UNITED STATES OF AMERICA

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

(8:07 a.m.)

DR. FREAS: Good morning. I would like to welcome everybody to this, our third meeting of the Transmissible Spongiform Encephalopathies Advisory Committee.

I am Bill Freas. I'll be the Executive Secretary for today's meeting.

I would like to now go around and introduce the members who are seated at the head table. I will be announcing the names of people who are possibly stuck in a snow bank that will be here and sitting at the table very shortly.

In the first seat -- he's on his way -- is Dr. Richard Johnson, Professor of Neurology, Johns Hopkins University.

In the next seat, Dr. David Bolton, head of the Laboratory of Molecular Structure and Function, New York State Institute for Basic Research.

Next is Dr. Ermias Belay, Medical Epidemiologist, Centers for Disease Control and Prevention.

Next is Dr. David Gaylor, Principal Scientist and Vice President, Science International, Incorporated, Little Rock, Arkansas.

Around the corner of the table, Mr. Val Bias, Co-chairman, Blood Safety Working Group, National Hemophilia Foundation, Oakland, California.

Next is Dr. Elizabeth Williams, Professor, Department of Veterinary Service, University of Wyoming.

Next is Ms. Shirley Walker, the consumer representative for this committee. She is Vice President of the Health and Human Services' Urban League of Greater Dallas, of North Central Texas.

Next, Dr. Nick Hogan, Assistant Professor, University of Texas, Southwestern Medical School.

Next is our Acting Chairperson for today, Dr. Suzette Priola, investigator, Laboratory of Persistent and Viral Diseases, Rocky Mountain Laboratories.

Next is Dr. Arthur Bracey, Associate Chief of Pathology, St. Luke's Episcopal Hospital, Houston, Texas.

Next is Dr. Sidney Wolfe, Director, Public Citizen Health Research Group, Washington, D.C.

Next is Dr. Stephen DeArmond, Professor, Department of Pathology, University of California, San Francisco.

Next is Dr. Rima Khabbaz, Acting Associate Director for Epidemiological Science, National Center for Infectious Diseases, Atlanta, Georgia.

In the empty chair we will soon be joined by Dr. Lisa Ferguson, Senior Staff Veterinarian, U.S. Department of Agriculture.

Next is Dr. Pierluigi Gambetti, Professor and Director, Division of Neuropathology.

Next is our industry rep. for today, Dr. Stephen Petteway, Director of Pathogen Safety and Research, Bayer Corporation.

I would like to thank all of our members for coming. There is one member who will not be joining us today, and that is Dr. John Bailar, University of Chicago.

I now would like to read into the public record the conflict of interest statement written for this meeting.

The following announcement is made part of the public record to preclude even the appearance of a conflict of interest at this meeting. Pursuant to the authority granted under the committee charter, the Director, Center for Biologics Evaluation and Research, has appointed Drs. Ermias Belay, David Bolton, Arthur Bracey, David Gaylor, Nick Hogan, Rima Khabbaz, Elizabeth Williams, and Mr. Val Bias as temporary voting members for this meeting. Based on the agenda, it has been determined that the committee will not be provide advice on specific firms or products at this meeting.

The topics being discussed by the committee in open session is considered to be a general matter's issue. To determine if any conflicts of interest exist, the agency reviewed the agenda and all relevant financial interests reported by the meeting participants.

The Food and Drug Administration prepared general matters waivers for special government employees who required a waiver under 18 U.S. Code 208.

Because general topics impact on so many entities, it is not prudent to recite all potential conflicts of interest as they apply to each and every member. FDA acknowledges that there may be potential conflicts of interest, but because of the general nature of the discussion before the committee, these potential conflicts are mitigated.

We would like to note for the record that Dr. Stephen Petteway is serving as the non-voting industry representative member for this committee. He is employed by Bayer and thus has interests in his employer and other similar firms.

Listed on the agenda are speakers making industry and federal government presentations and/or updates. These speakers have financial interests associated with their employer and with other regulated firms. They were not screened for these conflicts of interest.

Members and consultants are aware of the need to exclude themselves from discussions involving specific products or firms for which they have not been screened for conflict of interest. Their exclusion will be noted for the public record.

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

Waivers are available by written request under the Freedom of Information Act.

So ends the reading of the conflict of interest statement.

Thank you.

Dr. Priola, I turn the microphone over to you.

CHAIRPERSON PRIOLA: Are we going to do the presentation of the plaques?

DR. FREAS: Yes.

CHAIRPERSON PRIOLA: Yeah, right now we're going to present plaques to members of the committee who are rotating off.

DR. FREAS: Dr. Lester Crawford is coming here to present the plaques. He is Deputy Commissioner for FDA.

DR. CRAWFORD: Thanks very much.

It's a pleasure for me to be here this morning. I was a sometimes member of the committee myself off and on for a couple of years, and I want to congratulate all of you on the work that you've done and also the important work that you are to do in the future.

We just convened, as some of you know, the CODACS commission meeting of the U.N. this past week in Geneva, and as all of you would know, but many people don't know, this is a leading priority of that organization.

We also just went through getting the budget stuff ready for the next cycle at FDA, and it is amazing the train that TSEs, in general, and BSE, in particular, in chronic wasting disease to a somewhat greater extent even has to do with our budget.

The Center for Veterinary Medicine that I formerly headed up has a budget item for BSE which basically exceeds all of their other budget items in terms of enforcement actions. So we're going to be needing a great deal of advice from you. Your work is really sort of just beginning.

For all of you that have served and that we are about to recognize, I am very appreciative on behalf of the Commissioner and the Secretary of Health and Human Services to Dr. Bolton for steering this committee. We're very, very grateful, and I did see you steer even me one time, and I appreciate that.

So why don't we recognize them one by one? First Dr. Ermias Belay. Dr. Belay, thank you very much. Congratulations.

(Applause.)

DR. CRAWFORD: I was your special government employee and, therefore, you're available for other assignments. I was entitled to other assignments for 25 years, and unfortunately they used me. So as long as you're there, we'll be knocking on your door.

Dr. Beth Williams is next, and especially appreciate a dear colleague like Beth serving so long. Thank you very much.

(Applause.)

DR. CRAWFORD: And, Dr. Bolton, here is your battle award right here. Thank you very much, indeed. We appreciate it.

(Applause.)

DR. CRAWFORD: And I hope you have a good meeting, and please know that you're on our minds, and your deliberations and conclusions will be considered Dr. Priola, very, very carefully by us in our office.

Thank you.

CHAIRPERSON PRIOLA: Okay. Thank you very much, Dr. Crawford.

We might as well go ahead and open this meeting up. I'd like to welcome everybody and congratulate you for making it through despite the weather and the flight delays. I know that I live out in Hamilton, Montana, and it makes me appreciate all the more those nice mild winters that we have here versus what you have to deal with when you get these big storms.

I think we'll go ahead and start. The first topic is essentially an update on the implementation of the variant CJD guidance and how that has affected the current blood supply and the current monitoring situation.

And our first speaker for that is Dr. Alan Williams, who is the Director of the Division of Board Applications at the FDA.

DR. WILLIAMS: Good morning, and thank you.

Either we need to hit the lights or boost up the projector a little bit. Thanks.

What I'm going to do is give a brief overview of the topic, the blood supply in general, and various programs that have been put into place to monitor the blood supply and then move into ways that some of these programs can perhaps be conducted in a more coordinated fashion, capture a wider breadth of data and perhaps capture data more in real time that will be useful both on a predictive level as well as defining when responses might be necessary from either governmental or industry appeals.

First slide, please.

So the title of the talk is monitoring the blood supply. What do we really know, and I think the aim of this talk is really to hone in on some of the areas that we currently don't have data on and for which data would be very useful.

In terms of overview, because I know some committee members are new, the data currently available are from 1999 comprehensive survey conducted by the National Blood Data Resource Center. I should say right off that there was another data collection activity in 2001. A comprehensive report of those data is, I understand, complete and release is imminent. So there are updated data available.

But based on the 1999 data, roughly 13 million whole blood units are collected. About 12 million of those are actually transfused. About 92.5 percent of those collected are given to patients, and therefore, the margin of available whole blood to what is transfused is quite narrow. It is about five and a half percent in a similar survey conducted in 1997, about 7.5 percent in 1999. The margin in 2001 is unknown, and I think we're anxiously awaiting the see exactly what that figure is.

We do know that the demand for transfusion increases roughly four percent per year, and in large part this is due to increased use of chemotherapy and major increases in organ transplantation, particularly organs which require a large use of transfusion.

The whole blood and red cell supply increased about ten percent between 1997 and 1999, roughly five percent per year versus rising demand of four percent per year. Again, a continuing rather slim margin, but recognizing that in many instances the supply does increase proportionally to the need for additional blood and obviously collectors want to prevent wastage of that blood.

This survey that I referenced is a published report, comprehensive report, on blood collection and transfusion in the United States, and the 2001 report should be available very soon.

Next slide.

To characterize the blood donor base, nationwide approximately 80 percent of the blood collected is from repeat blood donors, and this is an important fact because these are really the backbone of the blood supply in the nation, and when deferrals occur either due to screening tests or travel or other medical deferrals not only do you lose a donor, but you lose a donor who potentially would be repeating into the future on a regular basis.

So that is one aspect of deferrals that really does impact the donor base.

Donors are changing in demographic characteristics as well as other characteristics. One of the most notable factors is many of the donors who had experience with the World War II and Korean War era became regular donors, and these individuals now are less prevalent in the donor base. Younger donors, at least on the whole, seem to donate with a little less enthusiasm. So it's tougher to get people in to donate blood, a little less civic spirit, if you will, and I think perhaps some of the industry speakers will further comment on that.

In addition, work sites which used to be a mainstay of collections, work sites are becoming smaller, and many of the large work sites because of economic pressures are permitting fewer and fewer blood collections in house.

The blood supply has been historically elastic. We recognize that at various times three to five percent of the donors could be lost by a specific deferral implementation or a test, such as the anti-Hepatitis B core test implemented in the mid-'80s; that we knew the elasticity at least could tolerate a three to five percent loss.

We also know that traditionally supply is limited during the summer and the holidays. I think this shortage both during the summer and the Christmas period was especially acute in 2002, and I'm going to discuss that a little bit.

We know that the public responds to appeals, and they respond in times of crisis. However, we do not know the long-term impact, how many of these donors continue to make regular contributions to the blood supply. Limited data that are available indicate that the long-term impact is, at best, neutral; that individuals who come out in times of crisis tend not to become long-term donors.

Next slide.

Now, the impact of donor travel deferrals was discussed intensively by this very committee over the past years. The deferrals for travel to countries which had endemic BSE in which donors may have had dietary exposure were implemented in two phases, in May and October of 2002.

The projected loss of donors was five percent nationwide, and this figure came from survey research done at ten blood centers nationwide, and I'll say a little bit more about that data collection in a few moments.

Unfortunately, however, the actual loss from this deferral is not measurable for a couple of reasons. We certainly know what donors come into the blood center and are deferred based on that deferral and are unable to donate.

However, certainly there's a lot of advanced notice of this deferral both through the media and through blood centers specifically, advising donors of this new implementation. So many donors would simply never appear at the blood center who might otherwise.

So the actual loss to the blood supply is difficult to measure.

We know that a major component of the industry implemented variant CJD deferral that is somewhat more restrictive than the FDA recommendation, and we also know from both the survey as well as the characteristics of donor bases around the country that the coastal cities would be hit harder by not only the variant CJD deferral, but other travel related deferrals; that the real United States mid-America would have somewhat less of an impact; that the New York Blood Center would have been particularly hard hit both by the higher travel rates in New York, as well as the loss of blood collected in Germany, Switzerland, and Holland which could not longer be imported due to the deferrals.

Military bases. We presumed that losses would be relatively high on bases. Unfortunately, while military individuals tend to be very good donors, they are not good survey respondents. So we weren't able to predict that too well.

And finally, in the many discussions that we had, this committee requested that FDA specifically work toward establishing supply monitoring systems and assess the impact of this deferral and report back to the committee on a regular basis.

Next slide.

There's also been much recent oversight of the blood supply at high levels. In November 1999, the Public Health Service and Health and Human Services committed to maintaining the blood supply and monitoring that supply by creating an amendment to the blood action plan entitled "Monitoring and Increasing the Blood Supply."

In July of 2002, the Government Accounting Office issued the second of two reports on the blood supply, specifically related to the variant CJD travel deferral, and it was entitled blood supply generally adequate despite new restrictions, and I think that's kind of the bottom line of the GAO report; that while the blood supply is tight, their view is that it's generally adequate and still has some resiliency.

On September 10th, the House Subcommittee on Oversight and Investigations held hearings on blood supply and blood supply monitoring, and similarly in November, due to New York's unique position with respect to supply, the New York State assembly held hearings.

Next slide.

And then in September of this year, the HHS Advisory Committee, Blood Safety and Availability, discussed blood supply and blood monitoring in considerable detail and issued some specific recommendations to the department.

The first was that the HHS should promote increased public awareness of the ongoing need for routine blood donations by healthy persons, and some of the subcomponents of this involved periodic PSAs and visible blood donations by high level public officials; funding of demonstration projects related to behavior and educational approaches to improving blood donation patterns; supporting specific initiatives to encourage routine donations by young persons and minorities, and playing a leading role in increasing participation of federal employees in donating blood.

Secondly, that the department should maintain and/or increase funded support for blood supply monitoring systems. And the characteristics of these systems should include monitoring long-term trends, data on daily inventories, indicators of blood shortages and excesses, and I think most importantly predictive models to identify trigger points ahead of when a blood supply crisis might occur so that a coordinated national donation campaign might occur.

And then finally, DHHS should support initiatives to improve management of blood inventories by defining roles for both liquid and frozen reserves of red cells to moderate fluctuations in supply and to improve disaster preparedness.

Next slide.

So what I would like to do with the balance of the talk is focus specifically on some of the things that we would like to know in terms of monitoring the blood supply. And I set out three goals, and I'm going to try to integrate those goals with the current programs and then the speakers that follow will give more detail about the programs.

So goal one is to monitor the breadth, impact, and duration of blood and reagent shortages. Such a program should have the sensitivity to define regional and local shortages through data derived from the entire population or a large representative sample. Clearly, to get a local perspective and cover the entire country one needs to have quite a large area from which to draw data.

The data should be available in real time. A program could be targeted to Group O red cells and platelets. I think most people will agree if you have Group O red cells, typically you would have blood as needed. So it's not necessarily to monitor other blood types in detail, that this probably would serve as a surrogate for overall supply.

Shortage alerts can be derived both from blood collection centers and from transfusion services. As I think you'll see from some preliminary data, blood collection centers tend to give an earlier warning of an impending shortage, while transfusion services, when they're short, you know there's an acute situation.

And I think, again, importantly the trends and characteristics of shortages need to be applied to predictive models so that one can anticipate when a problem is going to occur instead of looking at it historically and trying to address it.

Systems currently in place that produce data related to shortages include the DHHS supply monitoring program, the National Blood Data Resource Center monthly sampling, the American's Center Stoplight Program, the American Red Cross system inventory data, and the FDA Transnet system, which is under development and recently piloted, which I'll show a few slides on.

Next slide.

The second goal, monitor the long-term blood collection and utilization trends. I start off the talk by showing some nationwide data on overall collection and utilization. These data come from a single source, and clearly, these type of data are absolutely crucial to understanding the blood supply in the country.

Components include total collections of both whole blood and components, transient component use, monitoring of outdated products, capacity for ad hoc collection of operational data, data available to the public and to other agencies as needed, and perhaps something that hasn't occurred yet but would be certainly very useful is to be able to match blood utilization to patient ICD code or similar indications so that we can get an idea of how much blood is used for individual procedures and start to establish some benchmarking data around the country.

Such a survey, currently it's done bi-annually and would be an advantage to have data like this available on an annual basis and available sooner rather than a year to a year and a half after it's collected.

There is a single source for these types of data. It was collected every two years by the Center for Blood Research under the National Heart, Lung and Blood funding. That program then switched to the National Blood Data Resource Center, which continues to conduct a biannual survey of blood collection and utilization, but as I suspect will be mentioned shortly, this program is not now externally funded, and I think we would probably move much more quickly toward having a comprehensive, stable survey component if some funding were available.

Next slide.

And a third goal is to monitor blood center inventories both at major facilities and reserve depots. This would provide a continuous knowledge of inventory available for emergency distribution and early warning of impending shortage.

And both the ABC stoplight system and the Red Cross inventory currently provide those types of data, but not in an integrated national way.

Next slide.

So what are the opportunities at this point? How can we build a better monitoring network?

It needs to be sensitive to local shortage. It needs to be representative. Data need to be available in real time. I think there's value to be able to integrate blood center and transfusion service supply data to build a model regarding the dynamics of supply and utilization.

As I mentioned, both blood centers and transfusion services encounter shortages, but they might be saying different things when they occur.

Ability to use the data for prediction. It hasn't occurred yet. It's gotten a lot of discussion. It has not been usefully made available at this point.

And data access to a broad based set of customers would be an advantage to both public and federal agencies to respond as needed.

Next slide.

What I'd like to do in the balance of the time is discuss the pilot program that FDA has developed, and this is a shortage indication system. It's entitled Transnet, primarily designed as a system to measure blood shortages, but as I'll address at the end of the talk, it could be easily adaptable to reagents or other shortages throughout blood centers and transfusion services.

I'd like to briefly acknowledge colleagues from the FDA involved in the development of this program. We're also working closely with Voxiva Corporation, which hosted the server for the pilot program and has been developing a telephone data input component for the program, as well as numerous colleagues in industry who have both helped guide us in the design of the program and have served as pilot sites.

Next slide.

So what are the design characteristics for a shortage monitoring system? Keep it simple. We designed a program to utilize data routinely compiled by the facility, and the report done on a Monday through Friday basis is simple with a no shortage option for a typical situation, a single keystroke through the Web-based utility to indicate no shortage.

However, if a shortage does occur, we do allow that site to provide details as to the characteristics of that shortage through a drop-down list.

Defining shortage is a little bit tricky because if you use strictly objective criteria, the definitions might vary from one institution to another, depending on supply networks and other variables. So one needs to recognize local variations in a shortage definition.

Transnet was originally designed as a Web-based program. We anticipate that about 40 percent of transfusion services would not have ready access to the World Wide Web, and so we have developed a telephone reporting system which uses the touch tone keys of the phone to actually input data right into the server, a very neat system.

Sites need incentive for participation. The program is not designed to directly pay sites for data, and we feel that by having data from the program rapidly available to the participants it will be a big advantage, as well as having the data put to meaningful use to help improve what blood centers and transfusion services deal with on a daily basis, which is adequate supplies of blood.

We also hope that the data might be turned around and used as support for recruitment efforts in the future.

Next slide.

Transnet is population based. The pilot study was based on nine sites. We anticipate a second phase pilot of several hundred sites, and then ultimately we hope to roll out the program to all blood centers and transfusion services in the country, about 145 blood centers and 5,000-plus transfusion services.

And the capacity of the system could handle all of these data and I think would clearly reflect local trends as needed if we have full participation.

There is a denominator. Sites that wish to participate will register and will be expected to respond on a daily basis. Data entry, as I mentioned, is through a single keystroke or through a telephone response, and there is a reminder sent to the sites each day to input their data.

Next slide.

And the program does allow for shortage definitions to vary by hospital, by blood center, by geographic region, to vary over time, and based on changes in supply network or differences in supply network between facilities.

Next slide.

The data as received currently is received and mapped at a central FDA monitoring site. As I mentioned, the registration of reporting facilities allows not only establishment of a denominator, but verification and follow-up of selected reports.

Data access is through software that puts the data actually on a map as well as into a database so that one can see on a geographic basis just what the shortages are and where they're occurring, and I'll show you some examples of that.

And the software can also be set up to establish an electronic alert message when there's a predetermined threshold that's crossed in a certain area of the country.

Next slide.

The data is anticipated to be available on a public Web site. It's maintained in an Oracle database. The mapping software provides a public view of real time color coded maps reflecting both regional and national blood supplies, and these maps actually by double clicking you can drill down to individual areas as close as the country level, but we haven't established a rule that just for privacy of individual institutional data that we can't drill down to an area that's less than five participants.

So for the pilot site obviously with nine sites, it's difficult to use the map utility, but we anticipate it will work well when there are a lot of data available.

Next slide.

So this is the user screen for Transnet. This has actually been in use by the pilot sites. It operates through the FDA server, and on a daily basis they simply input a user ID and a pin number.

Next slide.

And the first question is very simple: does your facility have a blood shortage today? This is intended for transfusion services.

If you click "no shortage," it's "thank you very much," and that's the end of the program. If there is a shortage -- next slide -- there is a drop-down list of characteristics, and for transfusion services there include delay of elective surgery, delay of emergency surgery, delay of medically necessary transfusion due to lack of supply, incomplete orders for Group O, incomplete orders for platelets, user of supplier not used typically, support of RH negative patients with Rh positive blood, use of a strategic reserve of frozen inventory, double cross-match blood, early release of cross-matched blood and "other," which is a write-in column.

Next slide.

This is a mapping function. We haven't been able to use it effectively yet because of the minimal data available, but this is what it will look like if areas are color coded from insufficient data to no shortage to serious shortage. It can be done at the first level, at the state level, and then drilled down -- next slide -- to more details, such as individual counties.

Next slide.

Now, as I'm sure all of you are aware, we had a prolonged shortage of blood both during the summertime of this year as well as through the 2003 holiday period. There were two mid-week holidays, Christmas and New Year's, which disrupted the work week.

And around the first few days of January it was recognized broadly that a blood shortage was occurring. There were appeals shortly thereafter from Secretary Thompson and industry appeals, joint appeal from American Association of Blood Banks, America's Blood Centers, and the American Red Cross.

And the recovery from this shortage has been somewhat prolonged, as well as at least in the East Coast area now complicated by both the blizzard of '03 and the quarantine of units related to white particulate matter, and these are things that -- you know, everyday stressors to the blood supply that, you know, make complex any sort of long-term blood shortage problem.

Now, the Transnet pilot study at nine sites was due to start January 6th. We actually wanted to start it with artificial data so that we could use the mapping function and not worry about the n equals five rule.

However, two days after the program started, we said, "Wait a minute. We're having a shortage here. Why no use the system to try to produce actual data?"

So we go the sites to submit two days of retrospective data and then actually collected shortage data in real time.

Next slide.

And this is just a very simplistic version of what we found, but I think it shows some important information. The green line down at the bottom are the number of sites out of the nine that actually reported shortage on a given day. This is pooled for transfusion services in blood centers, and you can see that it ranged from five up to eight, and approximately the worst time of the shortage was around the 10th of January.

And then coming down, and at the current time there are one to two blood centers that are indicating shortages, and there haven't been any transfusion reports since the 10th of February.

The red line here I think is also an important one to consider. What we did was take the characteristics from that drop-down list, provided some weighting to them, just, you know, our best estimate for a pilot study as to weighting their medical importance, and then considered those from the sites reporting shortages and considered those on a daily basis.

So not only do you see the number of sites reporting shortages in a very mild shope coming down, but you can see that clearly from these indicators the characteristics show that around the 10th to the 12th of January the shortage was at its worst.

Next slide.

Here's a more complex version of that. It simply separates out the blood centers and the transfusion services. The point to make here is that the red line which indicates the characteristics from transfusion services ends about the 4th of February, and the blood centers are still reporting shortages.

So we feel that as we suspected from the beginning that the transfusion services reported more of an acute situation. The blood center is perhaps more sensitive and would have a longer tail to an event like this, and in fact, a tail that might be almost continuous over time at some blood centers.

Next slide.

So from these data we were able to define a few characteristics of the post holiday shortage. Of 148 transfusion service shortage reporting days, 54 of them reported shortages. So that's 36 percent of the time that these transfusion services were participating that reported a shortage. The recovery was a little sooner than the blood centers.

In contrast, blood centers reported shortages 53 percent of the time. So we think blood centers are a little more sensitive monitor. Transfusion services are a better reflection of acute need.

Next slide.

So where do we go with the program? I think this will be something that's determined over the next several month.

Phase 1 was the pilot studies, which we're just completing now.

There is tremendous interest at the Department of Health and Human Services in building an integrated blood monitoring program. There will be a report of the current HHS monitoring system today and we anticipate discussions over the next month or two as to how to integrate the systems and build a single monitoring system.

We do need review by the Office of Management and Budget before we can go further. Phase 2, as I mentioned, it would probably make sense to roll out to a larger, but still limited number of participants and allow the mapping function to be developed, as well as allow monitoring at the regional level.

And then finally at an unknown date we would hope to roll it out to the full population of blood centers and transfusion services.

There's, I think, a lot of expansion potential to the program. One could use it to capture blood center inventory as well as shortages. I think that would be a useful thing to do as well. One could capture platelet supply. One could use it for adverse event reporting and supply and reagent shortages, things that are specifically important to FDA.

One could even use it so that if a transfusion service had an acute need for blood, they could use the electronic system to actually E-mail other participating sites and have those sites call if they had the supply available.

And finally, one could anticipate use of a public blood thermometer, for instance, in the newspapers to keep the need for blood constantly in the public's eye and hopefully help improve donation.

So while there are existing systems and you'll hear data from them today, I think there is some room for improvement. I anticipate that will occur over the coming year, and just finally I'll say this committee discussed over the past two years survey data related to the blood donor base that helped lead to the decisions on the variant CJD deferrals. That survey was done without funding. It was finance by the private sector and through the NHLBI REDS study for data analysis.

There wasn't much time to do it once industry was asked to conduct the survey by the committee. We're not even in that good of shape right now. Should there be, you know, recognition of a major outbreak of BSE somewhere in Asia, we don't have a clue as to the proportion of donors who might be impacted by a deferral, and the skeleton to capture those sorts of data is not even as strong as it was several years ago.

I think that's something that the committee needs to be aware of. We base, you know, so strongly decisions based on those survey data and our ability to produce those has been compromised, and it would be wise to have some discussion about a comprehensive system which would take us into the future and produce data of that sort.

Thanks very much.

CHAIRPERSON PRIOLA: Okay. Thank you, Dr. Williams.

I'll open up the committee to any questions or comments for Dr. Williams. Okay.

DR. BRACEY: I have one question. Often what we find in the hinterlands is that there may be an element of inefficient use of blood inventory. Blood inventory management is something that's not typically taught in many areas, medical technology, pathology, et cetera.

Has your group thought about trying to use this model in a way to further educate people on the most efficient mechanisms for inventory management?

DR. WILLIAMS: I think you're addressing an area that's critically needed. So far this is designed as an interactive data exchange capability. I think to the extent that one of these data collection mechanisms -- and I would look more probably toward the annual of biannual survey -- needs to establish some benchmark data that institutions can use to define how appropriate their transfusion practice is.

That's something that's sorely missing right now, and I think that would provide, you know, a good baseline to compare local institutional use of blood.

I know there are strong messages from, you know, both federal agencies and the community to make the most appropriate use of blood, but that is difficult if you don't know what the other guys around the country are doing, and your surgeons or physicians have their own, you know, way of conducting medicine and feel that's the most appropriate use of the blood product.

So there's a need for improvement there. Will this Transnet system address it? I would say probably not in the near term, but I think a large scale expansion of the survey capability, particularly tying that to blood utilization for a specific procedure would go a long way toward meeting that need.

CHAIRPERSON PRIOLA: All right. Thank you, Dr. Williams.

Next we have a series of short talks on blood supply monitoring, and the first talk is by Captain Lawrence McMurtry if he's here.

(No response.)

CHAIRPERSON PRIOLA: All right. It doesn't appear that he is. So we'll go on to the second talk, which is Dr. Celso Bianco, Vice President of the American Blood Centers.

DR. BIANCO: Thank you.

Good morning. I'm with America's Blood Centers. That is a national network of locally controlled, not-for-profit blood centers that provide nearly half of the U.S. blood supply from volunteer donors.

Collectively, we operate in 45 states and serve about half of the nation's 6,000 hospitals. America's Blood Centers' total blood collections exceeded seven million donations in 2001 and 2002.

We thank FDA for asking to comment on the impact of the implementation of extended donor referrals for vCJD participation, particularly sine October 2002.

All 75 ABC member centers in the U.S. implemented the deferrals by the deadline. Actually it should be in my write-up October 31st, not October 1st.

Surveys of ABC member centers show that from 2001 to 2002, the average percentage of donors who came in and were deferred for travel related to vCJD increased from .2 percent to .7 percent of donors registered or over 56,000 donors.

The deferral rate for individual centers in 2002 ranged from .5 to 1.1 percent. These deferred donors had donated an average of 1.9 times a year for an estimated total loss of over 107,000 donations.

More worrisome is the fact that 75 percent of the individuals deferred were present donors. They are the most dedicated and safest blood donors because they have donated many time and have been tested and found negative for known transfusion transmitted diseases in each donation they made until they were deferred for a vCJD risk.

The next slide, please.

The donation frequency of these donors is shown in this figure, and those figures are attached to the back of the paper copy.

The donor loss, there is among the individuals, the first time donors were 26 percent. The individuals that donated three times or more represent 41 percent of the donations that were lost, if we add up 63 or two thirds of the donations, and these last for the future.

These donors lost is even more pronounced among those that donate platelets. ABC surveys so that the deferred platelet phoresis donors had donated an average of 4.2 times a year. We lost almost 2,800 aphoresis platelet donors in 2002.

We did deferrals fully implemented in 2003. We expect to lose over 13,000 aphoresis donations.

As expected, vCJD related deferrals now became a substantial component among all of the deferrals.

The next slide.

In 2001, they constituted only 1.2 percent of all referrals as compared to 5.5 percent in 2002.

Figure 2 highlights the increase, particularly aphoresis donors, those in yellow.

A previous ABC survey conducted in October 2002 showed that deferral of military personnel constituted a major proportion of vCJD deferrals because military personnel and their families have been dedicated contributors to the civilian blood supply.

Some ABC centers are also seeing substantial deferrals of high school and college students because they were military dependents, were born or lived in Europe for extended periods of time.

Other blood centers have told us that they have seen significant attrition of donors at corporate blood drives with multinational companies whose employees and families routinely are posted in Europe for extended periods of time.

It should be emphasized that all of these recorded deferrals occurred at the donation site, as Dr. Williams just emphasized.

It should be emphasized that all of these recorded deferrals occurred at the donation site as Dr. Williams just emphasized in the course of pre-donation medical history. These donors showed up to donate and were sent back home. Unfortunately we are unable to measure the number of potential donors who learned about the vCJD deferral through the information provided by the blood centers or through the media and did not show up to donate, nor do they capture donors deferred during pre-screening by blood center telemarketers.

We strongly suspect that these deferrals exceed the number of donors deferred at the donation sites by several fold. A survey performed by one ABC member center in California showed that 5.3 percent of active donors were affected by the implementation of the vCJD guidance.

The impact on the blood supply of ABC member centers. We are just coming out of the worst January blood shortage we have seen in many, many years. The volunteer blood staff are already stretched by the impact of vCJD deferrals; was confronted with two weeks of holidays. Christmas and New Year's Day fell on Wednesdays, making blood drives at offices, churches and schools almost impossible.

This is clearly shown in the next figure. The next slide, please.

Taken from our Web site that can be accessed by anybody with a computer and access to the Internet, we have something that we call the Stoplight, and Dr. Williams mentioned it. ABC members every morning receive an E-mail, and that E-mail has three lines: green, yellow and red.

Green means that they have three days or more supply; yellow, that they have two days' supply; and red when they have one day or less of their supply. And they just click on the right color, and this feeds immediately into a database and updated the screen in the Web site.

And so what you can see is very much what in other ways Dr. Williams showed, is the tremendous increase -- I don't have a pointer, but off the red centers -- oh, than, you very much -- off the red centers during the period of the holidays in the first -- right following the holidays, and a substantial number of centers with three days of inventory or more.

And a lot of this is because there is normally reallocation of blood between blood centers, exchanges so that they can provide those that are being lucky or are very effective in collection. There are certain regions of the country that are much more productive, much more effective in terms of collections than other regions, and now we are coming back to a little bit more stable, but we still have days with 21, 18 centers out of our 75 centers that are under a shortage.

I suggest that you just go to our Web site and click on the little stoplight, and you can have a very good idea.

At this point I'd like to ask the committee to think that it is time to reconsider the vCJD travel related deferrals. We all share a common purpose, the availability of safe and adequate blood supply that meets patient needs. Concerns about the transmission of vCJD by transfusion of blood and blood products have led us to implement strict donor deferral policies based on the precautionary principle.

However, while we recognize that evidence about the potential for transmission of TSEs in animal models continuously developed, we still lack evidence of actual transmission among transfusion recipients in the U.K. or elsewhere, including patients that receive multiple transfusions and recipients of clotting factor concentrates.

The incubation period for reactogenic cases of CJD is, in general, less than 13 years. The predicted incubation period of vCJD based on a cluster of five cases in Queniborough in the U.K. is 15 years according to Dr. Chris Prowse that is the Director of the Scottish Blood Transfusion Service, whom I consulted before preparing this statement.

The BSE epidemic in the U.K. cattle herd peaked in 1988 by date of birth. Thus, we should have already seen at least some cases of transmission by blood at this time.

The ongoing U.K. follow-up studies of donors who later developed vCJD and recipients of blood from these donors -- next slide, please -- has not yielded a single case. They took 114 vCJD cases, 20 that had been said to be donors. Eleven of these were identified. Nine had donated and they could trace the records. Those 56 components were traced.

None of these individuals are in the CJD records, and the CJD/vCJD reporting is very good in the U.K.

On the other side of the project, there were eight reported, eight vCJD patients that were reported to have received transfusions. Four were identified. Two actually were not transfused.

There were 103 components transfused to one case, 117 components. Again, there were none of these individuals that donated the blood had developed CJD.

