we had to kind of go through in a newsletter just what is the level of detail. But, certainly, this community will need that type of education and we appreciate that. We need to do a good job on that.

DR. LEUTZINGER: Thank you very much, Dennis, for pointing this out to us. We had no intention of making this sound like a format for GMPs. But your application must have some sort of written-out procedures that you are following. That is what we really mean by standard testing, the standard testing method you have shown to be suitable for its intended use. That is what we mean by validation.

I would say that we would expect you, down the road, to use these same procedures that you have written out in your application and that would be verified at the GMPs and inspection.

MR. SWANSON: Again, it is all in the level of detail. You are dealing with a group of people here that--I have said this before, too. We are not manufacturers. We did not come up through that environment so you are dealing with people that don't understand CGMPs, that are always going to take a look at these things with a critical eye unless you give us some very good guidance.

The other thing I say, we are always, as a regulated community, going to want to avoid prescriptive compliance activities versus performance compliance activities. I urge you, as you think about developing new CGMPs for PET to really focus them at a performance-based type of approach.

I realize that is different for you but unless you do that, you are always going to be butting up against this kind of an acceptance of it.

DR. LEUTZINGER: Dennis, we intend to do that. I just want to make one other comment about this validation business and just emphasize that we are talking about, in validation, you showing us that it is suitable, the method is suitable for its intended use. That has maybe a different connotation than validation, in this sense, than it has in the past for regulated industry.

We would like to have you have a system-suitability test. A system-suitability test can basically be any kind of indication that you see during the test that indicates that your method is working. We need to have that sort of thing written into those procedures.

We are going to talk about that at length later on, but that is the sort of thing that we are looking for because we need to know that you know that your procedures, your methods, your instrumentation is working properly.

That is very important because that is part of the assurance that we have, a trust that we will have of you, that your product will always have the quality that you are expecting and we are expecting.

MR. FERRIS: So you are going to be coming away from the standard accuracy, sensitivity, specificity type of validation?

DR. LEUTZINGER: Not entirely, but it will be a less rigorous, probably a less rigorous, kind of performance because that depends a lot on the intended purpose that you have. We are going to deal with each of those issues test for test.

DR. LOVE: Of course, people are using all these procedures right now for ABGNS. Would you accept whatever record they may have to prove the point?

DR. LEUTZINGER: We will. We will accept some of those records definitely, for sure. The history of its performance, of its use, is part of the history of performance and gives us an assurance that there is unlikely to be any problems in their use. We certainly will accept that as part of it.

DR. KASLIWAL: I also just want to comment to alleviate some of your fears that if you submit a method that is incomplete, somehow that is counted against you. If the method is deficient, that is the review process. We come back and ask you, "We need this additional information."

DR. BARRIO: Okay, Ravi. Okay. There is no procedure that I could send you that you wouldn't be asking me ten or fifteen questions. It doesn't matter how much I send you.

DR. KASLIWAL: That is not true.

DR. BARRIO: All right. I don't think that would be a good answer for our concerns.

DR. CONTI: I have some specifics on the CMC section, if you want to go through some of those. There are probably a couple of typos throughout the document but I don't want to nit-pick. There is one thing I do want to point out on the Kryptofix, the standard reference of 50Êgrams per milliliter. I am sure that is--

MS. AXELRAD: Do you have a page number?

DR. CONTI: Yes; I'm sorry. I think it is 25, the Kryptofix STP number. That box. That is not correct, I would assume. That is a typo. Is that 50 milligrams or something like that?

DR. KASLIWAL: I apologize. Some of those boxes, yes, it is micro. Some of the symbols didn't come out when it got transcribed.

MS. AXELRAD: What happened was we created the document in whatever word-processing program and, when we converted it to PDF to put it up on the web, the symbols changed.

DR. CONTI: Another thing, and these are important, I think, instructional issues, for example, on page 22 when we are talking about flow rates and synthesis times and things like this, I think we need to instruct the community to deal with either ranges or averages somewhere in the documentation, because having a specific flow rate may change and may not affect the process whatsoever.

So those types of things we need to address.

DR. KASLIWAL: I agree. It is not intended to be a single number. A reasonable range.

DR. CONTI: I had a couple of other issues on acceptance criteria. Some of the things are probably doable and others are going to be very difficult. I am just not sure which ones to deal with, particularly because we don't have the GMPs. This may be a factor and it may not. I don't know.

For example, doing gas chromotography on certain organics can be done, but when you get into inorganic salts and things like this, and validating the purity of these agents or their composition or dealing with resins and things like this on columns, I am not sure, at this point, what level of system we are going to have to go through to prove or disprove some of these agents without the full, I guess, spectrum of the documentation at this point.

So I am not sure how to address that in an application at this stage. For example, the hydrochloric acid for the FDG production. Can we just rely on the manufacturer, if it is a reliable manufacturer with a certificate of analysis? What if there is no certificate of analysis? How do we handle that type of situation?

DR. KASLIWAL: You are asking when you receive your hydrochloric acid--

DR. CONTI: Say, from Fisher Chemical, whoever you get it from to use, there is probably some information on the label. If you document that, would that be sufficient or do we have to do independent testing of those reagents, things like that.

DR. KASLIWAL: At this point in the application, what we are requesting is where you get it from and what your acceptance criteria and the certificate of analysis--acceptance criteria means, once you get it, what, exactly do you do to accept it.

DR. CONTI: If you order a box of gloves, do you try them on and pull the different pressures until they break and then that is some threshold; we can't accept that box of gloves? You can get down to that kind of level.

DR. KASLIWAL: I don't think we will look for that kind of level.

DR. CONTI: Obviously, it borders on the absurd, but I am trying to get some feel, based on what I have in front of me, as to what to provide you. I can certainly say that if a product comes from a reputable manufacturer and we have a track record, the community has a track record of using a particular product without incident, I don't see, other than referencing that, what we actually have to go on and do to validate that.

DR. LEUTZINGER: We do have to be realistic about it. We wouldn't expect you to do things that are kind of dumb like put the gloves on and see if they fit.

DR. BARRIO: "If it doesn't fit, then you have to quit."

DR. LEUTZINGER: That's right. But if you have a reputable source of a reagent, that certainly means a lot to us. But you need to show us that you have done this process, this testing method over and over again in your facility with that reputable source and that it has always stood up.

You need a certificate of analysis for it. We are going to have to do some thinking about additional testing, whether it is required or not. It may be, in some cases, not required. In some cases, maybe it should be required because that particular reagent is very fundamental.

DR. CONTI: The calibration of a syringe; that is another good example.

MS. AXELRAD: We have not yet decided. I think you are absolutely right, Peter, that this definitely connects with GMPs. We will have to be, obviously, consistent in terms of what we are asking you to show on the application and what you are going to be required to do under GMPs.

We haven't decided yet. At the last meeting, we had an extensive discussion about identity testing and other kinds of testing of incoming compound. I don't think it has been resolved yet within the agency as to--we had talked about differentiating some things you would want to check more closely, perhaps, than others and there might be, perhaps, some defined list of things that you would want to do something special with and other things that you wouldn't have to do it.

But we have not yet resolved that. I think that the next time we produce something on GMPs, we will put out whatever our position is on some of these things and then we will discuss it in that context.

DR. CONTI: I think the issue of critical components or critical issues that we have identified, and then the reliance on the performance and the historical data for the others would probably be the most appropriate way. That would certainly streamline a lot of these--

MS. AXELRAD: I don't know if you have ever done this, but have you given us a list of the critical components, the things that you think would merit that?

DR. BARRIO: We did. We discussed this in the USP context, during that time.

MS. AXELRAD: I don't think we did.

DR. BARRIO: Also for the CGMPs, we discussed the pluses and minuses of potential problems that every component or study material may have. We have discussed this. Now, we have not submitted any documents on that particular issue.

MS. AXELRAD: It would be useful if we had some idea of what you all would view to be the critical components and what kinds of acceptance testing or whatever you normally do because manufacturing practices, the CGMPs, are supposed to be current good manufacturing practices. They are supposed to be sort of representative of what a good manufacturing, or excuse me for the word, a good production, a good compounding. You have made me sensitive--a good compounding or production operation for PET would do.

It would be good to hear other than just--we have touched on some things that we think are and other things that we think are so obviously not, but we really haven't done a comprehensive list of what should be controlled and what the tests or controls would be.

I think that would be very helpful, both in the application context and the GMP context.

MR. FERRIS: Do you want that distinguished from, for example, with FDG, components that have been identified in the NDA that you have?

MS. AXELRAD: Yes.

MR. FERRIS: You want a separate list?

MS. AXELRAD: You could work off of this. Some of the things, the lists of things in here. You could say, "This is important. This isn't. This is important. This isn't. This is the kind of test that we would do for this." You could use this as sort of a way of going through, like a check list, identifying which things are important and which are not, from your perspective.

DR. KASLIWAL: Let me just get a feel. Bob, what you are saying is that the components that were identified in the DMF was the critical components? Is that the feeling?

MR. FERRIS: If I recollect, I think they were identified relatively clearly.

DR. KASLIWAL: As critical components. So the same component is the community feels ought to be--

MR. FERRIS: For FDG. Well, I am not sure of that. I am asking you, are you asking us to create a different list, a new list?

MS. AXELRAD: If you think it is different now than the way it was when the application was submitted in 1993, whatever it was, then you could tell us that. But we also would need a list for the other drugs because it will be different for those than it is for FDG.

MR. FERRIS: Right. There are also additional processes right now in terms of making FDG. Okay.

MR. SWANSON: With regard to your current development of CGMPs, are you working actively with any advisory groups in the development of those CGMPs? I think we had a good experience working together in coming up with final USP monographs. Do you have a group that you are actually employing in the process of developing appropriate CGMPs?

I would recommend that you do that rather than just working on it and throwing it in front of this group and waiting for us to respond to it and then have to go back and work on it again. If you had some active involvement from the community in your actual thinking of CGMPs, it might help us through this process in a more rapid manner.

MS. AXELRAD: What we are actually trying to do is respond to the comments that we got from you all at the last meeting and react to them and determine what the agency's position is going to be with regard to, like, at least five major issues identified. Tracy Roberts is not here today. She had a family emergency. Basically, she is the person who is doing the GMPs.

We have been trying to resolve some of the issues that you raised the last time to determine whether the agency is going to change its position. Then we will put that out at the next meeting. Basically, we have been using you all as the group that we are consulting with on this.

We have been doing it really in the meeting. So we will come back the next time and tell you what our position, our current thinking, is on identity testing, reserve samples, the kinds of things that we talked about the last time and then see where we are.

MR. FERRIS: On page 29, I would also support Dennis' contention to the PET community that you really couldn't submit an application today or tomorrow without a connection to CGMP because I think what you are asking for is a list of SOPs there.

