UNITED STATES

P-R-O-C-E-E-D-I-N-G-S

1:10 p.m.

EXECUTIVE SECRETARY FREAS: I would like to thank everybody for joining us this afternoon, for this our 19th and most technically advanced Allergenic Product Advisory Committee Meeting.

We are broadcasting from Room 4NN15 in Building 29 on the NIH campus. In Conference A on the first floor of this building there is a speaker phone reserved for the public. The public is welcome to participate in the entire portion of the meeting, however they must participate through the speak phone and the video machine down in Conference Room A.

Those people who have been on the line for the last 30 minutes realize why this is necessary. Those people in the public cannot understand it.

Today's entire meeting, of course, is open to the public, as advertised in the Federal Register.

At this time I would like to go around and check with each of our video stations first, and then I'll check with the audio stations to make sure that we can both see the visual and hear your audio.

So I'll go around and introduce you. I will start with our Chairman, Dr. Samuel Lehrer, research professor of medicine, Tulane University Medical Center.

Dr. Lehrer, could you say good afternoon and count to ten slowly.

CHAIRMAN LEHRER: Good afternoon.

EXECUTIVE SECRETARY FREAS: That's good. You don't have to count ten.

Actually, we did not pick up a video. I'm sorry, could you talk to ten?

CHAIRMAN LEHRER: So you're picking me up on video now?

EXECUTIVE SECRETARY FREAS: Your audio is loud and clear, but we do not have a video. We're checking on it right now. And we'll go on to the next committee member, with your permission.

CHAIRMAN LEHRER: Certainly.

EXECUTIVE SECRETARY FREAS: And see if we can pick Dr. Burks. Dr. Burks is professor of pediatrics, University of Arkansas for Medical Sciences

Dr. Burks, can you count to ten for us?

DR. BURKS: Wesley Burks, and I can see you now but I couldn't see Sam when he came on earlier.

EXECUTIVE SECRETARY FREAS: Okay. We appreciate you being there.

And, again, you know hopefully everybody has the audio number. If you give up on the video, we do have this audio number. But, hopefully, we're going to get this video to work but we're still trying to get the bugs out of the system.

Dr. Rebecca Gruchalla, you were the first one who joined us this morning. Thank you very much.

She's associate professor of internal medicine, University of Texas, Southwestern Medical Center.

Can you say good afternoon, Dr. Gruchalla.

DR. GRUCHALLA: Good afternoon, Dr. Gruchalla. Sorry. Well, I just had to do that.

EXECUTIVE SECRETARY FREAS: Oh, I appreciate a little levity. Okay.

PARTICIPANT: Like good say good night, please.

EXECUTIVE SECRETARY FREAS: Dr. Melvin Burger, are you there?

DR. BURGER: I'm by telephone, but apparently I'm not hooked up by video at all.

EXECUTIVE SECRETARY FREAS: Okay. And, of course, Dr. Burger is professor of pediatrics and pathology, Case Western Reserve School of Medicine.

Dr. Burger, I'm sorry you weren't able to join us by video, but we greatly appreciate you joining us by teleconference.

Dr. Susan MacDonald. I saw you earlier. She's associate professor of medicine, Johns Hopkins University School of Medicine.

Dr. MacDonald, can you speak to us?

DR. MacDONALD: Good afternoon. I can't see you at all, but I can hear you fine and you can see me.

EXECUTIVE SECRETARY FREAS: Yes. We're very fortunate. We hear you loud and clear, and see you, and everything looks good. Thank you.

On the audio portion only, is there -- Dr. Harold Nelson, have you joined us?

DR. NELSON: Yes. Good morning.

EXECUTIVE SECRETARY FREAS: Good morning.

Dr. Nelson is senior staff physician, Department of Medicine, National Jewish Medical Center.

Also on the line is Dr. Maria Soto-Aguilar, physician in private practice in Hudson, Florida, specializing in allergy, rheumatology and immunology.

Dr. Soto-Aguilar, can you hear us?

DR. SOTO-AGUILAR: Yes. Good morning to everybody.

EXECUTIVE SECRETARY FREAS: Thank you.

Next on the telephone our industry representative, Peter Hauck, Executive Director for Scientific Affairs Allergen Products and Manufacturers Association and works for ALK-Abello. Mr. Hauck?

MR. HAUCK: And good afternoon everybody.

EXECUTIVE SECRETARY FREAS: Good afternoon. Thank you for joining us.

EXECUTIVE SECRETARY FREAS: And next I would like to welcome a brand new member to this committee, Dr. Lynelle Granady. She's an associate physician with ENT and Allergy Associates in New York, New York. Welcome, Dr. Granady.

DR. GRANADY: Good afternoon. Thank you.

EXECUTIVE SECRETARY FREAS: I'm sorry to announce that our consumer representative, unless I'm mistaken has not joined us, and that's Delores Libera. She called me this morning. She has a high fever and was not feeling well and sends her disappointment that she could not join us.

Again, hopefully you all have this telephone number. If you get dropped from the video portion of the teleconference, you can dial in with the -- excuse me. The video portion you can dial in on the teleconference.

Now, I would like to just quickly go around and introduce the members in the room, starting over here with Dr. Karen Midthun. Dr. Midthun, if you would raise your hand once the camera gets to you.

Dr. Midthun is the new Director of the Office of Vaccines, Research and Review. Thank you very much.

Sitting next to her is Dr. Jay Slate, Chief of Laboratory of Immunology, Immuniobiochemistry.

Next is Dr. Richard Walker, Director, Division of Allergenic Products and Parasitology.

Next is Dr. William Egan, Deputy Director, Office of Vaccines, Research and Review.

In the corner over here is our technical chief.

I will start around here. WE have joining us Dr. Ronald Rabin, who is Senior Fellow in the Laboratory of Immunochemistry.

Next we have Ms. Jennifer Beidgewater, Consumer Safety Officer.

And we also have Jane Brown, who is the Committee Management Specialist responsible for getting this putting this meeting for us.

In the other corner with Dr. Robert Stumphey, our technical expert who is helping us with all the audio and the sounds.

I have been asked to remind you that this video conference is voice activated. And so please try not to all speak at the same time. And if you just wait a second, it'll be your turn to talk. And whoever speaks, the cameras will zoom in on them.

Now I'm going to read the conflict of interest statement which is required to be read for all advisory committee meetings.

The following announcement addresses the conflict of interest issues associated with this meeting of the Allergic Products Advisory Committee on April 8, 2003. To determine if any conflict of interest existed, the agency reviews the submitted data and agenda. The Committee agenda addresses update issues only. The Committee members were screened for their financial involvements with firms that could be mentioned in this updates.

We would like to note for the record that Mr. Peter Hauck is participating in this meeting as a non-voting industry representative acting on behalf of regulated industry. Mr. Hauck's appointment is not subject to 18 USC Code 208. He is employed ALK-Abello, Incorporated and thus has a financial interest in his employer. Mr. Hauck also serves as Executive Director for Scientific Affairs Allergen Products and Manufacturers Association. In the interest of fairness, FDA disclosure that his employer is a manufacturer of allergen extracts.

In the event the discussions involved specific products or firms not on the agenda and for which members have a financial interest, the members are reminded that they need to exclude themselves from such discussions. Their exclusions will be noted for the public record.

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

Dr. Lehrer, that takes care of the administrative introduction to the meeting. I would like to turn the microphone over to you.

CHAIRMAN LEHRER: Thank you very much, Bill.

I'd like to welcome everyone to the Advisory Committee broadcast.

Earlier this year I spoke to Dr. Slater about the format of this meeting. And since there were no regulatory issues that required the Committee's formal advice, we thought it best to hold a video conference to keep the Committee up to date on FDA's research activities and related allergen standardization issues. And certainly if you'll remember from the last meeting, allergen standardization was of importance to the Committee.

So Dr. Slater will be making up the update presentation.

I'd like to encourage of the Committee members to make comments on these presentations. We will have time for questions at the end of each presentations. And a time following all of the presentations for our comments.

And it's important anytime you make a comment, please identify yourself.

Another housekeeping issue, we would like to --

DR. MacDONALD: Excuse me. I can't hear anything either.

(Off the record while adjusting audio equipment)

EXECUTIVE SECRETARY FREAS: Dr. Lehrer, why don't you just talk for one second. And if you don't mind, I would just like to recap that you welcomed the Committee members and possibly maybe we should turn the host microphone over to Dr. Slater.

CHAIRMAN LEHRER: Yes.

EXECUTIVE SECRETARY FREAS: Would that be acceptable? Here I can only apologize to you for the technical difficulties that we are encountering. And if there's some way we could check and see what the matter is to get your audio back, we will. And also, if you would check in your local station to see if there's a problem down there that can be correct. And if not, I guess I will try to relay what you say to the rest of the Committee members. If that's agreeable with everybody?

Okay. So, at this time, Dr. Lehrer, I'm going to turn the microphone over to Jay Slater, who will be giving us the initial presentation.

Again, Dr. Slater is the Chief of the Laboratory of Immunochemistry.

DR. SLATER: Thank you very much.

First of all, again, I want to thank all of the Committee members for putting up with all the technical issues here. And I want to thank everybody for their patience in participating in the meeting in this manner.

We do have some interesting things to talk about today. But before we do, I just wanted to show this. This is a copy of today's newspaper. This in case, you now, given the unusual times, there was any doubt in anyone's mind that this is in fact a live broadcast. This is not being videotaped at an undisclosed location. We are all here and we are ready to go on.

DR. NELSON: For those of us not on the video, can you just tell us what it is?

DR. MacDONALD: What paper is it?

DR. SLATER: It's April 8, 2003. That should be enough, right. Okay.

So let's go ahead and proceed.

I'm going to start showing the slides. Now, we do have people that are viewing this conference in several different technological formats. Those of you who are viewing it on the monitors that's being generated by us don't have to listen to anything that I'm going to say, because I'll be controlling the slides.

Those of you that are looking at it on your own PCs, please set telepoint to slide show because there are some situations in which it will appear differently in slide show than it will if you're just looking at it in the normal view. So if you're looking at it on the PC, hit F5 now so that you'll be at the beginning of the presentation.

If you're looking at it in hard copy, just bear with me. When I say next slide you can advance visually to the next slide. If I say click, stay on the same slide. It just means that there's an animation that you're not going to see. There aren't too many of those, so bear with me.

