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

Combination Products

  • Print
  • Share
  • E-mail

Section Contents Menu

Meetings, Conferences, & Workshops

Transcript of Workshop.

DRUG INFORMATION ASSOCIATION
and the
FOOD AND DRUG ADMINISTRATION
CROSS LABELING WORKSHOP

COMBINATION PRODUCTS AND MUTUALLY
CONFORMING LABELING

Tuesday, May 10, 2005

8:35 a.m.

Marriott Bethesda North Hotel
5701 Marinelli Road
Bethesda, Maryland

 

C O N T E N T S

Welcome and Meeting Introduction:
Mark Barnett

Opening Remarks:
Murray M. Lumpkin, M.D., M.Sc

Perspectives on Cross Labeling:
Suzanne O'Shea, Esq.

Public Health Panel: Speakers

Public Comment and Open Discussion on Public Health Issues
Moderator: Mark Barnett

Legal Panel Discussion: Speakers

Public Comment and Open Discussion on Legal Issues
Moderator: Mark Barnett

Closing Remarks: 
Mark Kramer

 P R O C E E D I N G S

Welcome and Meeting Introduction

MR. BARNETT: I would like to welcome you to this FDA/DIA Cross Labeling Workshop on Combination Products and Mutually Conforming Labeling. I am Mark Barnett. I am going to be serving as your moderator today.

Let me first briefly discuss the issue that we are going to be talking about today and then I will let you know something about the format we are going to be using for this meeting.

We are here, basically, to explore the cross labeling of combination products; that is, products that may combine a drug and the device or biologic and a device or a drug and a biologic. The basic question is what kind of labeling should be required when a new product is intended to be used with one that is already approved and the two of them are going to be continued to be separately manufactured and marketed.

We are going to talk about two aspects of that issue. One of them is the public-health concerns and legal ramifications that may arise if the two manufacturers are not cooperating and jointly developing adequate labeling fort new combination product.

We are going to do this in the context of a hypothetical situation involving two manufacturers, Company A and Company B. This hypothetical is outlined in the Federal Register Notice announcing this meeting and you will find that in your handouts. Dr. Mack Lumpkin is going to be reviewing and further explaining this hypothetical in just a few minutes in his opening remarks.

There are also some related questions in the Federal Register that the speakers today are going to be addressing in their presentations. What we need from everybody here today, and I am talking about the panelists, the people who have signed up to speak from the audience, and anybody else in the audience who wants to speak up later on his advice on whether the FDA ought to be requiring that labeling of Product A and Product B conform to each other and, if so, when that should happen.

So, in a sense, your assignment for today is to put yourself in FDA's shoes and tell us what you think our policy ought to be about this. FDA's assignment is listen to what you have to say, go home and digest it and to later take it into account as we develop a policy on this cross-labeling issue.

Now let me explain the format we are going to be using today for this meeting. As I just mentioned, Dr. Lumpkin, the Acting Deputy Director Commissioner for International and Special Programs who is sitting up here on the platform is going to being with some opening remarks. Then he will be followed by Suzanne O'Shea, the other person up here on the platform, who is a Product Classification Officer in the Office of Combination Products and she is also the hard-working, diligent, multi-talented Program Chair for this workshop.

Suzanne is going to give you a little more background on the Agency's perspective and why we think it is important to develop a policy on cross labeling.

After that is over, we get into the meat of the program. That is going to consist of two separate sessions. The first one, this morning, is on the public-health issues in cross labeling. That is going to take up the rest of the morning. Then, in the afternoon, we will have the second session on the legal issues.

In each session, we are going to begin with prepared presentations by speakers who represent the FDA, the pharmaceutical industry, the Combination Products Coalition and AdvaMed plus a few other speakers from government and industry who have some experience in these areas. These formal speakers are going to be joined by a panel of FDA managers who whose programs are directly affected by these issues and who are working on developing the policy for the agency.

After these formal presentations, we will have a short break. You will be able to get coffee and cookies outside. Then we will come back and hear from a number of people who registered to speak in advance to make a statement. You will find a list of these people in your conference materials.

If you have made a reservation to speak, please be available at the right time. Come up to the microphone. You know the drill about that. After we have heard from everyone who is registered to speak, we are going to open the floor to anyone else in the audience who has a question or a comment for the panel. I am hoping, at that point, that we get some lively discussion.

Again, we are going to follow this procedure twice, once in the morning and then again in the afternoon. Lunch is going to be served from 12:30 to 1:30 right out here and, at the conclusion of today's meeting, we are going to hear some closing remarks by Mark Kramer who is the Director of the Office of Combination Products at FDA.

So, at this point, we are ready to begin and ask Dr. Lumpkin to give his opening remarks.

 Opening Remarks

DR. LUMPKIN: Good morning to all of you. I often feel like, at this hour of the morning, you need to be like that show--I don't if many of you have seen it; The Extreme Home Makeover Show. It was on last night. They have this huge great bullhorn and they go to the house, and they go, "Good morning, Johnson Family." And everybody comes running out and all of this.

I have always thought that would be nice, something like this at 8:30 in the morning to wake everybody up. We do, on behalf of Dr. Crawford, on behalf of Mark Kramer and myself and all of the colleagues that work with us in the Office of Combination Products.

I do want to welcome you here today and want to sincerely thank you for taking time out of your schedules to come here, as many of you have, to fill this room and to work with us today on this issue of combination products.

Obviously, you guys have looked at this issue. This is an issue that is important to you or you wouldn't be here today and it is clearly an issue that is important to us. It is not a new issue. As many of you know, the issue of combination products is one with which, I think, both the private sector and the agency have struggled for many, many years.

It is one of those quintessential FDA conundra where I think we often feel that we are trying to fit a square peg into a round hole both in terms, often, of science, in terms of regulation, in terms of the law, in terms of proprietary interests. It is what makes it interesting but it is also what makes it very, very difficult and a struggle to try to come up with some consistent policies and policies which are good for the public health of our country but also policies that recognize all these other interests that are clearly at stake in this particular issue.

So what we would like to do today is look at the issue of cross labeling. I want to thank Mark and the people in the Office of Combination Products for taking this issue on. As many of you know, several years ago, Congress mandated that the FDA form an Office of Combination Products. It gave certain statutory responsibility to that office. Mark and Suzanne and Patty and Patricia and all the others whom you meet here today from that office have been working diligently to deal with a host of very, very complicated issues involving combination products.

Many of you are seeing the fruits of their work, whether it is there proposed rule, and, hopefully the final rule on primary mode of action. It is dealing with issues like adverse-event reporting. It is dealing with user fees that are associated with combination products if it is one application, two application, the advertising regulations, what applies.

There are, it seems, an unending array of issues with combination products that really need focusing on and this group is trying very, very hard to do that. Today is part of their effort and, hopefully, it will now be part of your effort to help us all deal with the issue of cross labeling and what this means to these products.

In order to help us focus our discussion on this and to try to look at it from all the different perspectives that play, as you know, in the Federal Register, there was a hypothetical problem that was put forward. I think what we want to do is try to focus our discussions around that hypothetical problem this morning.

It is really kind of the worst-case scenario. Obviously, if you have got a combination product and the two manufacturers are working together and everything is wonderful, you can make the square peg fit the round hole much more easily.

But, in the hypothetical that we presented in the Federal Register Notice, it is very, very difficult. When we have been faced with these kind of situations in the past, obviously, they are issues with which we have struggled and with which we have not yet come up with a final policy and that is why we are here today.

The hypothetical is as follows. Here is Company A. It could be a drug company. It could be a biologic company. It could be a device company. But, for the sake of argument, let's just call it a drug company.

Company A has a drug that has an authorized new drug application. They are on the market. They have a label. They are legally out there marketing and selling their product and people are using it. Company B, which has no relationship whatsoever with Company A, develops a device with which one can deliver the product that Company A makes in a way that is different from what the approved labeling allows.

It could, perhaps, be a new indication. It could be a new method of delivery. It could be a new population. But Company B has this device and they are more than willing to develop the data to show that using this device and this drug together, at the end of the day, is safe and effective and they want to market this product and they want to be able to make claims about using their device with Drug A.

But Company A says no. They are not interested in being in any kind of business relationship or any kind of regulatory relationship with Company B. They simply have their product. This is the way this product fits in their corporate portfolio. This is the way they wish to look at this product. They know the liability that this product, when it is used, is the way that it is authorized.

They are willing to assume that. They, for their own corporate proprietary reasons, have no desire to be involved on this new device.

But there are those who think that, perhaps, there is a public-health benefit for having that drug used in combination with the device from Company B. Company B, clearly, has its own proprietary interests and its own desires as far as marketing their device and having an ability, legally, to make claims about how to use that device with drugs that are available on the U.S. market.

So that is our conundrum. The issues that surround that particular hypothetical are ones we would like for you to think about today. You are going to be hearing from a host of different perspectives. I think, for those of you have already looked around the room, you have got very senior leadership from the Center for Drugs, from the Center for Devices, from the Center for Biologics.

You have got the Deputy Counsel with us today from FDA. And you have got people from the private sector, from big companies, big PhRMA companies, big devices companies, from smaller device companies, people in the Food and Drug bar within our country.

There are people who are going to be presenting to you a whole series of perspectives. I think, from our perspective at FDA, there are really kind of four different elements that we have here. These elements, generally, we tend to think of them as being complementary interests. But, as the hypothetical has pointed out, there are times when they can become competing and then that is where it becomes a struggle for us. How do we balance (inaudible).

These are interests. Clearly, there is our interest in promoting innovation because we believe there are clearly times that this promotes the public health and our interest to protect the public health.

There is clearly the interest that we all have making sure that we all observe the law. Even if, at the end of the day, we don't think the law takes us to a good public-health outcome, it is, nonetheless, the framework within which we have to work and if, at the end of the day, the public health is not served by the law, we obviously don't just decide to break the law. We have a system for changing the law to make us give us the good public-health outcome that we want.

That is one of the questions we have for you today. Perhaps, at the end of the day, that is where we are, the sum of the problems that combinations confront us with.

The third interest is, obviously, the proprietary interests of the companies. There are certain proprietary rights and interests that clearly we, as a community and we as an institution have to respect and how does one respect the proprietary interests of companies and also deal with these other complementary or, perhaps, competing interests.

Finally, the fourth one is, I think, none of us at FDA want any one to short-cut good science or to short-cut statutory intent by what, at the end of the day, might be seen as regulatory gain (inaudible).

So here we have got all of these what should be complementary interests at times being competing interests and that is what we want to spend time on today. We want you, as Mark said, to put yourselves in our shoes. I mean, the many times I know those of you who are not from FDA here must sit back in your office and say, those people in Rockville, if I were at FDA, this would be easy. This is what I would do.

This is your day to do it. Okay? Pretend like you are one of us. These are issues that you have before you. What would be your way of approaching this? We are here to begin our discussion with the larger community on this particular policy item and we are, indeed, not bringing to you today answers. We are bringing to you today real-life situations that we have to deal with and we are very keenly interested in what you have to say.

Some of the questions in the Federal Register, just to remind you, we would be very interested in knowing how big a concern cross labeling really is to those of you in the private sector and, perhaps, would shed some light for us on why there are situations where companies choose not to cooperate.

I know, when we sit in Rockville and we look at these situations, it is easy for us to say, well, if I were at this company, this is what I would do. We would just cooperate with these people and this would be easy and we would go forward. But, perhaps, there are, indeed, legitimate reasons why there are times when companies choose not to cooperate. That is something that we have no expertise on. It would be very helpful to us if you could shed some light on why companies, perhaps, at certain times choose, for their own corporate reasons, not to cooperate.

Is there anything that FDA could do to encourage cooperation. If there are reasons that companies choose not to cooperate, if there are regulatory impediments, if there are regulatory policies, if there are things FDA could do to address some of those issues, we would be keenly interested in what you think those are.

From a public-health perspective, we would be interested in knowing how consistent you believe labels have to be in order to meet the public-health goal of not having consumer or practitioner confusion. So when we talk about conforming labels, when we talk about consistent labels, what do we mean? Do we mean word-for-word consistency? Do we mean message consistency?

What do we mean by conformity when we are talking about labeling, cross labeling, combination products? There are some special issues that people have brought to our attention about combinations and some of the critical problems that can arise after marketing. One of them is what happens when a drug is reformulated, if the companies don't have a working relationship and the reformulation actually then changes the characteristics of the way the device delivers the drug.

What happens in that situation? How do you make sure that doesn't become a public-health problem because the drug-device combination is now not performing with the characteristics that the safety and efficacy trials that were done would lead one to believe should happen.

So what happens when drugs are reformulated and there is no business relationship between the drug and the device company? From the device side, what happens, for example, when you have a reusable design and the labeling gets lost? It simply doesn't exist anymore. What is available at that point in time? How should one handle the instructions then for using a reusable device on combination with either a drug or a biologic?

There are some very interesting tricky eagle problems that we are interesting in hearing what people have to say and that is, if, in the situation where there is no relationship, no propriety relationship, between Company A and Company B, and Company A specifically says you cannot refer to the data in my NDA.

If FDA chooses to approve Device Product B, are we, indeed, relying on Company A's proprietary data inappropriately in way? And, in approving Product B, if we don't change anything in the labeling of Product A, has FDA misbranded Product A by its actions on Product B?

There are clearly issues relative to exclusivity. There are clearly issues related to the authority under which a product is approved and whether, and when, and if, and how, any generic version or follow-on version of that particular product might be authorized by FDA.

We would be very interested--particularly you will hear both the private sector and the FDA legal authorities talking about these issues and thoughts on how we might begin to deal with those issues.

Just a couple of things to, I think, keep in mind as we go forward so we can stay focused on the policy issues and the legal issues that we are talking about here. Let's just assume two things. Let's not get into big debates on whether, indeed, there will be adequate clinical safety and efficacy data to support the combined application.

There are clearly issues of how you get that. There are issues of whether one has to meet a substantial equivalence standard or whether one has to meet a substantial evidence standard or a reasonable evidence standard or whatever the legal and scientific standard is.

But let's assume that, for the combination product, that there will be adequate clinical safety and efficacy data because that is really not the focus of today's workshop.

Let's also not worry about where the product will be reviewed in FDA and whether FDA will have the expertise to review the product. Let us assume, as all of you know, that we actually talk to each other across centers at FDA and that we have the authority to do consults across centers and that all of our centers have the authorities to use the different regulatory authorities of all the other centers.

So let's not worry about will FDA have somewhere within its remit the expertise to review the product. Let's assume that we will for purposes of today's workshop.

As I mentioned a little bit earlier if, at the end of the day, you think this square peg is simply too big for this round hole and that we need to figure out a new way to go forward, we need a new statutory paradigm for this particular situation, then we would be interested in hearing that. Clearly, we can't change the statute. Only Congress can do that.

Obviously, we are trying to make this situation work given the regulations that we have, given the laws that we have, given the good science and practices that we have at this point in time. But, if you think we just can't get from here to there, that we need a fundamental change in the paradigm, by all means, we would be very interested in that particular comment.

That is kind of the remit for the day. This is a very challenging issue. This is one that, were it easy, I think we would have solved it many, many years ago. But it is one that is becoming more and more apparent and more and more acute because our impression is that these combination products are one of the waves of the future, that, indeed, this is where a lot of the innovation is going, where a lot of the new technology is taking us, of how we use older products in new ways and new ways of delivering them that make them, hopefully, more efficacious, more safe and, perhaps, able to be used in situations where they could not be used in the past, all marvelous public-health outcomes but outcomes that are often difficult to get to from a regulatory and statutory perspective.

I would like to finish by, again, thanking, first of all, Mark and Patricia and Suzanne and Patty and all the people here from OCP for having the courage to take this on and to have this meeting here today and try to begin to work through with the broader community coming up with an answer that we all can live with on this particular problem.

You are also going to be hearing from people beyond the Office of Combination Products. We have, as I mentioned, people here from the Combination Products Coalition, from AdvaMed, from PhRMA. There are speakers from all three human medical product centers at FDA, people from the Office of Chief Counsel and other parts of FDA.

I want to thank them for taking time for being with us here today because this is, obviously, as you can tell from the perspectives, something that all of these different perspectives have a very important role to play.

Last but not least, I want to thank you. If you were not here today, we obviously would be in even a deeper struggle than we are at this point because, indeed, you represent a perspective that is clearly important for us as we begin to go forward in this area of policy development.

So let me encourage you to have a lively discussion. I don't want to have to get the bullhorn out. We want people to be awake. We want you to use these microphones. We want you to challenge each other. We want you to put your ideas out here so that we can begin to synthesize the discussion and the ideas and begin to come forward with some policies and some new policy guidelines and guidance on these particular issues.

I want to thank Mark Barnett. For those of you that don't know Mark, he has been a marvelous M.C. for a lot of our internal discussions that we have had and a lot of our public discussions. I think he will do a very good job of keeping you focused and, when you kind of begin to get out into the periphery of the issue, Mark will be one to bring you back. So I want to thank Mark for doing that.

Again, thank you all for coming and we look forward, very much, to hearing from you today.

Thanks very much.

MR. BARNETT: Mack, thank you for your usual articulate and very well done preview of what you are going to be doing today. Mack kept referring to the courage of Mark. But the Mark with the courage is the one down there, and that is Mark Kramer, the Director of the Office of Combination Products. I am the one without the courage.

Anyway, our next speaker is Suzanne O'Shea.

 Perspectives on Cross Labeling

MS. O'SHEA: Good morning. Thanks, Mark. It is good to be back. I am Suzanne O'Shea from the Office of Combination Products. Mack mentioned to you that this problem has been around for a long time. I thought I would just confess that I came to the Ombudsman's Office in November of 1992 on a detail from CDER. Amanda Peterson and Steve Unger, at that time, handed me a stack of paper about an inch high and said, see what you can figure out about this cross-labeling problem.

So I am delighted that this conference is happening and finally getting this item off of my "to do" list. But, in the ensuing 13 years, that inch-high stack of paper has turned into about a 2-foot high stack of paper. So I thought I would just give you a little bit of background about what is in some of those sheets of paper that will, hopefully, help us frame the discussion today.

Why do we think cross labeling is an important problem to work on. As Mack alluded to, we think there is certain amount of confusion in the world whether cross labeling is required in the situation laid out in the hypothetical; that is, where there is no cooperation between Company A and Company B.

We are a little concerned that the confusion, itself, may deter development of innovative products and so, hopefully, we would like to clear up the confusion. Is cross labeling required or isn't it? We would also like to get the process to be a little bit more efficient on deciding if cross labeling is required because, when the problems do come up to our office--usually, these days, it is the Office of Combination Products, it is time consuming for all of you and time consuming for us to figure it out. So we would like to be a little more efficient in it.

A quick note on terminology, just to clarify. Sometimes you will hear the word "cross labeling" used. Sometimes you will hear "mutually conforming labeling" used. Is there any difference between those terms? As far as I know, there isn't any difference. They mean, to me, exactly the same thing which is not completely clear in itself.

But I think mutually conforming labeling may be slightly more descriptive in that it does give you the sense of you want the two products to have the labeling conform. The cross labeling sort of sounds like the old scarecrow thing; he went that way. So, just to be clear, I think we are talking about the same thing when we use those two different terms.

I would like to just run through some of the different situations where we have seen cross-labeling issues arise. This may give you a little bit of a perspective on why it is so confusing, because the situations are different and different resolutions have come up depending on what the situation is. Throughout the day, I think it would be helpful, if we want to refer to Product A and Product B and Company A and Company B as a shorthand for what we are talking about, even though we all know that we don't really mean a drug company and a device company, but that does seem to be the way it works most often.

But on one end of the spectrum, sort of at the least troublesome end of the spectrum, is where Product A enhances the safety or efficacy--or Product B enhances the safety or efficacy of Product A. Same indication. Same patient population. Same dose and same route of administration. But it is a cross-labeling issue whether Product A should refer to Product B.

Slightly one step harder on the spectrum is where Product B uses Product A in a new route of administration. That can be a fairly small change such as between intramuscular and a subcutaneous route of administration but may have impact. Or it could even more dramatic from a Product A approved for I.V. administration where it would be administered intrathecally with Product B. Those raise obvious safety and efficacy concerns.

It is different even when it gets a little more complicated is when Product B uses Product A for a new indication. These indications, again, can be fairly closely related such as if Product A is already approved for treatment of a certain condition and then Product B comes along and wants to use Product A for prevention of a the same condition. That may be an example.

They may be wildly different indications which makes it even harder. You may have a product for heart disease that Company B wants to use it for wound healing or something like that. Those present very different issues.

You may have a new patient population involved where Product A is indicated for Stage 4 cancer and, with Product B, they want to use it for Stage 2 cancer and there are obviously different kinds of concerns there. It is very much dependent on the facts.

We have seen cases where Product B is a new component of an already existing combination product. Sometimes, we will have a combination product that is reviewed under two applications. The riboviron interferon product is an example of that. What would happen of Company B wanted to come in and make its own riboviron product to be used with the interferon product but the interferon product won't be labeled? That is cross labeling.

Another example is when a drug product is approved with a specialty delivery device and they are both approved under one NDA. What happens when a new device company wants to come in and make its own delivery device to be used with Product A? We are back to cross labeling again.

The differences in the labeling can be an issue for us. Sometimes, the labeling between Product B and Product A will just be slightly inconsistent. For example, we have seen cases where Product A contains a caution against using a product in a certain way and Product B comes along and wants to use it in this way that is cautioned against.

We have also seen cases where Product A is contraindicated for the way it would be used with Product B. So the labeling of the two products would be just clearly inconsistent, contradictory. Those might pose different situations with different solutions for us. So you have to think of all these different possible situations when we come up with our policy.

Another way to slice this problem is by considering the categories of issues. Mack alluded to these but it has been helpful to me to consider the different categories. The three main categories that we will go through in just a minute in more detail are the labeling issues, the no ongoing relationship between the two manufacturers issues and the pathway issues.

The labeling issues come up--they are the issues that derive from the fact that the physical labeling, the piece of paper that comes with the products, are inconsistent or contradictory. Some of the concerns might be that the end user could be confused by that contradictory labeling.

One of the things that we consider in this situation, or we might consider, is how awful it would be if the end user inadvertently used Product A as directed in Product A's labeling and so whether that would cause great harm to the patient if they used the wrong set of instructions.

The labeling for Product B could be lost and then there might not be any directions anywhere for how to use the two products together. Co-packaging has sometimes been offered as a solution to the labeling set of issues. I might ask you to consider that today.

To a big extent, I think mutually conforming labeling or cross labeling is sort of a short hand for the whole set of issues that arises from the fact that there is no ongoing relationship between the two companies. This is a concern not only during the approval process but through the life span of the two products.

So we have the issues, the approval product's proprietary information, whether we are inappropriately relying on Company A's data in approving Product B, the issues that Mack alluded to, the device redesign, drug reformulation may be an issue. Is it even possible for Company B to appropriately monitor Product A to make sure that there are no changes that make a difference.

So there will be degrees of cooperation that two companies can engage in. They might engage in cooperation for the approval process. They might give a right of reference. There are lots of different creative possibilities so we would like to hear about them.

Then the pathway issues that we have. 21 CFR 3.2(e)(3) is sort of the central pathway issue. I am sure everyone in this room has memorized this regulation. But, in case you haven't, it is included in your package. But, buried within this regulation is sort of the core question here of when does the labeling for Product A need to be changed.

That is something we would like to talk a lot about today is when it is necessary for this labeling to be changed. And then the issue becomes, according to this regulation as we read it, when the labeling needs to be changed, the two products become combination products with all the implications of being a combination product including assignment to a center based on primary mode of action.

When the labeling for Product A does not need to be changed, then the two products are two separate individual products that can go their separate ways.

Then the other pathway issues are all the legal issues that we will hear about this afternoon. They are sort of bumps on the road on the way to figuring out whether we are able to permit Product B to get onto the market without conforming labeling from Product A.

What is not an issue today--we ran through these earlier today. I will just reiterate them here; a conclusion that cross labeling is not required is not a data shortcut. We will require adequate--it is a presupposition here today that adequate safety and efficacy data will be available. The differences in the types of marketing applications we prefer not to get into today. We would like to assume that whatever information is required we could get under whatever kind of application it is. And we would also like to assume that there will be active consultation and collaboration across FDA centers.

Why is this such a challenging issue? Company A's proprietary interests. One of the little oddities of this whole situation for us is that we don't hear, at FDA--we don't hear from Company A very much. The people who come to us are Company B who wants to get on the market. So we would be delighted to hear from Company A about what their interests are.

We think we have our ideas but we are not sure. We think that it may be having to do with the confidential trade-secret information in their application. It may be broader property interests of why should Company B be benefiting from my product. We would like to hear that.

Also, one of the things that makes this difficult for FDA is that we have certain core beliefs about labeling. When you talk to people in FDA, it really is very disconcerting to think about FDA endorsing a product to be used in a certain way and that information is not included in that product's labeling.

We struggle with that a lot. So it may be a difficulty. Then another reason that makes this so difficult is because it is very hard for FDA to tell Company B, don't even bother trying because, unless you can get Company A to play with you, there is just no way you can get there from here. Then that hurts also, that we don't want to--we don't like to say, just go away. We want to find a way to give Company B an opportunity to try to prove that its product is safe and effective.

So we do have the dual missions of promoting and protecting the public health. In that connection, we agree up front that we prefer companies to work together. It just makes things so much easier for us. But, in the absence of that cooperation, our goal here today and our goal for the Office of Combination Products is to identify a pathway to enable Company B to try to obtain approval of Product B while ensuring adequate regulatory oversight.

One of the things that we might consider today, ask you all to consider today, is what should FDA's default position be in these situations. Should we take the position that cross labeling is required unless we say it isn't or should we say it isn't required unless we say it is.

So, today, I am saying to you what Amanda and Steve said to me 13 years ago; see what you can do about this cross-labeling problem. We want to ask your help in inventing the box. We need to develop the policy and we encourage you to be creative and bold and we want to hear what you have to say.

Thank you.

MR. BARNETT: Thank you, Suzanne. Creative and bold. Those are the key words here.

I want to call up the first panel which is the public-health panel now. I will introduce them once they are up on the stage.

 Public Health Panel

MR. BARNETT: Let me introduce our speakers. I am going to ask you to--because people can't see the name tags. Raise your hand. Dr. Ramzi Dagher is Medical Team Leader in FDA's Division on Oncology Drug Products in the Center for Drug Evaluation and Research. Dr. Miriam Provost is Acting Director of the Division of General and Restorative Devices in FDA's Center--this is a misprint here in the program--Center for Devices and Radiological Health in the Office of Device Evaluation.

Leighton Hansel is Manager in Global Standards for the Medical Products Group at Abbott Laboratories. He is going to be speaking on behalf of AdvaMed. Dr. David Eveleth is Executive Director of Medical and Developmental Sciences at Pfizer. Dr. Paul Goldfarb is Consulting Medical Director at Genetronics. He is going to be speaking on behalf of the Combination Products Coalition.

Now our FDA folks are filing in. The FDA people are--and I have told people to raise their hand when I introduce you. Dr. Donna Bea-Tillman is Director of the Office of Device Evaluation at the FDA Center for Devices and Radiological Health. Dr. John Jenkins is Director of the Office of New Drugs in FDA's Center for Drug Evaluation and Research. Dr. Celia Witten is Director of the Office of Cellular Tissues and Gene Therapies at FDA's Center for Biologics Evaluation.

So let's begin. We will simply take them in the order that they are on the program and ask Dr. Dagher to start out.

 DR. DAGHER: Good morning. I am Ramzi Dagher from the Division of Oncology Drug Products. I am asked to give a perspective from a drug division. Being from the Oncology Division, I will provide that perspective although I hope that some of the principles I describe will apply obviously not just to oncology but across other disciplines.

Whether it is in oncology or other disciplines, often the goal of combination therapy--and, again, whether it is a drug-device or even a drug-drug combination, there is a design to optimize efficacy and limit toxicity.

Just out of curiosity, I polled around in our division. I asked folks, what are the different approaches that we are seeing where these issues can come up. These are the four general areas where folks responded. Some are fairly obvious, the light activation where you have a drug and there is a device that is needed to activate the drug in order for it to have its activity. There are some maybe where it is not so obvious, the novel delivery formulations. What we are referring to here is that there are some delivery formulations that have been designated or could be designated as devices and the combination of the active drug and the novel delivery formulation. The Clinical Pharmacology group reminded me of that one.