DR. KASLIWAL: Are you talking about the table?

MR. FERRIS: Yes; the table. And the documents that you are requiring there are SOPs; correct--which would have a basis in GMP.

MS. AXELRAD: Some people may have them. You all have been operating for some period of time. We have been told, in multiple meetings, that you have procedures and that you are using those procedures. I don't see that there would be a big loss in having you submit the procedures that you have been using so that we can determine--and, frankly, if we could get a feel for--we have our consultant, and they are going to be filling out an example application, and they will be getting data from real sites to be able to do that.

But the more we see, the better we will get a feel for what is out there and the better we can have a dialogue talking about real specifics about what we see to be adequate and what is not, and to talk about it in a specific context.

If you have the procedures and they are already written down and you are already doing this microbiological testing, just submit them and see what happens.

MR. FERRIS: My concern here is the background on environmental testing and environmental monitoring with respect to aseptic areas. People do microbiological testing of the final product; yes. The background that supports that testing is what is really defined in GMP and this may not be the appropriate forum for that at this time.

MS. AXELRAD: I would ask David Hussong to comment on this, but we did spend a lot of time working out the USP chapter with regard to how it would apply to sterile products. It seems to me that is sort of the overlay of that.

David, you might want to say more. You were involved in developing the USP.

DR. HUSSONG: The USP does provide certain tests and methods and it does recommend that additional controls be part of your manufacturing or production process. They have to be sensitive to manufacturing.

Not all of these documents need be strict SOPs. Clearly, things like endotoxins tests, sterility tests, filter-integrity tests, those have to be iron-clad and they have to work and certainly need to be a part of the product application.

Many of the others can be, rather than a strict SOP, can be a narrative description. Those will get you by for this. But what you submit in an application provides the basis for your ability to achieve your GMPs.

Typically, we don't view the USP as a GMP document but rather we look at what you do and how you test your product, test your process. If these methods that you submit are workable and in good shape, then I would be willing to accept the drug.

There is a link between what you submit in an application and your ability to meet GMPs. We are trying to get the information there to see that you are going to be able to achieve those.

DR. CHALY: I am Dr. Chaly from Northshore University Hospital. In our center, the microbiology tests are done in our hospital. We give our sample. Every day it is done, and they are going the test and they are giving us the results. So, in that case, what do you want us to do with that? They have a standard operating procedure over there. They are doing all those tests based on that.

DR. HUSSONG: In a situation where you are contracting out the testing, it is common to ask for the test procedures. And then you can submit them or you can reference them, assuming that the contractor, in this case, the hospital, is willing to say what test method they are using.

References are common in drug applications. What we are trying to avoid and the reason we simply ask for that sort of thing is a lot of hospital methods are not suitable for testing drug products. We want to make sure that that is exactly the situation because hospitals test for infections and drug products are tested for contamination. There is a difference.

MR. FERRIS: Typically, those tests are directed at a final product rather than an environment in a hospital, is what I am saying, and especially in a aseptic environment.

DR. HUSSONG: Yes.

DR. BARRIO: Then what do we do when the hospital does not have another method and the investigator or user doesn't have any other way to do it?

DR. HUSSONG: I think there is always another way to do it, but cost is a factor, too There are plenty of contract testing laboratories out there. Relying solely on an inappropriate method is the reason we are here, to make sure you don't.

MS. AXELRAD: I think I see the problem here. I think I see the issue. I think we need to talk about it among ourselves and I think particularly, specifically, because I know it is a big issue is in the sterility context. So why don't we talk about that and make sure it is on the agenda for the next time we do GMPs.

MR. SWANSON: I have a specific comment, page 25 on your acceptance criteria for FDG, you continue to have an acceptance criteria of isotonic. Again, that was something that we discussed at length in the USP and we removed the requirement for that agent to be isotonic. I think we agreed that you do need to address the osmolality of that agent based upon calculations but there is not a specific requirement for that product to be isotonic, nor should there be a specific requirement for that product to be isotonic.

DR. KASLIWAL: I guess you can calculate it if you so wish, but the reason we left it here is because the currently approved product is like that. But if your product is different, that is how it is going to be labeled and that determines which type of application you can submit or not.

MR. SWANSON: Back to the same problem, Ravi. You know, we are trying to develop appropriate procedures for the approval of these agents. We have given you a piece of advice on that. You are back hung up on your existing requirements when, in fact, the FDAMA tells you you ought to create new procedures. This is another example of that.

This is a prescriptive requirement that you are hanging onto because of your existing regulatory framework and thinking. Think outside the box. We are telling you; it is not a requirement.

You, in fact, have agreed with that when you sat with us in the USP. You are hung up because this is what your current NDA says. That is going to, in fact, then require everybody to either submit suitability petitions to address this issue or a new NDA that is going to have to deal with a non-isotonic product.

Why are you creating these kinds of situations for us?

DR. KASLIWAL: What are you recommending with respect to that, that we change that, take it out?

MR. SWANSON: There should not be a specification for the product to be isotonic.

DR. KASLIWAL: Okay. And to conform with the labeling, if somebody wants to submit an ANDA, then they can have it if they want. Okay. We will probably address that in the wording.

MS. KEPPLER: Comments on the CMCs?

MR. BRESLOW: Ken Breslow, PetNet Pharmaceutical Services. On page 21 of the FDG CMC section, you ask for the acceptance criteria for the target body and the target windows that come in contact with the target material are provided in attachment so and so. Could you expand on why we need to have an acceptance specification for the target body and the windows? Thank you.

DR. KASLIWAL: Oh, basically, because these materials--it is intended that the materials that come in contact with the target material--now, if you are going to change the target, whether your radionuclide impurities might change. It doesn't have to be extensive.

If you are using a particle or target, you basically need to say that is what you are going to be using.

MR. BRESLOW: Do you think it would be adequate to state the specification rather than having to undergo some kind of acceptance testing of a target body and a window?

DR. KASLIWAL: Yes. Your acceptance testing could be very simple.

DR. CONTI: This gets into what are the alloy components of the foils or what is the purity of the silver in the target. These are manufacturer-provided items. I think it is enough for us to tell you what we are using and use historical data that it is successful.

DR. KASLIWAL: That is all we want to know, that that is what you have been using and in the future, when you get it, that is what you will use. That is what, basically, it boils down to.

DR. CONTI: Or if the manufacturer provides a different component that they have already tested in their own facilities and say, "Now the alloy is a different percent than the foil; it works perfectly fine by the manufacturer."

MS. KUHS: Nobody tests foils.

DR. CONTI: It is a test of production. If you use a certain type of target system, you get a certain yield. If they change the constituents of target system, you might change the yield, but your product is identical.

DR. KASLIWAL: So, in that case, if the manufacturer changes, they would tell you that they have changed it; right? Or you will never find out?

DR. CONTI: You may select a different manufacturer for the material. All I am asking is that we just tell you what we are doing as opposed to going through an acceptance criteria for components such as this which is just not necessary.

DR. KASLIWAL: That is what we want to know. We want to know what you are doing.

DR. BARRIO: I believe, Ravi, we discussed this. The acceptance-criteria issue for the target body also implies that you should know where any metals or impurities are leaching out from target. I thought we had decided that this is not an issue on the preparation of FDG and maybe an area that we will have to address in the preparation of fluoride ion that directly is being very much directly used in the injection.

I think the USP monograph addresses the issue of these metals on one side that are important to be considered and, in the case of FDG, that is not an issue at all because you are going to remove them during the synthesis process.

I think there should be clarification on that part based on what we have discussed, namely that we are not really talking about the leaching of metals and that, to make sure, which is a level of them being released every time. As Peter indicated, you change your manufacturer and you might change the amount of metal that may be leaching and so on.

And that is measured because that is not the thing that anybody, any center, could do. That has to be contracted outside and things like that.

DR. KASLIWAL: For FDG, because you have a purification system, the theory goes that you eliminate everything and then the batches that you will submit to us. And those batches, you would have shown that it is not a problem, obviously.

For something like sodium fluoride, because it is produced in the target and then you will have either--and I don't know; there could be various configurations. In that case, also, it is the whole thing; this is what we use, this is how we made it and this is what at results are. That is what they show, show to us that you are able to manufacture that product.

MS. KUHS: I think that the issue of the targets and the target windows that we are trying to get at is having some knowledge of how the manufacturer actually purchases the material for the windows and distributes them out to the people who own targets.

I know of no in-process testing anywhere along the line--if we say it is a Havar window, the manufacturer ordered it from some place that said it was a Havar window. It is purported to be so many microns thick and it is packaged into inventory. When someone needs a new window, it is sent out to them and they put a new window in their target.

There is no in-process testing or identification anywhere along the line that I am aware of in that there is really no significant way that an end-product user on the cyclotron end can actually verify that that, in fact, is a Havar window other than looking at that it came from the supplier of the Havar window who set it was a Havar window.

MR. FERRIS: Do you care about that? Do you care about that or do you just simply say--or is it enough to simply say, the provided descriptive content of what the manufacturer says that is and then, if you did a fluoride run or a radiation, and then did radionuclidic purity on it and evaluated any impurities that would be present, that would be enough as an acceptance criteria.

DR. KASLIWAL: Yes; radionuclide impurity will be enough but the very first batch that you will make after that, you don't do radionuclidic impurities until after you have used it.

MR. FERRIS: Right; but I am not talking about doing a synthesis with it. I am just talking about making the fluorine and then letting it decay and then evaluating it for radionuclide impurities.

DR. KASLIWAL: That is perfectly fine.

MR. FERRIS: That is what you are saying needs to be done? But there is not anything further with regard to the manufacturer's composition of the target?

DR. KASLIWAL: No. It is not intended that you will take and determine the elemental composition of the target. The specification could very well be a dimensional type of specification, a particular geometry of the target and things like that.

DR. HOUN: I still see some frowns.

MS. KUHS: I am thinking of a very practical experience about you go in in the morning and you start up the accelerator and the target blows and you have validated that particular window and that target and you know you are getting FDG. You take off the target and rebuild it and put a new window in, you still have to make that run that day.

Are you going to do three validation runs to make sure that that particular target window isn't giving you any impurities? How far does this go?

DR. KASLIWAL: No. Hopefully, you have bought the same window from the manufacturer so the window, what you said was the same window, you don't have to. When you suspect that your windows are changed, your targets are changed, that is when you do that kind of thing.

DR. BARRIO: You can see, Ravi, what the fear is. When you go through all these validations that we discussed many, many times, I guess very clearly in one of the CGMP meetings we had is that validation for you, as a chemist, is not necessarily the validation that an inspector or somebody else in the agency, or even us, may think is appropriate.

I think that is the problem. I think you will see, in any example we touch, you have these kinds of questions, essentially. I think, perhaps, some genuine implementation of a common-sense rule that is easily readable by everybody could be helpful.