We're now on the second slide of the presentation in which I'm just giving the outline of what we're actually going to do today.

First we'll have a very brief lab overview. Then we'll talk about a couple of interesting operational issues that have come up. And finally in the longest portion of the presentation, I'll be giving you a research and regulation update covering some of our recent endotoxin studies and covering some of our work in our attempts to go forward with cockroach allergen standardization.

Next slide, please.

So for the lab overview, we're going to talk about staffing issues, lot release issues and reference replacement issues.

Next slide. You should be on a slide that shows that I'm the lab chief. The two principal investigators in this laboratory are myself and Dr. Ron Rabin, who is starting his third year as a Senior Staff Fellow in the laboratory.

We have two post-doctoral fellows, Jonny Finlay and Bo Chi. Jonny is also starting his third year and Bo just started with us, since she's working with Dr. Rabin, and he started earlier this year.

Next slide, please.

We have research technicians in the laboratory. Al Gam, Mona Febus, Marc Alston, Cherry Valerio and Katia Dobrovlskaia. At this point if we were all the same room, I would have them all stand up and bow. But you'll understand that that's not possible in the format today.

Next slide, please.

This is a graphic that shows the staffing. This is the biologist staffing in the laboratory. And each of the colored arrows at the bottom covers a 12 month period. So what you can see from this is that after sort of ups and downs of lab building, really until the last year and a half we've been at a fairly stable position with 5 biologists performing both our research and regulatory work. This is a good number for our purposes and we're very happy that we've been stable at this point.

Some of you may have noticed that we have had some shifts in our staffing, though, that aren't reflected in this. Dr. Melissa has moved to Cedar and the past year and Melissa Patterson, who is a biologist in the lab for several years, left the lab to become a full time mother.

We maintained our full compliment by moving Katia Dobrovolskaia from a predominately research position into the visiting associate position, in which she can do both research and regulatory work.

Next slide.

Our "routine" regulatory activities, routine is in quotes because it often is anything but routine, include lot release activities, reference distribution and reference maintenance activities.

Next slide.

In the past year we have reviewed 357 protocols. One of those protocols was withdrawn by the manufacturer. We have distributed nearly 2,000 vials of reference materials in over 100 shipments sent to the manufacturers. To put that in context, please switch to the next slide.

This table shows the protocols that have been submitted in the past four years. And you can see that this year is somewhat lower than previous years, but my guess is this is just a normal year-to-year statistical variation.

Next slide.

You're see the referenced distribution in numbers of vials over the past four years. And you can see that that has varied as well with the peak in 2000. But basically our level of activity in 2002 is roughly where we have been for the last several years since I came here.

That ends the brief introduction to the overview of the lab's activities. I'll be happy to entertain any questions on that short section of the presentation.

If there are no questions, then we're going to proceed to the next portion of the presentation, which is operational issues in the Laboratory of Immuniobiochemistry.

Go to the next slide. You should be now one slide beyond the title slide operational issues.

What we're going to cover in this is a presentation of some issues that came up with the replacement of our cat and ragweed antisera. And we're also going to talk about our transition to ISO compliance.

Next slide, please.

We use sheep antiserum to measure, M being one content of ragweed allergen extract and one content of cat allergen extract. The antisera are raised in sheep and are sent out to our manufacturers for them and for us to use in the measurement of these allergen for lab release.

And this past year it became clear that both of our antisera needed to be replaced. As to a cat serum, it was originally released in 1988. S6 ragweed serum was released in 20000. And we replacement programs for both of these in the spring of 2002.

Next slide, please.

When we embarked on this, we thought this was a fairly standard procedure that was going to go uneventfully, but as you'll see we had some difficulties with it.

Our plan of what is after a prebleeding of the sheep that we were going to immunize them with 25 micrograms of allergen. Then a month later we boosted them with another 25 micrograms. We did a test bleed 3 weeks later. Following the results of that bleed, we determined that we needed to give another boost. And then the plan was to go ahead and plasmapheresis them in September of last year.

The next slide gives you a graphic representation of that plan with the three immunization doses followed by a plasmapheresis that we thought would give us a good supply of useable plasma sometime in mid-October.

Switch to the next slide, please. Next slide, please.

What we learned, though, in August was that the farm where the sheep were kept notified us that another sheep, not one of our two sheep, that had arrived in the farm in mid-July developed neurological symptoms at the end of July and it was sacrificed in the first week in August. But the presumptive diagnoses was that sheep was scapies. If that wasn't bad enough, next slide, please.

We were then notified about a week later that a sheep that had arrived in the farm in late June also developed neurologic symptoms, was sacrificed with a presumed diagnoses of scrapies. On autopsy both of these animals were found to have that disease.

Next slide, please.

So now we're going to review the transmissible spongiform encephalopathies, a topic for those of you are on this Committee 2 years thought and hoped I'm sure you'd never hear from again. But let's just review it very briefly.

There are several animal forms of the TSEs. Scrapie occurs in sheep and goats. Chronic wasting disease appears in mule, deer and elk here in North America. There's a transmissible mink encephalopathy. And, of course, there's the bovine spongiform encephalopathy. The feline is lightly considered to be derivative from the bovine form, contracted by cats who eat contaminated bovine material.

Next slide, please.

Of course, there are several human forms of the transmissible encephalopathy, and the one that's of greatest concern to us is, of course, Jakob disease, especially in the new variant that's associated with bovine TSE.

Next slide, please.

Scrapie is really one of the best described of the TSEs. It has been described in Europe for over 250 years. And it first appeared in the United States in 1947 when a sheep fogger imported sheep from Europe for local use. It now infects at least 1,000 flocks in the United States and is considered to be endemic.

It is transmitted both vertically and horizontally. Horizontal transmission is presumed to be by contamination of animals with placenta and blood during lambing season.

As with the other TSEs, there's typically a long incubation period and there's been no evidence of human transmission at anytime.

Next slide, please.

Sheeps are of varying susceptibility to scrapie associated with polymorphisms in their prions, especially at codon 171. The form of the animals that are most susceptible are QQ or glutamine/glutamine at the codon 171 site. The animals that are arginine/arginine at that site are highly resistent. And the animals that are QR are the intermediate resistance, although they are pretty resistent, they're not halfway between susceptible and the resistent animal.

Next slide, please.

The USDA has a well established scrapie eradication program that consists of three essential elements. The first is preclinical testing and surveillance. This includes a live animal test. It turns out that a third eye biopsy can reveal scrapie before neurologic symptoms appear. However, even the third lid biopsy may take months to years to turn positive in an affected animal.

The second key element is the tracking of infected and exposed animals. And the third key elements is cleanup strategies which involves identifying and genotype exposed animals. When there's an animal in a flock that develops scrapie, all of the remaining animals are supposedly genotype. The QQ animals are destroyed immediately. The RR and QR animals that are exposed are tracked for the development of the disease, but they may be slaughtered for human consumption.

Next slide, please.

Scrapie is not believed to pose any risk to humans. There's been no recognized human transmission in three centuries of exposure in Europe. And our sense is that these serum that we obtain from our animal, which you'll recall are only in the same farm with the affected animals, our feeling is that these sera was probably safe. Again, because there's no documented transmission to humans. The contact of our sheep with the affected sheep was limited. They were not in adjoining pens at anytime. They were at least 30 feet apart. And finally, because normal BSL 2 precautions should be in place for all work with all animal sera. And this should be certainly more than enough to give an extra element of safety. However, it was also our sense that in order to maintain a serum reagent that is as safe as possible, we were going to have a somewhat different approach.

Next slide, please.

So what we began to do, is we began to immunize two new sheep. We went ahead and processed the plasma from the two exposed sheep, and we initiated an immediate process of conserving our current and existing stocks of both of those previous antisera.

Next slide. Our approach is that if the new sera are available before we run out of the old sera, then the sera from the exposed sheet will be safe frozen for possible future use. However, if we have a shortfall, then sera from the exposed sheep will be used until the new sera are available.

Next slide, please.

The advantages of this approach is that we're offering the highest degree possible of safety to the workers that are using the sera for diagnostic purposes. And in addition, we're collecting what we think are going to be vast supplies of these sera for the future that we certainly will be able to use. The disadvantage was the possibility of two serum switches in a relatively short time, and obviously the time and expense.

That ends that segment of the presentation. Let me just tell you that we actually are going to be able to go forward with the new ragweed serum fairly soon. We've tested that and we will be able to go forward with that.

With the cat serum, we've continued to have some technical problems with even the new cat serum that we're using. And what we're in the process of doing in order not to have a shortfall is we're actually going back and retrieving some of the original sera that -- the components of the original sera that we had used in previous years. We're testing that out and we're going to -- we think we'll be able to have about a year's supply of that that will give us some time to work out some technical issues that we're having within the serum. So we're hoping that we will be able to avoid -- we certainly will be able to avoid shortfalls with both of those new sera.

That ends that portion. If anyone has any questions, I'd be happy to answer.

MS. LEE: Jay, this is Dr. Lehrer. Could I ask a question about your immunization?

DR. SLATER: Sure.

CHAIRMAN LEHRER: You used advance -- 25 microgram doses and I think there were 3 doses and complete/incomplete. It's just 25 micrograms struck me as being somewhat as a low dose for a kilogram basis with such a size animal. But I haven't had a whole lot of experience with sheep, so I just wanted to ask you about that in terms of your choice of antigen.

DR. SLATER: It's a good question. We actually got very, very good titers with the exposed sheep and with the subsequent sheep. So I think you're asking a good question, but we did end up getting very good titers.These were some previous protocols that were used in the laboratory.

CHAIRMAN LEHRER: I see. Thank you.

MR. HAUCK: A question, Jay.

DR. SLATER: Yes.

MR. HAUCK: This is Peter Hauck.

Maybe two questions. So you anticipate there would be no disruption of protocol release or manufacturers being able to use these materials for lot release purpose?

DR. SLATER: No. We're not anticipating any delay at this point. Both the new ragweed serum and the retrieved old cat serum are in the process of being lyophilized over the next couple of weeks. We'll then do some brief testing and we should be able to send it out shortly.

MR. HAUCK: Okay. And, Jay, one last question. It doesn't sound like you'll need to, but at least two manufacturers have their own antisera for ragweed that they use for stability testing. In a pinch, could they possibly be used for lot release it came down to that? That's kind of speculation, but --

DR. SLATER: Yes, that's speculation I'd rather not get into to. I think we'd have to know more about those sera and we'd have to really be in a pinch. I think we've avoided that.

