UNITED STATES OF AMERICA
FOOD AND DRUG ADMINISTRATION
BLOOD PRODUCTS ADVISORY COMMITTEE
FRIDAY, OCTOBER 22, 2004
This transcript has not been edited or corrected, but appears as received from the commercial transcribing service. Accordingly the Food and Drug Administration make no representation as to its accuracy.
The meeting came to order at 8:00 a.m. in the
Ballroom of the Gaithersburg Holiday Inn, 2
Montgomery Village Avenue, Gaithersburg, MD 20877,
James R. Allen, Acting Chairman, Presiding.
James R. Allen, M.D., M.P.H., Acting Chairman
Kenneth Davis, Jr. M.D., Member
Samuel H. Doppelt, M.D., Member
Harvey G. Klein, M.D., Member
Judy F. Lew, M.D., Member
Charlotte Cunningham‑Rundles, M.D., Ph.D., Temporary
Jonathan C. Goldsmith, M.D., Temporary Voting Member
Liana Harvath, Ph.D., Temporary Voting Member
Blaine F. Hollinger, M.D., Temporary Voting Member
Matthew J. Kuehnert, M.D., Temporary Voting Member
Kenrad E. Nelson, M.D., Temporary Voting Member
Keith C. Quirolo, M.D., Temporary Voting Member
George B. Schreiber, Sc.D., Temporary Voting Member
Michael D. Strong, Ph.D., Non‑voting Industry
Linda A. Smallwood, Ph.D., Executive Secretary
A.Summary of Plasma Workshop
held on 8/31‑9/1/04
Mark Weinstein, PhD.. . . . . . . . . . . . .5
Draft UDHQ Acceptance Guidance:
Review of Public Comments
Judy Ciaraldi, BS, MT (ASCP) SBB. . . . . . 18
FDA's Current Thinking on Monitoring
Weight in Source Plasma Donors
Linda Alms, BS. . . . . . . . . . . . . . . 33
Open Committee Discussion
FDA's Current Thinking on Donor
Deferral for Potential
or Documented Infection with
West Nile Virus
1. Introduction and Background
Hira Nakhasi, PhD, Director,
Division of Emerging and
Diseases, OBRR . . . . . . . . . . . . . . . . 44
2. Summary of 2004 Epidemic
Theresa Smith, MD, MPH,
FACP, FIDSA, Centers for
Disease Control and
Prevention . . . . . . . . . . . . . . . . . . 52
3. Duration of Viremia/Experience
With ID NAT
a. Michael Busch, MD, PhD,
Blood Centers of the Pacific. . . . . . . . 77
b. Susan Stramer, PhD,
American Red Cross. . . . . . . . . . . . .109
Public Session. . . . . . . . . . . . . . . . . .138
Questions for the Committee . . . . . . . . . . .179
Adjourn . . . . . . . . . . . . . . . . . . . . .212
DR. SMALLWOOD: Good morning, and welcome
to the second day of the 81st Meeting of the Blood
Products Advisory Committee.
I'm Linda Smallwood, the Executive
Secretary. I will be reading a brief announcement
that pertains to the proceedings for today.
This brief announcement is in addition to
the Conflict of Interest Statement read at the
beginning of the meeting on yesterday, and it is a
part of the public record for the Blood Products
Advisory Committee Meeting on October 22nd, 2004.
This announcement addresses conflicts of
interest for Topic 3. Drs. Charlotte Cunningham‑
Rundles, Jonathon Goldsmith, Liana Harvath, Matthew
Kuehnert, Kenrad Nelson, Keith Quirolo and George
Schreiber, have been appointed as temporary voting
The Food and Drug Administration has
prepared General Matter Waivers for the special
government employees participating in this meeting who
required a waiver under Title 18, United States Code
Dr. Michael Busch is employed by Blood
Systems. He has contracts, is a researcher, speaker
and an advisor for firms that could be affected by the
Dr. Theresa Smith is employed by the
National Center for Infectious Diseases, in Fort
Collins, Colorado, and Dr. Susan Stramer is employed
by the American Red Cross.
In addition, there maybe regulated
industry and other outside organization speakers
making presentations. These speakers have financial
interests associated with their employer, and with
other regulated firms.
They were not screened for these conflicts
of interest. At this time I am asking if there are
any declarations to be made by any of the participants
at this meeting, please do so at this time?
DR. SMALLWOOD: For those who were not here
yesterday, I just wanted to announce the tentative
meetings, the tentative meeting dates for 2005, for
the Blood Products Advisory Committee.
Those dates are March 17th and 18th, July
21st and 22nd, December 1st and 2nd. Again, these are
tentative and you will be notified when these dates
are confirmed through the normal, appropriate
At this time I will turn over proceedings
of this meeting to the Acting Chairman, Dr. James
DR. ALLEN: Good morning. We'll start our
deliberations this morning by listening to a series of
updates. The first is the summary of the Plasma
Workshop held August 31st, through September 1st, this
year, by Dr. Mark Weinstein.
DR. WEINSTEIN: Thank you, we have the
slides, please. You'll be controlling the slides?
Okay, thank you.
I would like to review topics that were
discussed at the Workshop on Plasma Standards. I will
give you a review of the, next slide please. Of the
objectives of the workshop, a meeting summary, a
summary of the agenda, and some of the highlights that
were addressed during the meeting.
And some of our future actions. Can I
have the next slide? The objective of the meeting was
to obtain information to aide us in the development of
regulatory standards for plasma.
Particularly for recovered plasma,
including labeling, freezing, storage and shipping
conditions. We also wish to review the scientific
data, regulatory requirements and current industry
practices, regarding freezing, storage and shipping of
Another objective was to see whether we
could help to harmonize our regulations with those of
other regulatory bodies. And fourth objective was to
ensure that any regulatory decisions that are made,
are based on the science, the need for change and the
practicality of implementing any change in
regulations. Next slide.
Regarding our, the goals of the, with
regard to policy making, we want one to be able to
identify the quality of plasma through labeling, that
indicates the conditions under which the plasma was
prepared, including conditions of freezing.
We want to remove barriers to conversion
of plasma collected with the intention of its use in
transfusion, to its use in fractionation. Current
regulations reduce the flexibility to do this.
While relaxing some barriers, we need to
retain some distinctions, but only those that are
important. The distinctions that are being considered
include labeling that would distinguish plasma coming
from a whole blood collection, versus an apheresis
collection, product characterization based on intended
use at the time of collection, and conditions of
We also wish to have our regulatory
standards conform to the scientific state‑of‑the‑art.
Next. Now to review the agenda of the meeting.
On the first day of the workshop we have
a presentation about recommendations of the June,
2003, BPAC, that addressed recovered plasma standards,
and we also had an overview of current FDA
In brief, there was a lack of regulations
for recovered plasma, and there was a need to develop
specifications for allowable storage conditions and
We had a presentation from the consumer
community that emphasized the need for high quality
plasma products in the United States and
We also have a very extensive review of
the scientific literature that covered the effects of
freezing, of rate of freezing and storage temperatures
on the integrity of plasma proteins.
The purpose of this review was to help
provide us with a scientific rationale for regulations
that might be proposed. Next slide, please.
We then had presentations from the
international community on their standards and the
rationale, and their rationale for freezing, storage
and shipping conditions of plasma.
This included standards presented by the
Council of Europe, European Pharmacopoeia, Canada and
Australia. Representatives of plasma fractionation
and blood collection industries, reviewed their
current practices about freezing, storage and shipping
of plasma, and raised their concerns about the impact
of potential changes on their operations.
The panel discussion followed these
presentations, which further clarified regulatory and
industry positions. Next slide, please. Here are
some of the major points that came about from the
review of the scientific literature.
And I think these are very important. It
gives a frame work for at least the scientific basis
of some of our thinking. Loss of factor activity, as
reflected in lower product yield, may be regarded as
one measure of a reduction in plasma quality.
Loss of activity indicates that proteins
are being altered, potentially through aggregation,
proteolysis or conformational change. Now is a
surrogate marker for proteins that can be altered
during this shipping, freezing, storage process.
Factor 8 is currently regarded as the most labile
therapeutic plasma protein.
Conditions affecting Factor 8, may affect
other plasma proteins in unknown ways. Again the
notion that Factor 8, can be considered as a surrogate
marker, and that the yield of Factor 8 can be
considered as a measure of plasma quality.
I mention that delayed freezing decreases
Factor 8 activity in plasma. Preservation of labile
components in plasma is optimal up to six hours after
Factor 8 loses about 15 percent of its
activity when stored from 16 to 24 hours before the
plasma is frozen. An additional losses can occur if
it is stored for longer than 24 hours.
A very important point that was raised,
emphasized the number of times during the scientific
presentation is that the rate of freezing is very
Rapid freezing, such as freezing two minus
30 degrees in 30 minutes, gives a better Factor 8
yield than freezing it at minus 30 degrees over a much
longer period of time, say three to four hours, or
Storage within minus 20, to minus 40
degrees, appears to have little affect on product's
quality, as long as freezing, as long as the freezing
rate is optimized.
It's more important to maintain a steady
storage temperature in this range of minus 20 to minus
40 degrees, than an absolute temperature.
And finally it is uncertain whether the
time to freeze, way to freeze in storage or shipping
temperatures, affect product safety. And this is an
area that needs further investigation. Next slide,
The chart shows the current U.S. FDA
standards for plasma. One of our objectives was to
see about the chances of potentially harmonizing our
regulations with those of Europe.
I'll point out some of the areas that are
in contrast, that are now in contrast with the
European standards. First of all, our source plasma
is to be frozen immediately upon collection.
It is to be frozen at minus 20 degrees or
lower. Our regulations say nothing about the rate of
freezing. It can be stored at minus 20 degrees for
ten years, and it can be shipped at minus five
One fact that emerged from the workshop,
is that the current shipping of plasma, that plasma is
generally now shipped at minus 20 degrees or below.
And so this standard of minus five degrees
is not really what is the industry standard at
present. Fresh frozen plasma made from whole blood or
plasm apheresis, should be frozen within eight hours.
It can be frozen, stored and shipped at minus 18
degrees or lower, and stored for a year.
The freezing, storage and shipping
temperatures of recovered plasma are not defined.
Next slide. In contrast, the European Pharmacopoeia
makes a distinction between plasma use to make labile
proteins, such as Factor 8, versus the so‑called non‑
labile proteins, like immunoglobulins and albumin.
The time to freeze from collection to
freezing, to the time to freeze can be within 24 or 72
hours, depending on the product to be made. And
again, this is in contrast to our source plasma which
is supposed to be frozen immediately.
Plasma is to be frozen at minus 30 degrees
or below, at, if the product is to made, that is to be
made is a labile protein. Or at minus 20 degrees or
below for non‑labile proteins.
Storage and shipping conditions are at
minus 20 degrees or below. For plasma for
transfusion, the Council of Europe recommends freezing
to minus 30 degrees, within one hour, and storage
temperatures at minus 18 to minus 25 degrees, for a
three‑month dating period, and minus 25 degrees and,
below minus 25 degrees, if there is a 24‑month dating
So the idea of labile proteins freezing to
minus 30 degrees, the rapid rate of freezing are in
line with some of the scientific data that we heard
earlier on in the meeting, this idea of labile versus
non‑labile proteins is reflected in some of these
regulations and standards. Next slide, please.
The fractionation industry presented their
perspective on potential changes in the regulations
for freezing and storage and shipping of plasma.
These summarize a number of the points that were
raised by the industry.
Final products manufactured under current
storage and shipping requirements, are safe and
effective. Increased yield of plasma‑derived Factor
8 is not a driver for manufacturing. Yield is not a
Our current regulations that allow for
temperature excursions give flexibility to
manufacturers, changes in allowing for these
excursions would limit the availability of plasma for
use in manufacturing, and add to compliance
Changing freezing temperatures would be
costly and increase the cost of plasma. And resources
spent in changing freezing and storage temperatures,
could be better spent elsewhere. Next slide.
The blood collection industry also
presented their perspective on proposed changes.
There was a wish not to change the definition or
expiration date of source plasma.
Most plasma is used to make non‑labile
proteins. Factor 8 activity decreases the time to
freeze, but there's no change in its efficacy. There
is no reason why preservation of Factor 8 activities
should drive the standards, since it is a small part
of the market.
Manufacturers specify the requirements of
plasma according to procedures they have already
validated. FDA should focus its efforts on donor
safety, donor qualifications and good manufacturing
Labeling can indicate expiration date,
anticoagulant time to freeze, freezing and storage
temperatures. And finally, there's no compelling
reason to change requirements for freezing and
The next day, meeting, next slide, please.
The second day of the workshop, we had a review of
concepts of regulations, what regulations of the
And we had presentations by FDA, the blood
industry, the plasma industry, and this was followed
by a panel discussion. Next slide. This slide
summarizes some of the points made at the June, 2002,
BPAC meeting and FDA proposals for recovered plasma.
First of all, it was recommended that FDA
should develop standards for recovered plasma. FDA
proposed the term component plasma to replace the
terminology recovered plasma, because recovered plasma
has a negative connotation.
Component plasma would be defined as
plasma that is collected manually or by apheresis,
either separately or concurrently with other block
components from donors who meet all whole blood donor
Source plasma would be distinguished from
component plasma by defining source plasma as being
frozen immediately after collection.
Questions were raised at the 2002 BPAC
meeting, about having a ten year expiration date for
component plasma, and developing a time to freeze
standard for plasma used to manufacture labile
Again, reflecting the scientific evidence
that was available at the time. It was hoped at that
meeting that a workshop would provide data to address
the questions. Next.
This slide shows some other AABB proposed
standards for recovered plasma. These proposals were
derived in conjunction with America's blood centers,
the American Red Cross, ECA America, the Canadian
Blood Services, the Department of Defense, European
Blood Alliance and *(8:53:25).
PPTA, for the most part, endorsed these
recommendations, although they questioned a
recommended two‑year dating period for recovered
plasma. AABB proposed the name change for recovered
plasma to be plasma for manufacture.
The donor qualifications would be the same
as for allogeneic whole blood, including the
qualifications associated with infrequent plasma
Plasma for manufacture would be prepared
from plasma separated from whole blood, infrequent
plasma apheresis or by converting plasma for
transfusion to plasma for manufacture.
The expiration date is recommended to be
two years, and the label should state frozen within X
hours after phlebotomy and that the plasma should be
stored at minus 18 degrees and colder.
Next slide. There were some additional
comments, AABB proposed that freezing within a
certain, a specific time frame not be specified
because there are multiple types of products that can
become plasma for manufacture.
The fractionator can decide what plasma is
best, what is best for the manufacture of its product,
based on the labeled time to freeze. And short supply
agreements would not be necessary.
Regarding our future activity, last slide,
please. This workshop was only one opportunity to
collect information about standards for plasma. We
will continue information gathering through one‑on‑one
discussions with industry, particularly regarding
confidential or proprietary information.
And policy proposals will be developed
through a public dialogue process of notice and
comment. We are preparing a docket site together and
share comments about this workshop, and I anticipate
that that docket will be available in the very near
The web site for this conference, that
will give you access to the slides and transcript, and
notice of the docket opening, is at
www.fda.gov/cber/whatsnew.htm. Thank you.
DR. ALLEN: Thank you very much. Comments
or questions from the Committee with regard to the
workshop report? Just to clarify with regard to the
proposed name change, if I understand the process
correctly, you're going through a decision making
process, which, as you indicated on the last slide,
will be open ?
DR. WEINSTEIN: Correct.
DR. ALLEN: ? for public comments? Also,
you've not yet made a decision on that?
DR. WEINSTEIN: That's right. I will just,
for whatever it's worth, I will just make one simple
comment. And that is I tend to agree with the FDA
proposals, at least the component, the term component
plasma to me, seems to be more descriptive than plasma
for manufacture, which sounds as though it's primarily
being collected for manufacturing purposes. Other
comments or questions on this report?
MS. GREGORY: Kay Gregory from AABB. I
just want to explain why we did not particularly care
for component plasma.
In our way of thinking, we normally talk
about components as being things that we are preparing
for transfusion to patients. And we wanted to
distinguish this plasma, which is going to somebody
else to do something with, from the components that
we're working with and the terminology that we're used
to working with throughout our industry.
DR. ALLEN: That's a good rationale, thank
you. Okay. We will move on to our, thank you very
much, Dr. Weinstein, move on to our second update,
which is a discussion of the draft UDHQ, Uniform Donor
History Questionnaire Acceptance Guidance, review and
public comments by Judy Ciaraldi.
MS. CIARALDI: That's pretty good. Good
morning. Before each donation, blood and plasma
donors are asked questions concerning their medical
history and their high risk behavior.
This is because FDA has stated, in
regulations and in guidance documents, that donors
must me certain criteria and the donors are asked
these questions to determine if they are eligible to
Historically, the blood centers have been
responsible for developing their own questionnaires.
In the `50s, AABB, formerly known as the American
Association of Blood Banks, but now known as AABB,
developed their own uniform donor history
questionnaire that was used be most, or many, if not
most, blood collection centers.
And the number of infectious diseases
increased and other problems that are associated with
transfusion increased, so did the complexity of the
A task force was created from multi‑
organizations to review, evaluate, revise and
streamline the AABB questionnaire. The task force
submitted their questionnaire to us for your review.
We completed the review of the full length
materials, and published a draft guidance document
accepting it as a tool to collect donor information
consistent with our regulations and recommendations.
Today I'm going to discuss the comments to
the docket for the draft guidance document. Next
slide, please. The donor questionnaire process has
been discussed at several BPAC meetings, as you can
In the early `90s, the FDA commissioned a
report, a study by the American Institute of Research,
to look at the donor interview process, and their
results were presented at two meetings of the BPAC.
Later on we discussed validation of donor
questions and the task force got a chance to present
their materials at two BPACs. Afterwards we discussed
our review process and then the abbreviated
questionnaire and the self‑administered questionnaire
was presented and discussed.
We at FDA, really thank the BPAC for their
attention to this particular topic. Next slide,
please. In June of 2002, FDA did discuss its review
process of the task force materials. This is a
graphic representation of the review time line for the
Just to highlight a few points. In May of
2001, we received a full‑length questionnaire from the
task force that they asked us to review. This review
was conducted by six individuals within FDA, the
different offices in FDA.
It took us four months to complete this
review, and at the end of four months we submitted
comments back to the task force.
In March of 2002, we received the revised
full‑length and six additional documents to complete
a full questionnaire interview process. This
particular review was very complex, very broad. It
included eight FDA individuals and four of your BPAC
colleagues, for a total of 12 on the review team.