Electroporation is one that you will hear more about this morning, but I just wanted to give you an idea that, even within one discipline, we have got a variety of approaches where these issues can come up.

Now, I have to admit, based on this morning's speakers and the advice they are giving--I have to say, I took some of their advice and, in some ways, I didn't take their advice entirely. One thing that I think has been reiterated is that we are asked today to assume that the issue of regulatory jurisdiction--i.e., is something going to be reviewed in one center or the other, what is the interaction between the centers, et cetera--we are asked to assume that that is addressed and that is not an issue. That is why I have here the first bullet, that there are some considerations that are taken into account regardless of the regulatory jurisdiction. So let's assume regulatory jurisdiction is not an issue.

Having said that, however, to me, once you have answered the question of whether or not you need cross labeling or mutually conforming labeling, the next challenge is going to be--let's say you have decided you need it, for whatever reason. To me, the next challenge is how do you go about it? What judgments do you make pursuing that labeling?

I think, before we answer those things, I do think we need to at least go through some of the basic issues that come up from a multidisciplinary standpoint from the drug viewpoint in trying to answer those questions.

So, actually, what I did is I pulled, again, some individuals from the different disciplines we work with. I asked them, well, when you are dealing with these approaches, what are the basic challenges that you have had to deal with and you think we are going to have to deal with, from chemists, from pharmacologists, from toxicologists.

I apologize for this part where this is maybe a little bit veering off the focus of the problem of A and B, but I think there are some basic principles to go through very briefly.

So, starting with the chemistry and manufacturing issues, obviously, chemists think in terms of a drug substance and the drug product. If we are talking about a drug substance, many times, with the combination approach, the drug substance is the subject of an approved NDA or NDA supplement in which case all that is required is the right of reference to that NDA.

However, if it is a new molecular entity and no NDA exists, you have an unapproved source, a reference to the drug master file or an active IND is the way to go. If there is no active IND, then complete EMC information will be needed over that and the next (inaudible). For the drug product, again, if you have an approved NDA or IND for that drug product, then simply referring to that is enough.

If there is an active IND for the proposed formulation, if there is a new formulation, then, obviously reference to that IND is needed. Finally, if there is no active IND, then the CMC information should be provided.

What are those critical chemistry issues? For the drug-substance side, they would relate to the manufacturing, controls, specifications and stability. If we are talking about the drug product, then you would have to address those same issues and edited onto that would be the description of the components or composition of the drug product and the sterility issues where we are talking about parenteral products.

From a toxicology standpoint, again, I asked, what are sort of the big picture issues that you have to deal with. One obvious one is the isolation of the drug versus the device effect. Many times, that can be a challenge from the preclinical standpoint because we don't have models that easily address.

Then, one recurring theme, whether it is for the toxicologist or for the clinical pharmacologist, is the requirement for establishing the systemic versus local effect and the concerns of systemic versus local toxicity. Many times, again, with approved drugs and drugs where there is a substantial previous use in humans, we already know a lot about the systemic toxicity. So, many times, the requirements for any further nonclinical testing would be waived if they are already addressed.

From a clinical pharmacology perspective, the first route is one that often comes to mind when we think about these approaches. Obviously, from a drug-division review, we are always concerned about what is going to be the effect of the device on the drug. How is the device going to affect the pharmacology, the metabolism and the elimination?

Obviously, that seems pretty obvious. But another side of this, and I have to admit it is from my colleagues from industry during our discussions about preparing for this that reminded me of this. I also went back to some in our division. The device, itself, has some characteristics that could have an influence from a clinical pharmacology perspective.

The device, itself; if we are talking about these novel delivery formulations, the device, itself, could be metabolized or eliminated and the characteristics of that could be important. Even when the device, itself, is not metabolized or eliminated, there could be some consequences.

For example, if the device is going to be implanted in the body for some period of time, there could be local effects from that. Or could it affect the general care of the patient. In oncology, we use, for example, imaging quite frequently as part of the care and decision making for the patient. So that is another issue that might come up, whether or not the presence of the device influences (inaudible) and, finally, as we described before, even if you have a drug-device combination where the intent is to provide the drug locally, there would still need to be some information on systemic exposure or lack thereof.

From a clinical standpoint, again, Dr. Lumpkin warned us against going to the details of our standards. But, again, I wanted to simply discuss some of them briefly because, again, they do provide some challenges in terms of how you decide on whether you need conforming labeling. And then, if you have, how would you address that?

So, obviously, whether it is for an oncology or any other drug intervention or device, for that matter, defining the patient population is important. Now, that seems obvious, but I want to emphasize one way in which oncology is different, perhaps, from some of the other disciplines is that we actually have at least 100 different diseases. So that is one thing, whether it is folks from devices or other areas find when they come to talk to us is that we often are used to thinking in very specific ways about the indication and the patient population.

Trial design, obviously, can take a number of forms. But one key area where we see a challenge we have had to deal with in the past and also where we see this becoming more and more of a challenge is in deciding on appropriate comparator, how that is described and how that is evaluated.

There may be multiple interventions, for example, that are available that clinicians and patients accept and medical team has accepted, but there may be no clear standard. There is no one clear intervention that provides the best survival outcome, let's say, in a specific cancer setting. But if there are multiple interventions that are available and we can't identify one that is a clear standard, you know, there may have to be a design where that comparator arm has multiple interventions described. You have to keep that in mind early on when you think about what is going to happen in the future when we talk about the labeling.

Obviously, there has to be some need to isolate the contribution of the device for the treatment effect. There are two sides to that coin. People think of, when they hear treatment effect, efficacy. That's true. But there is also a safety component. You can't always assume that the major adverse events, let's say, that are seen are being contributed mainly by one component or the other.

In oncology, we use multiple different endpoints that we consider of benefit. Some of these can be considered ones that sort of require sort of a systemic effect where you might see them, such as survival, time-to-progression, maybe, and other ones that even local effects could influence such as response rate or alleviation of certain symptoms.

However, it is not impossible that interventions which have local effects won't have any influence on the ultimate outcome such as survival if you are influence a very important component of the disease process.

One thing I want to touch on which I think we have to deal better with from the safety standpoint is that, in oncology, at least, we have had a certain systematic way of thinking about decategorization of safety events. To be honest about, it has been limited in the sense that it has been based on a view that we are really talking mainly about drugs. I think that it doesn't--perhaps, that classic approach doesn't always do the best service when we are talking about combinations.

For example, we have this National Cancer Institute Common Toxicity Criteria grading system where adverse events are graded by severity. Those categories are usually either laboratory parameters that can be influenced like liver function or also there are clinical categories such as rash or allergic reaction or cough or pain.

However, again, because those were designed mainly with the view in mind that we are evaluating drugs, they tend to focus on systemic toxicities. They don't always do a good job of potentially evaluating the issues that come up with device interventions as well. They don't really have a very good way necessarily of teasing out local effects so much as they should. That is something we need to keep in mind.

This is just, again, an example from oncology. I don't know, in other disease settings, whether the paradigm in terms of evaluation of safety also has to be modified when we are talking about a combination of issues.

This is an example of how we have classically thought of safety in oncology of the drugs is that, you know, patients come to us with advanced disease, often. They often have a lot of other comorbid conditions when they come for treatment irrespective of the cancer that they have. So when you have deaths occurring during the evaluation, it is hard to tell whether those deaths are due to drug intervention or the natural history of the disease or an infection or some other process.

So, classically, the way we try to get at this is to say that if a death occurs within a certain period vis-a-vis the administration of the drug, then may be that is more likely related to the drug.

In our classic paradigm where we have had classic cytotoxic chemotherapy drugs, we have used a 30-day time point. The reason for that is when you give cytotoxic drugs like doxirubicin or cisplatinum, et cetera, usually you get most of the effect from the side effects in the first week or two after administration and, by the third and fourth week, you have recovered. So you have made this empiric judgment, not just us at FDA but across the scientific community, that, somehow, if the death occurs within those 30 days, maybe that is more relevant to the effect of the drug than a death that occurs afterward. A death maybe that occurs three months out is probably not likely due to the drug.

Now, that has probably served us well so far. But, again, as we evaluate combination approaches that involve the device component, I think that has limitations and we need to be more creative about how we think of that. That is why the length of follow up is just as important as how you define the safety.

So, before we talk about the program of drug-device combinations, I wanted to just step back and say that a lot of disease settings and especially in oncology, we already have a track record in dealing with combination approaches to treatment, whether it is drug combinations where you have a chemotherapy regimen that involves two, three or more chemotherapy drugs that are given simultaneously or in sequence, or where we are talking about multiple modalities.

Even in the classic paradigm, there are disease settings, whether it is in pedestrian, Wilms' kidney tumor in children or adult breast cancer where, for many years now, the standard of care for some of those patients is, actually, a combination of surgery, therapy and radiation. So we have had to deal with this for quite some time.

The other part of this that, I think, from oncology we have some insight on is that we have to use aggressive supportive-care methods because our drug is going to be so toxic and, in order to assure the best outcome, we need supportive measures such as other drugs, growth factors, transfusions, et cetera.

We have had to deal with how we describe these in the Clinical Study Sections of our labeling for quite some time. Now, in the drug labels, if we are talking about supportive-care measures, obviously we usually don't refer to specific products. But we do, because of the safety concerns that can arise, we do want to make sure that there is some description of the supportive care that is needed for that.

If there are specific safety considerations which require us to describe a specific approach, we will try to do that as well. Obviously, drug combinations, where two drugs affect the treatment outcome, those have to be approached differently.

So I have provided one example which, hopefully, will just help to sort of kick off the discussion for this morning. As I mentioned, we have a lot of different combinations that we use in oncology. There are settings where simply using a single agent would not be considered appropriate standard of care and the standard of care would involve combination therapy.

One example we have is cisplatin. This is a drug that has been approved for many years. It is a DNA-binding. It came to fame in use with germ-cell tumors that now have a very high cure rate with the use of platinum and other drugs. What also, with combination therapy established a role in a variety of other settings, one example being the treatment of non-small-cell lung cancer in adults.

So here we have a situation, actually, where we have four different drugs whose labels describe an indication from the study description of a combination use for cisplatin with gemcitabine, with paclitaxel, commonly known as Taxol, vinorelbine and docetaxel commonly known as Taxotere.

With all of these individual labels, there is a combination use with cisplatinum described for the treatment of non-small-cell lung cancer. Yet, the cisplatinum label, itself, although it includes several oncology indications, does not specifically refer to these combinations.

So how have we dealt with that? Basically, the approach was that we made sure that the label with those individual drugs that I listed describing the combination use we had enough information regarding the safety considerations for cisplatinum and not just describing the safety considerations of the proper use of the other drug. So we have information about monitoring patients.

Cisplatinum is a drug that could have significant renal toxicity and ototoxicity. That is not just described in the cisplatinum label. That is, obviously, also described in combination labels. Also, the dose modifications that may be needed for cisplatinum and for the other drug in question, those are all described in those individual labels.

Another perspective that helps us address this challenge to some degree is that, whether we are talking about cisplatinum or some of these other drugs that I mentioned, or others, that, with many of the cytotoxic drugs, we generally try to make sure that there is labeling in the Warning Section and other parts of the label that emphasizes the need for the monitor for safety and supportive care and actually has wording about the need for having practitioners experienced in the use of therapy to give these drugs is true of the cisplatinum label, itself, of these other labels and many of the other drugs that are the cytotoxic drugs.

Now, where this could become somewhat of a challenge is where we now have potentially combinations where, obviously, we have one cytotoxic and another drug that really doesn't have those same (inaudible) or a device, for that matter.

So, in summary, again, as this morning's folks reiterated, there are multidisciplinary considerations that we have to think about when a component of the combination approach includes a drug. This is regardless of the "regulatory" jurisdiction that that applies

Where those challenges lie depends on the nature of the drug-device approach and the effect on the drug characteristics.

MR. BARNETT: Thank you, Dr. Dagher.

Dr. Miriam Provost, you are next.

 DR. PROVOST: Good morning. Well, I am very happy to be here today to give you the CDRH perspective on the public-health issues related to mutually conforming labeling. I think some of the topics that I am going to talk about are going to be somewhat repetitive. I think, by the end of the day, we are all going to be a little bit sick of hearing about all these issues.

But I would like to come at the topic, some of the topics mentioned by Suzanne, with a little bit of the real-world perspective. I say that because I believe that CDRH is often the point of entry for people who are developing innovative new device technologies.

In my experience, many times they come to us. They submit a 510(k). They think that, in 90 days, their product is going to be on the market and then we have the unhappy task for calling them and telling them that they have a "drug" issue. That, often, is not a pleasant conversation. It often is a shock to them and a surprise and can make for some very unhappy investors, or so I understand.

So what I would like to talk about a little bit today is about a little bit of the perspective of how things are right now in terms of what is clear to us, what are the clear regulatory tasks, where some of the questions arise and then a little bit of a perspective from the public health, looking at the public-health side of it.

So I would like to start with just going over the FDA mission. I think it is interesting because there are three bullet points up there. I think all three bullets actually arise when you consider this whole issue of mutually conforming labeling.

So, stated most simply, FDA's mission is to promote and protect the public health by helping safe and effective products reach the market in a timely way, monitor products for continued safety after they are in use, and to help the public get accurate science-based information needed to improve health.

So, as I said, really, all three of those bullets have some impact on this issue. But what I would like to focus on, as Suzanne did, is the first bullet which is promotion and protection of the public health. That really is the balancing act that we deal with every day at the FDA.

When we think about this issue, if I am on the side of the scale that says FDA should be promoting the public health, then, clearly, we want to expedite approval of innovative products that have potential for positive impact. I think we are going to hear a lot about that today, what can FDA do to help expedite getting some of these innovative products on the market.

But I would like to suggest to you that there is another side of that scale and that is the protection of public health. If we are looking at it from that side of the scale, then we want to ensure that all medical products are adequately tested for safety and that labeling provides adequate directions for use.

So I am going to talk a little bit, I think, just to give equal time to both sides of the scale today. I want to talk a little bit, at the end of my talk, about the protection-of-public-health side of the scale.

First, I would like to start with explaining some of the flexibility that we already have at CDRH. I am going to start by explaining the whole product area of general-purpose drug-delivery systems. I think, as many of you know, CDRH has a long history of clearing unfilled general-purpose drug-delivery systems under 510(k).

I don't think you can read those small words on the screen, but you have them in your handout. These are quotes from the Code of Federal Regulations that describe device classifications that are general-purpose drug-delivery systems; external infusion pumps, I.V. administration set. What I would like to point out is that, if you read these, it is very descriptive of the mechanics of what this product does.

So, for example, an infusion pump is a device used in a healthcare facility to pump fluids into a patient in a controlled manner. No mention of any drug. No mention of any disease state. Then it goes on to actually describe the device.

You can see that in the next two device classifications as well. The nebulizer is a device intended to spray liquids in aerosol form into gases that are delivered directly to the patient for breathing. So, again, there is no mention of the drug or a combination of the drug and the device. It is just describing the device.

So, in general, for these products, mutually conforming labeling is not an issue. They are not prefilled with drug. There are no indications or claims for delivery of a specific drug or biologic. And, and this is very important, there are one or maybe many drugs that have already been approved by the FDA for delivery by this route.

So, when we get a 510(k) like this, it is fairly straightforward. The submitted of the 510(k) needs to show adequate safety and performance as a tool for delivery of drugs or biologics. So basically they just show they are substantially equivalent to a legally marketed predicate device as is done with any 519(k).

The submitter of the 510(k) does not need to show that the combination of this device and any drug, in specific, is safe and effective. So that is one kind of clear area.

The other end of the spectrum, we have more specific drug-delivery systems. For example, products that have a CDER lead are devices that are prefilled with drugs. You can tell them from CDRH because I think of them as devices prefilled with drugs instead of a drug in a delivery system. But these are things like pre-filled syringes, metered-dose inhalers. Those are straightforward.

Photodynamic therapy devices--this is an example of a product where the components along are not effective, that only together are they effective. So it would be difficult to think of how you would actually clear a device that is not effective on its own unless it is used with a drug. The regulatory path for these is also clear. Generally, they have either been done with A CDER lead or, perhaps, a two-application scenario in which there is an NDA and a PMA.

Then the last example is implanted infusion pumps. These are done with a CDRH lead with a consult to CDER or, perhaps, again, a two-application scenario where an NDA might be needed.

So those are specific drug-delivery systems. Now, let's talk for just a few minutes about some of those more discussion situations that I alluded to earlier. When I go through these examples, I want to point out that I don't have the answers or I am not going to say that, for every case like this, things go--cross labeling is required or cross labeling is not required. I just want to point out that these are examples that are not straightforward in which we have had internal discussions, or would have internal discussions, at FDA and with sponsors.

One example is cases in which the proposed indication for use for the device would necessitate a change in the route of administration for approved drugs. This would be an example, for example, of a device indicated for infusion of therapeutic drugs into the neurovasculature. There are no drugs approved, therapeutic drugs, approved. So, in this case, it would be difficult to clear or approve a device for that indication.

Another example that I think we are going to hear a little bit about are devices indicated for targeted delivery of chemotherapeutic agents. That is not on the label. If it is labeled for I.V. administration and this is a targeted administration, and a device application comes to us, we are going to have questions about that.

Another example, a device is intended for enhancement of drug effects in which the combined use may have a different safety or efficacy profile than the already approved use. One example might be devices used to prepare skin prior to administration of lidocaine. It wouldn't be clear, just upon inspection, that stripping off the stratum corneum would or would not affect, perhaps, the safety and effectiveness of the lidocaine.

A final example would be using devices as labeled may necessitate a change in dosing of approved drugs. A couple of examples here; the first one is breath-actuated nebulizers. With constant-flow nebulizers, a lot of the drug is just aerosoled and goes out into the room. Well, an innovative technology comes along that only activates the nebulizer when the patient takes a breath.

If you take that same dose of drug and put it in the breath-actuated nebulizer, the patient, in the end, could end up with a lot more--inhaling a lot more drug. The old dose was designed knowing that some of it was going to get lost, so questions can arise.

Then, finally, the product area of patient-controlled infusion pumps for pain medication. By allowing patients to dose themselves with a bolus of drug, are you, in fact, changing the dose such that the dose that the patient is receiving is one that hasn't been shown to be safe.

Okay. Now, those are some of the examples of the grey area. What I would like to suggest, something for everyone to think about, that, if we are the side of the scale that the FDA deals with--if we are on the side of the patient-protection scale, I would like to suggest that mutually conforming labeling can help FDA ensure patient safety.

This policy can do this by making sure that adequate information is provided to users on both drug and device labels, postmarketing safety information is collected on the combined use, and if important findings are made, both labels can be updated accordingly.

Finally, although this is really not supposed to be discussed today, the concept that, if we have mutually conforming labeling, the new uses are thoroughly evaluated by FDA experts at both centers according to their respective authorities. I certainly agree that we can work together and we do work together. We don't need to have mutually conforming labeling to ensure that we work together by any means.

But, if you are on the patient-protection side of the scale, mutually conforming labeling is certainly a sure-fire way to make sure that such cooperation and collaboration goes on.

Another issue that has been alluded to is manufacturing concerns. I think Suzanne mentioned this and it is the idea that, if there isn't this relationship, there can be situations that would be of concern. I think an example, kind of a classic example, is an I.V. drug that has a change in endotoxin level and an innovative device that using that for intrathecal delivery.

Or another example is a device that changes the laser output for a photodynamic therapy product and what effect will that have. So, if I am concerned, again, about public-health protection, then these are going to be concerns that I have and mutually conforming labeling and that cooperation between the two companies addresses those concerns.

So, finally, in summary, I just want to say, first, that I think the FDA regulations do allow flexibility to clear devices as general-purpose drug-delivery systems when appropriate. Questions arise when the device use requires an unapproved drug use. We have heard some examples of that. I think we will hear more about that today.

My suggestion to think about is that mutually conforming labeling can help FDA ensure safe and effective use of the combination products and also help us achieve our mission of promotion and protection of the public health.

Thank you.

MR. BARNETT: Next we will hear from Leighton Hansel speaking for AdvaMed.

MR. HANSEL: I would like to echo Dr. Lumpkin's comments about the Office of Combination Products and the staff. I predate Suzanne a little bit in Combination Products when I was at CDRH. Those were the days before it was mentioned in the statute--product like took three years for the agency to reach a conclusion on the status of medicated wound dressings. So people who have come into this area now really have a much better chance of getting a more timely and more coordinated view at FDA. I would also like to congratulate you--

MR. BARNETT: While we are waiting, Leighton, I just want to mention that our guideline for the speakers is 15 minutes and everybody has been great so far.

 MR. HANSEL: AdvaMed, the Advanced Medical Technology Association is the world's largest association representing manufacturers of the devices and diagnostic products, medical information systems. More than 70 percent of our members have less than 30 million domestic sales annually.

We have paraphrased the questions on the slides and there is some repetitive statements since they aren't solutions to several of the questions are similar. Before we address FDA's questions, though, we think it is important to state the overarching principles AdvaMed supports with respect to the regulation of combination products.

Those are we believe that to encourage innovation and promote public health advancement, FDA must be flexible in applying its regulation to combination products. Inherent in this approach is the requirement for fairness, specifically with regard to the protection of proprietary information and intellectual property rights of a non-cooperating party.

Further, all potential issues of safety and effectiveness related to the development of a combination product must be addressed. Risk-management principles must be considered when making decisions on when and how to apply mutually conforming labeling. I would like to thank Suzanne for saying that cross labeling and mutually conforming labeling in her opinion are the same. We have had a lot of internal discussions about whether they were or they were different. We try to avoid using terms other than responding to the questions.

In those instance where a determination is made that a combination product exists and mutually conforming labeling is required, the impact on development could be significant depending on whether or not the drug or biologic company is willing to work with the device company.

It may result in the device company not proceedings with development of the device. This may result when clinical studies are required to prove that the device and drug labeling are mutually conforming, new drug application would be needed to revise or update the labeling to provide for mutually conforming labeling, or the public safety would be compromised is the companies do not cooperate.

Of course, there are examples of this. However, it is a very difficult document because the requirement for mutually conforming labeling may be one among many factors that would lead a company not to proceed with the development of new product.

Further, it may not be possible to assess the impact on product development. The device industry is make up of a large number of small but innovative companies that may not have the resources to proceed with development of a product if they are unable to establish a relationship with the sponsor of an approved or cleared product.

Additionally, the life cycle of a device is shorter than that of a drug as well as the market size. It is almost impossible to assess the loss of venture capital available for the development of novel products.

The labeling of a separate marketed drug and device does not need to be identical but should be not contradictory. There can be broad agreement of the intended use, dosage and route of administration to provide for the adequate directions for use of the drug and the device. If the result of a systematic risk analysis indicates that there are no issues with regard to safety and effectiveness, then minor differences such as in dose, dosing schedule should not be an issue.

We are in support of this position. We look to the principles of the Inter-Center Agreement between the Center for Devices and the Center for Drugs characterized as the ICH. Section 7(a)(1)(a) of the ICA states that there are three essential parameters required for mutual performance. The flexibility extends to the following; indications, general mode of delivery and drug dosage schedule equivalence.

If device labeling is consistent with these key parameters of drug labeling, the essential elements of mutual conformance are generally assumed to have been met. From a regulatory perspective, such products are classified as devices, in our opinion, not as combination products.

The ICA provides flexibility for two of the three key parameters, specifically the mode of delivery simply must be the same, general mode of delivery and the dosage schedule must be equivalent. The ICA also provides for all three parameters to be examined for significance of their change. The regulations likewise suggest flexibility by identifying a significant change in dose as triggering the need for labeling changes.

Examples of the application of this flexibility include FDA's clearance of lasermeric infusion pulse for continuing infusion with local anesthetics in either a hospital or home environment where drug labeling did not specifically address home use and scleronychia of continuous delivery systems for delivery of insulin where insulin was labeled for bolus administration by syringe.

Even if there are changes in the three essential drug parameters provided in the ICA guidance, nevertheless it affords CDRH further flexibility to consult with CDER and resolve these issues through device labeling. More specifically, the ICA support device labeling as a mechanism to resolve labeling differences because it expressly refers to CDER consult process which does not count out a separate drug in drug review.

The need for mutually conforming labeling should be determined on a case-by-case basis by the Office of Combination Products during the request for designation in the RFP process. To determine whether mutually conforming labeling is required, FDA should look to the flexibility established by the ICA. Further, decisions should be based on sound science and a thorough risk assessment.

Consistent with the concepts outlined by the ICA, the labeling of two products does not need to be identical but should be consistent where it matters. The ICA is flexible requiring that labeling must be mutually conforming with respect to, as noted, prior indications, general mode of delivery, drug dosage and schedule of the drug.

In addition, Section 7(A)(1)(a)(ii) of the ICA on mutual conformance does not purport to address any other aspects of drug labeling beyond the three stated parameters. For this reason, flexibility has historically been afforded with the secondary aspects of drug labeling; for example, clinical pharmacology and precautions.

Consistent with the ICA principles, wherever possible, device labeling alone should include adequate directions for use of a drug and a device. All relevant information would be included in the device labeling. Consistent with the flexibility afforded by 21 CFR 3.2, FDA is permitted device labeling alone to describe use of the device with the drug where the drug is not individually specified.

Examples of this approach can be with drugs and devices that have broad therapeutic classes such as off-patent generic drugs, U.S. Pharmacopoeia, U.S.P. Monograph drugs, grandfathered drugs, over-the-counter drugs, and drug-efficacy-implementation, DESI, drugs where there is public information to provide the device company with adequate information to develop appropriate labeling.

Finally, device labeling traditionally will contain more detail with respect to how to use the design with a drug or a biologic. The PCA and insulin pumps are the best examples of generic drugs. Changes in dosages and locations for use with no changes in drug labels with the instructions being provided by the device labeling.

The loss of instructions for use does not have any greater implications for combination products than for any other FDA-regulated product. Further access to current labeling is made easier for both drugs and devices as a result of the current electronic labeling initiative. There is not need for an FDA policy specifically directed as separately marketed combination products to address the possibility that labeling for a--component might be lost over time.

For a device being used with a branded proprietary drug, there should be collaboration between the parties to gain access to the proprietary information. For a variety of intellectual property and commercial reasons, the device manufacturer generally will not proceed to develop a device for use with a branded proprietary drug absent a contractual agreement before and between the parties.

Thus, it is in the interest of the drug manufacturers to identify changes in drugs. Notification processes covering these issues are incorporated in the standard commercial arrangements. The device sponsor would be aware of the changes to the drug as a result of the contractual agreement between the drug sponsor and the device manufacturer.

Consequently, there is no need for FDA to impose new regulatory notification requirements on these commercial drug-device manufacturer relationships. Moreover, irrespective of contractual agreements presented to the Quality and Systems Regulations, device manufacturers already have the obligation to routinely monitor changes, including changes to the drug, biological products or components of or used with their marketed combination products including changes in impacting the device's safety.

As part of their product life-cycle management, device manufacturers could conduct risk analysis and to make appropriate changes meant for any necessary regulatory approvals or notifications. Additionally, to the extent that there are concerns regarding postmarket changes to drugs that might affect the device or combination product regulated by CDRH, FDA has the adequate authority to require a sponsor monitoring and reviewing drug changes through conditions of approval and other postmarket mechanisms.

As such, AdvaMed believes that the agency need not create new regulatory requirements related to notification, postmarket changes to biologics, drugs used with devices or combination products regulated by CDRH or devices with biologics or drugs regulated by CBER or CDER.

For devices intended to be used with generic off-patent drugs, the device sponsor already has an obligation of the quality systems to independently monitor changes associated with the products. In addition, FDA can require the device manufacturer, the B Company in the hypothetical, to monitor and review changes to the drug via other postmarket mechanisms such as conditions of approval or special controls. As such, direct collaboration between the drug and the device manufacturer may not be necessary.