MR. HAUCK: Okay.

DR. BURGER: Jay, this is Mel Burger. Maybe I'm missing something, but why not go over to Mil's monochronials?

DR. SLATER: You're not missing anything. That would involve designing a new assay and then validating it. It's certainly something that can be done, and it's certainly something that you sort of think about a lot when you go through this process of raising a new polyclonial serum.

You know there are -- we would have to make sure that it performed comparably. When you make that kind of a big switch, you have to make sure that we're not going to have a shift in what the unitage actually ends up meaning. So it's something that certainly can be done, and I'm always looking for new and better ways to allergens. But you have to understand, it's not a process that one can use in two to three months to make up for a shortfall.

DR. BURGER: No, I wasn't advocating it as a shortfall. To make up for the shortfall, I was wondering why -- since we're beginning to focus more and more on the single most important allergen, why we don't have a long term program of shifting over to monchronial antibodies for standardization?

DR. SLATER: Well, it's certainly something that we could consider. I think it's a good suggestion. But we're going to have to make sure that it actually performs the same way as our existing assays so that the actual unitage doesn't really change.

DR. BURGER: You know, maybe that should be an experimental goal of the laboratory to develop that for one antigen and see how it works.

DR. SLATER: I appreciate the suggestion. I think that's worth thinking about.

Okay. Dr. Lehrer, if there are no further questions, with your permission I'm going to go on to the next discussion.

CHAIRMAN LEHRER: Yes. That's fine, Jay.

DR. SLATER: Thank you.

Let's go to the next slide.

I'd like to talk briefly about CBER's Laboratory quality management initiative. And I'd like to acknowledge that in the audience down in Conference A is Bill McCormick from OVRR. He is the office quality manager. And he has provided me with the information that I'm using to make this brief presentation.

CBER has committed to becoming ISO-170235 compliant in its office product testing. So for those of you who don't know what that means, let's define a couple of terms here.

ISO is the abbreviation for the International Organization for Standardization. This is a nongovernmental agency founded in 1947 and based in Geneva, Switzerland.

ISO-17025 is a document that was generated by ISO in 1999 entitled "General Requirements for the Competence of Testing and Calibration Laboratories." This is a set of guidelines for labs that do testing and calibration to show operation of a quality system, to assure technical competence and production of valid results.

Next slide.

CBER's laboratory quality management initiative has several components to it. The first is to establish a policy with a center level quality manual. The second to audit the laboratories within the center for compliance with ISO-17025. And finally, to obtain test and laboratory accreditation where appropriate. And with accreditation, of course, is the consequence of a successful third party audit.

Next slide, please.

Why is CBER becoming ISO compliant? This is a complicated and expensive process. I think that's a reasonable question as to why we're doing this. And there are several reasons.

The first is to establish recognized competence of our laboratories that do testing. To assure the test and value of our data and our processes.

The second is, we require manufacturers' labs to be i GMP compliance. This is good manufacturing practices compliant. And although that compliance is intended for manufacturers not for us, it certainly makes sense that we would try to be compliant with ISO, especially since there are international efforts of harmonization that are attempting to equate GMP and ISO requirements.

The third is that ISO is an internationally recognized standard. And, again, this is a way of assuring the value and trust of the results that we obtain.

Next slide.

We are attempting to become ISO compliant in order to implement laboratory quality management policies and practices for all the official testing activities, to document a high level of training, competence and proficiency and to establish a consistent product testing process.

Next slide.

But are the elements of ISO compliance? ISO compliance consists of the management of 4 different features of any laboratory testing system. People, equipment, documents and processes. Let's go through those one at a time now.

Next slide.

The management of people basically covers all the people that would be involved in the laboratory testing of these products. All the people in the lab need to have defined roles. They need to have training that's appropriate for their roles in the process. They need to be demonstrated to be proficient in what they're supposed to, and that's both demonstration of proficiency at the beginning, initial demonstration of proficiency, and ongoing on a regular basis.

And furthermore, there needs to be an authorization process that assures that people are doing what they're supposed to be doing, what they were trained to do and that their training is ongoing.

The next slide, please.

The management of equipment follows from the same kind of reasoning. All of the equipment that's used in the testing has to be appropriate for the testing. It has to be appropriately calibrated. It has to maintained. There has to be a process in place to assure that only calibrated and maintained equipment is used. And there has to be assurances of measurement traceability to accepted standards.

Next slide, please.

ISO also requires that documents be -- and those documents include policies, procedures, specifications, equipment manuals and certifications. These documents all have to be reviewed and issued and controlled and they have to be approved and issued before they are used.

Next slide.

And finally, processes have to be managed. Those processes include the approval of testing materials, environmental specifications and monitoring, handing of samples, validation and suitability, the handling of data including non-conforming data, corrective action, preventive action systems, internal audits, management review and finally, although we never have complaints, the recording and handling of complaints.

The next slide, please.

CBER has committed to the implementation of this lab quality system. CBER has established a quality board. We've developed a quality assurance structure within CBER with the appointment of a center level quality manager, with the hiring of office quality manager. Bill McCormick, as I mentioned before in our case. The appointment and hiring of division quality coordinators. With the preparation of a CBER wide quality manual and with the purchase of an integrated quality management computer software system.

So what will this mean to my laboratory? My laboratory actually has fairly extensive written protocols for almost everything that we do. So there's really not going to be extensive substantive protocol revisions for our processes. There will be changes in terms of the documentation. There will be formatting changes to make the formatting of our documents consistent across the center. And there will be some substantive changes as well.

This will probably require a more formal separation of research regulatory equipment that we've had in the past, simply because the regulatory equipment will have more extensive record keeping and maintenance than the research equipment, although certainly the research equipment would benefit from the same maintenance and calibration as the regulatory equipment.

There will certainly be a series of internal audits that we will subject ourselves to as part of this process.

And then finally, in the near future, external audits as well as we approach certification.

Next slide.

The time table for this, of course, is tentative. But the idea is for implementation in stages of this process over the next 3 years with CBER -- with all labs seeking accreditation in the year 2005. The software is becoming operational this year. The polices and quality manual is being issued this year. We are beginning the initiation of compliance audits. These are internal audits which, of course, is starting and will be ongoing. And the training and process development is ongoing as well.

That ends this portion of the presentation. Dr. Lehrer, if there are any questions, I would be happy to take them now.

CHAIRMAN LEHRER: Are there any questions?

Well, it seems to me, Jay, that you've done an excellent job in spite of trying circumstances. And also I'm pleased to know that that is really you and not your double.

DR. SLATER: Thank you. I appreciate it.

DR. BURGER: How do we know that?

CHAIRMAN LEHRER: We don't know that for sure. But I think it's impossible to duplicate Jay, I believe. I mean that in the best of ways.

DR. SLATER: Well, thank you.

CHAIRMAN LEHRER: Next we're going to move the agenda to the open public hearing. Bill, could you --

DR. SLATER: Sam, we're not quite there yet. I think you're being overly optimistic. We have some more talking.

CHAIRMAN LEHRER: I'm sorry. I have my confused on there.

DR. SLATER: I apologize. We were going to be talking -- we're going to be doing the research and regulatory update.

CHAIRMAN LEHRER: Sorry about that.

DR. SLATER: No, I apologize. I wish I were faster, we could be done now, but --

CHAIRMAN LEHRER: Oh, that's fine. I think you've done an excellent job under trying circumstances.

So then the next presentation would be the research and regulatory?

DR. SLATER: Yes. Yes. Okay. Thank you.

So let's go to the next slide, please.

This is the research and regulatory update. And if you click to the next slide, you'll see the active research projects within the laboratory. Remember, there are two PIs in the laboratory, myself and Dr. Rabin. Those of you that were participating in the meeting last year heard about Dr. Rabin's project. We will not be talking about them at this meeting.

In terms of my projects, I'm going to focus on issues having to do with cockroach allergen standardization and endotoxin in allergen vaccines.

Next slide.

But before we get to there, I'm going to review some of our publications from this year.

We had three primary research publications out of the laboratory. The first and the third on your list are actually publications that you're going to be hearing about today. The first is our cockroach work, which appeared in Clinic Experimental Allergy last year. The last one is our endotoxin work that is going to appear in ACI within the next two to three months.

And the middle one is work that we did in collaboration with our colleagues in Australia having to do with the latex allergen Hev b 5. I'm not going to be covering that work today.

Next slide, please.

And on the next slide you see that the chapter in Meddleton's Allergy textbook still has not been published yet. But, hopefully, will be published sometime soon.

Dr. Rabin also authored a review article that is going to be appearing in the Encyclopedia of Hormones on CXCR receptors, which I neglected to put on this slide.

And next slide, please.

The following are the four abstracts that we submitted to the Academy of Allergy meeting this year, which cover the breadth of the work that we've been doing in the laboratory.

Next slide.

So before I start this presentation about the endotoxin and content of allergen vaccines, I'd actually like to thank the Committee for encouraging this line of investigation last year. I think you'll all recall that this was a topic of discussion last year and several Committee members strongly encouraged that we expand our then fairly small studies on endotoxin in allergen vaccines, and we did that. And we came up with some results that we think are interesting. And I want to thank the Committee for that extremely constructive input just about a year ago.

As you probably are aware, allergenic extracts are not required to undergo evaluation for the presence of pyrogens. And that is from CFR610.13 paragraph (b). However, in prior studies done actually out of this laboratory a confirmed variable endotoxin content in allergen vaccines. Now, those studies were not done with standardized vaccines. Largely they were done with house vaccines, but certainly there have been studies before that have shown that allergen extracts do have endotoxin content.

Next slide, please.

The products that are exempted in 21 CFR from pyrogen testing include blood products, horse serum, bacterial, rickettsial and viral vaccines, toxoids, toxins and allergenic extracts.

Next slide.

But we are interested and we were interested in looking at the endotoxin content of allergen vaccines. And we used sort of the standard workhorse method that's been used for the last 20 years in measuring endotoxin, and that's the LAL gel-clot method. The slide that you have before you has four pictures on it. On the left hand one is an imagine of the underside of the horseshoe crab, Limulus Polyphem, which the hemolift link, which is the basis for the LAL gel-clot method.

You see in the central panel a fisherman hauling in some horseshoe crabs.