And finally, the next slide and the last slide, the number of cases of vCJD is defined, as shown in Figure 4. After seven years of increasingly intensive surveillance, there are 130 cases reported in the U.K. and this is data for the month of January this year.

So we believe that it is time to reconsider and relax some of the geographical vCJD deferrals. While we understand the public desire for zero risk, we believe that the precautionary principle has been inappropriately applied in this case. It is time to analyze the available data and recognize that theoretical risk has not materialized.

Is it possible that the information will change? Sure. However, in this case, we could always reinstate the strict deferrals in whatever audit managers are appropriate at that time. We have a multitude of real blood safety issues to contend with at the present time. There is a worrisome shortage of blood donors, and we are discussing that increased demand because of blood intensive procedures.

We also urgently need to address other diseases for which transfusion and transmission has been demonstrated, such as West Nile Virus. As of February 5 of 2003, this disease has caused 4,008 cases of West Nile Virus and 263 deaths in the U.S. We're close to 600,000 individuals infected.

The epidemic has occurred in a single year, not in the 20 years of BSE or seven years of vCJD. And actually we are implementing screening under a research protocol for the West Nile Virus in the entire system by the end of June, early July this year when the mosquito season starts.

Many scientists and biotechnology companies are attempting to develop a test to screen blood donors for vCJD. Unfortunately there is no test on the horizon because the levels of prion protein in blood of experimental animals are too slow for detection with available technology.

Let's accept the fact that all epidemiologic studies performed in humans show that the risk of transfusion by transmission, if any, is extremely small. Let's relax the deferrals at least partially and help patients whose lives depend on a steady supply of blood.

Thank you for the opportunity to present our point of view.

CHAIRPERSON PRIOLA: All right. Thank you, Dr. Bianco.

Are there any questions or comments from the committee for Dr. Bianco?

Dr. DeArmond.

DR. DeARMOND: What would you recommend for relaxing the deferrals? What kind of relaxation?

DR. BIANCO: It would be, for instance, to extend the living in the U.K. back to three months and maybe making the European deferral less restrictive. From five to ten years, for instance, would be one.

I would decrease the theoretical. Dr. Alan Williams and FDA did a very nice study from survey data that was presented several rounds ago that was the basis for these calculations, and they estimated that about five percent of their supply would be -- we would lose and actually we are not finding too far from that, and we could reduce it to two percent if we took some of those measures.

CHAIRPERSON PRIOLA: Okay. Dr. Hogan.

DR. HOGAN: During the crisis in early January, were there any elective surgeries that were deferred because of the blood problem?

DR. BIANCO: Yes. There were many elective surgeries in many areas of the country.

DR. HOGAN: How about emergency surgeries?

DR. BIANCO: No, emergency, I think that in the history of blood banking, blood shortages were always there. So when there is a trigger, every blood center at every hospital starts a triage.

DR. HOGAN: Right, and in terms of the elective surgeries, how many days of elective surgery were canceled? Do you know?

DR. BIANCO: I don't have the data immediately here with me, but there are estimates made by some organizations that I could try to get.

Hospitals also don't like to report. There is a dissociation between the blood center and the collecting facility and the hospital. Hospitals don't like to report that they don't have enough blood or that they are postponing elective surgeries because obviously they don't want patients to be afraid of going to the hospital because they don't have blood.

CHAIRPERSON PRIOLA: All right. Thank you, Dr. Bianco.

Our next speaker will be Dr. Peter Page from the American Red Cross.

DR. PAGE: While the slides are being set up, I can say that the American Red Cross is a single national corporation with a single FDA license that collects almost half of the blood components from volunteer donors in the United States, and we operate through a series of 36 individual blood service regions, which I'll show you on a map later on in the talk.

Today I'll give an update on a presentation I gave at this meeting a while ago, which related our experience with donor losses that we could measure related to CJ criteria.

The first, not the absolute first, but the first larger impact study of related deferrals were in March of 2000, which related to not taking as donors people who had spent six months in the U.K.

Early in the next year, early in 2001, the American Red Cross had developed some geographic donor deferral criteria based upon the concern and uncertainty about the upcoming evolution of the epidemic and stating that the policy would be reconsidered annually and as new data accrue, and that remains the case.

So early in 2001, we went to only three months' presence in U.K. and added six months' presence in Europe as criteria for deferral, and having decided this in early 2001, we implemented it in October 2001, and I'll show you later the specific effect on our deferral rate then.

A couple of months after that, in January 2002, the FDA issued final guidance, which was to take effect in two phases later in 2002, the first phase in June of 2002 and the next in November of 2002.

It was shortly after that that in the spring of 2002, the American Red Cross decided to relax its criteria regarding the presence in the U.K. after 1996. Initially we had considered any time in the U.K. sine 1980 up through the present, but then in the spring of '02, we became consistent with the FDA's finding, recommendations in their earlier guidance, and due to the effectiveness of food chain controls in the U.K., we limited the deferral to presence in U.K. up to 1996 and not thereafter.

I have a long, complicated table that demonstrates the differences that have evolved in time over the deferral criteria by the FDA, as well as by the Red Cross, and their differences, and then some very minor differences that the U.S. Department of Defense has in its blood program. And then also to point out that Hemo Quebec is different yet, and that is different yet than the rest of Canada.

I have some slides that I'd really like to show you that --

PARTICIPANT: They're coming up.

DR. PAGE: Okay, good. I can skip over the early ones when we get to it, then -- that relate our donor deferral percents and our blood collections versus goal and then our blood inventory and distribution.

As has been stated by others, and I'm emphasizing the donor deferral data that I will present represent only those volunteers who came to the blood mobile, showed up, registered, answered the question yes, and were turned away. It does not include those members of the public and potential or previous donors who have heard through the media or friends who have donated of the traveler restrictions and don't come at all because those we can't count.

When the Red Cross in October of 2001 went from six months in U.K. as its primary deferral to just three months in U.K and added six months' presence in the rest of Europe, our deferral rate went up about a percent, and there's a slide that shows how it went up basically a percent the day that we implemented that.

Okay. Let's go through a bunch of next slides. Go through one, two, three, four, five, six, seven. Keep going. Stop.

This is the daily deferral rate for the month of October 2001. We implemented our deferrals, and as I said, we went from six months U.K. deferral to only three months and added six months for Europe, and the previous deferral rate was less than a tenth of a percent, and it went up to about a percent right away.

The next slide shows this same data by week and goes up through almost the end of 2002. So you see the abrupt increase of a percent in the middle of October '01, but you see that it has drifted downward, and it is about .38 percent. This is what we call the culling effect. The repeat donors who were deferred who know not to come back, and the spread amongst the public of the word and the people not coming in the first place.

It doesn't mean that there's a -- the people that are lost remain lost and it makes it more challenging and expensive for us to recruit new donors as replacements.

The next slide is a map of the United States. The gray areas Red Cross does not traditionally collect blood in, but each of the numbers in the colored areas represent one of our 36 regions in which we collect blood.

The next slide shows the deferral rate for each of those 36 regions and represent the period in March of '01 when they were only deferring for six months in the U.K., and then the back or higher (phonetic) bar is after October 15th, where I showed you the dramatic one percent increase, and you can see the variation in percents amongst each of the regions is substantial. There's a two and a half-fold or more difference amongst regions.

The next slide follows this into the future, and the high deferral rate at the end of October '01 is in the back, and as you move towards the front of the slide, we move to March of '02 and then the last six months of last year to show that the rate has continued to come down, but even as it comes down, there remains a several-fold variation among each of our regions in the measurable deferral rate.

Next slide.

This is just a graph showing how the rate has evolved over time, showing the range. Currently the range is .08 percent to .8 percent, with an average of .38 percent. That's for the last half of calendar '02, and that averages about 2,000 donors per month that come and are turned away from blood donation in the Red Cross system.

The next slide shows the deferral rate for all reasons, not just CJ, and it goes on the top from January 2000 on the left up through December '02 at the far right.

This increase in August of 2000 is a three to four percent increase, which is about the limit of elasticity that the system can tolerate, as Alan Williams mentioned earlier, and it is due for us changing from ear lobe to fingertip as to the source of the drop of blood to assess a donor's hemoglobin to make sure they have enough of an hematocrit to donate blood.

That still remains the largest reason for donor deferral. You'll see that the overall deferral is about 12 percent.

This is October of '01, where there was the one percent increase in deferrals related to CJ reasons, that then drifted down a little bit. So now the percent deferral remains about 12 percent. In the scheme of things it's minor, but remember we don't know the people who don't show up.

The bottom half shows our budgeted collections in blue or our recruitment goal or plan and in red our actual collections by month, and you can see that we were bouncing around a couple of months a little better, a month around Christmas in '00, that we were lower, and this is the September 11th effect where there was an increase in donations followed by a decrease in donations.

Subsequently we have, in general, been behind goal. I cannot attribute that all, and I cannot tell you how much of it, but a little piece at least is attributable to CJ. Alan Williams mentioned some issues that relate to the economy and the availability of donors and motivation.

So we have been below goal in this period as opposed to the earlier one, and as I say, I cannot tell you how much might be related to CJD, but at least some.

Next slide.

This is our distribution on the top and our inventory on the bottom. This is average daily distribution by month. You see that in September-October our distribution was down. There was plenty of blood. Hospitals that had donor programs were collecting a lot more than usual, and therefore, were ordering less and less from blood centers, and then you can see as that effect went away and donations were down, our distribution to meet patient needs increased.

But distribution has remained with monthly variations bouncing around about what it was before, maybe a little less.

Our average daily inventory of red cells by month peaked up at 300,000 after September 11th, and then dropped down. At the end of December it was about a total of 100,000. I can tell you more recently it has been much lower. It has been down to 60,000 or less.

We've been impacted by the holiday and post holiday shortage, by the storms, and as was mentioned, by withdrawal for another reason in some centers related to white particulate material.

Next slide.

This is my last slide and says that when anticipated and planned for, new donor deferral criteria should be accommodated by increased donor recruitment and collection efforts, all be expensive. It has become ever more challenging to recruit and expensive in hiring staff and advertising and particularly expensive for the collection staff that may need to work additional hours, have more hired and be trained, and collecting the last units of blood are much more expensive than collecting the first units of blood for any daily goal.

After the increase in deferrals from a new measure, there's a culling and the deferral rate for that reason drops. We have seen that with any number of new deferral criteria and new tests that we have implemented because the affected people don't come back. So deferral for TSE is one of several recently added deferral criteria, each of which increases the costs of recruitment and collection. And with West Nile there's others on the horizon.

The bottom line is I think all of our speakers agree, is that more regular, volunteer blood donations continue to be needed to prevent the seasonal shortages and develop a reserve for unanticipated occurrences, and we appreciate the HHS and federal government's encouragement and support in those regards.

Thank you.

CHAIRPERSON PRIOLA: Any questions from the committee for Dr. Page? Dr. DeArmond.

DR. DeARMOND: The deferrals that you talk about, these are people that showed up as before to donate blood and were told they couldn't donate.

DR. PAGE: That's correct.

DR. DeARMOND: So of those, the CJD component is a very minor component of all the reasons.

DR. PAGE: That's correct.

DR. DeARMOND: So do you have an assessment of how variant CJD threat has affected the entire system? Because that's a different measure.

DR. PAGE: That's something that I think we're all frustrated by not being able to measure because there's word of mouth and public information, and when we go to a company or a school to collect blood, we have the Bloodmobile sponsor who is usually an employee at the plant made aware of who's eligible or not so that they only sign up people who are eligible.

So there's a lot of pre-education or pre-screening before people come to show up, and we don't have a measure or we don't have a good measure of that. I think that's one of the reasons that as Alan Williams mentioned in the past there have been surveys in advance to try and assess the percentage of the population that would be affected and then our actual experience that we can measure is less than that, and I think it's because of the people who have learned and don't come.

DR. DeARMOND: Would you agree with the previous speaker that the deferral criteria should be made less restrictive?

I say that because the American Red Cross, of course, had the most restrictive criteria.

DR. PAGE: Yeah, we had the most restrictive, and we moved and became -- we decided to become the same as the FDA's recommendations and others for U.K. We examined the issue about rest of Europe where we remain more restrictive, and in our examination last spring and summer, with advice from people in Europe with more direct experience, we were advised not to change for the moment because we asked the question if we did change, would that mean that the risk would not change, and they were unwilling to say that because there was concern that there might be even more BSE in Europe than had been appreciated because of under reporting, no active surveillance.

We remain committed to looking at that again. A year later is coming up in the next couple of months. I'm glad to listen today to see what we can learn. There's a meeting in Canada next month that we'll go to as well, and if there's a basis to become less restrictive that we're comfortable with, we're certainly open to further modifications.

CHAIRPERSON PRIOLA: Dr. Bolton.

DR. BOLTON: How would you characterize your geographic distribution in terms of the total number of units collected? Are you affected by this sort of bi-coastal expected increase in deferrals or not? Are you more central-Midwestern?

DR. PAGE: Let's not go back on the slides, but on the map where I have each region and if you compare that to the higher rates, Alabama is one of the highest, and there's no obvious pattern of the major metropolitan coastal areas on either coast or each coast is reliably and consistently having a higher deferral rate.

Some say that that's due to military bases, and others say that's due to better pre-education and people not coming in the first place. I tried to make sense of it, and I couldn't.

CHAIRPERSON PRIOLA: Dr. Bias.

MR. BIAS: I was wondering if you have had much success in recruiting new donor bases, let's say, at the high school/young adult level and what impact that's had overall.

DR. PAGE: Traditionally about 20 to 25 percent of donations are from high school and college students, and they're a big piece and important piece of the current and future blood supply.

In the last year or two there have been additional initiatives in a variety of colleges. It's an area that we have, I think, all pushed and would look to do more.

Students are a good source. They are -- they donate quit well. I think our challenge has been more in industry and in the work sector that has been negatively affected recently.

CHAIRPERSON PRIOLA: Go ahead.

DR. BRACEY: Some say that the ability to recruit blood donors is highly dependent upon the amount of funds allocated for that activity. Noting your chronic shortfall, I'm just wondering: has there been a change in the budget for donor recruitment? And if so, has there still be absence of an effect so that the adage of increasing the budget to increase the number of donors may not necessarily be true?

DR. PAGE: I don't have specific budget numbers in my head, but I can tell you that it is a major impetus and expectation from each region to recruit and collect more, and the staffing level and allocation of consultants and computer programs at our national headquarters has increased in recent years.

It's not an automatic. Pouring money doesn't necessarily result in more donors, but that is an area that has not been cut, and the decree to which the funding has been increased I can get back to you on later. It's a fair question.

I think though that for the stability of a blood center, you don't want to cut from your donor recruitment budget because financially that's only going to hurt you and the patients in the end, and even as there have been financial challenges, that's not been a place that's suffered.

CHAIRPERSON PRIOLA: Dr. Gambetti.

DR. GAMBETTI: You said that it is difficult really to know why you have when you have new rules for deferrals whether, you know, people do not come to donate because of any specific reasons. It's very difficult to attribute the reason.

Do you know that people do not come because of that or could it be that above the deferral there is a trend to donate less that may confuse the issue of the deferral and contribute to the shortage?

In other words, is there the possibility that we are discussing the negative effect of deferral, but actually what we see is a general trend to donate less?

DR. PAGE: I think that with the aging of the population, there's a relative trend for lower percent people being eligible or able. There have been studies by a number. We've looked at why people don't donate.

There is a group that won't donate no matter what you do. There's a group that won't donate because they clearly are permanently deferred. There's a group that will only donate if we can them, and there's a group that will donate only if we come to their front door and make things more convenient for them.

I think it's an area of continuing interest to find out what other groups could be brought in to donate who are willing, but we need to figure out how to get the blood donor and the collection activity together.

That's an area that Bob Jones has worked with industry and the FDA and just wrote about in the recent Transfusion about an initiative to see who it is that's not coming and why and what we need to do to get them in. That's an important initiative that we support.

I don't have data as to whether it's increasing. I mean, there's anecdotes, and there are those who will say that in the post 9/11 era with publicity about outdating rates, for example, members of the public were turned off and didn't donate. Hopefully that's not a long-term effect.

DR. GAMBETTI: Well, yeah, I think that in order to make a decision it would be certainly useful to know what is the contribution of these two factors, deferral and less people offering to donate, fewer people.

DR. PAGE: We agree. Thank you.

CHAIRPERSON PRIOLA: Dr. DeArmond.

DR. DeARMOND: There hasn't been any example of a transfusion transmitted CJD in Great Britain over the, I guess, what, last 15 years or as long as they've been watching it? How do you evaluate that data? Do you still think that there's going to be a sudden explosion of transmission related CJD or just how do you view that and how do you assess the negative data in terms of risk?

DR. PAGE: Well, there's negative data in the United States as well, with a CJ look-back period, which the last I checked, which was not that long ago, remained negative, and now with years passing there are more of those recipients that have lived longer after their transfusion, which is what we need.

Many of them had not lived that long after their transfusion and were transfused more recently. I think that is one of the things that we will look at again more formally in the spring. The concern is that the incubation period can be quite long even if the risk is quite small. You might not see it in 100 or so cases.

With time there should be relatively more reassurance, but I think we're all concerned about the safety for patients, and after the experience in the '80s, the public expects us to do too much too soon rather than too little too late.

Now, I think as data accrues, we ought to continue to look at this, and if we overreacted, we should be willing to reexamine as Celso pointed out. Red Cross isn't prepared to make a change today, but is going to be actively pursuing more data and information and suggestions in the next few months.

DR. DeARMOND: Just following up on that, the HIV problem with transfusions and the prion problem with transfusions, the CJD problem, seem to be two different, apples and oranges, as one group would say. Clearly HIV transmitted very rapidly, and it was a clear connection between individuals who received transfusions and the fact that they got an HIV contaminated blood product.

With CJD that isn't clear at all, and even incubation times that they say they can extend out to 40 years, those are the extremes. Most of the incubation times are in a much tighter region of time.

And so even if there is a 40 year incubation time, you would expect it to be very, very small numbers. Do you factor that in that we're talking about two different diseases essentially, two different etiologies, two different mechanisms?

DR. PAGE: I certainly acknowledge that they are different diseases, and we're frustrated by lack of a test and lack of information regarding CJD. I think some of the animal experiments which I'm not qualified to speak of in detail have suggested that there's more than a theoretical risk, and I think that's something that we have to take into account.

We are not unwilling to relax our criteria. We just want to be very thoughtful and have a justification for doing it when we do.

CHAIRPERSON PRIOLA: Dr. Bracey.

DR. BRACEY: Do you have any information about the ability to apply the rules? That is, the number of donors that for one reason or another have passed through the system and donated when they are at risk?

The question is really kind of focused on how well is the system working, the screening system working, within your system.

DR. PAGE: It's not perfect. There are a fair number of examples of donors who donate once and then come back again eight weeks later and are asked the same questions and only on the second occasion will they remember or admit travel to Europe or U.K., and so on the second occasion they are deferred, and then we get what we call post donation information related to the earlier unit, which we'll then withdraw if it remains in in dated inventory.

So it's clear that we ask donors a lot of questions, and a lot of them are very complicated with differing periods of times, and honest donors may not remember the specifics on one occasion but remember it on another.

I believe the FDA gets a large number of such reports from us and from other blood centers as well.

CHAIRPERSON PRIOLA: Okay. Thank you, Dr. Page.

DR. PAGE: Thank you.

CHAIRPERSON PRIOLA: Our next speaker will be Dr. Robert Jones from the New York Blood Center.

DR. DAVEY: Thank you.

I'm actually not Dr. Robert Jones, who can't be here today. I'm Dr. Richard Davey. I'm the Chief Medical Officer of the New York Blood Center, and I'm here to share with you our thoughts from the New York Blood Center about these issues.

The New York Blood Center is the nation's largest urban blood center. We serve a metropolitan area with over 200 hospitals in the New York-New Jersey area, encompassing a population of 20 million people.

Over the past two years we've presented to you our perspective and experience on the controversial issue of blood donor deferrals as a precautionary measure for the possible risk of transmitting vCJD.

Today we present our experience to date and our recommendations to deal with the difficulties produced by the current policies. I'll get to that slide in a bit.

First, I'd like to, again, review with you what you've already heard, and that is we feel the American supply is under severe stress. Hardly a month goes by without a new issue that erodes public confidence in the blood supply and diminishes the donor base.

In the past few months, for example, we have had to deal with the West Nile Virus quarantines and the impending testing that's going to come up this summer. We've had to deal with smallpox vaccinations and deferrals for all of those that will have the vaccinations for smallpox. And we've had to deal with these elusive white parties that you've heard about from other speakers.

Now, this has been compounded, we feel, by three other major negative forces that have affected the blood supply nationally in the New York metropolitan area specifically. Those are the aftermath of the terrorist attacks in November and September of '01; the resultant poor economy; and most importantly, we feel, the loss of blood donations in imports to New York due to the precautionary vCJD blood donor deferrals.

These forces have combined to result in a national blood shortage in January that you've heard about that was very, very severe. Emergency appeals were common throughout the country. Blood types, such as O negative, were especially short, often less than one day's supply.

The New York Blood Center was able to fill only 50 percent of O negative orders. Many O negative patients have been given O positive blood due to these shortages, more than I've ever seen in my professional life. This is not good medicine.

In many other areas elective surgeries have been canceled. The recent snow storm has placed the nation in another precarious situation regarding the blood supply.

Again, just as an aside, today in New York all standing orders for platelets are canceled. Only emergency release for platelets are now in place.

We've lost a number of donors for a number of reasons, but certainly vCJD is a big one, and right now our hospitals, like Memorial Sloane Kettering, Cornell University, Columbia and all of the others are working very hard to triage platelets which are not very -- are a product, as you know, which cannot be very easily managed in terms of what the patient requirements are. Patients need platelets every day.

And one of the outcomes of this, I think, is hospitals are now, at least in our area, being forced in many ways to begin collecting on their sown, sine the New York Blood Center, due to our restrictions, were not able to always fulfill their needs. This is not good.

We had a very stable situation in New York. Many of our hospitals now are being forced to look elsewhere or draw their own blood. That implies quality assurance issues. They have to go through regulatory issues. I feel that the best way to manage a community blood supply is to have a strong community blood center, as we have had in New York City up to the recent time.

Well, nationally there have been certainly severe blood shortages and depressed donations. As I've mentioned, the impact of these forces has been especially severe in New York. We've made extraordinary efforts to increase donations and to import blood from other domestic supplies, and we thank those that have helped the New York Blood Center and the people of New York.

However, the impact of the reduced donations continues. The figure now that's on the screen and attached to the handout I think you may have seen from some presentations by Dr. Jones before to this committee, but this is an update. It shows the monthly blood collections at the New York Blood Center since March of '99.

I think as you can see, there was a long period of consistent, stable growth which we were quite proud of, which led up to September 2001 when blood donations surged following the disaster and then descended into a trough that continued into May of last year.

Between the drop-off in donations and the discards of units that were not utilized after the disaster surge, we now calculate a net loss to our available supply of over 21,000 units, or about five percent of our annual whole blood collections.

The economic impact of the disaster has left many of our corporate donors in the New York City area downsizing and low company morale. We feel it has had a resultant reduction impact on our blood donations.

You can take off the slide now.

So we feel we must make prudent decisions regarding the blood safety and the deferral of blood donors. We all, I think, agree that potential threats to the blood supply require our very careful attention.

Regarding vCJD, it seems appropriate to apply the precautionary principle to potential donors who have lived and resided in the United Kingdom. Ninety-eight percent of both bovine and human disease has occurred in that country.

Actually, when I was a consultant to this committee I think we discussed putting a bright line around the U.K., and I think that still applies.

So while the appropriate application of the precautionary principles is a worthy endeavor, creeping precautionism must be recognized and avoided. We ask this committee to review and continually reassess the rationale for exclusion of donors who have visited or resided in European countries other than the United Kingdom, and just to respond to some earlier questions, we would recommend that the U.K. restriction remain the same. That is where this unfortunate human experiment is playing out.

But we do feel that the committee should reassess the extension of these restrictions to other countries other than the United Kingdom.

There's been no link between donors, as you've seen, who subsequently developed the vCJD and recipients of their blood, and millions of transfusions have occurred in the U.K. and in other European countries since the start of the epidemic with no link to the development of disease.

As Dr. Bianco has pointed out, the projections for the extent of the epidemic have decreased, and the number of new cases has remained stable or perhaps even decreased.

The national loss of donors because of these deferrals is substantial and ongoing. The impact on New York remains severe. Aside from the loss of European imports, the New York Blood Center continues to defer one percent of our donors each month because of these restrictions. That seems like a small percentage, but it's every month one percent are lost.

These donors are lost, quote, indefinitely, and at present will never be allowed to donate again. Many, as you've heard, are repeat donors.

The many donors who have self-deferred remain uncounted. However, we do believe that these self-deferrals were substantial. Given the publicity regarding the deferrals and the clear drop in collections, it coincided with the implementation of each phase of the deferrals.

In fairness to those who make the difficult decisions to protect the safety of the blood supply, public expectations, political logic and, I think, a clear understanding of some of the medical uncertainties behind these policies and the implementation of these policies is understandable.

These decisions presume, however, that the blood supply is an elastic resource that can expand to meet the public need. What we have learned in the past few years is that the supply is not as elastic as necessary to adapt to policies of this impact. Patient care has been jeopardized, and as a medical care provider, this is where the medical logic of these policies falls short.

We urge examination of the vCJD deferral policy against the realities of risk to patients. We feel that the status of the current and future blood supply and the ability to adapt to new restrictions, such as West Nile, present a greater danger to the population of patients in need of transfusion than the possible transmission of vCJD via transfusion.

We urge this committee and the FDA to regularly examine the policies regarding vCJD restrictions as noted in the guidance issued pertaining to this issue. We ask that these deferrals be balanced against the recurrent and increasing stresses in the blood supply and the resultant impact on patient care.

I thank the committee for the opportunity to present our views on this important subject.

Thank you.

CHAIRPERSON PRIOLA: Questions or comments from the committee for Dr. Davey?

I apologize for not identifying you correctly.

Dr. Bias.

MR. BIAS: Do you know if the restrictions were rolled back for those outside of the U.K.? Do you know what percentage of donations that might result in for you?

DR. DAVEY: Well, I think we can perhaps look at our imports for starters. If there were any kind of reinstitution of the ability to bring blood into the United States from Europe, that would have a major impact on New York because for 30 years we relied heavily on that.

We're not ready to go to that stuff again, but that would obviously be a major increase in the blood supply in the metropolitan area.

In terms of how many we would gain from domestic deferrals, if you will, I don't have those data exactly. It would be substantial.

MR. BIAS: I'd love to see some of the figures sometime.

DR. DAVEY: Yes, and I think your question was U.K. versus all other countries.

MR. BIAS: Right.

DR. DAVEY: I think we defer many from, quote, all other countries. I can find those figures for you later.

MR. BIAS: Good.

CHAIRPERSON PRIOLA: Dr. Bolton.

DR. BOLTON: Obviously the loss of the Euroblood supply was a big impact. If those restrictions were relaxed, is that source of supply readily available again or was that sort of ramped down and lost?

DR. DAVEY: Well, clearly we've had to make major adjustments in our relationship with our European suppliers, and if we were to consider reestablishing that relationship, that would take a ramping up of that relationship.

We were on a trend to lower our reliance on European imports even before these restrictions were in place, and over the next several years, our plans were to go to total self-sufficiency in New York.

However, the loss of the European imports have had, obviously, a major impact, one of which we've lost considerable market share in the New York area because our prices have had to go up substantially.

We've relied heavily on domestic imports. We're working hard to work on our own local collections, and we hope that that would be the ultimate solution.

However, if we had the option to reestablish a connection with our European partners, we would certainly explore that.

CHAIRPERSON PRIOLA: Dr. Williams, you had a comment?

DR. WILLIAMS: Just to respond to the previous question about the proportion of losses from non-U.K. endemic areas. The deferral basically had three major components: the U.K. component, those who were potentially exposed to beef from Britain on military bases, and the non-U.K. European component. Nationwide, of the total estimate of five percent, between two and three percent would be attributable to the non-U.K. European component, but obviously this would vary a lot with the demographics and geographic characteristics of the donor base.

CHAIRPERSON PRIOLA: Dr. Hogan.

DR. HOGAN: Do you have any data on the morbidity and mortality of this blood shortage? That is, you've told us a lot about elective surgeries, et cetera. What about impact on patients? And specifically, has anyone died because of lack of blood? Have leukemic patients bled out because they don't have platelets or strokes occurred because of anemia and they didn't get transfusions?

Do you have any data on that?

DR. DAVEY: Well, thing that I've been concerned about that I mentioned, just for an initial comment, is the number of O negative patients that are getting O positive blood. While this is not immediately life threatening, this is a medical practice that we would clearly want to avoid. It sensitizes patients. It restricts them from getting properly type blood in the future, and we've had to had -- we've had to on many instances on a weekly basis use O positive blood to O negative patients because of the shortage of O negative blood.

Now, at the moment our platelet situation is of quite a concern in New York where, like I say, no standing orders are being filled. Emergency orders only.

Has this resulted in clear morbidity or mortality? I think as some of the other commenters may have mentioned, it's very hard to get our hands around that because hospitals are reluctant to be entirely open with what their situations are in their respective institutions.

I'm not aware of any specific patient direct mortality or morbidity because of this.

DR. HOGAN: The reason I'm asking that is because in some way to gather that information it would certainly help answer the question for this committee of what the effect of the decision has been. If we have to balance the disease that only has at maximum 28 people a year dying in Britain versus significant mortality or morbidity in this country, it might be important.

DR. DAVEY: Yes. I think maybe on the broader issue to your point while we have had elective surgery canceled, we've had O negative patients getting O positive blood. We know that. I think in the larger picture, we are a nation under considerable stress in terms of the blood supply. That stress is multi-factorial: West Nile Virus, white particles, smallpox vaccinations.

The vCJD deferrals are clearly a part of that. In New York, we feel a major part. All of this is impacting, I think, on a blood supply that is highly volatile and from a day-to-day, week-to-week, month-to-month basis is really not very predictable in terms of our hospitals' customers. They're very unsure on what they're going to be able to get and how they're going to be able to manage their patients.

I think that's the main crux of the issue. It would be much better if our hospital customers could rely on the blood center to provide blood on a stable, daily, regular basis so that this material is available for their patients without concern of its fluctuating availability.

CHAIRPERSON PRIOLA: Dr. Belay.

DR. BELAY: Yes. Obviously one of the ways to deal with this issue is increasing recruitment, and I was wondering if we have done everything that we can. Is there any room for improvement for our recruitment efforts? Are we running out of eligible donors? I mean, have we a shortage of money, or is there any room for improvement for our recruitment efforts?

DR. DAVEY: Well, there's always room for improvement. We're looking always for new ways to attract donors. As a matter of fact, we've come across, I think, a very innovative way in our blood center where we now are approaching jury rooms. We're aware of people sitting around not doing anything where we can approach them to donate blood.

(Laughter.)

DR. DAVEY: And so we've had very successful efforts now on Long Island at the courthouse where our blood drives now are very successful.

So these are innovative ways to look at getting blood donors that pay off. We're working very closely with New York State to have a very focused attention on high schools. We're also looking to even work with New York State, as a few other states have done, to lower the age for blood donation from 17 to 16. We feel the data are now substantial enough that this is a safe thing to do, but this is a regulatory issue that has to be dealt with with the state.

In New York we also have a program you may be aware of where we are screening donors for the hemochromatosis gene, looking for folks who might be at some risk for that disease, where a regular blood donation focus could be healthy and therapeutic to them and would add to the community blood supply.

So I think the answer to your question is yes. We're doing all we can to look at innovative ways to increase the blood supply. However, the losses so far are exceeding our efforts to replace them.

CHAIRPERSON PRIOLA: Dr. Bracey.

DR. BRACEY: Yeah, one question about your relationship with the Euroblood group. In essence the demographics of the donors relates to the genetics, relates to the distribution of the blood groups, and we have a problem with Group O, both negative and positive, and I was wondering with that organization was there an exclusive Group O import or is it mixed? And what percent were the types?

DR. DAVEY: Actually that's a great question, Art. For background, the New York area especially, I think, and in other areas of the U.S. we do have A, B, O imbalance. We tend to collect from ethnic groups, primarily Caucasians, that are a higher percentage of Group A. We transfuse to groups in New York, African Americans, Hispanics, that have a higher percentage of Group O. So we always have this imbalance of not enough O and too many A. This is a problem for us that we have to deal with.

The Euroblood in a way helped us quite a bit on that because we were able to import blood that was enriched in Group O. It wasn't entirely Group O, but it was enriched in Group O. So we could pass on this A, B, O -- this A, B, O importing percentage was able to compensate for the A, B, O imbalance that we had in our demographics.

We've lost that ability. So now we have to target especially Group O donors, and we're doing that as best we can, but this has, again, limited success.

CHAIRPERSON PRIOLA: Dr. Wolfe.

DR. WOLFE: We were told at our last meeting by someone from the New York State Health Department that along the lines of captive audiences, such as juries, that there was some effort being made to use the driver registration process to recruit more blood donors. I'm just wondering whether or not there are any more data on that for New York and New York State, and secondly, whether other jurisdictions have tried that.

It would seem that this is a typical example of a large group of people who predictably have to come in every year for something when they ‑- or every three or four years, and it just seems that there are a lot of examples like this that have not been, to my knowledge, fully exploited in terms of trying to get more recruitment.

A, do you have more information? I would certainly, when we have some time for discussion afterwards, ask other people whether this has been rolled out nationwide, and I would bet very highly that it works very well.

DR. DAVEY: We've not done that in New York. I think any opportunity we have to approach certain population groups and have access to population groups specifically like drivers, we would explore.