What process should be applied? We believe that a framework for the process has already been outlined in the ICA. To provide a predictable pathway, FDA should make the determination of whether mutually conforming labeling will be required early in the process at the RFD stage. Timing is particularly important for small, resource-limited companies. It is critical that the sponsor understand the regulatory path and the develop required early in the process for effective development of the products.

It is undesirable and potentially financially disastrous for a small company if additional significant data and regulatory requirements are opposed late in the product development after much of the work has been completed.

New combination products may raise new issues. OCP should be ready to address these new issues as they become apparent so the decisions FDA makes should remain consistent across reviewing divisions, encourage innovation and promote the public health.

We believe that OCP is best positioned to facilitate resolution of new controversies. AdvaMed would like to thank you for the opportunity for stating our position. We look forward to the Question and Answer Session.

Thank you.

MR. BARNETT: Thank you, Leighton.

Dr. David Eveleth from Pfizer.

 DR. EVELETH: I would like to thank the organizers for putting this together and giving me the opportunity to speak.

Before I get into this, I want to say that--what I am going to say doesn't necessarily represent an official position of Pfizer, Incorporated. It does represent a lot of bias and I want to make sure that we identify the bias that we are bringing to this up front for everyone here because I think it is bit different from some of the other speakers.

The first part of this is that we look at this obviously, as in the hypothetical, we are Company A. We have a drug and somebody has come to us with a device that will enable the use of the drug in some fundamentally novel way, a new indication, a new patient population. We also look at this in the sense of not a minor change; that is, we are not really approaching this from the perspective of, for example, the insulin pump in diabetes but rather somebody has come to us and developed a device that will enable us to use a small-volume parenteral in an aerosol where the small-volume parenteral, for instance, is used in multiple sclerosis and the aerosol, or the nebulizer, is going to be used to cure the common cold.

If anyone doubts that these sorts of things come up, I mean, you are underestimating the creativity of the device companies. We get these all the time.

Another important part of what we bring as a bias is that these are not mature products. These are not generic products. These are products that are early in their life cycle where we look at it from the perspective of a product that we have launched maybe a year or two ago. We are diligently satisfying our postmarketing commitments to the agency. We anticipate changing the label with postmarketing safety surveillance information.

We either have it as an intent or already have programs in place to develop new indications for the product, and we are constantly changing our processes for manufacturers, trying to drive down our cost of goods.

Since it is early in the life cycle, many of the aspects of how the product is made are proprietary. They are not proprietary just from the perspective, necessarily, of it being something that we would rather--that is confidential to the agency, but are still the subject of not-yet-issued patents.

So, we see this as something for a product like that that is very difficult. There are issues that don't lend themselves to a regulatory solution.

I want to jump to the second bullet which is, I think, very important and that is if these companies are not cooperating, there is a reason. If there is an unmet medical need, if there is a public-health reason why this device, the use of this drug and this device, is good, then that is an opportunity for us to serve our stakeholders including our shareholders.

So there may be a fundamental disagreement among the companies about whether there is, in fact, an unmet medical need, but if there is a strong public-health reason for this to come forward, it doesn't make sense why the companies are not cooperating.

Over the long term, the fact that these are going to be evolutionary--Leighton mentioned that, I think, device labels, or device life cycle, is typically short and drug product that is early in its life cycle is typically going to have a lot of labeling changes as they go on.

A lot of this has to do not so much with whether or not you can get a snapshot in time and do this right but whether you can maintain this and develop a process wherein this works overtime for the life of the product.

I do want to stress that we don't see FDA's role as a broker business deal and urge a lot of caution.

That caution has to do with the fact that we see a lot of risk in some of these. The first step really is a risk assessment. It is not so much whether mutually conforming labeling is required or not. That is one solution to some of these problems. But the first thing that you have to look at is what is the risk in allowing this product to go forward.

The first is what is the consequence of a device not working as it should and how does the combination--how much is the combination linked in the patient or the physician line. This gets, I think, partially to what happens when you lose the label kind of thing.

If something goes wrong, is the use immediately going to think about the device or the drug? Or are they going to think that it has to do with the two together? Really, in a combination product, it is the two together.

Then there is a potential for variation within specifications to influence efficacy and safety. We will get into that more later. But the first issue really is what happens if something goes wrong.

If that is not a great big deal, then there are a lot of easier-solution kind of things. But some of these indications, certainly some of the ones that we have been approached with, if the device causes the drug--if the variability in the way the device-drug combination works causes a failure of efficacy, that is a severe problem and certainly, in oncology, that is a severe problem.

So some of the clinical-safety issues that we worry about are how do you manage safety reporting, how do you manage attribution, who is going to attribute an adverse event of the device or the drug, remembering that the assumption here is that the original drug approval is for a different patient population, maybe a different physician population, certainly a different indication.

How are you going to attribute the adverse event to the device or to the drug? Is the agency prepared to attribute? How do you manage a single safety database when you have got different under-reporting bias, different kinds of physicians looking at adverse events and different risk-management plans. Certainly, we have a lot of products for which the risk-management planning include a specific data-collection tool, specific types of postmarketing surveillance which are going to be different in these different indications and different for the device versus the drug.

I am not really sure how you would deal with two different risk-management plans if they were the kind of intensive risk-management plans that some products have.

Safety issues related to the chemistry, manufacturing and controls section are probably one of the biggest worries that we have. Specifications, as was mentioned before, of each component are not transparent to the other manufacturer. We, as Pfizer, typically manufacture products well within the safety specifications that are in file and changes to those specifications by either Company A or Company B are not going to be transparent to the other company.

When that has the potential to impact safety, efficacy and the performance of the combination, especially when failure of efficacy represents a serious safety risk, that is, to us, a big problem. We just don't see a practical way to determine that variance within a specification doesn't pose a safety risk, especially given that Company B is not going to know what the specification is.

So then we get to the issues around the evolution of the label. If we add indications into the label--that is, we, as the sponsor, add indications into the label and there is a potential to use the device for that indication or in a population that crosses into that indication, how do we define if it is necessary to conduct clinical studies.

If, for example, we think combination studies are required--that is, we need to test the device in that new population--whose label is out of compliance? Who is the enforcer here? We think that if you are going to have a mutually conforming labeling, you better make these issues relevant, the sections of the label, relevant to all the indications. That is going to be a big problem, is the labeling.

So I think there are significant safety issues here that arise in the absence of cooperation. Conforming labeling has an impact but I think it has a limited impact. There are issues that don't lend themselves to regulatory solution and, again, if companies are not cooperating, ask why.

It may be that Company A has every intention of developing the drug in Indication B without the device. That is just one possibility. We just don't see this as something--at least, early in a product's life cycle, that can be worked through. Again, the agency's role is not to force cooperation between the companies as in a broker business deal.

Thank you.

MR. BARNETT: Thank you, Dr. Eveleth.

Dr. Paul Goldfarb speaking for the Combination Products Coalition.

 DR. GOLDFARB: Good morning. I am a medical consultant for a company called Inovio Now which Genetronics that works with a technology called electroporation. Although I have worked with Brad in developing this with the Combination Products Coalition, I would have to say that my view of collaboration on this is somewhat similar to my son's which is, "After me, you are first." So, basically, this is a presentation of where we have gone.

So the first question I have had, how many of you have heard of electroporation? This is the reality of the technology that the people who know about it are biochemists who use it in labs and regulatory people who have heard me do presentations. The challenge we have is it is still a very novel technology.

It is a technology that uses a reusable box that generates electric energy with a series of disposable hand pieces that is single-use for effecting ablation of solid tumors. The applicators are shaped in different sizes. The therapy, though, is a local therapy directed at tumors and causes a local ablation of the cancer.

The effect of the electroporation is you create an electric field around cells and that creates pores in the cells that facilitates the delivery that facilitates the delivery of drugs. So, in this case, we are using the drug bleomycin. I am banging through this part of it as quick as I can so we can address the regulatory issues.

The way it is used clinically is you would inject the drug directly into the tumor. After you have injected the drug, you then electroporate it with a series of pulses and you allow for the uptake the bleomycin specifically into the malignant tissue. The thing that makes the technology unique is you get ablation of malignant tissue without injury to the surrounding benign tissue even if the drug enters the tissue.

As a technology, we have used it in a whole variety of different histologies. That is significant. So this technology is really not histology-specific so it is not acting as a drug in the sense that this is a drug that is only active in certain tumors. It is really acting much more as an ablation technology that uses a drug.

If you can get the drug and the needles to the tumor, then you can actually necrose the tumor. I apologize for a few clinical pictures, but we will get through it quickly. This is a patient with a head-and-neck cancer. This is the typical cancer that you would approach. Surgically, we would remove half the tongue. What we are looking for is an alternative.

You inject the drug. You put the needles in and within ten minutes you can treat the whole tumor. The patient can go home the same day. Over a period of weeks, you get a necrosis of what is the tumor with no injury to the surrounding benign tissue.

We did a series of studies with the FDA. Initially, the first study we did is, on the yellow, what it shows is that, if you inject bleomycin into the tumor and you don't use electroporation, you have no impact. You don't necrose the tumor. You don't get any local reaction. Nothing happens. When you add the electroporation to injecting the drug locally, you get over a 50 percent objective response rate.

So, at the end of that, when we finished Phase 2, we felt that we had a technology which offered equivalent disease control to surgery. Potentially, it offered better tissue preservation which meant that you would have better function preservation and, theoretically, it should be less expensive than doing major surgical resections.

So the mirror of what was said at the beginning of the meeting, we are truly the square peg that has been trying to fit into the round hole provided by the agency for approval. So where we started was I would say I have been doing this since Mark came into position in the office. My impression of how long I have been here is when we had that first slide that the DIA put up, I thought that was me sitting up there with the brown hair and the clear face thinking that was the first meeting I went to on this topic. I can remember when that happened.

We felt that we had a novel ablation system that uses a well-characterized drug in a previously approved route of administration to ablate malignant tissue and spare non-malignant tissue. But it was used on conjunction with the device that uses brief pulses.

The negotiations with the FDA, we were seen by them as being a combination product with primary review by the drug group with consultative review and a recommendation that we use a 510(k) to get the device approved and do a standard approval which would be a standard survival-based study for people with advanced head and neck cancer which we were looking for an enhances survival because that is basically how you get a new oncology drug approved.

The issues that we had with that approach that made us the square peg was we had a local therapy that has not benefit in total tumor burden and is not addressed by what we do. A study looking at the survival endpoint really doesn't address the clinical benefit of the technology. This is not clinical benefit--i.e., survival--to local therapy when used in people with end-stage disease. The therapy would not be offered to patients in this clinical setting.

So we were being asked to do a study that was basically addressing a group of people for whom we would not be using it routinely and used in a model where you were looking to look for a local therapy to have overall survival advantage which seems inappropriate.

In negotiation with Mark's assistance, we moved forward. As you can see from these wonderful graphics, we are now sort of changing the square peg slightly and changing the round hole slightly. We have a study that is comparing the efficacy of electroporation to surgery. We are doing this on people with localized recurrent disease or second primary disease.

Now we are really asking the appropriate question. Does this local therapy work as well as another local therapy in a group of people for whom a local therapy would routinely be offered? Of course we are doing the study in a model in which a clinical benefit has to be defined. We are doing a study in which we are using functional outcome as the clinical benefit that we expect to demonstrate in the study.

We will also demonstrate that local control is the same and that survival is the same.

The issues that came up in discussing this trial as we have moved forward surround the drug bleomycin. So, specifically, bleomycin is a well-characterized generic drug. It is a Class 1 generic drug for parenteral administration which means that this drug is fixed in time and that anything called bleomycin has to meet USP standards, that the device acts with a class of drugs in the sense that there are several manufacturers of bleomycin and not a single manufacturer, that we are not the M.A. holder for bleomycin and we will not bring the drug into commerce at any time in the future and we have already explored with current drug manufacturers their interest in participating in this project, and there is no interest among the multiple manufacturers for them to participate in going forward with the combination approval.

The resolution to this that usually we have resolved with the FPA which, in a sense, is the real-life answer to the issues that we are dealing with today in a specific set of events, but, at least, I come here to say to you these things can be resolved in human lifetimes.

What we have agreed is that the label that is included with each of the disposable applicators will incorporate all of the relevant information regarding the use of the drug bleomycin when used as part of the procedure. So that means that, for each of the disposable applicators, there will be a set of instructions that reflect the use of bleomycin.

It addresses the issue of what happens when you lose the instruction because every time you open up a new applicator, there will be the instructions. We will continue to monitor the available formulations of bleomycin on a routine basis to make sure of this continued constancy of formulation and that the final approval of this will be through a device approval, a PMA.

It will be granted by CDRH, but there will be a collaborative review with CDER. It will review the clinical trial.

The implication for this, and so this last diagram shows that now we fit. Of course, one interpretation of this is that this is a company that got beat up, ground down and screwed by the agency. But, in fact, we see it as something that we have worked out together and we are comfortable with the result.

We think the implications are that new indications may not require relabeling of the drug and will be reflected in changes in the device label. So, as we look for new indications, and we are using this technology in treating other tumors, we can use this approach to move that forward and get supplements to a PMA to get new approvals for new indications.

Once initial safety issues related to bleomycin have been reviewed with CDER, future efficacy assessments of the ablation technology might be reviewed by CDRH and approved as supplements to a PMA. So we are moving, I guess, what I would say is the regulatory path more to be consistent with other ablation technologies because that is functionally what we do to patients and more away from drug approvals because we are really not using it as--drug.

In summary, I would say that the model provides an innovative, flexible pathway to review and approve and drug-device combination product that use specific generic drugs in a specific way.

Thank you for the time.

MR. BARNETT: That concludes the formal presentation in the first panel.

 Public Comment and

Open Discussion on Public Health Issues

MR. BARNETT: Before we get to the FDA folks and ask them for questions, we had two people signed up in advance to speak. I am wondering if they are here now. One of them was Mason Diamond. Raise your hand if you are here. Yes. And the other was Lee Ann Chambers. Good. You are both here.

If you don't mind, I will ask you to do your thing now. Let me first call on Dr. Diamond. If you don't mind, come up to microphone. Identify yourself and talk to us.

DR. DIAMOND: Thank you. I am Mason Diamond. I am the Vice President of Clinical and Regulatory Affairs at TyRx Pharma. I would like to thank the agency for allowing me this time to address you. I do not intend to discuss what the FDA can and cannot do. I think that we all understand that.

Similarly, I am reluctant to propose that the agency approach Congress to enact legislation given what is going on up there on Capitol Hill. While there is a spirit of cooperation in Rockville, it would nice if that same spirit extended itself a little further south.

But, in any case, I intend to focus my comments on the practical question, is mutually conforming labeling a need to have or nice to have? I think we all agree that there are many reasons why mutually conforming labeling and cooperation between companies would be of benefit as well as many reasons why they might now.

While it is true that the FDA cannot mandate that companies cooperate, the agency, in an effort to achieve the most ideal outcome, are in a position to ask the hard questions and make very pointed suggestions that we may not be able to because of internal company politics.

For example, in my previous life, I was working for a company that marketed a number of OTC anti-inflammatory drugs. During our pre-IND meeting, the FDA certainly asked, why do you need another OTC NSAID for, which was a very good question. I have always appreciated the agency sharing their insights, but these sorts of statements have never constituted a mandate in any way.

The topic of mutually conforming labeling seems the most significant aspect of 21 CFR 3.2(e)(3) relates to the section that states, labeling of an approved product needs to be change in some way, to reflect a change in intended use and so on. Given the change in how the existing product would be used, and in light of the current regulations, it is my understanding that an NDA or ANDA or similar type of filing would be required.

As a result, new labeling for the existing product would have to be drafted in any case, whether mutually conforming or not. So, if the cooperation between the relevant parties is not possible, that company developing the new use for the product would have to file to cover both the combination product and the new use for the existing product.

As well, they would also be responsible for any related labeling and any related product liability. One possible thing that the agency could do to make going the NDA route attractive to small companies possible would be revising the user-fee schedule similar to something that is done in devices where a small company has a benefit. It is less expensive to file a PMA, 90,000 as opposed to two-hundred grand, assuming the first PMA is free. So you really can't do a deal like that.

So, to get to the point of the question as relates to the public health, mutually conforming labeling would be nice to have but I don't think it is absolutely necessary.

Thank you.

MR. BARNETT: Thank you, Dr. Diamond. Does anyone on the panel, either the FDA folks or others, want to respond in any way or make any further comments on that?

Okay; if not, let me ask Dr. Lee Ann Chambers of Eli Lily to come up.

DR. CHAMBERS: Thank you very much--it has been very helpful. It has been, I think, a very useful and timely thing and we need this.

Just to preface my comments a little bit, I come from the trenches of trying to put some information on a drug-product label so my comments really come out of that experience. Lily is concerned about the current situation with general dose-delivery systems where a drug product is removed from its primary container closure and placed in a device for an extended period of time such as an infusion pump.

It is our understanding that review of the device submission for these products focuses only on the delivery accuracy and biocompatibility of the device components. It is not clear that the biocompatibility testing performed according to the ISO-10993 standard adequately addresses the performances of the device materials when exposed to various drug formulations that may be used in the device.

Could the drug formulation elute different substances from the device materials that are seen during the biocompatibility testing? A drug product with proven efficacy through discrete subcutaneous injections may not have the same efficacy when the drug is administered for continuous infusion.

Finally, there is no consideration of the impact of holding the drug product in the container other than its primary container closure for a time period of more than a few hours and at temperatures that may be higher than its regular storage conditions. While we agree that many devices may be considered general-delivery devices, we recommend that when a device is intended for a general class of drugs that the division at the Center for Drug Evaluation and Research who is responsible for that class be consulted.

They should be asked to review the indication, the mode of delivery and the drug dosage assumptions for concurrence that the device can be cleared for the drug class. If CDER agrees, it seems it should also be possible for the drug manufacturer to indicate the general class of devices that are appropriate for use with the drug product without generating additional data to support this information; that is, the device clearance should be sufficient evidence.

Thank you.

MR. BARNETT: Again, I will ask if anyone here wants to respond in any way or has a question or a comment. If that is the case, thank you very much.

Yes?

DR. WITTEN: I am wondering if the last speaker can give some examples of where these issues have come up and been shown to be problems.

DR. CHAMBERS: Our experience was trying to put infusion pumps on our drug-product label. The studies that we were requested by the Center for Drugs, we had to repeat some information that we found would have been addressed, or possibly should have been addressed, by the CDRH approval process.

MR. BARNETT: Anybody else? Let me ask if anybody in the audience has a question for this speaker. Yes; come on up. Identify yourself first.

AUDIENCE PARTICIPANT: (Inaudible) I have a question. Is it possible to ever submit a 510(k) for a device that will be a component in a combination product?

DR. JENKINS: The short answer to that is yes.

AUDIENCE PARTICIPANT: How would you handle that? Would the 510(k) is submitted to the division (inaudible) and the drug component would be submitted later?

DR. JENKINS: It depends on the determination of who has jurisdiction over the product. So I think that the short answer would be it would depend on the primary mode of action of the combination product.

AUDIENCE PARTICIPANT: Suppose the primary mode of action is the drug, would the device component still be submitted as a 510(k)?

DR. JENKINS: Yes; it really depends on whether or not you are going to have a single application or have several applications. There is a guidance document that is being held to address this. Mark, is there anything you would like to say about that? It just depends. There are too many factors and you would have to give a specific example. But it is certainly possible.

DR. PROVOST: If I could add that--the 510(k) couldn't be cleared if the drug was not approved yet. So if you are in a two-application situation, we wouldn't be able to clear the 510(k) if it is specifically for use for a drug and the drug is not approved.

MR. BARNETT: Let me take one more and then I want to go to the--go ahead.

AUDIENCE PARTICIPANT: To answer her question, there are some companies that have filed NDAs with pull-out sections of the 510(k) that related to it. There is no hard and fast rule, but that seems to have worked.

MR. BARNETT: Let me ask the FDA folks up here now. If they heard anything this morning from any of the other speakers that they would like to question, or to get clarification on. Celia?

DR. WITTEN: I have a question for the gentleman from Pfizer and that is that most of your talk was aimed at talking about novel drug-delivery products, devices. But many of what we see or have seen when I was at CDRH, anyway, or drug-delivery devices, actually what manufacturers came forward with were devices aimed at specifically doing what physicians had already been creatively using off-label, jerry-rigged, other kinds of devices, they used drugs off-label.

So, in other words, there might be a family of drugs that are used off-label for a long period of time by a new route and a sponsor wants to manufacture a device specifically aimed at accomplishing that which certainly is an advantage, I think, to have something that is designed to do what you want it to do.

How would you address that in the case where there were no drugs labeled for that use? It is a common route of administration, nonetheless, and the manufacturer wants to make a device specifically for that already commonly used purpose.

DR. EVELETH: We have had those kinds of experiences. The fact that the drug is commonly used off-label certainly indicates that there are a group of physicians that believe that there is an unmet medical need to be satisfied. It doesn't always mean that we and, in fact, in some cases, the agency would agree that that is a safe and effective use of the drug in question.

So I think that needs to be dealt with on a case-by-case basis. Again, if the company is manufacturing the device for the specific use with that drug and that drug only, they are facilitating a use that would sometimes would like to see and sometimes we are not necessarily supportive of that use because there are safety concerns.

There are examples today, in fact, of that. I can think of a specific example where there is a device that I know some people are working on to deliver a drug that, in fact, has the black-box warning, thou shalt not do this. So we, obviously, are not supportive of that kind of an effort.

I'm sorry; am I answering your question?

DR. WITTEN: Yes. So now I have a follow-on question. Now suppose--it is similar, just a different wrinkle on the same question. Device Manufacturer B, to use the terminology in the session, makes a device to deliver Drug A and does a study to show that you can do that. You have already expressed your position on that.

But now supposed Device Manufacturer B also does a study that it can work with Drugs Q, R, S, T, U, V, W, X, Y, Z. But none of the drug manufacturers want to participate in this labeling. Are we still in the same position that we were in with Drug A?

DR. EVELETH: I can think of cases where we probably are not. If there are that many other drugs, and one can think of a situation where there are many drugs in the class, and the drugs are well understood--that is, the EMC characteristics of the drugs--then, in some cases, we would--there are degrees of cooperation. There is a degree of cooperation where we say, that is okay, and we will provide you with access to information.

I think one of the problems with some of these generic drugs is that, even that involves an investment on the part of the generics company--I don't want to speak for my generics company colleagues, that it is a resource that they don't have and they don't seem to benefit in.

So if we see no safety issues ourselves, there are situations where we would not not cooperate, but we wouldn't invest a lot of money in it.

MR. BARNETT: Anyone else from the FDA want to question the panel?

DR. BEA-TILLMAN: I guess I have a question that I would like to direct toward the speaker from AdvaMed. One of the issues that, I think, concerns many of us is the ability of the Company B and the scenario we are talking about to adequately be able to monitor in the postmarket setting for changes that Company A makes in their product.

I know there was sort of a parallel issue that some of you who are familiar with the device world may remember when we dealt with re-use of single-use medical devices and the question of to what extent the third party reprocessors could adequately determine when the original equipment manufacturer had made changes to their product and redo their validation.

So I guess one of the questions I had for the speaker from AdvaMed, because he addressed this, is, realistically, is there enough information in the public domain about changes that the drug company might be making to their product that it is reasonable to expect that the device company would be able to determine when important changes were made.

MR. HANSEL: There are so many variations in the systems and it is hard to give a "one answer fits all." Certainly, in our internal discussions, we--my experience was, at Food and Drug and, as I think we are pointing out in our presentation, that a lot of this has to be on a case-by-case basis. When we were talking internally, we believe there are certain drugs that the very nature of how they are on the market as drugs and the controls they are under to be in a certain category would not likely make a change in the drug that would say, for instance, affect some of the more generic-type drug-delivery systems.

You could be using it in the context of activation of treating a disease condition where that might not be adequate. So I think, here, again, it is up to the device company to look at the intended use of the drugs that are out there and, as we acknowledged on a proprietary or traded drug, we don't believe it is likely that the device company could, in fact, have access to enough data to really make an adequate risk assessment.

MR. BARNETT: I thought that, in Suzanne's opening comment, she posed a very important, very simple but seminal question. I jotted it down here. She was saying, from a public-health perspective, what should FDA's default position be--that is, a mutually conforming label is required unless FDA says that it isn't or that mutually conforming labeling is not required unless FDA says that it is.

I think that is really an important issue. Anybody up here want to talk about that, any of our panelists?

MR. HANSEL: I think we were saying that, let's face it. FDA has access to information that no one company or group of companies would have access to based on analyzing that, they may determine that there is a need for mutually conforming labeling. What we are, I think, saying is that that should be the default that FDA concludes there is a need, not to say that the default is, you know--that, automatically, you need it, that it should be on a basis where FDA has concerns enough that they feel it is absolutely necessary.

Here, again, I think it depends on the product. I think this is something Food and Drug is going to have to address. The drug labeling, even in a generic drug, may be based on a traditional mode of delivery with the products that are in the marketplace.

I am a device company. I come in with a way to do it more safely or under more control. Then the public-health issues may be less even though the drug remains the same because there is new technology available to address the labeling limitations or precautions that were on the drug when it was introduced to the marketplace.

MR. BARNETT: Anybody else want to comment on that one? Anybody in the audience want to comment on that?

AUDIENCE PARTICIPANT: I am Ray Tessy from Celeron Therapeutics which is a microscopic biotech company. I think the issue of conforming labels, I think, is, quite frankly, a dream. I come from the side of not being a device company. The previous company, public company, I was with was huge. It was 300 people. My current company just hired its 21st on Monday.

So, compared to many in the room, it is a very different scale but I have dealt with the issue where we have drugs or biologics that are trying to be developed in combination with other drugs or biologics. It has been a bit of a sticky wicket.

In fact, asking another company to make changes in their labeling is universally impossible for whatever reasons that are out there. Quite frankly, it is usually about, I will say, risk management but I really mean money because if, in fact, it is financially attractive for the company to make a change, they will. Otherwise, they don't want the headache or the risk.

So I think having conforming labeling is going to be very difficult in general, certainly in the drug-biologic combination, or drug-drug or biologic-biologic. Maybe in the device world, it is different because of the nature of the beast.

But, if the FDA were to take the position that that is needed, I think they will stifle innovation. I think that, although it is not part of the FDA's mission statement to promote innovation, I think that the fact that they are having a meeting like this implies that their goal is to encourage innovation within the industry while supporting the other aspects of their mission which is safety and scientific merit.

So I would argue that conforming labels is a dream, particularly when you have the usual (inaudible) relationship equivalent to the Great Dane dating a chiuaua.

MR. BARNETT: Anyone want to respond to that? Yes?

DR. DAGHER: Not respond but just make a comment that we are talking about--just a comment that if we are talking about generic drugs, if there is consideration of having that default of that mutually conforming labeling will be required every time, there are disease settings where (inaudible) "off-label" use. It become a resource question also about whether there are resources there or if the effort, again, you have to balance safety--I am reserving that for generic drugs because, as Dr. Lumpkin said earlier, there are issues regarding the law that (inaudible).

MR. BARNETT: Any other responses from the FDA folks or anyone here to what this gentleman just talked about? Let me ask if anybody else in the FDA wants to respond to what they heard on the panel? John?

DR. JENKINS: I can see now why Susan has been working on this for 13 years. I think the guy who spoke earlier from the audience is correct. It is a dream but, in some ways, it is a dream would probably like to achieve because, if you can think of public-health goals, you would try to inform people about how to properly use the product. It would only lead to confusion if you have got disparate labeling.

On the other hand, the ability of FDA to force people to change their labeling if they choose not to, the ability to keep up with all of the changes that occur in the innovator product versus the device, is mind-boggling. So it is really hard to figure out how to get your arms around this.

I was thinking about the scenario that Ramzi described earlier in oncology where it is very frequent that there is not conforming labeling for the various regimens that are used in oncology and things seem to go okay. But then you also have to realize, and no offense to Ramzi, but the practice of oncology is really not always rational.

So people are often doing things in oncology that are not particularly rational, that are driven towards trying to find something that works. That is not the way we practice medicine in all the other areas and probably wouldn't be acceptable in a lot of other areas.