In spite of the complexity and broad
nature of this review, we were able to turn the review
around and provide comments to the task force within
After some exchanges back and forth, to
get extra clarifications and revisions to the
questionnaire, in July of 2003, we were able to inform
the task force that we had completed review on the
In addition, we were deep into the
development of the draft guidance document, accepting
it as a tool for screening donors.
We were preparing the draft guidance
document during the rest of 2003, and in the beginning
of 2004, when in March of 2004, the task force called
us and asked us, if necessary, to delay a little bit
the publication of the draft guidance document,
because they wanted to insert a new validated
question, into the questionnaire and they wanted to
make sure that we had the most current version
included in our draft guidance document.
They submitted those materials to us in
April of 2004, and we finished the review very quickly
and were able to say that we were now done. At the
same time, our draft guidance document was published.
The total review time for the full‑length
questionnaire, in FDA's hands was 13 and a half
months, in the task force's hands 14 and a half
months, independently of each other.
So this was a very big project by both
parties. Next slide, please. The draft guidance
document was published April 23rd, 2004, with a 90‑day
The draft guidance includes information
about the development of the task force materials and
our FDA acceptance of it. It also includes reporting
instructions for licensed blood establishment that
want to implement the new questionnaire.
The task force materials are included in
the guidance document as attachments. Next slide,
please. More specifically, the draft guidance
document states that FDA believes that the task force
materials will assist both licensed and unlicensed
blood collectors in complying with donor eligibility
It also states that licensed blood
establishments may report, in their annual report, if
they are going to implement the questionnaire without
modifications or with more restrictive modifications.
And we are also recommending the self‑
administration of this donor history questionnaire be
reported in the annual report. On the other hand, if
blood establishments wish to modify it, as otherwise
mentioned, they would have to send that in to us as a
prior approval supplement, so that we would have an
opportunity to review it.
Any new questionnaire that has undergone
major revisions by the blood establishment, have not
undergone this FDA review like the one that we are
We also stated in the guidance document
that blood establishments should report to us as a
change that's being affected in 30 days supplement, if
they would like to implement this process using a
computer‑assisted interactive procedure. Next slide,
There were 11 comments that were submitted
to the docket as of last week. Four came from
industry groups representing both the blood and the
One came from a task force themselves.
Three came from blood collection centers and blood
collection, blood suppliers. One came from a
One came from a computer‑assisted donor
history software vendor, and one came from a private
citizen. Next slide, please. We received some
positive comments to our particular draft guidance
These included their appreciation of FDA's
acceptance of the donor history questionnaire material
from the task force, including that they would be
allowed to self‑administer it.
The also appreciated the annual reporting
category, if they implemented without modifications.
There were no dissenting comments on the prior
approval category for major modifications.
One commentor asked if we could expedite
the CBE30 supplement review category for the
implementation of the computer‑assisted process.
Just to respond to this, all changes being
implemented within blood establishments, come with
some level of risk. And it is the responsibility of
the blood establishment to minimize this risk by
following good *(9:05:58) and process validation
before these procedures are implemented, regardless of
FDA approval. They also asked for clarification on
what we meant by without modification, and what was
required or recommended for using the accompanying
The things like the education materials,
medication list and so forth, that the task force
developed. More specifically, they wanted to know if
they must use a flow chart format that the task force
had prepared for the follow‑up questions.
We discussed this a little bit. We
haven't completed our full evaluation of the comments,
but we did discuss this, and we agree that some of
those materials that were prepared by the task force,
do contain formats that it is important for the blood
establishments to keep.
Specifically, the questionnaires
themselves. But some of the other documents, a blood
center may use a different format that is consistent
with their procedures.
Comments also asked us how to submit
comments or concerns that they may have to the
attachments. Now the DHQ materials belong to the task
force themselves. They are the property of the task
And they have changed control
responsibility over them. So comments about the
attachments or the materials themselves, should be
forwarded to the task force. Next slide, please.
The comments included, whether or not FDA
would discuss new questions with the task force before
we put them into draft or final guidance documents.
We would like to do this whenever our
policies allow. We have been discussing internally
about one possible way to develop new questions is to
conduct focus groups, whenever our resources and time
One comment asked us to change our donor
eligibility regulations to allow the position to
evaluate close contact with hepatitis and then the
Medical Director would determine deferral.
Right now the regulations do not allow for
this flexibility. Questions or comments like this, in
anything dealing with changing our regulation, is
beyond the scope of the draft guidance document
accepting the questionnaire.
A couple of comments asked us if we could
accept the abbreviated questionnaire in our guidance
document. At FDA's request, the task force is
continuing studies on the abbreviated questionnaire.
Once their revised product comes into FDA,
we will need to review it, and this process will delay
the publication of the questionnaire.
There were several concerns about a
comment in the task force material, a standard or a
need to complete the full donor history questionnaire,
but before determining eligibility. In other words,
if a donor answered a question early in the interview
process, that would defer them, why would they need to
complete the rest of the questionnaire.
That standard is not an FDA requirement or
recommendation, but it is included in the task force
materials. So this particular comment was forwarded
to the task force.
And all comments contained questions or
comments having to do with clarification of
information that was contained in the attachments
Because the attachments are the property
of the task force, all of these were forwarded to the
task force for their evaluation. And we don't
consider them relevant to the content of the draft
guidance itself. Next slide, please.
There were several concerns stated in the
comments to the docket. The donor history
questionnaire contains questions related to issues not
currently recommended or required by FDA. These
include a history of cancer, transplant graft and
questions about pregnancy.
FDA had stated in its draft guidance
document that it will allow these non‑required, non‑
recommended issues to be omitted from the donor
history questionnaire if the blood establishment so
This is because FDA does not have the
legal authority to require or recommend industry
standards where we've not come out in our own document
stating such. Next slide, please.
We also got some concerns that FDA did not
require or more strongly encourage the use of the task
force materials, and we also stated that we would
allow blood establishments that had previously
approved questionnaires, to use those even though they
were not tested and validated to the extent of the
task force materials.
Again, the FDA does not have legal
authority to require this particular standard and
require use of the task force material. Also, FDA
does not have the authority to rescind previous
approvals in the absence of data showing a potential
risk to the public health.
The task force is comprised of
participants from all the major blood establishments,
to ensure that it would be used widely.
And I think this is the hope of the task
force and that's the reason they composed or
constituted the task force with those members. Next
The process of preparing the final
guidance includes evaluating all the comments and
revising the document, if it is necessary. We also
are going to consult the task force about revision to
their materials based on the comments that came to the
We've informed the task force that we
should review these materials, because our guidance
document states that this is the version that we
reviewed and have looked at and agree with.
We have informed the task force, also,
that we feel this review is going to be much more
streamline and involve only the three liaisons to the
task force committee.
Lastly, we will prepare the guidance
document according to our regulations. The time to
complete this process will depend on the complexity of
the changes that are needed to be made to the draft
guidance document. Thank you very much for your
DR. ALLEN: Very nice summary, thank you.
Any questions or comments with regard to the donor
DR. ALLEN: Okay, I know that, at least my
perspective is that this is a very important step
forward and I look forward to it being completed. I
do have one quick question.
Has the task force or people working with
the task force, discussed updating of the history
questionnaire as new guidances come out. We discussed
*(9:13:02) virus yesterday. There was an update in
the last couple of years on, to try to detect symptoms
of West Nile Virus and so on, which I know will
probably come up again later this morning.
But as these new issues come up, is there
a way that the organizations that comprise the task
force propose to try to handle that and add another
uniform question to the questionnaire to keep it
MS. CIARALDI: The answer to that is yes.
They are, they have discussed it and they're still
discussing the most efficient way to do that. It is
the, and Kay Gregory is a member of the task force, so
she can finish up where I've left off.
But they have, they want to make sure that
the integrity of the questionnaire, that it's been
validated and all the questions on it have been
tested. They want to keep that integrity.
So, as new issues come up, they want to
have the opportunity to find a mechanism to quickly
test them. And then incorporate them into the
questionnaire so they are developing of that process.
I'm not sue it's been 100 percent
finalized, but they have been actively discussing it.
It's important to them as well.
DR. ALLEN: Do you want to make a comment
on that process?
MS. GREGORY: I think Judy summarized it
very well. And we're actually sort of testing the
process by testing the abbreviated questionnaire in
some additional ways, so we'll know whether the
process works very well or not, and we may need to
modify it if that's the case.
DR. ALLEN: Good, I'm glad the issue has
been addressed. Dr. Epstein.
DR. EPSTEIN: Let me just mention one
concept that has been discussed as a possible way
forward. Which is that as a new issue emerges, where
there appears to be a need to screen the donor for
medical or risk history, that we might provide
guidance to blood establishments to defer donors for
that risk, but not to frame a specific question.
We would then have a process whereby
questions were validated independent of that guidance,
and then only later integrated into the uniform donor
history questionnaire, as they were validated in their
own right, and in the context of the questionnaire.
So in essence, a two‑tiered process is,
you know, one concept that can be pursued.
DR. ALLEN: Thank you. Any other, yes?
DR. SCHREIBER: Does this uniform donor
history questionnaire also apply to the source plasma,
or is there another activity going along parallel, and
that's a naive question.
MS. CIARALDI: The questionnaire that is
currently in our guidance document, could be used by
source plasma, there's no restrictions on it.
But the source plasma industry has
determined that because of some of the differences in
donor eligibility criteria, that they have separated
into their own committee and they're working on their
They had submitted a first draft to us,
and we finished our review and have submitted those
comments back to them, and they are working on those
revisions that we've asked them to look into.
DR. SCHREIBER: Thank you.
DR. ALLEN: Okay, thank you very much. In
our third update for the morning, is FDA's current
thinking on monitoring weight in source plasma donors,
MS. ALMS: Good morning, I'm Linda Alms, a
Consumer Safety Officer in the Division of Blood.
Next slide. The issue that I'm going to speak briefly
about is the tracking of the ten pound weight loss
over a two month period of time in source plasma
Tracking of the ten pound weight losses in
donors over a two‑month period of time, is considered
a cumbersome process by industry, and it's an outdated
and ineffective procedure to reduce the risk of HIV in
plasma products. Next slide.
Tracking donors for ten pound weight
losses over a two month period of time, commenced
following CBER's revised memorandum dated December
As stated in the memorandum, the existing
cumulative records of each source plasma donor's
weight should be examined to assure that any weight
loss of ten pounds or more, in less than two months,
The December 14, 1984 guidance, was
superceded by a memorandum dated February 5th, 1990,
which also includes the statement requiring the
tracking of the weight loss for ten pounds or more
over a two‑month period of time.
A subsequent memorandum, dated April 23rd,
1992, addresses the additional possibility of HIV2
exposure, but no longer made mention of the ten pound
weight loss, tracking obligation of the source plasma
This memorandum does not specifically
state whether the February 5th, 1990 memorandum was to
be superceded. However, the current guide to
inspections of source plasma establishments, revised
April, 2001, still requires that the source plasma
donor's weight be examined to ensure that any weight
loss of ten pounds or more, in less than two months,
is detected. Next slide.
Since the early 1980s, improved testing
technology has reduced or eliminated the predicted
value of weight loss tracking with respect to
HIV/AIDS. Although, unexplained weight loss remains
a general indicator of possible ill health.
Source plasma donors are currently weighed
at each donation, in order to determine how much
plasma to obtain. These weights are recorded in the
plasma donor's records and they are available for
review as deemed appropriate by the center's medical
staff. Next slide.
Current requirements pertinent to source
plasma donor eligibility includes the following, 21
CFR 6040.63(a), states the suitability of a donor for
source plasma shall be determined by a qualified,
license physician or by persons under this supervision
and trained in determining donor suitability.
Such determination shall be made on the
day of collection from the donor by means of a medical
history, tests and such physical examination as
appears necessary to the qualified, licensed
And as stated in 21 CFR 640.63(b)1, each
donor shall be examined by a qualified, licensed
physician, on the day of the first donation or no more
than one week before the first donation, and at
subsequent intervals of no longer than one week.
Therefore, FDA's current thinking is that
it's appropriate for the active tracking of ten pound
weight loss among source plasma donors, to be
performed at the time of the annual physical, and that
other donor informational materials should be
harmonizes with those in places for the whole blood
donor eligibility. Thank you.
DR. ALLEN: Thank you. Comments or
questions on the, this presentation?
DR. ALLEN: All right, thank you very much.
I understand that we do have a request for an open
hearing statement from the Plasma Protein Therapeutics
Association, is that correct? Okay.
Please come to the microphone, I need to
read the public hearing announcement, so if you'll
bear with me for just a second, and then if you would
introduce yourself and make your statement.
Both the Food and Drug Administration and
the public believe in a transparent process for
information gathering and decision making, to ensure
such transparency at the open public hearing session
of the Advisory Committee meeting.
FDA believes that it is important to
understand the context of an individual presentation.
For this reason, FDA encourages you, the open public
hearing speaker, at the beginning of your written or
oral statement to advise the committee of any
financial relationship that you may have with any
company or any group that is likely to be impacted by
the topic of this meeting.
For example, the financial information may
include the companies or groups payment of your
travel, lodging or other expenses in connection with
your attendance at meeting.
Likewise, FDA encourages you at the
beginning of your statement to advise the committee if
you do not have any such financial relationship.
If you choose not to address this issue of
financial relationships, at the beginning of this
statement, they will not preclude you from speaking.
MR. PENROD: Thank you. Good morning, my
name is Josh Penrod, I'm a salaried employee of PPTA,
so that I hope that suffices as my disclosure.
The Plasma Protein Therapeutics
Association is the international trade association of
standard setting organizations for the world's major
producers of plasma derived an recombinant analog
Our members provide 60 percent of the
world's needs for source plasma and protein therapies.
These include clotting therapies for individuals with
bleeding disorders. Immunoglobulin is to treat a
complex, a complex of diseases in persons with immune
Therapy is for individuals who have alpha
one anti‑trypsin deficiency, which typically manifests
as an adult onset emphysema and substantially limits
life expectancy. And albumin, which is used in
emergency room settings to treat individuals with
shock, trauma, burns and other conditions.
PPTA members are committed to ensuring the
safety and the availability of these medically‑needed
PPTA welcomes the efforts made by the Food
and Drug Administration in reviewing the necessity to
monitor, at each plasma donation, records for the
donors weight measurements over a two‑month period of
time for the purposes of detecting an unexplained ten
pound weight loss.
The recommendation to monitor donor
weight, using measurements obtained to determine the
amount of plasma that can be donated by the donor, was
instituted prior to the development of tests able to
detect HIV infection.
We agree with FDA that such monitoring
today does not add a margin of safety with respect to
HIV/AIDS. For source plasma collection centers, the
repeated review of these weight loss records, over a
two month period, rather than adding to the protection
of public health, has instead become an onerous and
difficult task that frequently results in auditing
pitfalls rather than protecting the plasma donor or
the plasma supply.
PPTA agrees with the FDA assessment of the
utility of new and improved testing technology such as
NAT. We also agree with the FDA that unexplained
weight loss could be an indication of poor health,
that we would add that it could indicate a change in
physical activity, dietary habits, employment or
FDA has focused on the usage of the word
unexplained as being the operative turn in this
analysis. But this predisposes that any weight loss
has one cause, and it is either explained or not.
This binary approach may be suitable for
determinations of objective testing criteria and
standards, but it distal, surrogate marker, such as
the weight loss tracking, which never was truly
determinate of a disease state, is not subject to such
an interpretation, due to its inherent subjectivity.
We also agree, in large part, with FDA's
historical review of the blood memoranda issued over
the past 20 years, given today by Ms. Alms and its
briefing materials to the committee.
And the recommendation is contained
therein. He weight loss tracking criterion is
contained only in the current guide to inspections,
which is categorized as a level‑two guidance, and is
not subject to comment before implementation.
Our reading of these past memoranda, is
that while the April 23rd, 1992, memorandum, quote,
did specifically state whether, did not specifically
state whether the February 5th, 1990, memorandum was
to be superceded, close quote.
We would like to point out that the April
23, `92 memorandum, states that it replaces the
February 5th, 1990 memorandum.
Since the February 5th, 1990 memorandum is
replaced by the later memorandum, the earlier
memorandum should be considered to be superceded. We
also not that the 1984 and 1990 memorandum are not
generally available to the public on the FDA web site,
which indicates that they are, in fact, concerned by
the Agency to be obsolete.
PPTA appreciates the efforts of the Agency
in this regard. We also encourage the FDA to continue
review of the regulatory requirements and
recommendations that do not add to the safety profile
of product manufacture, plasma donation or public
While PPTA supports requirements and
recommendations that can add measurable improvements
to donor health and final product safety, outdated,
valueless requirements add burdens without benefit.
PPTA supports the FDA's review of
requirements that had become obsolete and FDA's
efforts to examine the regulations and the guidance
criteria to limit efficiency and do not generate
On behalf of PPTA and our member
companies, I thank the committee for hearing us this
morning, thank you.
DR. ALLEN: Thank you, any questions or
comments on the statement, Dr. Epstein.
DR. EPSTEIN: Well, Josh, you may be right
on a technicality, but the compliance program document
made it perfectly clear that it was still an FDA
policy to monitor the donor weight.
And I think FDA is concerned that if
source plasma establishments are in fact weighing the
donor then never to examine the weight records is not
appropriate. And we feel that we're providing
significant flexibility and reducing burdens by
recommending or proposing to recommend that this be
done only at the time of the annual physical, and as
a general, medical matter.
In other words, that's then within the
domain of medical discretion, how to deal with weight
trends. So, you know, I would just caution you that
because the `92 memo did not make specific mention,
didn't mean it was dropped.
Our intent in that memo was to supercede
the previous geographic referrals for HIV2,
recognizing that we now have testing for HIV2 and well
as HIV1. And perhaps there is an omission in not
capturing, you know, all previous recommendations.
But the compliance program makes clear
that we have not desisted from that recommendation.
MR. PENROD: We do appreciate the
flexibility we've been given, thank you. Although I
think we'd have to debate for another day, the role of
the compliance as policy making documentation.
DR. ALLEN: Dr. Goldsmith.
DR. GOLDSMITH: I was just concerned about
your third paragraph statement in which you refer to
weight loss as a subjective measure. Is there any
kind of a system for, and showing the accuracy of the
scales at the donor center. Is that why you refer
this as subjective?