However, I think we have to be very careful, and I think some of the jurisdictions we deal with are very reluctant to share this information because of issues of confidentiality and access to public information, to information that is not generally available to the public.

So we have not done this. Maybe Dr. Bianco could comment on whether he's aware of other blood centers.

DR. BIANCO: There are some blood centers around the country that do set up blood drives at motor vehicle facilities. It's not a very common practice. It depends very much on the demography, the type of people that come, and one of the intents -- and that's the reason, Dr. Wolfe -- is that we want to make a donor, a first time donor a regular donor, and so the effort has to be a continuous effort, and in some of those environments that is not an easy task, but is the follow-up of those because about 50, 60 percent of first time donors never come back.

DR. DAVEY: So do any blood centers use driver's license records to recruit?

DR. BIANCO: No, no. They don't use. The only thing that they use is access to the facility, at the facility where they have. The driver's license records, in general, are not available.

CHAIRPERSON PRIOLA: Okay. One last question from Dr. DeArmond.

DR. DeARMOND: This is kind of a generic statement. Maybe it's a silly statement, but the impression we get this committee is that between the variant CJD, CJD problems, travel to Europe, white particles, HIV, hepatitis, any number of others, West Nile Fever --

PARTICIPANT: Ear lobe versus fingers.

DR. DeARMOND: -- is only about five percent of the population potentially eligible to give blood in the United States? What proportion of the population is eligible?

DR. DAVEY: Well, that's a great question. It's one that concerns us all, not only the current issues, which seem to have come in a flurry, West Nile, smallpox, white particles, vCJD, but we have other things coming down the line, perhaps Chagas disease screening, Babesiosis, other issues that are going to impact on the blood supply. Bacteria in platelets is a huge issue that we're now addressing very actively in New York and across the nation.

All of these things are eroding the donor base. All of these things are impacting on the number of people that can donate blood. For those of you who have donated blood, and I hope most of you have, if you can, realize the inquisition that you have to go through to donate blood these days. It's a long list of questions, and then a very substantial battery of tests on your blood.

That's good. Most of these things are clearly indicated and have resulted in, we feel, the safest blood supply in the world. However, we have to balance these restrictions and these new impacts on the donor base, which is eroding.

In New York, only about three percent of the population donates blood. Nationwide it's a little higher, four or five percent. I believe, and others can maybe correct me, I believe that the number of eligible people, if you took 100 people in the U.S. that could donate with all of the restrictions is now, I believe, about 50 or 60 percent. Maybe it's a little lower than that, roughly.

So we're already excluding roughly half of the population, and much of the other half for reasons that you've heard earlier for one reason or another don't show up. So we're working on a very small percentage of very dedicated donors, and we're very concerned, as you've heard from me and others, that whenever we lose a donor, we don't lose one donation. We lose the donations for the rest of their life, especially if they're already repeat donors, which are the safest donors that we have.

So we have concerns here. It's a complex issue, and we think that CJD is certainly a part of it.

CHAIRPERSON PRIOLA: Thank you, Dr. Davey.

Should we move on to Dr. Kay Gregory from the American Association of Blood Banks?

DR. GREGORY: Thank you.

The American Association of Blood Banks is the professional society for over 8,000 individuals involved in blood banking and transfusion medicine, and represents approximately 2,000 institutional members, including blood collection centers, hospital based blood banks, and transfusion services as they collect, process, distribute and transfuse blood and blood components and hematopoietic stem cells.

Our members are responsible for virtually all of the blood collected and more than 80 percent of the blood that's transfused in this country. For over 50 years, the AABB's highest priority has been to maintain and enhance the safety and availability of the nation's blood supply.

On June 27th of last year, the AABB delivered a presentation before this advisory committee stating its belief that lack of appropriate data is a major barrier to determining the state of the blood supply in the United States.

Although public health experts and the transfusion medicine community recognize the need for such data and committees such as this one routinely ask for information, no agency is willing to fund a comprehensive, ongoing collection and analysis of data by an independent, non-government organization.

In particular, this committee has regularly mentioned the lack of data about travel to various countries and expressed concern about being able to predict the effect of expanding vCJD travel deferrals to other countries on the availability of the blood supply.

This is but one example of the need for data that AABB certainly agrees is necessary, but for which no funding is available. It seems that everyone wants data, but no one is willing to fund it.

This committee asked for information concerning the effect of the new vCJD deferral criteria that went into effect on October 31st in 2002. Quantifying the effects of the new vCJD deferral is next to impossible, as you've already heard.

The same difficulties in measuring the effect of new donor policies that were discussed during previous meetings of this committee with regard to vCJD deferrals also apply to this new application.

That is, it is possible to measure how many donors appear at the blood center and are deferred because of the vCJD criteria. However, we cannot measure how many donors self-defer because of the advanced publicity, including significant effects on the part of many blood centers to notify donors of the change so that they wouldn't show up and have to go home.

The committee should also keep in mind that the number of deferred donors does not equate to the number of blood components that are lost. Aphoresis donors can donate more often than whole blood donors and can donate multiple components. Deferral of such donors increases the number of blood components that cannot be collected.

You also heard this morning about a disproportionate geographical impact of the deferrals.

In addition, because there's no systematic, scientifically valid, routine collection of data concerning supply and usage, there is no established baseline, and thus, it is impossible to measure the effect of policy changes, such as the new vCJD deferral criteria.

The AABB encourages this committee to define data that it believes are essential for making sound policy decisions and to vigorously urge government funding of data collected by an independent, non-government organization.

Thank you.

CHAIRPERSON PRIOLA: Thank you.

Are there any questions or comments? Dr. Bolton.

DR. BOLTON: I maybe surprise myself for asking this question since I'm genuinely in support of government funding of lots of things, but I'm just curious as to why it would be the government's responsibility to fund this as opposed to the industry's.

DR. GREGORY: I think, quite frankly, the industry doesn't have the money to fund it. You know, we do fund as much data collection as we can, and we're more than willing to do it, but I can tell you that the National Data Blood Center that Alan referred to in his presentation is in great financial straits.

The AABB has willingly bailed it out for a little while, but probably cannot continue to bail it out indefinitely. So what we probably need is a partnership between the industry and the government, but right now since you're a government committee, we're asking the government.

(Laughter.)

CHAIRPERSON PRIOLA: Okay. Thank you, Ms. Gregory.

Our final speaker is Captain Lawrence McMurtry from the Department of Health and Human Services.

CAPT. McMURTRY: We've saved the least for last here.

Bill, where am I? Ah, thank you.

As the title said, we're going to talk about the use of monitoring community and hospital sites. This is the -- let me back up.

First of all, I apologize for being late and missing the earlier presentations. I don't know if this has been referred to or not by any of the other presenters, but this is the Department of Health and Human Services' monitoring program where we have a series of sentinel sites spread more or less generally across the country collecting hard data on the blood supply that are in those units.

We've broken them apart into community sites and hospital sites. We have three community sites. The rest of our sites are hospital sites.

In any event, this is data from February 1, 2002 through -- I'm going to say through yesterday morning, although you'll see over here what we call in our office the drool. Because of the weekend and because of the snow in the Northeast, we have several hospitals, several sites that have not reported. So all of this right in here is an artifact of tardy reporting.

The thing that we feel is important with this is that this does show right in here the summer slump that everybody expects to see and right here is the holiday shortage that everybody expects to see.

Yeah, let me have the next slide.

This is total red cells from the hospitals. Once again, there is a summer shortage, and here is the holiday shortage.

I need to explain this right here to you. We have had, because of some contractual issues, we've lost a couple of hospitals temporarily. I say "contractual issues." We have one woman who reports to us that's out on maternity leave. So we're going to get her data back, and then we have another hospital that we're having a contractual issue with, and so this shortage here is, as I said, a result of two hospitals that are not reporting.

This slide, once again, shows a summer shortage, a summer slump and a holiday shortage, the same trend as we've seen before and the others.

Let me have the next one.

Same thing here. This shortage once again is the holiday shortage. The reason it doesn't come back up still, the two hospitals that are not reporting.

Next.

O positive shows the same thing. O positive for the hospitals, once again, shows the same thing.

This is a slide that is not hard data. These are our -- this is our anecdote slide. Here we have hospitals that are giving us verbal reports, comments. This reflects the summer shortage. This reflects the summer -- I mean the summer slump here, holiday shortage here. And you'll see that even though we have hospitals that didn't report, and you saw the fall-off, this is what is critical right here. It's that our comments, our shortage comments fell off after this spike.

So I'd like to have a big grand finale and say this really means something. It's not that it doesn't mean anything. It just reflects what is going on out there with the blood supply as it's being reported to us.

So that's it.

CHAIRPERSON PRIOLA: Okay. Thank you, Captain McMurtry.

Are there any questions or comments from the committee for Captain McMurtry or any of the other speakers? Dr. Bracey.

DR. BRACEY: Yes. One of the panel members commented earlier on the need to have data related to adverse outcomes, i.e., cancellations, et cetera. It would be helpful if your data gathering could -- I don't know if it collects such data now, but if not, if it could be retooled.

CAPT. McMURTRY: Could I have that last slide back?

I sort of have an answer for you. These shortage comments, the way our system is set up, we have a space where the reporting sites can click on a comment, whether it's cancel surgery, delayed surgery, whether it's physician consultation, those are comments. We don't break them down. All we do is just add them up.

So I can't say that this right here represents X number of hospitals that have had to cancel or delay surgery. I'm saying that this line, for example, represents all of the comments that were made.

One of the things that we're thinking about doing is trying to figure out a way to break down the different comments so that we know which are cancel surgeries, which are delayed surgeries, which are consultations. We're in the process right now of laying a contract that's going to examine our data set.

I mentioned to you earlier that we're presently sentinel sites. We're trying to take this or not trying to. We are going to change this so that we have a representative sample, and we feel will be more reflective of the blood supply, and in the examination of our data sets, that's where we're going to look at trying to quantify anecdotes.

While that sounds strange to do that, that's what we're going to try and do.

CHAIRPERSON PRIOLA: Dr. Belay.

DR. BELAY: How does this compare with the transmit system that the FDA is setting up? Is there an overlap or are they complementary?

DR. WILLIAMS: I'm not sure which came first, the chicken or the egg. I was actually on the steering committee for the HHS program as well.

I think just to compare the two sites, you could consider that Transnet takes the characteristic information that Mack is referring to right now and captures it from a much broader source as well as I think we're thinking about doing a little more with those individual characteristics, for instance, the slide I showed, weighted them according to medical importance and showed their dynamics during the recent shortage.

That can be done with both systems. Certainly the data exists, but I think I guess I would argue that the values from getting it from a much broader base allows you to characterize much more at the local and regional level across the country. But those two components are similar.

The other major component of the HHS program is an inventory monitoring system at transfusion services, and as I mentioned in my talk, I think it's arguable that measuring inventory at blood centers might be a better way to go because it gives an earlier indicator of impending problems, and by the time it hits the transfusion service level, you're already realizing, you know, a shortage situation that needs to be dealt with quickly.

CAPT. McMURTRY: We are going to be getting together here, Alan and I and some others, to try and mesh these systems. We'd like to think that our systems will come together and will be complementary and provide data that FDA can use and data that the Secretary's office can use.

CHAIRPERSON PRIOLA: Okay. Thank you very much, Captain McMurtry.

We have a little bit of time for general questions or comments from the committee addressed to any or all of the presenters, if anyone has a general issue or question.

I know I have -- oh, Dr. Johnson, do you have one? No. You just looked like you were ready to say something.

I know I have one. I'm still a little confused in my mind as to I mean, it's clear that there have been significant shortages and problems with the blood supply, but how much of it is truly attributable to these deferrals? How much recruitment, whether or not it's been tremendously successful or not has alleviated that, has sort of, you know, overcome some of those losses.

And getting back to Dr. Gambetti's question about just is this, again, due to the deferrals or to a general negative trend, I don't have a good idea of how recruitment, in particular, has affected these numbers. Can any of you comment on that?

Because this was a huge issue a meeting or two ago where there were a lot of talks on this, and it sounded as though all of the speakers that had programs in place to do this.

DR. BIANCO: Yeah, this is a very difficult question to answer, and I agree. The CJD deferrals are one component of a much bigger picture. There has been, if we look at the data collected by National Blood Date Research Center or the AABB, there has been an increase in collections over the years. We see among our members a steady increase in collections of three percent a year to five percent a year over the last five years. There has been an increase.

And that's obviously the result of recruitment efforts. It is the result of increased use of advertising. Our members at America's Blood Centers got together and created actually a program of advertisement, market research and using everything that is used to sell cereal boxes to recruit blood donors, and that has been successful, but has not been able to keep up with the what is chewed up on the other side in terms of the deferrals and in terms of that negative trend that I have to agree there is a certain distrust for the system. The donation process has become much more difficult.

If we look at a questionnaire, there are imbedded in there about 40 questions that we ask, and there are about 75 questions. It's a question of time availability, many factors working together, but deferrals are very important in not only vCJD.

We estimated about 56,000 donors lost with the vCJD. We lose the same number with deferrals for malaria for which we have maybe one or two cases as year because people go to Mexico, go have their vacation, visit one of the ruins. they have been in an area defined CDC as a malarious area, and they are deferred for a year.

And usually a donor that is deferred doesn't come back even if he's a temporary deferral. The donor feels rejected, made the effort, and it's easier to say no the next time.

CHAIRPERSON PRIOLA: Dr. Bolton.

DR. BOLTON: Yeah, I have two questions that I'd like to get any reaction from any of the speakers, and one is how to phrase this. The very negative public reaction to the post 9/11 large donation and then loss of units that were donated, and the second question is at this point in time we have no case of transfusion related CJD transmission. How does your outlook change when the first case or the second case of transfusion transmitted CJD is reported?

DR. DAVEY: Well, Dr. Bolton, certainly the reaction to 9/11 was extraordinary. As you've seen from our graphs in New York, we had that surge in donations and then a subsequent trough.

We, and I know other blood centers, have made major efforts to identify those donors that came out in response to the disaster that day and see if we could reengage them in the blood donation process.

We've had minimal success, and I think others have likewise had difficulty in recruiting those donors, and I think what we've found is that there is a group of donors, and I think Celso or maybe some of the others mentioned this, Peter, that respond to a major disaster such as that, but are not particularly motivated to come out and be regular donors to maintain the daily supply of blood.

We've worked very hard to try to actually get away from the term "blood bank" in New York and instead try to educate our population that blood is really a pipeline. We're not a bank. The blood isn't put in there for year. It's a pipeline of 42 days for blood and five days for platelets from donor to patient that needs continually to be replenished, but some donors just do not seem to be able to respond to that.

So we have to identify what donor groups are most likely to become engaged, high school for sure, and engage them in the blood donation process.

DR. BOLTON: But it seems to me, if we look at the graphs, that you not only dropped back to the pre-9/11 rate, but you dropped below and have never recovered. There's obviously some donors that have gotten turned off somehow, and is there any way to get them back?

DR. DAVEY: Well, again, I think, as Celso mentioned, the impacts on blood donations and blood deferrals are complex, but we feel that at least, again, in New York there was a negative response to blood being thrown away. We were quite concerned about the continued media attention to donate blood after 9/11, when clearly we didn't need the blood, and that gave a negative message to the public when the media focused on the thousands of units that had to be discarded, and that clearly was a message that we certainly regret and had hoped not to be conveyed, but that's what happened.

But we also, again, in New York, and I think my comments may have focused on this, the drop in donations after 9/11, which certainly were a response to blood being discarded, then also moved right into the CJD deferrals, and we feel that that continued decrement from what our projected increase in donations was, now the 9/11 phenomenon has passed, and we feel that now we're looking at the ongoing impact of other impacts on the blood donor base, including vCJD deferrals, and we have to compensate for that loss.

CHAIRPERSON PRIOLA: Dr. Hogan.

DR. HOGAN: One impact of blood cell storage is need. The other is cost. Now, you had mentioned that there were 107,000 donations potentially lost because of the vCJD that you can quantify. How does that translate to cost for the blood?

DR. BIANCO: Well, the cost of blood has increased substantially, but, again, not only because of the losses of donors and increased investment and recruitment, but a lot of audit manager that have been introduced in recent years by increased percentage of leukoreduction (phonetic).

DR. HOGAN: Is there any way you can quantify the amount of cost specifically because of the deferrals, the CJD deferrals?

DR. BIANCO: No, I don't have these figures. Most of the costs that I'm aware of that increased came from leucoreduction filters that cost about $25 a piece and the implementation of nucleic acid testing for HIV and HCV in which the companies were licensed in the past year and started charging the full price. That is, the research study, the IND study was over, and so that was another substantial cost.

But for the recruitment what has happened over the years, in the early '90s, late '80s, because of the HIV, the pie remained about the same. So what many centers did was to shift a lot of the investment in recruitment back into quality testing and all of these other things.

Now I see a trend. There is a high investment in recruitment, a higher focus on donors as the basis of the system, but I cannot quantify without looking more

DR. PAGE: I agree with Celso in the major contributions to increasing cost by leukoreduction and nucleic acid testing are increases in costs for units already collected. I think you question is what would the cost be to collect more units than we have been.

It is more cost effective for us to collect blood at a large company where donors come in and meet their appointments every 15 minutes throughout the day that is not too far from the blood center because a biggest part of our cost is our collection staff, many of whom are nurses.

When we are short, we will collect at companies with smaller numbers of employees. We will drive further away from them. We will do more on Saturdays and Sundays, where the collection time is on overtime if it's a longer drive because we have to pay for their travel time or weekend or evening differentials.

So those have already been engaged to a certain extent, but that's a substantial amount of increase in cost. If there were more schools and colleges or major companies and factories to whom we could go and get the support of their CEO to donate, that would be a more, relatively move cost efficient way to collect the more blood that we're always trying to do.

DR. DAVEY: Just a quick comment. Again, New York is a unique situation because of Euroblood. We were able to import Euroblood at a very reasonable cost. It came into the United States at a very reasonable rate, and we were able to pass that on to our customers.

New York City, as you know, is a very high rent neighborhood. If you go there to see a Broadway show, you know what it costs. That also impacts on our ability to collect blood. We have a union situation in New York. Other cities do, too, but it's a very high cost effort in New York to collect the unit of blood.

So Euroblood allowed us to pass on some cost savings to our hospitals which we can no longer do. That has resulted in a substantial increase in price in what we've had to charge to our customers in New York, substantial, and they are feeling that pinch very, very acutely right now.

CHAIRPERSON PRIOLA: Dr. Bracey.

DR. BRACEY: Yeah, in these times of tight budgetary restrictions, there's one area where clearly the budget is growing, and that's the Office of Homeland Defense. Blood is a resource, one of many resources that might be needed in a disaster, and clearly there's some debate about whether it would be needed in all disasters, but my thinking is that it would have a role to play in selected disastrous situations.

And I guess what I wonder is what has the effort on the part of the blood organizations to stand up to the plate and to really get the resource issue addressed to eliminate the fragility of the supply?

DR. GREGORY: Let me address the issue on behalf of an interorganizational task force that has been put together by the AABB, but consists of AABB, ABC, ARC, CDC, FDA, practically everybody we can think of who would be interested in this particular issue.

And we've been working very actively over the past year at trying to figure out not only what we should be doing, but how we should be doing it and how we can finance those, and in particular, looking at is there any money in the Homeland Security Department that we could be able to divert to use for making sure that we do have blood available.

I don't have any specific answers for you yet, except to tell you that this task force meets on a very regular basis and is currently looking at a number of plans and how specifically to ask for additional funding and how we can handle this.

CHAIRPERSON PRIOLA: Go ahead, Dr. Bianco. You had a follow-up.

DR. BIANCO: I just had a follow-up. Dr. Bolton had asked another very difficult question that none of us answered. That is, what would happen if there was a case of CJD transmission by transfusion?

And we actually just had an example in the past few months with the West Nile Virus where very quickly there was information gathered by CDC, extensive epidemiologic studies done that were the basis for a lot of activity within the Food and Drug Administration and the blood banking organizations, and ultimately very fast not only came guidance for the hospitals to attempt to restrict use of blood only to the minimum necessary in areas where the epidemic was very serious; also, to the point that later in the year we even withdrew all products that were frozen products so they have extended dating plasmas that stay a year. They have a shelf life that were collected in regions of high risk and all of that.

So I think that historically there are responses in many ways that can be implemented very fast, and that we have done.

CHAIRPERSON PRIOLA: Dr. DeArmond.

Just a second. Sorry.

DR. DAVEY: I kind of wanted to comment on Dr. Bolton's comment. I agree with Celso, but I think that the way the public perceives transfusion risk can vary considerably. I think Jim AuBuchon has shown a slide recently that if you ask a transfusion recipient what does he worry about or she worry about the transfusion, it's HIV, HIV, HIV, maybe hepatitis, HIV, HIV, HIV down the line, maybe CJD, HIV, HIV.

Well, clearly the major risk for transfusion recipients these days if you're getting platelets for sure is bacteria, for sure, and I think we're aware of this and we're addressing this. But perhaps to Dr. Bolton's point I would suspect while I think the response to a case of vCJD would be appropriate and the proper agencies would mobilize, I think the public and media response to something like that would be extraordinary, and we would have to address that in whatever way we could.

DR. BOLTON: I guess the reason I make that point is that in the West Nile Virus situation, you have an acute epidemic or focus of infection that can be recognized quickly and dealt with. The first case of CJD or variant CJD transmission indicates an underlying potential of possibly millions of donors, not something that necessarily could be responded to quickly and effectively.

So that in our possible contemplation of relaxing the restrictions, that's what I want to put out there. The difference between now talking about no cases versus one case, I mean, either we never will or one day we'll walk into a meeting in this room or some other room and begin talking about the first case of transfusion transmitted CJD, and it's an entirely different view of the world at that point.

DR. DeARMOND: I've almost forgotten what I was going to ask. I worry about Alzheimer's disease more than --

(Laughter.)

DR. DeARMOND: It seems like as I'm hearing all of this and over the last couple of years also, that it's becoming clearer to me, I believe that there are really two factors that are key in this decrease. Part of it is the safety of the blood, of which it looks like the variant CJD, the CJD component that Sue asked about how much of a component is that to the decrease, is a component but may be a relatively minor component considering all of the deferrals that are there.

The second issue is the economics and sociology of acquiring and donating blood, which is a completely different issue, and I didn't think of the economics of it as much, but we hear that New York City gets blood from Europe because it's far less expensive and that it's unions and all of the rest of the wage side that's making it very difficult economically to acquire blood.

So those are two completely different issues, and it's our charge to worry about the safety of the blood supply. I don't think our job is to worry about the economics. I presume that's at a higher authority.

But certainly in terms of the safety of the blood supply and thinking about how I would vote at this meeting and raising the possibility that some day there will be a transfusion related CJD, that we would have to say we have to err on the side of safety, and that the CJD component is a relatively minor component of the entire problem of acquiring blood.

So the question is: am I wrong on that or has my thinking muddled?

DR. BIANCO: No, I cannot say that you are wrong on that. I think that is not a minor component. I think it is a substantial component, but when we think of zero risk, that is what has driven a lot of the precautionary principle and what we do. I would have to say yes.

But the ultimate safety, Dr. DeArmond, is no transfusion ultimately, and all of us when we look at transfusion and when the media is there, I'm a transfusion recipient. I had my GI bleed, and I needed 40 units of blood, and at that moment if you ask me -- well, I wasn't able to talk --

(Laughter.)

DR. BIANCO: -- but certainly I'm here because that happened and happened in '86, many years before HCV and all of those things. I'm lucky I came out okay.

I think there is that balance that we are not being able to do with the precautionary principle. One side of the precautionary principle there is let's do whatever we can do to prevent it, but we didn't look at the impact side because there aren't as many elements there, and that has paralyzed us.

Ultimately I feel that we are coming to a point where we are unable to make decisions because that one case is possible, and that's where we need the whole strength of this committee. You have to help us to make that decision. You are the experts, and you have to help us say in that balance this much is reasonable; this is not reasonable.

CHAIRPERSON PRIOLA: Dr. Johnson.

DR. JOHNSON: Just to follow on exactly the same subject, since 50 percent of those who are available are deferred, maybe one should look -- and there is a shortage -- maybe one should look at the entire 50 percent and say: do you really need to worry this much about somebody who's taking anti-malarial drugs?

What about 75 year olds who would be happy to donate or 16 year olds? I mean, it seems to me there are a number of issues that might be looked at in the entire list.

DR. BIANCO: We agree, and some have, as the example of the hemochromatosis, that is, there were changes that FDA has made that now allow us if the donor fulfills the criteria of an allogeneic donation to take that blood and transfuse when before we could not do it because of labeling requirements and all of that.

But that being enough, it is because the number that we talk about, about 50, 60 percent of the population is eligible to donate, but there is a difference between eligible and there is a difference and the ones that are willing and ready to donate, the ones that respond to the thing without a major disaster, without a major event in a city where people rush to donate blood.

It's that blood on the shelf, that is, what you see is that we are at the edge. We would like to have five, seven days' inventory. That would resolve the questions that Dr. Bracey asked, that is, in terms of preparedness or for any event.

That blood on the shelf is the one that saves lives. That blood that comes after the event, it will fuel back the pipeline, but is not the one that is going to resolve. How we get from this three to five percent of the population to donate to six or seven, there are other questions that are very important.

I said 1.9 percent of the donors that were deferred in this survey that we did. It's 1.9 donations a year. They can make five. If we could get one more donation from those people a year, we would increase it by 20, 30 percent, but it's very, very difficult because there are many other factors in the life of an individual, and blood donation is one small event in their lives.

CHAIRPERSON PRIOLA: Dr. Epstein, do you have a comment?

DR. EPSTEIN: Yes. I just wanted to comment in response to Dr. Johnson's point. You know, we have another committee, the Blood Products Advisory Committee, where we routinely review the safety issues for blood collection. We recognized back in 1999 that part of the strategy in dealing with the blood supply was to try to remove barriers to safe donation as part of a vehicle toward improving availability, and we have systematically in cooperation with the Centers for Disease Control reviewed many of the standards that result in the donor deferral.

And just to rattle off a quick list that I jotted down, we've looked at the value of or lack of value of the syphilis antibody screen, the anti-Hepatitis B core antibody test, history of hepatitis, the current exclusion factors for malaria risk, the hemoglobin standard as it applies to males and females, the value and safety use of deferrals for body piercing, tatoos, acupuncture. We've specifically looked at whether we could use hemochromatosis donors without special labeling or other restrictions.

We looked at male sex with male history, and we've looked at strategies to improve the questionnaire process itself by simply streamlining it, improving comprehension, et cetera, and the possibility to use abbreviated questionnaires in certain contexts.

So we do accept your point. We have been aware of it. Now, we could have a debate over how much real progress has been made removing barriers. The bottom line is that it's very difficult to remove safety measures that are in place because it tends to require a demonstration that you haven't made blood less safe by removing it, and that's always a difficult demonstration.

DR. JOHNSON: But, Jay, that's going to be one of the problems in stepping back from the CJD issue. It was one of the issues that was brought up when we got into this.

DR. EPSTEIN: Yes, but we face that problem all the time, and it's particularly acute when measures have been put into place on a precautionary basis rather than being data driven, in other words, with known benefits. It's a very difficult problem in the area of CJD because of the very long incubation period and the absence of a diagnostic test on blood that could be used in asymptomatic individuals.

But it's not a novel problem, but it will be a difficult one in this area.

CHAIRPERSON PRIOLA: Okay. Thank you very much. I'd like to thank all of the speakers for their presentations, and we can move on to the open public hearing portion of this meeting, and Bill could update us on any requests for speaking in the open public hearing.

DR. FREAS: As part of the Advisory Committee procedure, we hold this open public hearing session in order for members of the public who are not on the agenda to make presentations before the committee.

At this time I have not received any requests to speak at this morning's open public hearing session. I see that we have one request right now, and we welcome you. Would you please come to microphone?

While you're coming to the microphone, I would like to add that we did receive a written statement for the open public since it was not related to an agenda topic I had made copies. The copies are in the members' folders and out on the desk and you can read them at your leisure.

Go ahead.

MR. CAVENAUGH: Thank you.

My name is Dave Cavanaugh. I'm on the Government Relations staff for the Committee of 10,000, which is a national organization of people with hemophilia who contracted HIV from their medications.

I'm pleased to be here, and I'm pleased to see Dr. Bianco here because in 1986 when he needed that blood, we weren't all sure it was safe yet. It was 1987 before all of the protections were in place.

I know that this has been an update, and I know that the focus of it has been on supply, but it's been about geographic donor bans, and I would think that to really evaluate whether we need to keep a ban in place we also have to look at is there still danger there.

I think I heard a comment very early about minimizing the risk in Europe outside of the U.K., but I just would really want to look at some of the things we've looked at before, which is new countries who are affording BSE.

I looked a little bit about the problems in Portugal, which has a terrible slaughtering mechanism and meat safety procedure, and it's a very under reported situation before relaxing too much about the lack of a spread of cattle born disease.

There's also an issue of the scientists in the U.K. are unwilling to say the curve is over. We do have a Bell curve. It does seem to have a down side, but as we all know, Bob Will has been saying for years that there could be a mini of a larger curve, and somewhere in between there may lie truth.

And so I think prudence once again is indicated. I think we heard this morning, and I thank Dr. DeArmond for saying so, that there is increasing evidence in the animal research community about transfusion of this prion, and I think we have to watch out in this country unless we say we don't have it here, that we pay more attention to the CWD explosion, the chronic waste and disease in deer, which has gone from being constrained to one or two deer and elk ranches in Colorado and Wyoming to North Carolina, Wisconsin, and lots of other states of the country, and is being monitored very closely. It's a phenomenon that is subject to suppression because of the tourism costs that it brings to these states' economies.

And there are some rumblings of new findings on the phenotypes of CJD indicating that what we call variant CJD doesn't always have to be linked to cattle. It may be coming to us in other ways.

So I just would like to lay those issues before the committee before we conclude this update.

Thank you.

DR. FREAS: Thank you for your comments.

Next Chris Healey.

DR. HEALEY: Good morning. I'm with the Plasma Protein Therapeutics Association.

I know a lot of what you heard this morning was about blood supply monitoring, but on behalf of the plasma collection and fractionation industry, I just wanted to make you aware that there's also a plasma therapeutics monitoring system currently in place.

PPTA established this system back in 1997 when there was a shortage of IVIG that grew out of a number of factors, one of which was deferrals regarding classic CJD, and we were pleased that the agency was able to take prompt action and change its perspective and its stance on those deferrals and help relax some of the supply constraints that we have faced.

But our plasma supply monitoring system has been in place since then. We publish monthly data on the availability of the major product categories. At present we, like many people in the country these days are using a red light, green light, yellow light system to indicate the current availability.

Our numbers are based on a ratio of the first day of inventory to the total month of distribution across the entire industry, across all of the major fractionators in the U.S., and anything over a ratio of 1.25 is considered a green zone. We think that roughly equates to about five weeks of available supply.

Anything from about 1.25 to .5 is considered a yellow zone, which is a watchful area. We need to be aware of what's going on and track it closely.

And then, of course, anything below a .5 is a red zone.

Whenever we're in a yellow or a red, we publish actual data so that all of the stakeholders and all of the community has the actual data on availability.

When we're in the green zone we feel comfortable that supplies are available, and so there's no need to publish those figures.

So I'd encourage you all to log onto ppta.org and click on the red light/green light and check out the supply.

Thank you.

DR. FREAS: Thank you, Chris.

Is there anyone else would like to address the committee at this time?

(No response.)

DR. FREAS: If not, we will hold another open public hearing in the afternoon.

Thank you.

I turn the microphone over to you, Dr. Priola.

CHAIRPERSON PRIOLA: All right. Thank you.

I think we'll all take a short break here and return at about 11 o'clock to resume with the second part of the morning session.

(Whereupon, the foregoing matter went off the record at 10:45 a.m. and went back on the record at 11:05 a.m.)

CHAIRPERSON PRIOLA: Okay. If all of the committee members could return to their seats, we'll go ahead and start the second part of this morning's session.

Our first speaker is Dr. David Asher, and he's from the FDA. He's going to be talk about BSE epidemiology, food chain controls, and updating us as to how that's going.

Dr. Asher.

DR. ASHER: Thank you, Sue.

The Food and Drug Administration has accepted that effective food chain controls, including BSE control in cattle has substantially reduced the risk of human exposure to the BSE agent in the United Kingdom after 1996, and the question arises how will we know when a similar situation applies to other BSE countries.

If the number of people infected with the BSE agent after exposure is less than had been feared or if the risk that their blood contains transmissible amounts of agent is less, then it follows that the overall risk should be less, but how will we decide when the risk becomes acceptable or perhaps better justified to maintain supply?

As you know, the Department of Health and Human Services is committed to reevaluate TSEs and blood risk and the risks posed by the deferral policies frequently, at least every six months, and we've been attempting to do that in this committee as much as possible.

As part of that effort, I was asked to review the BSE, BSE situation, in large part because Lisa Ferguson is a member of the committee, and it's been our tradition not to ask committee members to be distracted by having to give talks themselves.

So first I want to speak briefly about variant Creutzfeldt-Jakob disease because that's what we're all afraid of, the cases, trends, and implications for the safety of the blood supply. Then I'll speak about BSE itself, world distribution, briefly about new countries, risk assessments that have been attempted, a warning from the World Health organization, a general consideration of measures that might be expected to reduce human exposure.

And finally I'll closed with some comments about actions taken by some of the European BSE countries, those in the European Union, and one example of a national action that has been taken, and I'll make a brief comment about compliance, although I must say that I really have no special information about the levels of compliance that are being reached in BSE countries.

First, variant CJD. As of last week, I'm told that the count in the United Kingdom is actually up to 131, of whom at least 119 have died, six cases in France, at least five of whom have, I believe, no history of travel to the U.K., and one case in Italy with no history of travel to U.K.