Remember, we are here around this table. We met with you I don't know how many times, as Jane said, we already forgot how many times. We have questions. Can you imagine what happens out there with the PET community receiving these things and saying, "Do I have to do this, or do I have to do that?"

People have no idea what the requirement will be.

DR. KASLIWAL: George, a lot of the comments I am hearing for the first time. We have every intention to clarify the requirements.

MS. AXELRAD: We are here to hear from you. If you are done with the CMCs, somebody in a audience?

MR. MATTMULLER: Hi. I'm Steve Mattmuller from Kettering Medical Center which would say the law doesn't address it. But I am in an "in-search-of-profit" institution. We are small PET center and some of this is scary. Some of this even looks like it goes beyond what was required for Peoria which I think some of us would say is excessive.

A very good example; on testing your target window, it is, like, for us, we are performance over here. We are not prescriptive. It is a big difference that I don't know if it is sinking in yet. If our foil breaks, we put in a new one and it works, we are happy. Likewise, also, on page 15, you are talking about acceptance testing for mannose before we even use it.

At the last meeting, we discussed this at length. At that point, it seemed like you were just going to require us to test melting point. Now I see, on page 15, there is a test for appearance, there is a test for specific identity. There is a test for purity and a test for melting point even before we use this product in our synthesis of FDG.

In the five years that I have been making FDG, we look at our mannose and our COA and we are happy with that. It works for us. I have got to tell you, I do an appearance test when I weight it out every morning, but I don't do any IR, NMR, or melting-point testing on my mannose before I use it.

DR. KASLIWAL: You can release the mannose triflate on the basis of identity alone provided you--the way it goes is for a component, you set your specifications which include identity and other aspects. And then those specifications can be satisfied in different manners.

You are saying you compare your specifications that you set, the acceptance criteria, on the basis of certificate analysis, the analytical data that is in the certificate analysis, and make sure that batch that you received is within that specification and you can accept it through that identity test.

That is all that it is intended to be. So I fail to see there is extra requirement here.

DR. CONTI: Let me pick up on this, too. I kind of agree because this is the type of situation where you have a certificate of analysis from an established product. That is enough, as long as you comply with the storage criteria from the manufacturer and you use it as you would according to your procedures.

You use it and you get final product. Those types of information are sufficient to deal with this. If you are making it in-house, this is pretty good because these are things you would want to do.

But if you are purchasing it, a certificate of analysis should be enough.

DR. KASLIWAL: The thing is that, a lot of times, what we have seen is a certificate of analysis, basically it is the manufacturers who report their analytical data. Now, whether that data conforms to your specifications, that is what you have to make sure.

A lot of times, what we see is the certificate of analysis will report analytical data with a little asterisk that, well, it kind of failed there, but we are going to sell it anyway because it is not intended to be for this, this, this, this purpose.

But whether that value meets your requirements, that is what you have to make sure.

DR. CONTI: I think we can do that. I think we can actually say what our specifications for acceptance are. But to repeat tests and to do some of these analytical procedures I think is unnecessary.

DR. KASLIWAL: I don't think we are requiring that here. All it is is you have to cite your specifications and tell us to accept that this is what you do. For this particular test, we simply accept the value from COA, this, this, this, and to release, we do an identity.

MR. SWANSON: I appreciate that. But then you have to go back through this and be real careful about what your wording is because, if you look under Page 15, Section 2, Item 4, specifications, it says, "provide tests to control the identity, purity."

So the logical way to interpret that, which is what you are hearing, is that I actually have to test for each of those when, in fact, what that wording should say is address your specifications for the following things.

Do you understand what I am saying? You have got to be very careful in how you word this so it is not misinterpreted such as what you just heard here.

DR. KASLIWAL: Okay. We can address the language there.

MR. BRESLOW: I just want to point out the continuation of that part of the form on Page 16, Item 6, continues then to indicate requirement for an identity test performed to confirm the structure to release the lot for production use. So you seem to be adding an additional layer there for the mannose triflate.

I am not really sure what the difference is between Item 6 and Item 4. I just want to tack on a little bit of historical information. In the past, when an NDA had been submitted, one of the specific identity tests that was included in that application for mannose triflate was to perform and FDG production run.

If it was mannose triflate, then it would make FDG. That would be a highly specific identity test. If it was not mannose triflate, it wouldn't make FDG. We are not talking about purity. We are talking about identity.

The agency rejected that premise, that that wasn't a measure of purity. So the agency had recommended the use of reaction-based qualification for something like O18 water and the concept that we used here was reaction-based qualification for mannose triflate, as far as a specific identity test, and it was rejected.

The intentions were good on the part of the applicant, but it was misinterpreted and rejected without logic by the agency. So even when the intention is good, and I can't think of a more specific identity test than doing an FDG run with mannose triflate.

So I just wanted to throw that comment out.

DR. MARGOULEFF: I am Dr. Margouleff from Northshore University Hospital. Just a comment that I think was brought up a second ago about the foils. I am not aware, at our institution, that we change the foils on a schedule. Very frequently, there will be a foil failure--not that frequently, but occasionally, for whatever reason, the foil will develop a perforation and it gets changed.

Periodically, when the product appears not to meet our other specifications, the target will be taken apart and cleaned and the foil will be changed. But I am not aware, and I asked Dr. Chaly who works with me, that we systematically, every so many runs or every so many days or weeks, would change a foil that seemed to be functioning perfectly well and where the product was coming out the way we expected it to.

So it feels, reading that line there, that you are asking us to change it even when it is working. That would just slow down our activities and cause us to go through the whole business of putting in a new foil and checking to make sure it worked when it hadn't failed.

I don't see the logic of that. Dr. Barrio?

DR. BARRIO: Would you help us with a specific statement, in which page--

DR. MARGOULEFF: I'm sorry. You had the page there in front of you. There is a line there that says something about, "We change the foil on the following schedule." I think it was page 21.

DR. MARGOULEFF: There is a line there. Forgive me for not bringing it up with me. But it sounds like you are asking there for us to have a regularly scheduled foil change. That is not a schedule. That is an event.

DR. LEUTZINGER: We acknowledge that. In the past, we have been concerned about what happens if you have a foil failure, whether that would impact the product in any way, introduce an inferior product. That was our thinking at the time that we wrote this in. We need to have your input on these kinds of matters.

I guess we just have to have your experience in those regards. You can see where our thinking came from.

MS. AXELRAD: People in the audience, you are welcome to speak, but please come to the mike so that the transcriber can get it.

DR. CHALY: Thomas Chaly, again, from Northshore University Hospital. On page 15, 4, again, in the test, in the bracket, you can see IR, NMR and all those things for the testing for the components. That is a difficult task. Most of our hospitals are small hospitals. We don't have IR and NMR or anything.

If you put it there, when the inspector comes over there, they don't know. They don't have to know about those thing. So when it is specifically there, IR and NMR, and we are supposed to do that--

MS. AXELRAD: We understand that there is an issue between what is required in these things. I think we have a process for dealing with this. It is going to be clarified. We are going to try and agree on a list of critical components and the tests that are going to be necessary for those and some issue in terms of the level of detail that we expect to see in the application for these things.

DR. KASLIWAL: If I can just respond. If you notice, in front of IR, there is a little example, for example. So that is an example you could choose. You could have anything other than that if you want. That is up to you.

MS. AXELRAD: Maybe there is something that I should say here. This is just a guidance document. It is to try and show you one way that would be acceptable to us. But it is not binding. You can do it different ways. There may be many different alternatives that would be acceptable to us. That is something that is really important to understand about all of our guidance documents.

We have a flexibility with regard to those that we don't have with regard to regulations in that people can propose to us alternate tests or alternate ways of doing it or even alternate formats.

We put this together like this just simply to make it easy. But it is not the only thing that you can do. You can do other alternatives and they may be perfectly fine.

DR. HOUN: I guess I also hear the comments about imprecise language. When we say "acceptance criteria," what do we mean? If we list things, does it mean that everyone has to do all those things listed? So I think there is a lot of clarity that we need to bring back to the document and also in educational materials as we go through the next few years.

DR. LEUTZINGER: I would just add something to what has already been said. We are saying, for example, an IR or NMR test for purity and so forth." We are just talking about a test that is relevant for that particular material and is suitable. We would just need to have you provide us the information that proves to us that it is suitable.

It is just the presence of that one test that does do the sort of thing that we are asking for in that testing procedure.

DR. HOUN: I think we will look for your comments on what are suitable tests because, if we are listing ones that are inappropriate or, yes, they are possible but they are too expensive, or impractical, I think we will look for your listing to help us.

DR. BARRIO: I think, for the most part, in my own experience with chemistry in general, acceptance criteria based on the certificate of analysis is more than sufficient for the most part provided that we go through a qualified buyer or groups who can provide this identity as indicated.

If I remember correctly, the old NDA--and, Bob, you might like to help my memory--one of the acceptance criteria tests for the triflate was melting point and only melting point. I could give you a compound that is white, is a powder and has a melting point of 121, benzoic acid. It is a perfect example. It will meet the acceptance criteria you have, and it is not the compound.

MR. SWANSON: It was with the certificate of analysis, too.

DR. BARRIO: Right. If you have both things, then, the certificate of analysis makes a difference. That is where maybe IR, NMR, other data is important. As chemists, we understand how valuable this information is. I think there issue that was raised over there is do we have to do it every time, blah, blah, blah. Certainly, it is not necessary if you have a good certificate of analysis coming from a manufacturer that is trusted.

DR. LEUTZINGER: We acknowledge a certificate of analysis, but you have to be careful with a certificate of analysis, too, that it really is a reputable one, hasn't been displaced or lost or interchanged, you know, at the site of manufacture.

Maybe it is a low risk for that happening, but it is an important issue. We have to treat a certificate of analysis very carefully, too, in that context.

DR. BARRIO: You will know, two hours later, whether you have the compound or not. But this is what Ken was indicating. Of course, I understand the logic or maybe the criteria, perhaps, on not accepting this as being the certificate of analysis, the fact that you have produced the compound and they say, "Well, then I have it."

But I think it is a very important element, here.

DR. KASLIWAL: George, if I can just elaborate on the melting point issue, the issue of whether you can accept the melting point. If you take a melting point of a compound, let's say, 120 to 122. Then, usually, what we see is the manufacturers want to propose 100 to 150 range as the acceptance criteria. But that is obviously not acceptable.

MS. AXELRAD: Is that the completion of the chemistry and manufacturing controls issues? Anybody have any other comments?

MR. LAVEN: David Laven, University of Kansas Medical Center. On page 24, the little box at the bottom, radiochemical no more than 4 percent fluoride 18. I just want to bring you back to Dennis' excellent comment of the arbitrary nature of that 4 percent value.