The example he gave in oncology for cisplatinum and there being other drugs that had mention of cisplatinum in their labeling, all that has been studied in clinical trials and you are at least aware of whether there is a drug-drug interaction. You are not always going to be aware of whether there is a drug-device or a drug-biologic or a device-biologic interaction. Even if you had studied it initially, later on, at the formulation changes or as the device changes, it is hard to keep track of all that.

So it is a huge problem to try to get your arms around the--I do think it is important to correct the speaker from the audience a moment ago. It is part of the FDA's mission statement to promote innovation. That was added in 1997. It is part of FDAMA. So we do have part of our mission to promote innovation of drugs and devices, et cetera.

MR. BARNETT: Anyone want to respond to that?

DR. DAGHER: Just to reiterate, too, that, although in the example I gave, I mentioned that there is some--toxicities to cisplatin, for example. Most of the toxicities for the cytotoxic drugs are actually the same across the board in terms of the myelosuppression, et cetera. Again, because it is a specific practice population that is also different, potentially, the issue that comes up with you have a device-drug or drugs that are used--

DR. JENKINS: Mark, I would also like to mention, this concept of general-use devices is also an interesting one that we have out there and yet there are issues that can arise with general-use devices as well. I recall one of the early reviews I was involved with when I joined the agency back in the early '90s was the pulmazyme application for DNA in cystic fibrosis where it was administered by different nebulizers.

We were struggling at the time when we were deciding to approve it how to label that product for use with nebulizers. It had been studied with a couple of nebulizers and, while nebulizers are approved as kind of general-use devices under 510(k)s, I think it is safe to say that performance characteristics of one nebulizer versus another is likely to be different.

So we thought we were doing the right thing by putting in the labeling which nebulizer was used in the clinical trials but, of course, with the life-cycle of nebulizers, those devices have probably changed dramatically in the past ten years. Yet, I bet the name of that nebulizer is still in the label today.

I may be wrong. It is probably not the same nebulizer that was used in the clinical trials to get the product approved and we really don't know if those changes over time have had no impact on the delivery of the drug or may have made it better, or may have made it worse. So, even the general-use devices can become problematic, particularly when they have unique performance characteristics. It may not be so critical for an I.V. infusion pump, but it may be much more critical for something like a nebulizer.

MR. BARNETT: Thanks.

Anyone else? Yes.

DR. PROVOST: I have a question for Dr. Eveleth, or maybe an expansion of what you said or, if anybody in the audience--I think one of the points that I was really hoping to learn a little bit more about is the whole of question of, well, why not. Why don't companies work together more easily? From where I sit at FDA, boy, that would be the best thing, as Suzanne alluded to, because we wouldn't have to deal with all these really difficult problems.

I know I heard, well, there might be something about safety regarding the new use and also some general things about economics. But is there any other reason, anything else, that can help illuminate this for us.

Then I would also follow that up with the kind of question of, well, is there something FDA is doing that stands in the way of cooperation. If anybody has a comment about that, I would very interested to hear that.

MR. BARNETT: You have to get very close to the mike.

DR. EVELETH: I don't think that FDA does anything to inhibit the cooperation, at least from our perspective. There are a myriad of reasons why companies might choose not to cooperate. But most of those, at least from our perspective, are easily overcome if there is authentic public-health need; that is, if there is big public-health need, there is an opportunity for us, in every sense of the word.

So the way we have become successful is overcoming obstacles to satisfying those needs. We work with lots of companies that we see obstacles with but we work with them because we see it as a way to satisfy an unmet need.

I have been through this before and I have seen situations where we might not cooperate with a company because we actually have plans to do the same thing ourselves and we are going to reveal that. We might not cooperate with a company because we have some information that would suggest that we really don't want to go down that road because there are safety issues.

I could see how some companies might choose not to cooperate simply because the cost of cooperation is larger than the benefit they see themselves getting out of it. An example would be the one raised by one of the FDA colleagues around an indication for which physicians are widely using a drug off-label.

For us, physicians are already using the drug off-label. We don't want to encourage that nor, necessarily, do we see it as--I mean, this is within the physician's--within their purview. Where is the benefit to us to cooperating? Where is the benefit to us in cooperating for that unless, by cooperating, that market expands a great deal and we can serve a lot more patients.

Usually, that is not the case although sometimes it is.

MR. BARNETT: Anyone else? I'm sorry; go ahead.

DR. EVELETH: I would like to ask some other folks because we also have this tendency--I have to say that we look for big opportunities to satisfy unmet medical need. Often, device companies come with what we see as opportunities that are not so big, or maybe opportunities that are not so clear. I do know that a number of times when we have disagreed is the situation where our view of the magnitude of the unmet medical need is not the same as the company in question.

MR. BARNETT: Yes?

DR. GOLDFARB: To take a moment and address this, I suppose, from the opposite end of the spectrum, it seems to be that when you embark upon--

MR. BARNETT: Get a little closer to the mike. I'm sorry. I want to make sure everybody hears.

DR. GOLDFARB: I think it seems to me that, when you embark upon these negotiations, the sticking point is never the money. It is the money. So, really, when we talk about unmet clinical need, I am not sure if it is a clinical need or an economic need. But if we just take Genetronics as an example. We have a novel way of using a drug that is a generic drug in a dose that is significantly lower than what would be used as a systemic dose.

Now, granted this is in indications where it wasn't being used before but, still, as the drug manufacturer, we are not coming to them with this great unmet clinical need because, if you have a company that makes Tagamet, then making a drug to treat recurrent head and neck cancer, you can't even figure out if there is a conference-room small enough to have people come and talk about it.

So those, I think, are the issues that we face. So what do I feel the FDA's role is? I grew as what I was, which was a New York City liberal person. So my view of what the role of the government is, if nothing else, you are an honest broker that can bring these people together to sit in a room to discuss this face-to-face with you to look at is there truly an unmet clinical need and does it pose a risk to the drug manufacturer because it is really going to put them at risk.

The people who manufacture bleomycin really don't want to be involved with us for a whole variety of reasons, not least of which is the drug was approved 30 years ago and they really don't want to be opening those files and going back and looking at all of them again.

So the way we have resolved it seems to meet the clinical need without threatening the people who are manufacturing the drug and makes the assumption that the drug, as a generic, will remain constant over time. I must say, I thought that was an elegant resolution.

I think there are other resolutions like that out there but it only occurred because we actively involved you in reaching that resolution. I think if we had only worked with the drug manufacturer, we never would have reached that resolution. So, yes; I think there is a very active and positive role that the agency can take in this even in the current administration.

MR. BARNETT: Ignore that last comment. It is an interesting area for getting FDA educated about this issue of cooperation. Anyone else want to shed some light on that issue?

AUDIENCE PARTICIPANT: Suppose that Pfizer has a particular drug and they own that drug exclusively. As a device manufacturer, I go to Pfizer and we try to work out a deal. It will either happen or not happen.

The situation where, let's say, I have a design that will be used with a variety of antibiotics, who do I talk to? There are so many players in the scenario, it just makes it too ungainly to work with.

MR. BARNETT: Thank you. We are due for a break.

(Break.)

MR. BARNETT: Mark Kramer, I think you had a question, did you not?

DR. KRAMER: Thanks, Mark. I am Mark Kramer from the Office of Combination Products. I had a question for Leighton Hansel. The question is really more from Leighton's experience at working at FDA as one of the drafters of the CDER-CDRH Inter-Center Agreement and also Part 3 of the regulations.

I was just wondering if he could give that perspective rather than now as an AdvaMed member on what course the Inter-Center Agreement, the treatment in the Inter-Center Agreement, of cross labeling, what the founding fathers and mothers at the time really were thinking and also in terms of the language of 3.2(e)(3)--that is when you have lost a drug label.

My question again is for Leighton to comment on, from his FDA experience, the meaning of the CDER-CDRH Inter-Center Agreement's treatment on cross labeling and also the definition of a combination product 3.2(e)(3) which is the critical provision we are talking about today in terms of cross labeling.

MR. HANSEL: I will give that a try. I was on the writing group for Part 3 and I was also one of the coordinators of getting the CDER-CDRH Agreements in place. For writing Part 3, in retrospect, I wish it hadn't gone direct to final but, in the preamble, I think we made it clear. We gave some examples of the last two--we gave an example of each of the last two items on those list of four. I, unfortunately, thought I had a copy with me and I don't.

In the preamble, we noted that most of the drug-device matters would be concomitant use and those did not fall within the area of combination products. So the goal, as I mentioned, at the RAPS Summit, was to keep the number of products that were subject to combination products with respect to labeling limited in number and only where it needed to be in place because we, as a writing group, felt that we didn't know what all combination requirements would be applied in the future but it was highly likely that it would be more than what would apply to something that was either solely a drug, device or a biologic.

In the Inter-Center Agreements, you have to be careful because two things were trying to be accomplished. They were finalized about the time the final reg was going to publish so we were all aware of content and we thought we made sure there was nothing in the agreements that would conflict with what was in our 3 Regulation.

But we were fixing a couple of things. We were trying to identify groupings of products that were combination products, those that were not, but also fixing some internal administration process where a center would have to approve another center's approval submission or whether there would only need to be a consult because there had been differences in agreement, whether it should be a consult or whether it should be an approval or whether there should be multiple submissions.

In that document, I think we also tried to keep the concept that--in retrospect I found, and I think it may still be somewhat true, that, given the regulatory processes that CDRH uses and CBER and CDER use, the same term may mean different things to people, especially like "no change in intended use."

Well, I think in the device world, that allows a lot of latitude which might not be there if you are looking at a drug approval. Consequently, the idea was to try to indicate that there would be some latitude in labeling conformance and still not require like a drug NDA to be approved. It was more along the lines that the drugs people could work with devices, identify issues that really needed to be addressed in the device labeling with respect to, say, some unique element that was going to be in there.

As mentioned by Lily, insulin was a perfect example of where the labeling clearly had the product in a refrigerator right before it was injected whereas, in an infusion pump, an external reservoir, that was a different storage condition.

Well, let's take technology and I come in--you know, that is one set of circumstances. Let's take technology and I come today and I say, oh, I have got a mini-refrigerator built into my reservoir and it will maintain the insulin at that normal storage temperature on my external thing on the patient and will warm it up right before it is injected to meet--so it is completely consistent with the labeling for the drug.

I mean, that would be a different set of circumstances. So we were trying to allow for--we figured we would be seeing all kinds of new technology. We didn't want that technology to necessarily be stopped dead by, say, traditional interpretations of existing requirements and that would allows some latitude.

MR. BARNETT: Any response to Leighton? Any comments on that?

MR. HANSEL: There is no one left alive that was involved so I can't be--

AUDIENCE PARTICIPANT: There are a couple of things. Consistent with Hansel's iteration of what the original intentions were and also in relation to two things that Suzanne said earlier about what FDA's role should be and regarding whether they should take a role to assume that things should be always be changed or should be changed only if it is required, and also to bring in the definition of cross labeling versus mutually conforming.

I do think there is a difference. I think that cross labeling implies some sort of active role and that I don't think that the FDA, at least in Hansel's iteration, wanted to have an active role to force that the labeling be changed even if it was mutually conforming which I think assumes a passive--mutually conforming means that there is no adverse or specific contraindication.

When it comes to looking at precedents and looking at what has already happened, there are numerous precedents that you would be able to see where the indications for use or the route of administration or all of these slight differences were overlooked and products, drugs, were approved with other drugs referenced or devices are approved with general or very specific indications but they are not exactly the same.

So I do think that there is a role for the FDA in assuring mutually conforming labeling but that mutually conforming labeling should have flexibility. It doesn't need to be exact and it does need to take action that would require and exception, not everything be required to have mutually conforming labeling because if you took that action, there are thousands of devices on the market that wouldn't be on the market right now; auto-injectors, injection pens, infusers.

Most of these associated injection and infusion and delivery devices, if you looked at them closely, they wouldn't meet this requirement and you would have to force the drug company to specify the syringe; is it only glass syringes? Is it glass and plastic syringes? Somebody may come up with a paper syringe, an elastomeric syringe. You would stop the drugs and everything would be off-label.

I congratulate Hansel for his forward thinking in writing these things, in helping write these things, because I do think that the FDA can be flexible and can look at what is already there to come up with a policy that would allow a flexible use of the existing labeling to advance the delivery of drugs.

MR. BARNETT: Thank you. Before we get a response--well, go ahead, respond to it, Leighton, if you would like.

MR. HANSEL: But it does bring up a point. I think it is important for Food and Drug to look at the things that are working, that drug company does clinical research to support their submission with a particular device but there are all kinds of different devices and manufacturers out there and other things where labeling is cross-referenced, like antibiotic sensitivity discs, things of that nature, and find out what is working and why it is working and don't do anything that would cause that to stop working.

MR. BARNETT: Let me throw in a question I don't think you have maybe discussed yet. What if Product A starts out as a brand-name drug but, several years later down the line, it goes generic and now we have 12 or 15 Product A's. Does that pose a problem at all?

MR. HANSEL: I thought about that because I think there is a change over time. An example that comes to my mind is a drug for treating psoriasis. When the research was done, the drug company determined what was the appropriate U.V. wavelengths to reach the best result for the treatment. That company chose to put those criteria in their labeling.

So, conveniently, there was equipment in the marketplace that could do that. But the device companies decided that they wanted to make that claim and they, in order to make that claim, had to come to Food and Drug, to CDRH, and show that their equipment had those wavelengths and capabilities.

If you had started with one device company and, over time, multiple drugs came out, how would you handle the fact that there would be competitive device companies that would want to come out with "me too" devices. So I think you can have an original approval process but, over time, like more drugs coming on the market or more companies wanting to see "me too" type devices, you are going to have policies and procedures that allow for that change in environment.

MR. BARNETT: Anyone else want to comment on that question? Yes?

DR. DAGHER: Just to say, as I mentioned earlier, not so much a problem or not but that the answer to the question I think could be slightly different when we are talking about generics versus when we are talking about the other scenarios. I think that was obvious from the presentations--all I was saying was, as you saw from the different presentations, sometimes the solution to the question we were asked today can be found somewhat easier when you are dealing with a generic drug than when you are not. So there is not one straight answer to the question we were tasked in answering today.

I think formulating the answer could be a little different when we are talking about a generic drug or device versus we are talking about something that is not generic or a new drug.

MR. BARNETT: Thank you.

Anyone else out here want to bring anything up? Someone in the back. Come on up.

AUDIENCE PARTICIPANT: My name is Jenny Greenhorn. I work for G.E. Healthcare. I would just like to share some experience and make some comments of the interesting experience over the last year where Company B has actually acquired Company A.

For those of you who are not familiar with G.E. Healthcare, G.E. Medical Systems, as it was, who make big scanners, imaging equipment, bought Amersham last year. Amersham makes the contrast media regulated as drugs that go with the equipment. Previously, we went on quite happily without speaking to each other and never the twain shall meet.

I am now sitting across these two businesses. What has been very interesting for me to see is that, actually, some of the collaboration, even when you are one business, is quite difficult. A lot of this is down to, I think, the process, the philosophy behind drug labels and device labels is actually different and there is also the postmarketing processes of reporting, pharmacovigilence, medical device reporting.

One of the really key issues is advertising and promotion of the difference in which they are handled. So, even when you do have collaboration, it is not always going to be plain sailing. I would just like to know whether anyone has got any comment on process rather than just the actual content of the label.

MR. BARNETT: Any reaction to that? Anybody up here or in the audience? We had another gentleman in the back, I think.

AUDIENCE PARTICIPANT: Hello. My name is Anthony Watts and I am the Chief of the General Hospital Branch. I guess a lot of the discussion revolved around general-use devices and General Hospital Devices Branch, that is sort of where I live. So I feel like I have to say something.

The first thing I wanted to talk about was--or actually I just have a couple of comments. I believe it was Mr. Hansel mentioned, in his speech, about the need to communicate early if FDA feels like mutually conforming labeling is necessary.

I think I agree with that, that what sometimes strikes us is we don't know until we get a submission that this is going to be an issue. So it helps a lot if we can get some early collaboration. I would say if this came up five years ago, I would be saying I don't see any early communication and we would be finding out at the last minute.

It is tough for us to sit there and say, have you talked about this with CDER, at the last minute because usually we get funny looks and comments and things of that sort. But, over the last, I would say, couple of years, it has been much better. But it can be better and I think, as much as we realize that a lot of the device companies are small and they are usually working from limited funds, if we can get that information early, that would really help us inform that process better.

So that would be my first comment. The second comment is regarding cooperation. I find this discussion about cooperation kind of interesting because, from my perspective, it is nice to know why companies don't cooperate but it really does not answer the question of can we still get the information we need whether or not there is cooperation.

From my perspective, some of those questions can probably be answered without cooperation, but that doesn't help for the business aspect of what goes on between the companies. I think that is sort of like the white elephant in the room. If we do ahead and get the information without the cooperation of Company A who might make the drug, we get that information from Company B, we don't know what the impact of that is going to be and the question is really should FDA really even be considered that.

I think, at some level, we do if we want to be in the real world. So those are my comments.

MR. BARNETT: Thank you. Any response? Yes.

DR. DAGHER: Not as much a response but another comment somebody in the audience brought up during the break and they may want to come to podium. They asked about the issue of diagnostic tools or kits. At least for me, in preparing for this, although there could be cross-labeling issues there, I didn't perceive that these would usually be combination products.

But if that person wants to discuss that or now wants to give their opinion on whether they viewed that as something for discussion or not. That might be worth--

MR. BARNETT: All right. Anybody?

AUDIENCE PARTICIPANT: Hello. I am the person that brought that up. I work for Doco-Cytomation[?] We manufacture cancer diagnostic kits. The reason I am attending is because I feel that--whatever the ground work that is laid here will certainly set a precedent in how we conduct business because we have a lot of pharmacodiagnostics and our portfolio is growing in that direction in the spirit of personalized medicine which I think there are several other people here that share that concern.

So I suppose I can give a hypothetical example. We have a kit where our labeling actually refers to a drug. I guess I just wanted to go on the record and say, what happens when a new drug is added for the same indication especially in cases where the drug manufacturer has used the already approved kit to enroll patients in their clinical studies in support of their NDA or BLA.

Does that go on the onus of the drug manufacturer to add that kit into their labeling or is it on the side of the diagnostic manufacturer?

The second point I just wanted to say because I don't plan on coming back up here was that, in the spirit of--there was some discussion about cooperation between companies. In the case I mentioned to you, our patient--you know, the patient population for what the diagnostic kit is manufactured for might not increase because all you are doing is adding a new drug. But it is the same indication.

So is there an incentive for the diagnostic manufacturer to go ahead and supplement a PMA in this case to add the drug to the labeling?

Thank you.

MR. BARNETT: Does anyone respond to that, either in the audience or up here? Yes?

AUDIENCE PARTICIPANT: That actually is the same that Dr. Jenkins brought up earlier when he was talking about pulmazyme. It has to do with when you approve a drug and a device together, in this case it would be a drug and a nebulizer, and the case that you are talking about is a drug and a diagnostic kit, they may cross-reference each other because they are the first on the market.

Then you get, in the case of a diagnostic, more drugs, or more drugs that are being indicated for the same indication or new indication which still require that diagnostic kit or, in the case of the device, maybe more devices that identify that analyte or that are predictive of that analyte or diagnose or bind that analyte that may be also used with that same drug. And you are stuck with drug labeling saying only for this device and the device labeling saying only for the drug.

In the case of pulmazyme, it was one of the nebulizers. That is where you have a problem with many of these which are truly combination products only upon approval but then generic devices or generic drugs come out and there is no incentive for the drug manufacturer or the device manufacturer to take the time or the effort to change that labeling because it is going to be used anyway. They are going to use that drug for that indication or they are going to use that device for all the different drugs.

In this case, in pulmazyme, I think Genentech is now in the situation where they have one or two or three nebulizers some of which are obsolete. Some they have been notified are no longer available and now they have to go through a whole process of an SBLA or an SNDA and there is no indication or no ability of the FDA to allow that to happen very quickly or very easily.

They will be asked to go and do the same types of clinical studies and I don't think that is reasonable. So these are truly combination products but they are only combination products because they can only be approved one with the other because they don't work together. But, once they are approved, it is fair game and everybody comes out with their generic drugs or their generic devices.

I think that when you are approving these combination products that some thought process needs to go into the fact that this is one of these products that will, in fact, generate generics and so don't name the brand. Try at least to be very specific about the physical characteristics, the particle size, the nebulizer plume, the specific analyte that is being identified and not the specific brand name of the drug or the device.

MR. BARNETT: Thank you.

DR. JENKINS: I would like to follow up on that because I would be interested in knowing if we had taken that approach in pulmazyme, how useful would that information be to the practicing physician? If we had said it was studied with a nebulizer that delivered X particle size and--if you gave the characteristics, I would guess that most physicians are not going to know which nebulizers have those characteristics.

So how useful would that be as an alternative?

AUDIENCE PARTICIPANT: Just to say that it can be used with nebulizers that are specifically indicated for this drug and let that be handled by the device side which is exactly what has happened now. Then they get the labeling in the device labeling which says that this particular nebulizer has been tested and has been shown to be able to deliver this drug with the appropriate characteristics, whatever that may be.

DR. JENKINS: That approach seems to be even more burdensome. You are asking the device manufacturers to have to show that their--

AUDIENCE PARTICIPANT: They already have to show that they can deliver the drug in the manner in which it has been--substantially equivalent to the manner in which--

DR. JENKINS: These are general-use nebulizers. They are not approved--the nebulizers don't have any specific reference to what drugs they can be used to deliver.

AUDIENCE PARTICIPANT: No. But if they are going to specifically indicate for that drug, they will be required to do that. In fact, it was an AIDS drug in the early 90's that was only to be delivered by one specific nebulizer, and it will come back to me later. Immediately upon hitting the market, you had 50 nebulizers that mentioned this unnamed drug. I mean, there were ways that device companies got around it, but the drug company was restricted to that one labeled but never ever--they would have change their drug labeling.

So I think there ways to allow the device side in this particular case to limit the access and the information to the doctors as to which devices, in fact, have been cleared and get more information that way. They won't get people trying to get around it.

DR. JENKINS: I still don't understand how the pulmazyme example fits into that because the pulmazyme studies were done by Genentech and, unless they were cooperating with the device manufacturers, the labeling for the device wouldn't say that it has been specifically shown to work with pulmazyme.

Basically, what we ended up doing, as I recall, is we said in the labeling which devices were used in the clinical trials that demonstrated efficacy and we said that it had not been studied with other nebulizers. But we didn't specifically say it had to be used only with those nebulizers.

AUDIENCE PARTICIPANT: That was close, but it still puts them in the position of now those nebulizers are no longer available or have been modified or discontinued, and now they have to change their labeling and they may be in the situation where, unless they are given some consideration, they will have to redo clinical studies.

DR. JENKINS: But it still reflects the reality of what we were facing at the time we approved the drug which was it had been studied with particular nebulizers and had been shown to be effective. We didn't know if it would work the same with other nebulizers.

If those nebulizers are now not available, I am not sure how the Center for Biologics--I guess, actually, probably we have that drug now in the Center for Drugs. I don't know how we are going to handle the issue of whether they can now name a different nebulizer.

AUDIENCE PARTICIPANT: I don't represent Genentech, so, please, do not look at pulmazyme. But that is the same idea that the Center for Drugs always has is that, just because you studied it with that particular delivery system, that means it is the only one with which it works. In order for you to go and have a different delivery system, whether that be a different electroporation device, a different infusion pump, a different syringe, that you would have to study that with all of them.

And there are specific physical characteristics that CDRH is more accustomed to dealing with that can be specified, reproduced, and determined to be equivalent that will produce the same results.

MR. BARNETT: We had two more comments up here. I think Miriam, you had your hand up first.

DR. PROVOST: Yes. I was just going to say that is a very interesting discussion. I was going to point out that approach that was suggested in which, perhaps, the labeling just lists the specifications for the product that the device is to be used with is something that we do in Devices.

An example that comes to my mind is radiofrequency-ablation catheters don't have to name a specific brand or model of RF generator. They just have to explain what kind of generator are they compatible with. But when I think when you get into the realm of a pharmacologic agent, it sounds like it is not quite that straightforward.

MR. BARNETT: And, Paul, you had a comment as well?

DR. GOLDFARB: I would agree with Lee. I think the example you also could use would be photodynamic therapy in which the drug was approved to be used with a laser that had certain physical parameters. The laser got its PMA to be used with a drug that had certain physical parameters, that they were paired approvals but neither mentioned a specific product.

So, in this setting, you would say this drug can be used with the nebulizer that has what are perceived to be the significant physical characteristics that made that nebulizer good for that therapy. If I want to bring out a new nebulizer to use with that drug, I would reference those physical parameters, I assume, and that would be the way to get the approval.

So I think the point is that, instead of approving it with a named design, we described the device and I think that now you have got two or three different examples how that could work.

MR. BARNETT: Thank you.

We have another comment here.

AUDIENCE PARTICIPANT: I hope you can hear me. Patricia Love, Office of Combination Products. I really appreciate this discussion and the range of examples that are coming up on pharmacogenomics, imaging, the photodynamic therapy as well as a range of things for us to consider.

Sooner or later, going forward, though, there will need to be either guidance or something to clarify what does flexibility mean, what are the principles that we would use. I am curious whether our representatives on the panel and industry representatives in the audience might have some suggestions on where can we begin to draw a bright line.

For example, it has mentioned generics but we probably could go down a pathway and say, well, for some generic products where you have a change and a salt or a buffer if it is with an iontophoretic device, that might be a problem. Another set of generics with another device, maybe it isn't a problem.

Pharmacogenomics sometimes is an issue. Yes; and there was a whole meeting a couple of days long with DIA and FDA trying to sort through a lot of those questions. So, are there some bright lines that we could consider for principles for when we may determine that mutually conforming labeling, cross labeling, would be necessary, maybe sometimes when it would be desired or encouraged, and times when we may actually find these other creative approaches to be most useful, actually--not just possible but really the best way to address a problem.

So, do you have some suggestions from that perspective?

MR. BARNETT: That is a really good general question. I am going to ask members of the audience to answer that as well, if they will. So, anybody on the panel first?

DR. DAGHER: The one situation where you would want to be more proactive would be if there are overriding safety issues. I don't know, from CDRH, how easily you determine that, but I think we all have examples where the overriding safety or safety versus also sort of maintaining some treatment effect that is already established, whether those judgments are different than the other judgments.

MR. BARNETT: Anyone else?

AUDIENCE PARTICIPANT: Thank you. I'm Carol Reed from Genaissance. It seems like you have already started to define some of those bright lines where you have said that--I think what I am hearing a lot repeated is physical characteristics of the device, for example, is something that carries over and, maybe on the drug side, bioequivalence might be something similar to that.

I guess, from my perspective, where I would advise some caution is on the pharmacodynamic side and getting back to this issue of drug-diagnostic kit combination products where the kit may be approved for a particular member of a class of drugs. Another member of that class comes along and maybe EGFR inhibitors are a good example where the pharmacodynamic effect of one EGFR may be very, very different from another even though they are in the same class and you would think, by their mechanism of action, they would be quite similar. When you drill down to the sensitivity and specificity and other predictive values of the diagnostic test may not apply from one drug to another.

So I will just toss that out, a couple of places to start thinking.

MR. BARNETT: Comments on that?

AUDIENCE PARTICIPANT: Since I have the mike, I can speak?

MR. BARNETT: You can walk around.

AUDIENCE PARTICIPANT: I just want to bring up a couple of--

MR. BARNETT: Why don't you identify yourself.

AUDIENCE PARTICIPANT: Dave Schriver from LifeCell. A couple of regulatory principles. I think the FDA is in the position of doing risk assessment and that, in terms of having conforming labeling, that labeling ought to be specific to whatever the drug or the device is and shouldn't conform unless there is an overwhelming need to have conformity based on a risk assessment.

I think, also, that labeling ought to be specifically to what is needed by the practitioner to use whatever is being presented properly. We all have discussed that it is not necessarily a dynamic thing that keeps getting updated. You have to be able to give people an appropriate reason and a method of using whatever the product is.

MR. BARNETT: Thank you.

Yes?

DR. JENKINS: That is actually a point I wanted to hear more from the audience about is we clearly are trying to provide the information that is informative to the end user and, in many cases, for device combinations, it is going to be physicians. In other cases, it may actually be the patient.