There have also been cases in Ireland, the United States, Canada, and possibly one in Spain among people who live for a prolonged period of time in the U.K.

The former Hong Kong case has been reclassified as a U.K. case from this list.

The disease remains strikingly different from sporadic CJ, and it's puzzling to me that the mean age distribution has remained the same since the disease was first described in 1996, about 28 or 29. I had sort of expected that it would be like kuru, that after the major exposure ended, that the cases would become older and older, but that has not occurred. These are numbers from Bob Will for the first 130 cases. The mean age at death is still 29. The mean age at onset, 20, 27, a much longer duration, 14 months and sporadic CJD with a range of duration of illness that extends to 39 months.

Thus far all cases tested have been homozygous for methionine at Codon 129 of the prion protein encoding, coding gene both in the U.K. and in other countries, which raises a problem in that we know for sporadic CJD and I-effergenic (phonetic) CJD that people with that genotype appear to be at increased risk, but the other genotypes are not completely protective, and it does raise the question of how many people with the other genotypes have been infected. When are they going to become ill, and are we going to recognize it as a new variant of CJD?

As was mentioned by Dr. Bianco earlier this morning, the onset of variant CJD in the United Kingdom appears to have peaked in 1999. Deaths peaked in 2000 so much that the epidemiologists there had to abandon their trend analysis using an assumption of an exponential risk, and when they shifted to a quadratic analysis, they've noted a trend that is downward, although, of course, one doesn't know how long that will last, and the same thing for deaths which have been trending downward.

Not trending downward is the overall death rate of Creutzfeldt-Jakob disease in both the U.K. and in Switzerland. Rates of sporadic CJD recognized have increased. The CJD surveillance unit staff thinks that that's most likely attributable to increased recognition and reporting which the Swiss authorities are not so sure. It bears continued watching.

There were some comments made this morning about the studies of blood of patients with variant CJD, and I must say, and it's a personal opinion, that I think it's really too early to take great comfort from those studies.

The epidemiological studies for recipients of labile components from donors who later came down with variant CJD, there were 33 of them identified from ten donors. Eighteen of those cases have died, and only four lived longer than one year.

So you really can't draw any conclusions from the patients who have died about whether there was the possibility of transmission. Of the 15 still living, all but one are less than ten years after transmission, and if the projections of incubation period after oral exposure are really 15 years and if the amount of agent in blood were to be very small, which animal studies suggest it would be, then I think that ten years is really not long enough to draw a great deal of comfort from the failure to transmit, although it would have been terrible, obviously if some of the patients had come down.

Furthermore, the attempts to transmit infectious agent from the blood of human beings, although they're negative so far, but they've been on test for less than five years, and they're, of course, plagued by the small sample size, by the small volumes of blood that can be tested, and the fact that there is a species barrier between monkeys, monkeys and human beings, and again, the amount of infectivity to be expected in blood if it's like the blood of experimentally infected animals would presumably be very small.

So we're still left with the same concern, that is, agent in blood is a regular phenomenon in a number of experimental models, and it would be premature to conclude that a phenomenon that occurs so regularly in experimental systems could never occur in human beings.

Okay. The bovine spongiform encephalopathy, and arguably of all the things that could be done to protect human beings from exposure, control of the disease in animals is probably the single most important. Twenty-two countries have now reported BSE in native born cattle, all of these in Europe except for two in Asia, Japan and Israel.

The United Kingdom continues to report the largest number of cases. In 2001, they had over 1,000 cases and 755 in 2002, followed by Ireland, Switzerland, France.

The cases in the United Kingdom peaked in 1992 and there has been a gratifying fall-off in cases. There have been more than -- I've lost count -- but more than 30 cases born after 1996, so called born after the real ban or born after the reinforced ban, when very stringent control of food to which ruminants were exposed went into force.

There's fortunately no evidence that I've heard that those cases represent either maternal or contact transmissions and the conclusion of John Wildsmith and other U.K. epidemiologists is they probably represent some leaking of contaminated food, contaminated animal feeds either from other BSE countries or retained in the U.K.

The incidence of BSE per mission cows greater than two months old remains highest in Britain, followed by Portugal, Ireland, Switzerland, and as we've mentioned, all 22 countries have it. The latest countries of concern would be Japan, which now has seven or possibly eight cases. Israel still has detected only one case.

In addition to the countries that have recognized BSE, there are almost certainly a number of other countries that have BSE, and at the last meeting at which we addressed this subject a year ago, there was some confusion about the different ways in which the USDA and the European Commission address the question of risk.

I think that the risk is in the last analysis not so different in the way it's addressed, but the management decisions based on the analysis are different. Both of them basically base their analyses on the recommendation of the International Epizootic Office.

The USDA does a conventional assessment in which groups of experts get together and look at available information both in the public domain and provided by the country and reach a decision about whether the country does or does not have a risk of BSE -- this is not articulated -- but greater than that's present in the United States. If they do, they're unacceptable to export live cattle or most bovine products to the United States, and little attempt is made to stratify the level of BSE risk in those countries. The management decision is they're either acceptable or they're unacceptable.

In Europe, of course, many, perhaps all countries with the possible exception of Norway have some risk of BSE, and it's not feasible to restrict all products to those from BSE-free sources.

So the EC, through its Scientific Steering Committee, has attempted a qualitative, almost semi-quantitative method. You're going to be seeing later in the day an example of a true quantitative, probabilistic risk assessment, but the EC decided to use a qualitative risk assessment of the probability that any country has BSE.

Their risk assessment led to more nuanced decisions because, of course, they can't eliminate all BSE countries from their sources of supply. Their system was not intended to assess human exposure risk. It depends on voluntary submission of dossier by various countries, which explains why you won't be seeing the names of lots of countries like Mexico on the list, because they have voluntarily submitted information to the EC, and the EC does not reach out and assess risk in countries that don't ask to be assessed.

The basis for the assessment is to convene an expert committee members nominated by the country being assessed, as well as by the EC, and one should note that the risk, of course, is dynamic. As countries take steps to control the risk, presumably it's going down.

The EC originally began by looking at eight risk factors, which I won't go through because they really boil down to two major concerns: one, what the EC calls the external challenge, which is the probability that BSE agent had entered a country either in live cattle imports or through imports of meat and bonemeal or materials contaminated with ruminant meat and bone meal, and then the so-called internal stability, which is the probability that the BSE agent once entering a country would be able to amplify and spread, and that's determined by the feeding practices in the country, specifically whether meat and bonemeal was used as a dietary supplement for cattle, the rendering practices, recycling of cattle into eye meat and bonemeal, removal of so-called specific risk material, the high risk tissues at the time of slaughter, and their sanitary disposal, the surveillance of BSE to be aware of whether there is a problem, and then controls across contamination, such as the prohibition of most forms of penetrative stunning for slaughter or certain meat cutting practices.

The system is called the geographical BSE risk or GBR, and it's based on certain assumptions. First, it assumes that BSE is restricted to bovines among the food animals, although the theoretical risk of infection of sheep and goats is acknowledged and steps are being taken to investigate that in Europe, but as part of the assessment, it's generally ignored.

The only source of BSE agent into a country are assumed to be imports of infected cattle or bovine derived material, and the theoretical risk of spontaneous generation of infectivity is ignored.

The transmission vector is assumed to be contaminated feed, and the theoretical risk of material contact and field contamination, that is, the so-called third mode of contamination which might be contaminated animal waste is ignored.

And finally, it's recognized that the fact that a surveillance system has failed to reveal cases doesn't mean that there are no cases. Risk assessment cannot rely strictly on reports of cases in a country.

The current version of GBR sets up four levels of risk, the risk being the risk of having one or more cattle clinically or otherwise present in a country. GBR-1 is highly unlikely. GBR-2 is unlikely, but not impossible. Three is likely, but either not confirmed yet or confirmed at a lower level, and four confirmed at a higher level.

There are only two countries currently listed as GBR-4 countries by the EC. Those are the United Kingdom and Portugal. Thirty-one countries are in GBR-3, and that includes most of the countries of Western Europe with the exception of Norway and Sweden. Nine countries are listed in GBR-2, including Canada and the USA. The USA imported over 400 cattle from the U.K. in the 1980s. Some of those cattle were recycled, and our ban on meat and bonemeal did not go into effect until the summer of 1997, and compliance, although increasing, has not been perfect.

Seventeen countries are currently listed in GBR, GBR-1. The only European country listed in GBR-1, Western European country, is Norway. The Eastern European countries, in general, have not submitted a dossier.

How successful has the EC's GBR been in predicting countries with BSE before they knew it? It correctly predicted the first case, that there would be a case of BSE in native cattle in nine countries. They would have predicted it in Japan and Israel, one thinks, if the dossier had been available by the time the first case was seen.

They did not predict BSE in three countries that subsequently recognized it: Austria, Finland, and Slovenia. So that their record for predicting BSE risk is pretty good, but certainly not perfect.

A year ago this committee reviewed measures that had been taken in the U.K. that we thought -- that you thought would be reasonably effective in preventing human risk. The first of those was effective control of BSE both in cattle and in small ruminants based on national surveillance programs, which should include active surveillance, meaning the testing of brain tissues from animals that increased risk of BSE.

Some countries have taken to testing brains of young, healthy slaughter animals, which seems to be a fairly pointless, if reassuring, exercise; prohibition on the feeding of most mammalian protein into ruminants, so-called feed ban, with steps to prevent accidental feeding; condemnation and destruction of animals with signs of BSE; preventive culling of animals at increased risk; and adequate compensation to owners to encourage cooperation.

Second, age based slaughter schemes which reduce the risk by prohibiting the consumption of meat products from ruminants slaughtered after an age when substantial amounts of BSE agent are likely to be present in tissues, which has generally been taken to be 30 months or even 24 by super cautious countries.

An example of that is the United Kingdom's so-called over 30 month scheme. Separation of high risk materials from edible meat products, again, prohibiting slaughter methods that embolize brain tissue, removal of the specified risk materials, mainly CNS, lymphoid and intestinal tissues from carcasses at the point of slaughter, and their effective segregation and sanitary disposal, and prohibition of advanced or mechanical meat recovery systems, particularly involving the skeletal and spinal column that might contaminate meat with ganglia spinal cord, perhaps even brain.

The measures applied to protect the domestic food supply should also be applied to imported food and compliance should be verified by inspections and audits. Those were the general conclusions, as I took them, of the meeting that this committee held addressing this issue a year ago.

It's interesting that the World Health Organization has recently published a little brochure suggesting eight questions about BSE that consumers might ask their national authorities that are very similar to those that we discussed a year ago: what have cattle been fed? Is there an active surveillance system? Have cases of BSE been imported or born in the country's herd? Does meat come from younger cattle?

High risk tissues removed and destroyed, procedures to prevent cross-contamination, any other meat products that could contain BSE agent by which they meant mechanically or advanced recovered meat, and then strict control.

On the 30th of January, the World Health Organization released a statement warning countries that have not recognized BSE, that they were almost certainly at risk of those, and those countries, they warned, were those in Central and Eastern Europe, in Africa, and in Southeast Asia.

The reason for that was they have been on record, large amounts of U.K. meat and bonemeal that were exported to some of those countries, and the fate of the meat and bone meal in the countries is not clear.

Also, some of the countries are unlikely to have an adequate surveillance system to recognize cases of BSE if they do occur.

What have the BSE countries of Western Europe -- I guess the term of art now is "Old Europe." Is that --

(Laughter.)

DR. ASHER: -- have done, and it's gratifying to see that some of the same procedures that we in the WHO have been concerned about have actually been encoded in legislation in the European Union as Regulation 999/2000-2001, which went into effect on the 22nd of May and which binds all of the 15 members of the EC.

Again, I cannot comment on how -- I will just briefly on how compliance efforts might be going. The regulation applies to live animals and animal derived products intended for food, feed, and fertilizer. It does not apply to cosmetics, medicines or medical devices for which there are other rules, and its rules applied to all EC member states.

Member states are to monitor BSE and scrapie with annual active programs that include rapid PrP tests, designate, remove, and safely dispose of SRMs, prohibit feedings of certain proteins to certain animals, notify authorities of suspect animals, quarantine carcasses, destroy carcasses if they're identified as infected, and identify the other animals at risk, restricting movements, compensating owners, prohibiting the marketing of certain products from bovines in high risk regions.

But they allow for certain products produced under controlled conditions from animals demonstrated not posing a high risk. And I might interject that that's very similar to what this committee concluded when addressing the issue of capsule gelatin from Western European countries in 1998.

Establish reference laboratories for TSE testing, including the rapid PrP tests. Those are Western Blot and ELISA based tests.

Carry out inspections of both EC member states and third countries exporting to the EC, and reassess the BSE status when OIE develops suitable procedures.

It appears from -- this is a very -- it's an 11 page act with 29 pages of annexes, 26 articles, and I'm a lawyer, let alone a European lawyer. So I can't comment on all of the nuances, but it appears to me that they're proposing a somewhat more complicated scheme for stratifying BSE risk in the various countries. They are proposing five layers of risk from a country that has no BSE, no risk, met all of the requirements for at least seven years, including no feeding of meat and bonemeal for more than eight years of the high BSE incidence countries that have more than 100 cases per million animals over 24 months old.

When they proposed to implement this system perhaps Lisa knows. I don't.

We were concerned about the likelihood that this legislation was actually being adequately implemented in various European countries, and I couldn't possibly review even all 15 EC countries, let alone all of the countries of Europe, but what I did was look briefly at what the French National Food Security Agency, AFSSA, had done in the past couple of years.

This agency provides advice to the government of France, and at least what I saw suggests to me that I can't comment on their actual compliance, but they're certainly well aware of all the issues that we have discussed and that the World Health Organization has warned of.

In the year 2001, AFSSA released 33 opinions. I laboriously tried to translate just the titles of the 26 opinions that they issued last year, everything from animal feeds, TSE small ruminants, farm game meet, traceability of sheep and goats, slaughterhouse sanitation, importation of Swiss cattle, fertilizer, comment on the article reporting TSE infectivity in rodent muscle, screening tests for small ruminants, survey of retrieval of specified risk material in slaughter houses, mechanical meat recovery, spinal cord removal from small ruminants, slaughter house inspections, risk assessments for beef, modification of U.K. imports, scrapie control, marketing of contact sheep and goats, that is, sheep and goats exposed to scrapie, and a proposed rule regarding the direct sale of sheep and goat carcasses containing spinal cords.

So clearly, the French agency has been concerned about quite a number that we've been concerned about, and if they're not compliant, it's not because they've not received adequate advice on how to become compliance.

There is an inspection and auditing organization in the EC structure, the Food and Veterinary Office, that has at least twice audited France. They audited them in March of last year and found that their system was generally compliant. They had some problems with traceability of animals with multiple sales and some minor problems with labeling, but at least as I read the report, the auditing organization was fairly confident that the system itself appeared to be adequate.

I did a small personal survey of non-French citizens living in France to ask about their opinion on compliance and was reassured that if the newspaper articles complaining of noncompliance are any evidence, that compliance is probably pretty good.

Thank you, Bill. I made it farther into the talk than I thought I might.

These are some useful Web sites that I used in preparing this summary, and of course, there are many others. I didn't list the CJD Web site.

I think it's fair to agree with the World Health Organization that BSE remains a worldwide problem. A number of BSE countries have undertaken effective responses to the problem. It's too early to conclude that compliance is adequate to reduce the risk to negligible levels. Unfortunately a number of other countries are almost certainly involved and so far as one knows have not taken any steps at all.

Thank you.

CHAIRPERSON PRIOLA: Thank you, Dr. Asher.

Are there any questions or comments for Dr. Asher? Dr. Belay.

DR. BELAY: They were having BSE cases in the United Kingdom in animals born after 1996. Are those cases still occurring or are there an increase in BSE cases in animals born after 1996?

DR. ASHER: My impression is that there's not an increase, but that they are continuing to occur. The reassuring thing is in those cases where the mothers of the animals survive, that is, were not slaughtered for other reasons.

There has been no case in which a BSE born after the real ban, a reinforced ban cow has a mother which itself came down with BSE.

You know, there are two patterns really with TSEs. The scrapie-chronic wasting disease pattern in which there's horizontal spread and perinatal spread at least for scrapie and then the kuru pattern in which there doesn't appear to be any spread perinatally.

DR. BELAY: So mother-to-child transmission is not an issue in those cases in the U.K.?

DR. ASHER: The evidence is increasingly reassuring that there's no evidence for maternal transmission. The original reports were that maternal transmission could account for as much ten percent of the cases. The problem was that all of the establishments in which the cases came from had a history of some feed contamination, and it was not possible to tease out which infections were occurring mother to calf and which ones were occurring from contaminated premises.

The contaminated premises is now the overwhelming favorite as the main, if not the only, mechanism by which BSE spreads, which makes it more like mink encephalopathy and kuru in its pathogenesis, and we obviously hope that it remains like that because it makes it easier to eradicate.

DR. BELAY: Are you still having the over 30 months for the United Kingdom? The over 30 months, are they still implementing it?

DR. ASHER: Well, as you probably know, they've been discussing for more than a year the possibility of relaxing it. My personal view is that it would be premature to do that.

Israel, also in my own view not a good idea, has elected to use the testing of 30 month cattle rather than eliminating them from the food chain. I personally think that's a mistake.

They didn't ask me.

(Laughter.)

CHAIRPERSON PRIOLA: Dr. Gambetti.

DR. GAMBETTI: David, having a review of this, and it's a very interesting review, what is your understanding of the relationship between the prevalence of BSE and of vCJD?

Because looking at the data that you have of BSE incidence, you don't see any correlation. Of course, there are so many factors that may influence, but even comparing, for example, Germany and France, you have six cases of vCJD in France and none in Germany, in two countries that are more or less of similar size and probably have a type of surveillance that is comparable.

So what is your understanding of the correlation between the two diseases?

DR. ASHER: Thus far, the correlation that seems solid is correlation with exposure to U.K. beef. France imported at least five percent of its beef from the U.K. during the bad years. Germany presumably did not.

CHAIRPERSON PRIOLA: Lisa, Dr. Ferguson?

David, are you okay?

DR. FERGUSON: I just wanted to add a comment, I suppose, on the incidence as reported in cattle. I would really caution everybody in putting any type of weight in those figures. We honestly don't know if that is actually a true incidence. Most of the continental European countries have really only started doing active surveillance when they were mandated to do so at the start of 2001, and no surprise, obviously then there was an apparent increase in cases.

So I think it's really too early yet to say do we have an accurate reporting of incidents in all of these countries.

DR. GAMBETTI: You think that it is still under reported?

DR. FERGUSON: Not still. I think at this point in time because they're mandated within the community, at least within the 15 member states, to test every animal that is slaughtered over 30 months of age, and that's giving us a good idea of what is actually going on at least in those countries, but that has just been going on since 2001.

DR. BOLTON: David, that just sparked a question in my mind. David, I realize that you are not a European lawyer, but what do you think they mean by "should" as opposed to "must"?

DR. ASHER: Yeah, obviously that was a concern to me, and that's why I put it in quotes. In this country "should" means advice, not obligation, and I was concerned about that choice of words.

CHAIRPERSON PRIOLA: Thank you very much.

I think we're going to move on here unless -- I think we'll move on here, and there will be time for more discussion after lunch.

We're going to go on, and besides, I don't think David can answer many more questions.

(Laughter.)

CHAIRPERSON PRIOLA: Give him a chance to rest.

We're going to move on to the primary topic of discussion and actually the topic for voting today, and that's consideration of labeling claims for TSE agent clearance and plasma derivatives. We have a series of talks here, and sticking with the issue of diminishing returns, this leads us right up to lunch.

So we're going to try to go through the speakers and where there's time take questions, but keep in mind we have discussion scheduled for after lunch. So there will be time after lunch, as well, and we'll try to take questions as time permits and see how the discussion goes.

So with that, I'll turn it over to Dr. Scott from the FDA.

DR. SCOTT: Thank you.

I'm going to try to put this in context for you, the question first by reviewing the current measures that we have implemented to try to enhance safety of plasma derivatives with respect to TSEs.

Just very briefly to review, we have recommended donor deferrals for vCJD and classical CJD, of course, as well as for vCJD risk. The ones for vCJD risk are listed here: more than three months travel or residence in the U.K. for a defined time period; more than five years travel or residence in France.

I should mention I'm talking about deferral of source plasma. The deferral is slightly different for blood.

Greater than six months at U.S. military bases in Europe because they have consumed British beef to Europe; transfusion in the U.K. from 1980 to the present; and having U.K. source bovine insulin injections between 1980 and the present.

We have also recommended for quite some period of time donor deferrals for people who have CJD, of course, and risk factors for coming down with CJD, including iatrogenic risks and family history of CJD.

We've also recommended quarantine retrieval of plasma units from donors with either of these diseases, as well as withdrawal of plasma derivatives if there was a donor that came down with vCJD.

One way or the only way in which these deferrals are distinguished from the blood donor deferral are that donors of source plasma are not deferred for travel or residence in BSE countries with the exception of France and the United Kingdom. So we do not have the pan-European, as we call it, donor deferral that blood donors have and that you've heard about this morning, which has affected the blood supply.

The rationale for this is as follows. We were aware already that model TSE agents are partitioned away from or removed from plasma derivatives during plasma fractionation, and this all came from published studies, as well as information that was presented here to the committee.

The European risk of variant CJD, with the exception of France and the United Kingdom is believed to be significantly lower because of smaller BSE epidemic relative to that in the U.K.

In addition, we were concerned because the effects of the European donor deferral on nationwide and worldwide plasma supplies were quite uncertain, but potentially severe.

However, we continue to have ongoing concerns about the possibility that variant CJD could be present or transmitted by plasma derivatives. Certainly we have had, as David mentioned, multiple reports that are published of TSE transmission by blood and by plasma in animals, and this includes in rodents, lemurs, experimental BSE in sheep, and natural scrapie in sheep. The last two are the Houston studies.

We also, as outlined previously, have insufficient epidemiology to make us feel more sure that this won't happen, partly because the epidemic size of vCJD is unknown, and that means we don't know the prevalence of incubating donors, and the likelihood of transmission by blood products is unknown, although it is believed to be low.

In addition, with vCJD, we have the lymphoid tissue expression of the abnormal PrP protein, which implies, but does not prove that lymphoid tissue is more infectious than that of a CJD person or donor, but it also suggests that vCJD might have other unusual tissue distributions of the infectious agent. We just don't know.

Just to briefly put this in context, here I'm comparing what we can do to reduce the risk of viral transmission in plasma derivatives compared with TSEs. In both cases we have implemented donor deferrals which are going to decrease the likelihood that an infectious plasma unit will enter a plasma derivative pool.

Of course, for viruses we have extensive donor testing, which can eliminate nearly all donors who are infected with HIV, HCV especially, but for CJD we don't have any such test, and we don't anticipate anything coming very soon that is going to help us out now.

In terms of clearance by processing, we have FDA reviewed manufacturer studies of clearance in viral inactivation, and these are very well established types of studies. So that we have assurance that viruses are removed or inactivated during plasma derivative manufacture.

Now, this is also a possibility for the TSE agents, and that is really the subject that we are going to be discussing today, and the question that is being brought forward to the committee is whether or not the current state of science and our ability to perform and evaluate these studies is sufficient to ultimately permit a labeling claim that certain plasma derivatives for which these studies have been done have a reduced risk of transmission of the CJD-like agents.

Just a few words about labeling. In cases such as this, it may provide an estimate of the risk of transmission of infectious agents. It already does certainly for viruses. It may discuss measures taken to reduce risks in the manufacturing, and it can allow the public to make a more informed decision risk-benefit.

Labeling claims for vital removal, for example, typically undergo thorough scientific and regulatory review prior to implementation, and all labeling claims certainly for viral clearance in the past have been based upon such rigorous and scientific evidence that we've been able to review.

Our current FDA recommendation for labeling for the risk of CJD in plasma derivatives is as follows because this product is made from human blood, it carries a risk of transmitting infectious agents, e.g., viruses and theoretically the CJD agent.

I should mention that the labeling for albumen is slightly more extensive because albumen has never been known to transmit a virus, and so there's additional reassurance in that slightly longer recommendations.

The questions that we have to ask you today are whether FDA should consider labeling claims for TSE clearance in plasma derivatives based upon specific demonstration of TSE removal during manufacturing.

And just as a caveat, I want to mention -- and we don't need to discuss it today -- but that this would assume adequacy of decontamination procedures in product manufacturing.

If you do agree with that premise, please comment upon whether such data would support the following draft wording for labeling. Because this product is made from human plasma, it carries the risk of transmitting infectious agents, e.g., viruses and theoretically the vCJD agent. Obviously we might want to have CJD and vCJD here, but this is very similar to the previous labeling, and it still contains the risk of viruses.

However, then it would go on to say something like, "It has been demonstrated that the manufacturer's process provides substantial clearance of agents similar to those causing CJD and vCJD. Thus, the theoretical risk of transmission of CJD or vCJD is extremely remote."

To help you think about this question, we have a number of speakers. Dr. Farshid from the Office of Blood is first going to discuss virus removal from blood products, validation, labeling claims, and a possible paradigm for the removal of TSE agents.

Then I will discuss specific clearance issues and where TSEs may differ from viruses, and when we're thinking about these types of clearance studies.

Then Steve Anderson from the Office of Biostatistics and Epidemiology will discuss a risk analysis that he has generated, a model risk analysis for the presence of TSEs in plasma derivatives.

Then Dr. Ruiz from the Spanish Medicines Agency, who is also heavily involved in the EMEA, will discuss the European perspective on TSE agent clearance.

And finally, we will hear from the PPTA. Dr. Baron will speak about TSE clearance studies performed by plasma manufacturers.

Thank you.

CHAIRPERSON PRIOLA: Okay. Thank you, Dr. Scott.

Moving on to the next speaker, you'll be coming back in a few minutes, and that's Dr. Farshid.

DR. FARSHID: Thank you.

My presentation will be a brief overview of viral validation studies and the Sievers (phonetic) approach in assessing such studies, the intention being that the general principle of validation was currently used for conventional viruses could be applicable to TSE clearance.

The product that currently requires viral clearance can loosely be divided into three different categories. There are monoclonal antibodies and recombinant product produced in subculture, as well as we note they from extensively tested and highly characterized starting material, and they have excellent viral safety record.

Animal derived products, they're starting with a real variable, and there are -- essentially a variable test is done in the starting materials. Therefore, the viral validation studies are more important in such products.

And finally, the plasma derived products overall and other human derived product also fall in this category like tissues and so on and so forth, and these are, as we know, the starting material are variable, and they have been implicated in transmission of viruses in the past.

And there are a number of complimentary approaches which have been alluded to this morning in reducing the risk of viral transmission, and they include donor screening. I don't need to go in detail. These are very well known steps which we're all familiar with, and donor testing and farm worker vigilance and inactivational removal and that is to validate the manufacturing process for its capacity to clear the viruses.

Therefore, the end of viral validation studies to provide evidence that the production process will effectively inactivate or remove viruses which could potentially be transmitted by the product.

So these are basically clearance of the relevant viruses of those that potentially could be in the product, and also to provide indirect evidence that the production process has the capacity to inactivate or remove yet undetermined virus contamination, basically to show the robustness of overall clearance capacity of the manufacturing process.

These are the list of some of the more common and established viral clearance methodologies. This list is by no means complete, but these are the one that we frequently see mostly in plasma derived product.

That overall is divided into viral inactivating agent and viral removal. The inactivation could be chemical, which such as solvent detergent, or physical or it could be a combination of both like for the chemical product which it uses chemical inactivated by exposure to light, and removal which may be part of the manufacturing process, could be precipitation and chromatography, or it could be deliberate like inclusion of some other viral filter like nanofilters (phonetic).

The process of viral validation includes scaling down the manufacturing process steps, the steps that need to be validated. They're scaled down because the validation cannot be done in the actual manufacturing process. It is not desirable from the GNP standpoint and also is not technically feasible.

and then that step or steps which need to be validated spark with high titer of infectious virus, which preferably will be the relevant virus, and if the relevant virus is not available, that specific model virus will be used for this purpose.

Then the subsequent step will be the viral titer will be determined the subsequent step, and finally the viral reduction for each validated steps will be added together and total log reduction will be recorded, and that will be overall the capacity of the whole manufacturing process in clearing the virus.

Therefore, the viral validation study has a number of validation components which need to be looked at. One will be the choice of viruses, which is used for the spiking experiment.

Design of the small scale or laboratory scale model which is being used, and kinetic solvent activation and thermal inactivation and thermal removal, the mass balance, and also robustness, and that is the deliberate changes in the manufacturing process and its capacity to retain the clearance with these changes, and also the sensitivity or infectivity assay which is used to determine the viral titers. And finally, the total log reduction which is provided by the whole processes.

In terms of virus selection, the viruses that can potentially be transmitted by the product, which refer to as the relevant virus, or if the relevant virus are not available, the viruses which toxinomically are similar to the relevant virus and also in terms of their physiochemical characteristic will be chosen, but the first preference will be that the relevant virus needs to be used.

And in addition to that, viruses with wide range of physiochemical properties to evaluate the overall robustness of process, and these are usually small, highly resistant virus which will be included in the validation studies.

Sometimes using the relevant virus, the panel which is used, using the relevant virus basically will take care of this second goal of the viral validation studies.

In thermal virus selection, the selection of virus basically depended on the nature of the starting material, whether it is cell line or human derived or animal derived, and there are some technical considerations also need to be looked at, and that is availability of suitable culture system for that given virus and availability of high titer stock and also reliable method for quantification.

This is basically a panel which more or less is used for validation of plasma derived product. I don't think I need to go through the whole panel. I assume we are all familiar with these.

The one thing I need to mention, that the inclusion of HIV as a relevant virus, it is a requirement, and also we need to have a specific model virus for HCV, and it's the panel of viruses which they are used to validate for the cell line derived product, and here HULV -- NULV is also a requirement for this type of product.

Overall the panel which is used for the spiking experiment should include DNN, RNA, viruses, should include both lipid envelope and non-envelope viruses and also in terms of size, should include small size all the way to the large sizes, and in terms of resistance from low resistant to high resistant, and the intention here would be to basically cover a wide range of virus with different characteristic, and by that it will have the capacity to remove or inactivate some of the viruses which they are not included in a validation process.

In terms of experimental design, usually the small scale is one-tenth to one-hundredth of the large scale manufacturing process,a nd the principle here is that the small scale must mimic the actual manufacturing process in order for the result to be relevant to the manufacturing, the actual manufacture.

And that, for example, in the case of chromatography, the buffer pH and protein concentration of the product should remain the same.

There a number of component parameters which are used depending on which step is being validated in order to establish that the small scale may be equivalent to the actual manufacturing process.

And an assessment of the validation package, these are basically the essential component that we looked at. That is important because that was really established whether the results are relevant to the manufacturing process or not.

And as indicated here, the operational parameter, the critical both process and operational parameter must be kept in the smaller scale and must show the product in terms of the specification for that particular step or steps is equivalent to what we have in the actual manufacturing process.

The viral validation package are evaluated on a case-by-case basis depending on the type of product and also manufacturing process. For example, this is what we look at if we're evaluating the photochemical processes which are used in the labile blood component. Here the concentration of chemical with changes in the donor and the plasma cell volume and leukemia and other impurities in a donor, whether the presence of impurities will affect the overall inactivating capacity and also the degree of impurity removal. If removal of some impurity is required, how much should be removed and what is the minimum requirement for that?

And basically the light or the intensity of the energy which is used to activate these chemicals that need to be validated, and the plastic bag transparency has been shown to have effect on overall capacity to inactivate. That also needs to be part of the validation package.

And the sample depth and finally mixing efficiency. These are all parameters that need to be validated, and the residual level of chemical or its breakdown in the product should be determined and validated.

Overall criteria for an effective virus clearance step should generate significant viral clearance and we see what we mean by significant. It should be reproducible and controllable at the process scale and also modelable at the laboratory scale.

What this basically says, that the validation study should be relevant to the actual manufacturing process, and these need to be demonstrated by producing data in the lab and also from those in the actual manufacturing process and should have minimum impact on the product yield and activity and should not generate new antigen or leave a toxic residue; basically should be safe.

With regard to plasma derived product, most of the dealing that we have is with the plasma derived product. The manufacturing process must contain two effective orthogonal steps for removal and inactivation of viruses, and at least one of these steps should be effective against non-envelope viruses.

And at least one stage in the production process must inactivate rather than remove viruses. That removal steps are difficult to scale down, and they are not robust. Therefore, complete reliance on removal would not be acceptable, and inactivation needs to be included in the process.

Under the best of circumstances, there are some limitations. Laboratory strain may behave differently than native, and there may exist in any virus population a fraction that is resistant, and the scale-down process may be different than the full scale, and source plasma or presence of neutralizing antibody may result on overestimation of the viral clearance.

And the ability of steps to remove virus after repeated use may vary. The question that frequently asks how much viral clearance is required, and this is basically it says the total viral reduction should be greater than the maximum possible virus titer. That could potentially cause in the source material.

For plasma derived product, we require, as I mention, at least two major significant steps and significant is translated to a step which produced more than -- at least more than four log clearance for a given virus.

And there should be at least three to five log excess clearance for the whole process, for the whole manufacturers, and there are factors which will be visited later on which may influence the TSE clearance, and that is the spiking materials that are used and how relevant are to human disease and the infectivity assays which they all use to determine the level of titer and also the type of the spiking preparation which will be used for this validation.

Thank you.

CHAIRPERSON PRIOLA: Okay. Thank you, Dr. Farshid.

Are there any questions from the committee? We have time for some questions.

Okay. If not, we'll move on to Dr. Scott, who is going to talk in more detail about clearance issues regarding the TSE agent.

