DEPARTMENT OF HEALTH AND HUMAN SERVICES

Donna DiMichele; Dr. Catherine Manno.

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

Dr. Louis Katz will be participating as a non-voting industry representative acting on behalf of regulated industry for this meeting. Dr. Katz' appointment is not subject to 18 US Code 208.

With regards to FDA's invited speakers for today's discussions, the following disclosures will assist the public in objectively evaluating presentations and/or comments made by the participants.

Ms. Mary Beth Bassett is the Vice President for Quality Assurance and Regulatory Affairs, Blood Systems Incorporated. Ms. Debra Kessler is Director, Regional Services, New York Blood Center, and a member of AABB. FDA participants are aware of the need to exclude themselves from discussions involving specific products or firms for which they have not been screened for conflict of interest. Their exclusion will be noted for the public record.

With respect to all other meeting participants, we ask in the interest of fairness that you state your name, affiliation, and address any current or previous financial involvement with any firm whose product you wish to comment upon. Waivers are available by request under the Freedom of Information Act.

And that's the end of the conflict of interest statement. At this time I would like to ask if you have a cell phone, that you would put it in the silent mode. The person sitting next to you would really appreciate it.

Thank you.

DR. ALLEN: Good morning. Topic III for the meeting, the first topic for the day will be the study design for the abbreviated uniform donor history. This has been an ongoing development over several years. I'm excited that it is moving forward, and hope that that momentum will be continued.

Our first speaker with background and introduction will be Dr. Sharyn Orton.

Agenda Item: III. Study Design for Abbreviated Uniform Donor History Questionnaire - A. Background and Introduction - Sharyn Orton, PhD, OBRR, FDA

DR. ORTON: Good morning. I'm going to give you a brief background, a little bit of information on the work of the task force on the full length questionnaire, some of the history of discussion on the abbreviated questionnaire.

Now, the AABB Donor History Task Force -- in 2001 this task force submitted to FDA a proposal to revise the full length questionnaire. And this was a very intensive work group. They used focus groups and one-on-one cognitive interviews to assess donor comprehension of questions. And they developed a very, very large package that has everything to do with the donor history questionnaire.

So, in 2002, they submitted their final materials to the FDA, including the full length questionnaire that had been revised, medical deferral list, educational materials, and an abbreviated donor history questionnaire.

In 2003, the FDA completed their review and proposed to accept the full length questionnaire, the medical deferral list, and educational materials, but not the abbreviated questionnaire. And this is in the draft guidance for industry that was published in April 2004. Comments have been coming to the docket. FDA has received them. They have shared them with the task force, and revisions have been submitted. Right now, that final process is under review.

Now, as far as the history of the abbreviated donor history questionnaire, this concept has always been endorsed by FDA, certainly long before I even worked there. I went back and looked. In 1993 and 1994 there was the AIR report. It was part of the 1997 blood action plan. In 1998, the FDA approved the first abbreviated donor history questionnaire. In 1999, the FDA approached the industry to develop a task force to include the evaluation of an abbreviated questionnaire.

In 2000, there was an AABB/FDA workshop that also included discussion about an abbreviated questionnaire. There was BPAC discussion in 2001. In 2003, a second abbreviated questionnaire was approved by FDA. And in 2003, there was very extensive BPAC discussion.

This particular BPAC, the discussion covered abbreviated surveys in general, survey concepts and design, and survey participant behaviors. Particularly, John Boyle(?) presented some information and Paul Beatty(?) from the National Center for Health Statistics. We discussed donor motivation or lack thereof when answering questions; donor satisfaction and retention; impact on blood safety; focusing donor attention to recent events; the use of capture questions versus a series of specific questions.

And a lot of discussion included: the methods of abbreviated questionnaire assessment; including comparison to rates for deferrals; infectious disease markers; post-donation information; biologic product deviation reporting; post-transfusion disease; whether we should look at things historically, prospectively.

We tried to incorporate information on the biologic deviation reports that FDA had received at the time. There was quite a bit of data given about that; what kind of data collection and analysis and interpretation might be appropriate; potential study designs. And there were many opinions given by the major blood organizations.

The question to BPAC in 2003, was does current knowledge support the use of an AABB uniform abbreviated donor history questionnaire as an alternative to the current donor screening process for appropriately selected donors? And the outcome was that the BPAC members voted 1 abstention, 2 no, 10 yes, and the non-voting member and industry representative voted yes.

So, why not accept the task force abbreviated questionnaire? At the time there was the use of one capture question for new medical events. And it was: since your last donation have you had any new medical problems, diagnosis, or treatments including vaccinations? And at that time Paul Beatty's group did do one-on-one cognitive interviews with four subjects. And one subject did not provide information on a significant medical event. And that particular individual was the only one that had a significant medical event.

So, albeit a small sample size, we were somewhat concerned that one out of four people in a capture question didn't provide the information that was trying to be captured. So, at that time we determined that some sort of assessment of the abbreviated questionnaire compared to the full length should be performed.