Do you have any research, have you done any work on trying to figure out what does the end user care about in these situations. We are hypothesizing situations where we think it is important to let them know which device or which drug or whatever, but have you done any research or done any surveys to try to figure out what does the end user want to know about how to use the device and do they care if there is conforming labeling or not.

AUDIENCE PARTICIPANT: Tessy again from Celeron Therapeutics. Dr. Jenkins, I think that is a perfect segue into one of the questions I had based on a comment that Suzanne O'Shea made at the opening about--I will use the word "fascination" with labeling that the FDA has because, in fact, the labeling is not as powerful a communication tool to the clinician who I am going to accuse as the end user in this discussion as, in fact, the literature.

I think three or four of you have mentioned the role that the medical literature has in setting both medical precedent and even standard of care. So, in many ways, at least in my experience, there is far more emphasis--my experience is not in primary--but far more emphasis on the use of medical literature whether it be randomized trial, well-done or otherwise, even non-randomized trials, well-done or otherwise, are frequently not audited.

But those very much drive clinical decision-making and use of products over the labeling. In fact, sometimes the labeling becomes often a secondary consideration if there are other sources which are felt to be high-quality sources.

That is my experience, both on the drug side. If I can ask a question that you can answer later, Dr. Witten, since you are from Cells and Tissues, et cetera; where do cells fit into this because I am very curious--you know, if you take a cell therapy and probably most cell therapies you are going to have to use some other drug or something to let's say to get them, to use a term of art.

It may be that you just choose drugs based on, let's say, immunosuppressive properties, let's say totally unrelated to my business. Let's say you have a drug for Parkinson's or a cell for Parkinson's. You give a six-month course of immunosuppression to get them in. I mean, does that become a combination product or concomitant--I think that is the term you used--for products that are used together but unrelated.

DR. BEA-TILLMAN: I am going to answer from sort of a design perspective and then I will let Celia answer. I mean, that comes up a lot. We have a lot of devices, for example, that the labeling requires six month of anticoagulation or things like that.

My understanding is that is more this concomitant use and isn't the combination-product issue.

DR. WITTEN: That would have been my answer two months ago when I was the Center for Devices. And now that I am at the Center for Biologics, I think we are aware that we have to look at these questions. I think, in many cases, that is the answer but I am learning, as I have been there, that there some wrinkles--

MR. BARNETT: This is FDA speaking with one voice, folks.

DR. WITTEN: I'm agreeing with Dr. Tillman there. So I would say that if there are some--there are some wrinkles to these questions so, if there are some of these questions that you think need to be answered early in your product development that you would like to come talk to us and you want to e-mail me, I would certainly be quite happy to talk to you and actually quite interested to talk to you because I think we need to try to figure out what the questions are we need to solve in order to create a path for these products to move forward.

MR. BARNETT: Yes?

AUDIENCE PARTICIPANT: I want to address this to Dr. Jenkins. In the device side, our development is guided by design controls which is very logical and rational because it is out of the engineering discipline. Part of that is design input. One of those things is to figure out why are you developing this, what are the needs of the end user.

So, I am not saying that everybody does this but part of that is to go to the end user for the need for this, what are the things that you need to provide, what do we need to do with regard to the labeling so that we can use this adequately. So that is the intent of this as far as a diagnostics.

You know, I think David had mentioned, that innovation, it is pretty much a juggernaut. Things are going, putting together things that have never been thought of before which creates a challenge for all of us and we all have to be our toes and look at see what is going on in biologics, drugs and devices because we need to keep abreast of what impact that might have on what we do. So that sort of goes along with being in this business.

MR. BARNETT: Thank you.

Leighton?

MR. HANSEL: I was just going to mention that, certainly in the device world, human factors is an element during product design. I could argue that, certainly on the PhRMA side, looking at the clinician, if there is going to be a need to relate the devices to their product, that that certainly would be a strategy that would be very useful.

MR. BARNETT: Thank you.

Paul?

DR. GOLDFARB: It is the only time in my career where someone has actually asked me what do I want as a clinician and so I hesitate not to take the opportunity to leap in. I think, for many of us, we look at the indications for use that come with the products that we receive in two ways. One, usually the print is too small to deal with it and, secondly, it becomes the rate-limiting step. It is really not what we can do but it is sort--your reference to it is, am I going to get into trouble if I do something at variance.

So that is how it has evolved, I think, in clinical practice. Whether that is for better or worse, I can't say, but I don't think we use it as going to it to know what to do with the product so much as we go to it to figure out what not to do.

DR. EVELETH: I think a lot of the things we have been discussing have to do with how you generate appropriate labeling for a combination where there is a drug and a device or where there are two drugs that work together. But what we can't lose sight of is that the challenge is really that this innovation is now happening constantly and divergently with the drug and the device and how to deal with that.

We have to bear in mind that the situation, at least the hypothetical that is in front of us, is that you have a drug or a device that has an established use. And now you are going to do something very different with it, and how do you deal with that because you have--I mean, I agree with the speaker that said that the literature, at least in many of the specialty marketplaces or in oncology, for example--physicians are going to be guided by the literature far more than they are going to be guided by the label, with all due respect to the FDA.

The dangers here are, as the product evolves for the indication that it was originally developed for--and you could imagine a situation where you have a specification for that product that is compatible with the device on Day 1. But the manufacturer of the drug is now modifying that specification for what it is interested in, not what the device manufacturer is interested in or that use.

What if that diverges to the extent where they are now incompatible? How do you deal with that? I mean, I certainly would not want--I know my colleagues at Pfizer would not want--us to be in a situation where a device manufacturer had made a device for one of our drugs and we decided to change the formulation, the preservation, the pH, or whatever, and, suddenly, the FDA concludes, oops, now it doesn't work with the device and that is a life-threatening event for the people that are using it.

What do you do?

MR. BARNETT: Thank you. Does anyone want to respond to that? Yes?

DR. GOLDFARB: I think Ramzi actually touched on the perfect example and that is cisplatinum was approved originally to be used as a drug. It was a rather dangerous drug that required, certainly, a lot of special care in how it was done. The work that was done in Memorial initially showing that you could use high dose with certain protection was really ground-breaking, but it really described the use in very specific ways.

Over the next 25 years, there have been innumerable new formulations of protocols using cisplatin in very different ways with different dose regimens, with different timing of dosage. And we don't all wring our hands and think we are killing people left and right. A lot of good clinical practice is you learn this stuff and, if you are using these drugs, you realize what the current state of art is.

Oftentimes, that is referenced in the approval of the subsequent drugs that got approved to be used with cisplatinum. If that model worked, and it seems to have worked well in that setting, there is no reason to assume that, if we expand that model from drug-drug to drugs-devices, devices-biologics and drug-biologics, that there is no reason to assume that the standard of care can't be maintained into the future.

Nobody has gone back. As Ramzi said, nobody has gone back and relabeled cisplatin. Nobody is having meetings to say how are we going to survive with the original label for cisplatin? It has evolved and it has worked out fine. I think that is how we should be basically looking at all of these products as these are all there to help people using other products as well and it is your responsibility to read the appropriate label.

One of my arguments initially to FDA was that, in fact, you change the label on bleomycin and say, now, this could be used with electroporation to treat head and neck cancer by injecting it directly into the tumor, that one of the hypothetical concerns would be, you would have physicians that didn't want to spend the money on the electroporation device who figured they could get away injecting it into the drug and they could get away injecting it directly into the tumor and expect to see at least part of the result that they would have seen if they used the electroporation device which, in a sense, would be doing a disservice to patients.

But I think the way it has worked out has been fine. Certainly, the way in oncology that it has worked out in terms of drug-drug interactions, it has been fine. So I would disagree that--I don't think these are major issues. I think the clinicians who are going to be embarking upon using these new products, it is incumbent upon them to know all of the indications and labels that are appropriate and not just what is out there from the old product.

MR. BARNETT: We have a comment up here and then we will go down to the floor.

DR. DAGHER: I gave the example, but, again, one of the caveats was that you had the same group of specialized physicians doing that. When we are talking about the different cytotoxic drugs, as the example I gave, you do have the same population of specialized physicians making those judgments. I don't know if that is always going to be the case in the different drug-device situations that arise.

The other thing I think--one of the things I brought up and the person from the audience brought up, the issue of the diagnostics, is that, there, we are potentially redefining the way we view patient populations. At least for oncology, that can have tremendous implications.

Although we do look at the literature in terms of our clinical practice and not just what is in labeling, I think there is something different in that when the public views what is the marketing agreement and what is in the labeling, what FDA says is the evidence of benefit in the population that is defined, that data has been reviewed by the agency. The primary data has been reviewed by the agency.

Now, if we are going to be in a situation where the diagnostic is going to define the population--it is not just going to say, lung cancer, potentially. It is going to say, if you have a lung cancer of a certain stage and you do or don't have over-expression of this or that marker, you are now defining the population.

Dr. Jenkins will remind us all that, in the regulations, you have to demonstrate safety. You have to demonstrate effectiveness. And, last but not least, you have to provide labeling that instructs folks on the use, in a well-defined population.

That is one of the reasons why I think the generalities that we were discussing I would be more careful, especially when we are talking about the diagnostics and defining the populations.

MR. BARNETT: Thank you. Let me remind you, we have about eight more minutes before we close this session. So let's have this question and maybe a couple more after that.

AUDIENCE PARTICIPANT: Hi. I am Arlene Pincus from FDA. I have a couple of postmarketing questions for Dr. Goldfarb. We have been talking about the drug changing in some way after it is approved and put on the market and that your postmarketing plan that you outlined was to monitor the formulation of that drug.

Practically, what is the lag time in between when that formulation--actually, I have three parts. What is the lag time in between when that formulation might change and when you would actually pick it up and then, if you did pick up some type of change and you were not sure of the clinical impact, what would you do about it? I mean, would you actually go to the extent of a new clinical trial in some way?

Then the third part of the question is what would you do with the devices that are already out there while that trial is going on or even after you do the trial and you find out that there is a modification that needs to be relayed to the physicians?

DR. GOLDFARB: It is my fondest hope that I live long enough to deal with postmarketing problems. It is my understanding, from working with my regulatory counsel and the FDA, that, because bleomycin is a Class 1 generic drug that anything that is marketed as bleomycin has to remain consistent in all formulations as well as the excipient products within the drug.

It is a specific model because it is a Class 1 generic. Theoretically, it should not change over time and still be called bleomycin. But what we had agreed is we will continue to do chemical analysis to show at least physically that the drug is the same going forward in time and that is how we plan to deal with it.

So that is why I was careful in the presentation to characterize the drug in that way because the issues are germane. I mean, if you had a drug that was under an NDA it could be modified in some time, and I will defer. But I think if you change the drug in some way, you have to notify people that you are changing the drug certainly if it has biologic significance.

DR. EVELETH: I would like to follow up on that a little bit in the sense that it is clear that for something like bleomycin there is a system of control in place. But for drugs that are under an NDA in a situation where information can't be shared--so, for example, we might have somebody come to you and say, well, we will test ten lots of the drug and prove that all ten lots of the drug are compatible with our device.

Well, specification might be pH of 5.0 to 9.0 and we might manufacture every single lot at 6.6 plus-or-minus 0.1. We sometimes have internal specifications that are much tighter than the actual specification that is in the file. Then, occasionally, you will have a lot that goes off.

Well, it is within specification. We know it doesn't have an impact on our use, but does it have an impact on his use? I don't know. You would only pick that up after the change was made. In addition, as Paul said, as the NDA holder, we can change that formulation and will do so to optimize its performance with the indication that we are interested in without regard, necessarily, to that secondary indication that the device is interested in.

Again, this is predicated upon sort of the bias that, as I said in my talk, that we are talking about something that is pretty innovative. So you are not talking necessarily about a drug that is normally used in the indication that the device combination is used in.

MR. BARNETT: Yes; I was just going to call on you. How about that? I was going to say, in the few moments remaining, Suzanne, do you have any final questions.

MS. O'SHEA: I would like to follow up on that point because one of the things that always comes up when we talk about this in FDA is, well, what happens if the drug gets reformulated. I guess my question is, based on what you have said, the possibility that there would be a reformulation at some point down the road--I mean, if you take that, it means that, then, you would always have to have a relationship between the manufacturer so that they could work together.

My question really is can you--can FDA, with your guidance, figure out when that relationship would be required and when it really wouldn't and the device manufacturer could monitor sufficiently and is it possible for FDA to figure that out soon enough in the development process of the device in order to make it a useful decision.

DR. EVELETH: I am not sure that there is a blanket, or there is a bright line here. But, to me, it gets back to the issue of a risk assessment. If something were to go wrong, what is the consequence of that and would we detect it and be able to correct it before significant harm had been done to patients or a threat to public health.

So there are situations where I, at least, could see that there is adequate control. We know that the drug product is not likely to vary much and the consequences that are of variance are probably not great and/or we could have a system in place to detect that.

So, if efficacy failed, it would be apparent there are some combinations where it is immediately apparent. You would pick it up in postmarketing. You could even implement a risk-management plan for the device.

There are other situations where we might ascertain that failure of efficacy would be hard to detect and/or a serious public-health threat and is, or is not, likely. That would be difficult. I think a risk assessment by the agency is first.

Now, I would caution that there are situations that I deal with, for example, where, from a device perspective, small changes in the composition of plastic in the devices make the difference for the products that are on the market between efficacy and failure for some products and not others.

So you have to be pretty careful about this. But I believe the expertise does exist.

MR. BARNETT: Thank you. Anything else, Suzanne, before we close it? If that is the case, it is now 12:30 plus or minus. We will see you back here at 1:30.

(Whereupon, at 12:30 p.m., the proceedings were recessed to be resumed at 1:30 p.m.)

AFTERNOON PROCEEDINGS

(1:34 p.m.)

 Legal Panel

MR. BARNETT: If you would all have a seat, please, I think we are ready to start. We are going to follow the same procedure this afternoon that we did this morning. The format is the same but the panel is different. This time, we are going to be talking about legal issues as opposed to public-health issues.

Let me introduce this afternoon's panel. Again, I will ask you people to raise your hands when I introduce you, because they can't see your name tags, so we know who everybody is.

Nancy Stade is a general attorney with the Office of Chief Counsel of FDA. David Fox is a partner at the Food, Drug, Medical Device and Agriculture Practice Group at Hogan and Hartson. Anna Longwell is an attorney practicing food and drug law. Kathryn Gleason is a partner at Morgan Lewis.

Then, on the FDA panel, we have Ann Wion, who is Deputy Chief Counsel for Program Review in the Office of Chief Counsel at FDA. We have Diane Maloney who is Associate Director for Policy at FDA Center for Biologics Evaluate and Research. We have Jane Axelrad who is Director of Regulatory Policy at FDA Center for Drug Evaluation and Research. And Dr. Joanne Less who is Associate Director for Clinical Research and Government Affairs with FDA Center for Devices and Radiological Health.

I think we will begin in the same order that we announce people, so I guess that is you. Thanks.

 MS. STADE: Good afternoon. I am Nancy Stade from the Office of the Chief Counsel. I am going to be talking about legal considerations in developing a cross-labeling policy. So, as you can see, I have 20 questions. So a lot of questions. Hope you weren't expecting to hear answers here because we have got a whole bunch of questions.

The basic one is can we do it. Can the FDA legally approve Product B--usually, it is going to be your device--without the cooperation of Company A and without conforming the labeling on Product A?

What we are looking for, of course, wouldn't it be nice if we could get a very simple, easy-to-apply answer that everybody understands and we can apply in the field and makes a whole lot of sense. The problem we have with that is that that brings us to no. That is the easiest answer, the answer that really avoids a lot of the questions and the problems that are going to go into this discussion that you are going to hear from the other panelists.

But I think what we heard this morning, maybe not uniformly but from a lot of folks, is that, if you are really asking for the best public-health results, no might not always get you there. So let's see if, instead, we can make friends with the complications and just dive right in to the difficult questions and see if maybe there is some way we can craft a policy that is a little bit more complicated than the policy that just says no but is still legally defensible.

As I said at the beginning, there are a whole bunch of questions that come up under the big umbrella of legal considerations. I have tried to divide them into three; can we make cooperation more attractive? The second one, which I think is really the big one, and that is what are the legal issues that arise when the labeling does not conform? The last category is changes to the regs. So those are sort of the three parts of my presentation.

The first one I really am not going to talk about too much. I think some of our other speakers are going to address that further. I will just start off by saying what I think Suzanne has said, a whole bunch of people have tried to drive home from FDA which is that cooperation is the best result and that, really, that gets rid of a lot of the problems we have in the public-health arena and in the legal arena as well.

But, as we have also heard, the drug companies--generally speaking, the drug companies don't always to play. Does FDA have any arrows in the quiver that can bring the drug companies to the table? Two come to mind and that is, again, when we are talking about Company A as a drug company. Well, maybe we have this user-fee waiver or user-fee reduction under PDUFA.

In some instances, at least, maybe not always, but sometimes maybe we can say, well, this user fee is going to be a significant barrier to innovation. So, can we say to the drug companies, guess what; come on in. We are going to give you a discount.

That is one question, the user fees. The other one is maybe, in some instances, the drug company--the particular indication that we are looking at would have an exclusivity term available to it. But I guess the question I have about both these things, user fees and the exclusivity, is really do they make a difference.

If you have a drug company that is saying, we don't want to play. We are not interested. Are those things going to get anyone to the table? Is that enough? It might be interesting to hear some more about that from the drug companies.

We probably have these tools available to us at the agency but I just don't know if that is going to be enough. The reasons it might not be enough as the next part of my presentation which is all these legal issues that come up with labeling doesn't conform.

I will just dive into what I think, what I suspect, from the perspective of that drug companies is really the big one, and that is the protected interests, the proprietary interests, of the drug companies. What is FDA doing to those proprietary interests if the agency goes ahead and says, you know what, we are going to approve Company B. Company B is going to have labeling that refers to Company A's product and, you know, Company A, you don't want to cooperate. Well, we are doing it anyway.

I have a reference to 505(b)(2). The reasons I have that is just trying to draw an analogy here. In 505(b)(2), the position of the agency is, if we approve this second drug after we have the pioneer out there, we are making a reliance not on the proprietary data but on a previous finding of safety and effectiveness.

Does that analogy work in this context? If yes, or even if no, what is the data that has to be included in Company B's application to overcome the concern about Company A's protected interest? So I think and I hope that one is going to be played out further this afternoon, so I suspect that is a really big one, the concern about Company A's proprietary-data interests.

Adequate directions for use. This is another concern in the context of the non-cooperating Company A. Is the agency's approval of Product B contributing to the misbranding of Product A? Hmmm. Well, are we creating a new intended use for Product A? Does it make a difference in this context that the approval of Product B has the imprimatur of the agency? Does it?

Exclusivity; now this is one that comes up a lot in this context and maybe it is because I am not a drug lawyer. But I sort of say, and I don't know if this one is real, when you are talking about this, by definition, you Product A. Company A is a drug company in our hypothetical. So, say Company A has some exclusivity term. We are talking here about approving a Product B that is not a drug, by definition, under the terms of our hypothetical. But does it really make a difference for purposes of exclusivity? I don't know. We will talk about that.

Then the last one that is of great concern to the agency, of course, is what mechanisms do we have to assure the safety and effectiveness of the two products used together postmarket. This came up during the floor discussion at the end of our public-health component this morning. This is obviously of very great concern. If we don't have control over the drug company--or whoever the Company A is--can changes occur to that product that affect the safety and effectiveness and affect the determination that the agency made initially that, in fact, the two products can be used together safely and effectively and can we do anything about that.

In the device context, we do have some tools that might be helpful. We have conditions of approval. When we are talking about a PMA and the 510(k) context, you have special controls. Now, are those mechanisms we can use so that Product B has some kind appropriate control over the product A going forward once the approval has happened, or the clearance.

The last category of issues I am going to touch on deal with the regulation. If you are here, you probably are familiar with it. Here it is. I won't read it but meditate on it a little bit.

The question is are there changes we can make to the regulation to help us along here? Can we change either the actual language of the regulation or is there some way that it can be interpreted that make the approval of Product B easier without, of course, harming the interest of Product A. We care about you, too.

One of the obvious part of the regulations that you might want to tease out and see what it means is individually specified. Now, how specific is that language. Does it mean the exact brand the dosage and everything? Or does it mean a chemical entity or how broad or how narrow is that language.

Of course, the result of reading that more narrowly is that it removes one obstacle from the approval of these Product Bs because it is allowing the agency to treat Product B as a single-entity product rather than requiring us to review it as a combination product.

Then the next part of the regulation that is an obvious candidate for interpretation, query whether it is a candidate also for some sort of change to it is the labeling of the approved product would need to be changed. Okay; well, this is kind of--this is really the heart of the beast, isn't it, because deciding what this language means is what gets us to whether or not we are going to be able to approve the Product B without the cooperation of Company A.

So the question is, can we develop some criteria to sort of thin out this part of the regulation short of amending it?

Just in summary, discussing the regulation, I would just like to point out that these interpretations of the regulatory language will affect whether a particular combined use will be considered a combination product. This is not the same thing as saying that it removes all the other issues that I have discussed before which have to do with Company A's proprietary data, the misbranding question, et cetera. So tinkering with the regulation might get us past one set of issues. I don't think it, by any means, clears the way.

So this isn't going to be the whole fix. It might be part of it.

So just in conclusion I would like to say we could take the position, and save me a lot of work--I think it would be a position we could probably defend legally--that we can never, ever approve Product B without conforming labeling on Product A. Very simple. Easy to understand. Easy to administer.

We don't think it is the best public-health result in all instances but we would like to hear if there are ways that we can arrive at some legally supportive alternatives to this position so we can arrive at the best public-health outcome.

 MR. FOX: Thank you very much for inviting me and especially to my friends at the Office of Combination Products for inviting me and talking about the legal issues that the agency has to contend with as it tries to shape a combination-product cross-labeling policy.

I just wanted to comment, I am very impressed, Nancy, that you are able to get the number of questions down to 20. That is an achievement, in itself. I think there are probably hundreds of questions underneath this.

I can't help but mention how timely this issue is. I was driving over this morning along Montrose Parkway now, or Montrose Road. I was deep in thought on cross-labeling issues and there, off to the side of the road, is an enormous tractor-trailer that says, Mobile Hypobaric Chamber and Blood Donation. Just a light-bulb moment. Cross labeling is everywhere, even on the tractor trailers.

As I said, I am going to talk a little bit about--I hadn't captured it this way but I suppose it is--about the "no position" where the agency starts from, some of the very, very basic first principles that the agency has to contend with the create a cross-labeling policy.

The agency, for certain, as we have heard today, is dedicated to facilitating innovation. I have seen that in both my agency experience and on the outside. But, at the same time, the agency has a very important enforcement role to play and they really can't just take the provisions of the Food, Drug and Cosmetic Act and, on the fly, riddle them with holes that may dilute the agency's ability, later on down the road, to enforce the act.

So there is a very, very careful balance here. I think it is important, as we have this discussion and move towards creating a cross-labeling policy that we appreciate some of the first principles that the agency has to protect as it goes through on a case-by-case or on a policy basis.

The basic issue I am going to wrestle with is does the Food, Drug and Cosmetic Act require mutually conforming or cross labeling of combination products. Is no not only the default position where it is "no, but," but is the agency constrained in every instance to require cross labeling. If they are not constrained, where does the Act allow the agency some discretion?

The first place--you can approach this from multiple angles, but the place that I would start with is the Prohibited Act, a section of the Food, Drug and Cosmetic Act, the commandments of the Act, what thou shalt not do. There is a long list but the ones that are the most relevant here is Section 301 which states that, "The following acts and the causing thereof," which, in lawyer-speak can be read to mean directly or indirectly, a person shall not introduce a misbranded drug into commerce--this is all double for devices--misbrand a drug or device while it is commerce or receive a drug or device in commerce that is misbranded.

Should there be any doubt, there is the famous 301(k) that say, "Or doing any other act while an article is held for sale that results in the article being misbranded." We will get to misbranding in a moment but these are some pretty rough rules. They are blunt instruments. They are designed to enforce and prevent bad actors from putting misbranded drugs in commerce or facilitating them as branding of a drug while it is in commerce.

We will see in a moment why, even for good actors, this does implicate cross labeling.

The other provision, the other prohibited act, is it is absolutely prohibited to introduce into commerce a new drug that has not been approved under Section 505 of the Food, Drug and Cosmetic Act, an unapproved new drug or the unapproved-new-drug charge as we would refer to it at the agency.

So what is misbranding. Again, there are multiple statutory provisions. The one I focus on is that a drug shall be deemed misbranded, equally a device shall be deemed misbranded--no discretion there, if it is labeling is false or misleading in any particular, or if its labeling fails to bear adequate directions for use or adequate information by regulation for each intended use of the product, the fundamental, first principles that the agency has to protect against.

There is a regulation with a rather expansive view of what intended use means. Certain parts of it are, I might say, dormant from a regulatory perspective. The agency tends not to go too far down the road towards speculating about possible uses that a product may have.

But, clearly, the intended use reg says we go beyond just the immediate labeling in front of us for the product that we are regulating and we look more objectively at the circumstances surrounding the distribution of the product and we want to make sure that it bears adequate information for each of the uses for which we know it is going to be used; so, a fundamental first-principle Food and Drug law.

Then, complementing that, under Section 505, the new Drug Approval Provisions of the Act, once again, any new use, any new condition of use, for a drug that is not otherwise generally recognized as safe and effective, anything that makes a new drug new again, to the extent it is described in labeling, it has to be approved for that use.

I am holding back the punch line as to how this relates to cross labeling, but the final piece of this puzzle is to understand just how broad the term "labeling" is. It is not just the immediate container. It is not just the package insert. It is anything, really, that the agency can say is textually related to the drug as it is moving in commerce.

It is intentionally interpreted very broadly to prevent people from trying to separate the labeling from the product as is moves in commerce and try, through manipulation or deception, to put into the marketplace labeled uses of the product that are disconnected from the actual product, itself, in order to avoid a prohibited act in the Food, Drug and Cosmetic Act.

There is an old Supreme Court case that brought this together, the Kordel case that we always point to, that says, no, no; if FDA can find a textual relationship between the written word, the written printed or graphic material that is moving in commerce, and the drug, FDA can describe to the drug the uses that are described in that written, printed or graphic material.

Now, here is the interesting question with cross labeling. Can a third party's written, printed or graphic material create extended uses for another party's drug. I don't want to get into all the iterations. We have drug, biologic, device and the whole circle, but let's just, for the sake of simplicity, just look at it from the third party being a device manufacturer and the first party being the drug manufacturer.

Can a person--for example a device manufacturer--misbrand or cause the misbranding of another person's marketed drug product? Can a person suggest uses of another person's drug that, in turn, triggers the labeling and drug-approval requirement of the Food, Drug and Cosmetic Act?

Now, it is not really realistic to think that, in every instance, a third party who talks about the use of a drug would then become, in essence, a drug sponsor and have to go and get approval for a drug before that third party could continue to talk about that use.

I think the key question for FDA, as it creates a policy, is to what extent can FDA, itself, develop a policy that, in effect, would allow the agency to authorize, say, the marketing of a device that puts a drug to an unapproved, unauthorized use, a use for which the drug would be misbranded. I think that is one of the fundamental questions.

I think Nancy did a great job of touching upon it and raising that question. I apologize that I don't have a great answer to the question. But that is, I think, one of the things that, from a legal perspective, putting aside the very important public-health and business issues that were talked about this morning, the legal perspective, the fundamental question the agency has to wrestle with.

Now, for the sake of time, I am not going to go through each of these examples. But there is a fairly long legacy of the agency taking action, not just against the fringe actors but against main-stream companies that have tried to market a device for an unapproved use of a drug.

There is a 1989, the terbutaline letter--I think it is in wide circulation. If anybody has a copy, if you could send it to me, I would appreciate it, but it was issued to MiniMed. Interestingly enough--I think, if I have this right, it was issued by CDER to a device company, MiniMed, saying, "You are marketing your product for unapproved uses of the drug. Please stop."