You want to say provide a procedure, you are kind of nice in giving us an idea for everything else that you want. But you leave that box conspicuously blank. So if you don't have an idea, I certainly would like to entertain how you want me to go about that.

DR. KASLIWAL: And idea about--

MR. LAVEN: Determine the 4 percent. Why is it just 4 percent? I go back to Dennis' comment and I think that needs to be addressed.

MS. AXELRAD: We will address it. We intend to address that. I don't think we need to repeat the entire discussion. I think we understand there is a concern here. We understand you feel that it is arbitrary. We are going to revisit that.

MR. BRESLOW: I have one other request. Maybe Eldon or Ravi can describe to the audience the reasoning and usefulness of reference standards as indicated. Thank you.

DR. LEUTZINGER: I would just say maybe one thing about reference standards. In any kind of analytical test for identity or any other kind of testing, really whether yo have good results, whether the results can be relied upon, whether they are trustworthy, really depends--it is only as good as your reference standard is.

A reference standard is very important. It always has been very important. We are not asking you to do special things with the reference standard, just that you have something that is an authentic material that you can refer to so that you know that the data that you get out of that analytical test and its performance is, in fact, trustworthy. That is the meaning of a reference standard.

MS. AXELRAD: I take it that somebody has an objection, that there are places in here where we require the use of reference standards or suggest the use of reference standards that are inappropriate.

I would hope that, when you give us your list of the critical components and the other comments on this that you would indicate those places where you think that the use of a reference standard is unnecessary and we can revisit those.

DR. KASLIWAL: I would just like to point out this is in line with what the current USP monographs require.

MR. MATTMULLER: I think our concern with these reference standards is that right now they don't exist. If they don't exist, does the responsibility then fall on us to take something from a manufacturer and go through extensive documentation that it is, indeed, what they say it is?

DR. LEUTZINGER: Which reference standards don't exist.

MR. MATTMULLER: I believe for Kryptofix and mannose, CODG.

DR. LEUTZINGER: But wouldn't you be able to--you purchase Kryptofix. Actually, you can set up a standard, yourself, simply by you have been using--you can do some testing on it or, if you don't want to do that, maybe there is some other means by which you can establish it as a reference standard.

You can set up a reference standard in your own lab. It doesn't mean that you have to buy that reference standard from NIST or USP or some other reference-standard body. Some reference standards you can establish, yourself and then you have that reference standard. Maybe that is not a primary reference standard in the sense that we talk about an official reference standard. But it can still serve the same purpose.

DR. BARRIO: For your information, the USP is providing, or will be providing, the reference standards--at least we worked this out for the five radiopharmaceuticals that we review. And then all these reference standards are USP, the source may be there.

I am not sure Kryptofix is part of it, but it may be. I think, conceptually, we all will agree that we need to understand any chromatographic procedure we use, we need to know, if we say this is FDG with a reference standard, this is FDG, of course.

I think your concern is how far this will go. That goes back to my comment a few minutes ago in which it is always the fear that--but I think, eventually, we will have to apply common sense in the whole process and this is what we are intending here.

MR. SWANSON: Again, you might want to give us some guidance on how to go about establishing our own internal reference standards. It would seem to make sense--let's take Kryptofix as an example. I am using Kryptofix from a supplier. The real concern is is that Kryptofix carrying over into my final product.

So it would seem to make sense that the Kryptofix I am getting from the supplier could also be my reference standard to test for that because that is really what you are--that is the goal of doing this test.

DR. LEUTZINGER: We understand that. I was an academician, too, before I came to the FDA and I have made many, many reference standards. It is very simple. They don't have to be, necessarily, bought from NIST or USP. I have done sufficient testing on them and established them as a reference standard.

I put them in a bottle and kept them in a very secure place and checked on its integrity from time to time and made sure, in fact, that that reference standard will still be integritable. It is very easy to do. It is not rocket science.

MR. SWANSON: Give us some help on that, Eldon. Give us some guidance on how we should go about documenting it as a reference standard. I think that would be a big help.

DR. LEUTZINGER: Will do.

MR. FERRIS: Clearly, with the idea in the reference standard that the centers really would prefer being able to buy a reference standard from NIST because they really don't want to be in the business of making them.

DR. LEUTZINGER: Of course. But sometimes it is not always possible.

MS. AXELRAD: What else do we need to cover?

MS. KEPPLER: I know that I had a couple of general comments that were sent into us about the guidance and then some comments on the labeling. I don't know if others of the committee had--

MS. AXELRAD: We have to make a decision. Do we want to keep going or do we want to break for lunch and then finish up.

DR. CONTI: We have gone through the CMC.

MS. AXELRAD: I am just trying to get a feel for how much you want to cover. If we are going to go for several more hours, I think we should break for lunch.

DR. CONTI: I think we should break for lunch.

MS. AXELRAD: Okay. Let's be back in an hour, back by 1:00 or 1:10.

[Whereupon, at 12 o'clock p.m., the proceedings were recessed to be resumed at 1 o'clock p.m.]

A F T E R N O O N P R O C E E D I N G S

[1:15 p.m.]

MS. AXELRAD: Let's get on to the next subject here.

MS. KEPPLER: What about labeling, maybe go through the labeling.

MS. AXELRAD: Attachment 2.

DR. CONTI: I will just kick it off. There is one concern I had with regard to pediatric uses. One issue relates to FDG. I guess what you are referring to here is the epilepsy data. It is on page 20. There is a statement in there, in pediatrics, the recommended dose is 2.6Êmillicuries.

I don't recall. That may be in the documentation for the NDA, but I really would avoid that specific sentence altogether. You might want to provide a range based on other commonly used radiopharmaceuticals, but it is a real problem in terms of that particular calibration for all pediatric cases.

The dose is adjusted based on the patient's weight, primarily, anyway. Also pediatrics, I am not sure if you specifically define the pediatric patient population, a sixteen-year-old also may be considered pediatric but weigh 240 pounds and, therefore, would require an adult dose.

So I think there is some discretion on the part of the physician that needs to be applied here as well.

DR. LOVE: Right. The 2.6 is in the existing label for the epilepsy indication. That is why we use the 2.6. In recognition, though, of some of the things you are talking about, the next sentence talks about, "the optimal dose adjustment has not been determined," because we didn't have data to specifically describe what adjustment should take place based on age, body, weight, size, any other issues that might be relevant.

But the 2.6 is in the existing labeling for the epilepsy indication.

MS. AXELRAD: Is there any way that you could indicate what that 2.6 is based on, based on body size or weight?

DR. LOVE: There is not. It is almost identical to this in the existing labeling where it describes the dose.

DR. CONTI: I would move to just strike the sentence and leave it that you don't have the optimal dose perhaps identified in the literature but allow the physician to do this under the practice of medicine to make that decision.

DR. LOVE: The next sentence hopefully would allow that. This says that this dose was established, and it is just the recommended dose. Recommended is--it doesn't say, "Should not exceed." It doesn't say anything else. It doesn't even say it should be, that this is the recommended dose. So this labeling language would allow a physician to make judgments according to either dosimetry or any other relevant issues.

DR. HOUN: I think we can see what we can do in the next revision in terms of making it less precise, just letting physicians have considerations that other doses are possible. I think that was just the general sentiment.

DR. CONTI: Jerry Kuhs is just making another point. He is just mentioning that it also is equipment-dependent to some degree as well because of the differences in doses that are used, even within the adult populations for the different cameras.

DR. LOVE: We certainly could put something in there about equipment.

DR. CONTI: It also applies to the other isotopes, the other pharmaceuticals. I don't recall a specific quote for the other ones off the top of my head. I would have to douple check but, again, I would like to leave this as vague as possible and leave it to the discretion as much as possible of the physician.

MS. KEPPLER: I don't know if this is something you could add or not. Maybe this is just so minor it is not worth mentioning, but, on page 21--I guess I view the label maybe different. I view the label as being as a consumer. My training was as a nuclear-medicine technologist so I would get these out and look at them and say, "Give me guidance."

The patient should be at rest during the uptake period for the FDG. I don't know if you can add that or not, but that would be a useful addition.

DR. LOVE: Thank you.

MS. KEPPLER: In fluoride, the other comment that I had, I would say, with existing equipment, the 2.0Êmillicurie dose is probably not appropriate either. I don't know what, if anything, can be done about that. It is page 31.

DR. LOVE: I'm sorry; what are you recommending that it is?

DR. CONTI: With fluoride, we use very similar doses to with FDG, 5.0 to 10 millicuries. So again it is the amount of radioactivity for an adult patient that is an appropriate level of this type of positron-emitting isotope.

You can do the studies with less but I don't think that we should recommend something as 0.5 to 2.0

DR. LOVE: I think this was an area where we were having just a little bit of concern because the label that was approved for the original product has this as the dose. So, actually, if there are some other data, other references or something that you could provide, we could certainly look at it and change the dose.

DR. CONTI: That may be because of the technology that was used to image fluoride in those days.

DR. LOVE: We understand. But I am just saying, if you had some other information you could send us, we would be glad to take a look at it to address the dose.

DR. CONTI: There aren't many papers. George, what was the dose used for the fluoride in that paper that you were taking about? Do you remember?

MS. KEPPLER: It was mixed 50:50.

DR. BARRIO: Peter is asking me what the dose of fluoride ion was in the paper from Freiburg on that mix, 50:50. I don't remember the dose, really, but it is probably 10 millicuries total and 5.0 on 5.0. That is what we use at UCLA.

But I think in the papers that have been recently published in the last ten years with fluoride ion, I think the dose was 10 millicuries, if I am not wrong. If I read this correctly, it gives me the impression that the dose is based on the impurity of strontium 85 that is there, maybe, in that particular preparation, limited by this.

DR. CONTI: Wasn't that reactor-produced fluoride, though?

DR. BARRIO: Right. Therefore, you don't have to--they are different issues.

MR. LEEDHAM: Part of this has to do with the fact that this is a less reference-listed labeling and that language currently is in there.

DR. CONTI: Can we add--

DR. BARRIO: Is it, at least, for limitation, the impurity?

MR. LEEDHAM: I am not sure whether it is or not without looking at NDA.

DR. KASLIWAL: I am not sure that this product is reactor-produced. In this labeling, it seems like it is produced by particle acceleration. It may be a different nuclear reaction.

DR. CONTI: That is probably the case.

MS. KUHS: It think it is deuterons on neon 20. And so I don't think it is based on impurities but it was actually based on the ability and the count-rate ability of the equipment at the time.

If you administered too large a dose to the patient, it wasn't that it was harmful to the patient. It was that the equipment used to image it would flood and you wouldn't get a good image. Now there is much higher count rates available, better imaging equipment and you can produce a much higher quality scan in a much shorter time with a different dose.