In the upper right hand panel you see beautiful Delaware sunset with horseshoe crabs on the beach.

And in the lower right panel you see a worker bleeding horseshoe crabs for their hemolift to be used.

You should note that this is a survival procedure. The horseshoe crabs do survive the bleeding, although no one to my knowledge has asked how they feel about the insertion of foreign components or biomedical health.

On the next slide you see, if you look closely, a clot forming in the bottom of the tube that has been inverted. This is what you're looking for when you add endotoxin to the limulus emulase.

Now for those of you who are viewing the next slide on a hard copy, there would be -- there is a small animation that's not terribly interesting, but it's there and you won't see it.

Next slide.

Those of us that are watching it animated are watching a wonderful imagine of the cascade where endotoxin activates factor C, which in turn activates factor D, which in turn activates a pro-clotting enzyme. And finally generates the coagulant that forms the clot.

If that were the only mechanism by which this cascade could occur in the limulus licate we would be in much better shape. But, unfortunately -- and those of you that are doing this on your own PCs, please click now -- there's another route by which the clot can occur, and that is when (1,3) ß-D-glucans activate factor G. Activated factor G can also active the clotting enzyme and form the clot made out of coagulant.

Next slide, please.

So there are two possible ways that we can imagine that there might be interference with the LAL gel-clot determination of endotoxin content in allergen. And the first, of course, is that non-endotoxins, the glucan content may induce clotting by an alternative pathway. And also we were concerned since this is a proteolytic cascade that proteases especially in cat and mite extracts, might also induce clotting.

Next slide.

So our approach would determine the endotoxin content using the standard gel-clot method, but then independently to attempt to assess the contribution of non-endotoxin components.

Next slide.

We started with only standardized allergen vaccines which we obtained from all of the allergen manufacturers. We started out by just performing LAL gel-clot assay. And then we attempted to control for these confounding factors by selectively absorbing the allergen vaccines with ENP-silica resin, followed by a repeat LAL assay.

ENP is endotoxin neutralizing protein. It's a 12 kDDa, cationic, amphipathic protein that binds to and neutralizes the biological activity of LPS.

Theoretically that adsorption would help us control for both glucans and for the proteases, neither of which would be expected to bind to the ENP. In addition, as an added control for protease activity, we pretreated selected allergens at 95 degrees centigrade for 15 minutes, and that treatment would be expected to inactivate any proteases. Of course, that was followed by a repeat LAL assay.

Next slide, please.

On this first table you see 24 grass pollen extracts, so I'm just going to take you through this slowly since we have a few of these tables that are like this.

There are 24 grass pollen extracts that were assayed. The manufacturers are coded in the second column.

In the third column you see the total gel-clot activity, which is given in a twofold range. And this was before any treatment or adsorption.

In the fourth column you see the depleted gel-clot activity, this is after the ENP-silica adsorption.

And finally you see the corrected endotoxin content, which is actually the difference between the geometric means of the first ones, and if there was a range the geometric mean of the second one or if it was at or below zero, it's essentially the geometric mean of the third column.

So the numbers you really need to look at the last numbers, but I will direct your attention to the fact that there were really only a couple of extracts in which a significant amount of the gel-clot activity remained after ENP absorption. That was in ryegrass, the second ryegrass pollen extract for manufacturer C where you could see that perhaps half of the activity was not absorbed out by ENP. And in the last ryegrass extract where perhaps 10 percent was not absorbed out.

What you can also see by looking at the right hand, far right hand column is that the range was pretty variable. There were some that really had very little or no endotoxin, and there were a few that had certainly measurable and appreciable amounts.

Those of you that are working on your own PCs, if you would just click twice now you'll see arrows that will appear indicating which of these extracts are glycerinated, which one is in phenol. And it turns out that's that one that had significant amount of non-endotoxin LAL activity. And the GP stands for both glycerin and phenol.

The short answer is there really is no overall pattern in any of these tables associating the solvent with the amount of endotoxin we were able to measure.

Please go to the next slide.

This is a slide that shows 9 ragweed pollen extracts, two of them are ragweed mix and 7 of them are ragweed -- are short ragweed. Again, we covered a range of manufacturers. Again, with some of them there was some activity that was residual after depletion with ENP, but you can see here that for the most part the lion's share of the activity seemed to have been removed by ENP-silica indicating specific endotoxin. And, again, in the form -- you see that there is quite a bit of variability among these different products.

If you go ahead and click three times, you will once again see which ones are glycerinated, phenolated and which have glycerol and phenol. And, again, no obvious pattern.

The next slide shows our result with four cat pelt extracts and with seven cat hair extracts. And here you see a pattern that I think is interesting, and will comment on it again later.

Again, on a percentage basis, there's almost no contribution by any activity that doesn't get completely absorbed out by endotoxin neutralizing proteins. So, again, the lion share of this seems to be true endotoxin. But you're starting to see a difference in patterns between the cat pelt and the cat hair that's suggests that the pelt results are significantly higher than cat hair, and I'll show you those data represented graphically in just a couple of minutes.

Those of you on PCs would go ahead and click twice now, you'll see again which ones were in glycerol and which were in glycerol and phenol. The other two were extracts.

Next slide, please.

And this slide shows perhaps our most surprising results. And this is the result with 7 d.farinae extracts and 7 d.pteronyssinus extracts. Again, covering the range of manufacturers. Again, very little contribution from anything that appears to be other than endotoxin, true endotoxin. But a real dichotomy between the results of d.farinae and d.pteronyssinus.

If you'll go ahead and click twice, you will see that most of these were glycerinated and one in each category had glycerin and phenol in it. Again, no major differences could be observed.

Next slide.

This graph summarizes the -- basically the percentage of LAL gel productivity that was not adsorbed out by endotoxin neutralizing protein. The ordinate is actually not quite accurate. It says gel-clot activity from ß-D-glucans, but it really would be gel-clot activity from any ß0glucans or proteases because that would be revealed in this analysis.

And what you can basically see is that with the pelt, the hair and the dust mites, there really is an insignificant amount of what's not adsorbed out.

With the pollens, both the ragweed and the grass pollens are you saw in the tables, there were some in which there was some percentage that was not adsorbed by the endotoxin neutralizing protein. But in general we think the LAL gel-clot assay is a pretty reliable assay and it doesn't seem to be interfered with by these other things that we were concerned about.

Next slide.

Shows our heat inactivation study. This is actually not all of our heat inactivation studies. We've done more since. But you can basically see here that with the ryegrass pollen, short ragweed, a couple of cat pelt extracts and four dust mite extracts the results before and after heat inactivation are completely identical. And we've actually confirmed that with several other extracts that we've studied since then.

PARTICIPANT: Jay, have you worked with mold extracts at all?

DR. SLATER: No, that's a very good question. This study was limited to standardized extracts. I think you're raising a very good point. I think we might actually come up with very different answers if you looked at molds. But we haven't done that yet.

Let's go on to the next slide.

And the next slide, again for those of you -- no actually, the next slide graphically summarizes all of the corrected endotoxin data. Please note that there are six different categories at the bottom. The grass pollens, ragweed pollens, cat pelt, cat hair, d.farinae and d.pteronyssinus.

Please note as well that the numbers are on a log scale, so these differences are actually quite large. And if you'll go to the next slide, you see the same data but with the significance analysis on the top. This is based on a rank order sum test, which is the appropriate test because these data are not normally distributed.

And what you can see is that cat pelt and cat hair are significantly different with a p - 0.02. And it'll be no surprise having looked at the table with the dust mites, just how significant the difference is between D.farinae and D.pteronyssinus, the p value is .0003.

Next slide, please.

So, we have a series of conclusions and discussions.

The first conclusion is that we think that for the most part the observed LAL gel-clot activity probably represents real endotoxin content and not either the ß-glucan question or protease activity.

Next slide.

So the next obvious question is is this amount of endotoxin physiologically significant? If you took all of the 58 extracts that we studied and took them together into a single allergen immunotherapy mix, he mean endotoxin content of all of those products is 1900 endotoxin units per mL. Let's keep in mind that allergen immunotherapy doses can range anywhere from 0.5 to 2.0 mL per month. In some cases even more.

Keep in mind as well that depending on the actual distribution of the mix, that number could be much lower or it could be higher. Obviously, a mix that consisted only of pollens would not contain nearly this much endotoxin based on our serving. Likewise, if it contained D.pteronyssinus, it would contain much, much less.

So the question is how significant is this dose and what does it actually mean. And in order to find that out, you really have to go back and dig. Because there aren't that many studies in which endotoxin has been administered by injection to humans. Looking at the two papers that are indicated

at the bottom of this slide from Wolff in 1973, which is really a review of many years of previous work that Dr. Wolff had done and looking as well at a paper by Dr. Michie that appeared The New England Journal in 1988, you can estimate that about 40 to 50 EU per kilogram averaging to 2800 to 3500 EUs, when administered intravenously can elicit a rise in temperature, heart rate and white blood cell count.

Now, please note that this is when administered intravenously. There's every reason to believe that the effects of endotoxin injected subcutaneously would be much less. But there are no data on those responses to endotoxins administered subcutaneously that I was able to find.

Let's go to the next slide, please.

How are specific endotoxin limits set when they are set? Well, they are product specific and they're generally based on the actual drug dose, and on an overall limit of 5 endotoxin units per kilogram per dose.

Next slide please. This table just gives you some examples. This is not an exhaustive survey by any means. Vaccines can have a wide range of endotoxin content, mostly in the lower range of that range. But there certainly can get somewhat higher.

The remaining four are drugs, not vaccines. And we determined those numbers based on published USP limits at the estimated maximum therapeutic doses. For example, Penicilloyl polylysine has a USP limit of 5830 endotoxin units per mL, and a typical injected dose intradermally would be about .01 mLs.

Doxorubicin we also calculated in a very similar way. Fentanyl and insulin.

The purpose of this is really to give you an idea of what the range of endotoxin is in other products.

Next slide, please.

The fact is that the clinical consequences of endotoxin and allergen vaccines really have not been studied. There are no data that adverse events from immunotherapy are associated with endotoxin level. And furthermore, even though some people who have seen this data have raised this possibility, there is no data to support a beneficial effect, either, of endotoxins.