I tried to come up with examples from different contexts. There are other similar regulatory letters. There are instances where the activity suspended or held in abeyance, essentially, device 510(k) clearances until there was a drug approved for use with a device, saying, you know, your device meets the requirements for device clearance, but we can't allow you to market this device until you get a conforming drug approved.

There are a number of RFD, request-for-designation, decisions that are also consistent with this in which people came in with an RFD seeking to be cleared solely as a device where the agency said, that's fine, but bear in mind, you are putting a drug to an unapproved use; please take care of that side of the house as well.

Aerosolized talc and insulin pumps, among others, got some public attention. Iontophoresis is another one that got some public attention where the agency took a stand under some of the first principles that I went over saying, we, the agency, can't be in the position of clearing devices that put drugs to unapproved uses. As recent as 1999, we have received warning letters.

So, how do we get out of the "no position?" Are there offsetting statutory provisions that will give the agency some flexibility. Again, the agency comes to the table with a lot of flexibility. It has a lot of discretion to interpret the act. It also has enforcement discretion. It does not have to take action against every single violation. In fact, it doesn't take action against most violations of the Food, Drug and Cosmetic Act.

But I think there is a line between recognizing the agency's enforcement discretion and the fact that it doesn't take action against every technical violation of the act. On the other hand, expressing affirmative policies that would create exceptions, essentially exceptions, interpretations that allow for exceptions, to some of the first principles that I went over.

The most obvious place to look for some offsetting statutory authority to help us out of this conundrum is Section 503(g) which is one of only two statutory provisions that expressly cover drug-device-biologic type combination products. 503(g) authorizes FDA to assign combination products based on their primary mode of action.

I am not going to go into the details of primary mode of action, but it is, at least on the surface, a statutory provision that looks more administrative than substantive. It talks about how the agency, once it determines primary mode of action, can send the review of that combination product to a single center, a single lead center.

Then it has a short provision that says the agency can then avail itself of any available agency resources if things are necessary in order to assure the safety and effectiveness of the product. So, despite sending it to a lead center, the statute makes clear the agency is not prohibited or constrained from going and getting other expertise from another center to help with the review of the product.

Now, the question is is that a substantive provision. Does it put combination products under a new set of laws, essentially, that, for example, would allow the agency, under one application, to approve both a drug and a device use. That is very, very significant from the cross-labeling perspective because, in theory, a 503(g) creates a unique regulated article and provides sort of unlimited flexibility to the agency to draw from drug and device law to clear that article as a whole.

It would allow, for example, a device manufacturer to incorporate into its product a new drug with a new use and, if you will excuse me, to get the entire gemisch approved under on device application. It might even go so far as to allow a device manufacturer not even to have to manufacturer the drug but just incorporate into its device approval or clearance essentially drug labeling that instructs on the use of the drug.

I am going, for the moment, and maybe somebody will hold my feet to fire during the panel discussion--I will be agnostic on whether I think 503(g) creates substantive rights and allows FDA essentially to approve drugs or drug labeling under the device provisions of the Act.

We do know, though, that the agency, with the implementation of Section 503(g), and the promulgation of the Part 3 regulations, did say that it now had the authority to, in many instances--I think, actually, it would be well to repeat this phase--the default position in the regulations is that most combination products will be approved under a single umbrella application and only as appropriate will the agency require separate applications.

There are scores--I am sure the audience can reel off their own list of favorites, but there are scores of single-application combination products that have been approved quite comfortably and stood the test of time by the agency. The albuteral metered-dose inhalers. Somebody can correct me on this but I don't think there was a separate 510(k) at each instance for the device apparatus that accompanies the drug.

Transdermal patches; again, no separate clearance. Then we have the interesting drug-eluting stent where, essentially, you have a single-source drug, Sirolimus--I think it is marketed under Rapimmune, is the brand name--applied to a device. Essentially, if you are classical Food and Drug scholar, you might say that is a new use of the drug. It makes it a new drug and all new drugs need to be approved under Section 505. Yet we see the entire product was approved just under a PMA with copious drug labeling, or drug information, in the design labeling but no separate NDA.

I think that is kind of leading us in the direction to show that FDA has found ways to accept new drugs out of the New Drug Provisions of the Act based, presumably, on a combination-product theory, that there is something in 503(g) that allows the agency not to require a New Drug Application for what is, in essence, a new drug or a new use of a new drug which gives newness.

Now, I am going to wrap up here and skip over the last two slides, but, if you drill down even more, if you really want to have fun with this or torture yourself with this, you can, very clearly, see a statutory authority for the umbrella NDA. The law was changed in 1990 such that there used to be an exclusion, under the definition of drug, that said that if something is a device, it cannot be regulated as a drug, oversimplified.

That was eliminated in 1990, presumably to allow device delivery systems to be regulated, essentially as drug components, to call the whole product, the whole combination, a drug and then approve the whole thing under an NDA. There is not, however, a reciprocal provision under the definition of device.

In fact, the way the definition of device is written, it excludes most every drug. Anything that relies on chemical or metabolic activity is excluded from the definition of a device. Nevertheless, as I pointed out, for the stent decision, the agency did find a way to approve a new drug under a PMA. I think that is an important leading precedent and the question, really, is, now, how far can we take that.

There is a whole other way of doing this. You could try not to avail yourself of 503(g), to call yourself a single-entity device and get up to the line but not cross that line in the regulation that Nancy put up, that section 21 CFR 3.2(e) and say that, my device doesn't require an individually specified drug and, therefore, it is not a combination and, therefore, it can be approved solely as a device.

I think that is another key part of this exercise is determining what it means to have an individually specified parallel product. For example, when a drug has gone multi-source, when it is generic and widely available for multiple generic sponsors, is it individually specified if the device manufacturer says, I am going to use my device with USP-grade, or pharmacy-grade, generic drug X and doesn't specify a brand, a specific brand, of that drug, is that enough to get out of the combination-product regime and seek approval solely under the device statute.

I think that is another key question is articulating what it means to be individual specified and whether that means brand-specific, formulation-specific and whether that is not triggered by reference to a drug that is available as a generic.

If you are really interested in the extent to which third-party labeling does or does not misbrand another product, there is an interesting citizen petition--it is a very complicated scenario. I won't bore you with it now. But, in April, 2004, the agency was forced, under a petition process, to examine some of these issues about the extent to which third-party labeling can misbrand a first-party product. If anyone is interested, let me know and I can forward you the agency's response. It is a good response.

These questions speak for themselves and, hopefully, we will get into them in the panel. Thank you very much for your time.

MR. BARNETT: Thank you.

Anna Longwell, you are up next.

 MS. LONGWELL: Hello. I am subbing today for Danielle Miller who couldn't make it. I am another member of the Combination Products Coalition and also an attorney. I was working with her on this presentation so she recruited me. So that is why you are seeing a strange face.

We start with an aphorism. This is actually a composite of many people in the Coalition's ideas that we would like to bring before this whole group, first talking a little bit about the question that FDA had is, why might not people want to cooperate, then going into what we believe the scope of FDA's authority is--that is, what they could do and what they may not able to do--some tools that they could use now without changes to the regulations or the law, which is even harder, to encourage innovation and then policy options as well.

So, first the barriers. Why don't companies want to work with one another. This comes from a discussion with members of the Coalition who are biologics firms, who are drug firms, who are drug-delivery firms and who are device firms. So the coalition is pretty broad.

So these are the issues. First of all, the potential impact on the product. If you have an established product, you know, not everybody in the commercial world is a wild-eyed innovator. If you have got something that is going and it is going well, your plans may not include introducing a radical change to that. Believe me, if anybody is trying to change labeling for a product that is either been PMA'd or NDA'd or BLA'd, they know they are in for a big job, so just simply resistance to change, especially if things are going very well.

Secondly, one of the things, and as somebody who works with V.C.s a lot, one of the things you really want to find out about a company, especially a small one, is, okay, do they really understand their freedom to operate. Have they had a freedom-to-operate analysis?

Usually, this means I.P. But, with a regulated product, it may mean more than I.P. But even in the I.P. world, you know, do they understand that they are going to maybe have to get some licenses or they are going to have to make some arrangements in order to operate freely once this becomes commercial.

If they don't even understand the issues, the V.C. is likely to say, bye-bye. If they do understand, and they have a plan for dealing with them, the V.C. is for more likely to go ahead with that company.

So there are definitely both intellectual-property and confidentiality concerns. Once again, Company A may very well have a fairly well-thought-out strategy that they have spent thousands of hours with million-dollar-a-year executives contemplating and developing and this has no part of it, This particular new little project just isn't in the strategy. It doesn't fit.

Believe me, not being a good fit, for you that are not in the world of commerce, is one of the things that really just completely wipes out a project. Maybe it could work. Maybe it could make money. But it is not a fit. You will hear that.

Research priorities, of course. That gets down to money. We could do this but we have already assigned all of our budget for the year in that area to Projects A, B and C and they definitely have a higher priority. Costs. Product liability. This is going to, perhaps, open up an entirely new patient area. We have no idea how our underwriters are going to look at this. How is going to affect our corporate insurance?

Interesting competing products. Once again, oh, yeah, well we just signed a contract with Company X to develop something nearly identical to this. You are not maybe going to discuss this openly because you didn't want to discuss it openly. It is a top-secret project. But it happened and you are stuck and you are not going to be able to contract with another person for virtually the same thing. And existing partnerships, which is very similar.

So those are a number of the reasons, all of which we have heard from some of our members for not going ahead with what might otherwise look like a completely reasonable thing.

You don't want somebody else's issues in general to control your resources. If this is a new indication, if this is a new patient population, you have issues, then, with their product. Product B is somehow maybe affecting your product, Product A, and you don't have complete control. Again, the limits on flexibility. If Product B changes, are you going to have to change your Product A labeling?

Manufacturing concerns have been done to death here. I don't want to talk about them at all. Once again, the internal expertise issue. A lot of your investment is in your people, in what they know and how they know it, and you don't want to share that frequently to a great extent.

Once again, another one is we have never heard of you, Company B. You are not on the stock exchange. We have never seen a financial report. How do we know you are even real? Different perceptions on market size. What looks pretty big maybe to a start-up Company B might not look all that big a deal to a rather largish Company A.

Of course, in every case, the new indication, whatever it is, is going to give you more additional regulatory exposure. You are going to have a whole new set of A.E.s out there, perhaps.

MR. BARNETT: Excuse me. After switching slides, the general guidance is about 15 minutes per session.

MS. LONGWELL: Okay.

MR. BARNETT: Thanks.

MS. LONGWELL: We have already gone through this some, so I will run through it. FDA derives its authority from the FDC Act but it also has its authority from the Administrative Procedures Act and, of course, we always have the Constitution there in the background and a couple of other laws like the Public Health Service Act.

You do have broad authority. You can promulgate regulations for the efficient enforcement of the Act. There are some limitations, of course, but you approve drugs and devices. license biologicals, regulate the use, and I really mean use, now, not sale. FDA can, in various ways, actually regulate the use of products and it does require manufacturers to include adequate directions for use in labeling. It can take enforcement action against misbranded drugs and devices.

These are quotes from the regulations for which various products have been found to be mislabeled if you don't have adequate instructions for use, as David said.

FDA doesn't have the authority to require companies to work together and it doesn't appear to have the authority to mandate that a manufacturer seek an approval for a new use. At least, it certainly doesn't have that authority under its labeling authority.

The Association of American Physicians and Surgeons was a good example of that. FDA did try to use its labeling authority to enforce developing pedestrian uses and the courts seemed to find that that is not a good way to use the labeling authority.

They don't have the authority to force manufacturers to mutually conform. As a practical matter, they don't always even force manufacturers to mutually conform. And we have seen some of that. Finally, they can't reference proprietary information in a drug sponsor's file for another sponsor seeking approval of a combination product except under the 505(b)(2) which has already been discussed.

Actually, in terms of our membership, anyway, nobody thinks it is a very good idea for FDA to provide this matchmaker role. However, there are some tools for encouraging innovation that we think are quite possible. They can't require them to work together but they can encourage them.

FDA can't require drug companies to recall a product, can they? They can do it for devices but they can't do it for drugs. However, most drug companies that are approached will voluntarily recall products. So FDA just simply suggesting things or encouraging things frequently is very effective. Existing tools, user-fee waivers--we have talked about that--expedited review and approval times and the power of persuasion, as i just said.

They can seek additional tools from Congress but, once again, this means revising the law, an additional period of market exclusivity, tax incentives for voluntary collaboration. There are also policy options.

When companies refuse to collaborate, they have options for proactively supporting innovation while protecting the public health and one of them is 505(b)(2). I am hoping that our discussion will include a little extensive discussion of how that might be applied.

The other thing is they can require labeling changes if they are related to safety issues. They can facilitate the independent operation of the combination product sponsor and this is where we get into the redefinition issues. I mean, is this, in fact, an individually specified drug or not. Where is the line?

They can require a sponsor to assure that the combination product is effective under the circumstances described in the combination-product labeling and that it will continue to be effective. This may very well involve a series of protocol agreements with how do we test new lots of the drug as they are acquired.

Apply post-approval requirements to combination products. Once again, you can do change controls or studies or even registries. All these can be required as a condition of approval of that second product.

They also should consider a mechanism for notifying combination-product sponsors of changes to Product A that are not under the sponsor's control; that is, you may not be able to tell them what it is but you may be able to tell them your drug or your device has undergone a change which we have been notified of which would trigger a testing sequence on the part of the sponsor.

So there are a number of options that FDA could consider that are practical options that wouldn't involve either changing a regulation or law at the moment.

And that is the end of the story.

MR. BARNETT: Thank you.

Kathryn Gleason, please

 MS. GLEASON: Good afternoon. My name is Kathryn Gleason. I am here today representing AdvaMed and its 1200-some innovator and related health-technology companies. We are gratified to be here. AdvaMed has a very real, a very vested, interest in these issues because many of its members have spawned the innovation that has, in turn, spawned the questions that Nancy and David and Anna have all raised today for us to reflect on.

:A couple of comments on sequence. I think, to Suzanne, my commendation to the sequence of panelists. We heard from Nancy and David the tough questions and some of the legal impediments that stand in our way to an easy solution. Anna has begun to articulate some of the issues that are relevant to the questions that FDA itemizes in its Federal Register.

What I would propose to do, time permitting, is to go through each of the questions with actual proposed particularized responses, not so much to tell you that these are the dispositive answers but to have a straw man for the audience to comment on, to critique, otherwise support as appropriate, simply to move this process forward.

Let me say one other introductory comment and that is a double-up on something that I believe Leighton talked about this morning and that is the methodology that AdvaMed used in looking at these issues. They very much saw the solutions as a three-legged stool, if I could use that analogy.

I was initially charged by Suzanne to come up and speak on behalf, in effect, of the agency. I was not going to be to AdvaMed's interest but somewhere through the process, after I charged my own self with formative thoughts, AdvaMed asked me to weigh in. When they did, we began to swap notes and I think we were all gratified as to how our thoughts coalesced and our interests were consonant.

What the AdvaMed member companies told is that, in order to have this as a workable solution, there had to be due attention to the tri-part interest, due attention to Company B and its quest for innovation, due attention to Company A and its vested rights and interest in its own proprietary information and obligations and exposures and due attention to the FDA in its mission both at the premarket and postmarketing stages.

So what we did, before answering the questions, was to actually throw up on the white board some overarching, to use David's term, first principles that would guide us in answering each of the seven questions. We did it with particular attention to the hypothetical, what would be Company B's interest, Company A's interest and FDA's interest.

The first is innovation and advancement of public health. This really is Company B. Company B is emblematic of those objectives. There were two things that we stipulated at the outset. First, and obviously AdvaMed does represent the innovator interest here, it is optimal to have as much flexibility as possible to encourage the innovation and advancement of public health that Nancy referred to.

We think, and simply to get the ball rolling one step beyond David, that the law does accommodate this flexibility, that there is a broad framework in place now upon which we can build and I offer here, for example, aggregated into three bullets.

Let me take the second one first. We believe that combination law is a separate and distinct authority. It gives a separate and distinct regulatory clout to the agency and the sponsors and, in turn, offers separate and distinct solutions. We believe, moreover, that Congress, in creating this fourth class of category, has told FDA that they may have flexibility in solving problems, specifically under 503(g), we believe a very substantive instruction has been given to FDA that it may use any agency resources necessary to accomplish the objectives.

Indeed, I think the agency has already begun to regard this as substantive authority; for example, mixing and matching postmarket controls to accommodate this fourth category of products.

We see flexibility, or at least the beginning of a flexible system, in the wording of that provision that deals with cross labeling. As Nancy said, how those words are cast and interpreted creates a doorway to flexibility. We would like to work with that understand our options.

We also, particularly in recent years, particularly in recent years, particularly since the OCP has been looking at these issues consistently and critically, we see flexibility in our ICA documents. Those documents have been in place now over a decade. I would argue that they create substantive rights under administrative law principles and, if you parse those words, you see flexibility in treating premarket issues, particularly in the device-drugs ICA document.

On to the interest of Company A. We summarize their interest in a word, "fairness." They didn't ask to come to this party. We must abide by and respect their roles and responsibilities accordingly. I think our members felt that we needed to protect this company against undue responsibility and liability. It is not fair to do otherwise.

We also want to abide and uphold the proprietary information that they have spent time, money, investing in and seek, themselves, to protect. Those overarching principles we factor into our responses to the seven questions.

Then comes the third left of the stool, FDA's interest. Here we stipulated that FDA very much needs a workable solution that is sustainable legally so they will not be tortured by endless challenges and we think, as I will carry through in specific answers to questions, that a new regulatory paradigm is not needed, that the fundamentals of the legal framework are there and we can simply build on them.

We also think that FDA, obviously, needs to protect its four missions, both at the premarket stage stipulating that all safety and efficacy issues will be addressed and, postmarket, that there would be adequate mechanisms to give the oversight that FDA needs to get its job done in that context.

Question 1 talked about the reasons in the Federal Register--it talked about the reasons why manufacturers might not cooperate. I think Anna covered this very able so I am going to breeze through this. We list here, and cast your eyes over it, the myriad of reasons why a company might not want to cooperate. There 101, 201 reasons when you really get down to it.

But they are categorized roughly into two buckets; the legal regulatory exposure and the commercial concern. We give here, from the Supplement, what Anna said, and from that duplicated But we give this matrix, this mosaic of possibilities, because we think it colors and informs the next part of Question 1 which is what should FDA do.

At the premarket stage, I think we feel, as we heard from Anna, that, at least in terms of statutory requirements, there really is no clear statutory authority to mandate solutions nor would our members want new statutory authority to how FDA can do this, so the relationships between Company A and Company B are largely commercially driven and I think that is the way it should be.

We take it, though, a step further. Because the relationships between Company A and Company B are commercial, we think third-party intervention really should be driven by Company B. If Company B should want the cooperation and assistance of FDA, there should not be a barrier to that intervention but they also do not want that forced encouragement absent their concurrence. These relationships are sensitive. It may disrupt ongoing discussions. It may create undue inflammation of an otherwise fragile interaction.

We also had that same pragmatic approach to incentives. Here, again, we went back to those two categories of reasons why companies don't cooperate. Yes; you can have user fees and user-fee reductions, but we are not entirely convinced that this is an able use of the agency's time to think about ways to incent Company A. The reasons for that, again, going back to the previous slide, is that user-fee reductions and similar forms of solutions will not fundamentally solve the core imperatives they have that got them to know. If there is major product liability exposure that they are concerned with or fundamental issues with respect to intellectual property, these tweaks are not going to solve that problem.

There are so many other issues that need to be addressed in the cross-labeling paradigm that I think AdvaMed members would prefer that you cast your attention to other solutions.

We do feel, however, that you have ample authority at the postmarket stage to encourage Company A's involvement. To drive that point home, there will be another slide later on that makes that point, perhaps not cooperation by Company A but certainly active involvement to various postmarket mechanisms that are available to you. I will discuss those more in a bit.

So, to summarize, then, Question 1, we think those solutions are largely driven commercially and it really ought not to be cornerstone for a cross-labeling paradigm.

The next question talks about whether or not there is adequate protection to ensure that FDA does not improperly rely on Company A's data and what mechanisms ought there to be to further shore up that and address that concern.

We took that question and actually recast the issue a little bit in order to fully respond to the question. For this, we turn to and give two slides on what, under our existing law, may not be relied upon and what, under our existing law, may be relied on and then, from there, we built the next question, are those laws adequate. That is how we answered this second question.

We, again, begin with the overarching stipulation that any solution must protect Company A, protect their data to which Company B should not a right to reference under existing law. Bear in mind that many of the innovator companies in AdvaMed are also Company A so they see these issues both ways. They are very nonparochial about responses to this issue.

They believe that the law has very clearly and, at some length, stipulated about reliance. 505(l) makes it very clear that Company A's data is protected through the date of approval of the first-day NDA. I think AdvaMed members very soundly supported the concept that those laws should be upheld and that they drive conclusions on reliance.

There is an acknowledgement among the members that ongoing debate continues, that 505(b) debates continue to unfold. But I think that the appetite of our company is that we ought to allow those debates to unfold under 505(b)(2) authority and drug authority, that the combination law permutations are not the forum for that. There should simply be respect for the law as that ultimately gets interpreted.

In terms of what information may be relied on, our companies drew from drug, device and combination authorities to give three principles that they would like factored in to any discussion on reliance. Laws both under drug and device authorities very much acknowledge the use of public-domain information to support approvals coupled with any data and documentation that needs to be developed by the Company B.

They also--that is, our members--acknowledge and want to factor into this issue that drug laws accommodate for separate applications including for fairly significant changes, dosage, route of administration and indications of the type embedded in the hypothetical that you offer in our Federal Register.

This would be based on data that they would have a right to refer to, public-domain data and data that they, themselves, generate, but that system is available and can move separately from an NDA. They also wish to remind the agency that, for now a decade, there has been some discussion of reliance embedded in the ICA.

The ICA, in particular, says that the FDA may acknowledge that prior approvals of drugs could reduce the efficacy requirements specifically stating that an additional showing of clinical effectiveness of a drug, when delivered by a specific device, will generally not be required and they feel that point is extremely important to this hypothetical that you offer and ought to be upheld.

So, again, to summarize on the information issue, much is said under current law. This has been vetted and debated for years. The debates unfold. They don't think the combination laws is the forum for that. They do not want to reinvent solutions. They believe that looking to these other authorities is the way to answer Question 2.

Question 3 asks about adequacy of labeling. Here the group dealt with the question in two ways, first from a regulatory perspective and then from a commercial or product-liability perspective. The answers they got a little different depending on the analysis. First, again, at the risk of redundancy, they believe that, because combination authority is separate and distinct, it offers distinct solutions including concerning adequacy of labeling.

Congress has given us flexibility to interpret adequacy of labeling and resources needed for safety and effectiveness. We think the authority is there. We also acknowledge, though, that Company A may have real questions. David asked the question, and I think Nancy did, too, what about misbranding. Does awareness impute a new obligation to revise my labeling under either the misbranding authorities or the supplemental authorities.

Here, again, if you stipulate and accept the notion that our combination laws allow distinct solutions, then we ought to be able, through some form of clarification, to protect Company A from needless regulatory exposure. We offer examples here.

We offer that FDA could more expressly acknowledge that combination law represents a regulatory category with unique solutions. FDA could stipulate that Company B will be solely responsible for premarket review of labeling adequacy. FDA could stipulate that approval of Company B's labeling would not, itself, trigger misbranding exposure.

FDA may want to retain some misbranding authority--say, for example, if Company A jumps into the game and wants to, itself, promote the drug-device use. But these kind of clarifications, we think, are permissible under the existing legal framework.

Product liability perspective; here we think you need to bifurcate the issue still further into those drugs that are individually specified and, for today's discussion, we identify those as branded proprietary-specific drugs and those that are not individually specified.

Where a specific branded proprietary drug is identified, there will be questions about product liability. Company A will wonder whether the foreseeable risks render their warnings inadequate. They will wonder whether any confusion or conflicts will dilute the adequacy of their warnings and their instructions for use.

They will wonder whether or not they will become more of a target for product liability than the small innovator company that actually put the delivery system on the market. It is often the case that Company A is considerably larger with deeper pockets than Company B.

We asked the question of whether FDA could do anything to protect that interest. I think, at the end of the day, we could mitigate those concerns. but not substantially. We don't draw solace from preemption themes. Those continue to be narrowly construed. There was a Supreme Court case just a week and a half ago that gave us further unrest in that context. We don't see FDA's stipulation that Company A will assume full responsibility will fully resolve product-liability issues.

At the end of the day, really in the spirit of fairness and comedy and avoid excess challenge, we felt that, for individually specified drugs, it would have to be a contractual arrangement defining the rules and responsibility of the two parties. We saw no other way around it.

The picture was different, however--we have interposed on the hypothetical a product that is not individually specified, that is not branded and proprietary. Here, we felt that, since Company A would not be cited specifically in the labeling, Company A liability would be significantly less and Company B's obligations would be adequate from a regulatory perspective and a product-liability perspective.

Company B would assume the responsibility of providing clinically relevant information, all clinically relevant information, pertaining to A and B.

So, again to summarize this issue for specific branded products, there may be adequate regulatory solutions but contractual arrangements are needed to double up to address all commercial issues. For products that are not individually specified, regulatory stipulations and clarifications of existing law would be adequate.

The next question relates to exclusivity and the implications of cross labeling where exclusivity remains. Here, again, I will share with you the analysis. What we first said was is it legally defensible. The second part of our question was could there be challenge. Then, based on those two answers, we articulate a proposed response.

First, as to whether or not it is defensible, in the abstract, I think it is. And we have heard comments from Nancy and I think David as well on this, but certainly Nancy. The principal purpose of our exclusivity laws is to protect against inappropriate generic drug competition and inappropriate use of proprietary data.

We have stipulated that there should be no inappropriate use of proprietary data. As for generic drug competition, that would not happen in this hypothetical. If Company A has remaining exclusivity, it would be Company A drug that the users would look to use in this A-B combination.

One might ask for patent certification. I would only say, in response to that, that that is not the ordinary course of business under FDA-regulated products. Yes; it exists under a particular drug law but most people, most companies, duke intellectual-property issues out in non-FED legal fora. So we felt, at least theoretically, there would be a defense for using a product that had remaining exclusivity.

We then asked the question of challenges and we felt that, at the end of day, there very possibly would be challenges. The bigger the franchise, the more likely the challenge. Companies have spent much in developing their exclusivities and history has it that they will go far in protecting those interests.

What could they assert? They could assert that PMA process is a end run around, a less burdensome end run around 505(b)(1). They could assert that device process is an imperfect surrogate to 505(b)(2) because there are no patent certifications. They could assert that the exclusivity laws are very defined, specific and all-encompassing and you simply can't move beyond those solutions.

Would they prevail? Hard to say but, at the end of the day, we felt that FDA would receive challenges, would have to devote resources to them. At the end of the day, the Company Bs would not have a predictable pathway because there would be challenges. For that reason, products that had exclusivity remaining, again, we felt that the contractual solution was the way to go to mitigate, to drastically reduce, the risk of legal challenge on cross labeling.

Our answers were far more expansive and accepting on the generic side. Here, just to tell you what we mean by generic, it is really those drugs where there is a nonproprietary database and moves beyond generics and not just the ANDA drugs but others off-patent without market exclusivity, USP drugs, those deemed grandfathered by FDA, those OTC drugs, broad categories of therapeutic use. You see some combinations, say, for use with neurovasculature agents and things like that, broad categories of use.

We felt several things, that there ought to be optimal flexibility there because, A, FDA would not have the risk of challenge. That is because there is not an access to data and the exclusivity interests that would be entering this fray; B, that the law accommodated--the law now accommodates for flexibility through 3.2 and C, when you look at these categories of drugs, the change management is far more limited.

Once you begin drastically to change these drugs, you are in another category. So change management provides visibility for the agency and also some level of consolation that, post-market, you are not going to need to worry too much about changes, certainly less so than the new

May I get a time check, Mark? I may be that I won't be able to go through all seven questions.

MR. BARNETT: You began at 2:25. It is now 2:50.

MS. GLEASON: So I am pretty much to my amount of time. I like to talk.

MR. BARNETT: Let me take order from the boss.

MS. O'SHEA: I think you have got plenty of time.