So I think the dose was, at the time, based on the ability to image it, not necessarily any impurities or radiation dosimetry to the patient.

MR. FERRIS: How would you accept more current papers?

DR. LOVE: If you just either provided us the paper, ideally, or just the reference and we would take a look at it. What we did for sodium fluoride, as you know--our major issue was to look back at the adverse events that might have been in the available literature to see if there is some reason why we would not extend the existing indication and existing labeling.

But if you have some information that can update and revise this, and we would be glad to take a look at that and can do so and make any changes that are needed in the next labeling cycle.

MS. AXELRAD: I would suggest you do that as part of the comments on the guidance document so that it will sit in the docket, the comments will go to the docket, as opposed to just sending us a letter or something with the stuff attached. It would be better if you submitted it as part of your comments so that it can go in the record in the guidance document and everybody can see it.

DR. CONTI: Do we have the capability of qualifying--do we have to include this in the labeling and provide you new data?

DR. LOVE: If you provide something that would allow us to strike it, then we can certainly consider that.

MR. SWANSON: How do you see this labeling being used? Again, this gets back to the issues of commercial versus not-for-profit. Assuming that I am a commercial distributor, I would see this labeling, package-insert, information going with the vial of the product to the pharmacy for subsequent dispensing of doses, as we do traditionally.

But if I am a facility that is producing these agents only for use in my patients, how do you foresee this labeling being used within that type of an environment?

MR. LEEDHAM: With an approved NDA, you would have the label, and that is usually distributed with the product when it goes into distribution.

But if you are producing it for your facility, I am not sure that it has to go out with each product. It would just be your standard labeling.

MR. SWANSON: So this would be our labeling. It would be on file someplace and that is basically it.

MR. LEEDHAM: If it was sent outside the institution, certainly labeling should accompany it.

DR. CONTI: I had that question on the adverse reactions under FDG. It is page 20. The database is from, I guess, the epilepsy data but what is the documentation to the effect that hypotension or hyperglycemia or alkaline phosphatase increase was secondary to the drug as opposed to other processes in the patient?

DR. LOVE: I guess one thing is just a clarification of the labeling. When the labeling that we report, we report all adverse events. That is the standard practice for any drug. We usually report whatever occurred during the conditions of the trial. Many times, the trials are not designed to separate what was because of the drug or what was not.

If there is something that is obvious, then usually it is either there is a comment or it is not in the labeling, if it is obviously not related to the drug. Where there are not data to clarify it, then we list it. But it does not necessarily mean it is directly related to the drug or not.

But, yes, this is the database in the labeling from the original NDA approval for that first paragraph, essentially.

DR. CONTI: Would there be a way of clarifying that for our new label, or no?

DR. LOVE: Again, what that would basically mean is going back and seeing whether or not there were data. I can't say specifically off the top of my head the extent to which this was researched during the original NDA, to say what was done, was there a question? Is there a database that can answer it.

My recollection is that what is here is based upon what could or could not be resolved during the NDA phase. So, if it here, that means there were not sufficient data in the NDA to exclude the cause/effect relationship for these things that are listed.

DR. HOUN: We put out a statement, because we do that a lot in labeling.

DR. LOVE: Right. But I guess what I am saying is, to specify specifically whether this was or wasn't known to add that label, what we would have to do, at this point, is to go back and look at the original NDA database to see whether or not there really is a reason to say it was or was not causally related and we don't know, or something like that.

DR. CONTI: It was a literature review so it would be from the literature rather than a clinical trial.

DR. LOVE: Whatever is available, we would have to look at it--you are saying how would we do it.

DR. CONTI: Understood. All I am asking is can we put something in here to the nature that it is not determined, or cannot be determined from the data, if we review it, that it was a cause-and-effect relationship.

DR. LOVE: If that were, in fact, the case; yes.

MS. KEPPLER: I don't know that it falls under this section or not, or I was going to bring it up under a general discussion of the guidance, I had a question posed to me that it was unclear to this person whether or not the guidance allowed for continued off-label use, if the sponsoring organization was an institution and an employee of that institution was using the drug for an off-label, was prescribing the drug for an off-label use, whether that would be allowed or prohibited in this scenario.

MS. AXELRAD: Generally, once a drug is approved, it can be used for off-label uses, for uses that are not approved. It is within the discretion of the physician. If they want to prescribe the drug for an off-label use, that really is usually up to them.

We would rather they not go around advertising or promoting it for off-label uses.

MR. SWANSON: I think the question boils down to, though--and let me phrase it differently. If I were to submit this application, how would you recommend I submit it. The University of Pittsburgh owns the PET facility. If I submit this as the University of Pittsburgh being the applicant, then are the physicians that are faculty at the University of Pittsburgh considered agents of that manufacturer and, therefore, not allowed to promote off-label uses or would it be better for us to submit it as coming from the PET facility which would, then, keep the physicians separate from being considered as agents of that manufacturing operation?

MS. AXELRAD: I don't know if I can give you a definitive answer on that. I would say it might be better that the application was held by the PET center and that would sort of separate them from the people who were using it. We would have to talk to our Division of Drug Marketing, Advertising and Communications to sort of get a definitive answer on that.

MR. SWANSON: That is probably something that, again, it would be wise for you to take a look at. I get a little concerned because federal agencies tend to want to have the responsible institutional official be the university. In other words, most regulatory agencies would want to see this come from the university as the responsible institutional part of it.

Then I get concerned about how that would affect issues related to promotion. I think it is something we need to take a look at.

MS. AXELRAD: I think that what we can do is give you our analysis of what we think the implications would be if it was the broader entity with regard to any constraints you might have on promoting. Why don't we sort of commit to do that maybe at the next meeting. I am not sure it is something we would directly address in the guidance document, but it will certainly come up when we start talking about this issue of registration and whether you are a manufacturer or something else and all of that.

So why don't we provide that at the next meeting and see where we go from there.

MR. SWANSON: Okay.

MS. KEPPLER: That's all I think we had on the labels. If anybody else has any questions--

DR. LOVE: Any other areas to cover?

DR. CONTI: Do you want to go through the other two attachments, I guess 3 and 4. Maybe we could take those together if there are any comments or questions. My sense of these is relatively straightforward. I am not sure if there are specifics.

MS. AXELRAD: I was going to say, I really hope so. We thought it was going a little too far to fill these out, but we decided that we would be complete.

MS. KUHS: This might be an appropriate time to revisit something that I brought up earlier on about this kind of carrot-and-stick approach to the waiver or user fees. Again, my concern is the long-term growth of PET in being able to introduce new chemical entities.

I would like the agency to at least consider possibly even a specific time-frame waiver of user fees for new chemical entities so that we give the opportunity of the field to grow and to really get new PET compounds through the system.

When we are faced with an arbitrary decision on whether or not user fees are going to be applicable, whether or not for a commercial entity or private industry or academia or a hospital, there isn't any one of those groups that really has the wherewithal to look at user fees at this point in the growth of the industry.

I think it would be a gesture on the agency's part to foster new technology development to at least look at a specific time frame for a waiver of application fees for new chemical entities.

MS. AXELRAD: I don't think we have the legal authority to do that. We have to evaluate waiver requests for user fees under the statutory criteria and the Prescription Drug User Fee Act. Those criteria were amended at the same time as a part of FDAMA. So if Congress had wanted to put in to the User Fee Act which was being considered at the same time of FDAMA some kind of a blanket waiver, they could have done that.

Now, we got to the waiver that we have given for these three products by going through an analysis of the statutory factors but it would not now enable us to just give a blanket waiver for five years for all PET.

Frankly, I am not sure, given just the way the industry has changed, our understanding of how the industry has changed over the two-plus years we have been working with this, that we think it would be appropriate to waive the fees if a commercial manufacturer, for example, were going to come in with an application.

There are a number of ways that a small institution, an academic PET center, for example, could qualify for a waiver under our normal statutory criteria. For example, we have a small-business provision that if a small business submits the first application for a human drug product, which is what this would be, they can qualify for a complete waiver of the application fee for the first application that they submit.

The criteria are they have to have under 500 employees including affiliates. Depending on the facility and all of that, there may be issues of that or if it is part of a big university system, or whatever. We could deal with that.

There are also exclusions for federal and state governments that are not getting profit out of it. So, for example, if NIH were to sponsor it, depending on how the agreement was set up, they might qualify.

Even if you don't qualify under the small-business waiver, then we get to the criteria that we actually used for this which was whether the fee would be a barrier to the protection of public health or a barrier to innovation.

There we get into, and we have already dealt a little bit, with the innovative aspect of this and PET. So the issue would become part of the financial test. But we are actually working on a rule on waivers to explain what our criteria are.

There are some guidance documents that are referenced in here, I think, on how we do waivers for user fees. But, again, we have no statutory authority to completely treat PET separately. If we gave a commercial entity, for example, a waiver of a user fee, I don't know how we would explain to the next innovator-manufacturer of some other kind of drug product why they shouldn't get the same kind of treatment.

So here we are concerned about the precedent that we might set.

MS. KUHS: I think you answered part of the question that I raised to begin with in that you do have a mechanism with this barriers to innovation. I think that there are two sides to that issue about whether fees are waived nor not depending on the size of the entity.

Representing a commercial supplier of this, it looks a little tenuous to me about a university or a hospital being able to introduce a new chemical entity into the PET drug market that might be prohibited from a commercial agency doing it. It is just the other way around.

So I think if we look at some ways of treating all of the people equitably in here, and maybe this five-year period, or whatever, would be a way to sort that out so that if those issues come up, we don't stifle the innovative process during this initial period while we are trying to figure out how this industry is going to sort itself out.

MS. AXELRAD: We are willing to work with you to try and figure out how these factors would be applied to whoever it is that would come forward with a new chemical entity, for example. But I just don't think we have the legal authority to agree to a blanket five-year waiver, period.

DR. CHALY: As I understand, if the production of a FDG increases millicurie per ml, compared to what was done in the NDA in Peoria, we may have to apply for the NDA rather than the ANDA.

MS. AXELRAD: It would not, however, be subject to a user fee because it would be a 505(b)(2) application in which you would be simply addressing the change in the product from the approved product to the new product. Those are not subject to user fees. That would not be assessed the user fee.

That is something that everybody has to understand, that it is only the very first application for a new active ingredient or a new indication that is assessed a fee. This would not be a new active ingredient. It is still FDG. It would have slightly different parameters that you would have to address in the CMC section, but it would not be subject to a new application fee.

DR. CHALY: So, suppose we are producing--because, in their applications, if it is 50 millicurie per ml, and they are now producing, because we have a better machine and we are producing 200 millicuries per ml, we don't have to apply for an NDA?