MR. PENROD: Well, we think weight loss is
a measurement of weight loss, rather than of
necessarily being symptomatic of HIV. I'm not sure I
DR. GOLDSMITH: Well, you say that weight
loss is a subjective measure. Weight loss is an
objective measure if the balances have been checked
MR. PENROD: Well, weight loss certainly is
DR. GOLDSMITH: Right.
MR. PENROD: However, the extent to which
you are using it as a surrogate for another disease
state and its interpretation of the meaning of the
weight loss within that context is open to
DR. GOLDSMITH: But it is a general part of
medical practice to assess the health of individuals
by monitoring their weight over time. So I guess it
would seem to be appropriate to use it in this
context, even though it's not good for HIV, it might
be good for something else.
MR. PENROD: Well, we're not abandoning
weight loss or weight measurement. Thank you.
DR. ALLEN: All right, thank you. At this
point the public comment section is closed, this
session is closed. We will move on to our open
committee discussion, the third topic for BPAC for
this meeting, FDA's current thinking on donor deferral
for potential or documented infection with West Nile
As we will hear, you know, we are in our
second or coming to be close to the conclusion, I
hope, of our second season of screening with nucleic
acid testing for West Nile Virus.
We've learned an awful lot and we'll hear
the updates and recommendations for changes in
practice. Our first introduction and background will
be by Dr. Nakhasi from FDA.
DR. NAKHASI: Thank you, Dr. Allen. Good
morning. I sort of sound like a broken record. Every
BPAC I'm up here and presenting you the update of the
West Nile, but I think I hope next time we'll have
that, you know, we will see how it turns out to be.
Well, I that the topic of discussion is
today's, is the, we would like to see if we can have
our *(9:31:23) on the donor differential for potential
and documented infection of West Nile Virus. The next
The issue today is on the table is under
concentration, updating our current guidance on West
Nile, based on the recent reports that extended
*(9:31:41), which came out from our, that schedules
them under INDs to revise the current deferral period
which is in the current guidance physician and the
revised one on May of 2003, from 28 days to 56 days
for blood donors.
We want the positive screen by NAT or
reported symptoms of headache and fever. Also we
would like to, the question on the table is to revise
the guidance to have donors which are deferred with
either the positive test, screening test for West
Nile, or suggestive symptoms to be entered after
testing negative by ID‑NAT on a follow‑up blood sample
prior to re‑entry after 56 days.
Now, next slide, please. Just to, a quick
and brief background, but because Dr. Alan Williams
will give a detailed background about what the current
guidance talks about and how the questions have been
changed and that, you know, what we would like to
change and we'd like to make the changes and also the
question is on the table, which, you know, he will be
asking at the end. Just to re‑orient you about the
current recommended donor deferral criteria, they are
based on the donor deferral based on the reactive NAT
Currently, if a donor sample is tested
positive on individual donation, FDA recommends a
deferral of 28 days, which is based on the known
longest period at that time, which was known at that
time, which was the in 1950s, and so, you know, cancer
patients, and that was based on that, on 28 days at
This was before the testing was initiated.
And what is happening under this, currently under
clinical trial and IND donors are asked to enroll in
a follow‑up sample, those who have tested positive.
And then they are re‑entered based on
documented IgM conversion, seroconversion and
additionally a negative NAT result after 28 days is
required for donor re‑entry.
In some cases, you know, if you want to
re‑enter the donor earlier, before 28 days, it is
retested, the individual sample and donation, and if
it is negative it is re‑entered after 28 days.
Or, if it is positive, then the donor is
deferred again for 28 more days. Next slide, please.
The next criteria is based on donor deferral based on
the West Nile symptoms. This is basically on the
potential, again, based on the known knowledge at that
time having the extended period, you know, donor
period of 28 days.
The potential donors with medical
diagnosis of West Nile infection, including diagnoses
based on symptoms or laboratory results are deferred
for 28 days from the onset of illness or 14 days after
the conditions are resolved.
The other question is also asked regarding
the previous symptoms are included as part of the
current donor selection criteria. This was based on
the hypothesis ‑‑ not hypothesis. This was based on
the thing that during the ‑‑ some of the
transfusion‑transmitted cases which were negative on
NAT later on to show that they had symptoms reported
to be symptoms before or after the donations.
So in that question, what is happening is
donors are asked about the fever and headache in the
past one week and if yes, they are deferred for 28
days from the day of interview.
Next slide, please. So that's the current
guidance. Now, during the last year's study and
testing and this year some of the testing done, ARC
and BSL studied West Nile RNA dynamics in a number of
reactive blood donors from 2003 epidemic.
They followed. The follow‑up was to
determine the rate of disappearance of RNA as well as
the seroconversion of IgM and IgG. What they found
out, surprisingly, is that in rare cases, some of
these West Nile viremia may last up to 49 days and
that in those cases, RNA it coexist with both IgM
So that sort of raised our flags that the
virus can be found as long as 49 days, even though it
is very rare. But you will hear more about the mean
days of duration of viremia from both ARC presentation
and BSL presentation by Sue Stramer and Mike Busch.
Next slide, please. So the questions to
the committee are, do the available scientific data
support extending the currently recommended default
period of 28 days to 56 days: one, for blood donors,
the positive West Nile NAT screening test; and, two,
for blood donors who report symptoms of headache with
fever in the week before donation?
Next slide, please. The next question
would be, do the scientific data support a
recommendation to obtain a negative result by ID‑NAT
prior to reentry of blood donors who are different
either on the basis of reacting to NAT and/or on the
basis of symptoms?
Third is to the committee. Are there
other alternatives that should FDA consider regarding
criteria to reenter donors who are deferred for West
Nile based on that or symptoms? So those are the
questions which Dr. Alan Williams will present at the
end of the discussion.
Next slide, please. So quickly to update
you, but you will hear the more expanded, extended
update from CDC. Just to reorient you while you are
listening to those presentations, as of October 19,
2004, we have this year so far 2,151 cases and 68
Forty‑seven states are endemic for West
Nile virus, and there was one case reported, one case
of transfusion‑transmitted case, in Arizona. This
happened before the ID‑NAT was instituted in that
region because, as you remember, this year, as soon as
the native area became hot, that means that you found
more cases, you know, a lot more than four cases in
certain regions, the blood establishment changed from
Mini‑Pool NAT to ID‑NAT. So this case happened before
the ID‑NAT was instituted in that, just 12 days before
the ID‑NAT was instituted in that.
And, as we confirm with NAT, the IgM
reactivity donor recipient follow‑up, you will hear
more about this case from Dr. Theresa Smith's and Dr.
Jennifer Brown's presentations later on.
Next slide, please. So now how do we
stack up in the interdiction of the asymptomatic
donors since we started testing in the ID West Nile
NAT by Mini‑Pool NAT as well as ID‑NAT now this year
in certain areas?
Last year, 2003, in last year, 2003, 880
West Nile presumptory donors were reported to CDC
ArboNet. Underlining the CDC's ArboNet, there are
more than those cases, approximately 1,000 cases,
which found the blood establishments.
As of October 19, 2004, this year, we have
191 presumptive donors. And, you know, look at the
comparison between the two numbers, even though the
year is not over yet, again officially reported for
CDC ArboNet using both Mini‑Pool as well as ID‑NAT.
Then this testing, ID‑NAT testing, started in May '04.
Next slide, please. So what are we doing?
We are still continuing working closely with the test
kit manufacturers to see how we expedite the test
licensure. And we are still continuing to participate
in biweekly, this year biweekly at least, meetings of
the task force established by the blood community and
blood bank community, which includes CDC, NIH, and
coordinating and monitoring the infection throughout
Next slide, please. So today's agenda
will be as follow. First, the summary of the 2004
epidemic will be presented by Theresa Smith and
Jennifer Brown. And the duration of viremia and
experiences with the NAT testing, both Mini‑Pool and
ID‑NAT, which is going under IND, will be presented by
Mike Busch and Susan Stramer. And the current
thinking on the deferral extended and donor deferral
guidance will be talked about by Dr. Alan Williams.
And the questions will be again presented to you by
Thank you very much.
ACTING CHAIRMAN ALLEN: I am extremely
impressed. You wrapped up right at the zero second.
I have just one quick question. And I
suspect that this is information that will come out
later. But if you know it, you reported the number
for both 2003 and 2004, the number of presumptive
viremic blood donors. Do you have a rough estimate of
the proportion of presumptive positives that are
DR. NAKHASI: I think that Theresa and
Jennifer will talk about that.
ACTING CHAIRMAN ALLEN: Very good. Any
other questions or comment on this introduction before
we move to the full presentations?
ACTING CHAIRMAN ALLEN: Thank you.
As introduced, our next speaker
summarizing the 2004 epidemic is Dr. Theresa Smith
from CDC. Welcome.
DR. SMITH: Thank you. And I appreciate
the opportunity to talk to you about what we know so
far about the 2004 epidemic.
B. SUMMARY OF 2004 EPIDEMIC
DR. SMITH: Go ahead and go to the next
slide, please. I will quickly go over the virology of
West Nile virus, the epidemiology from 1999 to 2004,
some of which you have seen last year during this
update. We'll go on to the 2004 update and blood
donation surveillance events.
During these two portions of the talk, I
am going to be underlining the fact that the data that
you're getting is not the last word. We are still in
the midst of transmission. We are still in the midst
of gaining surveillance information.
Next slide, please. West Nile virus is a
flavivirus in the Japanese encephalitis sera group.
West Nile virus and St. Louis encephalitis are the two
members of this serogroup that are found in the United
These organisms are primarily bird
pathogens. And they amplify in avian host. That
means that an infected mosquito that causes an
infection in a bird has a great deal of change between
how infected material goes into the bird versus how
much infected material is available in that bird once
it has a full‑blown infection.
The common method of transmission amongst
nature is from birds to mosquitos to birds. Mammals
are a dead end host for this virus with only low‑level
viremia occurring within mammals before an illness
Next slide, please. I think that you are
fairly familiar with some of what has happened over
the last few years.
Next slide, please. But you might not be
familiar with where some of the data is coming from.
ArboNet is a national arbovirus surveillance system
that is a Web‑based passive system begun in 2000. It
includes 57 area health departments that report to the
Division of Vector‑Borne Infectious Diseases in Fort
Collins. They report mosquito, bird, horse, and other
animal surveillance data, including the year, state,
county, and date of collection of the specimens.
For human cases, state and county of
residence, clinical illness, and onset date, age, sex,
race, ethnicity, and risk factors for developing West
Nile virus infection are collected, including the
questions of blood donations and receipt.
The next few slides I think you're
familiar with and I will go through quickly. They
will show you the spread of West Nile from 1999
through 2004. One of the aspects I would like to
concentrate on is the difference between the map that
you saw at this time last year and the map that we
then created once we had all of the data in for this
If you would show the next two slides?
Next, please. Next. Next. Next. Here is what you
received last year about this time. Next slide,
please. And you can see that by the time we had
received all of the data for 2003, we had added two
new states. Idaho and Nevada now have activity in
this slide. It has become a fuller, more dense slide.
And areas that originally had only non‑human West Nile
virus activity now were showing human cases, which are
in red. Thank you.
Next slide. Here is our most recent as of
the time of the printing of these slides set of data
for 2004. As you can see, this is as of September
27th. And I would like to point out again that not
only is transmission still occurring, so, too, is
reporting quite a bit behind that as well.
Next slide. The 2004 surveillance update
I'm going to again take use of the numbers of last
year and compare them so you can have a basis to
understand this year's numbers.
Next slide, please. In 1999, there were
62 human cases of West Nile virus disease in the
United States; 2000, there were 21; 2001, 66; 2002,
4,156; 2003, 9,862.
I want you to note that in each of these
cases, these are the reports that we received with an
onset before December 31st of that year. That
contrasts with what data you will be receiving today.
Next slide, please. If we look at what we
had at the time of the printing of these slides, there
were 4,137 cases of human West Nile virus illness that
had been reported to CDC. And, again, thinking of the
previous slide, this is only 42 percent of what we
ended up understanding had occurred during that year.
At the time of your report last year, you
were told that there were 36 states and the District
of Columbia that were affected. West Nile meningitis
and encephalitis had had 1,153 cases reports. West
Nile fever had had 2,414 cases reported. There had
been 80 deaths, with a median age of 79 years.
Eight states last year had over 100
reported cases. Almost 90 percent of the reported
cases occurred in these states. That included
Colorado, South Dakota, Nebraska, Wyoming, Texas,
Montana, North Dakota, and New Mexico.
Now, if we contrast this to roughly the
same period this year, we had at that same period
1,784 cases. If we assume that this is, again, not
quite half of the cases for this year, it would appear
that we are not going to have quite as many cases this
season as last season.
However, we do already have 39 states and
the District of Columbia affected: meningitis and
encephalitis cases number 632, West Nile fever cases
number 721. There have been 56 deaths at the time of
this report, with a median age of 75.
At the time of this report, 3 states had
had over 100 reported cases, accounting only for
two‑thirds of all of the cases: California; Arizona;
and, once again, Colorado.
Next slide, please. Here you see the West
Nile virus human cases by week of onset, 2003 in pale
blue versus 2004 in burgundy, I guess. And you can
see that in 2004, we had an earlier rise in the number
of cases and that through the beginning of July.
There were actually more cases per week of onset than
there were in the previous year.
Next slide, please. For the blood
donation surveillance events, I am again going to go
ahead and show you some maps comparing what you
learned at this time last year to what the ultimate
reality of the 2003 season was.
Next slide, please. Here is what you were
shown last year with 495 donors reported as of
September 17th in 2003. You can see that they are
predominantly central. There is some crowding in
Next slide, please. By the end of the
year, it has become much more dense throughout the
Midwest. And you now have coast to coast events.
Next slide, please. Here, as the
information we had on presumptive viremic donors as of
October 4th, 2004, you can see that we are already
coast to coast but not particularly dense in the
number of cases that have occurred in any one area.
Next slide, please. Last year at this
time, there were 495 presumptive viremic donors
reported in 20 states. This turned out to be
approximately 60 percent of the ultimate total that
were reported to the CDC, which was 818. The top four
states for reporting presumptive viremic donors were
Colorado, Nebraska, South Dakota, and Kansas.
This year, at roughly the same period of
time, we had 157 presumptive viremic donors that had
occurred in 20 states again. The most common four
states for reporting were California, Arizona, Texas,
and New Mexico, in this case an entirely new set, as
opposed to the West Nile virus illness in general.
Next slide, please. How have we done in
terms of our ability to prevent transfusion‑associated
transmission? Well, we have decreased both our
numbers as well as the viremic load of the donations
that have been affected. In 2002, plasma from 16
implicated donations had virus titers ranging from 0.8
to 75.1 plaque‑forming units per milliliter, with a
median of 10.5 plaque‑forming units.
In 2003, plasma from four implicated
donations had virus titers ranging from 0.06 to 0.5
plaque‑forming units per milliliter, with a median of
0.11 plaque‑forming units per milliliter.
This year, at the time of this report, we
had one implicated donation with a viral titer of
approximately a .12 plaque‑forming units per
Next slide, please. I'm going to give you
a summary. And immediately afterward, I'm going to
give you more information through Dr. Jennifer Brown.
Overall what we have seen is that widespread West Nile
virus activity has covered almost all of the
continental United States, with New York, the original
site, still reporting human cases. There has been
continued westward expansion with human cases reported
from all states except Alaska, Hawaii, Maine, and
The concentration of presumptive viremic
donors has occurred in those areas that have the
highest concentration of infection rates in general.
We do continue to investigate possible
transfusion‑associated transmissions. And we have not
seen this year that our West Nile virus
transfusion‑associated transmission rate is at zero.
Next I would like Dr. Jennifer Brown to
give you the update as of earlier this week. Thank
DR. BROWN: Thank you.
So as Dr. Smith pointed out, we are
continually receiving new surveillance information.
And I put a few slides together just to update you on
what has been happening over the past couple of weeks.
These data are current as of October 19th,
which was Tuesday of this week. And as of that day,
there were only three states left that had not
reported any West Nile virus activity in 2004:
Alaska, Hawaii, and Washington State.
In the Northeast, we have seven states
that have reported West Nile virus activity in birds,
mosquitos, or in horses but have not reported any
human cases in 2004.
Next slide, please. So the current human
case count is 2,151. And those cases have been
reported from 40 states and the District of Columbia.
About 35 percent of these cases have been cases of
West Nile neuroinvasive disease and about 41 percent
have been cases of West Nile fever, but there's a
substantial number of cases that have not yet been
classified. So we will be looking for those case
classifications to be updated as we receive more
information from the health departments that are doing
Sixty‑eight of those cases have been fatal
so far. The median age of the decedents has been 74
years. And no one under the age of 43 has died as a
result of West Nile virus infection.
Next slide, please. So here is a map, to
give you a visual. You can see that we have had a
quiet year in the Northeast in terms of human cases,
but that does not mean that West Nile virus has been
absent from those areas. We have evidence of
transmission in birds and mosquitos in all of those
states that are colored in green.
The states that are colored in blue are
states that have reported human cases. And, as you
can see, Washington has reported neither ecologic
activity nor human cases, but with newly reported
ecologic activity and human infections in the State of
Oregon, it seems likely that either late this season
or next year, we will start seeing some West Nile
virus activity in Washington State.
Next slide, please. So this is the top
ten in terms of reporting of human cases in 2004.
And, as you know, California, Arizona, and Colorado
have reported the highest numbers of human cases.
They currently account for about 62 percent of that
2,151 cases that have been reported so far.
One of the things that I wanted to point
out to you as you look at this slide is that several
of the states shown here are states that have
experienced epidemic activity in past years but are
still continuing to report substantial numbers of
In particular, Louisiana and Illinois are
states that were foci of the epidemic in 2002. Each
of these states reported hundreds of cases in 2002 but
then continued to report substantial numbers of cases
in 2003 and 2004.
So, for me, this illustrates the need for
continued vigilance, even in areas that are not
currently experiencing epidemic levels of West Nile
Next slide, please. As of Tuesday, we had
191 presumptively viremic donors reported to CDC from
23 states. And, as Dr. Smith reported to you, the
highest numbers of donors had been reported from
California, Arizona, Texas, and New Mexico. Three of
those presumptively viremic donors had gone on to
develop West Nile neuroinvasive disease or meningitis,
encephalitis, myelitis, or other CNS pathology.
Forty‑five have gone on to develop symptoms of West
Next slide, please. This is the
presumptively viremic donor map updated as of Tuesday.
It's not much different from the one Dr. Smith showed
to you. The one thing that has been added is that a
green triangle marks the county of residence of the
transfusion‑associated transmission case that was
reported in the September 17th MMWR.