MS. GLEASON: Okay.

MR. BARNETT: Okay; do it.

MS. GLEASON: Then I will proceed through the remainder. Thank you, Suzanne. Think at least it will give a launching point for active discussion.

MR. BARNETT: Thank you for asking.

MS. GLEASON: My pleasure. I adhere to protocol. So, again, looking at the issues of exclusivity versus generics, for the generic drugs and the public-domain drugs, I think our members very much felt that the law accommodates flexibility. We should build on it and the solutions are far more readily acceptable to the agency and sponsors because there won't be the challenges and there will be postmarket control.

The fifth question relates to the arsenal of regulatory tools that might be available to FDA to ensure appropriate oversight. Let me interpose on this the stipulation that, for individually specified branded proprietary drugs and drugs that remain on exclusivity, we are advocating contractual arrangements where Company A and Company B would define rules and responsibilities and that helps the agency with postmarket management issues, both in terms of notification to FDA and interaction with FDA, labeling and so forth.

Even so, we list, over the next several slides the wide array of authorities that FDA has that it could tweak to solve postmarket problems. We see this not in the slightest as an impediment to getting the cross-labeling job done. Company B certainly could have conditions of approval imposed on it, particularly if they are proceeding through the PMA process. Both Company A and Company B would be responsible notifying you if there were specification and manufacturer changes or other changes warranting premarket review.

The monitoring obligations of GMPs and QSRs, the postmarket vigilance mechanisms available under law a broad misbranding authority--I know Anna referred to a recent case that it seemed to clip a bit of the wings of 201 and I would suggest that, if there were postmarket concerns in asserted misbranding misgivings, that it could look to labeling and misbranding authority to change labeling as needed.

All in all, our conclusion is that the agency has ample oversight and authority over Company A and Company B to allow cross labeling to proceed, although these tools would need to be further explained in any kind of clarifying guidance.

The optimal approach for clarifying relevant guides, relevant law. If you assume, as we have, that the framework is there, then you can similarly assume that clarifications could be accomplished through mechanisms and allow for clarification and further interpretations.

I think one of the reasons why AdvaMed members thought that guidance most agreeable is that it allows for innovations and combination law to change. Combination law is not a static law and there has been some sense that, as time progresses, so, too, should the interpretations that help us get the job done administratively, we feel that guidance potentially could accomplish because, as we said earlier, the four elements of law are there. Distinct authority allowing distinct solutions. Flexibility; C503(g). 3.2 cross labeling allows for words that lend themselves to flexibility should you interpret in that fashion and the ICA document that gives great flexibility.

I am going to stop a second and tell you that, when we think about the flexibility of combination law, not all products, as someone said, analyzed for cross labeling will be combination products. As 3.2 allows, if there is not cross labeling, you could very much be, as David said, a device as opposed to a combination product.

Nonetheless, the framework for cross-labeling authority resides under our combination authority. So even if the ultimate output is a device, we believe that the combination authority gives us the solution track to cross labeling.

For clarifications, again, we went back to that three-legged stool and we asked what kind of clarifications would be helpful with innovation in Company B. We feel that optimal flexibility should be stipulated and encouraged with permutations to be driven on a case-by-case basis, of course.

We think, and I think Leighton mentioned this this morning, that terms absolutely need to be defined. We drop over ourselves in trying to understand what these terms mean. We think that there should be a stipulation that device labeling, alone, can address these issues if the drug is not individually specified.

Now, why is that important? It is not just important because it may be under device law but, remember, this triggers primary mode-of-action analysis. If you become a combination, suddenly you trundle up under the PMOA algorithm.

For delivery systems, that is essentially a rerouting of all products that are in that category because there will never be the primary therapy reaction. So, to avoid the perniciousness for that category of products, in particular, we feel this kind of stipulation should be there.

If any of you didn't quite understand that nuance, I would be happy to repeat it but it is an important one to our members.

Further, for Company B, a stipulation of the law to establish that it is sustainable, stipulations as to reliance and how you interpret that and some level of acknowledgment of ICA principles. These documents that have served the agency not withstanding some of the draftsman problems. They have served the industry very, very well and I think it is a solid consensus of AdvaMed members that they want that language to be sustained and, indeed, further encouraged.

For Company A, the two issues that must be protected, the affirmation of their proprietary law, we think, consistent with current authorities, is an able way to describe, a sufficient way to describe that, a clarification does not suffer from undue responsibility and liability. Company B alone is responsible for premarket review concerning adequacy of labeling, testing and design of the delivery-system drug.

There will be imposed on that contractual solutions for some of these other issues that we have talked about where you would acknowledge that, for certain drugs, companies would have to interact and stipulate responsibility.

Then, representing the FDA's interest, again, I think if you compromise between the public-domain drugs and the drugs that have exclusivity and are specifically identified by branded names, you draw solutions that are different, that dramatically reduces your risk of challenge. I think your position, therefore, becomes more legally sustainable.

If that is an important part of your protection when you go forward, I also think full stipulation and explanation of postmarket authorities is important for your mission and, of course, the fact that regulatory resources need to be brought to bear to answer all science in the premarket stage.

There are other issues like roles and responsibilities when you begin to parse down. You know, what happens if Company A begins to promote the product even though they didn't go through it. What happens with the interrelationship of recalls or postmarket vigilance. We would work with you and other interested parties to try to define that more

I think, in the interest of time and in due deference to Mark, I am going to stop at this point except to say three points that I hope I haven't driven home too much that are very, very important to the members.

One, existing law really can be--can serve as the framework for solutions. Relatedly, flexibility comes from existing authority. With each of the examples I have given, there is a pathway for optimal flexibility.

And, three would simply be a sentiment; that is, I am drawn by something Nancy said which was the easy answer is no. I think each and every one of the companies that participate in cross-labeling review had the strong sentiment that you could have said no.

But you looked to explore options that would allow for innovation and, for that, they have asked me to express their gratitude.

Thanks.

MR. BARNETT: Is there anybody that wants to volunteer from the FDA panel to respond to anything they heard here this morning?

MS. STADE: This is a comment that might have a question buried in it. I think any of my Food and Drug colleagues from the private bar can address this and that is I heard a lot of discussion about the flexibility that FDA has under 503(g). For those of us who are lawyers and deal with these issues, we could just talk about that issue all day long, whether it really gives substantive authority, whether it sets a series of guidelines.

But I think there should be some agreement that that authority is available when what you are dealing with is a combination product. Now, the definition we have been dealing with today in the 3.2(e)(3), I think it is. That helps us decide whether what you have got is or is not a combination product.

So the first thing the agency does, we have got an application comes in. Does the labeling need to be changed on the product that is already out there? If we decide no, under our current regulatory definition, it seems to me, we have decided we don't have a combination product here.

So I guess I would put the question out, in that case, do we have any additional flexibility or are we just stuck with the old drug-device authority.

MR. BARNETT: Anybody want to respond on the panel? Yes?

MS. GLEASON: I will take a stab at it. I think, Nancy, that you are right that technically the output of the analysis is that the product is not a combination. But I would argue that the analysis for cross labeling is very much under combination authority and it is combination authority which allows for distinct solutions. So I am not troubled by that nuance. I don't know what my fellow colleagues think.

MR. BARNETT: Okay. Anyone else? If that is the case, let me go back and ask the FDA panel again. Any questions? Shall I start down there with Joanne?

DR. LESS: My question, I guess, is for any of the panel members that would wish to address it. A couple of you have mentioned the use of postmarket controls, postmarket studies, special controls for 510(k)s. How do you envision that working? I mean, what types of special controls are you referring to or what types of studies would you envision that would actually capture some of the safety concerns that we heard expressed this morning about if the drug is--if there are changes made to the drug so that the drug is still within specifications, how would you pick that up in a registry or some type of study since it is going to be a fairly small nuance where you might not even be able to see a change in safety or effectiveness of the combination.

MS. LONGWELL: I can try some of that. One of the things that came up in discussion about the kind of postmarket controls would be a simple sequence described to us. Normally, in a 510(k), supposedly you do not address GMP issues in a 510(k) review. However, that is a portion of the law that is often not entirely--well, it is reinterpreted in terms of this has to do with the safety of the device.

Tell us how you are going to qualify your device for use with this other product; that is, do you have your own specifications for the other product?

DR. LESS: So, in a sense, you would be relying on design controls?

MS. LONGWELL: That's correct.

DR. LESS: And the design manufacturer setting specs for the drugs?

MS. LONGWELL: Setting specs for the--yes; one possibility would be, if it was a drug-device combo, that these people would be responsible for specking a drug every time they got a new lot of drug.

DR. LESS: But, if they don't have the cooperation of the drug manufacturer, how will they, then, know that that drug manufacturer hasn't made a change that could be detrimental to their product?

MS. LONGWELL: Ah. Well, as part of design control, it appears they would have to look at that drug from a chemical--I mean, they would have to have chemists that set their internal specifications for that drug.

DR. LESS: Okay.

MS. LONGWELL: So that it would work with their product.

DR. LESS: Okay.

MR. BARNETT: Jane, did you want to make a comment?

MS. AXELRAD: Yes. One comment, and then a question. I think that a lot of the discussion has confused me because I think that people are confusing combination products and--I think that this has been somewhat confusing to me because I think a lot of the discussion has been confusing combination products with cross-labeling issues. A number of the speakers, for example, talked about the use of the 505(b)(2) mechanism for resolving or solving or approaching some of the cross-labeling issues.

I was wondering if any of them could elaborate on that because I find it difficult to see what the role of a 505(b)(2) would be in simply a cross-labeling issue as opposed to an actual combination product where you actually were combining the drug and the device, or two drugs in some way.

MR. BARNETT: Response to that before we go on, because you had a question, as well; right?

MS. AXELRAD: That is--I mean, I think that people need to be really clear in their discussion of when they are talking about combination products where the products are actually being marketed together somehow and cross-labeling issues where, for example, the device manufacturer isn't purchasing the drug or isn't selling the drug with the device or anything, all they are doing is simply cross labeling it.

I think the issues of exclusivity 505(b)(2) and a number of the other issues that were talked about are different in those two situations. So I wanted someone to talk about 505(b)(2) in the context of how it would actually apply or play out in the cross-labeling area

MR. BARNETT: Go ahead.

MR. FOX: Let me try a few scenarios. I think the most common one is the instance where a device manufacturer wants to develop a device for what would be a new condition of use for the drug. Under past precedent, the agency hasn't wanted to clear or approve a device that essentially would put a drug to an unapproved us.

So Option 1 is now to go over to the drug manufacturer and try to work with the drug manufacturer to have them file an NDA supplement and get that condition of use integrated into their labeling.

When that is unsuccessful, Option 2, for the device manufacturer, is to think about, well, maybe I should try to manufacturer the drug myself. And if they are going after a drug that is 505(b)(2)-able, in the easiest instance, one that is already generic, and the change that they want to make is beyond what they can under a suitability petition, then 505(b)(2) becomes a really good place to work because, for those who are not familiar with this area, one way to look at 505(b)(2) is it is the equivalent of an incremental NDA.

You only have to submit the data, and sometimes that data can be literature, needed to justify or support your incremental change to the product. But you start with the product as it is approved so you don't have to do all--the theory is, and there is a lot of debate, Kathryn alluded to. But the theory is you don't have to repeat a lot of the preclinical work and you essentially start with the existing labeling.

Now, I think the really hard next question that we have looked at quite a bit, and, unfortunately, haven't found a pathway, is whether a device company in that instance could do the equivalent of a labeling-only 505(b)(2) submission where you use 505(b)(2) not with the intent of manufacturing the drug but with the intent of getting just labeling approved that you could ship with your device.

My read of the statute is that it is a non-starter. The statute, in too many different ways, contemplates a specific formulation to be taken through the panel process. But I think that has been talked about over the years. People have different euphemistic terms for it. But I think that is the next question because I think, oftentimes, even though you go through this thought process with a pioneering device company, particularly a small company, they just realize quickly they don't have the resources to take even a (b)(2), even a skinny (b)(2), through CDER.

I just want to make sure, Jane, because I think it is really important to make sure we are on the same page. Does that clarify it somewhat?

MS. AXELRAD: I think that is one scenario.

MR. BARNETT: Does anyone have any direct response to what David just said? If not, let's do our refreshment break now. Do we really need 30 minutes? Is 15 minutes okay for you guys? All right. It is now ten after--say, at twenty-five after 3:00, let's be back here and we will continue.

By the way, two things before you go. If you don't like to get up at the microphone, there are cards on the table. You can write your questions down and Patty Sauer with the FDA will watching. Hold your hand up with your card, and she will come over and get it from you.

The other thing is want to remind you the docket will be open until July 8 for written or electronic comments. The docket number is in your handout material.

So, see you back here in 15 minutes.

(Break.)

 Public Comment and Open Discussion on Legal Issues

MR. BARNETT: I think we will start by asking the two FDA panelists who haven't spoken yet to make any comments they would like to.

Do you want to start, Ann?

MS. WION: I want to bring us back our basic question, I think. I think we have heard a lot today, both in this morning's session and this afternoon's session about flexibility that folks think FDA has with respect to combination products and cross labeling and flexibility that folks want FDA to exercise.

I would like to try to get back to--we are trying to distinguish between the legal restrictions and flexibility and the policy restrictions and flexibility. So I would like to ask the non-FDA folks if they could--and I know this is a challenge. It is not to put you on the spot, but I think this is really why we are here today; if you could say whether there are any situations in which you think FDA should require cross labeling as a matter of law.

MR. BARNETT: First we will ask the panel and then we will go out to the audience. Anyone on the panel want to respond to that critical question?

MS. GLEASON: Ann, where FDA, as a matter of law, should require cross labeling--if you believe that 3.2 informs your analysis in this respect, I think it turns on whether or not the drug is individually specified. After analyzing that regulation at some length, our members believe that where a branded specific proprietary drug is individually specified and where it does not mutually conform, cross labeling may be needed. And FDA has the authority and can invoke that authority to require mutual conformance.

Then you get into the question of what kind of mutual conformance should there be and that was discussed this morning, and the devil is in the details on that. I think I would revert, for example, to AdvaMed's position on what Leighton said and the flexibility afforded, frankly, through ICA documents concerning the parameters that they would look to decide whether, in fact, the label does have mutual conformance.

I don't know whether that answered your question, Ann.

MR. BARNETT: Yes?

MR. FOX: From a pure legal perspective, I think the law, as it currently exists, requires that each new condition of use for a drug renders the drug a new drug. I think the law is straightforward and I think that each time the agency has tried to create express exceptions to the new drug provision, it has been sued and it has lost, starting with the Hoffman LaRoche case, implementation of DESI. It has been a pretty consistent track record where they have tried to be flexible, I think, for public-policy reasons and for practical considerations but have put one company at a disadvantage to another sending one through the new drug system and giving a pass, even a temporary pass, to another.

So I think the starting point is in the Newness Regulation already on the books--it has been on the books for a long time--that, when you create a new condition of use for a drug and it is not GRAS-GRAE, it is a new drug and Section 505(a) is straightforward. It requires approval.

Now, I think the interesting question is whether there are other statutory provisions that the agency can use to approve a new condition of use for a drug. There is a certain unity to the standard, the ultimate standard for approval for a new drug and the ultimate standard for approval for a device.

They both ultimately require safety and effectiveness by adequate and well-controlled study or studies. I think there are some examples we have gone over where the agency has found the ability to approve new conditions of use for drugs and biologics or brand-new drugs and biologics within the PMA regime.

So I would turn it back to the FDA panel members to help us understand what the legal theory was, or is, that allows the agency to approve, essentially, or allow new drugs to move in commerce based on a PMA approval. Is it 503(g)? Was that the source of law? Was it a definitional issue? Did the agency essentially recharacterize the drug component as constituent of the device?

Or, third, did the agency not consider that to be a new drug? There is a pretty finite series of options that the agency had to have run through to reach that conclusion.

MS. WION: We will have to get the people who gave the legal advice back in the room before we could answer that question.

MS. AXELRAD: If they had any.

MS. WION: But if I could have a follow-up question, I think both those answers were very helpful. But I think part of the confusion, too, is that if there is something that is required as a matter of law so we are trying to focus in on, now, again, under our hypothetical, it is Product A is the drug and Product B is the device.

But, if you are saying there might be some situations under which cross labeling is required as a matter of law, then you kind of get back to, if you have got the two parties, Company A and Company B, then it is back to the question of whose responsibility is it to do what.

Here, Kathryn, you have, on your Slide 6--I thought it was a very interesting slide that said that FDA has some authority to require more active involvement by Company A. I kind of got the impression that you might have been saying that the device company might have a lot of control and ability and that FDA might be able to do something premarket working just with the device company.

Then, once the approval happens, then FDA can make the drug company do something. Now, again, I don't want to put words in your mouth but that is kind of how I read your slide. I think that is an interesting approach. So, now, FDA maybe has to do something and maybe, under this slide, would be in a position of having to force the drug company to do something after the device company has independently got approval.

I want to give you an opportunity to tell me how I have got it wrong or right or whatever. Thank you.

MS. GLEASON: I think it is our view that, either premarket or postmarket, FDA should not be forcing Company A to file premarket approvals. However, in the postmarket scenario where Company B has gotten, say, a PMA approval, if, for example, Ann, adverse experiences land on Company A's doorstep, they have an obligation, under law, to deal with them under their postmarket vigilance requirements.

They have an obligation, if they decide to change their product, to notify FDA of those changes. They have an obligation to undertake the general monitoring controls embedded in GMP regulations.

I tried to differentiate between cooperation and the active involvement that simply comes from the panoply of postmarket controls available in affecting companies in that context. A product is on the market and they want to do something postmarket, FDA will be aware of them. I am saying that those postmarket mechanisms are available to give FDA more oversight tools than perhaps they had premarket.

MR. BARNETT: Anyone want to comment on this? This is a crucial question. Anybody have any opinions on this? If that is the case--Diane. Yes; you are the only person from the FDA that hasn't spoken yet.

MS. MALONEY: I would like to maybe have two things. One is to take David Fox up on his offer where he at least said he would be agnostic until we asked him about his opinion on 503(g) and whether or not it creates substantive rights. So if he could comment on that, and then I can maybe ask my second question.

MR. BARNETT: Good. David?

MR. FOX: Thanks, Diane. I can count on you.

MS. MALONEY: He is my buddy.

MR. FOX: I wasn't a Food and Drug lawyer at the time, but my reading of the history of 503(g) was that it started out as substantive, as an effort to require, or allow for, expressly allow for, a single application to approve both products. And there was that amending of the definition of the drug to really at least allow it for an umbrella NDA, to allow the device components to be characterized as drug components.

But, in the end, my sense was that Congress backed off and went with something that was more administrative and housekeeping than substantive. Now, maybe, in retrospect, you could look at that as being actually Congress' effort to give the agency even more flexibility beyond just a single application issue.

But, when I read the plain language of the statute, it talks about resources and it talks about people. It doesn't connect up with specific statutory authority for governing labeling, manufacturing and premarket approval. I think one of the things we learned in the tobacco exercise was that you can't just read the definitions of the statute in isolation. You have to be able to connect them up with the substantive provisions of the Act.

I get questions routinely from industry about how to do things like labeling and manufacturing controls for combination products. There is just not a really good fit. There is not a Section 9 or 10 of the Act that describes combination products and how you regulate the labeling, manufacturing and premarket review. It just doesn't connect up.

Now, I guess you take away from that that the signal is--then it is a free-for-all, part of what we tried to take advantage of in tobacco. That means you draw from drug and device authorities as you wish. But I think we learned in tobacco there are repercussions there. It is hard to sustain that argument all the way through a really careful and deep challenge.

Now, there is another provision of the Act, Section 563, that came in later under FDAMA that you could read it to recognize a separate category called combination products and it does give people the ability to petition the agency and ask for kind of a specific regulatory designation.

Maybe you can look at that as enhancing the authority together with 503(g) and then you can create a specific regulatory regime, on an ad hoc basis, for a given combination product. But that has the flexibility that I think Kathryn was trying to articulate, but straight up under 503(g) I don't think in a plain language. I see it as recharacterizing the components of the combination. I still think they are subject to drug and device law as drugs and devices. I think a combination, in most instances, is no greater than the sum of the parts and each of the parts is subject to its home set of regulations, drug and device.

You know, I think the precedent is there for, on a case-by-case basis, being able to fold the review of a drug into a device review and apply the safety and effectiveness standard kind of under one statute to the whole. I think you are effectively approving the drug when you do something like the coated stent. I think you are approving it as safe and effective for use in this combination.

Then, what you are basically doing it, it is not true cross labeling but you are allowing that device label to carry all of the needed drug information, all of the needed device information. I think you could probably articulate that you have satisfied all the necessary provisions of the Act when you have done that.

I think that is a good--I think, on a case-by-case basis, that is a good solution. It is an alternative to cross labeling. It is the concept of that umbrella approval.

MR. BARNETT: I see down at the end of the table Joanne and Jane have been scribbling away madly. Are those questions or a grocery list you are working at? Do you want to throw anything in?

MS. MALONEY: Can I just make a comment on that?

MR. BARNETT: I know, Diane, you had another one. Diane?

MS. MALONEY: I wanted to say, David, thanks very much, and I guess kind of relates to my next question which is basically I can see that things are not so simple. I was kind of looking for a yes or a no.

This goes to Kathryn. Your presentation which I think gave a lot of supposed answers made it seem like there were a lot of simple answers. I guess I would like you to maybe comment on maybe that it is not as simple as it might have come across and then maybe ask the others on the panel and, actually, in the audience as well, to comment on anything that maybe you think might need to be added to that.

MS. GLEASON: Fair enough. Diane, it is virtually impossible. I must confess I felt off more than I probably could have digested by trying to go through and aggregate my answers for all seven of the issues in the Federal Register. Less there be any misimpression, I think AdvaMed members realize that this is very complicated and that the answers really lie with far more detail that will follow in their written comments.

I do think that we have a blend of law and policy at the end of the day by way of solutions. I think there are a variety of things that you could amalgamate, David, to perhaps get at the right answer without going to Congress and opening up for new legislative authority that has unintended or intended broader adverse consequences or could serve to redesign a system that, at least for AdvaMed members, seems to be working pretty well.

It is fact-certain that precedence, particularly those recent, have built on the premise of flexibility under our combination laws. What those combination laws are and how they are articulated is the obligation for the agency with us working with them.

I take your edification well, David, that, in addition to the 502 authority, you can look to the specific statutory citation for that fourth category, separately recognized, frankly, together with an amalgamated understanding that safety and efficacy would be dealt with under essentially the evidentiary standards of both drug and device and you can get at the answer that is right.

I think, if you then pile on with policy imperatives, for example, reducing the risk of challenge for those that remain on exclusivity because they will sue. There will be challenges. Some of these other things that we talked about, eventually you could arrive at a mosaic of admittedly complicated legal framework that continues to help us out.

I think that the problem that our members had was that, if you want an easy, simple solution, it would drive you to a new legislative amendment. I think, in today's day and time, there is simply no appetite for that avenue.

MR. BARNETT: Thanks. Anyone want to jump in from out here? Anything you have heard you want to question? Okay. We have a written question somebody sent up here. It is kind of interesting, so I will read it. It says, "To stop a manufacturer of Product B from commercializing a combination product utilizing the A product, could the manufacturer of A contraindicate the use of the B product and could they promote not using the combination?"

Anybody want to try their hand at that?

MS. LONGWELL: They would have to assure that their labeling was neither false nor misleading; that is to say if, in fact, the contraindication was called for and there was data to show that it was, they would be obligated to put in the contraindication. If, on the other hand, they simply didn't want these jerks using their stuff and making money off it and they had no evidence that it was causing any harm, I believe that one could find that that was a misleading contraindication, therefore that they misbranded their product.

MR. BARNETT: Anyone else want to add that?

DR. LESS: Mark, I was going to say, in Devices, this issue comes up a lot where you have two different device manufacturers putting products together. If Company A doesn't want their product to be labeled for use with Company B, what usually happens is one will say that it voids the warranty of the other. So, even though they probably can be used together, the one company will say that it is voiding the warranty rather than trying to go down a contraindication route.

MR. BARNETT: That is what they said about my t.v. set when I tried to mess with it.

DR. LESS: Exactly. I guess that is what is striking me about this conversation today. When you look at the definition under 3.2(e)(3), we are reading that as you have a device and a drug, or a device and a biologic. But we have been facing this issue for years and years with two device companies. I am sure, in our sister centers, it is same thing. Starting back years ago with contact lenses and solutions, companies didn't want to cooperate because they wanted a bigger share of the market and they didn't want to have somebody else come along and copy their solution for a certain contact lens.

So this isn't unique to two different types of products being used together. I am hoping that, maybe within the centers, there are some ideas that we have come up with over the years for handling this.

I know, in our center, most of the time we have been doing it on a case-by-case basis but I think we probably have come up with some principles that we might be able to use when it is two different types of products that will carry over.

MS. GLEASON: Joanne, remind me what those principles are using the contact lens and the solution as a case study. Contact-lens Company A does not want to cooperate with solution Company B. What would be some of the abiding principles that you would consider as you reflect on cross labeling?

DR. LESS: Well, I guess I give you more of a historical perspective than maybe principles that have sort of evolved over time. I think, originally, when the contact lenses came on the market in different solutions, we weren't sure that all of them could be used together. So we went through this series where every single time we cleared a new lens or a new solution, they had to prove that they could be used together and that they weren't going to harm the other product.

Then, eventually, we came up with different types of lenses, ionic, non-ionic and, I guess, low-water, high water, so there were basically four different types of lenses. So the solution companies had to show that they could be used with any type of lens within one of those four and then they got that whole category.

So they only had to test themselves against four different types of lenses and then they got all of the lenses that were in all of those four categories. That worked for a number of years until the lens manufacturers decided to tint their lenses and then we had to start all over again because we weren't sure that the lenses could be used with the tinted contacts.

MS. GLEASON: Right. But, from a Company A perspective, let's interpose the hypothetical that Suzanne and colleagues gave us. Was the contact-lens company forced to change its label? I think not, but you need to refresh my memory.

DR. LESS: I think we tried. I think we tried to get both of them to have mutually conforming labeling. In some cases, it worked because it was to the advantage of one company or the other. In some cases, the company didn't want to because it was not advantageous to them simply because they were making the same--they were making their own solutions so they didn't want somebody else to be labeling their solution for use with their lens.

But it also became a huge resource issue because, every time we cleared a new lens and we required new labeling, they were printing their labeling out in bulk and we were approving those PMAs really quickly. So they had to keep changing their labeling like monthly. So, eventually, we got into the state where we said, okay, then, next time you update your labeling, change it to include this next lens instead of making them do it on a continuous basis.

MR. BARNETT: I was wondering if Suzanne had any questions for the panel.

MS. O'SHEA: I have a question about the proprietary information aspect of this. I would like feedback on the whole thing. As I understand it, if FDA reviews Product B for use with Product A, inevitably, then, if you get beyond whether we even have the authority to do that or whether it is a combination product, but, assuming we get beyond all that, does FDA inevitably rely on the manufacturing information for Product A when it reviews Product B.

If this is a new drug use, a new use for a drug, then it is a new drug and the CMC information, which will be the trade secret information that Company B just has to access to, since it is not going to be a manufacturer of the drug, that, then, will FDA inevitably be relying on the proprietary information in Product A's NDA if it does whatever it is doing in the Product B application?

MR. FOX: I am going to give another tedious answer. I'm sorry. I think the answer to your question is yes, that Product Company B is relying on information that could be characterized as proprietary relating to Product A, but they would only be doing so to the same extent that they could if they opted to make a generic of their own of Product A under Section 505(j); that is, it is akin to, I think, the agency's interpretation of this--back to this concept of 505(b)(2)--as well as akin to traditional generic drug law which does allow other sponsors, later in time, to come and rely on a pioneer's proprietary information once that pioneer is no longer protected by patent or exclusivity.

We approve follow-on or generic versions, or FDA approves them, rather routinely under the concept that, once the pioneer has had its run and its patents have expired, and it is not protected by exclusivity, the product can be referenced in an ANDA or a 505(b)(2) application and the starting point is as if you could have gotten an ANDA.