DR. LOVE: That is more or less--it depends. In your application is already approved, then you would just be submitting a chemistry supplement to update. If you wanted to change in the future, if it is your first application that you are submitting, it would probably come in as a 505(b)(2).

DR. CHALY: We haven't applied.

DR. LOVE: I understand. I am just saying, right now, it would be a 505(b)(2). In the future, if you got a 505(b)(2) with one process and you wanted to change to what you are talking about, not a process change but a strength change, then you would submit just a chemistry supplement to update that.

But it would still not be subject to the user fee issue.

MS. AXELRAD: It would not be subject to a fee in either event. It would be a 505(b)(2) application but, because it is for FDG, albeit at a much higher strength, it is not assessed a user fee.

DR. CHALY: Thank you.

MS. AXELRAD: And no ANDAs are assessed user fees in any event, period.

MR. FERRIS: To go back to the guidances, did you say that they were referenced in this? Is that the interim guidance, separate marketing application, that one includes the definitions of small business and the barrier issue?

MS. AXELRAD: I am not sure. I will have to check and see if we actually included this. Because we gave the blanket waiver, we may not have included citations to the guidance. If you hang on, I will look. It may not be in here, but we can provide it to you. It is actually on our website.

For one thing, there is a PDUFA page, if you look at the FDA's website, if you can find it. I have a hard time finding it and I don't have the site here with me. If you look around, there will be a page on Prescription Drug User Fee Act. The guidance is there.

MR. FERRIS: Would you be able to post that on the PET dockets?

MS. AXELRAD: Yes; we could put a link on that to the PET page. How about the PET page? There is a PET page, also, a whole separate page on that where the guidance is and everything. We can provide a link to that waiver document.

DR. CONTI: Are we going to move to the guidance, itself? Do you want any more user-fee issues? Maybe we can cover the guidance document, itself, at this point.

Maybe I will start it off by asking a question in terms of some specifics. I know that there are probably some comments about general issues on this document, but I think we have probably discussed them ad nauseam already.

In any event, the issue of cross references and patents. It says in the guidance that you may want to reference other applications or you may refer to investigational new drug applications. Then, under the patents, it says, "You are required to disclose any patents of yourself or anyone else."

One is a sort of philosophical question. What is really the difference there in the sense of if you have to find out what everyone else is doing anyway through a patent process, why wouldn't you be required to find out what else has been done with INDs and DMFs and is this really more of a requirement than you may want to do this.

MS. AXELRAD: The patent provision is put in here, and was put in the statute, under 505 specifically under the Waxman-Hatch Amendments, the 1984 amendments, where they made it possible for there to be generic drugs. As part of that whole setup, the patent holder had to notify the agency of any relevant patents and the potential generic applicant or a 505(b)(2) applicant was required, if they were coming in with an application, to make a certification as to whether the approval of their application and the marketing of the product would infringe the patent.

There is this whole elaborate process for suits and everything else. But it is, like, totally standard, according to that. The reason is that if you are going to come in with either a 505(b)(2) or a 505(j) application, we have to be sure that it is not going to infringe the patent. So that is the whole basis for that.

DR. CONTI: I understand the patents. I guess my question is really more related to, say, the drug master files and things like this. Potentially, if you are going to submit an NDA and you wanted to reference a DMF, what you are saying here is that you don't necessarily even have to reference the DMF.

MS. AXELRAD: If you can supply all the information yourself, in your application, you don't have to reference a DMF. A DMF is essentially a convenience established by the agency, for where you don't have personal access to the data because the manufacturer, for example, wants to keep it confidential, the cyclotron manufacturer--this may not be a great example--the cyclotron manufacturer might not want to tell you about all the workings and everything of the cyclotron or the box manufacturer might not want to tell you anything about that.

You buy the box but you don't know how it works. So you would reference the DMF for the box and, if we had any problems with the way they described the box, we would go to them. But they wouldn't have to share with the sponsor of the application what those details are.

Like stopper manufacturers, that is used for stoppers, and things like that. They don't want to share with you how they make that rubber stopper and what its composition is, but we need to know because we want to make sure that nothing is going to leach off the stopper into the vial.

So you would reference a DMF. We would go look and make sure that the components in that stopper are okay. If we have any problems with that, we deal with the stopper manufacturer and you are given a letter of authorization to reference it.

But if you have all the information about that stopper, yourself, and you want to put it in your application, that is fine. You don't have to reference any DMF.

DR. CONTI: I have a couple of concerns. One is that we would have to be aware of all these various DMFs, looking at the entire spectrum of operations. You wouldn't hold us to that?

MS. AXELRAD: You just have to supply the information that is in this format and that is required by our regulations. You can supply that any way you want. If you have access to it, you can supply it direct and it doesn't matter if there is a DMF out on the same thing. You supply us what you are relying on, period. If is only if you want to rely on something that you don't have the ownership of, then you have to get a letter of authorization to reference it. You may not get to see it, but we, then, could review it.

It is like filling in a piece of the application with data that somebody else owns and you get permission to use it.

MR. FERRIS: Typically this would happen with the membrane filter, for example, if one is going to use the millipore filter. Rather than doing the validation for pore size and pressure-hold test and all that type of thing, just reference the DMF.

DR. CONTI: If, in fact, the NDA, the current NDA for FDG, does reference the DMF, and I am sure it does, another person submitting an NDA at this point would have to use what as a reference; the current NDA? Would the information be within that document automatically accessible?

MS. AXELRAD: Which DMF. I understand that there were a couple of DMFs associated with the original application. One of them dealt with safety and efficacy data which doesn't need to be in any of these applications.

DR. CONTI: It is the chemistry.

MS. AXELRAD: The chemistry only? Well, in a way, we are saying--I don't know. Is there a chemistry that they would be relying on that would be separate from what would be in this model format application? What would be in that DMF that somebody might want to rely on that they wouldn't be able to supply themselves?

DR. CONTI: The reference-listed drug.

DR. KASLIWAL: For example, if I can intercede here, for container closure, you reference a DMF. And for your NDA or ANDA, you referenced that DMF. Another applicant comes and they would reference the DMF unless they can--

MS. AXELRAD: Ravi, I think he really wants to know--the question is whether another applicant would need to reference I guess it is the ICP DMF that established the chemistry for the original FDG. I think not because I thought that what we did here was tell everybody in the guidance document what that referenced, what the parameters were of that reference-listed drug so that people did not have to go to any other document.

They know when you look at the column here that says reference-listed drug, that is what they have to match if you want to be an ANDA, for example.

MR. FERRIS: I think what Peter was asking is whether or not, in the NDA, whether the NDA contained portions that reference the DMF held by ICP. I think that, and Jim you may be able to help me on this, what happened was we originally filed a DMF but then came back and rewrote the entire NDA so that it really didn't have to rely on the DMF.

MS. AXELRAD: We don't think anybody has to reference or rely on, in any way, the original NDA, period, or DMF. We think that the fact that there is an application out there approved for FDG is just there and that people now need to just tell us how they are manufacturing it in accordance with this format and we will evaluate it, period.

There should not be a need to rely on information that is either in the NDA or the DMF.

DR. CONTI: I have another question. Under certifications, page 13, for example, safety reports. Do you have examples that you could share with us, either templates that we could use as far as how to file safety reports or what the contents are?

MS. AXELRAD: I am sure we could do that. I am sure we could. If we don't have any, we could develop a model or something like that, although I am told that for Paperwork Reduction Act purposes I am not allowed to use the word "model" or "template." We changed the titles on these. I think "sample" is okay, but model and template are not okay.

MR. LEEDHAM: Usually, the safety report involves when an NDA is submitted in the time period there are still studies that are ongoing. For these particular applications, we don't see that there will be any safety updates.

DR. CONTI: There has to be some mechanism if there is an adverse event or something.

MS. AXELRAD: You mean postmarketing safety reports.

DR. CONTI: Right.

MS. AXELRAD: Yes; we can give you models of those, or something, that shows what it is supposed to look like.

MR. LEEDHAM: The MedWatch and so forth.

DR. CONTI: I have a couple of minor comments on the tables. Under the sodium-fluoride table, page 24, as we discussed with that dose, you have 2 millicuries per milliliter at calibration. So we will need to revisit that within the guidance document as per whatever we decide on as far as the labeling as well.

Also, in the FDG, let me get to the page number here, on page 34, strength of the generic product in tableÊ6. You have 4.2 to 0.22. There probably is a typo someplace.

DR. KASLIWAL: I think that is right out of the labeling, approved labeling.

MR. LEEDHAM: What you are talking about, Dr. Conti, is the reference-listed drug and the strengths. So those will remain the same.

MS. AXELRAD: You seem to be asking the question of whether it makes sense to say 4.2 to something considerably less. That is not usually the way you would express it.

DR. CONTI: I don't care if you reverse it or whatever, but it is not right.

DR. KASLIWAL: I think what that might mean is, at the time of calibration, it was 2.0 millicuries per ml and ranged starting from a higher to the lower.

DR. CONTI: It is confusing. It confused me.

MS. AXELRAD: Perhaps we could drop a footnote. I am not sure we will be able to change it if that is the way it is expressed for the reference-listed drug. But we could certainly drop a footnote explaining what we just said.

Does anybody in the audience have any other questions on specifics on the guidance document or the labeling, I guess we didn't ask.

MR. MATTMULLER: On labeling of the immediate container or vial--would you want all that information on a small 10 ml vial label? It is possible to print it that small, but I don't think anyone could read it. I think, in some cases, it has been allowed that there is a small label on the actual glass vial that has a lot number on it and maybe a current date calibration information. But then, the bulk of this information would be on a larger label that is supplied. That is actually what people look at because they typically don't pick up a radioactive vial and read the calibration data off of it.

MS. AXELRAD: What page? I'm sorry; do you have a page?

MR. MATTMULLER: Page 25 in labeling.

MS. AXELRAD: Thank you.

MR. LEEDHAM: Steve, that would be on the immediate carton, versus the vial. Certainly, the vial would need to be identified but not to this extent.

MR. MATTMULLER: Thanks.

DR. CONTI: I'm sorry; just to go back to the strength issue again. I apologize, but I think it would be helpful to make these consistent. On table 3 on page 23, you talk about strength, here, end of synthesis. And then there is no parenthesis in there for this higher to lower value whereas, on page 34 on table 6, you have this additional information and it is at calibration time.

It might be helpful to make these consistent.

DR. KASLIWAL: We intend to. Currently, our thinking is the end of synthesis, obviously, but the calibration time came whenever, twenty years ago, thirty years ago. That is how it is.

We think it is highly unlikely somebody is going to make sodium fluoride at 2 millicuries per ml today. So the very first application, more than likely, is going to be a 505(b)(2) which will probably change things.

DR. CONTI: I guess we are finished with the documentation; right?

MS. AXELRAD: Right.