We do have some data that shows that the endotoxin content of allergen vaccines doesn't effect their diagnostic performance. There have been some skin testing studies that have indicated that it doesn't effect the size of the immediate skin test reaction. But in terms of immunotherapy there really are no data. And certainly we would think that future studies in allergen immunotherapy should always be controlled for endotoxin dose. This is something we now know is potentially something that could effect immunotherapeutic responses and should be monitored in future studies.

And finally, we think that, you know, the role of endotoxin and the safety and efficacy of immunotherapy should certainly be assessed in future studies.

Next slide.

What we conclude from these data in terms of descriptive results is it's clear that the pollen extracts have less endotoxin in them than the cat and mite extracts. It also appears that the cat hair extracts have less endotoxin than cat pelt. And it appears quite clearly that the D.pteronyssinus extracts have much less than the D.farinae.

We are in the middle of investigating why that is happening. There's several different possibilities. One is that there really is a difference in the bioburden between the two different kinds of bugs. Another is that there's an endogenous heat-stable ENP-binding LAL activator in D.farinae. That would be a first, but it's certainly a possibility.

The other is actually a little bit more intriguing. Endotoxin neutralizing protein is actually derived from the horseshoe crab. It's certainly possible that there's an endogenous endotoxin neutralizing protein in D-pteronyssinus that's confounding our measurements. But these are all things that are addressable in studies, and we're actually working with individuals who grow dust mites for a living and trying to work with them to find out what the cause of this difference is.

Next slide.

So our plan is to further expand our study of endotoxin content. We'll be doing additional standardized and nonstandardized extracts.

We're also going to look at different methods. I still think the LAL gel-clot method is a good method. I think we've shown that it works and that it's reliable in our hands. But there are other methods for measuring LPS that are quite sensitive and actually can give you a fingerprint of the kind of LPS in the product. In particular, we're going to be working with colleagues here at CBER on using gas chromatography, mass spec analysis. And that should be coming on board within the next couple of months.

In addition, we're going to be investigating further the differences between D.farinae and D.pteronyssinus.

Dr. Lehrer, that ends my discussion of endotoxins. I'm planning on going on next to talk about cockroach allergen standardization, but I'd be happy to take questions at this point.

CHAIRMAN LEHRER: Yes. I got a question. I think it's fascinating the differences in endotoxin content with D.farinae and D.pteronyssinus. I wondered if there's any differences in the way that these are grown or harvested or any processing procedure?

DR. SLATER: Actually, interestingly enough that was my first question. The media that they're grown on are absolutely identical. The temperature and humidity conditions are identical as well.

There is some difference in how long they need to be grown to get to saturation. But it's not that huge a difference. And I don't -- honestly, I'm sorry, I don't remember which one grows faster.

But I spoke to several people that are involved in growing dust mites and we were really unable to find any obvious growth condition type of situation that would explain this difference. It's a mystery.

The one thing I must tell you when I spoke to the entomologist, the only entomologist that I found whose an expert on dust mites, is he did assure me that these species really are quite different from each other. That we tend to think of them as just sort of a variance of the same thing, and they really are biologically very different.

So it wouldn't be terribly surprising, for instance, if one expressed a protein that the other didn't that somehow interferes with the testing.

It also might be possible that one of them actually has a bioburden that the other one doesn't, and we just don't know yet. We're working with the microbiology department over in the Clinical Center, who have been extremely helpful. And we're working with three different sources of these raw material. And I think we are going to figure out what's going on, but we haven't yet.

CHAIRMAN LEHRER: That might provide some interesting information just about the basis of the assay just because these differences are so dramatic.

Another question or comment about the assay that's used. I think you touched upon biological significance of some of these measurements, or tried to, these deal with that. But the -- I think -- I always bring this issue up, and you've touched upon it. In terms of the specificity of the assay yourself, and I know that you're trying to control for this in terms of the information that you presented. But there -- one could always say there may be other types of polysaccarides or -- that they cause reactions. And, indeed, the question concerning measurement of molds I felt was rather interesting. Because I wondered how active molds would be in such an assay.

And I know that it is accepted in the community. But I was wondering if there was any good studies that have actually used some of the old time endotoxin biological assays and showed a, you know, a good correlation? When you measure endotoxin in crude material it's not measuring pure endotoxin.

DR. SLATER: Are you talking about correlating this with a rabbit-pyrogen test for instance?

CHAIRMAN LEHRER: Yes, something like that. Some type of classically accepted endotoxin assay.

DR. SLATER: Well, you know, I must say I think now at this point the LAL gel-clot assay is probably considered the classically accepted assay. Certainly it's the USP standard now, although there are other tests that are newer and are considered to be, you know, standard as well. I don't think anybody uses the rabbit-pyrogen test as a standard anymore.

We actually very early in our study, we actually --

CHAIRMAN LEHRER: No. I didn't mean for you to use another assay. I just wondered if there was any correlations in the literature between these classical assays and this assay?

DR. SLATER: Yes. There are correlations in the literature between the LAL gel-clot assay and the rabbit-pyrogen test. I don't know whether there are correlations between the GS mass spec assay and the rabbit-pyrogen test. I'm not aware that I've ever seen that.

CHAIRMAN LEHRER: Okay. Are there any other questions from other members of the Committee?

DR. SOTO-AGUILAR: Yes. I'm Maria Soto-Aguilar. A question for Jay.

It seems like the D.pteronyssinus, which is maybe more protein genically than the D.farinae, would you would you speculate or would you suggest that the presence of endotoxin has something to do with it?

DR. SLATER: I'm sorry, I didn't catch that. That it has something to do with what?

DR. SOTO-AGUILAR: For potency between the matter D.pteronyssinus being more potent (technical difficulties) that D.farinae when you're doing an assay that before you may even give a little less dose of D.farinae versus D.pteronyssinus for somebody whose reacted to both the species. Would mean that maybe the presence of endotoxin is the one making one less effective than the other with the endotoxin in the extract vial may be effected (technical difficulties).

DR. SLATER: Well, you know, I think there are two different ways to ask the question. If you're asking is it effecting the potency of the allergens in skin testing, I think it's probably not. However, you know obviously one of our questions is whether endotoxin might be functioning as an averment, which certainly is not a thought that has originated with me. And you know, that's certainly a possibility. But, again, there really are absolutely no data, and believe me I've looked, to support either the argument that endotoxin is a problem in immunotherapy, that is that it's associated with either safety problems or decreased efficacy on the one hand, or on the other hand there's certainly no evidence that it actually potentates immunotherapy.

MR. HAUCK: A question and a comment. This is Peter Hauck.

Jay, when you just slightly talked about mean endotoxins content as in calculating the immunotherapy dose. You studied pretty much concentrates, 10,000 and 100,000 PAU, so did you assume that people would be getting between .5 and 2mLs of these concentrates per month? Because that seems a little high.

DR. SLATER: Yes, probably is a little high That probably was a bit of an over-estimate.

MR. HAUCK: Yes. Well, also with the cat and mite, it might -- you know, you could be approaching that dose, at least a .5 dose.

Secondly, I note some companies when they do their endotoxin testing on their water or stoppers, what have you, their components, they use a chromogenic assay for endotoxins. And I just wondered if -- I don't know the answer to this.

Number one, was it chromogenic because of the color of the extract be a problem, the test was that the employee didn't do it. And, two, does it suffer from the same maybe lack of specificity that the limulis test has?

DR. SLATER: Well, the chromogenic assay is also based on the same protein cascade, just generates a different endpoint, not a clot.

MR. HAUCK: Okay.

DR. SLATER: So, you know, I think --

MR. HAUCK: But it gives the same results pretty much?

DR. SLATER: You would think so. We didn't actually direct -- and I don't know about the issue of the chromogenic assay. I don't think it would be a problem in terms of the pigmentation of the product because you have to dilute these products quite a bit.

MR. HAUCK: Okay.

DR. SLATER: My guess is it would not be a problem. But we didn't have any experience with that.

MR. HAUCK: Okay. Thank you.

CHAIRMAN LEHRER: Are there any other comments, questions?

DR. MacDONALD: Jay, this is Susan MacDonald.

I enjoyed seeing your poster at the academy meeting. And I compliment you for taking this on.

I also seem to remember that there was a poster from a guy from West Virginia who had a mice model who gave -- made the mouse allergic and then gave endotoxin and showed that it switched to a quote/unquote TH1 type of -- kind of cytokin response. So maybe -- last year at this meeting I was thinking that endotoxin was -- after reading these toll receptors and we were causing problems, that perhaps we're actually -- we might be actually helping people if there are doses of endotoxin in these extracts.

DR. SLATER: Or at least we're helping mice.

DR. MacDONALD: Oh, mice. That's right. And being from Hopkins, we don't like mice.

DR. SLATER: It's a very good point, and I think it's a good theoretical question. Again, I think it's actually critical that it be looked at and certainly worldwide there are many immunotherapy studies that are going on. I think it's very important. And, hopefully, by disseminating this data and with the publication of studies like this, we'll indicate to people that they really have to be very careful and control the endotoxin content so we learn something.

DR. MacDONALD: I agree. Thank you.

CHAIRMAN LEHRER: Okay. I guess there are no more comments or questions, and we can continue.

Jay, with the cockroach allergen standardization.

DR. SLATER: Terrific. Thank you very much.

And, again, thank you for your comments regarding the endotoxin.

I've taken the liberty of saving the best for last. This is, I think, a very encouraging talk the next one. If you think endotoxin was good, this is better.

Next slide please.

We're going to talk now about cockroach allergen standardization. And this isn't the first time you've heard about standardization, those of you that have been on this committee over the past several years. But I actually think that we're sort of capitalizing on some of our previous work, capitalizing on some very important collaborations. And I think we're really going to make some good progress in a very important area.

I'm going to talk about -- there are a number of things that my lab is doing involving the standardization of cockroach allergen extracts. We're actually going to focus on the clinical studies and some of the depletion studies. I'm not going to talk about the IgE combinatorial library work that Jonny Finley is doing my lab. And I'm not really going to talk much about the correlation studies. But those are, obviously, important for our efforts.

The next slide.

Those of you who are on PC, go ahead and click twice and then click to the next slide that's "Current Standardized Allergens."

There's a danger in this part of the talk of preaching to the choir too often, but these are the currently standardized allergens. There are 19 of them. And they're listed on this slide.

The purpose of allergen standardization.

Go to the next slide, please.

Is to establish for each of these allergen vaccines a U.S. standard and to establish a testing procedure to measure the allergen content of each of those allergen extracts. The manufacturers, as you know, may use the established procedure or they may choose to develop an equivalent procedure that they relate to the established procedure.