There is this thing that people refer to as the Parkman letter as a concept of how a sponsor, later in time, can take a product and may want to make a rather substantial change to it from the way it is currently marketed.

Rather than go through the process of first getting and ANDA, getting a generic version, and then supplementing their application to change the use, they can directly to 505(b)(2) and just submit information on the new use.

But, what is underneath that application is all of the pioneer's proprietary information. So I would say, in that scenario, if you have a device manufacturer that wants to put an existing drug to a new use, and that new use is outside the existing labeling, they, through one of the mechanisms we have been talking about, either through an umbrella people on the PMA side or through going ahead and getting a 505(b)(2) application themselves, can submit the data needed to show that that drug would be safe and effective with the device.

And, underneath that approval, would be, in essence, the pioneer's proprietary information. But that is a version of reliance on proprietary information that is, again the subject of a lot of debate or a lot of open issues but may end up standing the test of time, standing up to challenge, because the reliance is no more and no less than what you are allowed to rely on under the generic-drug law.

But you are giving me quizzical look, so I must be way off the reservation on this.

MS. AXELRAD: I would like to comment, too.

MS. O'SHEA: In the 505(b)(2) and the generic drug, the applicant there is going to be manufacturing that product. So it will have its own CMC information or its own manufacturer processes that the agency will look at.

In this case, the one thing, assuming we can get all the necessary safety and efficacy data, it is really the manufacturing information that Company B just can't get without Company A's cooperation because it can't get it.

MR. FOX: Right. But, in that instance, they would be using the product as it is available in commerce. Company B would be using Product A.

MS. O'SHEA: So you are suggesting, then, that the manufacturing information isn't really necessary to approve Product B with Product A because it is just sort of being assumed that we are just going to rely on it because I don't think, in the generic, in the 505(b)(2) case who would be relying on the Company A's manufacturing information.

MR. FOX: Correct. I'm sorry. If you are going to manufacture your own version, you would generate your own manufacturing information. But I think, if I remember correctly, in the Help Section of Form 1572 for INDs, it talks about clinical studies in which you are studying your product with use with another marketed product.

It says, as long as you are not changing the other marketed product, you don't have to submit manufacturing information, chemistry information, anything on that existing product. There is kind of a pre-set mechanism for how the company can study its product for use concurrently with, in combination with, not a fixed-dosage form in a cocktail style with another product.

Again, FDA sort of says, as long your starting point is the currently marketed product, you are okay. So the two scenarios are you have a device where you want to use the product as it is currently available in commerce in which case you really only need to generate safety and effectiveness information.

MS. O'SHEA: Right.

MR. FOX: Or, if you are going to make your own, yes, you would generate your own information. Then I am not sure why you would have to rely on somebody else's manufacturing information.

MS. O'SHEA: You wouldn't if you were making your own. Did I miss a major part of that issue? I just want to make sure. I mean, it is an important question.

MS. O'SHEA: I thought it was a bigger deal.

MR. FOX: It can be if you are setting about to make a drug or biologic and you have questions about, say, assays that you need to incorporate into the process or certain steps, key steps, and you are trying to draw that information out of FDA by having FDA look at somebody else's application. It is a major issue. It is a huge issue.

MR. BARNETT: Nancy, you had a comment, I think.

MS. STADE: I will just follow up with you, Dave, because you did such a nice job answering those difficult questions. So I am just going to ask, when we are in the cross-labeling context and we don't have a 505(b)(2) application or a generics application, do we have the same ability to rely, to the extent we are relying, on data that has already been submitted by a pioneer when we don't have the benefits of the quid pro quo in the Hatch-Waxman Act, or are we in a different position?

As I understand it, again, and I am not an expert on the exclusivity and Hatch-Waxman protections, but, as I understand it, to some extent, the ability to approve a 505(b)(2) application or a generics application reflect that fact that there are certain protections in that legislation for the pioneer and for their data. We have exclusivity. We have patent protections.

Now, if we don't have that same set of protections in this, what you speak of as a umbrella PMA, so we are not going to have the same set of statutory protections, is the agency in the same position as far as being able to approve that?

MR. FOX: That is a really great question, if I understand the question. In a true generic setting, or in a 505(b)(2) setting, drug against drug, you would have to at least certify to the patents. But if you are proceeding solely through a device provision and essentially coopting somebody else's drug, you are not even required to certify to their patents and does that create a problem?

MS. STADE: Yes. And I will just clarify a little bit further. It seems to be 505(b)(2) and, also, the generics provisions can be looked as a specific legislatively crafted exception to the usual rule that you come in with a whole big old NDA. Here, we are giving you a lessened burden because we have crafted these specific protections and this is the way we do it by statute.

When we don't have the same precision in the statute, when we are talking about this umbrella PMA, do we still have the same latitude to do that?

MR. FOX: I am going to dodge that question to a certain extent only to say that I think, at least as a starting point, that it is non-issue if the target drug is not patent-protected, if it doesn't have a listed patent and it has already gone generic.

I think, if you take that whole category of products, you know, older new drugs that are in the Orange Book but don't have patents listed and have long-since passed their exclusivity, I think it is a non-issue. I think that is a rich source of targets for device manufactures.

For a drug that is still patent-protected and single-source, I think that is a really good issue. I have to think about more whether I could find a way around that issue. But I think that is a really good issue.

MR. BARNETT: Do you have another comment? Go ahead.

MS. GLEASON: I was going to augment what David said to some extent with two points. One, Nancy, I think, under combination law, I would try to answer the question under combination authorities. I would draw everyone's attention to the policy that has been there now for over a decade that does say that a combination sponsor does not have to reprove the efficacy of an approved product. That policy exists independent of our drug laws in the balances between innovation and generics. I guess one of the balances in that policy articulated in the ICA is that FDA wants to promote innovation. If you will, the balance is innovation versus innovation.

The second point, David, is I agree with you that you significantly diminish the risk of challenge but these are issues that the innovators, the Company Bs, are very, very interested in because it makes a difference as to whether you have to reprove something or not. So it very meaningfully affects the clinical data obligations.

MR. BARNETT: Yes. Come on up.

AUDIENCE PARTICIPANT: I realize this is a pretty messy issue, but I am going to make it messier since this is late in the afternoon. Let's suppose that Company A and Company B--Company A is an innovator, not a generics manufacturer, and Company B comes to them and says, I have this device that will enable you to get into a new indication. And they say, all right; I will do it.

Company C comes along and says, I am going to copy Company B's device, and I use the word "copy" loosely because there are protections. But it is not so hard to do. But Company A and Company B are already working together so they are not going to work with Company C. Company C comes to you and says, approve mine independently. You either say no, for some reason, or you say yes. If you say yes, you have just destroyed all the incentive you might have created for A and B to work together.

So, what would you say to Company C if you say no and what would you say to Company B if you said yes?

MR. BARNETT: That is a good one.

MS. STADE: Okay; so A and B are A and B, and then C is a device company also? I'm sorry; I just have to clarify.

AUDIENCE PARTICIPANT: The scenario is that A is a drug manufacturer. It is a branded drug. It is not a generic. B is a device manufacturer that enables the use of A in a new indication. A and B decide to work together. But C decides they can make a device that is just as good or even better than Company B's device. But, obviously, A is not going to work with everybody.

In fact, B has convinced them to enter into an exclusive arrangement. Device companies are creative and competitive and I am sure you have run into this. I mean, this can't be the first time.

MS. WION: I guess what I am not understanding is, I think your scenario is exactly what we spent the day talking about because, if you will remember, in the morning, everyone's theme was the easiest way, the best way, gee, if only the world were like this, the cooperation would get around the legal problems. They would solve the policy problems. So we applaud your Companies A and B.

The issue we all came together to talk about was whether or not A and B are doing their wonderful thing together. C wants to do something else. Now, I don't know if you were trying to bring to our attention, should FDA care that what C is doing isn't so innovative after all because FDA knows A and B are taking care of it. That breakthrough to the public health is going to be happening.

So, in that situation, should FDA, as a policy matter, say, Company C, go off and think of something that really matters to do and then come back and talk to us, or, as a matter of law, does FDA not get to say, well, I like what they are doing but what you are doing isn't to helpful. But, as a matter of law, we have to apply certain principles of law and principles that don't have to do with whether or not you are exactly doing something that is a breakthrough or wonderfully cooperative or whatever.

That is why we are here today.

MS. AXELRAD: Just to make it even more complicated, what if you have the Company A, Company B, working together and they get approved. They get the device approved and then Company A, adds the new indication to its label and gets exclusivity for that so that you have a drug-device not combination but cross labeling that is already approved.

Then along comes Company C with a different device that wants to deal with the same drug. I mean, it raises another whole legal complication in that you are working under two statutory schemes and you have implicated exclusivity under 505.

MS. LONGWELL: I'm sorry; I don't quite understand the issue there. If, in fact, the exclusivity is the drug's extended exclusivity for a new indication, they haven't lost that because somebody else is building a device for that same application. I think that is the way of the world. If, in fact--oh; take pulmonary delivery, for example. A drug that gets an indication for pulmonary delivery may have done it with one or more nebulizers that are on the market but other nebulizers will come in and will be used with that drug, whether it is on their label or not.

I think that that doesn't seem to hurt the drug company, particularly. They may have gotten exclusivity. They have still got exclusivity. There is no other drug coming on the market with that indication.

MS. GLEASON: I guess one of the ius that I would throw out is a factor to be considered there, Jane, is whether the company that seeks exclusivity, whether or not they have access to the clinical data that is essential to their approval.

If the clinical data was born from a PMA because Company A, in the first place, didn't want to cooperate with Company B and it proceeds through and gets PMA approval independent of Company A, does Company A have access to that clinical data that would serve as a basis for the essential clinical trials to sustain the exclusivity, I suppose, would be the question. Maybe not.

MS. AXELRAD: Also, what kinds of protections does Company B have against Company C from doing the same thing since exclusivity isn't covered under the device approval.

MR. BARNETT: We have somebody in the audience? Do you have a new topic?

AUDIENCE PARTICIPANT: No; this is on the same topic. I think that this particular situation exists and I think we have brought up a number or examples. Pulmazyme and Herceptin are two of them. The reality is that these products that need to come to the market that have to have a drug component and the device component where the drug component has no function--cannot be delivered without the device component, the design component, if it is specifically indicated for that drug has no use, are combination products by definition but are only combination products for the first approval.

After they are approved, then you have the section of the cross labeling that says that if there is an approved drug product on the market and I want to come out with a device and I won't change its labeling and, again, this is what we have been talking about all day, then I don't need to--I am no longer a combination product but I am a drug that can get drug approve--excuse me, a device that can device approval.

Herceptin and Her2Nu diagnostic tests were approved together as a combination, two separate approvals, but concurrently because one couldn't be approved without the other. But, subsequently, there have been a number of other products that are on the market specifically for that diagnostic analyte that are used in concert with herceptin. The same thing with nebulizers, the same thing with injection pens. And there a number of products on the market.

The reality is that the exclusivity is only for the first time through. But once that exclusivity--once approved, then you have fair game for the non-innovators, the "me too" device marketeers who will come out and develop substantially equivalent or PMA products that do the same thing.

Now, as far as the drug company, it is just more people using their drug. But, for the device company, that is fair competition. That has been fair competition since device law has been around with substantially equivalent and PMA.

MR. BARNETT: Comment on that? Anyone?

MS. GLEASON: It is definitely a general exercise, Mark, in self edification in the verification law assuming--it does seem to be a complicated series of what I would call second-generation issues that would probably benefit from and give-and-take that is more deliberative than this Q&A session today with specific questions asked and full thought given to what the answers would be.

It does reflect the possibility that conclusions today could have unwitting results for tomorrow and that there could be some maneuverability that we have not yet thought of.

MR. BARNETT: Mark?

MR. KRAMER: I have three questions. Two are related to individually specified. Kathryn, you explained AdvaMed's position that individually specified means a branded proprietary product. I am just wondering how others on the panel feel about that, because you could take the example that we had this morning that Dr. Goldfarb presented with bleomycin. Bleomycin is somewhat individually specified. They might not be the same brand name there but bleomycin is more specific than, let's say, a chemotherapeutic agent.

I was just wondering how much consensus there was about what individually specified really means.

MS. LONGWELL: We have thought about that, too in CPC. We thought about trying to define it. We thought, you know, this is obviously a grey area. Nobody has defined what individually specified means in any regulation or even in a guidance at the moment.

So, this is wide open. The definition suggested by AdvaMed is one that, you know, could be used and would serve to clarify a lot of area and would serve to justify a lot of what FDA is already doing. Those are a couple of good reasons to pick a definition.

So I would say you don't really have a definition now. We didn't really feel like offering a definition but, if you want one, that sounds pretty good.

AUDIENCE PARTICIPANT: Okay. The related question that I have is what you think about when a product must be individually specified. For example, we sometimes get new submission for devices that are intended to deliver drugs in new ways. They might not be intending to name any drug at all or even a class of drugs. Let's say, they want to deliver ophthalmic drugs in a new way, instead of having to use a dropper, some novel method.

It could be across the board for any one of a number of drugs. How do you think we should deal with cases where a company may not want to individually specify a drug product but we think that the device has no life outside of a specific drug.

MR. BARNETT: Anyone?

MS. GLEASON: I'll take that. Mark, I think, at the end of the day, we have to be intellectually honest about what the delivery system or the innovator company is trying to do. If, in fact, safety and efficacy is such that they need to demonstrate with a particular drug and that drug remains on exclusivity and/or is a branded, proprietary drug, then the company is, in essence, trying to do indirectly that which they could not do directly if this theory upholds that cross labeling is triggered by branded proprietary references.

So I think you do have to understand the details to answer the question whether there is a branded proprietary drug at the end of the story. I think, without speaking to AdvaMed, I would think--

AUDIENCE PARTICIPANT: I just wondered, when you proposed your definition of what individually specified means, if you were think it was a way to allow devices like this to really be control of whether they need to be individually specified or not; in other words, for them to say, look I am not going to name any drug product, therefore I am not part of this whole quagmire.

MS. GLEASON: It very much turns on the specifics. Let me take a flush catheter with, I won't say an embolic agent but I will see neurovasculature drugs. If neurovasculature drugs already use flush catheters and they want to have broad categories, it strikes me that the device mechanisms, that the evidentiary standards apply to that category of devices that would allow that broad, unspoken, nonspecific drug reference even if some of those drugs out there may be branded and proprietary.

So I really do think that you are driving at the right issue but you have to drill down and understand the particulars and that should be part of the critique. But if the details, both on the device side and the drug allow the flexibility.

Similarly, with Paul Goldfarb's example; I don't know whether that oncologic agent--is Paul in the room--is multisource or could be multisource, but I would argue, if it was multisource or could be, then it should not be deemed branded and proprietary.

MR. KRAMER: I have one final comment. This one is unrelated to the--specified. This is for Anna, but anybody else that would like to comment. This has to do with your suggestion that FDA should, in a sense, stay out of it between--in terms of engaging cooperation between the two companies. But, on the other hand, we should encourage it.

To me, that is somewhat of a mixed message because I know that when, let's say, I call a company and I call and the receptionist answers and they say, "I'm sorry, she is not here right now." If I say I am so-and-so from FDA, "Hold on," and they will find them.

What that means to me is sometimes when we are calling from FDA it is a little bit different. I am just wondering how can we walk a line, or how would you propose that we walk a line between encouraging without undue pressure.

MS. LONGWELL: I don't know if you can. We might be considered a vulnerable population in terms of our interactions with FDA so, therefore, you would have to sort of have a consent form or something like that.

But aside from that, I think that FDA can, in general, applaud people who are cooperating, if they are cooperating. I do think, and the whole group thought, that there was just no way that, even if there were regulatory authority to force cooperation that it should be done.

However, I think that you can exercise--what can I call it--moral persuasion at times when it looks like this might be a good idea. I think FDA has done that in the past and it hasn't gotten them into major trouble.

If anyone else wishes to comment, go ahead.

MR. BARNETT: Okay. Yes?

MR. FOX: I have a question, if that is appropriate. Assuming that cooperation is a positive goal, and I raise that question as to whether it is, but just assuming that that is in the calculus the better outcome and you want to encourage the device company and the drug company to work together.

One of the situations we have encountered is with a drug that is multisource, that has gone generic, so there are multiple makers of the drug and they all have the same labeling. If, from the device side, we approach the drug company and say, would you like to work on this program, cooperate in a clinical study and supplement your NDA or ANDA with new labeling, they will say, well, if we get new labeling approved and we have done a clinical study, we will get three years of exclusivity. But our drug is still going to be A-rated in the Orange Book and our competitors will all be substituted readily for our products.

We basically have no way of protecting our investment. I want to know, sort of following up on Nancy's slides, I don't think the user-fee waiver is probably that great an inducement for some of the reasons that others spoke about to get drug companies to cooperate.

But I think one of the disincentives to cooperating is the inability of the drug company who invests in the program to protect its labeling because every dollar it spends in the development program is a dollar in the pocket of the A-rated competitors of generics, particularly if the device company goes and approaches the pioneer.

The pioneer well knows it is not worth investing in the new indication because the Orange Book, essentially the way it is run right now, essentially negates the three-year exclusivity.

So one thought is whether you could develop, or the agency would develop, indication-specific listings in the Orange Book or maybe just in this limited instance. But I would be interested, to the extent exclusivity is an inducement to getting people to cooperate, what are some ways that we could make the existing exclusivity, three-year exclusivity, more meaningful, taking my representation that it is not meaningful today under the current version of the Orange Book.

MR. BARNETT: Who wants to respond to that very good question?

MS. GLEASON: Can I ask David further questions to understand the issues. In that context, what is Company A, the drug company, doing? Are they simply supplying the drug or are they providing meaningful R&D, financial assistance and--because it is ordinarily the case that the device company comes along and says, I am going to do it lock, stock and barrel. All I really need for you to do is to supply us the drug and to have an uninterrupted supply in that context, and, frankly, I don't even need a right of reference because you are generic.

So I don't know what they are really doing that they are so worried about exclusivity for. The licensing agreement or the contractual arrangement would be parsed so that it would be of admitted limited value to them.

In that context, why, necessarily, would they feel that the exclusivity option would be their remuneration, if you see my point?

MR. FOX: There just seem to be too many reasons that have been put on the table why the drug company doesn't come to the table. Some are good. Some are bad. Some we would consider maybe reprehensible. But there are reasons why they won't come. It is just time, resources, interest.

So, if you are looking for ways to bring them to the table, one way is if they jointly invest in a clinical study for this new use, the new condition of use, and add it to their labeling, a new indication and add it to their labeling, the reward is supposed to be that you get three years of data exclusivity. That data is protected and other drugs can't be labeled for that use.

But those other drugs will be given an A rating that is not indication-specific and, under most state laws, they will be substituted if they are a lower-cost drug. So I think you ask really good details about some of the practicalities, but it is sort of an--maybe it is sort of a hypothetical question.

MS. GLEASON: I think you are asking the right questions, but I would only say, as a practical matter, is we seek to solve cross-labeling issues as soon as we open the fray to permutations that involve exclusivity, a whole bundle of other maneuverings and interest irrespective of cross labeling will enter into this picture, and we will never solve cross labeling because I think, once you begin to talk about new administrative options for exclusivities and redefining those rights, there are a much broader audience with a different set of interests that will come to bear on that.

MR. FOX: So we can just cast aside three-year exclusivity as an incentive?

MS. GLEASON: No; I think three-year exclusivities exist but I think the devil is in the detail and the question is how much is that company doing. That is very similar to what either Jane or Joanne said about grabbing at an exclusivity after the fact. Well, do they have access to the data? Is it really their exclusivity and how much, as a matter of public policy, do we really make nod to that if, in fact, the innovator Company B, the device company, is doing all the hard work.

MR. FOX: It is an attractive idea. We would invest in the study if only we knew we could protect our labeling from our generic competitors.

AUDIENCE PARTICIPANT: I agree with Mr. Fox. I think the whole discussion has been about trying to get these two companies in bed with each other. Quite frankly, if you don't offer some incentive for, I guess, Company A in your scenario, then I think you bring up an incentive that a lot of companies will jump at and they will not make an investment unless they can get some return on their investment.

The bottom line is, of course, it is going to open up a can of worms. But there is already a can of worms here as far as I can tell. So, in fact, if the goal is to attempt to get people to work together, you have got to provide an incentive that makes it worthwhile for them to work together. Otherwise, it is going to be business as usual.

And, by the way, I am not in the generic business anymore. But I have fought in those fields in the past. So I actually think that is a pretty interesting idea that really doesn't cost the regulatory side much. I guess it does cost you in headaches, but it is a pretty interesting idea, if the main goal is to try to get Company A and B together, you have got to provide some incentive for that.

MR. BARNETT: Jane?

MS. AXELRAD: I would like to challenge David Fox to play that out as to how, exactly, you would fix the problem since, let's say, you have an innovator. You have generics that are A-B rated to the innovator on the market. The innovator does go along and does a study and they get three years of exclusivity, how are you going to make that meaningful? What are you going to do, take the generics off the market?

I mean, they are already out there and they are already A-B rated. How are you going to make the exclusivity meaningful if you decided you really wanted to.

AUDIENCE PARTICIPANT: Remember, this is not a regulatory issue. It is a reimbursement issue. So, if the reimbursement specifies that the drug has to be labeled and, somehow, that labeling makes you different from all the other A-B-rated drugs, then that is where the value comes in. So it is really as much a reimbursement issue, at least in my simple mind, and it scares me to death to talk to a bunch of lawyers. I can tell you that.

But it really is more of an issue to try to funnel opportunity to the person taking the risk.

MR. FOX: Yes; I think that is one solution is an affirmative statement in the products that didn't invest in the study, not for use with, a negative labeling statement which would allow the company that was rewarded with exclusivity to take that to third-party reimbursers. It is an off-label prohibited use. If the agency prohibited the use for the protected indication, it would solve the problem. That is one way to do it.

MS. AXELRAD: Without any science basis for that.

MR. FOX: A legal basis.

MS. AXELRAD: Okay, but it wouldn't be based on science.

MR. FOX: But it is legally accurate. It gives meaning to the three-year exclusivity to point out those people who didn't invest don't have a right to the data, and it shouldn't be used for that use.

MS. AXELRAD: I guess I have the same kind of comment that Kathryn was sort of getting at in that--and it is the same issue, I think, with the user-fee incentive--is that you want to encourage innovation. You want to encourage these combination products. But there are competing public-policy considerations that you don't want to lose track of.

If you start narrowing and changing all of the incentives with regard to exclusivity to encourage this one class of products, you may have all kinds of unforeseen implications and these wouldn't even be unforeseen in that if you are talking about reimbursement and CMS would be basically paying a lot more for drugs and device combinations.

Just to do this kind of encouragement, you really have to weight the costs and benefits of that. I just wanted to make the point about user fees. People say, okay, fine, just give the people a waiver under user fees and to provide the incentive.

But user fees, under the Prescription Drug User Fee Act, at least, is a zero-sum game. We are required to collect a certain amount of revenues and, to get to that revenue target, we divide by the number of fee payers. So the more people you give waivers to, we don't get any less money or anything.

Basically, you just redistribute who pays. So what you are saying is everybody else in the industry should bear the cost of paying for the benefits and incentives that you are providing to some small number of firms. So, again, there are sort of unforeseen public-policy consequences, or foreseeable public-health consequences, that you have to consider before you make any of these major changes.

MR. FOX: Sure. I don't mean to put you on the spot. It is way too complicated an issue to try to address here but I think it is an interesting--as we go through the catalogue of incentives, of ways to bring people together, that is just one that comes to mind.

MR. BARNETT: A comment from the floor.

AUDIENCE PARTICIPANT: I would like to say I think that, since the product that you are after is cooperation between the two companies, the commercial implications of that can be worked out within the context of the business arrangement the two companies have. But I think it is workable. It is certainly attractive outside of the generic side of the business.

But I would say, be careful what you wish for. If you give us three years of exclusivity for every indication we can figure out how to make a drug-device combination, your workload will increase.

MR. BARNETT: Anyone else? I wonder if the FDA folks on the panel would want to comment on a very simple question. Many Company Bs out there are waiting for the FDA to decide on something. What is your advice for them as they wait? What action? What strategy should they be using? Does that make sense? Nancy?

MS. STADE: Are you saying that they are waiting for us to develop a policy here?

MR. BARNETT: That is why we are here. We are here to develop a policy, but it isn't developed yet. And they want to move ahead. So what do you want them to do now?

MS. STADE: If you have a particular product you are trying to develop, one thing is to try to come in and talk to us. There have been instances where we have been able to go ahead and do an approval or at least to formulate a vehicle for approval, even if it hasn't happened yet, without the cooperation. So don't foreclose that possibility.

The other is, obviously, you should be seeking cooperation on your own before you come in because that is the preferred result and you don't know until you ask; right?

MR. BARNETT: Anyone else?

MS. GLEASON: May I just give a permutation to that question. As the policy unfolds, I would hope that those case-by-case decisions that have been made through negotiations with OCP and the various centers remain intact and are not altered by any new cross-labeling policy that gets articulated. So there should be attention to grandfather protection.

MR. BARNETT: Okay. Are you all running out of steam? Maybe we are, as well. I think maybe it is time to call the Open Discussion to a halt and to ask Mark Kramer to come up and give some closing comments to this meeting.

 Closing Remarks

MR. KRAMER: Thanks, everybody. At first I thought I had one of easiest jobs here, just to close the workshop. If you were expecting me to try to synthesize all of what we have heard today in five succinct minutes, I don't think I am going to be able to do that.

It has been a long day so I will be brief. I did want to start with a thanks. I would like to thank Mark Barnett and Mack Lumpkin and all the speakers and panelists and organizations who sponsored them. We had a lot of brain power, I think you will agree, put to this issue today.

It is going to go a long way in helping us make some progress on this issue. I would like to give a special thanks for Suzanne O'Shea who has done just an outstanding job not only developing and implementing this workshop but on all the work that has been going on for more than a year now in terms of developing policy in this area.

I would also like to thank all the conference organizers including DIA and Patty Sauer from the Office of Combination Products for making today go so smoothly and, finally, to all of the members of the audience who were not shy about expressing their views, your views, about what we should do about this area.

I want to make sure that everybody is reminded about the open docket that we have got on this topic. I think this is a very complex area and it is probably one that is as difficult as any to try to do orally. I think sometimes written comments on a topic like this might really be the best vehicle for you to really go in to the subtleties and complexities that we need to in the area.

That docket number again; it is 2005N-0098. That period for comments will be open until July 8.

Where do we go from here? I think, as you have heard today, what might appear to be a relatively simply determination that needs to be made about whether the labeling of an approved product needs to be changed to reflect its use with a new product can raise some fairly substantive public-health and legal ramifications.

The cross-labeling issue has been hanging around for a long time. Suzanne has got her stack of paper. I think it has grown after today. I am sure the transcript alone will add several inches to that. But if it was that easy, I think we would have fixed it quite a while ago.

But I think today's workshop is a really important milestone in helping us get to fixing this problem. It is a reality check for us. We have worried about a lot of issues for a long time and it gives us a chance to get input, both today and then in the docket that will follow, into which issues really are most important and what ways there might be to get the square peg in a round hole dealt with.

We are going to consider all the comments that we have heard today. We are going to pore over the transcript to make sure we heard what we think we heard. I think this is going to require several re-readings to really go through everything that was just discussed today besides the written docket.

Then we are going to evaluate our options and, ultimately, share our thinking in some form for public review and comment. I can't realistically give you a time-frame today on when you can expect to hear the next bit from us on cross labeling. But I really will give you a commitment on behalf of the Office of Combination Products to work through these issues.

I think it is a really important class of products that this issue confronts and our goal is to work toward a policy that we hope you will find will be common sense, that will promote public health by providing regulatory pathways for innovative new medical products, one that protects the public health by ensuring the appropriate premarket review and postmarket regulation and one that is also legally sound and protects companies' proprietary interests.

I think a very good question to ask is can we get there. I think, with your help, I am pretty confident that we can. So we look forward to continued input on this issue. This is just one step in the way.

At that, I would like to close the meeting and thank you very much for joining us today.

(Whereupon, at 4:41 p.m., the meeting was adjourned.)