DR. CONTI: I have to congratulate you guys. You did an outstanding job on all of this. It is very good.

MS. AXELRAD: Except for the title.

DR. CONTI: Except for the title. The title stinks.

MR. FERRIS: And the registration box.

MS. AXELRAD: Is that even in there? I thought we took that out. Is it in there or is it there by omission?

MS. KEPPLER: In the form, there is a spot that says, "Put your registration number."

MS. AXELRAD: You are kidding me. I knew it wasn't in the guidance because I said, "I am not dealing with this one on this particular document." If you notice, we fixed your GMP certification on the form, the 356(h) form. You were supposed to certify that you were in compliance with GMPs. But we fixed that.

To get the form changed takes like a year to go through OMB. So we just told you what to fill in instead of what you are supposed to fill in on the form.

MR. LEEDHAM: Versus certifying that you be in compliance with 210 and 211.

DR. CONTI: I have a topic I would like to discuss for a few moments. This is related to the continuation of our interactive process with the agency, specifically as it relates to the subsequent steps, which is the finalization of the CGMPs but also, beyond that, to the process of inspections and issues such as we raised last time as pertaining to things like inspections for cause versus routine inspections.

I think, for example, in that particular area, particularly in the interim phase, inspections for cause should be something that the agency considers in this interim period. But you still want to consider that as we move down the road because of the long historical track record for the safety of these pharmaceuticals.

The other thing I would also like to bring up is that we would like to continue to entertain the iterative process with the next level of PET pharmaceutical productions which is the RDRC and the IND mechanisms because a lot of this will have implications for the development of future drugs.

So I guess I am sort of reiterating a position and speaking for members of the committee here, and I am sure that they would concur, that this is a high priority for us, if anyone wants to say anything else on that.

MS. AXELRAD: We fully intend to have those discussions in the same way we have had the past discussions. We definitely want to deal with GMPs, I think, probably the next time we get together. We would like to talk about whatever the next version is of GMPs. Probably, we may as well, I guess, at the same time, discuss, perhaps, inspections, maybe, and the registration issue.

And then we will move into the other issues of the RDRC, INDs and how to bring the other new drugs into the process. So I definitely foresee this dialogue continuing for the foreseeable future.

DR. BARRIO: You mean the issue of profit/non-profit base as indicated in the law or--

DR. CONTI: Let's take this off the record for a second.

[Conversation off the record.]

MS. AXELRAD: We would be happy to hear it. Actually, there is something else I want to raise with you. We would love to hear how you think that we can take into account these differences between for-profit and not-for-profit at a future date if you would rather do it at a future date.

But we really would like to hear it. For one thing, we need to have a mechanism because we struggled with this, as David Horowitz said this morning, with how do we distinguish between a for-profit and a not-for-profit entity. We did some legal research on this and we found that it was a lot more complicated than you think, that somebody could have a not-for-profit number, a tax-exempt number, which is what is usually used to designate not-for-profit status, and, within the broad entity, say a university is not-for-profit, there could be a very profit-making PET center, for example, or some other group of doctors who donate a certain part of their services to a charity, for example, and make money off of everything else.

It is very complicated. We are a science agency. We are not--there is at least one economist here, but we don't have a whole large staff of either economists or lawyers who are going to be equipped to delve into the financial workings of these institutions to determine what these differences are.

So if we are going to draw any kind of a distinction, there has to be some relatively bright-line simple test that we can use that would truly distinguish a for-profit versus a not-for-profit entity.

Then I would like to know what consequences that would have, once you go through the distinction, what, exactly, would be different. One would register and one would not. One would submit an application and one wouldn't. One might follow GMPs and one might follow other GMPs.

Whatever they are, it would be really useful if you are going to tell us that you want us to make the distinction, to address both how we are supposed to make that distinction and what the consequences that would flow from the distinction would be, in as specific terms as you can, because otherwise you just toss the ball back to us.

If we can't sort it out and we are not able to do that in any practical way, we won't do it. So we need to be able to deal with it in a very practical way that would be applicable across the industry.

DR. BARRIO: We discussed this issue among ourselves, although not in great detail, but I think the intent of the law was to--I think the wording was bad, perhaps, at two or three different levels. But we have been asked this question many times, what does it mean, a non-profit entity.

I think the intent of the law was to say a cyclotron unit producing radiopharmaceuticals for physicians working in the same site is producing the radiopharmaceuticals not for profit, for the physicians who are going to use it in opposition to industry that is going to produce and distribute them and sell it for a profit.

I don't think the intent was to say whether we were, at the university or a hospital, or an institute that is for-profit or not-for-profit. In all honesty, probably most hospitals are for-profit. Probably most universities, even--in private universities, they have tremendous investments for profit in many places.

I don't think the intent was to define the institution, whether the institution is for-profit or not, but rather whether the cyclotron unit was producing that radiopharmaceutical at cost for a physician or for distribution.

That will simplify. If we use that approach, that will simplify the analysis a lot in my interpretation. I think it is going to be absolutely impossible to define, for most of working in different institutions whether these institutions are for profit or not. I think that is going to be very difficult.

DR. CONTI: I have a question. Maybe your attorneys have looked at this already. Within the context of what is actually written in the law, is there a flexibility for a consensus between the PET community and the FDA to actually subdivide things differently as opposed to profit versus non-profit.

Do we have that? Can we mutually agree that the segregation should not be based on that?

MS. AXELRAD: No. I don't think so. In fact, Dennis just told me this morning that we absolutely have to do what the law tells us to do which is distinguish between for-profit and not-for-profit entities. I heard that comment and we are going to go back and see if we can do that. And now you are asking if we can not do that.

MR. SWANSON: No.Ê Please let me put that back in the context it was made. You have got to address it. The law has spelled that out as something you have got to look at. In your materials that you put out in the public forum, you have got to address it.

It may be there is no solution to this. But you at least need to work through the considerations that led to that conclusion.

MS. AXELRAD: We have done that. If you want to suggest a paradigm or some way of cutting this that you think would be appropriate, we would certainly be willing to look at it and talk to you about it. The lawyers are not here anymore. We would have to look at that.

But, like I said, the thing that I would really beg you to do is make sure that it is easy for us, practically speaking, to draw whatever line you are trying to draw between the people that are one side and the people that are on the other side, so that, practically speaking, we don't have to hire a staff of economists and lawyers to do it.

It has to be something that we can practically implement and also that you are really clear on what the logical consequences of that division are in terms of how all these documents and things that we are producing would apply.

DR. CONTI: I think, though, the other key issue is identifying, within FDA what the concerns are traditionally in this type of discussion. For example, you are talking about true industry and the commercialization of drugs, what are the issues that you would want to follow or investigate in the commercial environment as opposed to a non-commercial environment, let's say.

For example, a large manufacturer that is producing pharmaceuticals all over the country, thousands of doses or whatever, there are probably certain things that you are worried about from a regulatory point of view or would be concerned about as opposed to the single cyclotron unit type of operation that might be run with direct hands-on supervision.

A large industry may have one regulatory person supervising a number of sites versus a single on-site person all the time. So there may be issues that, if you help identify, we might be able to work together to come to some sort of recommendation vis-a-vis your concerns.

DR. HOUN: I think in the Mammography Quality Program, we had to go through this similarly in looking at mammography facilities that see one woman a week or do four mammograms a week versus ones that do four an hour and what standards should those have, should they be different.

I think the concern was that quality for the patient shouldn't have to differ. So we, in that program, went through similar thinking about should we have some different standards because a person that does one mammogram a week versus one every fifteen minutes has a different type of operation, has a different type of load, has a different staff, has different finances, resources, et cetera.

I think we were stuck by the agency not being able to say the quality should be different. So I would be interested in hearing more from your community in terms of saying, if we only manufacture, we only produce, a PET drug once a week, twice a week, versus a large commercial entity that is producing lots and lots of vials, what are the relevant differences?

I think you are asking us, in our experience, where we have seen the difference. At least for this other field of radiology and production, we were not able to be able to tease a difference out that would not have a differential public-health impact.

DR. CONTI: I think you are probably going to find that to be true within the PET community as well because the quality really should not make a difference whether you do one vial, two vials, three vials a day, or batches. It is probably not going to make much of a difference.

Of course, the other unique aspect of our industry is that we do test each batch. So I don't think it is going to have anything to do, really, with the final product, per se, and the quality product that the patient is going to receive.

It is going to do more with issues such as supervision, such as compliance across multiple facilities, making sure that they are all the same, they are all following the same guidelines at each site and things like that.

MS. AXELRAD: In our GMPs, that is what we were trying to take into account. The new GMPs that we are writing were designed with the PET center who produces one vial a day in mind as opposed to the commercial manufacturer who is doing a lot.

I think we talked about this before, about whether we would have more stringent--we are trying to get to the sort of baseline GMPs for the small entities. If we wanted to have more rigid standards, perhaps, even affecting distribution, particularly, or wherever we saw greater risk associated with the operation, then we could do that.

But I think our aim, in terms of the GMPs that we are working on is to design a GMP that is reasonable for the person who is only producing one or two vials a day and then, if we need additional requirements for people who are of a broader operation or more of a distribution of the product, then we would go there.

DR. MARGOULEFF: I think you are entering into an area that is mainly semantic. I think for-profit/not-for-profit has much more to do with the IRS than it does with what we are doing. Just as you are not able to use the word "template," but talk about "sample," a non-profit institution can have a "surplus," but not a "profit."

Parenthetically, we have been running, for ten or twelve years, at a consistent loss. So if it were possible for to produce more than we used ourselves and we distributed it to other hospitals in our own hospital network so that they could share some of the costs, it wouldn't lead to a profit or even a surplus. It might lead to the fact that our deficit was not as large.

So I think you enter into a very foggy kind of an area if you try to do it based on these semantic considerations. I think it would be really best to look at how it affects, and I think you just dealt with that, quality and what we deliver to the patients.

Otherwise, I think you can get into a morass that you will never be able to get out of in defining, is my hospital for-profit? Is his not-for-profit? Is his semi-for-profit? A lot of us are making it a loss. Should we get special consideration because we always lose money? That would encourage people to try to show a loss every year.

So I don't think you can do it with the money end of it. I think you have to stay with the actuality of the production, safety and quality.

DR. BARRIO: I don't think it matters whether the PET center is making or not making money. Everybody is trying to make money. That doesn't make them for-profit. I think the issue, again, is not the PET center to be in for profit. I am trying to refer to the intent of the law.

The intent of the law was to protect individuals, a small percent of academic units and hospitals, from excessive regulations. And then the idea was to say, "Okay; if you are producing isotopes, radiopharmaceuticals, for your physician working next door, you are giving it to him or her not for profit because you are not trying to make a profit with the physician. That was the idea.