DR. SCOTT: Dr. Farshid has outlined for you what is essentially the state of the art in terms of viral inactivation and validation of such procedures in the study of whether and how you can remove viruses from plasma derivatives during processing. And that's the context we want you to be thinking about TSEs in, but clearly we are talking about something that's quite different in terms of perhaps the nature of the agent, being a protein. That is still arguable, but certainly it may behave differently from a typical virus, and also in terms of the amount of load that you have of a TSE agent in a pool that is used to make plasma derivatives.

What I'd like to do is outline for you some of the similarities and differences between TSEs and viruses with respect to clearance in plasma derivative processing, and hopefully this will help you to think about the question that we are asking.

Dr. Farshid has already shown you some nice slides about validation of viral removal and inactivation, and certainly many of these premises are applicable to the study of TSE clearance. In particular, the scaling down of process steps, the spiking of appropriate steps, that is, ones where you think you might have a reasonable degree of clearance with high titers of infectious agent. this can occasionally be problematic for TSEs, and I'll discuss that a little bit later.

The determination of a reduction factor for each step, I think, is pretty standard and certainly is applicable to TSE studies, and the summing of reduction factors ideally from orthogonal processes to give a total log ten reduction value.

As you will be hearing, the method of removal of TSE agents can be similar to that of viruses, that is, in some cases, for example, filtrations may work. However, we don't have a known method that can be applied to plasma derivatives to actually inactivate the infectious agent. The ways that in activate TSEs are clearly more rigorous than the proteins and plasma derivatives are likely to withstand.

Dr. Farshid has shown you that it's quite well established that one can scale down purification steps and have ways of checking to see if the lab scale model is close to or similar to the actual production scale. It's obviously necessary to do this because you can't have infectious agents in your facility. It would be completely impractical for many different reasons.

One thing that's really important in the TSE studies is how you detect your TSE agent, whether you use infectivity assays or in vitro assays which detect PrPSC or some combination of these, and these are important issues to discuss, but clearly the assay that you use to tell whether or not you've reduced the infectious agent is of critical importance.

This is just a schematic of how a study may be done for a single step. So, for example, you'd have a spike of a TSE agent. I'll go into what kinds of agents are typically used in a minute, and in this case it would go into starting plasma usually before it's frozen as it's thawed. You can make cryoprecipitate, and in that thawing step you remove precipitate at a very low temperature which contains Factor 8 ultimately or is refined to make a Factor 8 product.

The supernatant from this cryoprecipitation is used to make nearly every other plasma derivative, and so what you need to recognize is that when you have this kind of study, you'll generate a clearance value. However, it may be with respect to the cryoprecipitate or the other products.

So very frequently in plasma fractionation you actually have a precipitate with a product and a supernatant with a product, and that needs to be kept straight.

This is just an example of a typical Cohn-Oncley plasma fractionation process which is used beginning with plasma to make a number of different products. So, for example, here you have Factor 8, Factor 9, antithrombin III immunoglobulins and albumen, and what is happening here is that you're undergoing sequential precipitations under somewhat different conditions of pH, ethanol, and ionic strength or salt in order to isolate the proteins of interest that will be made into products.

So the point of showing this is really to tell you that almost all manufacturers, well, I would say all manufacturers do this somewhat differently. The separation of fraction two plus three can be one plus two plus three, but even in these fine details of pH, buffer concentrations, and so forth, we'll differ among manufacturers. Not only do you have these differences. Cohn-Oncley is not the only method used to make some of these products. There's a Kistler-Nichtmann method. There are purely chromatographic methods.

So I'm just showing you one example of one type of plasma fractionation.

Furthermore, there are refinements before you get to your final product which involve both viral inactivation methods, but also chromatographic methods and other ways in which the products are further purified beyond the Cohn-Oncley procedure.

So this gives you a sense of the complexity of plasma fractionation, and I also want to give the message that each manufacturer's process is different, and therefore, reliance on literature studies of removal of TSE agents by plasma fractionation are really only going to be useful if all of the conditions are met that are actually used in manufacturing.

So we don't think that you can take an article and apply it to an actual manufacturing process that is different in any way.

There are a number of challenges in the studies of clearance of TSE agents, and I'm going to outline some of them here. the first question is what source of infectivity to use. The obviously easily available material is from experimentally infected animals, usually rodents. It's clearly quite difficult to obtain a large amount of brain samples from infected people.

Furthermore, you need a BSL-3 facility to study variant CJD or BSE. It has been demonstrated in one published study that PrPSC partitioning is similar when the source is human infectivity or animal TSEs.

Another question is, and this is a very important one and very difficult to address, we simply don't have the information. What form of the infectious agent is most relevant to blood?

The forms of the agent that have been used in spiking experiments so far have been brain homogenates. Subcellular membrane fractions and a very popular one is microsomal fractions, which are obtained through a high speed centrifugation of brain homogenate, and membrane free infectious material that's been further purified. I'm sure Dr. Bolton could tell you a lot about it.

There's also a potential problem with the lower limits of assay sensitivity so that to detect infectivity using an in vitro method, usually you can only go down or the sensitivity of the assay is two to three logs or infectious -- logs of infectious doses, and there's an upper limit of titers available for spiking when you make a brain homogenate or another type of preparation so that the range of infectivity removal that you can demonstrate is typically around four to five logs. That makes throughput experiments or experiments where you want to couple steps somewhat more difficult because they're limited by the range of inactivation that you can detect.

It's also an important question to ask: which assays are best to measure the outcomes of your experiment. Certainly in vivo infectivity can be done. It takes a lot of time and an enormous number of animals and is quite expensive. But certainly I think that people here in the audience who have participated in these studies will tell you that you might need a year or two to get your answer.

There are several different in vitro assays for PrPSC, and in at least one case there's a published study which correlates an in vitro assay with an in vivo bioassay.

In addition, there are concerns about mass balance. So sometimes TSE agents can be retained by columns, and what you put into your experiment is not what you get out, and this is something that needs to be addressed in certain cases.

I didn't plan this. It's just coming out that way.

In addition to the challenges with the studies themselves, we certainly will have challenges in evaluating studies if and when they're submitted in a similar fashion to viral validation packages. We need to think about how much reduction of TSE infectivity is enough.

And to that end, we have the model risk assessment that Dr. Anderson will be presenting soon at least as a way of thinking about how much reduction is enough. Certainly as I mentioned, the loading amount or the amount of TSE infectivity in plasma, if any, is likely to be significantly lower than you would have for an infected or virally infected donation.

How many disparate clearance steps should there be? In other words, for viruses we have asked that there are at least two orthogonal steps of removal, in other words, two steps that work by different mechanisms, and they may be less achievable or less easily achievable in the case of TSEs.

And in a similar vein, what steps can be summed and which cannot? Which are redundant and which are not ?

I just want to reiterate the point and give some examples to demonstrate that TSE clearance depends upon specific characteristics of the starting material as well as the process conditions, and I'm just going to briefly mention three examples: partitioning of infectivity and how it may depend upon the pH ionic strength in alcohol concentration, cryoprecipitation methods and how those may influence the degree of TSE clearance and the effectiveness of depth filtration, which appears to depend both upon the filter used and the properties of the starting material.

Again, I'm trying to make the point really that these are not generic studies, that they need to be very specific, and this first example from a publication in 2002, scrapie brain homogenate was spiked into buffers that had various alcohol concentrations, salt concentrations or pH.

These were incubated and centrifuged to make a precipitate, and the PrPSC was measured in the supernatant, and here you're looking at Western Blots.

But the points, this is perhaps a bit difficult to read, but here you're looking at pH variations studied at five, 6.5, and eight; alcohol concentrations at 25 percent; and salt concentrations.

And basically what you can see is that indeed there are differences when you vary any of these parameters,a nd actually in this paper they construct a model which is predictive for partitioning of PrPSC in steps and plasma fractionation.

But if you just take one example and look at pH of eight with no alcohol and two different salt concentrations, you can see that the supernatant still retains a lot of PrPSC if you increase your alcohol concentration to 25 percent in the context of a higher amount of salt. then you have removed most of your PrPSC from the supernatant.

And if you look at this figure, you'll see two more examples, but I just simply want to make the point that the conditions under which you precipitate matter.

In another example, there have been varied reports of clearance of TSEs during cryoprecipitation, and two reports suggest that you get clearance in the effluent, in other words, not in the cryoprecipitate, and a different report suggests that you actually get about a log of clearance in the precipitate or for the precipitate itself, and this would appear to vary perhaps because of the method of cryoprecipitate recovery and also perhaps because of the spiking agent that was used in these studies.

And finally, depth filtration has been variably reported to be effective or ineffective in TSE clearance, and I think that's because not all depth filtrations are the same and not all starting materials are the same.

So, for example, here you have sequential depth filtrations for or two depth filtrations for Fraction 5. This was published in a paper by Foster, and you can see that the two depth filters have different degrees of removal of the TSE agent. This was done using a microsomal spike.

Here you have in a different study by Vey and Biologicals the same depth filter, but different degrees of removal of the agent, and so, again, not all depth filtration is the same, and you can't just assume that you're going to get removal without looking at the specific process.

So manufacturing processes are highly individual, as I've already discussed. Rigorous demonstrations of TSE clearance need to be based upon the specific manufacturing process, although published studies can be useful to identify steps that have potential for TSE clearance.

And I've just listed the studies for TSE clearance that have already been published for plasma fractionation. Some of these are quite detailed, not at the level of detail that we would want in a validation package, but that would make a paper much larger than any journal would accept.

But you can see that there are ten here and some are by the same groups, but they encourage us that these studies can be done, and they can be well done with the caveats that I've mentioned.

What would we do to evaluate TSE clearance studies from industry to support labeling claims of lowering possible TSE risk? We would need to have a characterization of the spiking agent used, where it came from, how it was made, and how the titer was determined.

There would need to be accurately scaled down processes with reproducible experiences. It's not enough to do an experiment once.

A well characterized assay that reflects TSE infectivity, although it would be reasonable to bridge binding assays to bioassays for certain steps. In other words, it might not be necessary to do a bioassay each time for each step, for each iteration. There would need to be an estimated log's clearance of TSE by processing steps, a reduction factor and a clearance factor. That just depends on how you measure your starting material, a demonstration of mass balance, and a demonstration where relevant that non-orthogonal, that means similar clearance steps are or are not additive.

When we come to the more difficult portion of evaluating submissions, we would want to provide a consistent evaluation, and the question that we need to address is what is the clearance belt line that would support labeling for products.

For example, similar to viruses, at least two orthogonal steps with four logs or more clearance for a total of eight logs. I'm not putting these out as decisions. I'm putting these out as current thoughts about this because it's quite a difficult issue to address, and in order to do it, I think that we will have to work further with our risk assessments and also with the capability of these processes.

At least two steps demonstrated to be additive with four or more logs clearance per step for a total of eight logs. Perhaps at least two steps, either orthogonal or demonstrated to be additive with more than three logs of removal per step.

An important question is: is a single clearance step of four logs or more sufficient if it's very robust and reproducible in the lab experiments?

And a question that I think has been asked around lately: are clearance steps of two logs or more -- I'm sorry -- two logs or less reliable if they are robust and reproducible?

I would point out that the clearance mechanisms for TSEs appear to be mostly involved in partitioning. As I mentioned we don't have an activation, and classically partitioning is thought to not always be quite robust or reproducible. So it would be important to demonstrate that in any case.

So I finish with the labeling for TSE risk. The current proposal which I've already mentioned, so this is exactly the same as what I had in the first set of slides, but I do want to point out that future improvements in risk assessment, understanding the nature of plasma infectivity and improved study methods could still provide a basis for even additional labeling content beyond this suggestion here.

Thank you for your attention.

CHAIRPERSON PRIOLA: Okay. Thank you, Dr. Scott.

Are there any questions for Dr. Scott? Comments? Dr. Belay.

DR. BELAY: You said the different companies have different manufacturing processes and that this could actually -- it may affect the clearance process. Do you have any idea how many of the different manufacturing processes actually were tested for TSE clearance, one, two, or, you know, the majority of them?

DR. SCOTT: I believe four, three U.S. ‑- I'm sorry -- five and two European. Not everything is published, however.

DR. BELAY: So five total?

DR. SCOTT: That's corrected.

DR. BELAY: How many different manufacturing processes?

DR. SCOTT: Steve is probably going to correct me though.

DR. PETTEWAY: No, I was just going to say Hank Baron actually has a complete breakdown for the --

DR. SCOTT: Okay. Well, actually I think there are more. In fact, I've already thought of another non-U.S., but most of the -- not most, but quite a few of these studies aren't published and are in process, and that's right. I think the PPTA slides contain more of a listing of the numbers of studies that have been done.

So in terms of published, four.

DR. WOLFE: This I'm sure we'll come back during the overall session when we're evaluating or voting on this, but from your perspective, what is the problem as in with the present label that leads you to propose to us that we change the label and that the present label is, I suppose, more alarming in some sense than the one that's proposed?

The present just says there's some risk, and then the new one says, but because this manufacturer has done ABC, this risk is remote, and so forth.

So just general answer to the question: what is the problem for which FDA is seeking us to solve it by putting a very different kind of label on?

DR. SCOTT: Well --

DR. WOLFE: Is it a regulatory? Does the FDA want to get more regulatory authority so that it can compel these companies to do the kinds of things that would allow them to have this other kind of label?

What's going on here?

DR. SCOTT: Right. Well, I think you've hit the crus of the matter actually. First of all, I would point out that permission to add something to the label is not compelling anybody to do anything, but I think what we see now in the literature is studies suggesting that you can remove TSE agents by probably many of the processes fortuitously that are used in plasma derivative measuring.

But what we don't have is the level of scientific data that we would like to have to more clearly tell us that the risk as best as we can tell is reduced, and that involves actually looking at raw data, the assay methods in detail, looking at things to a far greater extent than can ever be published in a paper.

So we would feel more secure in the risk reduction if we actually had the data to evaluate, and we would feel that if the data were good enough and the models were relevant enough with the current scientific limitations, that it would be reasonable to convey the possibility that risk is being reduced by the manufacturing process.

DR. WOLFE: Just to follow up, does that mean then that the companies who would want to be able to use this new label, let's assume theoretically that the committee agrees that this new label is better; it is less alarming, whatever. but if the companies want to be able to use it, they would have to go through whatever hoops are being described here. You have to describe their manufacturing process, but if companies chose not to use the new label and to use the current one, they would not have to do it?

I mean, again, you said you're not compelling anything, but unless this is a regulation across the industry, some companies may choose not to go through the various hoops necessary to get to the new label and would continue to have the old one.

And from a public perspective it gets fairly complicated. You know, am I getting plasma derivative from a company who has decided to go with the old label or am I getting it from one with a new label? I mean, what's the overall thinking on this?

DR. SCOTT: First of all, I just want to point out that the new label doesn't say there is no risk.

DR. WOLFE: No, I understand that.

DR. SCOTT: It just says that risk is reduced.

DR. WOLFE: But it's clearly a reduced -- I mean, the new label compared with the old.

DR. SCOTT: Right, right, and it would be -- yes, this would be an incentive perhaps to perform and submit these studies which are not inconsequential.

DR. WOLFE: Right.

DR. SCOTT: But it wouldn't be a requirement, and you're quite correct that it could result in labels that are different, but we've had this in the past. I'm sure that not all viral inactivation labels were slapped on at exactly the same time once viral removal or inactivation steps were implemented in plasma derivative processing.

But you're exactly right. There may be different products with different labels.

DR. WOLFE: Including ones with the current label because they aren't willing to go through what you're --

DR. SCOTT: That is correct.

DR. WOLFE: Okay. Thank you. That answers my question.

CHAIRPERSON PRIOLA: Dr. Epstein, do you have a comment?

DR. EPSTEIN: Yes. Also in response to Dr. Wolfe, I think it's important to make known that there is a driving force, that there are companies that have expressed interest in a revised label suggesting mitigation of risk.

Now, in fairness, FDA put a lot of pressure on the companies to generate the kind of clearance data that you have seen and more what you're about to see. We did this because there was a concern about the safety of the products and a lot of unknowns.

And these data have been used by the FDA in considering its policies, for example, with regard to removing the recommendation to withdraw derivatives if a donor subsequently was shown to have classic CJD or classic CJD risk factors, and as Dr. Scott outlined earlier, our decision not to extend the deferral for five years' exposure in any part of Europe to plasma donation.

So these data have been very material, but the driving force right now is that we're being asked whether to permit a revised label, and that provokes all of the questions that you asked. You know, what will be the FDA standard? Will we allow it at all? If we allow it, will we allow different labels for different products? Will we allow the old labels to co-exist with the new label?

And the bottom line is where does truth in labeling lie here? Are the studies a sufficient basis for companies to make what will clearly be interpreted by the public as a claim of supervisor safety for their product?

And of course, that implies competition over relative safety. That's a good thing to the extent that it produces real improvements in safety, but to the extent that it may not reflect true differences either in safety or the assurance of safety, then it would not be a good thing in the market.

So that's to figure out. What should be the scientific standard? And then we're not really asking you the regulatory question, which is therefore what should we do, but we will have to make decisions what labels to allow. Will it be uniform? Will it be boilerplate or will it be case by case?

I think what you've heard Dr. Scott argue is that it needs to be case by case based on the nature of the scientific data, but that's one of the questions we're going to ask you.

CHAIRPERSON PRIOLA: David, did you have a quick question?

DR. BOLTON: Yeah. That was a very good response, Jay. I like that.

My question is sort of about the products and about the label. It seems to me, and maybe I'm wrong, that most, if not all, of these products are not sort of an optional thing. I'm going to take this or maybe I won't take it, first of all.

And secondly, the ultimate consumer is the patient, but the person generally making the decision on which product to use is a physician. So how effective -- I mean, a physician is not going to say, "Well, gee, I'm not going to give this to my patient because this label says, you know, that this is not as good as this other one."

I mean, they may make a decision between source A and source B, but they're certainly not going to give the product because the label doesn't look good. Is that right?

DR. SCOTT: Well, it's an interesting question, and we haven't performed a survey, but my guess is that our industry representatives might have something to say about that. They may have more of a finger on the pulse, but I agree with you that the person who reads the package insert, if anybody, is usually the health care providers.

But we do have a population of well informed people who receive these products regularly. So I should modify my statement because there are a lot of people, I believe, who would pay attention.

CHAIRPERSON PRIOLA: I know that there are members that have other questions, but we have sort of a time limit in terms of how long they're even going to serve lunch. So --

DR. BOLTON: I appreciate your need to move the meeting forward.

(Laughter.)

CHAIRPERSON PRIOLA: My decision making ability dropped precipitously the longer I go without food. So I'd like to move on to Dr. Anderson, and please keep your questions in mind so we can ask them after lunch and have the discussion. And Dr. Anderson will talk about risk analysis for TSE and plasma derivatives.

DR. ANDERSON: Okay. While we're waiting, my name is Steve Anderson, and I'm actually the Associate Director -- and which one of these works? -- I'm the Associate Director for Risk Assessment at the Center for Biologics. I'm actually in the Office of Biostatistics and Epidemiology, and no risk assessor would actually give a talk without putting up the types of frameworks they work under, and I'm not exception.

Shown here is the WHO risk assessment framework that's commonly used in microbial risk assessment and chemical risk assessment.

And I'm just putting this up as a frame of reference. I'm not going to talk about some of the components. I think Dot has sort of given some nice -- Dr. Scott has given some nice background. So I'm not going to talk about hazard identification. What I'm going to really focus on is exposure assessment, the frequency and level of exposure that could occur specifically for variant CJD, the potential for variant CJD, and Creutzfeldt-Jakob disease agents in plasma and plasma derivatives.

And then finally I'm going to skip right to the end and do the risk characterization and look at predictions from the model, and I'm going to show you those results, and I'm going to go somewhat quickly in favor of the sort of lunchtime that we're all eyeballing, and basically the basis of quantitative risk assessment is a computer model, and that's what I've generated, and it's a way to link the relative data together in a meaningful way.

And I think it's very important and very useful in a way because we can understand how something upstream, for instance, a manufacturing process, a reduction step, affects something going on downstream, which is a patient or recipient receiving a product and something that happened upstream, meaning what that patient is going to experience.

Also, quantitative risk assessments, with these frameworks we can estimate potential exposure and risk. I'm going to do that for you in the results at the end, and it's a framework to identify critical elements where research will improve the model and specifically where research is needed, and there's a big need for research with this issue, in particular, and I'm going to talk more about some of the things that are needed just based on the findings in the risk assessment.

And then we can also with this framework understand key elements that drive the risk. What are the key things that contribute to risk in this case?

And I thought I'd put these two things out on the table because we deal with them constantly in risk assessment, uncertainty and variability, the uncertainty is what we don't know. We have little or no data available in some cases, few replicates of some of this data for some of the processing and reduction steps.

There's a lot of variability, variability in the reduction steps and processing steps. Detection and measurements are an issue, and these are especially an issue when the reduction is small, on the order of less than one log or less than two logs.

Now, the objectives of risk modeling is to quantitate the relative contributions or parameters in the model and identify -- I've just sort of said this -- identify critical elements for research, but importantly, sort of provide information about risk for making regulatory decisions, such as the one that's posed to us today.

Now, one value of quantitative risk assessment modeling is that it allows us to do what's called "what if" analysis. So what we're doing here today is we're looking at CJD in our donor population. We know that exists at about the rate of one case per million.

We're also interested in the potential for variant CJD in the U.S. donor population, and others before me have talked about that. I'm not going to go into that. We have populations in the United States that could be at risk, could donate blood, impose a risk to others.

What we did in our modeling was we separately modeled CJD and we also jointly modeled the influences of CJD plus variant CJD, and what we did was we assumed a rate of one per million for variant CJD as well. So we assumed these two things were equal, and we're looking at the impacts of those assumptions.

And in our "what if" analysis, Dr. Scott just talked about this, and Dr. Farshid. The standard for bioclearance is four logs for a processing step. Now, this is difficult to achieve, as Dr. Scott said, difficult to achieve that level for TSEs.

So what we were interested in is we were looking at two different scenarios in this case. One was steps that are greater than two logs reduction, what are those impacts. So let's just let me remind you that this is not only going to include steps greater than two logs, but the steps greater than three logs for clarification.

And then we also looked at steps above three logs clearance as well. So we had two scenarios. One looked at three logs and one that looked at two, and it should say plus three logs down here for clarification.

And the scope of the risk assessment. We tried to limit it. We limited it to three products: albumen, immunoglobulin and Factor 8.

There are other products that we could have looked at, but we limited it to these three, and what we were interested in again, CJD and variant CJD, and what's the potential exposure risk posed, and what level of reduction are we seeing through some of these plasma derivative processing steps?

Again, just sort of quickly moving along, I'm going to talk about some of the background for CJD and donor populations because those are an intimate portion and a major part of the modeling that I've done, but basically David Asher has talked about this. Basically he showed the same type of data that I'm going to be showing, but the mean age for sporadic, again, is 65 years, a mean of 29 years for variant CJD.

Now, if we look at that in comparison to our donor rate, rate of donation, so looking at the U.K. variant CJD, cases from 1995 to 2001, and those are here in red, and then the U.S. CJD cases in 2000, those are in the yellow. You can see the mean-median is around 65 to 70 here, 29 here.

But then I tried, in my limited ability to do computer stuff, I tried to match these up so that looking at the donor rates below and try to match these up, there is quite a large, about 18 or 19 percent of the population below 25 years are donors, and then it drops, and then it begins to increase.

But if you go and match this up to this area, a significant number of donors if they donate especially donors donating in this region, again, could be at risk if we did have variant CJD in this country.

And then the donation rate goes up, and then it starts to drop again. I think one good thing is this second peak coincides with this sort of valley here, and then it starts to drop down, and just as variant CJD is picking up or -- I'm sorry -- CJD is picking up.

So that is the type of information we're using in the model. We're using prevalence of CJD and presumed variant CJD assumption in our donor pool, and what we're also doing is we used a method called back calculation to look at the estimated number of people incubating the disease that could be asymptomatic, both CJD alone and then we did both together, variant CJD plus CJD with our assumptions of one per million for both of these diseases.

We also have to acknowledge again uncertainty in the model, and this is basically just a thumbnail sketch of what was done for the modeling and the components that went into the modeling.

So here's the donor population. We looked at that. We looked at the age and frequency as shown in that graph, and we also looked at the prevalence of CJD and variant CJD. We did back calculation to look at the number of asymptomatic individuals.

And what all of this information was trying for was if there was an arrow here, it was looking at TSE positive pools that could be present among all the pools of plasma that are used to produce products in the United States.

And of here, this side of the model is specifically looking at contaminated pools. So what's the characteristics of a contaminated pool? And what we were interested in was the amount of ID-50s that might be in a contaminated pool and further characterization of that.

So we're interested in recovered pools, in source pools. I'm going to explain this more in depth in a minute. I'm just sort of giving you a thumb's view.

And plasma processing steps. So what are the steps used? Now, this is one general model, but we did this for each of the products. We did one model for albumen, just like this sketch, one for Factor 8 and one for immunoglobulin. So separate models for each of those products.

And the plasma processing and reduction levels that we worked with for each product varied, and I'm going to show you a little bit of that in the results section. And what you get is contaminated pools coming down this side, and you're looking at the number of contaminator or percentage of contaminated pools coming down this side, and what we want is to know the percentage of contaminated pools potentially that could be out there and how much contamination could be in those pools.

So we used this to do this prediction for all pools, the amount of TSE agent that might be present for all pools.

We also then linked this to production data, how much albumen is produced, how much immunoglobulin, how much Factor 8 in the United States, and then how that product is utilized. How is albumen utilized? Who uses it? Who is Factor 8 utilized? How much of it is utilized by patients, et cetera?

And finally what we want to get to is exposure and somehow risk got cut off there, but it's supposed to be risk in here as well. So we ultimately want to look at predictions of exposure and risk.

Just to sort of give you an idea about the kinds of data we use, we use census data. We used eight specific CJD death data in the U.S., variant CJD data for the U.K., age specific donation data, and I should acknowledge that we got data from the RAD study from the National Heart, Lung, and Blood Institute, funded by the National Heart, Blood, Lung Institute, and we used that data specifically from the year 2001.

We made assumptions regarding screening and processing. There were data on the quantity of TSE agents in plasma. I'm going to talk specifically about these in a minute. Data and some assumptions on the amount of manufactured products and how they were utilized.

All right. This is just a few of the parameters that we used to give you an idea of what we did in the model. Some of the things that we looked at were effectiveness of the donor questionnaire for CJD. How effective was that questionnaire?

We assumed in this model that it was 20 percent effective. We have an effectiveness rate for the donor questionnaire for variant CJD. We assumed it was 90 percent.

Now, these numbers can vary. I didn't vary them in this particular run of the model, but I think in later steps we're going to go back and vary these to look at how variation in these questionnaires will affect the outcomes of the model.

And what we've got here are the distributions that we used specifically for CJD infectivity. So we used what's called a triangular distribution, and this is in ID-50. So one ID-50 was the minimum.

The most common or the median was ten ID-50s per mL of blood, and then a maximum of 100 ID-50s per mL of blood, and then for variant CJD, very similar. One, 100 ID-50s per mL of blood, and then 100.

Now, this is blood remember, and what we're interested in is plasma. So the percent of infectivity that could reside in plasma, basically our most like was about 30 percent. The left side of that distribution was 25 percent. The right side was as high as 40 percent of that distribution.

We had fixed pool sizes for donors. Sixty thousand donors went into a pool that was used to make product. For recovered plasma, if there were 60,000 donors that donated recovered plasma, that would lead to 12,000 liters as the total size of the pool, if it was 60,000 donors.

For source plasma, it would be about 48,000 liters.

We took a sort of conservative approach in our assumptions about variant CJD and CJD incubation. We assumed it was from four to 40 years, and we based our back calculations on this. So quite conservative, and that really just generates our lack of understanding of really what that distribution should look like.

So we need more data actually, especially for this parameters.

Just going on, I'm going to talk about albumen, and I'm going to just give you some of the background for albumen and then the results of the modeling.

And what I'm talking about here is from the Scottish National Blood and Transfusion Services, and all of the data I'm going to present is just based on that Scottish data, from Foster, et al., and Vox Sang.

And basically for albumen, we have three processes that I looked at, three reduction steps. The most likely for ethanol precipitation of Fraction 4, again, we assumed three logs reduction. It could go anywhere from two logs all the way up to 3.1.

We assumed 4.9 for Fraction 5 depth filtration. It could be as low on the left-hand side of that distribution, as low as four logs of reduction or as high as five logs on the right-hand side of that distribution.

Again, for Fraction 5 depth filtration by another filtration methods about 2.3 was the median value for that.

Okay, and then under the model results, and what we assumed, based on data that we had from FDA was that $95,000 kilograms of albumen are distributed in the U.S. annually. If you break that out, it's 25 grams of albumen per liter of plasma, and so you would need about -- this should be 2.5 million donations of source plasma. You would probably need more if it were recovered plasma.

The number of pools that were in the model that we developed, we did a model just based on all recovered plasma pools. You would need 880 pools of recovered plasma or 220 pools of source plasma.

Again, source plasma you're getting 800 mLs on average from a person. Recovered you're getting about 200 mLs.

And just going down here and skipping to the bottom, about to my surprise 72 percent of all pools just for CJD might end up contaminated and as many as 85 percent if we add in the variant CJD. Add in the variant CJD infectivity, and that includes the back calculation estimates as well.

So the results. Even thought you have a number of perhaps contaminated pools, you have a significant amount of reduction going on. So for the three log reduction, which is the lower level of reduction, I'm just going to point you to 2.7 times ten to the negative nine, and this is the potential ID-50s per gram of albumen.

I think the difficulty for albumen is we don't really know how much albumen people are getting on an annual basis, and that's really one of the big unknowns. So I sort of skipped that, but we don't know how much albumen is in products out there, how many medical procedures are using it. So we just did our estimate on a per gram basis.

So as far as the potential ID-50s, ten to the negative 11th for a two log reduction, about ten to the negative ninth for the three log levels of reduction. Remember this is going to have one additional step.

The CJD plus variant CJD. The numbers actually don't change much, surprisingly. We're, again, at ten to the negative 11th and ten to the negative nine, and I'm just going to go on to immunoglobulins. Again, the reduction steps that we looked at again, based on the Scottish data, Fraction 1 and 3 precipitation where you actually remove this, you're getting a most likely of 3.3 logs of reduction and then for Fraction 2 depth filtration about 2.8 logs of reduction, and then we've got on either side of those 1.8 to 2.9, et cetera.

So let me just sort of skip to the end. We have 23,000 kilograms of immunoglobulin distributed in the U.S. You get about three grams of immunoglobulin per liter of starting plasma. So approximately nine million donations go into that.

The number of pools again, with 60,000 donors for totally recovered plasma it would be 640 pools and 160 pools for source plasma. We did have some utilization estimates here, and I think we can definitely refine and get better data for this, but just some of the preliminary data we have is that recipients get immunoglobulins every three weeks at about 400 milligrams per kilogram body weight. That works out for a 120 pound person to be about 380 to 400 grams annually that they might receive.

Again, significant level of potential infectivity in the pools. Again, the levels that we're seeing predicted from the model, if you go to three logs, all steps three logs and above, about five times ten to the negative six, all steps two logs and above, again, more levels of reduction. You're getting about 3.3 times ten to the negative eighth, and here are the 90th percentiles to the right for the predictions.

Again, if we add variant CJD into our estimate, you get a slight increase in some of these numbers. Basically they're very comparable on both CJD and CJD plus variant CJD. So not much coming from the variant CJD, even though we had a conservative prediction.

And then finally if we have a potential for immunoglobulins, let me just go back. This was an estimate per gram basis of immunoglobulins. This is an estimate of per 380 grams, again. This is the amount used in the treatment of a person, 120 pound person, again, ten to the negative fifth down to ten to the negative third, ten to the negative fifth, ten to the negative third.

I'm going through these somewhat quickly, and this is the assumption that we use for Factor 8. Again, from the Scottish National Blood Service, the Foster data. We assumed a most likely value of three logs reduction, a minimum of two logs, as much as 3.1 logs, and basically the numbers for that.

There's 375 million units distributed in the United States, 150 units per liter. So approximately three million donations, 3.2 million donations are needed of source plasma. If you use recovered plasma, you would need more.

Number of pools again of 60,000 donors for recovered plasma, again, you'd need 220 pools to make up your 375 million units. You'd need 55 pools to make up your 375 million units up here as well.

Utilization estimates just for Factor 8, the product is used on an episodic basis, but also some people may use it prophylactically. The estimates go anywhere from 60,000 units to 100,000 units a year. We chose somewhere in the middle at about 80,000 for doing our risk calculations.

So one dose we figured was about 250 units. Again, I think it sold in units of 250, 500 and 1,000 or more, but we just picked the lowest volume and lowest amount. Again, we're getting about ten to the negative fourth here, ten to the negative fourth, a little bit higher for variant CJD when it's added in, and the 90th percentile is on the right.

Now, if you look at the annual utilization where a person might be on prophylaxis of like 80,000 units, the mean there was about .03 ID-50s that may be expected in a year's supply, about .1 if we consider variant CJD in the equation as well.

Let me just go on and just to sort of summarize because I've sort of just done a shotgun of data, and here's the summary of what I've just shown you. Again, albumen on a per gram or per unit basis we're looking at. Albumen per gram, we were in the range of about ten to the negative nine to ten to the negative 11.

For immunoglobulins per gram basis, we were at ten to the negative five to ten to the negative eight on some of our estimates.

For Factor 8, ten to the negative fourth for our 250 units, and then an annual treatment there would get us down to about ten to the negative second or ten to the negative one even.

And just to identify some of the data gaps, this list actually would be about ten slides, but these are some of the big things that are needed. The amount of CJD, variant CJD HM present in the blood, in plasma, when it's present, understanding more about the disease and when the agent appears in the blood is needed, progression of variant CJD and CJD, information on that that's needed, the variability and the levels of blood in plasma, as I said.