We certainly support measures to improve the predictive value of the donor screening process, including use of an abbreviated questionnaire. That is our primary goal as well. So, the AABB task force today, Debbie Kessler is going to propose a study design for assessment of two medical capture questions on the abbreviated questionnaire.

In addition, Mary Beth Bassett is here from Blood Systems. They were the second blood center that was approved for an abbreviated questionnaire, and she is going to update the committee and the audience on their experience with it.

So, the questions to BPAC today are: (1) does the committee agree that the proposed study design (exclusive of sample size) is adequate to reasonably demonstrate equivalence (or lack of equivalence) between the two capture questions on the abbreviated questionnaire, and the 17 specific questions on the full length questionnaire?

This is in fact 16 questions. It has been changed since I made my slide.

Two, and if yes, does the committee agree that the proposed sample size adequate? Three, if no, what alternative study design and/or sample size does the committee propose would be adequate?

Thank you.

DR. ALLEN: Dr. Schreiber.

DR. SCHREIBER: Sharyn, could you expand on what the one out of the four were deferred? You said it's a significant medical event. Can you tell me what was significant?

DR. ORTON: Debbie, are you going to address that? She's going to address that in her presentation.

DR. SCHREIBER: The other question I had is when we hear the ASCP proposal, are they using the same abbreviated form?

DR. ORTON: No, they have their own. And Mary Beth, are you going to address how similar it is? No.

MS. BASSETT: My name is Mary Beth Bassett from Blood Systems. I don't have in my presentation, an actual comparison between the abbreviated process that we use, and what the AABB is proposing. But we can talk to those similarities. I can let you know how many capture questions we have, what criteria we use for determining our frequent donors, and that kind of information after we hear from Debbie and my presentation.

DR. ALLEN: Let me just ask my question also, although it probably will be addressed in the later presentations. And that was in the four people, the one with the medical event or of the four that had the cognitive interviews, I assume that none of them also then had the full length questionnaire first. So, we don't know if the full length would have picked up the event.

DR. ORTON: No, we don't.

DR. ALLEN: So, there is that. It's hanging out there as it was something that was considered significant that came up on a cognitive interview that wasn't picked up.

DR. ORTON: I need to clarify too, Jim, in the cognitives, they go question by question, versus giving one questionnaire and then given another questionnaire. It is done somewhat differently than in real life the way the questionnaires are given.

DR. ALLEN: Other questions for Dr. Orton before we move onto the presentations? Thank you.

Our first full presentation descriptive of the new study will be by Ms. Debra Kessler, Donor History Task Force, and Director, Regional Services, New York Blood Center.

Agenda Item: III. Study Design for Abbreviated Uniform Donor History Questionnaire - B. Study Design - Debra Kessler, RN, MS, Donor History Task Force & Director, Regional Services, New York Blood Center

MS. KESSLER: Thank you for inviting me today.

I'm just going to give a brief overview of where we are today. As Sharyn mentioned, the process to design the full length donor history form in 2000. FDA asked the AABB to work on it. There was extensive collaboration between numerous stakeholders. We had blood centers, including independent blood centers, Red Cross, FDA members. There were experts in creating questionnaires. It was a very broad group of committee members. And a groundbreaking approach to design, and for the first time a blood screening document was systematically evaluated.

The old uniform donor history form, which was the previous one that AABB had was a very different document from the current donor history form. You notice we dropped the uniform part of it, just calling it donor history form. Questions were complex. There were compound questions. And it didn't follow any identifiable order of time when we were asking the questions.

The task force made many changes in the new DHQ. Eliminated a number of questions that were not required by either FDA or AABB. Local blood centers could decide to put some questions back if they felt it was important. We broke up compound questions. Pretty much every question asks about one thing. We simplified the language. And we restructured the format into time periods that begin at the present, and go backwards in time.

Changes were made with the input of these experts in the fields of questionnaire development and reading comprehension. We had focus groups to review the new questions, and individual cognitive interviews by NCHS were performed to validate that the questions elicited the information that we intended to get.

In June 2002, the Blood Products Advisory Committee unanimously endorsed the full length donor history questionnaire. And in April 2004, the FDA issued a draft guidance. In December 2004, there were questions to the docket about the draft guidance, and the committee submitted responses to these questions. And we had revisions, so that the donor history questionnaire is now Version 1.1, and revisions also to associated materials. And currently, the donor history questionnaire is in use in many collection sites.

So, the next steps, AABB will receive any feedback from FDA regarding the revisions and the new Version 1.1. We'll issue the revised materials from the AABB, and FDA should be ready soon to issue final guidance on the full length donor history questionnaire.

Just to let you know how it is being used so far, it is in use in at least 19 ABC blood centers. A survey was recently done, and currently in 2 Red Cross centers. An additional at least 28 ABC blood centers and the entire Red Cross system will go live with the full length donor history form in the next 12 months. So, a great number percentage of the country will be using the full length form.