Next slide, please. I do have a little
bit more information to report to you. We have
learned of a second probable case of
transfusion‑associated transmission. That is still
under investigation by the State of Michigan.
The donor was an Illinois resident who
donated blood in Iowa and subsequently became ill.
The donation was nonreactive by Mini‑Pool, reactive by
individual donation testing. The donor has
The platelet recipient is a Michigan
resident and does reside in an area where there is
West Nile virus transmission. And the recipient has
not developed symptoms of West Nile virus infection
but has seroconverted.
Next slide, please. The question that
everyone is asking us at CDC is, what is going to
happen in 2005? There are only a few things that we
can say with any degree of certainty.
Next slide. First, human cases will
continue to occur in areas where West Nile virus has
already been identified.
Next slide. Second, the geographic range
of West Nile virus will continue to expand through the
movement of infected birds.
Third, epidemics will occur in areas where
conditions are favorable. But, unfortunately, we
can't tell you right now in the Fall of 2004 where
areas of epidemic activity will be in 2005. And
that's why on the next slide we see that surveillance
is critical for early identification of epidemics.
That's why it's so important for us to look for West
Nile virus activity in birds, mosquitos, horses, and
blood donors, and to look for human cases as well
because that's the way that we learn where epidemics
are developing. And hopefully we can learn about them
in time to implement public health interventions.
Next. And, finally, I'd like to conclude
by showing you the faces of some of the people that
are responsible for the collection and analysis of
ArboNet data. Some of them are shown here, and some
are shown on the next slide with the ArboNet team.
Dr. Smith and myself are both available to
field your questions if there are any.
ACTING CHAIRMAN ALLEN: Thank you both.
Yes, Dr. Lew?
MEMBER LEW: Since we know reporting is
what I consider the tip of the iceberg, what do
serologic studies show in terms of how many people
will actually be infected every year? And when do you
think you will reach a point where the majority will
DR. BROWN: Well, we know from past years'
serosurveys that have been conducted in areas of
epidemic transmission in the Northeast, in New York
City, and Connecticut; in Louisiana, where an epidemic
took place in 2002; in Rumania, where a West Nile
virus epidemic occurred in 1996.
Population‑based serosurveys conducted
after West Nile epidemics in those areas showed that
overall at a population level, the seroprevalence of
infection was no more than two to three percent. And
so it's unlikely that at this point, even in areas
that have previously experienced West Nile virus
epidemics, that we have reached a level where
background immunity in the population would be
adequate to protect against future epidemics or future
ACTING CHAIRMAN ALLEN: It does seem that
we've got a slightly different pattern in the United
States than we have ever been aware of in any other
country. New York now is in its sixth year of
reported cases, even though it was a fairly small
number of human cases this year.
So we may find that if you consider the
United States as a whole, we may become an endemic
country for continued West Nile virus activity.
DR. BROWN: Oh, certainly. One of the
things that we can say with certainly is we will
continue to see cases of West Nile virus. What
remains to be seen, since the virus is so new, we are
still learning about its ecologic behavior.
And so what we don't know yet is whether
it will fall back to a level of endemicity where we
will only see sporadic cases, as we do with St. Louis
encephalitis, punctuated by irregular and
unpredictable outbreaks, or whether we will continue
to see what we have seen so far, which is sporadic
cases in some states, modest levels of activities in
others, and epidemic levels of activity in still
others. We will just have to keep watching to see
ACTING CHAIRMAN ALLEN: One other question
just for clarification. Of the total reported human
cases, that includes the asymptomatic virus‑positive
people if you become aware of them as well as those
with West Nile fever and West Nile
DR. BROWN: No. That's a very good
question. ArboNet ‑‑ when we discuss reported cases,
we are referring to the case definition for West Nile
virus disease that has been developed by the Council
of State and Territorial Epidemiologists. And that
case definition refers only to symptomatic cases.
We track presumptively viremic donors
separately. So the mechanism for tracking donors
allows us to track people who are asymptomatic, but
when I reported those 1,251 cases, those are only
cases that meet the national case definition for West
Nile virus illness. So an asymptomatic donor would
not be included in that count.
The donors that did, those 48 donors that
did, go on to develop neuroinvasive disease or West
Nile fever, they are included in that overall case
count. So that's why we present the case count
separately from the donor count.
ACTING CHAIRMAN ALLEN: Okay. There was
still a, however, category. If you add up the
meningoencephalitis and the West Nile fever, that
still doesn't total 100 percent, however. Are those
just not classified yet?
DR. BROWN: Right. Those are not all
asymptomatic donors. Those are cases that have not ‑‑
their clinical syndrome has not yet been classified.
And they're still under investigation by the state
health departments that are tracking them.
ACTING CHAIRMAN ALLEN: Thank you.
MEMBER DOPPELT: I just had a question to
follow up to that. On one of those slides, I think
you said it was 35 percent had neuroinvasive disease.
So depending upon how you're counting, what's the n,
the number infected? So I assume that that means that
the total percentage of neuroinfected is not really
different this year than last year or not?
DR. BROWN: That is hard to say.
Thirty‑five percent of the cases that have been
reported to us have been classified as neuroinvasive
illness. Because so many of them have not yet been
classified, it's difficult to say. That's kind of a
It's difficult to say what the final ‑‑
what proportion of neuroinvasive disease cases, how
much they will contribute towards the total number of
cases reported. And, as you have pointed out, the
proportion of neuroinvasive disease cases as a
proportion of the total number of cases reported is
not the same as the proportion of neuroinvasive
disease cases as a whole of the entirety of people who
We think that about one in 150 West Nile
virus infections will result in neuroinvasive disease.
So it's not that 35 percent of everyone who is
infected with West Nile virus gets neuroinvasive
disease. The actual number is quite smaller.
ACTING CHAIRMAN ALLEN: Dr. Lew?
MEMBER LEW: Have you had a chance to see
of the people who fit in the definition ‑‑ in other
words, how good is your definition for West Nile for
reporting when you have ability to test that they
actually are positive? I mean, has it been validated
some, the definition that you have?
Just like initially with the HIV epidemic,
there was criteria to make the diagnosis. But then
later we have testing.
DR. BROWN: Yes. The case definition that
we use ha two components. One is the clinical
component, and one is the laboratory component. And
so in order to meet the case definition, a case must
first meet the clinical criteria for diagnosis.
But then they must also have one of the
laboratory criteria for diagnosis. And these
laboratory criteria we are very comfortable have a
very high positive predictive value for being cases of
West Nile virus illness.
ACTING CHAIRMAN ALLEN: Dr. Williams?
DR. WILLIAMS: Alan Williams, FDA.
Pertinent to the questions being posed to the
Committee today, of the two presumptive transfusion
cases under investigation, the first my understanding
is the donor did not report having any symptoms prior
to the donation. Do you know what the situation is
with respect to the second donor under investigation?
DR. BROWN: I only have very limited
information about that case, but it is my
understanding ‑‑ and I'll ask Dr. Smith to jump in if
she knows more, but it's my understanding that this
was a case where the donor became ill following
donation and the investigation resulted as a result of
the donor notifying authorities.
ACTING CHAIRMAN ALLEN: Dr. Nakhasi?
DR. NAKHASI: Hira Nakhasi, FDA. Dr.
Allen, I just wanted to have clarification of what
Jennifer Brown said. You know, you were asking, do
you think in Europe or other countries, why the U.S.
now is sort of developed and Peter has it.
There was a paper last year in Science
where they described the differences between the
mosquito population here in the United States and
Europe is different. So that's why the difference
possibly could be, that they have much more endemic,
they have become better or worse of that. And you
have the better ‑‑ you know, you have epidemic
currently going on.
And because other of the differences are
hardly because the European population and the U.S.
population more or less are the same basically.
DR. BROWN: That is a very good point in
that the differences in the mosquito populations could
be one factor that influences the behavior of West
Nile virus in the United States. That may be one
thing that makes West Nile virus different in the U.S.
than in Europe.
DR. NAKHASI: Yes.
ACTING CHAIRMAN ALLEN: Okay. Dr.
DR. KLEINMAN: Yes. Steve Kleinman. I
have one comment and one question. The comment is
more for the Committee, just to be clear that the
number of positive donors reported to CDC through
ArboNet or whatever, AlterNet, whatever it's actually
called, are actually fewer than the number of West
Nile virus donors that will come up in the next
several presentations because not every state gets the
report and reports it on to CDC.
So that's just a comment, although I think
it is interesting that the relative proportion of
cases dropped significantly in 2004, both in CDC's
data and in the blood center data.
My question is a more general one. Have
you seen or have you been able to assess the effect of
mosquito‑spraying programs on the progress of West
Nile? I know it's a county by county or state by
state decision, but what is sort of the general
climate of whether effective places spray for
mosquitos or not?
DR. BROWN: At CDC, we do feel that
mosquito control is an important component to West
Nile virus case prevention, but it is very difficult
to do a scientific assessment or to quantify the
degree to which cases can be prevented by spraying.
That is because mosquito abatement districts tend to
vary by community.
And in order to answer that question, you
would have to find two mosquito abatement districts
with different vector control programs, but those two
communities would have to be similar in every other
way. It is extremely difficult to find that set of
circumstances where you could answer the question of
whether it was only the mosquito control that was
making the difference in cases.
So we are looking, our entomology group is
looking, at ways to answer that question, but it is
very difficult. That being said, we do feel that
vector control is a very, very important part of case
prevention, especially in epidemic areas.
DR. KLEINMAN: Yes. And do you have a
sense on at the community level how frequently
communities are actually doing this versus not
spraying or is that just so individual that it is hard
DR. BROWN: That is another thing that
tends to vary a lot by community. In Maricopa County,
for example, in some residential areas, there was a
high degree of resistance and some political
resistance as well to doing aerial application of
insecticide, where in some more rural areas, it's no
problem at all.
So that's another thing that varies from
community to community. And that's another reason why
it makes it so difficult to do scientific studies to
try to quantify the degree to which this is effective.
ACTING CHAIRMAN ALLEN: We are getting a
little afield here in terms of spraying. And I
realize the relationship. I personally would love to
continue the discussion.
We have got a schedule to adhere to. We
will take questions from two other people at the
microphone and any others from the Committee directly.
DR. BUSCH: Yes. Mike Busch from Blood
Of the two cases breakthroughs, probably
breakthroughs, issues, one of them, as you indicated,
is reported to MMWR. It was a Blood Systems case
where we had our system to turn on individual donation
NAT, but it basically was not completely ready to
operate in early June. The epidemic started earlier.
Had that system been in place, we're confident that
that donation would have been screened by ID‑NAT and
The second case you mentioned, you
indicated that it was ID‑NAT‑reactive. Was that
ID‑NAT performed by the test of record at the blood
center? And also I think, to my knowledge, all of the
transmissions from prior years and this year have been
IgM‑negative. Was that additional case tested for
DR. BROWN: I do not have the personal
familiarity with that case to be able to comment, but
perhaps Dr. Smith.
DR. SMITH: Hi there. We are in the midst
of getting this one settled. So I'm afraid that we
haven't shared all of our information. We tried to
give you enough to let you know that this has
occurred. So I apologize that I haven't given Jen all
of the information she could share with you.
This case came through during a time when
the blood bank was doing Mini‑Pool testing. There had
actually been no positive Mini‑Pools. So there was no
trigger that could have been sent off to switch to
ID‑NET. And in retrospective testing of the plasma,
it was IgG‑negative.
DR. FITZPATRICK: Mike Fitzpatrick from
America's Blood Centers. Just one question.
You stressed the importance of
surveillance on prediction and looking at what has
happened with the epidemic. A number of states and
counties have stopped surveillance of birds, and I
just wondered what the impact of that is on your data
and what the future holds for those areas that are no
longer doing that surveillance.
DR. SMITH: Many places have chosen to
stop surveillance for birds this season and will
reinstitute that in the spring. Once you have a
positive bird, it doesn't gain you more information to
have more positive birds in any one particular county.
I don't know of anybody that has said that
they will not be accepting for a new season reports of
dead birds that they would want to check.
ACTING CHAIRMAN ALLEN: Dr. Lew?
MEMBER LEW: Just as a follow‑up to what
Dr. Williams had mentioned. And I can stand for
clarification, but my understanding is about one in
150, as you mentioned, or one percent or less has
encephalitis, 20 percent with West Nile fever‑like,
but the vast majority of people with West Nile
infection are asymptomatic. So that is going to be a
DR. SMITH: Also, for the clarification of
the numbers, currently this is not a disease that is
required to be reported. So we're not going to get
100 percent of the neuron base of numbers or 100
percent of the West Nile virus fever numbers, which is
also going to make the percentages then different. In
the coming year, meningitis and encephalitis will be
ACTING CHAIRMAN ALLEN: Thank you, Dr.
Smith and Dr. Brown, for a very nice update. I hope
both of you will be available later in the day if
people want to engage you in discussions or we could
go on for hours.
Our next presentation we're going to get
back more directly to blood collection center
experiences, duration of viremia, and experience with
individual NAT testing, Dr. Michael Busch from Blood
DR. BUSCH: Thank you.
C. DURATION OF VIREMIA/EXPERIENCE WITH ID‑NAT
DR. BUSCH: This is a project that
obviously involved lots of collaborators to
characterize both the index donation and the serial
follow‑up samples as well as some other studies
correlating viremia with the total infection rates in
the population. So, again, the collaboration by
several companies as well as Blood Systems. And this
was supported by NHLBI and CDC and Blood System
Next slide. Actually, the insights into
the natural history of West Nile virus I think are
able to be significantly enhanced and expanded with
the implementation of donor screening because, really,
for the first time with donor screening, we're
detecting humans within the acute viremic phase of
infection and are able to then follow them to
understand better the evolution of viral immune
markers and pathogenesis questions.
So, really, we're very interested in
further studying these issues, both with respect to
the donor screening and deferral policies we're
talking about today, but also I think we're generating
data that has insights into the diagnosis of the
infection in clinical populations and also the
So I am going to summarize for you four
studies that we have been doing relevant to the
question of viral dynamics. The first is just
analysis of the index donations, the yield donations
themselves, then a study where we have correlated the
yield of Mini‑Pool NAT with the cumulative incidence
of West Nile virus in a particular state, an epidemic
Of relevance to this discussion, this
analysis has allowed us to estimate the duration of
the window period that Mini‑Pool NAT detects. That,
in turn, actually allows one to use that understanding
of that window period to estimate total infection
rates in the population.
The next analysis is a study that Blood
Systems did where we did a large amount of individual
donation NAT testing of samples that had been
Mini‑Pool‑negative from 2003. By analysis of that
data, we have been able to estimate the lengths of the
window period that is detectable by individual
donation that prior to Mini‑Pool‑detectable levels of
viremia as well as the subsequent windows that are
detectable by ID‑NAT with antibody, either IgM or IgG.
And then, finally, an analysis of the sero
follow‑up data from about 180 viremic donors in a
determination of the lengths of the window periods to
both seroconversion and to persistent detectable NAT
reactivity by replicate individual donation NAT.
Next slide. So in terms of the index
donations, all of the data I will be presenting is
based on Blood Systems laboratory screening using the
GenProbe platform in 16‑unit Mini‑Pools.
The viremia levels were determined with a
target capture real time PCR assay developed at
Chiron. And the serology is based on focus technology
Next slide. So at Blood Systems, we
screened ‑‑ this is all data from 2003 ‑‑ 680,000
donations, 230 confirmed viremics. Of those, you can
see about 80 percent of them were detected by
Mini‑Pool NAT and 18 percent were detected either
through the retrospective or prospective ID‑NAT
If you look at the index donations in
terms of their antibody status, overall 20 percent of
the viremic donations that we picked up had antibody
in them but a very different rate of antibody
depending on whether the units were detected by
The Mini‑Pool NAT screened units, only
eight percent had IgM‑detectable; whereas, the samples
that were ID‑only that were missed by Mini‑Pool but
detectable by individual donation NAT, the vast
majority, 75 percent, had IgM antibody, indicating
that most of those were in the post‑acute viremic
phase as IgM was developing.
Next slide. This is just a conceptual
window phase evolution of the primary viremia. I
don't know if anybody has a pointer. No. So, in any
event, the overall viral load of the Mini‑Pool yield
donations, which we're calling stage 3 here, the
samples that are detectable by Mini‑Pool NAT, are
about 2,300 copies median, mean of 37,000 copies. And
you can see that there are some lines drawn that
represent the limit of detection of Mini‑Pool NAT,
which is about 80 copies per mL; whereas, if you test
the samples individually, the viral load can be as low
as 5 copies per mL and be detectable.
Next slide. This shows the distribution
of the units that were detected by Mini‑Pool NAT,
either with IgM, on the left, or without IgM, on the
right. So what you can see is that the samples again,
all detectable by Mini‑Pool NAT, that had IgM had a
very low viral load. The median was 198 copies per
mL; whereas, the samples that lacked IgM had a much
higher viral load. So these are the tail end.
In fact, if we go back one slide, please,
you can sort of see that what we're picking up with
Mini‑Pool NAT, the vast majority of them are prior to
IgM seroconversion, but we pick up a small portion of
units that are on the down slope of viremia, so have
low viral load in the presence of IgM.
Next slide. Next slide, please. So the
next analysis I'm going to summarize is the
correlation of Mini‑Pool yield to infection rate in
the population. This is a study that was done through
the Reds Program, where all of the donations from
North Dakota that had available aliquots, which had
been screened by Mini‑Pool NAT and actually also
screened by individual donation NAT, we went back and
we performed IgM testing on all of the available
samples. And then we analyzed the data, the weekly
data, of Mini‑Pool NAT yield over time. We had 28
Mini‑Pool NAT yield units.
Next slide. And then again we performed
IgM testing on about 4,000 samples to understand over
the course of the epidemic how was the IgM conversion
evolving. We also went back about nine months later
and sampled another 1,000 donations from this same
region. And we performed both IgM and IgG testing on
those collections about six months out from the
epidemic, from the exact same donor pool.
Next slide. This slide is a summary of
the results of that testing. Again, from North
Dakota, we are starting over here in July and running
through October. And then we come back in May of '04
and have additional data six months later.