Now, whether the physician is making money in the PET center is not the issue, at least what the law intended because, otherwise, we would be all for profit and then we will be all regulated like industry.

I don't think that was the intent of the law. The intent of the law was just to differentiate clearly. But I agree. Everybody has a different interpretation, including me, of course. Therefore, it is very difficult to really come up with a single answer.

But I think the easiest way to deal with this is to consider the entity that this not-for-profit, the radiopharmaceutical unit or cyclotron producing the radiopharmaceutical and what was the purpose of that transference.

If I transfer it to the physician working next door, I am doing it at cost. If I go and distribute things elsewhere, then I am certainly trying to profit from that. But it is not the PET center. It is the radiopharmaceutical unit.

The other issue that comes with this is if I transfer radiopharmaceuticals to the physicians working in the unit, what does it mean? I think that is really more complicated because--what is a unit, by the way. The unit is all the physicians, for example, in the University of California from San Diego to Davis? Or the physicians working in the same medical center? Or the physicians working in the nuclear medicine or department of pharmacology, or whatever unit we have?

It may be that we have a nuclear medicine in three different places, controlled by the same physician group. Are they the same physicians? I think it is more difficult because it really is open to twenty different interpretations.

I think we really have a serious problem in terms of the language of the law and the interpretation. But I think--I don't know what your views are, but really simplifying this and having some sort of understanding of what the intent was and simplifying this and, perhaps, making clear to everybody what we are trying to do may be the way to deal with this problem.

DR. MARGOULEFF: If I could just make one more comment. I don't think there is any doubt that PetNet or Mallinkrodt are commercial entities and they are for-profit. I don't think there is any doubt that most of the hospitals and university centers such as the ones we work in have a logo that says that they are not-for-profit.

I think many of us have either contemplated or have made arrangements with either PetNet or Mallinkrodt or one of the other companies for us to make isotopes in excess of what we use, ourselves, transfer it to them and they, in turn, will distribute that in their for-profit organization.

That is a reality. I don't think that is a secret to anybody here. I think that is the nut that you can't crack by making these definitions. What becomes of that? It is made on one entity. It is transferred to another entity. One is for-profit and one is not-for-profit. Where is that?

MS. AXELRAD: We hope you will tell us the answer to that question if you want us to draw that line.

DR. MARGOULEFF: If I knew the answer to that question, I would patent it.

DR. CONTI: Interestingly, one of the answers is that it becomes the practice of pharmacy and medicine at that point because a lot of that is really compounding from a manufacturer. So to regulate that aspect of the business could potentially get into more serious conflicts with other entities, other agencies.

MS. AXELRAD: We have our own problems on the compounding side. I said at lunch, if you think PET is difficult, compounding is ten times more difficult because we are trying to draw the same line there between compounding and manufacturing.

When you have a pharmacy who is under--the practice of pharmacy setting up a small factory that is distributing albuterol around the country and saying they are doing it under the practice of pharmacy and they are 100Êpercent pharmacy compounding, is that manufacturing or compounding?

The MD Anderson Cancer Center, and I use this as an example all the time, or any of the other cancer centers who is mixing up large batches of chemotherapy agents, what is that, manufacturing or compounding? It is very difficult. We have not succeeded yet in drawing that line on that side. We are trying to do it there, too.

But they are really difficult questions. Just tell us what you want to do and we will see what we can do, is basically what I would say.

DR. CHALY: It is not very clear to me who should apply for an NDA and who should apply for an ANDA. I was reading on page 4 on this guidance for industry and it is not very clear to me who should apply for an ANDA and who should apply for an NDA.

MS. AXELRAD: You should apply for an NDA if your product is the same, within the way it is described in the guidance document, as the reference-listed drug. If, when you fill out those tables in there that say, "This is the reference-listed drug," and then there is a column for the generic, if the strength is the same, if the active ingredient is the same, if the route of administration is the same, if all the parameters in that table are the same, then you should apply for an ANDA.

If there is anything different, if your strength is broader--what are some of the other things that make it different? The composition is different, there are certain things that are different, then you come in and apply under 505(b)(2). They are virtually identical. It really doesn't make a huge amount of difference in terms of what you are completing in terms of the paperwork or anything whether it is a 505(b)(2) or a 505(j).

If you have any questions, I would suggest you contact the people who are in here and they can go through your particular situation and help you determine whether it should be an NDA or an ANDA.

DR. CHALY: This is the same question I asked you before because I said that, in developing this NDA, in the making of FDG, the percentage of the production was very low at that time. Now we have sophisticated synthesizers available and the strength has gone up considerably.

So it cannot be exactly an NDA because the mission is producing a better product, we ought to apply for an NDA. It is not fair. That is what I am saying.

MS. AXELRAD: It doesn't make any difference. It is virtually the same submission. The chemistry is exactly the same, whether it is an NDA or an ANDA.

DR. CHALY: But you said the strength cannot change.

MS. AXELRAD: The strength can change. If the strength is different than that for the reference-listed drug, then you call your application a 505(b)(2) instead of a 505(j). That is the major difference. There is almost no difference between the two things.

You say that the strength is different and you address the strength; end of subject. It is really not that much difference.

DR. CHALY: But in 505(b)(2), it says, that it is an NDA.

MS. AXELRAD: Yes. But a 505(b)(2) NDA and a 505(j) application are almost identical. There are no real substantive differences in what you submit, whether you call it a 505(b)(2) NDA or a 505(j) ANDA. We are not talking about a regular NDA where you have to go do a drug-development program. It is really almost a semantic difference between the two.

There are very small differences between what you submit. The chemistry is basically the same and that is the bulk of what you are going to be submitting.

DR. HOUN: Maybe it is time, John Friel, if you want to introduce yourself and your role in the agency so that if there are people who need help, they can contact you.

MR. FRIEL: I am John Friel. My position is Deputy Director of the Office of Training and Communications in the CDER. What we try to do is look out for the--we like to think of ourselves as an advocate for our industry, or our customers. Maybe that is better word today.

We are in charge of the website. We are in charge of taking calls that you might want to place to us to try to help you figure out what you need to do. A lot of times, we will be needing to triage your calls to other people. We are not necessarily scientists in our shop.

Our mission is education and communications with the outside world. So we are listed in the Federal Register notice as the contact point for this regulation. So we would be happy to hear from you, take your questions and, again, try to help you sort out who else you might need to talk to in the center.

We would also like to have any thoughts you might have on the website. I think we have had some good suggestions here for some linkages that we need to put onto the PET site. I think we can certainly work on that and get those up as quickly as we can.

Any other suggestions you might have along those lines are welcome. We are also in the business of helping put together the various meetings and outreach that we are doing. We are in the process of putting together a packet to go out to all the PET centers which will include various bits of information.

We are planning a number of visits, in the upcoming months, to get out to annual meetings and seminars to talk about this subject. We would be happy to hear any suggestions you might have along those lines, too.

My phone number is 301 827-1651. Again, it is John Friel. If you want to reach me by E-mail, it would be frielj@cder.fda.gov.

MR. SWANSON: A general question. Is the FDA, and maybe you don't want to answer this, but is the FDA really looking down the line--because it seems to me you have got a problem coming up with pharmacy compounding with PET with autologous cell therapy, with autologous gene therapy, where fundamentally, in the future, you are going to see classes of drugs which, by the virtue of those drugs, have to be manufactured or prepared at or near the sites where they are administered.

This is just the tip of the iceberg. So are you thinking along the lines of how you might solve this problem in general? Maybe you talk about a whole new registration of limited distribution facilities or limited distribution centers with a new set of standards for those limited distribution facilities.

Has anybody given any thought to the broader picture here? Are we trying to attack each one of these as separate individual problems because, if you think, we are going to see more and more of this down the line.

MS. AXELRAD: It comes together with me for both PET and compounding. Obviously, the two things are sort of complementary or related. People are looking at how to regulate these other newer gene therapy and some of the other new emerging technologies and things like that.

So I think that the agency is looking at that and will--sort of deriving lessons from our experience with PET and our experience with compounding to see if there is a broader--what the common issues are and the common problems and see if there is some way of addressing them that would solve the problem.

MR. SWANSON: It might actually solve a lot of our problems with orphan drugs also which is another consideration. It is basically a limited distribution problem, also. In the past, the problem is trying to produce those drugs and comply with CGMPs and produce them in some kind of a cost-effective, a reduction-in-cost manner.

So it just seems like this is just one category of many that we have similar problems with.

MS. AXELRAD: One of the things we need to do is bring in the patient groups because one of the people that hasn't really been at the table, and we may try and get some people to be at the table, are patient-advocacy groups that clearly should have some say here in what requirements are going to be relaxed in the area of these emerging technologies.

For example, for orphan drugs, they are treated like regular drugs. There are incentives provided in terms of exclusivity and things like that for the development of them but the reality is that they submit new drug applications just like everybody else. And they certainly follow GMPs.

I believe that the NORD and the patient-advocacy groups for patients who are suffering from rare diseases would not support anything less with regard to those. So I think that we need to bring in the patient-advocacy groups when we are dealing with some of these emerging technologies and how we are going to treat them and have them be at the table while we are making some of these crucial issues that are going to affect patient care.

MR. SWANSON: There is no question you have to have the quality of these drugs. But you take the cost of compliance with the NDA requirements and the CGM requirements and you spread that over a traditional drug where you are dispensing hundreds of thousands, if not millions, of doses, it ends up being much less cost impact on that drug than an orphan drug or a limited-distribution drug. That is fundamentally the issue that we are facing here.

MS. AXELRAD: Is that pretty much it? I want to thank everybody for coming--did you want to say something?

MR. MATTMULLER: I'm sorry. I have got an early flight so I won't be long. To mirror what Dennis is saying, probably the biggest concern I had coming here was determining what I would call performance GMPs, which most of us are actually operating underneath right now, versus what I would call FDA prescriptive GMPs.

I would challenge the FDA to show data that Peoria's FDG is any better than the FDG that I make or Pittsburgh makes. Given the lack of data showing there being a quality difference, and, by any means, none of us is arguing for lower quality FDG.

What we are arguing for is the way to ascertain that quality, whether it be performance versus prescriptive. If their FDG is no better than ours, why, as a small institution, non-profit, do we need to burden--or what additional gain do we get from the additional burden of prescriptive GMPs.

MS. AXELRAD: Thank you for that comment. We are grappling with that issue in the area of GMPs. We are trying to define a whole new set of GMPs--we are not using 210 and 211. We started from scratch--to try and define an appropriate GMP regulation to deal with specifically the small academic PET center that produces one or two vials a day.

We will be having a lot more discussions in the next few months about how is the most appropriate way to do that.

MR. MATTMULLER: Thank you.

[Whereupon, at 2:35 p.m., the meeting was adjourned.]