Variability in the reduction steps, CJD and variant CJD, and the amounts of variant CJD and CJD agent that might occur during processing, and also interested in how these products are utilized. I think we need better data to get a better sort of risk estimate of how these products are used.

Now, I should say that what I presented is a concept model. So we're using the Scottish data, but we are aware of the extreme variability in the processes for manufactured or manufacturer. So we're proposing this as a concept, but certainly manufacturers -- we could take a manufacturer's data and plug it into this model and get a risk estimate similar to what I just showed you as an example.

And I'd like to thank basically this risk assessment team that we had at FDA. Dr. Scott, Dr. Asher, and then Dr. Rolf Taffs, this was really the core team, and then we had a lot of help from John Finlayson, who taught a naive person like me a lot about plasma derivatives, and Mark Weinstein also helped us with some of the data.

And then finally, I just wanted to acknowledge that we got information from Westat for the red study, and that was what we used for the donor data provided, and that donor data, I should acknowledge that it really is just for blood products and blood donations. It's not specific for plasma donations. So I think that's another data. We need specific information on age related donation rates for plasma as well.

And that's it. Thanks.

CHAIRPERSON PRIOLA: All right. Thank you, Dr. Anderson.

I think we'll move on to the final talk. Dr. Ruiz.

DR. GAYLOR: I have a number of questions.

CHAIRPERSON PRIOLA: Oh, you do?

DR. GAYLOR: And comments on this, and I don't think we ought to do it before lunch.

CHAIRPERSON PRIOLA: You don't think you ought to do it before lunch?

DR. GAYLOR: I think it will take several minutes.

CHAIRPERSON PRIOLA: Several minutes. Okay.

DR. GAYLOR: To go through my comments and questions. I mean, you could go through the next talk, but I'd like to go through the comments and questions while this is kind of fresh with us. I think it will take several minutes because I don't understand a lot of what was said.

CHAIRPERSON PRIOLA: I have several questions about it, too, in terms of ho wit was said. Let's go ahead with your comments and questions as long as they're fresh because we can possibly move Dr. Ruiz's talk to after lunch.

DR. GAYLOR: But there's the possibility that while the rest of you ate lunch I could ask questions of Dr. Anderson.

(Laughter.)

CHAIRPERSON PRIOLA: It's important to discuss it.

DR. ANDERSON: I think it's important to acknowledge that there are three models here, and it is really technical. So I don't think it's unreasonable to have a lot of questions.

I had like 30 minutes to present three models. So --

CHAIRPERSON PRIOLA: No, I think that's fine. If it's all right with Dr. Ruiz, if we can move your talk to after lunch, then we can go ahead and do this. I'd rather do it when it's fresh in people's mind myself than delay it.

So go ahead, Dr. Gaylor.

DR. GAYLOR: Well, first of all, I would not call this a risk analysis. I would call it an exposure analysis.

DR. ANDERSON: That's true.

DR. GAYLOR: I heard nothing said about what's the probability of infection or disease. I just heard things about -- your sixth slide, you're assuming that the vCJD rate is one in a million when you know in the United Kingdom that it's .3 to .4. Why are you going three times higher?

You know, I could see later on doing a sensitivity analysis and maybe using one, but why not use your best estimate? We're trying to make some decisions here. Let's base those on our best estimates and then we can do the "what if" if it's three times higher or three times lower or whatever.

DR. ANDERSON: Right. We can go back and do that.

DR. GAYLOR: Well, I think that's why you're getting 85 percent infectivity in your pools. If you start down there with three times lower, I'm guessing it will be down around -- it will still be high, 30 or 40 percent, but you know, that may change your decision and it may not. But I would certainly try and I would start with your best estimate, and then you can do the worst case scenarios.

DR. ANDERSON: Actually, if I could just clarify something and go back to the slides here, what we're getting for estimates actually just with CJD infectivity is 72 percent of pools. So we're not even including our variant CJD as a possibility.

DR. GAYLOR: Well, that's true.

DR. ANDERSON: So this is actually adding very little to the model.

DR. GAYLOR: So it would add perhaps -- yeah, I stand corrected.

DR. ANDERSON: I think where we're going to get more infectivity coming up is if we change the assumption on the donor questionnaire. If we move that from 90 percent down to 50 percent, you're going to allow a lot more cases of variant CJD into your sort of virtual pools, and your numbers are going to go up.

DR. GAYLOR: Well, your 90 percent may be pretty good. You should have some idea of how effective your questionnaire is in keeping people out of the pool that you don't want in there. So to me I would guess that 90 percent may be about right, but certainly you can change that number.

Okay. Let's go over to -- well, let's just pick one of these. About two thirds of the way through your talk you have a slide parameters in TSE and plasma, where you're talking about your model parameters.

DR. ANDERSON: Okay.

DR. GAYLOR: Well, in fact, that was one of these right where we were at here. The questionnaire was 90 percent effective.

DR. ANDERSON: Yeah, right here.

DR. GAYLOR: Yeah. Now, right below that I don't understand this next line. You say that CJD blood infectivity per mL, the median is ten, and that is ten what?

DR. ANDERSON: That's ten ID-50s per mL.

DR. GAYLOR: An ID-50 to me is the dose that causes a 50 percent infection rate, right?

DR. ANDERSON: Right.

DR. GAYLOR: And you're saying you're going to use ten times that rate as your best estimate, as your median estimate of what causes infection? Is that what you're doing?

I don't understand.

DR. ANDERSON: Ten ID-50s. So (pause) ‑-

DR. GAYLOR: You're saying -- you're saying that you're using for your median infectivity dose, you're using a number that's ten times higher than what you think is the dose that gives you a 50 percent infection rate?

I'm probably misunderstanding this because that doesn't make sense to me.

DR. ANDERSON: Well, I think this is -- maybe David or somebody can explain this better about ID-50s and how they're used in BSE and TSE work

DR. ASHER: Yeah, that was taken from the most frequently observed titer of infectivity in rodents. Occasionally -- it's never been observed higher than 100 per mL and seven, ten intracerebral ID-50 has been the most common observation in rodents, although the number of assays, of course, is very small.

DR. GAYLOR: I still don't understand what this ten is. Is it --

DR. BOLTON: The number --

DR. GAYLOR: Ten what?

DR. BOLTON: -- it's a measure of the amount of infectivity. So in terms of if you put it in virus terms, it would be, say, equivalent to ten plaque forming units, but the assay measure is a biological assay in animals. So you can look at a 50 percent infection dose.

So that one ID-50 would be the amount of infectivity that was required to infect half of the animals that were inoculated.

DR. GAYLOR: Okay. And then so you're assuming that the dose in humans would be ten times higher.

DR. ANDERSON: Right.

DR. GAYLOR: Which would give you 50 percent. Okay.

DR. DeARMOND: But that's never been demonstrated. The ten ID-50 units of CJD blood per mL has never been seen.

DR. BOLTON: No, but if you go to the animal models where you're not crossing the species barrier and where, again, I think David made the very excellent point that there's not a good scientific basis to assume that things that reproducibly occur in animal models don't occur at all in humans; so if you just say if we take the animal model as a true model for human beings and we say that there is -- I don't even remember what the titers are, but it's about one or something like that, infectious units per mL of blood. Then we just translate that into human infectious units in human blood. That sort of gives you a ball park.

I'm not saying that that's accurate by any stretch of the imagination, but it's our best guess probably.

DR. DeARMOND: All of the transfusion studies say that that's not true. I mean, ten is a lot.

DR. BOLTON: Human blood to primate.

DR. DeARMOND: Human to human.

DR. JOHNSON: This is IC. If you inoculate human blood, IC into other humans, we don't have --

PARTICIPANTS: Right, right, right.

DR. JOHNSON: So it's a very funny comparison.

CHAIRPERSON PRIOLA: In any case, it is a ballpark figure.

(Laughter.)

DR. ANDERSON: I mean, we're acknowledging it is a big ballpark.

CHAIRPERSON PRIOLA: Right. It's a huge ballpark.

DR. ANDERSON: Yeah.

CHAIRPERSON PRIOLA: Exactly. So you don't want to take it too --

DR. ANDERSON: Literally.

CHAIRPERSON PRIOLA: Right.

DR. GAYLOR: Okay. In your next slide, then, when you drop down to vCJD, so you had 100 then for your median value. You went up ten times that to 1,000 for your maximum. You went down by a factor of 100; is that correct now when you're looking at the minimum or is that a misprint?

I mean you went factors of ten on the line above. Did you intend for your minimum to be one then?

DR. ANDERSON: Yeah, actually that median --

DR. GAYLOR: Is that correct?

DR. ANDERSON: -- is going to go up for variant CJD.

DR. GAYLOR: Well, the median went up, but you didn't raise your minimum. Is that --

DR. ANDERSON: I kept that right there.

DR. GAYLOR: Is that an oversight or is that the way you wanted it?

DR. ANDERSON: That's sort of -- that is the way I wanted to.

DR. GAYLOR: Okay.

DR. ANDERSON: That can be changed.

DR. GAYLOR: No, that's all I wanted to know.

DR. SCOTT: Well, Steve, I think the reason you went up to 1,000 is the Lancet papers, which have their limitations for sure, but suggesting that the assays where they did look at blood for people with variant CJD, that the limit of sensitivity of the assays was approximately 1,000 ID-50.

So that's where the 1,000 comes from. It's sort of a worst case estimate based on these very small studies.

DR. ANDERSON: Thanks, Dot.

DR. GAYLOR: Okay. Then three slides beyond this for the results for albumen.

DR. ANDERSON: Per gram or --

DR. GAYLOR: Your last albumen slide.

DR. ANDERSON: Okay.

DR. GAYLOR: Yeah, that's the one. Again, I guess I don't understand. You're saying ‑-

DR. ANDERSON: These are ID-50s.

DR. GAYLOR: Right. But you're saying if you do all steps to log reduction, you have this 5.47 times ten to the minus 11. If you do a better job of reduction, three logs, then you're saying you actually have a higher -- or am I thinking just backwards?

DR. ANDERSON: Well, actually on this slide here that I'm showing up here, what we're looking at is everything above three logs, which are these two steps, which would give you a total of about 7.9. Then if you include everything above two logs, you'd get what, 10.2 as a total reduction?

So it's above two logs.

DR. GAYLOR: Oh, okay. That wasn't --

DR. ANDERSON: And then steps above three. It is confusing.

DR. GAYLOR: Okay. Well, that makes -- I understand it now. So what you're saying is you're way down there one billionth of the ID-50?

DR. ANDERSON: That's what we're saying.

DR. GAYLOR: Is that what you're saying?

DR. ANDERSON: At least in the model estimate.

DR. GAYLOR: Well, then I would say you've done a risk analysis.

DR. ANDERSON: Well, I actually agree with your first point because we haven't really predicted disease here or cases that could arise. We're actually just predicting --

DR. GAYLOR: Well, if you're that far below the ID-50, I would guess it's pretty close to zero, mighty, mighty close.

Okay. That's all of my comments.

CHAIRPERSON PRIOLA: Dr. Petteway.

DR. PETTEWAY: I'd just like to make a comment relative to the frequency approvals. You're assuming 60,000 donors per pool, and say for the source plasma industry pools are much smaller.

DR. ANDERSON: Right.

DR. PETTEWAY: So pools can be 1,000, four or 5,000. In that case, --

DR. ANDERSON: I mean, very much --

DR. PETTEWAY: -- get tenfold or more less.

DR. ANDERSON: You know, we can really set this up and structure the pool size, and we don't really have a handle on that for the entire U.S. So we made that 60,000 as the max, but certainly if that data comes to us, we can incorporate it into the model, and that would be very useful.

DR. PETTEWAY: Okay.

DR. ANDERSON: Any others?

CHAIRPERSON PRIOLA: Okay. If there are no more, thank you very much, Dr. Anderson.

DR. ANDERSON: Thanks.

CHAIRPERSON PRIOLA: If there are no other questions, let's go ahead and break for lunch and return back here at about 2:10.

Thank you.

(Whereupon, at 1:20 p.m., the meeting was recessed for lunch, to reconvene at 2:15 p.m., the same day.)

A-F-T-E-R-N-O-O-N S-E-S-S-I-O-N

(2:15 p.m.)

CHAIRPERSON PRIOLA: I think we'll get started. We're going to resume with the last talk of this morning session, and it will be Dr. Ruiz who will give us the European perspective on TSE agent clearance from plasma derivatives.

DR. RUIZ: Thank you.

First of all, I would like to thank you very much, the organizers of the meeting and especially Dr. Scott for inviting me here today. I mean, I do appreciate very much being here today and listening also to all of the discussions and having this opportunity to present the results of our latest discussions at the MEA regarding TSE and particularly variant CJD in plasma derived medicinal products.

So we have had already, I mean, since 1998 a number of workshops to discuss about human TSEs and plasma derived medicinal products, and the last one was held in June last year, and the result from those discussions and the recommendations and proposals is what I would like to share with you today.

But first of all, I mean, for those of you who perhaps are less familiar with the structure of the MEA, I would like to take a couple of minutes just to show you which committees are involved in this discussion so you can have a little bit better perspective of where the opinion comes from.

As you all know, the MEA is based in London and started working in 1995, and there are three scientific committees. One of them is the CPMP, Committee for Proprietary Medicinal Product and deals with human medicine. So all these three committees are involved in the decision of marketing authorizations of new product and new applications.

So the second of these committees is the CVMP dealing with veterinary medicines, and the third one is the COMP dealing with the Committee for Orphan Medicinal Drugs.

So these are the main three committee. They are composed by one or two representatives of each of the 15 member states, and they meet once every month at the MEA premises.

Now focusing on human medicines, you can imagine that just to work for the CPMP is normal. So there are a number of satellite groups that provide some advice or report to the CPMP on different areas, and some of those groups are dealing with biotechnology, the Biotechnology Working Party; the Blood Products Working Group dealing mainly with efficacy and safety of blood products; and then a number of other groups on different areas, the Efficacy Working Party, pharmacovigilance, safety, mutual recognition facilitation group and vaccine expert group, and there is one group, the quality working party, which is common to the CVMP and the CPMP.

The Biotechnology Working Party, I mean, it's a group that besides dealing with the quality aspects of recombinant products also is where discussions regarding the quality and viral safety aspects of blood products take place and also is the group where in this particular case all of the discussions for dealing with variant CJD risks and plasma derived medicinal products take place.

And usually the biotech working party, which is also composed from representatives and experts from the different member states, what it does is reports these conclusions and recommendations for the CPMP approval. So it works as a link, you know, between the CPMP, the COMP whenever they need advice in any of these areas, and also to the Blood Products Working Group mainly on aspects dealing with viral safety.

So with this background, what I would like to introduce now, as I said at the beginning, is the results of the discussions after EMEA expert workshop that was held last year. The aims of the workshop were to review the latest information on human TSE; also to put the precautionary measures in place, and if there was a need to introduce additional measures, and also this was done with the view of, well, to consider if it was possible to harmonize all of these precautionary measures between the member states.

The topics that were discussed included, well, epidemiological data and risk factors, new data on infectivity and screening tests, ongoing work on the standardization, leucodepletion, removal and partitioning during the manufacturing process, and as you can see, there was a wide participation in the meeting. So it was the TSE expert group, the members of the biotechnology working party of the Blood Products Working group, and CPMP, as well as experts in the field and representatives from the manufacturers, the European Commission, and patient associations.

Well, as you all know and as has been said this morning, from an epidemiological point of view so far until the beginning of February there have been reported 140 variant CJD cases, and as you well know, and it's been also said this morning, most of them are reported in the U.K. or linked to states in the U.K. Only the six cases in France and one in Italy are considered as endogenous cases. So that supports the risk. I mean -- sorry -- the hypothesis that there must be another risk besides the residence in the U.K linked to variant CJD.

A good review of all the epidemiological data was presented by Richard Knight, and according to what is known, I mean, it's still considered that the most possible hypothesis for the origin of variant CJD's beef products, especially mechanical or recover (phonetic) meat, and they have found, they have identified this cluster in the north of the U.K. compared to the south, but so far no dietary factors have been identified as significantly different in the north from the south.

But as well as you can see here, we have to keep our eyes open to any potential risk. I mean, I'm just putting in this because I was so surprised that these days you can find a product with such a name on the market. The photography, I mean, the picture was provided by one of my colleagues, Peter Castle.

So a number of measures had already been taken in the past regarding TSEs, and as a residence in the U.K. had been recognized as a risk factor for variant CJD. That was what led in the U.K. to the decision of no longer fractionate their own plasma in 1998.

And consequently, I mean, with this decision some member states in the European Union introduced some exclusion criteria for donors based on the time spent in the U.K.

So all of these data and others were updated and reviewed at this workshop that was held in the AMEA, and the conclusions in the form of a position statement is what I'm going to present to you in the following slides, and they will be published at the AMEA Web site.

So the recommendations and proposals focus on three main areas. One of them is record policy. The second is donor exclusion criteria, and the third is the manufacturing process.

So regarding the recall policy, well, it was not considered justified that there had to be a change from the previous position regarding sporadic familiar or iatrogenic CJD. That is that if a donor is identified as having contributed to a plasma pool and subsequently develops any of these forms of Creutzfeldt-Jakob disease, it's not considered necessary to recall those products from the market.

However, considering that variant CJDs had a marked disease and that higher infectivity and wider infectivity has been found in tissues as compared to sporadic CJD and also the fact that it is known if infectivity is present in human blood, that the precautionary approach as has been established before in 1998 was maintained, and that is that plasma derived medicinal products should be recalled where a donor to a plasma pool has subsequently developed variant CJD.

And this would include plasma derived medicinal products used as the active ingredient, I would say, or excipient, and what it must consider is that in the case of plasma derived medicinal product used in the manufacture of further medicinal products would have to be considered on a case-by-case basis, depending on the nature of the product, the amount used, at which step of the manufacturer it was used, the downstream processing. So there would be a number of considerations that would have to be taken into account.

Then regarding the donor exclusions, there have been a number of measures or approaches taken in different countries in Europe based mainly on the two risk factors identified, and those are the time spent in the U.K., on the one hand, and the other, the endogenous risk, that is, either from the BSE incidence in the particular country or due to the BSE import.

And of course, the benefit of excluding donors due to their residence in the U.K. will have to be or had to be balanced with a shortage of medicinal products, and also taking into account that the traveling pattern for different countries in Europe to the U.K. was different. That led to different policies in different EU member states.

So then we have countries no exclusion criteria at all, such as Denmark, the Netherlands, Sweden or Norway; some that have five year exclusion period, I mean, five year residence, such as Ireland; one year, such as France and Spain and Luxembourg I cannot be completely sure; and six months the rest of the countries.

So the recommendation in order to try to harmonize the position in all of the member states and also to avoid any potential problems with the movement of plasma derived medicinal products obtained in one country and going to another, the recommendation is going to be that the donors who have spent a cumulative period of one year or more in the U.K. between the risk period are excluded from donated blood or plasma for fractionation.

Well, this would be a recommendation, but then the national policies will have to be -- well, it will have to take time to address them, of course. Then the risk for transfusion and the policy for transfusion can be different as, well, the different risks have been analyzed.

And regarding the manufacturing process, a number of aspects were considered, and all of them have been discussed greatly this morning. So I will go just briefly through all of them.

We still don't know if infectivity is present in human blood in the clinical phase because it is assumed that someone in the clinical phase of the disease would not be qualified as a donor.

Then the experimental data from animal models suggest a very low level of infectivity, and we've heard from, you know, all of the models available, mice, hamster, non-human primate models that in all of the cases it seems that really the infectivity level is very low.

And then we also have data from investigational studies showing that there is removal of infectivity by steps commonly used in the manufacture of plasma derived medicinal products, including centrifugation, precipitation, chromatography and nanofiltration.

But also we've seen that the effectiveness is dependent on a number of variables. So it's very difficult to evaluate what's the contribution of the manufacturing process in a generic way, and these were just two of the most recent published works that have been also quoted this morning.

So considering that there are a number of studies already published that will seem to indicate that there is, indeed, a contribution of the manufacturing steps to remove or partition a potential infectivity, our initial recommendation would be that manufacturers use this general information to evaluate the potential key steps of their specific manufacturing processes to reduce infectivity.

As I said, this would be like a first risk analysis evaluation, and as has been said this morning, although this would be a very useful background information, each manufacturing process is different. We also recommend that manufacturers send them product specific evaluation on key steps for plasma derived products as a precautionary measure.

We know that this is a difficult task that, you know, you need really a lot of studies behind the facilities and personnel and time. So at least on our side CPMP's biotechnology working party with the involvement of external experts is thinking about developing a points to consider document to provide guidance on this respect, and this is certainly a priority on the agenda of the Biotechnology Working Party, and we will hopefully stop working soon and probably having a first draft some time this year.

And while it's been the case with previous workshops, the conclusions of this one will be published hopefully in March at the EMEA Web page, and this address, and if you want either an update or the text of the recommendation, you can find it later in this one.

And thank you very much for your attention.

CHAIRPERSON PRIOLA: Thank you, Dr. Ruiz.

Are there any questions for Dr. Ruiz from the committee?

I have one general one. Do the Europeans consider that the standard level? One of the issues that Dr. Epstein brought up was is the level of science there now to begin to really look at clearing TSEs from all of these blood products in a way that's reproducible and reliable and effective.

Do you consider that the standard of science is there now?

DR. RUIZ: Well, as Ben said this morning, I mean, there are a number of published studies that can be used as a base, and just at least to have an idea what contribution of the manufacturing process will be.

I mean, certainly -- and that's why, I mean, if you see the wording that we have used and will be used in the recommendation, it's really, you know, terms such as highly desirable and recommended because it's not going to be a mandatory requirement. I mean, we would like and we have heard -- I mean, we are talking about precautionary measures, and we would like to be prepared just in case, you know, if, well, a case is discovered tomorrow that has been transmitted through blood or blood derived products or if, you know, a heterozygous for the 129 colon (phonetic), I mean, is find. Additional scientific information comes that will make us think that there are additional risks. So in that sense we would like to be prepared, but also we think that even if the manufacturing processes are different, there are a number of studies that are, in a sense, optimistic or support the idea that if there was infectivity at least part of it would be removed.

So that's why, I mean, we will take the first approach of telling the manufacturers to assess according to this general information what the contribution might be for that process and then in a second step to ask for specific data.

But we know that that is difficult. So that's why we also decided to take the compromise and try to develop some guidance about, well, we've heard that there are differences, you know, in using different spiking materials and processes, and if bioassays or in vitro assays should be used.

So we will try to get something before asking for specific data.

CHAIRPERSON PRIOLA: Okay. Thank you, Dr. Ruiz.

DR. RUIZ: Thank you.

CHAIRPERSON PRIOLA: This topic is now open for general discussion by the committee on all of the talks from this morning. Any comments, questions? No discussion at all? Oh, yeah, Dr. Gaylor.

DR. GAYLOR: I just want to make it clear that I am wholly supportive of the risk analysis that was done and I think this is a very useful tool, and even if they're off a factor of 1,000, the exposure appears to be so low that there's not a high level of concern or hardly a low level of concern.

So my comments might have sounded a bit negative. I was having more problem with the limited time for presentation and understanding what some of the tables were trying to tell us rather than the risk assessment process itself.

I think they've done a great job and certainly moved that along from where we've been in past meetings.

CHAIRPERSON PRIOLA: Mr. Bias.

MR. BIAS: I guess my question is for Dr. Scott and the FDA. And I'm wondering was there some reason that the term "extremely remote" was used as opposed to just "remote"; why we needed the adjective.

DR. SCOTT: Well, I think that you're right. It really is a matter of perception, and to say "extremely remote" is we think it's extremely unlikely. I can't give you a perfect answer to your question.

MR. BIAS: That's okay. That's okay.

DR. SCOTT: Other than to say we're trying to articulate that we think it's most unlikely, and that is a strong way of saying it.

Yeah, and we have a similar statement in the albumen label, as Dr. Epstein has pointed out.

MR. BIAS: You have a stronger track record with albumen --

DR. SCOTT: Correct.

MR. BIAS: -- than you do with BSEs.

DR. SCOTT: That's right, and I think ‑-

MR. BIAS: And that's why that statement exists.

DR. SCOTT: -- that's the case, and if we have stronger supporting information, we might be able to make a similar statement for other products.

MR. BIAS: Yeah. From what I've seen in terms of the test, and I'm not a doctor, but from what I've read, we're not quite there. Almost every research paper that they have sent me to review said that more testing needed to be done, that we were testing different phases of the manufacturing process, but at the end of each one, they noted that they needed more science. They needed more time, and I guess that's my point.

If you put something -- and I know that the labels are written essentially for the physicians, but representing a patient community that considers themselves enlightened, if you put the word "extremely" in there, they're going to take it off their list. It's going to be more difficult for me as an advocate and educator of people with bleeding disorders to get them to take this issue seriously if they consider it off the list.

And I think this statement is worded quite strongly, and people who are reliant on these drugs on an every other day basis or weekly basis or however often they need to take, they're looking for relief. They live with these drugs. They're dependent on these drugs. In many cases, based on their insurance, there isn't a choice about which drug you get to take. You know, you can say, "I want the recombinant one," or I want the best plasma base, but in many cases those choices are growing narrow and people look for indications from the government and from their advocacy agencies for relief from the pressure of not only living with the disorder, but then having to learn about all of the things related to the disorder.

And I think that we're overstepping based on where the science is today to have something as strongly worded as this.

So that's just sort of my comment from a patient perspective.

DR. GAYLOR: I also wondered what those two words meant. If I were a physician and had to make a decision, I don't know what "extremely remote" means. I don't know whether that's one in a thousand, one in a million or one in a billion.

Being a quantitative person, I would like to see more of a quantitative statement there. I realize this committee wouldn't be able to come up with a number, and I couldn't come up with one myself, but when I know there have been millions of transfusions made in this country, in the U.K. and so on, without a known case, you know, "known" in quotes, case of vCJD, I would say very extremely remote.

(Laughter.)

DR. GAYLOR: Whatever that means, but you know, it would be nice if we could make a statement if we think -- or maybe make a statement like that. There have been whatever, so many millions of transfusions made, and we're not aware of a single case, period. I mean, if I had that information, I think I could make -- I know what I'd do if I needed blood. I wouldn't worry about it if it was a high risk, one in 100, but maybe we can make a statement like rather than just saying "extremely remote."

CHAIRPERSON PRIOLA: Dr. Johnson.

DR. JOHNSON: I'd just like to go back to Mr. Bias' other concern, and that is that would there be two medications, which one would be labeled extremely unlikely, and one would not be labeled that way, and then the patient would not have a choice of which one to take because of insurance.

I hadn't thought of that, but I think that's a real concern. That would be an anxiety I would not think we ought to put on the patients.

CHAIRPERSON PRIOLA: Dr. DeArmond.

DR. DeARMOND: I would agree with the statement I just made that maybe the agent should be specified a little better. I was thinking a similar thing. In fact, in the statement a virus and CJD agent are really thrown into the same sentence almost as if they're the same sort of bird, and that any processing of them is going to be essentially the same, and that's completely wrong.

Perhaps the viruses for which a real risk is known because they actually have transmitted disease and the TSE agents for which no transmission has been identified should be stated, and that's not lying, and it's not saying that there wouldn't be any transmission in the future, but up to this point there isn't any evidence.

So that might be a modification of this statement. That may be strong enough for the whole process.

CHAIRPERSON PRIOLA: Dr. Bracey.

DR. BRACEY: Just a couple of comments. One is that in terms of having multiple agents with different degrees of infectivity, there is a precedent, and that precedent would be with the various clotting factor concentrates. The difference is that the users of those products tend to be quite familiar with the degree of risk and not totally subject to the -- I hate to use this term -- but the whim or the purchasing issues in the pharmacy.

And I would think that probably if there were two agents that what would usually happen in most hospitals is that the cheapest agent would be the agent that would be acquired, and really in terms of the end user knowing the details, I don't think those details would come out.

The second issue that I have is I guess not having been in the committee before I had always thought that the data on CJD, not vCJD, was good enough that we really weren't that concerned about CJD. In fact, I thought I sensed sort of double speak because where early on it talks about the theoretical risk of vCJD, but then -- and the continuance. CJD gets lumped in with vCJD. So I just wanted to make sure that we wanted specifically to lump the two together.

It seems that, you know, the early part splits out the vCJD as a separate category, but then it gets lumped back in with all of the TSEs, and I'm not sure if that's what we really intended to do.

CHAIRPERSON PRIOLA: David.

DR. BOLTON: Not to focus on he wording in the second part of the question, but to go back to the first part of this question, I'd to, I guess, ask either Dr. Epstein or Dr. Scott if my understanding of this is correct, that if we were to recommend this, that the FDA begin considering labeling claims, that the process would be that then the manufacturers would be invited to submit dossiers that would then very clearly specify all of the things that Dr. Scott described about their validation studies for existing clearance or existing fractionation procedures, and after review of those data and finding them acceptable, that then the question of changing the labeling would come about. Is that correct?

DR. SCOTT: That's correct. The labeling change would depend upon the submission and evaluation of data.

DR. BOLTON: Right.

DR. SCOTT: Yeah.

DR. BOLTON: And are any manufacturers currently prepared to submit such an extensive dossier? Do you know?

DR. SCOTT: We think it's possible.

DR. BOLTON: Okay. So then maybe consideration of the wording is somewhat important.

I mean, I don't want to get too -- I wouldn't want to spend a lot of time fiddling with the wording, but having sat for many years on human subject review panels and knowing that our understanding of language is not the same as the understanding of the general public, I might suggest that the term CJD is going to be understood by probably everybody in this room, but I'm not sure about the general public, and that the wording "provides substantial clearance" could be changed to "substantially removes," and it might make it a little bit more understandable.

DR. SCOTT: Well, I think some of the information or the discussion we need from the committee is exactly what you're discussing, and that is the nature of such a labeling claim. So, for example, we've already heard that some members might find something that sounds more quantitative, more helpful, and this other discussion on how to craft such wording is of great interest, and so I would like to hear more.

CHAIRPERSON PRIOLA: Dr. Gambetti.

DR. GAMBETTI: I introduce another aspect. If I understand correctly, the label, the fact that the existence of this label or actually two labels will introduce two categories of products, the one that are extremely -- in which CJD or vCJD is extremely remote and the one that instead is remote, and --

(Laughter.)

DR. GAMBETTI: -- and so we will have two -- if I understand correctly, we have two categories of plasma, product of plasma, and then they are -- do you expect, do we expect that it will cost -- the cost will be different? In other words, it will be a product that is cheaper than others, and therefore, you will create two kinds of plasma available to patients and, therefore, who can afford it will use extremely remote and cannot -- will buy the remote.

Is that a possibility?

DR. BOLTON: My understanding is that these processes are existing already and that the processes are not going to change, but what I suppose you could have is you could have the perceived higher quality product and the lower quality product, and whether that would result in a price increase or decrease depending on which side of that fence you fell on, I don't know.

Would the industry reps. like to?

DR. PETTEWAY: Yeah, yeah, I think you're right, David. I think, in the first place, all of the analysis that you're going to see and all of the analyses that have been done are on existing processes. They already exist. All that's been done is defining the potential of any of those processes to remove TSEs or not.

Okay? So there's not been one process where someone has innovated capability to make their process safer than another based on removal of TSE. This is what exists now.

That's one issue. The other is the question that you asked, David, which is are companies prepared to provide rigorous data supporting this, and I would say that most of the studies that you will see that Hank Barium (phonetic) will present and the studies that are published are kind of the tip of the iceberg of the rigor of the studies. I think Dr. Scott made that point.

So I think most of us that do these studies follow all of the viral validation principles essentially when we do the study. So I think that the studies do have significant rigor that could be addressed.

DR. GAMBETTI: In conclusion, there is one problem, the sense of it, that you think that all of the manufacturers follow a procedure that is rigorous enough that allows us to define the product extremely safe. In other words, all of the manufacturers, all of the plasma available now is being subjected to this kind of rigorous purification. We don't have two different problems. We have only -- or more than two.

DR. PETTEWAY: No, no. You can't say that. Remember that, again, as Dr. Scott, I think, made very clear, different manufacturers fractionate plasma in a similar way, but even sometimes differently, and then the purification processes that lead to the product are independent between manufacturers. So not all processes are the same from manufacturer to manufacturer. Each one has to be evaluated independently before you can make any assessment of any time.

DR. GAMBETTI: So we will end up with a series of, well, at least two different products.

DR. PETTEWAY: Well, you'll end up with sort of what you have now with viral validation because that's how viral safety is determined. It's process to process because all processes don't have the same level of viral validation, for instance, based on analysis for any given virus.

DR. BOLTON: Right, and for some processes for which the dossier has been submitted to the FDA, you will know to the extent that you can know anything in this field what that process does to remove prions and for the other ones for which they have not been submitted, you won't know.

It could be equivalent or it could be better or it could be worse, but you won't know.

DR. PETTEWAY: Exactly.

DR. BOLTON: And what I wanted to again focus on is that to me there is very little benefit in the label change in terms of public health. There is maybe reassurance to the public, but the benefit comes in the understanding of these processes and how they do or do not remove prions.

And secondarily from that is the question of some or maybe many of these processes may be proprietary, and so I guess another question is the dossier that is submitted to the FDA will be known; the process and the effectiveness will be known to the FDA, but it will not necessarily be released to the public and to the other manufacturers, and where there may be a benefit to having a particular process that's very effective, that may, in fact, be limited in its effectiveness because it won't be communicated to other manufacturers.

Steve, maybe you, since you're the industry rep., but what do you think about that?

I seems to me that there would be benefit not only of understanding the effectiveness in terms of removal, but also communicating that more generally.