You can see the goal here is the Mini‑Pool
yield occurring. So we begin with some early yield of
viremic donations. About two or three weeks later, we
begin to see IgM conversions accruing in the
And then after Mini‑Pool yield has
completely disappeared, the IgM rates in the
population begin to plateau. And they end up peaking
at 5.2 percent. So this is the infection rate in the
donor population that corresponds to this yield of
Mini‑Pool NAT that was observed in that same donor
When we went back six months later, the
IgG rates were 5.3 percent, so essentially identical
to the early IgM rates. But by that point, the IgM in
the donor pool had waned to 1.1 percent. So IgM, a
transient marker, had begun to disappear. These
numbers here give us the total infection rate in the
Next slide. A couple of observations from
this. I think that, as you can see, IgM really comes
up a little bit later than that, so wouldn't be an
early marker, a good screening tool for the early
viremic phase. It peaks three to four weeks after
detection with peak IgM rates about four times the
Mini‑Pool NAT yield peak rates.
If we are concerned about a tail end
low‑level viremia, actually, IgM could be argued to
have some value because it is picking up all of the
convalescent infections. And, as you will see later,
ID‑NAT itself even performed individual donations but
only once, is not able to detect all of the low‑level
viremic subjects. So some people are coming in at the
tail end of the epidemic who would be detectable by
IgM but have viral loads that might be so low that
even individual donation NAT wouldn't pick them up.
Next slide. Late in the epidemic, if we
were to screen for IgM, we'd lose a lot of donors.
Over five percent of donors in a high‑activity region
would be seroreactive. And, yet, the risk of these
units is extremely small, if any.
After six months, IgM rates have dropped
to 20 percent of their peak. However, what that tells
us is that IgM screening in a subsequent year would
still have some tail of prior year seroreactives
coming into the next year epidemic. So it tells us
that it's very difficult to use IgM or serology in
general to estimate risk once you have seen an
epidemic in a prior year.
And then, finally, even in a highly
endemic region, as discussed earlier, the vast
majority of donors were never infected. So there's
still clearly going to be a susceptible population
and, therefore, a potential need for continued
Next slide. From an analysis of that
relationship between the Mini‑Pool yield data and the
total infection rate in the donor pool, David Wright
at Westat was able to derive the length of the
Mini‑Pool window period. And that's 6.9 days with a
confidence bound shown here.
This is a very important parameter to
understand because it, in turn, allows us to benchmark
off the length of the Mini‑Pool window period to
estimate the lengths of other window periods.
It also lets us use that window period and
national Mini‑Pool yield data to estimate total
infection rates in the population. In fact, we have
estimated that during this year by compiling the
national data for Red Cross and CDC, that something in
the range of 750,000 people were infected with West
Nile virus in '03 based on the Mini‑Pool yield data in
the country and the length of this window period.
Next slide. I'm just going to skip this.
This is the statistical modeling that was needed to
derive that window period estimate from that
relationship between the Mini‑Pool yield over time and
the seroconversion rate.
Next slide. Okay. The next study I want
to summarize ‑‑ and this will all tie together at the
end ‑‑ is our large‑scale retrospective testing study.
So this was work that was done with strong
encouragement from FDA in 2003.
When we realized the epidemic was so
massive and there were some breakthrough
transmissions, we began to save samples from
high‑yield regions that had been negative by Mini‑Pool
NAT but, again, from regions that had high yield. And
these samples were tested by individual donation NAT.
If they were reactive, we immediately
tried to retrieve any untransfused product, confirmed
that the donors were infected based on both analysis
of the index sample and follow‑up of the donor, and
then collaborated with CDC in terms of investigation
of recipients, who were transfused with units that
came from donors who were ID‑NAT‑only reactive and
Next slide. So overall we tested 23,000
donations that had been Mini‑Pool NAT‑negative by
individual donation NAT. And in the three areas which
had high activity, we picked up 30 viremic units.
Toward the end of the year, we also turned on
prospective ID‑NAT in the Dakotas.
Once we started to see the data showing
significant low‑level viremia, an additional 4,000
donations were tested prospectively, yielding an
additional 17 viremic donations.
Fourteen of these 17 were ID only, meaning
that they were negative when retested at one to 16
dilution. So three of these would have been picked up
by Mini‑Pool. So we have 14 plus the 30. So we had
an overall 44 additional infected donations that year
detected by ID‑NAT.
Next slide. And this is data, then,
showing the evolution of the detection over the course
of the two‑week intervals over the course of the
epidemic, specifically focused on North and South
Dakota, where every donation was tested, essentially
every donation was tested, by both Mini‑Pool and
What you're seeing here are stages of the
infection. So the blue bar is units that were
detectable by Mini‑Pool NAT. This light blue bar here
is the units that are detectable only by individual
donation NAT but have no antibody; and then the units
that have IgM only, low‑level viremics; and then those
that had IgM and IgG.
You can see that at the beginning of the
epidemic, you have these low‑level viremics without
antibody, the ones that we know can transmit that are
seen actually kind of throughout at some low rate.
But what is striking is you get this high
Mini‑Pool yield, but then as the epidemic is moving
along, you begin to see large proportions of the
viremic donations are ID‑only units in the presence of
antibodies. So these are these convalescence
infections that still have very low‑level viremia in
the presence of antibodies.
Next slide. By analysis of the number of
cases in each of these stages ‑‑ so in this Dakota
region, we had 79 Mini‑Pool yield units. These are
the number of front‑end antibody‑negative ID only with
IgM only and ID only with IgM and IgG.
And using the 6.9 days that we derived
earlier, we can estimate the lengths of these other
window periods based on the relative frequency in this
sort of serial cross‑sectional analysis that we picked
up units in these stages. You can see that these are
fairly brief periods.
Overall in this fairly comprehensive
analysis, only 66 percent of viremic units were
detected by Mini‑Pool NAT, but the majority of those
that weren't detected were antibody‑reactive and we
believe probably had neutralized the virus.
Next slide. So this just takes that 6.9
days that we had derived earlier from the cumulative
NAT infection rate IgM data and uses that to estimate
the lengths of these earlier window periods, the .55,
.65, and 2.29 days.
Next slide. Okay. The next analysis is
the follow‑up of the donors, the last analysis. And
what we're looking at here is enrolling the donors per
the IND into the follow‑up study. It included a
symptom questionnaire, which you will hear about later
today, and approximately weekly sampling.
The follow‑up was to continue until the
donors had converted their IgM and tested negative by
single ID‑NAT. The follow‑up included RNA by TMA
quantitation and IgM and IgG. And then a subset of
over 60 of the panels were further tested to better
understand the low‑level persistent viremia by
performing five additional replicate TMA assays,
individual donation. And a number of these panels
were also studied for additional antibodies, including
plaque neutralization, by CDC, Rob Lanciotti.
Next slide. Overall 182 of our about 230
donors enrolled in the follow‑up study. You can see
that the follow‑up averaged about 15 days to the first
sample, but a number of the donors did come in fairly
early on to let us look at early events, an average of
2 and a half specimens per donor.
Just one factual point, which is that at
index donation, there were 140 of these 182 who were
negative for IgM on the index donation. On the first
follow‑up lead, 81 percent of them had converted their
IgM. In a second follow‑up lead, the remainder had
converted their IgM. So 100 percent of the people who
enrolled into follow‑up converted their IgM on
Next slide. Just one example. What we're
looking at here is the viral load of the index
donation dropping to negativity on quantitation, the
antibodies kicking up the IgM, the IgG. And here is
plaque‑neutralizing activity. In every case,
plaque‑neutralizing activity was observed concurrent
with the development of IgM antibody. So the antibody
is effective at neutralizing virus in an ex vivo
mixing type analysis.
What you see down here are the percentage
of the six replicate TMAs that were performed on all
of the serial bleeds. And you can see that the
viremia is detectable out to here. And then as you
out in time, only a small proportion of the six reps
may be reactive.
We had examples in our data by the singlet
follow‑up TMA of people who were negative and then
came back for another bleed and were positive. And so
we were seeing flip‑flops that were of concern.
By doing the six replicate TMAs, we no
longer had any of them. We could basically show that
what was really going on was just a waning viremia and
that it was the probability of detecting that
low‑level viremia that led to an occasional negative
followed by a positive. But by doing the multiple
reps, it was all a smooth transition down in viremia.
Next slide. Just another example.
Next slide. So the analysis of that data
was done by David Wright using what's called
interval‑censored longitudinal analysis modeling. And
what we looked at was the time from the index donation
to IgM and IgG seroconversion as well as the time to
loss of RNA from the index donation by a singlet TMA
assay and also the times between these different
seroconversion and RNA loss events.
And then for the subset of 56 cases that
we did the 5 replicate TMA assays on 580 follow‑up
samples, we were also able to look at time from index
to loss of RNA by 6 replicate TMAs, so a more
sensitive quantitation of detection of viremia.
Next slide. This just shows these window
periods. So this is these people are being detected
at some point in the Mini‑Pool NAT yield window phase.
We're assuming on average they're being detected in
the middle of that period. And then this is the time,
3.4 days to IgM, 7.6 days to IgG, 11 days to loss of
RNA by singlet ID‑NAT, but an additional 6 days if we
do the 6 replicate ID‑NAT assays. So these are the
Next slide just summarizes the statistics
around these estimates. I don't have time to go
through these, but you have them in your handout. And
you see confidence bounds. These confidence bounds
are confidence bounds around the mean. So this is how
accurate is this average time from Mini‑Pool NAT
positivity to IgM?
For this discussion, the most important
parameter is down here, how long after Mini‑Pool
positivity to negative ID‑NAT, again 11.2 days, or to
negative 6 replicate ID‑NATs, an additional 6 days?
And if you want a 99 percent inclusion bound, then you
would take the standard error times 2.3. And you end
up with about 31 days to negative RNA by singlet TMA
from the index donation date.
Actually, if you add any replicates
reactive, this gets out to about 38 days. So this is
the outer limit of detectable viremia, even doing six
replicate TMA assays.
Next slide. Then this is just rolling it
all together. One interesting observation actually
Steve Kleinman noted is our estimates for the length
‑‑ this is the data I showed earlier based on the
retrospective testing at Blood Systems. And the
window periods are a little bit shorter here than we
derive by following the donors longitudinally.
This may relate to some symptom‑based
self‑deferral after the people have gone through
primary viremia. They may be less inclined to come in
and donate blood because of symptoms. And, therefore,
that's why we're not seeing as many donors and this
window period is not imputed to be as long based on
the rate at which donors give in this tail end viremic
phase compared to what is seen when we actually follow
viremic donors prospectively.
I don't think these differences are
probably statistically significant, but it suggests
that there may be some symptom‑related deferral
occurring after the primary viremia, which is what is
understood. The symptoms are all believed to occur
after the primary viremia and reflect the immune
Next slide I think is just conclusions.
Oh, just the important question of lookback, how many
of these units are infectious. In our large ID‑NAT
study collaborating with CDC, we had 27 confirmed
viremic donations that components were issued and
potential recipients exposed. Twenty‑one of those
were low viremic antibody‑positive, 6
Unfortunately, despite extensive testing
and work by CDC, we were only able to ascertain
recipient outcome in four cases. Two of two
recipients that got ID‑only IgM‑negative units were
infected, and zero of two recipients of a donor who
was ID‑only and IgM and IgG‑positive were infected.
So, despite the extensive retrotesting to
trigger additional lookback, there were very few
outcomes defined. Really, the other idea, and it?s in
progress, is to look at animal inoculation studies.
There?s primate studies being planned by CDC and Darin
Maria with Harvey Alter.
We?ve done some recent Murine knockout
model ‑‑ next slide ‑‑ I think last slide ‑‑ just ‑‑
this is a model where these mice have been genetically
engineered to lack certain immune response functions,
interferon and alpha beta receptor knockout.
These mice are extremely susceptible.
Unlike wild‑type mice with 100 plaque‑forming units,
you only get a proportion dying. These knockout mice
are extremely susceptible down to .1 plaque‑forming
units kills these mice and they have rapid outgrowth
So we did infuse ‑‑ Michael Diamond
infused 500 microliters of plasma from five of our
units times two into these knockout mice. And we were
able to show transmission in this model using one of
the breakthrough transmission cases, the Nebraska case
from last year.
So we?re continuing to study this model to
put in the IgM reactive units or do mixing studies,
adding early seroconversion samples to these
infectious units to try to further determine the
infectivity of these convalescent donation samples.
Next slide. So in summary, you know, I
think what we?ve seen is that we?re seeing a logical
progression of conversion of antibodies, both IgM,
IgG, and plaque neutralizing activity immediately
after the viremic donations.
The low level viremia, though, is
persistent in the setting of the antibody for about 11
days and it actually extends another six days if you
do multiple replicates. And this critical question
still remains, are any of these infectious? Again, to
our knowledge, there?s never been a transmission
linked to any of these seroreactive low viremic units.
CHAIRMAN ALLEN: Thank you, Dr. Busch, for
a very elegant presentation. A lot of data collected
under sort of make the rules as you go kind of a
situation. Very nicely done and saved a number of
possible transfusion‑transmitted cases in the process.
Comments, questions from the Committee?
MEMBER KLEIN: Mike, you make the point
that there is really no scientific evidence that there
has been any transmission by any of the cases that
have endogenous IgM antibody. Do you think that the
passive antibody studies that are being done really
are going to help you in that regard given the fact
that with other diseases total prevention of infection
with passive antibody is contrasted with endogenous
antibody is questionable?
DR. BUSCH: You mean a high titer ‑‑ the
immunoglobulin prep that?s being developed? I mean it
all depends on when you give it. In animal studies,
you know, if you give it before you expose, you can
If you give it literally, you know,
concurrently or within hours of the inoculation, you
may be able to either, you know, abort infection or
suppress the viremia that occurs with infection.
So, yes, so I doubt they?ll be proven to
be effective. You know the majority of people during
the primary phase are completely asymptomatic. So the
only way you?d pick them up is with nucleic acid
screening. So it?s to me by the time you would
identify a case clinically, they?ve already
MEMBER KLEIN: I?m also thinking about
whether that would help you in terms of saying that
well maybe some of these IgM‑positive infections would
be infectious. And I?m not sure that demonstrating
the path ‑‑
DR. BUSCH: Right.
MEMBER KLEIN: ‑‑ of antibody does or
DR. BUSCH: Yes.
MEMBER KLEIN: ‑‑ is going to help you
DR. BUSCH: Exactly. The other problem
we?re realizing now after we?ve sort of worked with
Michael Diamond on this mouse model is these animals
are markedly immunosuppressed. So if we show ‑‑ you
know we may not see when we add antibody to a viremic
donation and then we put it into an animal that?s
completely immunosuppressed, the ability to clear and
eradicate that complex virus may not exist. So ‑‑
CHAIRMAN ALLEN: Yes, Dr. Kuehnert?
MEMBER KUEHNERT: You mentioned that
through your data approximately, I think it was 33 to
38 days duration of viremia from the time of donation,
and ‑‑ but you also sort of passed through quickly one
slide that showed one particular donor that seemed to
be beyond that.
And I wonder if you?d comment if that is
an outlier or what were you ‑‑
DR. BUSCH: Right ‑‑
MEMBER KUEHNERT: ‑‑ showing in the data.
DR. BUSCH: ‑‑ that?s ‑‑ right. We had
one donor who was initially ‑‑ it was one of these
flip‑flop cases where we had a sample ‑‑ I forget
exactly but something like 30 days it was negative but
the donor happened to come back in like at 42 days,
you know, before we had the results on the prior
donation that would have said you don?t need to come
back anymore, they came back and got another bleed.
It was reactive on one of two initial
reps. We went back and tested that six more times.
And one of six additional reps was reactive. So it
was overall, I think, five more times, so two of seven
And that is the outlier case. And this
distribution of the length of the tail of viremia,
again the modeling currently assumes a normal
distribution. Clearly there will be some people who
may have, you know, for whatever reason, a longer tail
MEMBER KUEHNERT: But it was just the one?
DR. BUSCH: But again, if you look at that
case, there were long bleed intervals between that
date ‑‑ so it was like 30 to 42 days. And then the
donor came back again like at 70 days and was
completely negative. So when in those intervals, you
know, viremia was resolved ‑‑
MEMBER KUEHNERT: Okay. And the other
question I had, and I?ll ask Sue this also, about a
question that came up earlier about presumptive
viremic donors and how many of those are confirmed.
I wondered if you could comment on that.
DR. BUSCH: Well, it?s an interesting
issue because if you screen through Mini‑Pool NAT, and
then you resolve the individual donation, they had to
have had a fairly high level of viremia and virtually
100 percent of repeat reactives, which we do all the
time. We do an initial and then we repeat it, which
is the definition of presumptive viremic.
Virtually 100 percent of donations
screened through Mini‑Pool NAT are confirmed of
presumptive viremics. If you are screening by ID‑NAT,
again if it?s repeat reactive, it has a virtually 100
percent probability of confirming.
But we also ‑‑ a lot of the donations that
are picked up by ID‑NAT that are real are initial
reactive only. When we repeat it, it?s negative.
And that?s because what we?re picking up
is this extremely low level of viremia that is
stochastically detectable by the initial ‑‑ it was
fortunate we got it once but, you know, that tells you
there?s ‑‑ and the other factors, there?s a lot of
donations that are being given in the tail of an
epidemic that are missed by ID‑NAT that had we done
four replicates, you know, we know that five percent
of the donor pool in these regions got infected and
yet only a very small fraction came in at exactly the
time of the viremic phase.
So there is a lot of people giving in that
downstream convalescent phase that a single ID‑NAT is
not picking them up. These units have been transfused
extensively and no infections have been observed.
MEMBER KUEHNERT: The bottom line is most
presumptive ‑‑ the vast majority of PVDs are
confirmed. And so that?s something that, you know, I
think health departments, we?re trying to communicate
that message and ‑‑
DR. BUSCH: Right.
MEMBER KUEHNERT: ‑‑ it would be helpful
for blood centers to communicate that also because
that presumptive sometimes throws people.
DR. BUSCH: Right. And actually Steve
Kleinman has a paper coming out soon that will really
MEMBER KUEHNERT: Great.
CHAIRMAN ALLEN: Dr. Kleinman, you want to
make a quick comment on that?
DR. KLEINMAN: Yes, in the 2003 data,
using the ABC?s presumptive viremic donation
definition, which is a little different than the Red
Cross, is actually 99 percent positive predictive
value for presumptive viremic indicating confirmed
And I think it was kind of similar in
Sue?s Red Cross definition. So it?s very high.
CHAIRMAN ALLEN: Thank you.
MEMBER LEW: Just as a follow up for what
was said, it sounds like the study design though, you
mention this person, if you all had known he was
negative at the last time, you would have told him not
to come back. But he happened to come back and you
all went ahead and drew blood. Is that correct? And
he happened to be positive the second time?