Next slide please.

The questions comes up as to which allergens should be standardized. And the sort of trivial answer is well all of them should be standardized. But we have to make choices and we have to do -- we have to make choices to which to do. And even if we were going to do all of them, we'd have to choose which ones to do first.

These criteria were discussed several years in a Advisory Committee Meeting. And I really want to focus your attention on the second and the last criteria, which I think are particularly relevant for today's discussion.

The second point is that we should choose which allergens to standardized based on the consistency of the currently marketed product. The idea there is that if the currently marketed products are very consistent and very high quality, should probably will not be that much impact if we choose to standardize. And we might want to put that toward the end of the list.

Conversely, if a product is very highly variable, then we might want to move that product up. Especially if the last criterion is met, and that is if there's a relatively high public health impact of the correct diagnoses or of adequate treatment.

Next slide please.

Among the allergens that have been associated with asthma are several indoor allergens, particularly dust mites, cat, more recently cockroach mold and dog. And among the outdoor allergens, molds have been associated with asthma as well.

The next slide -- you knew if I was going to talk about cockroach, I had to have a slide like this.

PARTICIPANT: Ugh.

DR. SLATER: Thank you. That's what I wanted to hear.

Now on the bottom you have panels with the standard, American, German and oriental cockroaches. But I just have to tell you, those two roaches that -- this really is -- this is not a -- this is not a trick of the digital age. Those cockroaches really are the size of a piece of birthday cake. Those are hissing Madagascar cockroaches, and they really do hiss. And they were actually on tour and they were here at the Smithsonian several years ago. You could actually go down and handle them. But we're not talking about that, so I'm going to get off that slide.

Next slide, please.

PARTICIPANT: Good.

CHAIRMAN LEHRER: Do they have cockroach allergy in Madagascar?

DR. SLATER: You know, I never asked the question. My guess is the problem with those roaches is more significant than allergy.

CHAIRMAN LEHRER: I suspect that's the case.

DR. SLATER: It may be competition for food source.

PARTICIPANT: And birthday cake.

DR. SLATER: And birthday cake, right.

So why is cockroach allergy important? Well, we know that especially in the urban environment cockroaches are ubiquitous. They're extremely difficult to control. And they appear to be associated with asthma, especially asthma in the inner city.

Next slide.

Why do we think that cockroach allergen standardization is important? Well, I think it's important to the patient to provide for more accurate diagnosis of cockroach allergy. And, hopefully, to provide for safer and more effective immunotherapy for cockroach allergy.

To the physician/scientist, I think it's critical because, as I like to say if you can't measure, you can't really study it. There are lots of studies going on and we're just going to have better science if we have a standardized cockroach allergen extract.

And, obviously, at the FDA our concern is to have safer and more effective products.

Next slide, please.

So the first phase in standardizing an extract starts in the laboratory, and that's work that we've done to a great degree, to develop and adapt methods for allergen determination and to compare the allergen content of different lots of allergen.

Next slide, please.

And this is just the front page of that article from the Clinical and Experimental Allergy from last year. This is work that Melissa Patterson and I did on looking at cockroach allergen extracts in the United States.

Next slide.

The goal of that study, which I'm going to review in the next few slides, was to determine the consistency of available U.S. products in protein content, in specific allergen content and in overall allergenicity. And to determine the best lot release measures that might be used.

Next slide.

In order to start at a starting point, we needed to use something as a reference. And we chose to use as a reference extracts labeled E2-Cg and E2-Ca. Cg is for German cockroach, Ca is for American cockroach.

These were both purchased by CBER from Holister Steer, which was then Bayer Laboratories. These were lyophilized relatively high potency extract that had been previously characterized by this laboratory.

There was some limited skin test data that we could use to try to relate these to. And most importantly, it was lyophilized and it was available in very large quantities.

Next slide, please.

We compared these extracts to 26 commercial extracts that we obtained from all of the allergen extract manufacturers. Six of them were aqueous -- I'm sorry. Did I put you on the next slide where the table is? Is everybody on the table with the American and German cockroaches?

There were 6 aqueous and 20 glycerinated extracts.

Now let's go to the next slide, which should be titled "Relative Potency of All Cockroach Extracts As Determined by Competition ELISA."

For those of you that ar working on your own PCs, if you would do me a favor and just go ahead and click three times. That'll highlight in red the 3 numbers that I want you to focus on.

That column is the relative potency scaled to protein content. So the third column, which is just the relative potency, is not controlled for protein content, which was highly variable among these different extracts.

What you can see by looking at the scaled relative potencies is that the potency of the glycerinated American extracts was on average about one-twentieth the potency of E2-Ca. What you can also see is that the potency of the German cockroach glycerinated extracts was about one-fifth of the potency of E2-Cg. And the aqueous extracts were similarly quite low.

Now, a very reasonable next question would be well how potent were E2-Ca and E2-Cg? And I don't have that data in this presentation, but based on the limited skin testing data that had been done before, these were both extracts that were less than 10,000 BAUs per mL. So they were not very highly potent extracts. And conversely, the German and American cockroach extracts that we tested, they were commercially available, were much less potent than those.

Next slide.

At the time we really only had good ways to measure Bla g 1 and Bla g 2 in the German cockroach extracts. We didn't have the availability of a Bla g 4 or 5 assay at the time.

And what you see here is a table that shows the mean and range Bla g 1 and Bla g 2 contents of the glycerinated German cockroach extracts. You see at the bottom the E2-Cg content for Bla g 1 and Bla g 2.

And what you can see is that the commercial extract had Bla g 1 than E2-Cg. And they actually had more Bla g 2 than E2-Cg. What you also see is that the Bla g 2 content was really highly variable, ranging from 8 to 66 micrograms per mL. The bla g 1 content actually was fairly constant, really only running over about a twofold range.

Next slide, please.

We did do some correlation analyses among the German cockroach extract between protein concentration, relative potency, Bla g 2 and Bla g 1 levels. And in general they were fairly correlated. In fact, the weakest correlation was between protein content and relative potency. But there were actually good correlations amongst several of the others, notably the Bla g 2 content and relative potency were fairly closely correlated.

Next slide, please.

So our conclusion from that study was that the commercially available cockroach allergen extracts very widely in protein content, Bla g 2 content, SDS-PAGE banding pattern, which I didn't show you, and overall allergenicity. And they appear to be less potent and contain Bla g 1 than the candidate reference extracts.

Next slide.

So our conclusion from this is that we think that from the point of view of the variability of the products as well as the public health importance of good cockroach allergen vaccines, our conclusion was that cockroach allergen vaccines need to be standardized.

Next slide, please.

What we needed next was to develop a new set of cockroach references to characterize those references to do ID₅₀ testing, which I'll be talking about shortly. And to proceed to the next phase.

Next slide.

So in terms of the clinical testing, skim testing is really the bedrock upon which we can move forward with standardization.

Next slide, please.

And that skin testing is the ID₅₀ testing or IDEAL testing that was designed and promulgated here at CBER and was used to successfully standardize grass pollen allergen vaccine. And there are four basic steps in IDEAL testing.

First of all, this is a technically demanding test. The testers need to be demonstrated in being proficient in the test. You need to recruit an adequate number of study subjects. You need to actually do the testing and then you need to analyze the testing.

The next slide in shows in very abbreviated form the essential elements of that testing. That testing involves serial three-fold dilutions of the extracts, performing multiple tests on a single individual. The measurements in the bottom of orthogonal diameters of the erythema of those test results, and finally the calculation of the D50, which is the dilution at which a 50 millimeter sum of erythema result occurs.

Next slide, please.

The fundamental point of the inclusion criteria as you want to have individuals who really are allergic to the product that you're testing. You want to have individuals that have some clinical history that indicates that they have an allergic disease, such as allergic rhinitis related to exposure to the allergen of interest. And you also want to have individuals that have a positive skin response to the allergen, otherwise the plot that you're going to get for the previous slide will just be flat.

We'd like to get individuals who demonstrate at least a minimal sensitivity of 30 millimeters of sigma E which is the sum of the erthyema diameters to the allergen concentrate.

Next slide.

The exclusion criteria are largely aimed at safety issues as well as the importance of being able to get valid reproducible data.

Next slide, please.

Those of you were at the Advisory Committee two years ago have seen a version of this formula before. This is a formula that's based on the two one-sided test analysis, designed by Schuirman several years ago. This particular analysis is described in a paper that Ron Rabin wrote in collaboration with Rich Pastor and Tony Lachenbruch here at CBER that is going to appear in the proceedings of the most recent Paul Ehrlich Institute.

From that analysis we see that n, the number of study subjects that we need to study, is a function of sigma, which is the standard deviation of the measurement in this case of the D50 measurement and delta, which is the acceptable difference that you might not be able to detect if your sensitivity was too low.

So number sigma is the standard deviation of the measurement. And delta is the minimal difference that you absolutely want to make sure that you're able to measure.

What you'll see here as well by looking at the above formula is that n is a function of the square of the ratio of sigma and delta. So for instance, so if the ratio of sigma to delta is one, at a beta of .05, n is 26 or 26 study subjects.

If you double the ratio of sigma to delta, and you go through a sigma delta ratio of 2, you don't double n, you actually quadruple n. You go up four fold.

Now, if you're looking carefully at this slide, keep in mind that n is always rounded up. So, for instance, where n is 18, it's actually 17.1, which is why 60 as you'll note is really not four times 18. But these numbers are rounded, that four fold calculation doesn't always work out precisely. But that's because of the rounding.

So now we're going to talk about what sigma and delta are or should be and how we're going to estimate the number of study subjects that we need.

Next slide, please.

I'm going to argue in this slide that delta for the D50 should be about ten percent. Ten percent seems like a little bit of a tight measurement, I suspect, to many of you. But I think we have no choice but to be able to measure the D50 within 10 percent. And that's because the number that we really want to come out with is the BAUs per mL of these products. And the BAUs per mL is an expediential function of D50, as given in the formula in the second line of this screen.

Just to give you an example of what ten percent difference is in D50 would yield, this table shows what ten percent differences in D50 would give you in terms of your BAL/mL as changes.

If you had a fairly potent extract with a D50 of 14, this is in the top row of the table, a ten percent decrease in the D50 would lead to an 80 percent decrease in the BAU measurement. Ten percent increase would be nearly a four fold rise in the BAU measurement.