I know then you lose some advantage as a manufacturer, but the public health benefits by having the better process understood of such.

DR. PETTEWAY: Well, that's unfortunately or fortunately a fact. There are proprietary methods for achieving the end products that each company has. Some are fairly standard and communicated. Others are not standard and communicated.

If we follow the paradigm for viral validation or viral safety, most companies at least in general publish or make available the viral clearance parameters for their processes and what those results are. But as far as having detailed information about viral removal from one company that benefits another company, that as yet is unclear as to whether that can be done or not, and it isn't just safety or what we want to do. It's how the law allows companies to share information and interact.

CHAIRPERSON PRIOLA: Okay. If there are no other comments, I think maybe we could go on to Dr. Hank Baron's presentation on what manufacturers have done in terms of studying TSE clearance from blood products.

DR. BARON: I would like to thank the FDA for providing the Plasma Protein Therapeutics Association, the PPTA, with the opportunity to present the data that has been generated on prion removal by the different member companies that have generated this data.

I would also like to make it clear that I am speaking here today as a spokesperson for the PPTA and not particularly as a representative for my employer, which is Aventis Behring, because this is a presentation in which I am basically going to provide you with a consolidated view of the entire industry's approach to trying to resolve this very, very difficult issue.

I also, by the way -- my heart goes out to David Asher, and I hope that I don't succumb to the same fate because I am also suffering from an airplane cabin sinusitis and dry cough. So at some point I might look as pathetic as he looked at the end of his presentation. I hope not.

(Laughter.)

DR. BARON: I think that it's always worthwhile to begin with a statement that puts the discussion into the perspective of the current state of reality, and that is that there is currently no scientific evidence to substantiate that persons with preclinical or clinical CJD, and that includes variant CJD, carry infectious prions in their blood or have transmitted them through blood or plasma products.

So we've heard the word expressed many times today. This risk is considered theoretical, and this statement is based on a wealth of information for what is sometimes called the classical forms of CJD, which we've known for several decades, and of course, we have a great deal less experience with variant CJD. We do not benefit from the vast epidemiological perspective for variant CJD that we have for classical CJD, and because of this the time window is just not quite long enough to rule out the theoretical risk.

But we do know a few things about variant CJD, and some of these things have already been mentioned, even in the last discussion of the group here, and that is that whole blood, red blood cells and platelets from U.K. donors have been and continue to be administered to U.K. recipients.

There have been an estimated 30 to 40 million transfusions in the U.K. over the past ten years. Yet there has been no evidence of any link of transmission of variant CJD or classical CJD due to transmission, and these are products that do not go through the fractionation and purification processing that plasma derivatives go through.

David Asher mentioned that several patients with variant CJD were recipients of transfusions. None of these could be linked to a donor who had CJD or variant CJD, and some 33 cases of variant CJD are known to be blood donors today, and there have been no cases of CJD or variant CJD noted among named recipients.

Now, again, David very clearly pointed out that many of these recipients died before their theoretical incubation time could be fulfilled, and others have been -- the time period since their reception of the product is ten years or 15 years or less. So we still need more time to be able to rule this out.

However, it is clear that in the U.K., a country which reports 98 percent of all BSE and 95 percent of all variant CJD in the world, there's no evidence that variant CJD has been transmitted through blood or plasma products. Again, it is still too early to close the book.

The other information that we have about the potential infectivity of bloods from variant CJD come from two Lancet publications, and these are the only published data researching prions in the blood of variant CJD patients, but two Lancet publications have reported that prions were undetectable in blood, plasma and buffy coat of four patients with variant CJD despite detection of prions in their lymphoid tissues.

One of these papers was published by Moira Bruce and her collaborators using a wild type mouse bioassay to detect the infectivity. I think it was Dr. Scott pointed out the limit of sensitivity of this method is quit limited.

Jonathan Wadsworth studied variant CJD blood samples using a highly sensitive immunoblotting assay, and he also was not able to detect pathogenic prion protein, PrP scrapie in any of these samples.

Now, again, these results are limited by the sensitivity of the methods that were used for detection. I personally believe that Wadsworth's Western Blot, which uses phosphotungstic acid precipitation is probably more sensitive than this wild type mouse bioassay.

Nonetheless, I think it's good news that these methods did not detect prions in the blood of these variant CJD cases, but I think there's also a message here that we really need to continue to work to develop more highly sensitive, rapid methods of detection of prions than those that are currently available.

So a responsible approach then to the manufacturer of these plasma protein therapeutics is to treat this theoretical risk as though it were real, and this is done in a number of ways.

One, through rational, science based, precautionary policies to minimize the theoretical risk. We've heard both the European perspective and the American perspective with regard to individual donor deferral criteria, geographic plasma rejection criteria and withdrawal and notification policies in the event that a donor afflicted with variant CJD is subsequently -- well, a donor of plasma is subsequently diagnosed with variant CJD. So I won't go into any detail on this.

The other way that we have been attempting to address this theoretical risk is through the development of extremely rapid and highly sensitive methods of prion detection, which are basically applied to two major goals.

One is research. To continue the research of prion infectivity in the bloods of Creutzfeldt-Jakob disease, including variant CJD cases.

I know that in addition to developing these highly sensitive methods of prion detection, there is also a great effort within the industry to use currently available methods, such as the study that's being coordinated by Baxter, to look for infectivity in the bloods of variant CJD patients using monkeys.

We at Aventis Behring have developed a sandwich version, a two site immunoassay called the confirmational dependent immunoassay, which also has a high sensitivity for detection of human prions.

And of course, the people at Bayer have developed a high sensitivity Western Blot which they use in their partitioning studies.

And this brings me to the second approach to attempting to minimize the theoretical risk further, the use of these detection methods in the assessment of prion partitioning in manufacturing processes.

Now, it has been abundantly pointed out already that the assessment of prion partitioning in manufacturing processes for plasma proteins represents a particular challenge, and this is because -- we come back to the statement -- prions have never been detected nor transmitted through human blood, plasma or plasma derivatives. Therefore, we have no clue as to the biophysical-chemical nature of the theoretical prion contaminant in plasma. We don't know what it is, and we don't know how much of it is there, if it's there at all.

So there's uncertainty then when talking about designing partitioning studies as to appropriate relevant prior spiking gauges for the study of the prion partitioning and manufacturing processes.

So this slide summarizes in one slide most of the challenges that Dr. Scott indicated were inherent in the designing of partitioning studies, removal studies for prions in manufacturing processes.

One, it has been stated a number of times you have to demonstrate the validity of the scaled down model for your manufacturing process.

The nature of the spike. Again, sine we don't know what the physical-chemical nature of prion infectivity in plasma is, there are questions as to what is the appropriate spike.

Detection methodologies. There have been a wide range of detection methodologies used across industry to measure the levels of prions in the input and output samples in these studies. Some of us are using immunoassays. Some of us are using infectivity bioassay. Some of us are using one and the other, one and then the other to correlate and to demonstrate the relevance of using the immunoassay.

And there's the question of using model mouse or hamster prions as opposed to human prions in spiking studies.

Finally, there's the question of the validity of evaluating independent steps and adding up removal factors versus looking at couple process step removal, and I'll try to address each of these issues before I actually go into the data.

This is a slide which was contributed by Steve Petteway of Bayer, and it essentially shows the design of one of these removal studies in which you take the manufacturing process; you scale it down; you miniaturize it to laboratory scale, and then you take your spike of a known level of prions. You put it into your input solution. You prove; in essence, you assay. You do dilution series to determine what the level of prions is in your spike.

Then you go through your separation procedure, and then you analyze both the supernatant sample and the precipitous sample to see what the levels of prions. The difference between this and this will give you the removal factor, and this will confirm the mass balance and tell you that everything is where it should be and that the results are logical and reproducible.

And of course, as has been mentioned, these scaled down processes have to be shown to be equivalent to manufacturing processes as they occur in the production facility, and this data shows you -- it's probably too small and difficult for you to see, but the three bars show you the protein concentrations in an output sample from a scaled down step as shown in the manufacturing facility, as shown -- that's the black bar.

The red bar is historical data from manufacturing in the buyer facility, and the white bar is the data that shows the scaled down process for the removal studies, and as you can see for this particular study, there's very good correlation between the manufacturing process and the scaled down process.

So then the first step is to perform the scale down correctly and to, as has been discussed abundantly, demonstrate equivalency for all of the key parameters of that scaled down process.

The second issue then is, okay, we've got the procedure scaled down and it has been validated to equivalency. Now, what do we use as a spiking agent?

Here you see a range of all the spiking agents that have been used in industry by different member companies of PPTA to evaluate this. You can see you have a tremendous range of physical-chemical characteristics here. You have brain homogenates, and even with brain homogenates there is a range in that you have some people using crude brain homogenates, some people using cleared brain homogenates, some people using solubilized brain homogenates with detergents.

So even within the brain homogenate itself you have a range of options. Then you have progressively more purified forms of infectivity, microsomal membranes, caveolae-like domains which are a specialized membrane compartment which is rich in PrP scrapie.

Then you have what has been referred to earlier as the Bolton preparation, purified pathogenic prion protein that maintains its infectivity.

And finally, some people have been using also prion fibrils or prion rods or scrapie associated fibrils. They've been given various names, which essentially are extracted from brain homogenate with detergents and through ultra centrifugation, but this differs from this in that this employs protease K so that these form fibrils composed of the protease resistent core of the pathogenic prion protein. So this is composed of PrP 2730.

So you can see a tremendous range of physical-chemical forms of prion infectivity have been used in these different studies, and there is also a range in terms of the removal behavior of these different types of spikes.

The unbound, purified PrP scrapie, for instance, doesn't always partition in the same way as the three membrane bound spikes. This is just to give you an overview of the wide range of forms of infectivity that can be used and have been used in these different studies.

Next, prion detection methods. So you've selected your spike -- here I go -- and you then must measure the level of prions in your input and output samples.

Now, what has been and what is still the told standard for measuring prion infectivity today is the infectivity bioassay, essentially using rodents. Some are using mice. Some of us are using hamsters. Some of us are using a human prion spike and using transgenic spike as a mouse bioassay.

However, the problem here is also that if we use these methods to analyze all of the steps involved in the purification and the fractionation of plasma protein derivatives, we can wait many, many years before we get any kind of removal results.

But since we know that PrP scrapie, which is the pathogenic form of the prion protein, is highly correlated with infectivity, there's been now some immunoassays for PrP scrapie. Again, some of using Western Blot assays. We at Aventis Behring are using the confirmation dependent immunoassay developed by Uri Saffar in a co-development with Aventis Behring.

And some of us, as has been mentioned earlier, are using, again, a combination of both, doing studies with the Western Blot or the confirmation dependent immunoassay, and then correlating the results obtained with one or the other with infectivity bioassay in transgenic mice to demonstrate that there is good correlation between the infectivity and the PrP scrapie immunoassay.

And then removal is determined -- well, this is what I just said.

And this is, again, data which was contributed by Steve Petteway from Bayer, and it shows the correlation that I was referring to between results obtained from the Western Blot immunoassay and the bioassay in hamsters, and over a series of steps, one, two, three, four, five, six, seven, eight, nine steps, they've been able to show essentially that the Western Blot removal factors correlate quite well with the removal factors that were obtained with the hamster animal bioassay.

So this is a very nice determination from the Bayer group. We at Aventis Behring have an ongoing study. I'll mention that in association with a slide coming a little later on.

The next issue asks the question is it appropriate to use model prions, hamster or mouse prions, as opposed to using the relevant, the real human prion, Creutzfeldt-Jakob disease or variant Creutzfeldt-Jakob disease prions, and again, this is data from Bayer showing for three different fractionation steps the removal factors obtained over a range of strains of prion. This is variant CJD. This is sporadic CJD. This is a genetic form of prion disease in humans called the Gerstmann Straussler Scheinker Syndrome. This is sheep scrapie here, prion protein, and this is prion protein from hamster experimentally infected with hamster prions.

And basically what the data here shows is that the removal numbers for these different steps are comparable whether you use the hamster prion or these various human or sheep prions. So the hamster prion seems to be predictive of the partitioning of the human prions in this study.

And this is further data that was developed by Aventis Behring. By the way, the previous slide showed data obtained with one spike form with brain homogenate. This is further data developed by Aventis Behring in which, as I mentioned, since we demonstrated a difference in the partitioning behavior of membrane bound prion preparations as opposed to the purified PrP scrapie, we now do all of our partitioning studies with two forms of infectivity. We use both microsomes and a purified PrP scrapie, and we express our removal factors as a range, as a worst case and a best case, but essentially the data here are for variant CJD and sporadic CJD and the hamster prion. This is a hamster strain, SC237, basically comparable for this particular step, which is 25 percent ethanol precipitation.

And they're comparable whether you're using the microsome spike or the purified pathogenic prion protein spike.

This is another step in which we analyzed this question in exactly the same way, microsomes and purified PrP scrapie, and again, you can see for variant CJV, sporadic CJV and hamster prions comparability of the removal behavior for these three forms of prions.

And in context with this study in which we used variant CJD prions to spike these two processes, we have currently ongoing a correlation study with animal bioassay in which we have used as a model animal, model for variant CJD, transgenic mice carrying the bovine PrPG, which are highly susceptible to variant CJD prion. So this titration study is ongoing.

So the conclusion that we can draw at least at this time from studies done independently by two laboratories and using three different spike forms either that shows removal of rodent prions seems to be predictive of the removal of human CJD and variant CJD prions.

This is another slide contributed by Steve Petteway from Bayer, and this slide shows the relevance of determining removal factors for independent steps and the relevance of using additive removal factors from independent steps as compared to doing a series of a couple steps.

And what you have here is a sequential series of steps, the initial spike one in here, 5.2 logs. You go through a number of steps until you get down to this step here, and so what you can see is you initially spike with 5.2 logs. Now, by the time you get to this step the pace of the precipitate contains four logs, and so what you have then is essentially a one log removal only through these steps.

But then when you go from this step to this final step separating into precipitate and supernatant, what you're getting is a 4.2 log removal or essentially the totality of what you had and the input at this stage, and you're getting no detectable prions in the effluent fraction.

And when this step was analyzed separately, and I don't have the data here to show you, but Steve Petteway can comment on this later if he wants to, when this step was analyzed separately, it had the same removal factor, essentially four logs, as it had in the combined series of steps.

So the bottom line then of this demonstration is that partitioning determined for independent steps is consistent with partitioning determined for a couple of processes.

We at Aventis Behring have also done similar work not as extensively and more limited in which we've taken two steps at a time and looked at each step individually, and we have found comparable removal factors, either additive or combined.

So now I am going to go through the range of data that has been generated throughout the member companies of the PPTA in assessing the removal of prions for manufacturing processes, and the major product categories that I'll be showing you data on are these four: Factor 8, immunoglobulins, albumen, and protease inhibitors.

So, again, remember that I remind you all what you already know, but I will remind you anyway, and that is, again, these steps are not the same across different companies. The steps can differ in terms of a number of parameters. The studies are being done with different spiking agents, and the results are being analyzed with different detection systems.

So there is absolutely no standardization in the body of data that I'm going to show you next.

Factor 8 products. So here are a number of steps involved in the production of Factor 8: cryoprecipitation, aluminum hydroxide absorption, polyethylene glycol or cylcene precipitation, ion exchange, chromatography, monoclonal antibody purification, and filtration, and these are the removal factors that have been generated in different studies from different companies.

You have the references here for each bit of data that's shown in the table. I think what you're seeing essentially is that you have fairly consistent values being demonstrated for these different steps in spite of the enormous variability in which these studies are being done.

And this table here or this line shows you the cumulative removal factors reported for the different Factor 8 products from the different companies analyzing these different steps. I'd like to point out that this 8.0, there are differences actually in the purification of these different Factor 8 products. This, for instance is a -- excuse me. No, this one. Where is the monoclonal purified? I'm not seeing -- here it is, four. Okay. You've got it. So I was right the first time.

This 8.0 removal factor was obtained with a monoclonal affinity purified Factor 8 product as opposed to this 4.8 to 5.5 was done with a Von Willebrandt's factor, Factor 8 product, which of course is not as highly purified as the monoclonal purified product.

So the range of purification of these different Factor 8s is different, and some of this may be reflective also in the numbers for removal factors.

Here are the numbers now for immunoglobulins. Again, the laboratories that generated this data, you see some of the references. Lee, et al., this is, of course, the Bayer group. Vey, et al., this is the Aventis Behring group. We have ZLB. We have Baxter. We have Biotest reporting data here, and these are all of the steps involved in the production of immunoglobulins.

Again, you see a nice consistency in the data generated by these different steps in spite of the fact that there are huge differences in the way these studies are being performed.

And here are the combined or the cumulative removal factors for these different immunoglobulins.

For albumen, which really comes down at the very end of the successive chain of precipitation with increasing levels of alcohol, again, you have the different steps here and, again, you see the consistency of the data. Obviously these steps, again, these fractionations are not done in the same way at all times.

The conditions vary, and here you do have some variability in the removal factors for this step, and albumen seems to have the greatest removal factors. You have a removal factor of 16 logs. If you add up all of the steps studied from the Bayer group, this is the data from the Aventis Behring group. Seven plus logs, the 14 plus logs depending, again, on whether you're using the microsomes or the purified PrP scrapie in our processes.

Eleven, point, five is coming from work done by Foster at the Scottish National Blood Transfusion Service, and this 5.8 log removal factor is coming from data provided by Baxter.

And finally, data on proteinase inhibitors. Again, here are the steps that have been analyzed among the different member companies of PTPA, and again, you see fairly good consistency in the data, and here are the numbers for overall cumulative removal of prions for this product category.

So I'd like to conclude by saying that the removal of prions by plasma manufacturing process is very significant, and it further minimizes the theoretical risk which is already addressed by donor deferral.

If you like, the FDA has shown us information today indicating that the reduction of the exposure risk due to the donor deferral is roughly 90 percent. So that is one log.

So we're talking about adding logarithms to further minimize the theoretical risk addressed by donor deferral. This removal of prions by a manufacturing process is process specific. It can't be stated enough. The process is different from company to company and must be analyzed separately, and we do see differences for similar processes in different laboratories, but it has been demonstrated across all prion spiked materials, different prion assay systems, and manufacturing step specifics, and I believe that and PPTA believes that this removal is very substantial as compared to the still theoretical level of risk.

Now, I'd like to conclude by saying that I do not pretend to be providing you here with definitive answers to any of the multiple questions that bedevil this whole issue of prions in blood or plasma and the safety of plasma derivatives.

Now, does the removal of prions by manufacturing processes, does it enhance the safety profile of these products? On the one hand, you could say not really because the safety of plasma derivatives with respect to prion infectivity is established and has been established over time, and there has not been a demonstration of the transmission of a prion disease through a plasma derivative.

So adding removal to zero, if indeed there is zero, does not necessarily enhance safety. Nonetheless, what these studies do do is they add a measure of reassurance in that they do add extra reduction of prions which are spiked into plasma and into plasma intermediate samples. They do add the assurance of knowing that fortuitously these plasma fractionation and purification processes can remove extra levels of prions.

And I would just like to leave you with the hope that this committee, the FDA, and the patient community recognizes and appreciates the enormous effort, really that has been put into trying to sort these problems out by the plasma proteins therapeutic industry.

And I will close there. Thank you.

CHAIRPERSON PRIOLA: Thank you, Dr. Baron.

Questions from the committee? Dr. DeArmond.

DR. DeARMOND: So that's very impressive. Starting with your ending comment that has all of this process helped, and the answer was, well, maybe not because you're already starting with a product that is pretty safe, to begin with. At least there's no evidence that it's not safe, but then you say also, yes, because it adds the reassurance. You and other companies, your companies have really put a tremendous effort into doing this, and I'm sure it costs quite a bit. It costs us a lot just to do our measly little experiments.

So one of the questions I have is with regard to the statements that we're supposed to discuss has to do with this process. Several companies have put out the effort and have shown that if there even is a risk, it's reduced tremendously by your process.

Do you think -- and I don't know how many other companies are out there making the same products. You've mentioned three companies, Bayer, Baxter, Aventis Behring, and how many others are involved?

DR. BARON: Steve, how many do we have total in the PPTA?

DR. PETTEWAY: Well, there are 11 companies. I think there are 11 that participate with us. I think all of the companies are doing these types of studies, and they're at one stage or another.

DR. BARON: That's right. I think some have begun later than others.

DR. PETTEWAY: Right.

DR. BARON: And so the data that I presented today was the data that's currently available. We scrambled very hard within the last three weeks, four weeks to get all of this data together and to try to put these composite slides together, but what you saw is data from about six laboratories.

There are others who either are producing data now or for whatever reason felt that it wasn't time for them to share their data, but I'm sure there's more data out there.

CHAIRPERSON PRIOLA: Dr. Johnson.

DR. JOHNSON: Just a clarification. I don't understand on the tables how you get reductions of 11 and 16 logs.

DR. BARON: I'm sorry. That's cumulative. That's by adding up steps. So if you have four steps with four logs, that's 16 logs. Those are additive numbers.

DR. JOHNSON: Oh, starting from the back again?

DR. BARON: You do a number of steps. You determine a reduction factor, and then you add them up to determine what the total additive reduction is, which of course is not going to have really any real relevance to what will come through the process in one shot, but it's extremely difficult to do these down scale and analyze for a product the whole throughput in one experiment.

CHAIRPERSON PRIOLA: Dr. Hogan.

DR. HOGAN: On one of your slides for albumen you had a difference of seven logs in terms of removal, 7.7 to 14 whether it was microsomal or purified. Is the lower number the microsomal or the purified?

DR. BARON: The lower number is the microsomal membrane fraction. The higher number is the purified PrP scrapie, which in precipitation steps we've seen consistently the purified PrP scrapie precipitate more readily than do the microsomes.

However, we've seen an opposite effect in absorption steps in which it seems that the microsomes are -- that the PrP scrapie is more difficult to remove than the microsomes.

CHAIRPERSON PRIOLA: Dr. Bolton, did you have one?

DR. BOLTON: I probably have many, but I'm not sure I should launch into them all.

Hank, I thought that was also an excellent presentation, and as we've discussed between ourselves, this is a difficult area, and I guess I've discussed my views in this committee meetings before.

Obviously I have some interest in the use of purified PrP for many things, but I'm less enthusiastic about it as a pike. As you know, except under certain circumstances in here, in these kinds of studies because it's so far removed from the probable natural physical state of prions that would be in blood or plasma.

I'm wondering in how many cases are the cumulative reductions actually measured in cumulative steps. I know in studies that we did when we were looking at the processing of lung lipid surfactant, we found that in some cases steps were additive with one spike added at a particular point, but in other cases with a spike added at a different point there was a surviving fraction that was then no longer removed by the subsequent step.

So I think it's very important to really -- especially when the additivity is significant that that actually would be documented in whatever way it can be in the actual physical process.

Could you comment on that?

DR. BARON: You know, that's fair enough, and I don't think any of us have done a full cumulative analysis of a process from plasma to product.

Steve, do you want to make a further comment there?

DR. PETTEWAY: Well, yeah. The data that Hank showed you for the cumulative removal was for a solvent detergent immunoglobulin product that we have. We also have a chromatography product, and we actually did that. We spiked the plasma, and we carried the entire process through and followed it for the reason you said, which is we were concerned that there would be some part of the infectivity or prion that would be refractile to any removal, and we never found that.

At least for those two -- and remember all processes are not the same -- but for these two, the additive or adding all of the steps together was born out by spiking at the beginning and seeing it all removed at the end.

DR. BOLTON: Well, I guess my comment is as much for Jay and Dot, that in the FDA's consideration of all dossiers, that especially in assumed additivity of significant steps that they demand at least an examination that that is, in fact, born out in the physical process.

DR. BARON: I think that was clear in Dr. Scott's presentation. I think she made a point of raising that as a key issue, and we all agree on that.

DR. BARON: I guess I --

CHAIRPERSON PRIOLA: Hold on just a second. Just one moment. Dr. Epstein just wanted to comment.

DR. EPSTEIN: Well, I just wanted to comment that we have, indeed, made that point to various manufacturers, and some experiments showing additivity of clearance have been performance, but the number of such experiments becomes very large if you look at all of the processes you might wish to combine, given the limitation that any one steps can only be studied at best to a four log clearance, and most don't get near that. They're more like, you know, two and a half to three maximum just because of the limitations to the input in the assay.

So you have, you know, a daunting problem because you're not going to be able to study all combinations of processes. So we've tried to focus on looking at the processes that have the best clearance and showing that they really are additive.

CHAIRPERSON PRIOLA: Mr. Bias.

MR. BIAS: I guess first I'd like to say on behalf of the patient community thank you for all of the work, and I hope you'll share those results with our medical and scientific advisory committee at the National Hemophilia Foundation.

But I also didn't want to get too grateful because the reason you're doing all of that work is because over the years plasma products have infected my community with a large number of diseases and things that we live with in addition to the bleeding disorder.

So you have to do the work at this point, and we are grateful that you have, but I just wanted to put that out there as a reality check, so to speak.

But I did hear a "but." So does that mean that because you measured it at different steps that this cumulative log thing that's going on here isn't quite reality or respective or it's just still speculation on some level?

DR. BARON: I don't think it's speculation. I think that it needs to be confirmed by analyzing successive steps and comparing the data. We've done it on a limited scale. I think the Bayer people have done it on a fairly wider scale, but I think that it needs to be shown for a number of steps that you can actually in reality add these numbers up and come up with the same removal factor as if you had run the steps successively.

So this is something that needs to be analyzed, not for all steps, as Dr. Epstein has indicated, but at least to show on a certain number of steps that this is what's happening and that you can then draw the conclusion that it is valid to add up removal factors.

MR. BIAS: So it's a great start, but not quite there yet.

DR. BARON: I think that there is always a great deal of work to do, and we still have a number of products that we haven't begun to evaluate. So I think, you know, this is an enormous amount of work. I think one of the things that we also had to do is get some baseline data as quickly as you can because we'd like to cover the whole gamut of products, and then based on that baseline data we go back and we try to refine here and there and try to confirm certain things.

So the first step is to generate the whole list of steps that are employed in the different companies and come up with the baseline data.

MR. BIAS: Once again, thank you for your effort.

CHAIRPERSON PRIOLA: Dr. Belay.

DR. BELAY: This is a totally different question. I was curious about the industry perspective on the proposed labeling. Do you see this proposed labeling an improvement over the current labeling?

DR. BARON: First of all, the current labeling I think is not entirely appropriate in that I don't believe it's correct to lump prions and viruses in the same sentence. As has been pointed out by a number of people here today, one is a true risk. HIV, HCV, HBV have been transmitted through food plasma derivatives, and to then include the theoretical risk of prion infectivity within that statement that implies that there's a risk I think almost puts them on an equal level and might create some confusion.

So I would, first of all, prefer that the occurrence of labeling actually separated these out into two separate paragraphs and clearly outline that the risk for prions is theoretical, but nonetheless that there is this theoretical risk.

As far as the labeling claim is concerned, I think I'd like to pass on that question. I think that to answer you now would be answering in a vacuum because I don't know how the FDA intends to establish the criteria for this, and I'd feel a lot better about it when I was more aware of how this process is going to develop and how the criteria for being allowed to have this labeling claim are going to be set down.

So I think that I'll pass right now. I'd like to see how the process is going to develop.

CHAIRPERSON PRIOLA: Go ahead, Dave.

DR. BOLTON: Let me rephrase that slightly differently. Assuming that Aventis Behring submitted a dossier that was acceptable to FDA, you would want to have some modified -- would you consider having a modified label appropriate or desirable at that point, assuming you could agree on the wording in some way?

DR. BARON: If this labeling claim became a reality, and having good data, which I know is rigorous and reproducible data, I would certainly provide it to FDA to be able to have this labeling claim.

CHAIRPERSON PRIOLA: Okay. Dr. Petteway.

DR. PETTEWAY: I don't want to prolong it, but I --

CHAIRPERSON PRIOLA: No, no. Go ahead.

DR. PETTEWAY: -- just think it's -- I think we're talking about balance, and I think if you're going to have a warning statement that says these products are at risk for CJD without providing information based on what evidence exists to mitigate that risk, that may not be wholly appropriate either.

So I think regardless of what the wording is, if there's data that's defensible and that has some effect on how you perceive the risk, then maybe it's not unreasonable to find a way to include that.

DR. BARON: But your condition, again, was conditional. My current position would be I would really have to see how this is going to be developed.

CHAIRPERSON PRIOLA: Doctor --

DR. BARON: That's why I don't have an opinion one way or the other. I think that my colleague from PPG would like to make a statement here.

DR. HEALEY: I know you're not taking public comments, but I think there were a number of questions directed at the industry at large, and I just felt it appropriate to inform the committee at this point I think there are kind of two sentiments within the industry.

On the one hand, I think there is a recognition that there is a need for fair balance with a warning statement already on the label offering some balance as appropriate in this circumstance.

On the other hand, I think there is another sentiment in the industry that says adding yet additional language about this theoretical risk is simply going to heighten or perhaps draw more attention to what is admittedly a very remote and theoretical risk at best, and therefore, any additional labeling language isn't appropriate.

So I just wanted to express to the committee that at this point there are a few sentiments out there within the industry.

Thank you.

DR. DeARMOND: Well, that kind of addresses one of the questions, the other question I had, and that is: would it be beneficial to the plasma purification industry to have this statement? And if so, then all of the companies that are involved, all 11 of them, should demonstrate that they can reduce infectivity even when it's spiked by a significant amount, whatever that turns out to be, anywhere from four to eight logs.

But if that's not the case, is it the desire of the industry, notwithstanding what was just stated, that those companies who have made the effort, that they should be recognized by a statement that their procedure reduces any additional possible risk and it's up to those other companies to come forward to be awarded that extra statement?

DR. BARON: Again, Steve, I don't think that I can answer that question for the whole of the industry. As has been pointed out, there are divergencies within the industry as to the answer to your question.

I would like to say though that getting back to the issue of how many logs of reduction are enough, I would like to remind everybody that we're still dealing with a theoretical risk, and if you could minimize that risk by adding one or two logs of removal, that's better than zero logs, and so any removal serves to add further reassurance as long as it can be demonstrated that it has been shown to be done by scientifically valid and reproducible methods.

CHAIRPERSON PRIOLA: I have one question that relates to a statement you made that 263K or -- I'm sorry -- hamster scrapie acts very similarly to CJD, GSS and variant CJD spikes, and this term "removal factor." You show large reduction in infectivity for variant CJD and CJD. How did you determine that? Was that by -- did you know the initial titer of the spike you put in? Was that done by Western Blot, by bioassay?

DR. BARON: This was done and most of these studies were done by Western Blot of by CDI amino assay. Essentially the way you determine it is you prepare your spiking material and you put it in your matrix to meet whatever medium you're going to be putting into your process, and then with that initial sample, you do a dilution series on it, and you determine your input quantity based on the last dilution where you had a positive signal.

So if you start with a five log of dilution and then in your output sample you've got one log, that gives you four logs, let's say, not removal but reduction is perhaps a more appropriate term when you don't have complete removal.

CHAIRPERSON PRIOLA: I guess I'm asking because this is what Douglas Lee did, right, for Bayer?

DR. BARON: Yes. That's what we do.

CHAIRPERSON PRIOLA: And that was based on the hamster scrapie agent. Has that been validated for the CJD agent and the variant CJD agent and the GSS agent that were used to compare the 263K?

Because if you're using the 263K technique for those other agents, that might not be valid because they're not hamster scrapie.

DR. BARON: I can speak for -- well, I'll let Steve speak for the bio data and then I'll speak for my own.

DR. PETTEWAY: If I understand what you're saying, I think the answer is yes. That's what we did for the CJD agents.

CHAIRPERSON PRIOLA: So CJD was bioassayed and compared to Western Blot dilution?

DR. PETTEWAY: No.

CHAIRPERSON PRIOLA: That's what I'm getting at.

DR. PETTEWAY: All of the CJD data is PrPSC partitioning.

CHAIRPERSON PRIOLA: Okay.

DR. PETTEWAY: All of it. There's no bioassay. The bioassay data was done with the hamster 263K where we compared infectivity with PrPSC partitioning. We just aren't able to do that with the --

CHAIRPERSON PRIOLA: Right. That's what I wanted to know because you're comparing an unknown human to a hamster standard essentially by Western Blot in terms of titer, in terms of --

DR. PETTEWAY: All of the sheep, the hamster --

CHAIRPERSON PRIOLA: Right.

DR. PETTEWAY: -- all of that was all PrPSC comparisons.

CHAIRPERSON PRIOLA: Exactly, but I guess I'm just thinking in terms of you have -- in doing these validation studies, what kind of spike to use. What could you use as standard agent to go through these spikes and, you know, show that it's valid, and I understand why you did and that you're using the Western Blot as a readout, but I guess what I'm saying is that you have the Western Blot readout but you don't know for CJD or GSS or variant CJD that that Western Blot correlates as well with the bioassay as your hamster stuff apparently does.

DR. PETTEWAY: If CJD infectivity is uncoupled with PrPSC, that's exactly right.

DR. BARON: May I add something to the answer? Because for the studies that we did with variant CJD, sporadic CJD and hamster prions in which it was the same thing, the samples once were a dilution series and we compared the CDI titer of the in-going sample with the CDI titer of the outgoing sample in terms of dilutions.

But those experiments, we actually are also doing a bioassay which is ongoing, and we're using for the variant CJD transgenic mice for bovine PrP.

CHAIRPERSON PRIOLA: One of the reasons I bring up the question is there is this transgenic mouse that overexpresses mouse PrP to a huge degree. They get sick very rapidly, die. You can barely detect PrPSC in them. So there's not always such a tight correlation between level of PrPSC and level of infectivity.

That's one of the reasons I bring it up as an issue, and I understand you're doing the bioassay at least in the bovine transgenic mice, yeah.