DR. BUSCH: Correct.
MEMBER LEW: So just by the study design,
you may have missed a number.
DR. BUSCH: Yes, there?s no question.
Again, had we done, you know, replicate NAT on further
follow‑up leads, on a lot of cases we would have
determined that that window was longer. So it?s all
dependent on the sensitivity of your RNA assays just
like the HPV discussion yesterday.
CHAIRMAN ALLEN: Dr. Williams?
DR. WILLIAMS: Mike, as you are aware, the
recommendation to screen donors for headache with
fever symptoms within the last week was largely driven
by the CDC studies of the 2002 epidemic, which found
that three of the 14 implicated donors had pre‑
And you?ve speculated that IgM both would
be related to symptoms and quite likely would result
in a neutralized non‑infective donation.
So how do you resolve those two findings?
DR. BUSCH: Well, again, you know, the
symptoms, if you look at people who are presenting
with symptomatic West Nile infection, with either the
febrile or the neuroinvasive symptoms, you know, 100
percent of those people are seroreactive. By the time
the symptoms occur, RNA screening with standard RNA
assays is not sensitive enough because the primary
viremia phase has been resolved.
And, you know, I think also in the natural
history studies, both of the donors that you?ll hear
from Susan and from Blood Systems, indicate that the
symptoms come on subsequent to the primary viremic
phase. That symptom complex is probably immune
mediated. So these people are, you know, the
neurologic symptoms are a reflection of the immune
response to the infected cells.
And so to me, that that plasma viremia is
neutralized isn?t inconsistent at all with the fact
that the symptoms are occurring concurrent with the
development of the immune response.
MEMBER NELSON: Yes, but I think Dr.
Williams was raising the issue that there were three
cases where transmission had occurred from people who
previously had symptoms. So that would suggest that
maybe the virus wasn?t neutralized in those three
people if the symptoms were due to the West Nile
infection. Isn?t that what you were talking about?
And I think there is a discrepancy there.
DR. BUSCH: Yes, it could be. And, again,
if you ‑‑ I think we?ll see from Sue, a lot of donors
who aren?t infected but who were caught by false
positive results indicate there were symptoms in the
week before. So unless you?ve got a controlled
population, it depends on the symptom complex you?re
I mean a lot of people will report after
the fact that they had a headache or fever in the week
or two prior to the donation. So that?s not
necessarily, you know, related to their viremia that
led to the transmission event.
In all of those cases, yes, the people, I
think, had detectable viremia without antibodies. So
that would suggest ‑‑ I mean that question of whether
viremia, in the absence of any detectable immune
response, can be associated with a syndrome, a fever,
headache syndrome, is ‑‑ I don?t think there?s
evidence that that does happen. But I?m sure it?s
MEMBER NELSON: You know the one that
leads to the really great data on ‑‑ I mean we know a
lot about the biology of this virus infection because
of the screening of blood donors, that?s for sure.
But the one population that we can
actually learn more about the length of the window and
that kind of thing would be plasma donors who donate
frequently. And, unfortunately, because of the viral
inactivation, they?re not involved.
But it would seem that if you could save
some samples from an endemic area, an epidemic area
from plasma donors where you?d have samples taken
every few days during the epidemic, if that could be
arranged, it might add to the data. It might be
useful. It would require, you know, cooperation and
negotiation and what have you. But I think it might
CHAIRMAN ALLEN: Yes?
DR. KAHN: Mike, you are talking about IgM
infectivity and so on, how sensitive is ‑‑ first of
all, how sensitive is the IgM assay that has been
used, the IgG assay? How low level of immunoglobulin
detection can reach?
And second, how sensitive in the fact of
discriminatory between IgG and IgM, how specific is
the test for IgG/IgM? Could you please comment on
DR. BUSCH: Yes, I mean these aren?t tests
that we were involved at all in developing. There are
four or five commercial assays, Focus, PanBio, Abbott
had an assay. And then there?s also CDC?s assays.
And we?ve done very rigorous comparative studies in
there. They are virtually identical. And Kyrone also
has a variety of serologic tests, both EIA and REBA
In terms of the time to detection of
antibody, they?re obviously picking up antibody,
particularly IgM prior to clearance of ‑‑ you know,
with significant ‑‑ I don?t have data with respect to,
you know, picograms of IgM or IgG.
DR. KAHN: Yes, that?s exactly where I
want to go because we don?t know if recurrence of
infection, not disease ‑‑
DR. BUSCH: Right.
DR. KAHN: ‑‑ in the presence of
antibodies possibly. And one way of demonstrate that
is that antibody can be treatable before outcome of
diseases Dr. Klein mentioned. And as we know, IgM can
last from one year to the next year.
We don?t know if some is left over in the
second infection from a different strain or so on
because it still needs to be done if what we are
detecting calling negative for IgM doesn?t have any
IgM at all. It?s still questionable.
DR. BUSCH: Yes. And two other points,
one is that these assays are IgM/IgG capture assays.
And, again, all of the different assays that we
evaluated had IgM and IgG configurations and they
identically paralleled one another. So I think they
are specific for IgM, IgG, and IgA.
The other thing we had, I think Sue will
show some wonderful data on ramp‑up dynamics, and we
had like seven cases which had two bleeds prior to any
antibody. And we thought we would get good ramp‑up
data. But in six of those seven cases, the viral load
actually dropped before the IgM kicked in.
So that suggested something else is
underlying the control of that primary viremia besides
these antibodies. Either they are complex and we
can?t detect free antibody because it?s all bound to
the virus or they are cell mediated or host, you know,
replication capacity issues.
CHAIRMAN ALLEN: Okay. We are running
well behind. Dr. Klein, a quick comment.
MEMBER KLEIN: Yes, just a quick comment
on Dr. Williams? referral to the 2002 donors that
While it is true that three of the 14 had
symptoms prior to their donation, I think two
important points need to be kept in mind. One is we
weren?t doing Mini‑Pool NAT testing in 2002.
It may have been that those three people
would have been detected by tests that are currently
in place. Therefore, we don?t know whether the
question of headache and symptoms is necessary because
we can?t compare them. It would only be necessary if
they were test negative. And we don?t know that.
Secondly, only one of those three donors
actually had the symptom of fever and headache within
the prior week. The two other donors had that
symptom, I think in one case two weeks before and in
one case greater than two weeks before. So our
question, presumably, would not have caught those two
So I think this becomes important later on
when we talk about the symptom question and what we
should do with it. I just wanted to make those
clarifications. I don?t think I?m misspeaking if
anybody else is familiar with the paper.
CHAIRMAN ALLEN: Okay. And that certainly
is an important question.
We?re going to change ‑‑ modify our
schedule slightly. Dr. Stramer will speak and have the
full time allotted to her to present the Red Cross
data. And then we will have a break as soon as she
completes her discussion. And move on to the rest of
the agenda right after the break.
So, Dr. Stramer, we look forward to the
Red Cross data.
DR. STRAMER: Thank you.
In order to consolidate the number of
slides, I combined my title slide and my outline.
DR. STRAMER: That?s about all the
consolidation that you?ll see. I?ll present similar
types of data as Mike did with some emphasis, though,
on some of the FDA questions related to the donor
I?ll review our donors identified in 2003
that were positive by prospective Mini‑Pool and
individual donation NAT. I?ll review our
retrospective individual donation NAT studies.
I?ll review our modeling viral dynamics.
We used a little bit different approach but the time
periods, as reported by Mike and me, will not be that
much different although we have to sit down and really
do a side by side.
Then I?ll go through our 2004 data to
October 19th by both Mini‑Pool and individual NAT
And then we?ve looked at some data for
efficacy of donor deferral based on the headache with
fever question seven days prior to donation.
Next. I?ve highlighted in red what I?d
really like to go through to move through these slides
In 2003, we had 415 confirmed positive
donors identified. We used the Gen‑Probe TMA
screening method as does Blood Systems. For
confirmation, we repeat TMA and we do PCR, a validated
assay at National Genetics Institute, and we do IgM
seroconversion in the retrieve plasma unit.
The method of IgM we used was Abbott and
we have found that to be a little bit more sensitive
than the CDC test and more sensitive than the Focus
test in our validation.
Our overall frequency was about one in
5,700. The range of positive donors last year was
from the end of June through the first day in December
and 74 percent, three‑quarters of our positives, came
only from two states, Nebraska and Kansas, or 307 of
Next. This was where we saw cases last
year, again emphasizing Nebraska and Kansas.
Next. And on the previous slide, as I?ll
show for this year, the red dots don?t indicate the
number of cases, they indicate the counties. So there
may be multiple cases per county. Last year we
triggered ‑‑ we had developed a trigger that we used
this year to initiate individual donation NAT and we
did that prospectively from August 20th through the
4th of October.
Now we developed the trigger but we would
have triggered earlier had we developed the trigger
earlier. So we were only able to do this through the
second half of the year last year.
Through our ID‑NAT studies, we confirmed
181; however only about half of those required ID‑NAT
for positivity. And of those ID‑NAT positives, 92
percent of them, or the vast, vast majority, were IgM
positive at index and only eight percent, or eight of
96, were IgM negative at index and, therefore, most
likely to transmit. So that was really the yield of
our ID‑NAT prospective screening.
And the viral loads, in most cases, well,
in all the antibody cases, were below the levels of
quantitation by the NGI assay, the same issue we
talked about yesterday with HB core where the NGI
assays only can quant down to 100 copies per mil. So
the eight that were IgM negative had viral loads
between 100 and 950 copies per mil.
Next please. We then also did
retrospective ID‑NAT based on the request from FDA so
that we could complete the entire season, at least in
Nebraska, with ID‑NAT screening. We did find an
additional 21 NAT confirmed positive cases by ID‑NAT.
And all of them would have required ID‑NAT for
detection. None of them were detected by Mini‑Pools,
which is good because it corresponds with our
Of those, we had two that were IgM
negative. So if you combine the eight and two for the
entire season, we had ten ID‑NAT positive, Mini‑Pool
negatives that were IgM negative. So our total
positives in the two states where we were epidemic was
328. And of those, which I?ll show you in a
subsequent graph, 38 percent were ID‑NAT detectable
only with ten ‑‑ or just under ten percent being IgM
Next please. Okay, this shows the entire
battery of cases we detected by ID prospective,
retrospective, and Mini‑Pool NAT testing from our
first case to our last case. What?s important here is
the difference between blue and all the other colors.
This is the methods of confirmation.
What?s blue here is those that confirmed
with RNA and were IgM negative. Here you can see, as
Mike showed, the ramp up of IgM positivity as the
season went on. And these two lines indicate the
period of time that we were doing ID‑NAT testing.
Next please. Now this shows for the two
epidemic states, Kansas and Nebraska, when we did see
cases either that were detectable by Mini‑Pool or
those that required individual NAT screening for
detection. So comparably to the IgM increase, these
were those donors that required both ID‑NAT and were
IgM positive, so increase of IgM reactivity and low
In orange here, I separated out those that
were ID‑NAT reactive but IgM negative, the ten I
showed you. So you can see that they occurred pretty
much evenly throughout the season.
Next please. Okay, using the slide Mike
showed, I?m not going to dwell on the numbers but just
shows you numbers that we had during each of the
periods to our total of 438 positive. And, again,
most of them detectable by Mini‑Pool NAT.
Next please. So for the seroconversion
studies and the viral dynamic studies, we used our 415
positive donors. Of those, 350 participated in follow
up with 335 seroconverting.
But of those 335, we could study ‑‑ or we
chose to study 186 in detail. And that was because
these had multiple closely‑spaced follow‑up samples.
And the time to the first follow up we chose for
analysis was less than or equal to 35 days so that we
could include the donor with the longest viremic
period at their first follow‑up sample.
Of the 186, 76 showed repeat TMA
reactivity in multiple follow‑up leads ranging from
two to 39 days. And of those 76, 12 have fluctuating
or intermittent viremia.
Next please. I?ll show you three examples
of profiles of seroconverting donors. Blue shows you
the loss of virus. This is the signal to cut off
ratio on the TMA assay. The boxes down here represent
the quantitative PCR at NGI. And then this is the
Abbott seroconversion to IgM followed by IgG. So this
donor?s pretty typical in viral clearance.
Next please. Here?s one where even though
the virus didn?t go below the cutoff of the assay, you
can see kind of a decrease as IgM is coming up and
then another spike before viral clearance.
Next please. And here you see one
actually that went negative. We didn?t have volume to
do multiple reps but at least in the rep that was
tested, it was non‑reactive, also non‑reactive by PCR.
Two more reps were positive in subsequent bleeds and
PCR was positive on this 19 days.
Next please. So on our modeling study,
what we did is we did find three donors who we termed
anchor donors. And this corresponds to what Ken had
referred to in your question before about studying
plasma donors where we could see closely‑spaced
intervals where these donors were undergoing ramp‑up
So we then were able to calculate, doing
linear regression, a best fit line for the ramp up of
these and then fix our other donors to this anchor
line. And then calculate events based on a
standardized time. So what we calculated on the three
anchor donors was a .46 log increase per day or a .019
log increase per hour.
The doubling time for these three
individuals, their viral infection, was just under 16
hours. And then if you back calculate, using the
doubling time to one copy per mil to indicate times
zero, and then use the lower limit of detection of the
TMA assay of ten copies per mil, you can calculate the
window period from time zero, that is one copy per
mil, to NAT reactivity using the lower limit of
detection of the assay.
So for ID‑NAT, we calculated a window
period of 2.2 days and for Mini‑Pool NAT, a window
period of 4.8 days.
Next please. So here you can see the
anchor donors. These individuals had a range of
viremia presentation between 1,400 and 3,600 copies
per mil. And then between 70 hours and 92.25 hours, we
actually have the times, you know, relative to
donation and their follow‑up samples, had progressed
to viral loads of 37,000 to 110,000. So here you can
see the best fit line.
Next please. Now if you apply that best
fit line and move it down to one copy per mil and
apply ‑‑ you can apply the ID‑NAT window period here
at 2.2 days, the Mini‑Pool NAT window period of 4.8
days, then if you use this line over time and look at
where our IgM non‑reactive donors had viral loads,
this is for 241 from our 2003, it took 8.2 days to
reach the median viral load of 5,800 copies per mil
and 12.5 days total to reach the maximum viral load
which we saw at 580,000 copies per mil.
Next please. Now for the duration of
viremia study, firstly we looked at the time the
donors presented from our one copy per mil to
presentation. And that had a mean and median of 7.9
days and a range from 4.3 to 12.5 days.
Using the time when donors cleared virus
and using an adjustment factor for donors that had a
very long inter‑donation interval to their first TMA
non‑reactive result, we calculated a range for viremia
from one copy per mil to the end of detection of
viremia as 6.5 to 56.4 days with a mean and median of
And according to the sample size used for
this analysis, it would represent 99 percent of the
Next please. So this graph now shows you
the viral clearance in this 186 donors here giving you
the 56.4‑day maximum and the median and mean of 20.5
Next please. Then to calculate from one
copy per mil to the time of detection of IgM and IgG,
we had IgM first coming up at 6.5 to 29.3 days. And
a mean and median of 15.7 days. And then IgG coming
up about four days later. But we had a smaller sample
set for this. But the mean and median were relatively
close but the IgG onset, at least the shortest onset,
was about four days after IgM.
Next please. And here you can see the IgM
duration from ‑‑ or the IgM detection that is starting
from one copy per mil with a mean and median of 15.7
days from one copy per mil.
Next please. So if you put all of our
times together, this is our timeline slide. So first
I said you have an ID‑NAT detection of a 2.2 point
estimate. Then adding the time it takes to detect by
Mini‑Pool NAT, you have 4.8 days.
And then the time of donor presentation,
when donors were picked up by Mini‑Pool or ID‑NAT
screening, we had a 7.9 day mean and median. I said
it was about eight days to the median viral load
detection so those two agreed.
IgM onset had a median of 15.7 days with
this range. IgG onset was a little bit later. And
then to show the 56.4‑day maximum, here you have the
viremic period only followed by IgM and IgG so I tried
to combine these two colors into purple with a range
of 6.5 to 56.4 days.
Next please. Okay, what happened in 2004,
using the same trigger that we developed last year, we
based our switch to ID on four hots NAT reactives,
which is defined as a signal to cutoff ratio in the
TMA assay of greater than or equal to 17 and a
frequency of one in a thousand.
The actions are listed here. We did
convene con calls with the regions and the labs when
we saw two cases to let them know to be ready.
And if regions wanted to trigger early, we gave them
that option. So we then converted to ID‑NAT and we
stopped production of frozen transfusables.
Next please. This is where our cases
occurred this year. This is only one county ‑‑ these
are single counties represented, not indicating the
number of cases per county. And our hot spot, as CDC
already referred to, was California although we did
see a few cases in southern Arizona.
Next please. Just to highlight here where
the majority of our cases occurred, we?re in four
counties that we screen in southern California, Los
Angeles, Orange, Riverside, and San Bernadino. Where
greater than one case per county was observed was also
in Maricopa County but we also had a case in Pima and
Cochise. We had a number of cases in Arkansas. And
a number of cases in Kansas.
Next please. Overall, we saw for this
year 106 presumptive positives and this is our
definition based on hot cases, 99 which have confirmed
which have an S/CO range of 2.8 to 37. So we will
confirm positives that are not necessarily in the hot
During this time, we also switched to a
new probe reagent from Gen‑Probe which significantly
decreased the number of false positive reactions we
Next please. These are the areas we did
ID‑NAT. We did ID‑NAT in southern California, in
Arkansas, the Greater Ozarks Region, and in our Kansas
region, Central Plains.
Of the 56 positives we had in southern
California, 50 were detected based on ID‑NAT. Even
though we triggered in Greater Ozarks, we never had an
ID‑NAT positive. And in Kansas, we did have three of
our seven that were detected by ID‑NAT.
We don?t know yet if these were Mini‑Pool,
you know, if they?re ID‑NAT only or Mini‑Pool
detectable. Those studies are still ongoing.
Next please. This is our epidemic curve
of 2003 versus 2004. Certainly the 2004 data firstly
are less and the curve is not as pronounced as it was
Next please. Similarly, with confirmation
we haven?t seen the big upswing in IgM but we?re still
missing seven cases. But I don?t know that that?s
going to change things dramatically.
Oh, on this slide, I did want to point out
we used the Abbott IgM test in 2003 and then in 2004
because, unfortunately, Abbott discontinued their
test, we switched to the Focus test. And based on our
validation studies, we used a reduced cutoff for Focus
of a .67 times the cutoff to detect reactivity.