Now, if you go to less potent extracts with D50s of 10 and 8, those differences are somewhat less, but you can see that you really don't want to get much beyond a D50 difference of the percent if you want to have anywhere near a good estimate of what your BAU/mL is going to be.

Next slide, please.

Those of you that have PCs, please go ahead and click one more time, which will focus your attention on the numbers at the bottom of the table.

To estimate what the sigma for D50 is going to be, we actually have some previous data. And those data come from an article by Smith in the Anals of Allergy in 1995 in which they looked at three different cockroach extracts in a number of allergic individuals. And I'm going to direct your attention to the bottom of the page where you can see that the standard deviation for their mean sigma E measurements were on the order of 12 to 20 percent. In the case of the German extract, the standard deviation was 1.21 with a mean of 10.6. With the American cockroach extract, it was 1.94 over a mean of 10.04.

Next slide.

We also have data from an earlier CBER study of German and American cockroach extracts that was performed at three centers, Cincinnati, Georgetown and Johns Hopkins. And I pulled together those data, and that shows you roughly, again, of 20 percent sigma for these measurements.

Next slide, please.

So based on a ratio of about 1.5 between sigma and delta, and based on a power of 80 percent, we estimate that for each of the extracts we would need an n of about 39 or 40 study subjects.

Now, there are two different allergen extracts we want to study, American and German cockroach. That would indicate 80 study subjects.

We also wished to have geographical diversity in our study. We know that there can be some differences in allergen recognition in different parts of the country. I'm not sure that we would necessarily find it from results in different parts of the country, but we do need to try to obtain that. And, therefore, we're estimating that we'll need about 240 study subjects for this study.

Now, we may actually need fewer because it is possible that we will be able to have some overlap between American and German cockroach allergic subjects. That is, if the study subject allows us to test them to both allergen. And if that's the case, we can anticipate that we may be able to do this with between 150 and 200 study subjects.

So at this point we think we've established that cockroach allergen vaccines need to be standardized, and our next step is really to go forward with this clinical study. And at this point I want to report on a fairly happy collaboration that seems to be going forward very nicely.

Next slide, please.

Some months ago I approached the NIAID Inner City Asthma Consortium to discuss the possibility of collaborating. I should note, even before you go any farther, that we're fortunate that Dr. Gruchalla is not only on this Advisory Committee, but she's also a member o the Inner City Asthma Consortium steering committee, and has been involved in this process for the last several months.

For those of you that don't know, this consortium was established in fiscal year 2002 to explore and evaluate promising new strategies for the treatment of asthma among minority children residing in the inner city. This consortium of basic scientists and clinical investigators will conduct clinical studies to elucidate the immunopathogensis and the natural history of asthma in this population.

Next slide, please.

This is a list of the members of the consortium steering committee, who I must say were very hospitable to this project. I think the steering committee really recognized that their ability to go forward and study the role of cockroach as a causative agent in inner city asthma was dependent on having good reagents and this was really a happy confluence of interests between CBER and NIAID.

Next slide, please.

So, where are we? Well, the steering committee approved the process and we're actually in the process -- they are in the process of identifying the study centers that will participate. You should understand that the Inner City Asthma Consortium has in place study centers in the inner city that are intended to perform studies on inner city asthma. So they are simply finding out which ones will be willing and able to participate in the study.

We need to, obviously, get many IRB approvals. We need to get IND approval, which will be an interesting experience for me. We need to order the extracts, distribute the material.

I think the proficiency testing should not be a problem. And then we're going to proceed with the study.

As we proceed with the study, we'll be generating data. The inner city has a consortium, has a consultant that will be analyzing the data. We will also be obtaining the raw data and analyzing it ourselves. We'll be doing correlative studies with in vitro tests. These studies will be done both in my laboratory and in other participating laboratory from the Asthma Consortium. I anticipate this is going to be a very, very productive collaboration.

I should tell you that our time table now is aimed towards actually starting the clinical part of the study by the end of the summer. Once that is starts, it really should not take too long to generate the kind of data that we all need to move forward.

Next slide.

CHAIRMAN LEHRER: Jay, could I just ask a quick question?

DR. SLATER: Absolutely. Sure. Yes.

CHAIRMAN LEHRER: That your citing this very important study. And this kind of came to mind in the beginning when you were talking about the variability of the commercial reagents and also the fact that they seemed to have somewhat lower activity than expected. And it brought to mind the studies of Bob Esch and others with the fact that cockroaches with proteolytic enzymes and that they can essentially digest a lot of the allergens.

What I'm asking is that it's -- your dealing with standardization of, and particularly this kind of study which is kind of, as you say, useful. Has stabilization been addressed at all prior to embarking on this?

DR. SLATER: I think it's a very good question.

We really didn't see any evidence in our previous study that there was tremendous instability in the products. We did look at several of the products longitudinally to see if we noticed any inherent degradation. In fact, interestingly enough, products from the same manufacturer even of different ages were actually fairly consistent among themselves.

So we didn't see a lot of evidence of that.

I'm not sure necessarily how active the proteases are in 50 percent glycerol. Certainly what we're going to be using are going to be glycerinated extracts at this point. And we are going to be looking at stability issues as we move forward. But at this point what we're going to do is we're going to obtain extracts that are glycerinated. We're going to characterize them as well as we know how in the laboratory. We're going to get them, do the clinical studies and then we're going to see where we are with that.

CHAIRMAN LEHRER: I think that sounds good, particularly that they're glycerinated. But something you might want to keep in mind, at least discuss with your colleagues in the study, is that everything is done the same way. For example, when extracts are brought out for skin testing or whatever, sometimes some people leave them out for a couple of hours, sometimes people leave them out for a whole day; those kinds of things that if you can try to standardize in some way or just keep them in the same time frame I think are important. Because they could make a big difference in your results. And I think that this such an important study, you want to minimize any of that.

DR. SLATER: Well, I think that raises a good point. The way we're actually going to store the XEX, is they're going to be stored in insulated racks which the investigators will keep in their refrigerators, and then they'll take them out in those racks. So we really should have fairly good temperature control, at least within the constraints of the study. But you're raising a good point.

CHAIRMAN LEHRER: Good. Thank you.

DR. SLATER: So I think I already directed you to go to the next slide.

It's not going to be any surprise to any of you, given the previous talk, that we're actually looking at the endotoxin content of cockroach allergen vaccines. It's also not going to be any surprise to any of you that there's a fair amount of endotoxin in these allergen vaccines, although again I'm often surprised at how variable it is. But some of the levels are actually quite high, and we are going to be looking at that as well.

Next slide, please.

So certainly the endotoxin issue needs to be studied in depth.

Dr. Lehrer, just to address a question I suspect is in your mind already, we do heat inactivation on these, and we didn't really find any evidence of heat inactivable activity causing the gel-clot to occur. Likewise, when we did the endotoxin neutralizing protein absorptions, we really found that virtually all of this was adsorbed out. So it appears to be that this is real endotoxin that we're measuring. But, again, the proteolytic enzymes could be there, they're just not contributing to the LAL gel-clot measurement.

Next slide.

The next question that I'd like to address as part of our cockroach allergen standardization protocol, and this is actually the last topic of the afternoon, is whether overall allergenicity measurements will be sensitive to changes in specific allergen levels. This is a question that really was first raised shortly after I came to the lab when we were looking over mite stability data. These were data that appeared in publication in JACI in 2000 looking at the stability of dust mite preparations over long periods of time.

And one of the interesting things that we found that we really didn't quite know how to interpret was that when we measured the relative potency of these products over 12 months, we found that they were absolutely rock stable, whether we stored them at minus 20 degrees or at 4 degrees. They were really quite impressively stable in terms of our measure of overall allergenicity. But when we looked at specific allergen, either by measuring group 1 and group 2 allergens in dust mites or by looking at specific bands in the SDS-PAGE we found significant loss of activity or of bands when the products were stored at 4 degrees. Not minus 20 degrees, but at 4 degrees.

So it raised the question that I'd like to address in the next couple of slides.

Next slide, please.

When you have an assay in which you're measuring a particular protein with a monospecific antiserum, this could either be a monochronal antiserum antibody or it could be polychronial antibody that was raised to a specific well defined protein. What you basically have is the situation in this cartoon. In this cartoon the multi-colored smiley faces on the right represent five different proteins in your allergen mixture and the blue arrow to the left of them is the enzyme that you use -- is that the antibody that you're using to detect one of those proteins. In this case, you're using the blue arrow to detect the blue protein.

And you can easily imagine that if the amount of the blue protein begins to drop out, that you're specifically be able to measure that dropout of the blue smiley face. But the situation we have with the competition ELISA is much more complicated.

Next slide.

With the competition ELISA we have a somewhat different picture. We're measuring, let's say in this carton, five different proteins with five different colored arrows, five different antibodies. So each one represents its own smiley face.

Well, it's not entirely clear that if you lose one of those proteins, you're actually going to pick up the signal. It's going to be sensitive enough to detect it, and that's shown in the next slide where you see that in the top part all five of the smiley faces are represented and all five of their colors. But in the bottom one, those of you that are looking closely, will see that the blue one is gone. And there's a white space in its place.

So the question arises as to whether when you deplete a complex protein mixture of a single protein and you're measuring with a mix of different antibodies whether the signal is actually going to pick up that loss of the specific protein.

Next slide.

We are proposing to address this question by doing a series of depletion analysis of German cockroach extracts. We have already raised very high titer specific antibodies to Bla g 1, 2, 4 and 5. We plan to selectively adsorb the cockroach extracts that we have with each of these. We're then going to test for specific allergen levels and then test for overall allergenicity using the competition ELISA.

Next slide, please.

This is not a very impressive SDS-PAGE. We've actually attempted this with both Bla g 1 and Bla g 2. I'm going to show you the Bla g 1 data, and I'm going to tell the Bla g 2 data, because that actually just came to me yesterday.

This is the SDS-PAGE. And what you see here in the left most lane is just the molecular weight standard. The next lane, which is labeled CR is just the crude cockroach extract. We then have an empty lake. The next lane that's entitled Sham was a cockroach extract that was adsorbed with just normal rabbit serum. And finally the last one is the one that was specifically adsorbed with the anti-Bla g 1 serum.