So are there any other questions?

(No response.)

CHAIRPERSON PRIOLA: Okay. Thank you very much, Dr. Baron.

DR. BARON: Thank you.

CHAIRPERSON PRIOLA: It was very nice.

The hearing is now open for statements from the public. Is there anybody in the audience who would like to make a statement to the committee?

(No response.)

CHAIRPERSON PRIOLA: I don't see anyone stepping forward. Perhaps we should take a 15 minute break here and then reconvene to discuss the question and go through the committee discussion and vote.

So return at four o'clock.

(Whereupon, the foregoing matter went off the record at 3:48 p.m. and went back on the record at 4:05 p.m.)

CHAIRPERSON PRIOLA: Okay. We're going to put up the question that the committee is being asked to consider, and it's a two part question. The first part can be answered with a yes/no vote. The second part we've just been asked to comment on the proposed label. We're not being asked to rewrite the label.

PARTICIPANTS: Aw.

(Laughter.)

CHAIRPERSON PRIOLA: Yeah. I knew that would disappoint you, yes.

Well, we would have enough time. We've got until Friday.

(Laughter.)

CHAIRPERSON PRIOLA: But definitely to comment on the label as it sits and what we think about it, just not to rewrite it. So Bill put the question up, I suppose, yes.

So the first part of the question, again, it would be a yes/no vote. Assuming adequacy of decontamination procedures in product manufacturing, should FDA consider labeling claims for TSE clearance in plasma derivatives based upon specific demonstration of TSE removal during manufacturing?

So that's open for discussion. Yes, Steve. Dr. Wolfe. There's two Steves.

DR. WOLFE: I raised this question before lunch, and other people were very articulate in their own takes on it, but I still am worried about this issue of a double standard.

Implicit in even Part A is should FDA consider labeling claims for TSE based upon specific demonstrations. This means that if companies, through the what has now become known as dossiers -- I think dossiers generally mean something else -- but a very good recitation providing data showing that you have gotten a bunch of logs down below, what the theoretical risk is.

I am still troubled by the idea that this is a voluntary process. Those that wish or don't wish to do it, don't wish to spend the money and the resources that have clearly been spent, and I thought these were good presentations as well, can then hang in with the original existing label and those companies will not have to spend as much money, and it is entirely possible that those products will be less expensive.

And in this horrible area of double or triple standards in medicine, it's entirely possible that some people without knowing it necessarily, patients, will wind up getting the lower priced spread, so to speak.

What I am objecting to in this whole process, and I think it has become even clearer to me after these presentations, is that the FDA is really implicitly proposing to allow two different standards for salable, quote, FDA approved plasma products. One is, okay, if you don't want to submit the data dossier, you can stay with the existing label, but if you do, we will consider it, and if you meet certain kinds of standards, you will be able to have the new label or some variant thereof.

So I would just like to raise the question about why is it necessary to have this be done on a voluntary basis in such a way that we will guarantee two different -- I mean, again, assuming that the theoretical risk is -- I think that the presentation that was made by Dr. Baron was the responsible approach is to treat the theoretical risk as though it were real, and I agree with that statement. That's the responsible thing to do.

And so if we are treating it as though it's real, some people are going to really do much more to reduce the theoretical risk. They will get credit for what they've done, and they will get to have this new label, and the others that chose not to do that will have the old label, but both of these products or sets of products, whatever, will be approved.

When I asked Dr. Petteway during the break what he estimated were the number of companies -- and I will not hold you to that. You said it was just a guess -- I don't know if it was the number of companies or the number of products or the percentage of the market that would actually submit these. He said about 50 percent.

Whatever it is, it is clearly not going to be the whole industry, and it will lead these others out there. So I would really like to hear from FDA why it is necessary to move in a direction that will guarantee a double standard for at least theoretical protection from the risk of these agents.

DR. SCOTT: I think as Dr. Epstein intimated, that we have been asked to consider this, but that isn't the only reason to consider it. I think that in terms of having actual data to support the possibility that there's reduction of TSE agents in plasma derivatives, this is something that we would like to see, and this is among the ways of beginning to see that.

Now, there are a lot of caveats to these kinds of studies. They aren't going to be scientifically perfect. Nevertheless, we don't want to wait until we know the perfect science, such as the exact form of TSE infectivity in blood.

The many years it's going to take for us t find out whether or not it's transmitted by blood or plasma, we'd rather be able to take measures and to encourage that measures be taken so that we can understand these products in that regard.

And I would also like to point out that it is by no means certain that we have substantial clearance of TSEs in all products, first of all, because we don't have the data, but second of all, just guessing based on what's been published, a risk estimate, and applying that to specific manufacturing procedures suggests that certainly there are going to be varied levels of clearance, and that might be something that's important. That's even something that might drive us further, all of us, in considering whether or not it's necessarily even to do more.

I don't want to alarm anybody. I don't think personally that I've seen a red flag, but I think that if we progress in these studies we'll find out a lot more about the products and TSE clearance, and I think among the ways to progress in these studies is to encourage the submission and generation of data.

DR. WOLFE: No one can disagree with the idea of encouraging the submission and generation of data, but you are pretty much saying, I understand, I think, clearly that the incentive for doing the study, submitting the data, the carrot, so to speak, is this label, which will be more acceptable.

I mean, there are two parts to the label. One is the part, the separating out the viral issues from the TSE issues, and clearly one is a theoretical risk, and one is a real risk. That can be done with a modification of the existing label.

But the second piece of this labeling change beyond that is giving these companies credit for the work they've done and the demonstration or documentation of the work through their submissions, dossiers.

There's two different kinds of things. Any company, whether they are willing to submit the stuff or not and get the new label, should and could benefit from a clarification of the existing label on the issue of the distinction between the TSE and the viral issues, but only those that submit would get the second thing.

So we still are left, even though you -- I mean I understand we always like to have incentives to companies that do things, but it's an incentive that starts separating out companies from those that want to and have done it.

David's statement before is that this is not going to be a public health advance because it's already being done by these companies. But if other companies did it, it would be possibly or theoretically a public health advance for them. If they don't, it's okay because they can keep selling their FDA approved product.

So I think I've made my point clear and you've made yours clear, but it still gets down to a double standard for products theoretically if this goes forth as planned or as described.

DR. SCOTT: Yeah, I think you're correct it would.

CHAIRPERSON PRIOLA: Dr. Johnson.

DR. JOHNSON: Can we go back to worst case scenarios?

I mean, I think there are two bad scenarios in this, and I just want to see if they can be avoided. One is you have two products that are considered to be of two qualities, not because they are. They may be just alike, but the label, the label looks different on one.

In a population like the hemophiliac population that was brought up, where they're very aware and they read their labels, and they really know what's going on, and they read it in their HMO or Medicaid or it says, okay, cheapest product, and they sort of get the feeling that they've been done as second class again.

I think that's a terrible worst case scenario that this could be. The other worst case scenario is what if there's a transmission of CJD by one of these products. What if the one that has the low level warning on its, there's a case and you say, "Oh, gee, you should have made them" -- you know, that's because of what all made the high one, or what if it's one of the people who has the high label who comes down and everybody says, you know, "What do you think you're doing around here?"

I mean, I don't see much advantage other than sort of -- I mean, I understand the idea of an incentive to the companies to sort of get on board and do this, but if it really becomes an issue of having two qualities of products at two prices, that becomes a problem or if you really do believe that there might be a transmission one day, which I don't think there will be, but there could be.

I think having two levels of products is just looking for trouble.

CHAIRPERSON PRIOLA: Dr. Bolton.

DR. BOLTON: I was actually going to say something else, but I think if there are two levels of products, there are already two levels, two different qualities or more different qualities of products. The products that are on the market now are not equivalent.

DR. WOLFE: But they all have the same label right now.

DR. BOLTON: They do have the same label. Okay, but I don't know that it's unfair to give the consumer the right to additional information if such information exists, and this is, after all -- the demand is for scientifically validated information that will be reviewed and passed upon by the FDA.

But I guess my point is I wanted to bring up something actually that Steve Petteway had said earlier, and not to be a spokesman for industry necessarily, but just so he doesn't have to say it again, is that part of the reason for having the amended label would be to mitigate some of the negative perception that may be conveyed by the existing wording which already is mentioning a theoretical risk of CJD.

And I sort of joked to Dot before. I said that the companies that submit these dossiers, that this may or may not be the correct word, and have it approved, they might prefer to just have all language with respect to CJD removed, and then they would be happy. Obviously that's not going to happen.

CHAIRPERSON PRIOLA: Dr. Hogan.

DR. HOGAN: Yeah. I was going to say something else, too. But you brought up the point there are two levels of products and how can you evaluate between those two. I think we have heard from Sue and Dot and other people today that there's problems with equating any of this data in terms of what's spiked, whether that's meaningful, whether or not you can equate that across processes, and then are the bioassays reasonable to evaluate across the board.

So what we're really going to ask the FDA to do is to look at a case-by-case basis and see is that real and then determine whether or not that company is going to be able to market their product at a higher level, because it's going to cost more to do that, than another company.

That's an incredibly difficult problem. That's their problem, but I would suggest that we have like in Factor 8 where there's one log removal, and you've done studies; I've done studies. One log is tough to measure. One is better than zero, but how do you know that that wasn't with purified PrP? How do you equate those?

DR. BOLTON: Yeah, I think -- Steve, do you want to go ahead?

DR. PETTEWAY: Well, you know, as it stands right now, we're kind of in a similar boat with viral validation, for instance. The difference is that we have a very well laid out set of guidelines that we all follow so that we do the studies all the same way, but we don't always use the same viruses, and we don't use the same spikes, and there's no standardization of assays.

So we're all independently on a case-by-case basis demonstrating the viral safety of each individual product already.

So this issue is not different than what we struggle with with viral safety anyway, not to say that that's all perfect. That's just the way it is now, and this follows a similar paradigm.

DR. WOLFE: Well, again, for viral safety there's one label now. WE are here talking, proposing two different labels.

DR. PETTEWAY: A different issue.

DR. WOLFE: Okay. Good.

CHAIRPERSON PRIOLA: Dr. DeArmond.

DR. DeARMOND: But as I see it though, I think we're getting ahead of the game here. We have two questions, and the first question is more of a scientific question, and that leads, depending on our answer to that, leads into the second one. The second one may not be acted on at all.

In fact, we could say yes on that one and say that we don't think you should work on or that that should be deferred. The second one should be deferred.

So I still would begin with the program that was sent to us and ask ourselves should the FDA consider labeling claims for TSE clearance in plasma derivatives based upon specific demonstration of TSE removal during manufacture. Have we got enough data to say yes or no about that as a general rule, based on this remarkable data that was presented to us?

DR. BOLTON: I wanted to go back to Dr. Scott's presentation. I think it's her second one, where at the end she brought up this question of what kinds of criteria should be used, and I guess you probably want some discussion from us on that, but I think clearly, as Dr. Hogan brought up, I would not think one log clearance in any one step is meaningful at all because we all know these assays in any type of assay bounces around.

And two logs is even questionable. Once you get above reproducible three logs or four logs, I think that's really meaningful, and I hope that when they evaluate this information that's submitted that they would take those types of things into account.

DR. DeARMOND: But that's true for the second statement, the second statement, because this product, which is all of the different products, that you fill in the specific name of the product there, and each one of them will have a different clearance, and each one of those are going to have to have their own judgment on them. That's part of the second question.

We begin with this first question: have they demonstrated any clearance.

DR. BOLTON: No, I think the first question is generic. Should the FDA consider --

DR. DeARMOND: Yeah.

DR. BOLTON: -- this information?

DR. DeARMOND: That's right. Based on the type of data we've received so far.

CHAIRPERSON PRIOLA: I think that this is exactly it. Should the FDA consider, based on the standard of scientific evidence we have, such a label?

And it seems like most people agree here that there's quite a bit of good evidence that you can clear TSE agents with certain of these manufacturing steps.

I think the second question has to come out of that because if you vote yes and then say, "But we don't want you to do a label," you're giving a mixed message. We don't want to talk about the label, but we think you should do it.

So I think that the questions are very closely related

DR. BOLTON: But I would like to just make clear at least my perception is -- and I would not vote for a generic, yes, you should allow the label to be changed based on the data that was presented here today, even though I'm impressed with the data today.

I'm talking about should the FDA go ahead and review individual submissions that are much more detailed than what we've got here on specific steps of specific processes by specific manufacturers and, assuming that that is acceptable, then they would allow that label to be changed on that product.

That's what we're asked to do, right?

CHAIRPERSON PRIOLA: Yeah, right. I think that's exactly right.

Did you want to say something, Dr. Scott? You had popped up there.

DR. SCOTT: I think it was said well for me.

DR. DeARMOND: Yeah. So that's what that first question is. Should they begin the process of doing this? That's what we're saying, and it's either yes or no. We don't think there's enough data for them to even begin to think about it or, yes, there is enough data that they can look at each individual case.

DR. HOGAN: I would suggest that it's not yes or no, that it could be --

(Laughter.)

DR. HOGAN: -- it could be, no, there's not enough data -- maybe I'm always a fence sitter. I don't know -- but there's not enough data to say they have to change their label, but you could encourage the industry to continue to do what they're doing and come forth with the data as it's developed.

DR. DeARMOND: But it's should the FDA consider. It says nothing about changing labels or anything. Should the FDA consider label claims? Is there enough data to even start that process?

It's a different -- it is a yes or no and not a maybe.

DR. HOGAN: Okay.

CHAIRPERSON PRIOLA: Dr. Epstein, do you want to respond to some of this?

DR. EPSTEIN: Well, I just wanted to make a comment back to the points that were made by Sid Wolfe about two qualities of product. It needs to be understood that clearance for TSE agents is not currently a standard of FDA review and product approval. So we're not talking about two product standards.

What we're talking about is whether the product label can communicate to the public -- that's the professional and patient community -- any distinctions in what is known about the manufacturing process.

Now, there are precedents for that in the diagnostics area, for example, is a screen sensitive for HIV-1 Group O as well as non-Group O, where we don't have a claim that it's a Group O sensitive test, but we allow performance information with some set of Group O samples to be put on the label. It tells the user something.

Now, I think that I agree with the point that if we think that that kind of labeling will create a misinterpretation of relative product safety, in other words, it's not based on real safety assessment, then we ought not to do it, and those who think that ought to advise in the negative on question one.

If we think that it's, in fact, real information, that it speaks to safety assurance, even though it's not an approval standard, then perhaps there's a role for it.

And I just want to echo a statement that was made earlier. It may have just sort of slipped by, but why this is happening is that there was a lot of historical resistance by particularly the fraction areas, but also product manufacturers of other things like vaccines, to put the generic statement encompassing the risk or CJD risk on the label because they felt that that was going too far considering that it was a theoretical risk, and at that point in time the balance of the discussion and FDA's decision was consumers ought to know about the theoretical risk if, you know, we, the experts, were concerned about it.

So the decision in the end was to put this statement on as part of the warning on the label, and so the consequence of that has been a desire by the companies to make that risk label go away, and that has been an incentive for many companies to do the kinds of studies that you're seeing. They don't want the products to be flagged as, you know, worrisome because that has a chilling effect on doctors' willingness to prescribe and patients' willingness to use.

Now, recognizing that many patients are obligate users, but still there's the issue of people being more judicious with the product. So these labels definitely have a driving force both on medical practice and on development within the industry.

So there's some sense that if we can say meaningful thing on the label that are reflective of real scientific data that has safety meaning, that we ought to because that's a net benefit to the public and to the patient.

But the countervailing argument here is if it's not scientifically meaningful, if it's not a real assessment of relative safety or validation, then it does more harm than good.

So I think that's where the line is drawn, and I tend to agree with Dr. DeArmond. First questions first. We are looking for a global assessment whether the kind of data that are available to us today do or do not rise to the level where we ought to be considering specific claims for specific products by specific manufacturers with specific processes.

DR. WOLFE: Let me just rephrase what I said before, which is the assumption is -- and it's part of what's imbedded in the first question -- that it looks as though, assuming that there is some TSE particle derivative in plasma, it looks as though these steps would reduce the theoretical risk. That's the operating assumption. If that's so, then part of question one follows that those companies that are willing to submit the data that show that they have been able to document the reduction in theoretical risk will get to have the better label.

And again, without ever getting to the second question you are left by saying yes to the first question with an almost guarantee of a double standard because you wouldn't say yes to the first question unless you believed that there was some scientific basis for considering a labeling claim.

If you believe that, then you consider the labeling claim from that fraction of the fractionating industry, so to speak, that is willing or able to spend the money and time to develop that, and you lead just with question one, without ever getting to question two, to this double standard of products, and this is really what I object to and why I think that even on question one you're already heading in that direction or you're essentially saying to FDA, "Yes, we think that the data we've seen today is impressive enough," which I agree it is with my less knowledge on this particular topic than some of you, "and we think you should go along the path of looking at it in a company-by-company or product-by-product," saying whether it does or does not qualify for some yet to be worked out language in question two.

And again, the reason that I object to it is that this essentially is giving the FDA license or permission or advice since we're an advisory committee to go in the direction of a double label, double standard rather.

CHAIRPERSON PRIOLA: Mr. Bias.

MR. BIAS: Probably just background, I guess I would agree with Dr. Epstein's statements, but in terms of background, in terms of patient use, I have not experienced a change in product in the however many years I've been using these products that didn't result in some kind of price change. That's inevitable. The business today is less about lower priced products, but more about theoretical safety levels that cost more.

I mean, none of us can prove. We're thankful that we've not had any problems with the protein products that are produced today, but theoretically we really can't prove they're any better than the plasma products.

So it is a sense of reassurance for the patients, for the health of people in the United States.

The other thing I wanted to say is people don't generally change, don't do product of the month stuff. Okay? You use the product that works for you, and you use it consistently, and the only thing that would make you change is if your insurer decided that they weren't going to pay for it anymore, which is rare.

I mean, that happens now that we're in the world of HMOs, but it doesn't happen -- it's not rampant throughout the world, and it is regional in some aspects.

So people don't change with the product of the day if a label changes or something like that. You use what works for you and the only reason you will change is that, you know, the FDA walks in the plant and shuts it down and you're forced to change. People don't generally change the flavor of the month. That's just not how these products work. They are really -- you become in tuned, your body becomes in tune with a particular product, and your body responds to it in a different way.

So I just wanted to give a little bit of that as background to the committee so you would be aware of how people used them, but if the FDA thinks this is a good idea, then we'll encourage all of the companies to go. And I agree with them. If one of the companies gets this new label, eventually they'll get the new label. They'll all do the work to do it.

They're not going to sit back and let somebody get a competitive advantage or be able to put that in their ad because our community is very small. So all of the ads run in the same magazine.

(Laughter.)

MR. BIAS: You just go to page 20, 25. They're all in the same place at the same time. If they run on the local level, it's the same ad that runs on the national level, and I've tried to explain to them that, you know, marketing is not the way it goes. People don't necessarily select a product just on marketing.

But you know, if they want to put money into that, it's their company. So I just wanted to kind of give that as a background.

CHAIRPERSON PRIOLA: Dr. Gaylor.

DR. GAYLOR: I think the key word here in that first question is the word "consider," and if I say no, I'm saying, "FDA, don't even think about this. Let's go home now. We're not even going to think about it." Don't think about anything that was said today. Just forget this whole meeting if we vote no.

I don't think voting yes is saying, yes, go ahead and change the label. It's just saying, "FDA, think about it."

I vote yes. I call for a vote.

DR. BRACEY: Yeah, likewise I'd be interested in calling for a vote, but I think that one thing that is important in terms of users, and the user out in the field expects that if the FDA really feels that this is an important issue, that it should become a standard.

Maybe we don't have enough information now, but it needs to be considered. We can't drop the ball here, but if it's important, then it should be applied to all of the manufacturers. We don't have that data. I think we need to continue to explore whether or not this needs to be the case, but I feel quite uncomfortable with the doubt standard because, in essence, it really doesn't offer the less sophisticated user, which there are many of us out there in the real world, the option of selection. It puts you in a box. It gives you no option.

So I think the idea thing is to continue to consider, try to determine whether or not it should become the standard and apply it to all if it really is that important.

CHAIRPERSON PRIOLA: Dr. DeArmond.

DR. DeARMOND: The term "double standard" has for me kind of a negative connotation. It's sort of like men and women dating and things like that. And there's a different way to think of it, I think, as more towards the free enterprise system and people do have a right to make advances and to be better competitively.

So I don't think being competitive -- it would be like saying a great tight end. There's a double standard. He could just catch balls better than anybody else and other people can't. It's a double standard for them. They should be out there catching also, at least be given the chance.

So the double standard I don't think is the right term in this case, and I think we are dealing with a system in which those who make the greater advances should reap the rewards to some extent.

And I don't think this question is dealing with those sorts of businesses or the sociology of how people purchase things. I think we're still being asked a relatively, I think, straightforward scientific question, and I think we should keep our thoughts to that.

CHAIRPERSON PRIOLA: If there are no other comments, would there be any objection to bringing this question to the yes/no vote?

Okay. Let's do the vote.

DR. FREAS: I will quickly poll the members at the door, if you don't mind, and then I'll go around the table as we normally would.

(Laughter.)

DR. FREAS: Dr. Gaylor.

DR. GAYLOR: I vote yes on number one.

DR. FREAS: Dr. Bracey.

DR. BRACEY: Yes on number one.

DR. FREAS: Now going around the table in the usual order, Dr. Johnson.

DR. JOHNSON: Yes.

DR. FREAS: Dr. Bolton.

DR. BOLTON: Yes.

DR. FREAS: Dr. Belay.

DR. BELAY: Yes.

DR. FREAS: Mr. Bias.

MR. BIAS: Yes.

DR. FREAS: Dr. Williams.

DR. WILLIAMS: Yes.

DR. FREAS: Ms. Walker.

MS. WALKER: Yes.

DR. FREAS: Dr. Hogan.

DR. HOGAN: Yes.

DR. FREAS: Dr. Priola.

CHAIRPERSON PRIOLA: Yes.

DR. FREAS: Dr. DeArmond.

DR. DeARMOND: Yes.

DR. FREAS: Dr. Wolfe. I'm sorry.

DR. WOLFE: No, for the reasons stated.

DR. FREAS: Dr. Ferguson.

DR. FERGUSON: Yes.

DR. FREAS: And realizing that industry doesn't have one position, Dr. Petteway, would you like to make some comments?

DR. PETTEWAY: Yeah, unfortunately I can't vote on this, but I'd just like again to say, and I think you guys have addressed that, that if you have to have a warning statement that it's not unreasonable to put it in appropriate context in a label.

DR. FREAS: The summary of my tally on voting is 12 yes votes, one no vote, and industry's comment.

CHAIRPERSON PRIOLA: All right. We can go on to the second part of the issue, which we've actually discussed quite a bit already, the wording of the label. Are there any further comments about that issue?

Dr. Wolfe.

DR. WOLFE: Yeah. Although I voted no on number one, I attempted to dissociate the two changes in the label that are imbedded in this proposed one. I still think that one of them could be accommodated without this risking what at least I still believe is a double standard, which is to clarify the existing label and the existing label without any further submissions.

I mean, again, I would love to see companies adopt a uniform standard as opposed to a double standard, but short of that, it seems to me that distinguishing between the viruses and TSE agents could be done by just a modification of the existing label because that is true for any company, whether they have or have not adopted the new technology to reduce the theoretical risk of logs.

So this is sort of an amendment in a way to the second question, which is I think we should consider or recommend to the FDA that before getting all of the data and dossiers, that we should think about changing the existing label which applies in a single standard to all products to distinguish between the theoretical risk of the TSE versus the real risk obviously of the viral infections.

CHAIRPERSON PRIOLA: David.

DR. BOLTON: I was just going to recommend that the FDA, when they decide what this wording is or should be, that they, first of all, keep the lawyers out of the room; secondly, read it to your grandmother and your 12 year old, and if they understand it, then you probably have good wording.

CHAIRPERSON PRIOLA: Sorry. Dr. Belay.

DR. BELAY: I don't know why the vCJD was slipped in the second part of the sentence. At least one of the advantages for the current labeling does not mention vCJD, and the mere mention of vCJD may scare even some physicians, and I'm concerned that this proposed labeling may actually be more concerning, more alarming than the current labeling because it mentions vCJD, and that's one of the advantages for the current labeling, because it doesn't even mention vCJD.

CHAIRPERSON PRIOLA: Okay. Dr. DeArmond.

DR. DeARMOND: Yeah, kind of following up on that and one of your comments, Sue, was that a lot of the data is based on hamster, and they haven't really shown yet that vCJD is cleared or sporadic CJD is cleared to the convincing level that they did with hamster.

So even that final statement would have to be modified to some extent or generalized to TSEs.

DR. BELAY: What I wanted to add was this is my recollection. I may be wrong, but the reason, one of the reasons probably vCJD was not included in the current labeling was probably because that issue has already been dealt with by the deferral criteria.

In other words, substantial risk that would come from vCJD was addressed in the deferral policy, whereas for CJD, I mean, the deferral policy will not address more than, as we heard today, more than probably 20 percent of the risk.

I just wanted to make that point.

DR. SCOTT: Can I respond to that?

CHAIRPERSON PRIOLA: Yes.

DR. SCOTT: Yes, I just wanted to point out that the negotiations about labeling, the internal negotiations, began before vCJD was really much on the radar screen and that subsequent to that we've left it as CJD for the time being, but that actually it was not linked to the donor deferral. Not including CJD in the labeling was not linked to the geographic donor deferrals which only just went into effect, whereas the labeling came long before.

So I just wanted to make that clarification.

CHAIRPERSON PRIOLA: Dr. Johnson.

DR. JOHNSON: Although I voted that FDA should consider it, I certainly don't think we want to change to what's worded here.

(Laughter.)

DR. JOHNSON: So I will vote against this.

CHAIRPERSON PRIOLA: Any other comments?

MR. BIAS: Just one. I'm just not going to wordsmith and assume that it's an extremely remote possibility that they'll use this wording.

(Laughter.)

CHAIRPERSON PRIOLA: And your major objection to it is that it's difficult? What's your major objection to it? Is it similar to --

MR. BIAS: As mentioned before, I'm having trouble, you know, figuring out what "extremely remote" means. A quantified explanation would be better, but as they said, they don't want wordsmithing. So I'll stop at that point.

CHAIRPERSON PRIOLA: Did you have a response to that, Dr. Scott?

DR. SCOTT: I just want to say that the more comments we can gather now, the most helpful it will be, and we are not going to fall on our swords over this kind of labeling at all, and in fact, that's why we didn't put it up to a vote.

CHAIRPERSON PRIOLA: Dr. DeArmond.

DR. DeARMOND: Would it be inappropriate with regard to the question about what has been demonstrated by the clearance studies to ask Dr. Baron to say whether it's inappropriate to categorize what he said as purely just a hamster clearance, or does he feel that he has also demonstrated human clearance, CJD clearance?

DR. BARON: Is it okay?

CHAIRPERSON PRIOLA: It's fine. Sorry. Go ahead.

DR. BARON: I think for at least -- let's see. There were three steps for Bayer and two steps from Aventis Behring in which I showed data which demonstrated, I think, very clearly that vCJD, sporadic CJD, and hamster prions all had very comparable removal or reduction factors for those five steps evaluated.

So I think the data is going in the direction that the hamster prions are predictive of the partitioning of vCJD prions. Now, I think that it would be wise to do this kind of bridging study over more steps. I don't think you need to do it for every step, but I think it would be also wise if other companies other than Bayer and Aventis Behring also generated some of this comparative data because I think that's what's necessary to convince that the use of the model rodent prions is relevant.

But I think the data so far is very strongly supportive of the fact that the hamster prions are predictive.

DR. DeARMOND: So does that change your view of this, Sue, at least with regard to a couple of specific products by a couple of companies?

CHAIRPERSON PRIOLA: Yeah. No, my objection was more to the quantifying a human based system using a hamster standard. I think that with what the user is probably showing by Western Blot that you get lower and lower levels of PrPSC as you go through some of these steps, and I think we can all agree that that almost always means reduction in infectivity.

So my concern is just the level claimed, which isn't really part of the discussion here. I think that they are clearing infectivity. Is it the amount that they're saying or is it different because of the different properties of the agent?

Dr. Belay.

DR. BELAY: This is just a suggestion. The phrase "extremely remote" could also mean it occurs, but it is extremely remote. In other words, it's better to characterize it as theoretical. In other words, it's not true. Nobody has identified it, rather than, you know, introduce another phrase saying it's extremely remote. The word "theoretical" actually probably conveys the fact more so than, you know, the phrase "extremely remote."

It's more remote than extremely remote.

(Laughter.)

DR. BOLTON: And more theoretical as well.

CHAIRPERSON PRIOLA: Oh, sorry. Go ahead, Dr. Wolfe.

DR. WOLFE: Well, I mean, the current labeling says "theoretical." So you're essentially advocating for the language there, which I was just simply saying the current labeling in mentioning only CJD and not vCJD is inaccurate, but as was pointed out, it was developed at a time before vCJD became the radar screen and of concern.

So that could be a modification, again. The current label now just adds vCJD to make it clear that it is a theoretical risk, and that would, again, apply in the interim. Because, I mean, I think it's possible for FDA to make a slight change in the existing label now even if you are inclined to go ahead and make further changes based on manufacturing submission, which I would oppose for the reason I said.

But I think that the easier step is just to change that now because that's just an historical aberration that it was missing vCJD in the first place, and it, again, is accurate and applies to anyone making plasma, the theoretical risk.

CHAIRPERSON PRIOLA: Dr. Bolton.

DR. BOLTON: Just in the interest of making sure we don't get out of this meeting on time ‑-

(Laughter.)

DR. BOLTON: -- and we said we weren't going to wordsmith, but I guess we are, but I would just suggest that a lot of people may not understand exactly what we mean by theoretical risk, and that if instead the wording is changed to something like undemonstrated or not proven or something along those lines, then it might be a lot clearer to the general public than theoretical risk.

CHAIRPERSON PRIOLA: Other comments? Oh, yeah, Dr. DeArmond.

DR. DeARMOND: I think the problem we have is that there are multiple products, in fact, more products that we even discussed today that would fall under this; multiple companies, and each with a different purification procedure; and that the FDA is going to have to modify this for each one of those, and really I think what I would suggest is that they have to look at each product, and they're going to have to modify this to adapt the wording to match that precise product in that precise company with the reductions that are demonstrated to them.

This is too generic right now.

CHAIRPERSON PRIOLA: I think my understanding of this labeling is it is intended to be generic, but that to get the label, you have to provide that your specific process clears the infectivity. So the label is intended to be generic, but it's generic for a specific process.

DR. DeARMOND: Maybe I should ask the wisdom. Do they mean it as a generic or is it supposed -- does the word "product" -- is it going to be substituted with Factor 8 or something else?

DR. SCOTT: The intention was that if we're satisfied with a certain beltline level of clearance, however that is obtained, that the label would be as you see, generic and not reflect exact amount of clearance for each step, for example, which is what I think you're asking.

Are we absolutely tied to that? Not necessarily, but that was the intent.

DR. DeARMOND: I'm not sure I'm asking for a specific clearance, but a reasonable clearance that would be acceptable. One log is probably insufficient. Four logs is far better, but each product and each process is going to be different. So that's where it makes a general statement more complicated to apply.

And you're going to have to review all of this data that comes in on specific processes.

DR. SCOTT: Yes.

DR. BOLTON: Dot, can I assume that given our vote on number one that what FDA will begin to do is to develop criteria that will allow a company to pass this? The criteria may be the same for similar products or may vary from product to product.

Jay is sort of nodding one way or the other, and then once you pass that, once that's determined, then meeting that would then allow the generic wording to be put on the label.

DR. SCOTT: Well, I'll try to read Jay's mind, but I think that it's important for us to provide consistent reviews and to the extent possible, to communicate to industry what we would like to see in terms of degrees of clearance.

Have we figured all of this out yet ourselves? We have not, but it's incumbent on us to do this in some fashion, I think.

CHAIRPERSON PRIOLA: I would like to make one comment about the separating viruses from CJD, and in looking at the way this label, which will likely not be the one that is used, is written, that it seems very easy that you can make the first part of the label about viruses so that you end it after, for example, viruses and then go into "it has also been demonstrated that it removes the theoretical."

So then you separate the two quite clearly. You're not trying to put viruses in the same light as CJD. It's still hopefully clear. It gets rid of "theoretically." But that's just a suggestion.

Any other comments on the labeling from the committee?

Dr. Baron.

DR. BARON: Thank you.

I just wanted to ask a question to Dorothy or to Jay on the side of the FDA, and that is that assuming this process moves forward and you get then into the process of developing criteria, will you be seeking the outside participation of experts, including people from industry in partnering with you to try to develop some of these criteria together?

DR. EPSTEIN: I would think so.

(Laughter.)

CHAIRPERSON PRIOLA: Dr. DeArmond?

DR. DeARMOND: So, Sue, what are we supposed to do now? Because I don't know how to word these things. My wife works in the drug industry, knows how to use words, but this is beyond anything I can do.

CHAIRPERSON PRIOLA: I think that what we've done is, I hope, what the FDA wants and that's discuss where we like it, where we don't like it, suggestions, and then they will have to deal with specifics in even considering it, if that's correct.

DR. EPSTEIN: Yes. I think we've heard a very good discussion and many very useful comments, and we're just going to have to go home and think about it.

CHAIRPERSON PRIOLA: So the FDA does all of the work now.

DR. DeARMOND: Great.

CHAIRPERSON PRIOLA: If there are no other comments or suggestions, I can declare this meeting adjourned. So thank you all.

DR. FREAS: Thank you for coming through the snow and attending this meeting.

(Whereupon, at 4:52 p.m., the meeting in the above-entitled matter was adjourned.)