And interestingly enough, using that
reduced cutoff if you compare 2003 and 2004, we did
get the same relative frequency of IgM negativity and
Next please. Okay, now I want to go into
the headache with fever question. That?s our Question
33 and the donor asserts on Question 33 if they answer
yes. So the question is in the past week have you had
fever with headache? And if it is yes, we defer the
donor for 28 days and enter them into our DDR.
The above question is required from FDA,
is asked from June 1st to November 30th each year, or
longer as directed by the Medical Director.
However, at the Red Cross, and this I have
no input in, our next software upgrade will make the
question required year round. It?s just not feasible
for us to turn things on and turn things off. The
potential for error is too great.
So as we?re going through this question
and the data was have, I ask you to review it
carefully because it?s important because we are going
to be doing a question that may not have any value
So to look at the efficacy of the
question, we collected data from five regions, two
that were West Nile prevalent in 2003, that is
Nebraska and Kansas, and then three non‑prevalent
large regions. I chose LA, Boston, and our region in
Portland. And we had a half‑million plus donations
that were looked at, donors that were looked at.
So we compared the positive cases, that is
detected by testing, with a yes response to fever with
headache question. You would think in epidemic areas
you would have more yes responses.
Next please. So the vast majority of
positives, I already told you, came from two states
but the vast majority of yes responses came from
Boston and Oregon and they were later than when our
cases, which were July and September. These positive
responses to the questions started in September
We only had some limited overlap in yes
responses in cases in September in Nebraska and
Kansas. And although the number of actual deferrals
that we had was low, a yes response did not agree with
West Nile cases by time or by location of the
And if we assume all yes respondents were
West Nile positive, then the sensitivity of the
question ‑‑ so this is best case ‑‑ would have been
Next please. So I?m going to show you now
each region very quickly. Red is where virus occurred
and blue is where a yes response occurred. So this is
in Kansas. So here we had positive cases. And here
we had positive responses to question. Seven versus
Next. This is now Nebraska. These are
our number of positive West Nile cases. And these are
the number of yes responses, five.
Next please. This is southern California.
We actually had two positive cases last year. They
were travel related, they occurred early, and in the
entire region of Los Angeles last year, we only had
one donor say yes to the headache with fever question.
Next please. Now in Portland, we had one
travel‑related case and these were the number of yes
responses in blue. So they were greater starting in
July and running through November.
Next please. And lastly, Boston is my
favorite. We had no cases but we had yes responses to
36 ‑‑ 36 donors answered yes. And you can see that
this probably represents flu rather than West Nile.
Next please. So if you put all the data
together, here are the West Nile cases and then here
is the onset of a positive response to the question.
Next please. Now another way of looking
at this was through our surveys of NAT‑positive
donors. And Sharon Oryton will present these data at
So all of our NAT‑reactive donors ‑‑ this
is from her abstract, and I?ll show updated data from
2003 and 2004, but from the abstract 2003, we
requested all NAT‑reactive donors to complete a survey
which was based on CDC?s survey that we used in 2002,
administered by a donor counselor. And it?s completed
at the first follow up prior to knowledge of
So every NAT‑reactive donor is given a
survey so we have built in controls into the study
because we have negatives and positives.
West Nile symptoms are stratified as
occurring prior to, or on the day of, or after
donation. Symptoms were more frequent among cases
versus controls. And at least one symptom was
reported by 78 percent of the cases versus 38 percent
of the controls.
So we had 78 percent cases reporting
symptoms which is certainly higher than one would
predict for West Nile. But if you look at the
numbers, we had 32 percent pre and 68 percent post.
That was significantly different and of controls, an
even split of when they answered yes.
Next please. So each symptom was reported
by over 50 percent of the cases of donors reporting
pre‑donation symptoms. Fever with headache in the
seven days pre‑donation was not reported at the time
of donation but on survey, it was reported by 4.5
percent of cases and 1.6 percent of controls.
The majority of donors? symptoms occurred
post‑donation. And of symptoms reported pre‑donation,
the fever with headache question, when asked pre‑
donation, did not elicit a yes response.
So we did have bias in the studies since
questioning of both cases and controls did occur after
a West Nile NAT‑reactive notification, which is why
these numbers are probably greatly elevated as far as
symptoms that were reported.
If you are told you have an infection
perhaps, you become creative in what symptoms I?ve had
or you?ve had.
Next please. So I?ll show you now four
slides for control ‑‑ cases and controls for 2003 and
2004. So here we have the donors who reported at
least one symptom, what the most common symptoms were
that were reported. This is the updated data set. So
it?s 33 percent reported prior to donation. On the
day of or post‑donation, 67 percent.
Next please. This is what our controls
reported, people who did not have West Nile confirmed.
And it was an even split pre and post.
Next please. This is then the 2004 data,
almost identical to what we see in 2003 where 31
percent pre and 69 percent on the day of or post.
Next please. These are the controls,
again virtually a dead heat.
Next please. So in conclusion, although
the number of actual deferrals to the above question
was low, a yes response did not agree with West Nile
cases by time or by location. And best case
sensitivity for the question was 3.5 percent.
And from our survey of NAT‑confirmed
positive donors, we showed that the majority of
donors? symptoms occurred post‑donation. And if
symptoms were reported pre‑donation, the above
question, when asked pre‑donation, did not
consistently elicit a yes response.
Again, there was bias in the study and so
what we conclude is that the above question has no
CHAIRMAN ALLEN: Thank you very much, Dr.
Stramer. I?ve got a couple of questions.
You calculated the best case sensitivity
for the question. Did you calculate a specificity for
DR. STRAMER: No, we had no way of ‑‑
CHAIRMAN ALLEN: Okay.
DR. STRAMER: ‑‑ well, there was no way to
really do that with any type of accuracy.
CHAIRMAN ALLEN: Appreciating the problem
of getting accurate symptom questions, you commented
on the bias. And I?m not referring this to the
question that is there but more to the laboratory
results that you got.
And my question would be for donors who
had asymptomatic viremia compared with those that had
West Nile Fever or Meningoencephalitis, was there a
different pattern in terms of the viremic data,
appearance of antibody, and that sort of thing? And
you probably don?t have all that kind of complete
DR. STRAMER: No, I believe in 2003, we
had five donors who actually were symptomatic. And
they ‑‑ I mean who developed severe disease. And they
did donate and they felt fine on the day of donation.
So that?s really the only information I have.
CHAIRMAN ALLEN: But in terms of the
duration of viremia or the ‑‑
DR. STRAMER: No, they were not different
than the other duration of viremic individuals. We
looked at that, yes.
CHAIRMAN ALLEN: Okay. Thank you.
MEMBER KLEIN: So I think you?ll find
fewer headaches in Boston now that the Red Sox won the
MEMBER KLEIN: But more to the point, do
you know of anyone who is doing any kind of testing of
the donors who report that they have headache and
fever a week before donation when they are screened
and then are turned away.
DR. STRAMER: No.
MEMBER KLEIN: Is anyone testing them?
DR. STRAMER: No, we haven?t done that.
But in the 3.5 analysis, we just assumed everyone who
answered yes was infected. And even then, it was only
3.5 percent sensitive.
CHAIRMAN ALLEN: Other questions or
comments? Yes, Dr. Kuehnert?
MEMBER KUEHNERT: Just wanted to turn to
the length of viremia question again. I wondered,
first of all, if you can tell us whether Red Cross has
had a situation where they?ve had a donor test
positive and then come back for their next donation
and been viremic just to sort of get a reality check
on whether that has occurred.
DR. STRAMER: No. I mean we?re deferring
the donors now who are viremic for a minimum of 28
MEMBER KUEHNERT: So when they come back
after 28 days ‑‑
DR. STRAMER: No, wait. Let me finish.
That?s one criteria. And then the other criteria is
that they must test ‑‑ I mean this is what the FDA is
asking, they must test ID‑NAT non‑reactive and have
seroconverted. If we can?t demonstrate
seroconversion, even though they cleared virus, we yet
require another sample to make sure that what we?re
seeing is an intermittent viremia in the absence of
MEMBER KUEHNERT: But if they actually ‑‑
DR. STRAMER: So it?s really the time of
when they would present for subsequent donation is
actually far longer, in reality, than 28 days.
CHAIRMAN ALLEN: Right. If they had come
in and donated a unit of blood, they were found to be
totally acceptable, donated a unit of blood, it was
positive on NAT testing, because they had just
donated, they would be deferred for at least 56 days,
DR. STRAMER: If it?s a whole blood donor.
CHAIRMAN ALLEN: Yes.
DR. STRAMER: Right. But a pheresis donor
or an autologous isn?t.
CHAIRMAN ALLEN: Okay.
MEMBER KUEHNERT: So you?ve had people
come back for ID‑NAT at 28 days and been positive?
DR. STRAMER: Yes, in the follow‑up study.
MEMBER KUEHNERT: Right, right, in the
study, okay, okay.
DR. STRAMER: Yes.
MEMBER KUEHNERT: The other question I had
was just to try to compare apples to apples with Dr.
Busch?s data. What?s the 99 percent confidence
interval for length of viremia? I think ‑‑
DR. STRAMER: The outer limit was 56.4
MEMBER KUEHNERT: So that was the longest
that someone was ‑‑
DR. STRAMER: Well, not observed but that
was calculated based on the modeling.
MEMBER KUEHNERT: Oh, okay.
DR. STRAMER: Observed was 39 days.
MEMBER KUEHNERT: So the 56.4 was a
maximum 99 percent? Okay.
DR. STRAMER: Well, that was what the FDA
requested, 99 percent.
MEMBER KUEHNERT: Okay. Thanks.
DR. KLEINMAN: Can I comment on that?
CHAIRMAN ALLEN: All right. Okay, Dr.
Kleinman, do you want to comment on this particular
DR. KLEINMAN: Yes. Because, Sue, that
was 56.4 days from your time zero.
DR. STRAMER: That?s correct.
DR. KLEINMAN: Not 56.4 days from your
time of actual detection which, if I understood your
data correct, you?d have to adjust by about ‑‑ you?d
have to adjust downward by about 7.9 days, I think.
So then your maximum would be 48 days from
the time of detection by NAT. The model would predict
a maximum viremia period of 48 days for 99 percent,
DR. STRAMER: Yes, Steve, you?re
DR. KLEINMAN: Okay.
DR. STRAMER: The 56.4 days is the entire
viremic period from one copy per mil to no more virus
or one copy per mil on the other end. So you would
have to deduct the time period from when the donor
actually presented which was 7.9 days. Steve?s
CHAIRMAN ALLEN: Dr. Williams?
DR. WILLIAMS: We?ll get a chance to
discuss this more after the break. And with a number
of card‑carry epidemiologists around the table, it may
But two observations. One is the
observation of onsite deferral for any question be it
male sex with other males or West Nile Fevers is just
a shadow of the total deferral impact which largely
occurs before the donors appear at the blood center.
So just to keep that in mind that it?s really a small
proportion of the total deferral impact.
And the second comment is what we?re
really talking about is predictive value for the
window period when the NAT assay is going to be
negative for the donors. And I would maintain that
you really can?t get there from the data at this
So that, you know, as sensitivity issue
determination using, as a gold standard, the window
period, donors who would not be detected by NATs, we
really can?t estimate at this point.
DR. STRAMER: Well, you can?t ‑‑ well, you
also can?t estimate the value of the question.
DR. WILLIAMS: That?s true.
CHAIRMAN ALLEN: Other questions for Dr.
Stramer from the Committee?
All right, Dr. Busch?
DR. BUSCH: Yes, just one comment, Sue.
In your follow‑up symptom data on the donors, you
presented that 78 percent of the cases indicated there
was a symptom whereas 30 percent of the controls. And
then you showed what percentage of those who reported
symptoms reported the symptoms before or after.
And I just ran the numbers to calculate
out. In the pre‑donation symptoms, which I think is
the focus of the question, you know how many prior to
the index donation had symptoms, if you actually
calculate out what percentage had any symptom in the
cases, it?s .24 percent. And in the controls, it?s 14
So 24 percent versus 14 percent had any
symptom. And none of them, I think, had both fever
and headache before the donation whereas after the
donation, it?s 53 percent in the cases and 14 percent
in the controls. So the controls had identical rates
of symptoms before and after.
DR. STRAMER: Right.
DR. BUSCH: And the cases really had
virtually identical rates of symptoms before as did
the controls where they had much higher rates
subsequently. So I think it?s a wonderful case
control analysis that to me argues that the symptoms
before are really background.
DR. STRAMER: Right. Because they?re
background, they blend into the controls. You are
right. That?s a good observation.
Okay, Sharon, the card‑carrying
DR. ORYTON: Having done the analysis
DR. STRAMER: Well, leave it to the
DR. ORYTON: ‑‑ the people that reported ‑
‑ one thing that?s ‑‑
CHAIRMAN ALLEN: Would you please identify
DR. ORYTON: I?m Sharon Oryton from the
FDA. One of the things that wasn?t in the abstract
and will be in our AABB abstract is there was a very
interesting combination. So people didn?t just report
symptoms pre‑donation or just post‑donation. We had
all kinds of combinations of that. And that will be
The other thing I do want to point out is
even in this population, fever with headache had a
positive predictive value of 69 percent. Now granted
those individuals pre‑donation didn?t admit to those
symptoms when they donated but the symptoms themselves
do have a good positive predictive value for West Nile
DR. KLEINMAN: Was it after the donation
DR. ORYTON: These were the ones that were
pre‑donation. Just looking at the 16 that did report
fever with headache pre‑donation, the positive
predictive value was 69 percent.
CHAIRMAN ALLEN: That?s 2004 data?
DR. ORYTON: That?s the combined 2003/2004
CHAIRMAN ALLEN: Okay.
MEMBER KUEHNERT: Could I just ask a
question? Sharon, was there ‑‑ I haven?t had a chance
to look at the abstract ‑‑ was there any kind of
multi‑varied analysis done to look at independent
DR. ORYTON: The data set really for the
number of symptoms that we had really wasn?t large
enough to do that. And I had hoped with the 2004 data
we would be able to. It didn?t increase that sample
size that large. And I just did that analysis really
two weeks ago. So no, I haven?t looked at that.
MEMBER KUEHNERT: Okay.
CHAIRMAN ALLEN: All right. We are well
over our planned schedule. Any other questions or
comments from the Committee?
CHAIRMAN ALLEN: Okay. We will take a 15‑
minute break here. I would like to reconvene at
We will then go into open hearing and then
Dr. Williams will make the presentations of the
questions and FDA?s thinking.
(Whereupon, the foregoing
matter went off the record at
11:26 a.m. and went back on the
record at 11:45 a.m.)
DR. SMALLWOOD: Dr. Allen.
ACTING CHAIRMAN ALLEN: We're going to
move into our open public hearing. I've got three
speakers who would like to speak: Dr. Jeffrey Linnen
from Chiron Corporation; Dr. Steven Kleinman, combined
statement from AABB, ABC, and ARC; and Dr. Brian
Custer or, Mike, are you presenting his day or is
Brian presenting data?
DR. CUSTER: I am.
ACTING CHAIRMAN ALLEN: Dr. Brian Custer
from Blood Systems, Incorporated.
** So I need to read the open hearing
announcement, and following that, we can move right
into Dr. Linnen's presentation.
Both the Food and Drug Administration and
the public believe in a transparent process for
information gathering and decision making. To insure
such transparency at the open public hearing sessions
of the Advisory Committee meeting, FDA believes that
it is important to understand the context of an
individual's presentation. For this reason, FDA
encourages you, the open public hearing speakers at
the beginning of your written or oral statements to
advise the committee of any financial relationship you
may have with any company or any group that is likely
to be impacted by the topic of this meeting.
For example, the financial information may
include the company's or group's payment of your
travel, lodging, or other expenses in connection with
your attendance at the meeting. Likewise, FDA
encourages you at the beginning of your statement to
advise the committee if you do not have any such
If you choose not to address this issue of
financial relationships at the beginning of your
statement, it will not preclude you from speaking.
** DR. LINNEN: Okay. First slide, please.
Okay. The first thing I want to correct
is I'm from Gen‑Probe, not Chiron.
But this assay ‑‑
ACTING CHAIRMAN ALLEN: Sorry. I'm just
reading what's on the paper.
DR. LINNEN: ‑‑ is the result of a
partnership between the two companies, Gen‑Probe and
Chiron Blood Testing.
Okay. Next slide, please.
I want to give you an overview real
quickly of the assay. This is an investigational
assay, and it's currently being run on two platforms.
The semi‑automated version of the assay is run on the
same platform that our licensed HIV HCV assay uses,
and we have recently started testing on the TIGRIS
system, which is our fully automated system.
Testing on the semi‑automated system
started in June of 2003. Testing on TIGRIS started in
August of 2004.
This shows the semi‑automated system. I
just want to comment on the throughput. This could be
considered a high throughput system. If one
technician is working, 182 individual donor testing
results can be generated in about five to six hours.
If pools of 16 donations are tested, nearly 3,000
donations, results could be obtained in the same
length of time.
This shows the TIGRIS instrument. This is
a fully automated system. It has a fully automated
sample in handling and assay processing. Since I
called the semi‑automated system high throughput, I'll
call this very high throughput. We can obtain 1,000
individual donor test results in 14 hours.
If pool testing is used, 16,000 pooled
results can be obtained in 14 hours.
The other thing I want to point out is
that it has reagent dispense verification which
monitors critical reagent addition steps.
I want to show some data comparing the
performance of the two systems. This is analytical
sensitivity data. It's a pretty large experiment. It
uses 90 replicates at each copy level. These are the
copy levels on the X axis. The bars are percent
reactivity. So we're looking at 100 copies to zero
copies. The lowest possible samples are at one copy,
and you can see at 130 copies the performance is very
similar, exactly the same. At ten copies, very
similar. You can see that the semi‑automated system
performed slightly better in this experiment, but you
can see then at the next lower copy level the results
So overall I think we would conclude that
the results between these two systems when compared
Next slide, please.
This is also a comparison of the two
systems. This shows in‑house specificity testing that
was done at Gen‑Probe. This experiment or these
series of experiments along with the analytical
sensitivity experiment was done with three lots. So
the results are divided among the three lots.
What we see here is about 3,000 tests for
each platform and two false positives were seen in the
semi‑automated system. One was seen in the TIGRIS
system. Eleven invalid results occurred with the
semi‑automated system, two with the TIGRIS system.