And what you can see, and really you guys will have to have an awful lot of faith in me, and I wouldn't blame you if you didn't, the only band that I could find that was really different between the sham and the adsorbed one was that band all the way at the bottom that you can see appears to be absent in the last lane and is not quite absent in the sham lane. There's really no other difference that you can see on the SDS-PAGE. And I'm not even that sure that difference is highly significant.

But let's go to the next slide.

What can you show, though, is that as measured by ELISA in our laboratory in the upper left hand panel, there's a big difference in the ELISA results with the product that was adsorbed with the anti-Bla g 1 as opposed to the one that was adsorbed with just the regular rabbit serum.

In addition, when we sent these same samples to a contractor to have it tested for Bla g 1, 2 and 5 content, you could see a fairly dramatic drop in the Bla g 1 content from 149 unit per mL down to 14.8 units per mL. The Bla g 2 content appeared to go up and it's probably not significant. And the Bla g 5 content appeared to go down somewhat, but not quite as dramatically as the Bla g 1.

So on the next slide we compare these two extracts with the competition ELISA. And you can see that there's really no difference between those two in this particular competition ELISA.

Now, let me just stop for a second. There are questions that we need to answer about the competition ELISA as well. We do know that the serum that we used, the serum pool that we used, does recognize Bla g 1, and I think that's probably the most important question. But we need to go back and actually qualify and perhaps retool the competition ELISA to try to optimize it better than it's been optimized so far. And what I mean is that these selectively adsorbed products that we're generating could actually help us in our effort to optimize the competition ELISA better. But it certainly does raise the question as to whether a measurement of overall allergenicity that looks at 5, 6 or 7 allergens really is capable of detecting a specific drop in a specific allergen of interest.

Next slide.

So, the final conclusion that we have that we're going forward with is we have to acknowledge the possibility that our best surrogate tests, our best test for potency, may in fact not be the competition ELISA. And it may not be the competition ELISA for two reasons.

There may be other tests that will simply correlate better. But in addition, if we decide that one or two or three proteins are particularly important, we need to make sure that changes in those proteins are picked up in whatever assay or assays we choose.

Last slide.

So in conclusion I'm very optimistic that we're going to be able to standardize German and American cockroach allergen vaccines in the near horizon, and this will facilitate definitive studies on the role of cockroach allergens in inner city asthma, and on the best methods for eradication and treatment, and it will make for safer nd more effective products.

And I want to thank you all very much for your attention.

CHAIRMAN LEHRER: Thank you very much, Jay.

I have a quick question concerning reactivity with the issues that you were addressing in your last several slides about the importance of reactivity to particular allergens within the mixture and trying to get at that looking at removal. And then when you started talking about your reagents, I think that really hit the mark in terms of the serum that you're using, and that's really going to in some ways define -- or in many ways define your results.

Do you think another way to look at this might be to just see what type of reactivity the populations have? I think a lot of this has been done already in terms of different allergens. Well, perhaps for the newer allergens it hasn't been. Just to see that the variability? I mean, do you have people that vary that differently in terms of these individual allergens? The reactive. Because those are the ones that, you know, could be most effected by using a single allergen or several allergens to standardize the extract.

DR. SLATER: Right. Well, part of the study -- I didn't really go through it in all of its glory. But part of the study that we're planning to sort of tag onto the IDEAL study, is we are going to be getting blood samples from each of the study subjects. And we're actually -- one of my intentions is to actually use that to correlate the skin test result with specific antigen reactivity in each of the study subjects. And then it's determined whether it's even theoretically possible to construct the serum pool that will detect each of the allergen specifically.

Another alternative, of course, is instead of having a single competition ELISA that measures overall allergenicity, to have a multiple individual test and to set multiple individual limits. But that has problems associated with it as well. But it is one of the objectives of the collaborative study to try to answer some of these questions.

CHAIRMAN LEHRER: Good. I think it's particularly important to know the variability of reactivity, and it may not be as variable as it could be. But I think that will provide very useful information.

Very nice.

Now are there other comments or questions in the Committee?

MR. HAUCK: I have a couple of questions and comments.

Jay, did you notice or did any of the technicians that were doing the assays with the commercial cockroach extract, did they notice sediment or precipitate?

DR. SLATER: Yes.

MR. HAUCK: Yes, they did?

DR. SLATER: Well, in other words in our work with the commercial cockroach extracts?

MR. HAUCK: Yes.

DR. SLATER: Yes. Yes, there were. And we made special effort to assay them as soon as we could after we got them. But some of them were actually precipitated upon arrival.

MR. HAUCK: It's a major practical problem that the manufacturers have, even with the glycerinated cockroach extract. And we don't know what is causing it.

Kind of the latest thing, you talked about comparing --

CHAIRMAN LEHRER: Could I interrupt for a second, Peter?

MR. HAUCK: Sure.

CHAIRMAN LEHRER: Is there any information that the allergens are actually in those precipitates or do the precipitates alter the allergenicity?

MR. HAUCK: I'm not aware of any data.

DR. SLATER: Well, we've looked at it with several other allergens, and we haven't really seen any evidence of that. We also looked at it briefly with some of our cockroach extracts, and we didn't see any evidence of that either.

But we certainly can't say that we've done it on enough extracts to draw a broad conclusion about that.

CHAIRMAN LEHRER: That would be important information.

DR. SLATER: Absolutely.

MR. HAUCK: May I continue?

CHAIRMAN LEHRER: Please.

MR. HAUCK: Sure. Thank you.

Another kind of related issue, when you compared the reference that you had from Holister Speer and you compared relative potency based on protein content between the two, the protein content found in the commercial material and the protein content in the reference, correct?

DR. SLATER: Yes.

MR. HAUCK: Okay. Be aware that, I believe the reference material was dialyzed and the dialyzed products, cockroach is very rich in urea, which will assay as either an enheightened protein or protein nitrogen.

And as far as I know, the commercial extracts are not dialyzed, so they're very rich in urea, which also may in fact be part of the problem that we manufacturers have with the precipitating issue.

CHAIRMAN LEHRER: Peter, I'm sorry. So the dialyzed product would not have much urea in it, is that what you were saying?

MR. HAUCK: Right. So it would have a lower protein content and therefore, if you compare it on an equal protein content, it's very likely that the reference would have a lot more allergen activity per mL protein.

Maybe another way to look at it would be if you could get milligrams of starting material.

CHAIRMAN LEHRER: Yes.

MR. HAUCK: Jay, I could double check on that. But that -- that's what I understand. But it also could be a possibility that maybe that's the way to manufacture the product, dialyzed.

DR. SLATER: I think we found that the E2-Cg and E2-Ca had a substantially higher protein content than the others. So I'm not really sure. But what you're saying is that that protein was actually protein whereas the protein that we were measuring in that cockroach -- most cockroach extracts was urea

MR. HAUCK: That's been my experience and my colleague's experiences.

DR. SLATER: Thank you.

MR. HAUCK: Okay. And then just a comment on the -- you're to be applauded on the sides of the clinical trial. And I just wondered if you thought about, and maybe Dr. Lehrer can respond to this as well, the American cockroach and the German cockroach don't necessarily hang out together except for maybe New Orleans. But that's probably a good reason to use a couple of different geographical areas.

I think you may have to demonstrate if you're going to use American and German on the same patient, that you're not just measuring activity, but you should demonstrate that those allergens from both species exist in that patient's environment.

CHAIRMAN LEHRER: I think that's a very good suggestion. And you really want to look at the diversity of areas in terms of reactivity.

DR. SLATER: Thank you.

CHAIRMAN LEHRER: And just a comment on New Orleans cockroaches. I think, as any of you know who have visited our fair city, that New Orleans' cockroaches I think will hang about with just about anybody. And they do that.

DR. SLATER: And they don't just eat birthday cake, right?

CHAIRMAN LEHRER: Not yet. Okay.

Any other comments or questions? Okay.

Jay, I wanted to thank you very much for an excellent presentation.

PARTICIPANT: Yes. Thank you, Jay.

CHAIRMAN LEHRER: Next on the agenda, I think I have it correct this time, is the public hearing.

Bill, could you make the appropriate announcements regarding the open public hearing on our agenda?

EXECUTIVE SECRETARY FREAS: Thank you, Dr. Lehrer. As part of the FDA Advisory Committee meeting procedure, we hold open public hearings in order to give anyone not on the agenda a chance to address the committee and make a statement concerning matters pending before the committee.

In response to the Federal Register announcement regarding this meeting, I have not received any requests today to participate in the open public hearing.

I would like to transfer to either Shiela Langford or Denise Louster down in Conference A. And Shiela or Denise, could you activate the microphones by saying good afternoon? It'll take a second. If you have your mute button on, would you take the mute off.

MS. LANGFORD: Hello.

EXECUTIVE SECRETARY FREAS: Yes. Can you keep talking?

MS. LANGFORD: There's no one here.

EXECUTIVE SECRETARY FREAS: Pardon?

MS. LANGFORD: There's no one here for open public hearing.

EXECUTIVE SECRETARY FREAS: Okay. We do not have any open public hearing at this time, Dr. Lehrer. We have a small group of people down in Conference Room A.

We appreciate your participation in the meeting. And that closes our open public hearing.

And Dr. Lehrer, I turn it over for you if there's additional Committee discussion.

CHAIRMAN LEHRER: Thank you very much, Bill.

So in the concluding part of this meeting then, are there any other comments or discussion concerning the information presented during this meeting?

Any closing comments to be made by anyone?

EXECUTIVE SECRETARY FREAS: Yes, I just would like to thank all the Committee.

First of all, I would like to thank you for chairing this meeting. And thank all the Committee members for putting up with the difficulties that we have encountered today.

I do apologize for the technical difficulties. I believe if you were aware of the cold, wet, rotten rainy weather that we are having now in Washington, D.C., you'd be very appreciative of the fact that you were able to attend by video conference and teleconference as opposed to experiencing the climate down here.

But, again, my thanks to everybody for participating.

I would like to remind you that the next tentatively scheduled meeting of the Allergenic Products Advisory Committee is November 19, 2003. We'll be contacting you in September, as soon as we start formulating any plans regarding the possibility of holding a meeting on that date.

So, again, my thanks and that's all.

CHAIRMAN LEHRER: Okay. I'd like to thank the members of the Committee for participating and also the staff of CBER for arranging the meeting and the presentation.

If there are no further comments, then we can adjourn the meeting. Thank you all very much.

(Whereupon, at 3:22 p.m. the meeting was adjourned).