Overall the specificity was very similar, 99.94
percent with the semi‑automated system, 99.97 percent
Now, this is similar to what we have seen
in the field. It's not quite as good as the
specificity that Dr. Stramer showed, but we think it's
representative of how the assay performs. So we think
specificity is really pretty much the same on both
Next slide, please.
Okay. Now, I want to update screening for
2004. This year we have a total of 29 sites. That's
compared to 24 in 2003. The first confirmed positive
donation occurred in the middle of April, and this
came from Florida. The confirmatory testing for 2004
is similar to what we were doing in 2003. There have
been some changes. We are using a different
confirmatory net assay. We're now using the Gen‑Probe
alternative TMA assay, which is a validated assay
that's been transferred to the Bayer Reference Testing
Lab in Berkeley.
We're continuing to use Focused
Technologies for IgM testing.
So this is an overview of the clinical
results so far. Based on testing starting in June of
2003, we've tested over 15 million donations with the
procleics WMB assay, and 1,100 positive donations,
West Nile virus positive donations have been
intercepted, and that's since the beginning of testing
If you compare 2003 to 2004, the numbers
are really quite different. Two thousand four, based
on our algorithm for confirmation, we had 885
confirmed positive donations with this test, and these
were primarily in Colorado and the upper Midwest.
In 2004, the numbers are substantially
lower. This number is actually confirmed, positives
plus probable positives, basically the same definition
that Dr. Stramer used for presumptive positives, and
these are primarily in the Southwest, as has been
Okay. I want to say a little bit about
the testing that has occurred on the TIGRIS system.
Currently, three sites are using the instrument. The
American Red Cross in San Diego started in August,
August 18th. Flood Systems started later in August,
August 26th, and then the Bonfils Blood Center in
Denver started August 30th.
Now, two additional sites are in the
process of preparing the starting testing on this
system. So the data that we have as of 10/6 is over
36,000 individual donor test results have been
generated. We are six initially reactive results, one
confirmed positive and five of the results are
pending, but based on the SSTOs, most of these will be
confirmed positive results.
Okay. Next slide.
I'd like to show you the confirmed and
probable positives for 2004 showing the number by week
on the X axis. As you can see, there's a definite
peak that occurred, 8/23 or the week starting 8/16.
Next slide, please.
What's really useful is to compare it to
the 2003 data, and you can see the data for 2004
almost appears like background compared to 2004, but
there's a couple of interesting things when you look
at this graph.
There's a peak occurs the exact same week
between the two years, and there's also this
phenomenon where there's a slight downturn in the
number of confirmed cases and then it goes back up
again. They're not exactly the same pattern, but it's
very similar and we don't quite ‑‑ haven't analyzed
that in detail to try to understand why that might be,
whether they're coming from different geographic
regions or what the case is.
I'd just like to recap what I've gone
over. This assay has been used to identify over 1,100
West Nile virus infected donations, and again, that's
since June of 2003. Testing on TIGRIS started in
2004, and based on our in‑house studies with the lots
that are being used for the pivotal clinical trial, we
think that the two instrument platforms perform
basically the same.
And one last slide. I'd like to
acknowledge the NHLBI for their support in the
development of this assay.
Thank you very much.
ACTING CHAIRMAN ALLEN: Thank you, Dr.
Any questions for Dr. Linnen, comments?
ACTING CHAIRMAN ALLEN: Okay. We will
move on to the second presentation, Dr. Kleinman.
** DR. KLEINMAN: Good morning. I'm Dr.
Steven Kleinman. I would like to announce that I do
have some financial consulting arrangements with
manufacturers that are involved in NAT assays.
Today I am here representing the AABB
Interorganizational Task Force on West Nile Virus.
That task force includes members of ABB, America's
Blood Centers, American Red Cross. It also has
representatives from FDA and CDC, but this statement
comes from the three blood organizations that are
represented on the task force.
So the Interorganizational Task Force on
West Nile Virus would like to comment on the available
scientific data regarding the deferral period for
blood donors who had a reactive or confirmed positive
screening test for West Nile Virus by NAT.
We will also comment on the recommendation
that donors who are deferred based on a reactive or
confirmed positive test should be tested and found
nonreactive by ID NAT on a follow‑up blood sample
prior to their reentry.
Based on the data presented to the BPAC
today from both ARC and Blood Systems, AABB supports
an extension of the deferral period from 28 to 56
days. Viremia has been found to extend for up to 39
to 49 days following a NAT positive donation, and
preliminary modeling predicts that the viremic period
would be less than or equal to we have 56 days here
from the time of one copy per mL, but it's actually to
48 days from the time of detection for 99 percent of
the West Nile virus infected donor population.
The data demonstrate that viremia beyond
28 days is at a low level and is accompanied by IgM
anti‑West Nile virus antibody. To date there has not
been a documented case of transfusion transmission of
West Nile in the presence of donor IgM.
Although the available data set supports
the absence of such transmission, it is too small to
provide complete assurance that transmission could not
occur. Therefore, during the continuation of donor
testing under IND, AABB recommends that in addition to
the 56‑day minimal deferral, donors who test West Nile
virus NAT reactive or confirmed positive must have a
non‑reactive ID NAT prior to reinstatement. This ID
NAT could be obtained any time after donation, could
be obtained prior to the 56 days, but the donor would
still be deferred for 56 days, but the donor would
still be deferred for 56 days. That's our position.
Data accumulated during the continuation
of current INDs can then subsequently be reviewed and
may prove to be sufficient to justify discontinuing
the ID NAT testing requirement and permitting
donations solely on the basis of an elapsed 56 days.
We recommend that FDA consider requiring
manufacturers to include this ID NAT retesting
requirement as part of their ongoing IND. Based on
the modeling that predicts that the vast majority of
West Nile virus NAT reactive donors will not be
viremic beyond 56 days, we additionally recommend
automatic reentry, that is, a procedure where no ID
NAT required for those donors who do not return for an
extended period of time, for example three to six
So what we're saying here is that if you
want to reenter the donor in 56 days, you would need
a negative ID NAT, but there are circumstances that if
you wait long enough you wouldn't need to obtain an ID
NAT and you could still reenter the donor. We think
that time frame should be somewhere in the three to
six month time frame.
Now, turning to the other issue in front
of the committee, AABB recommends that the use of the
pre‑donation question about fever and headache to
interdict potential West Nile virus infected donors be
eliminated. This question was added to the donor
history prior to the availability of screening tests
under IND presumably based ‑‑ and I think we heard
today actually based ‑‑ on the data reported by
Pealer, et al., for the 2002 West Nile virus season,
that three of 16 West Nile virus transmitting donors
reported pre‑donation symptoms.
However, these symptoms were not reported
in two of the donors within the seven‑day period
before donation. It was recognized by the CDC that
this question had limited value even at the time of
implementation. The data presented today by American
Red Cross for 2003 do not support the efficacy of this
question. The frequency of reported fever with
headache did not correlate with West Nile virus
incidence either by geography or by time.
Even in the unlikely event that all donors
reporting fever and headache had actually been
infected in the epidemic regions, the sensitivity of
the question would not have exceeded 3.5 percent.
Therefore, we advocate elimination of this question
which has no demonstrable value and which contributes
to an already complicated donor questioning process.
A further examination of the 2003 data
indicates that donors who tested confirmed positive
for West Nile virus had the majority of their symptoms
develop post donation. Based on these data, we
recommend continued encouragement for donors to report
post donation information about fever with headache
and for blood thinners to continue to retrieve units
that are in inventory from any such donor reports.
Finally, we would like to comment on the
final sentences in the agency's review of management
in the appendix section of the issue summary document
for this meeting. This section states that, quote, if
a master pool is reactive and all individual donations
are nonreactive, a fresh specimen from each of the
indexed donations is tested using the original NAT and
the alternate NAT method, unquote.
Under the current West Nile virus INDs,
reactive pools for which resolution testing has been
performed and all donations associated with the
samples found nonreactive by ID NAT are released
without further testing.
This is the same scheme used for licensed
HIV‑1 and HCV NAT assays. It is not realistic to
think that an alternate sample under the strict
handling requirements of the NAT assays will always be
available for testing and that results of alternate
NAT on this sample would be available in time to
release time sensitive components.
There are no data to support the statement
that I quoted above from any of the INDs. I think
that's the conclusion.
ACTING CHAIRMAN ALLEN: Thank you, Dr.
Any questions or comments for Dr. Kleinman
from the committee?
ACTING CHAIRMAN ALLEN: We will move on to
the third statement. Dr. Custer.
** DR. CUSTER: Hi. I'm Brian Custer, and
actually I'm an employee of Blood Systems.
What I want to do is actually talk to you
about our 2003 donor survey results. We've been able
to look at them in a little more detail, and they
provide some insight. There are some limitations to
what you can glean from the 2003 data and actual
survey and the way it was implemented, but I think
that it actually is informative.
So just briefly, BSI Medical Affairs staff
actually administered the questionnaire. It is based
on the CDC questionnaire, just slightly modified, and
then subsequently, of course, as we know, people
rather than confirmed positive or not necessarily
confirmed positive due to the issue with false
positives, particularly during 2003.
Next slide, please.
So the people who were interviewed who
ultimately then confirmed either negative or positive,
63 were negative and 141 were positive. So that's
just the lay of the land, the large numbers.
The next slide, please.
Brief information on sort of who these
people were demographically and also the time of the
interview in relation to actually the donation, and it
was fairly soon after the donation. So we don't have
a lot of information on, you know, symptoms 30 days
out after a donation, but in regard to age the people
who confirmed positive and the people who were
negative were essentially the same, and then for
gender, a slight suggestion that males were more
likely to be positive than females, but that's not
Next slide, please.
So this is a fairly busy slide. What it
does is it covers all of the various symptoms that
were actually inquired about during the interview or
during the survey, and you can see the first column.
This column is the people who confirmed negative, and
then the center column is the people who confirmed
positive, and then a comparison of ‑‑ when it's on,
it's on ‑‑ a comparison basically using chi square or
Fisher's exact test.
And fever and headache are not the only
symptoms that come out as being significantly more
likely in the people who confirm positive. In fact,
actually new rash was the one that was most
statistically significantly more frequent in people
who confirm positive, but there were other symptoms
also that were more likely, such as painful eyes
(phonetic) and chills and generalized weakness. So I
just wanted to make that clear. It by and of itself is
not going to necessarily discriminate.
Next slide please.
But to look specifically at fever,
headache, and headache and fever, once again now
actually the next slide I will present will actually
look in relation to actually the discrimination data,
but right now we're just looking at data without
regard to the onset date of the symptoms. So these
are people who will have donated and may have had the
symptom before or may have had the symptom after.
If you do look and see that actually with
regard to fever, it does seem that people who actually
ultimately confirm positive were more likely to
report fever than those who were negative. It's a
similar situation for headache and actually also for
both fever and headache, but once again without regard
to the onset date.
So now moving on to the next slide, the
next slide tries to break this out toward those
various periods of interest, and you can see at the
top actually is fever, once again, and then there's
headache, and then there's headache and fever
If you look at fever alone, you can see
that actually in the week prior to the donation, none
of the people who were positive actually reported the
symptoms in that interval. For headache, the
distribution, once again, you can look and you can do
the comparison between the negatives and the
positives, but you can see that for the positives it's
pretty evenly distributed when they're going to report
that headache symptom.
And then finally with regard to headache
and fever, once again, in that week prior to the
donation actually nobody reported those symptoms
whether they were West Nile virus positive or West
Nile virus negative in final confirmation in those
seven days preceding the donation.
There were people who reported the
symptoms prior to the seven days and also people who
reported the symptoms afterwards, and that's really
all I wanted to leave you with. We're just sort of
looking at that data. We don't see a strong
relationship between that particular seven‑day
interval in advance of the donation and the headache
and fever combination.
ACTING CHAIRMAN ALLEN: Thank you, Dr.
Any questions on these data for Dr.
One wonders whether some people consider
mosquito bites to be a rash.
DR. CUSTER: That's true.
ACTING CHAIRMAN ALLEN: Yes. Dr.
DR. WILLIAMS: While the study was in
place was there not a deferral question in place
regarding headache with fever and a weak prior
donation? So unless there were false negative
questions, you wouldn't expect to see that.
DR. CUSTER: Well, the question was in
place, of course. The simple thing is that all of the
people who would have been deferred for that were
deferred, but now going back in a sort of
retrospective questioning, then people do report these
symptoms. So actually everybody reported here would
not have been deferred for the symptom complex because
they didn't report it at the time of donation.
ACTING CHAIRMAN ALLEN: Ken.
DR. NELSON: Why did you ask about
headache and fever for more than seven days prior to
DR. CUSTER: That was the design of the
questionnaire, and the questionnaire asked
specifically about the onset date, and so those are
categorizations that were made after ‑‑
DR. NELSON: So you first asked if you had
a fever, headache and ‑‑
DR. CUSTER: If you had a fever and then
what was the onset date for that fever.
DR. NELSON: Because if you look at those
data, there were more people reporting fever more than
seven days among those who were West Nile virus
DR. CUSTER: That's true.
DR. NELSON: And you know, I don't
ACTING CHAIRMAN ALLEN: I apologize, sir.
Thank you very much.
We do have one additional speaker, Dr.
Michael Fitzpatrick, America's Blood Centers. It was
not on my list, but he does have a handout.
** DR. FITZPATRICK: Thank you, Dr. Allen.
I am Mike Fitzpatrick and I'm employed by
America's Blood Centers as their chief policy officer.
Just a couple of slides to correlate with
Dr. Stramer's information on the impact of the
headache and fever question.
Next slide, please.
We surveyed our centers and got the
results that you can see of 5.6 million donor
interviews compared to 4.8 million West Nile virus NAT
assays, meaning that about .8 million donors were
deferred prior to being tested for various reasons,
not just the headache and fever question, however.
The two blue lines, if you look at them,
indicate a dead battery ‑‑ no. We've normalized the
data as to rate per 10,000. So you're looking here at
the rate of positive tests per 10,000 samples tested
for West Nile virus testing. Here you're looking at
the rate of yes answers to the headache and fever
question per 10,000 donors interviewed.
The blue lines, this blue line is from
centers that actually had a West Nile virus positive
test. So they had a donor that answered no to the
headache and fever question, was subsequently tested
for NAT, and the test came out positive.
This orange line indicates those centers
who had yes answers to the headache and fever
question, but have had zero positive West Nile NAT
test results in this period, and this is July 2003 to
And you see that the headache and fever
yes answer lines track fairly well. They're getting
about the same rate of positive answers regardless of
other West Nile virus test are positive or whether
it's in the region, and so the point of this slide is
to point out that there doesn't appear to be a good
correlation between the West Nile virus test results
and the headache and fever question.
So from that we look at the ‑‑ we have
similar interview deferrals. We have no correlation
to season or the geographic distribution, and we don't
really see there's much value in that test. And we do
have regional data. For time interest I won't show
that to you, but the next slide shows actually a
region that Sue talked about also.
And this is Nebraska. You can see here
there were zero yes responses in 15,000 interviews,
14,953 tests results with 19 positive.
So even in an area where there was endemic
West Nile virus and there were positive test results,
there were zero yes answers to the fever and headache
So in regards to one other comment just to
Dr. Williams on the self‑deferral issue, yes, we did
see a lot of self‑deferrals for geographic travel when
we instituted deferrals for BSE. I think it's
unlikely that we're seeing a lot of self‑deferrals for
advertising about fever and headache and West Nile
virus. The downers are asked how they feel during the
interview. They're asked about their general health
conditions. They're also asked an additional question
about fever and headache. It's unlikely that we're
seeing a lot of self‑deferrals that are not being
counted with the fever and headache issue.
That's all I have. Thank you.
ACTING CHAIRMAN ALLEN: Thank you, Dr.
DR. NELSON: Apparently if somebody
answers yes to that question, they're not tested for
West Nile virus or followed up, right?
DR. FITZPATRICK: Correct. If you answer
yes and they're deferred, there's isn't a follow‑up
DR. NELSON: There is no follow up.
DR. FITZPATRICK: Correct. They're not
DR. NELSON: I mean that would be one way.
You could design a study where you took a bunch of
people who reported a headache and then controls and
looked for West Nile virus markers then and
subsequently. I mean, that might be the best way to
get the answer to this question.
ACTING CHAIRMAN ALLEN: Doctor.
DR. KUEHNERT: Well, I just wanted to
point out that, you know, all you're really saying is
this question has poor specificity because, you know,
your number of donors, you know, overwhelmed the
number of West Nile virus positive individuals where,
you know, you could see the possible value of it. So,
I mean, what you're really saying, they just have very
bad specificity, right?
DR. NELSON: It usually occurs in
December, too, right?
DR. KUEHNERT: We're not going to get into
that, but yeah.
DR. FITZPATRICK: Yeah, I mean, if you
look at the regional, even in the regions where as Sue
showed, where you had fairly high positive test
results and were considered hot regions by both CDC
and the blood donor industry, there was no increase in
the fever and headache yes answers.
So the raw correlation ‑‑
DR. KUEHNERT: Right, but even there the
rate is, you know, one in 1,000, you know. So looking
at a graph like that I don't think you could really
evaluate anything except specificity.
DR. FITZPATRICK: Right. When you have
very, very low prevalence, its difficult to draw a
ACTING CHAIRMAN ALLEN: Dr. Stramer, a
quick comment and we need to move on.
DR. STRAMER: It's bad sensitivity and bad
specificity because NAT in Nebraska, the frequency of
West Nile positives was one in 143.6 percent of those
tested, and even there during the epidemic period we
only saw five positives. If you take all of the
positives, the yes responses, and you assume all of
them are infected, as Ken, you test all of the yeses.
Let's assume all of the yeses are
positive. Then the sensitivity of the question was
only three and a half percent.
ACTING CHAIRMAN ALLEN: Thank you.
DR. LEW: I think it might be worthwhile
just pointing out the retrospective study that they
showed where there was no positives within the time
period, the one to six days, because it is
retrospective, there is inherent bias in that if I had
donated blood and I initially said I didn't have a
fever and headache then and then now I'm asked to come
back because I'm positive, I think I would try to
remember. If I had a fever and headache, it was a
long time ago rather than within the time period I
should have deferred myself.
I mean I think it's natural for people not
to want to implicate themselves.
ACTING CHAIRMAN ALLEN: Yes. Potential
biases in the way in which we unfortunately need to
Okay. Any other questions or comments?
ACTING CHAIRMAN ALLEN: Fine. Dr.
Williams, would you present FDA's current thinking in
the questions, and let's move on with our discussion?
** DR. WILLIAMS: Thanks.
Next slide, please.<