Now to the abbreviated donor history questionnaire. Just in terms of what the response was from blood centers about whether or not they would like to use it, at least 39 ABC centers and the entire Red Cross system plan to use the abbreviated donor form when it's available.

Currently, all donors answer all questions at every donation, including non-repeatable events. The abbreviated questionnaire eliminates non-repeatable event questions; it identifies recent changes in health, travel and behavior; and it retains questions about risk-associated activities that might have changed since the last donation.

There is precedent. As Sharyn mentioned, there are two blood centers that have approved abbreviated donor questionnaires. Mary Beth is going to talk about Blood Systems. But their model is exactly the kind of model that we are proposing for the AABB form. They do have a medical capture question that eliminates a number of medical questions on the full length form, and Mary Beth can explain more about that to you.

What we hope to accomplish is that the safety and efficacy is equivalent to the full length form; that donor satisfaction is increased for frequent blood donors; and that maybe, and nobody knows if this would happen, there would be an increase in the availability of blood, because people would feel that the process is simpler.

The abbreviated form is based on the full length form, and the full form is, as you know, a validated document. The testing of the abbreviated capture approach for recent changes in health status was performed by NCHS on four subjects in one-to-one cognitive interviews. And at that time there was only one capture question. This is what you were asking about before. Since your last donation have you had any new medical problems, diagnoses or treatments including vaccinations?

The capture questions in the cognitive interviews were performed only four people due to the lack of resources, but we wanted to get some testing done. And while this was not a full, large scale test, the preliminary findings generated optimism that the approach was worth pursuing.

Now, the event that wasn't reported was a miscarriage. The question, pregnancy in the last six weeks, is not one of the questions that is being rolled into capture questions. It is asked on the abbreviated form, as well as the full length form. The person who answered the question could have just as easily missed it on the full length form as the abbreviated form. So, while that was a concern, it wasn't exactly the issue that was captured in this interview process.

We want to have equivalent safety to the full length form. We want to increase the donor focus on recent events and behaviors, instead of directing their attention to things that have happened long ago. And as you can see, some people you may know, Alan Williams and Sharyn Orton, have said in writing that, "behavioral risk screening needs to be optimized to query donors about factors that best predict recent infection."

Again, with equivalency of safety, we want to increase the donor attention by reducing the time it takes to complete the screening. Whether or not this will increase the availability of blood, we don't know. Maybe these factors would influence that, increase donor satisfaction, decrease phenomenon of lapsed donors, and increase frequency of donation. That's yet to be seen.

Now, which donors will be eligible for using the abbreviated form? The definition of a frequent donor is someone who has successful completed the full length form at least twice, so they have seen all the questions at least twice, and that the next donation after, or at least one of the donations are within the previous six months. So, they have been exposed to all of the questions the last time within the last six months.

The donor will answer all the questions. Anecdotal evidence indicates that donors may remember information at the second donation that they didn't remember at the first, however, this is less likely to occur after a second experience with the questions.

Now, in the development of the abbreviated form, again, the full length validated form was used as the basis. We did question by question consideration. We deleted questions for a single risk in the past. For example, from 1980-1996 were you a member of the military, civilian military employee, et cetera? So, once they have answered that, that time period is gone now. So, we eliminated questions like that.

We consolidated travel questions into a single capture question about recent travel, consolidated medical questions by use of two capture questions. After the cognitive interviews we went from one to two separate questions. The HIV and hepatitis risk questions remain intact. Those will not change.

As you may remember, in December 2003, as Sharyn mentioned, the question was put to BPAC about whether the current knowledge supported the use of the AABB abbreviated form. And the vote was pretty strongly yes, that we should be able to move ahead with it.

Nevertheless, here we are. FDA requested that the task force develop some kind of pre-implementation study to compare the full length and abbreviated form as related to the medical capture questions. And this was based on that concern with the cognitive interviews where the pregnancy was not brought up. But again, maybe it wouldn't have been anyway.

So, the abbreviated donor history form study. A subgroup of AABB Donor History Task Force met and we had representatives from AABB. We had FDA on the committee, NCHS, and various blood centers. Actually, we had CDC on the committee too, I just recalled.

The study objectives are to determine whether the two medical capture questions identify deferable risks as well as the full length form, compare responses to the two medical questions with responses to the questions they cover in the full length questionnaire, and identify reasons for discrepancies between the answer provided to the capture questions, and the full length questionnaire using cognitive interviews. That is our main source of analysis is cognitive interviews.

The two medical capture questions are since your last donation, have you had any new medical problems or diagnoses, and had any new medical treatments?

This is the list of the questions from the full length form that are being rolled into these capture questions. We're asking if they are on an antibiotic, if they are taking any other kind of medication for an infection. We're asking in the past 12 months have you had a blood transfusion, transplant, graft, been treated for syphilis or gonorrhea?

And have you ever had a positive test for HIV, and basically the risk questions for infectious disease, dura matter transplant, cancer, problems with your heart or lungs, bleeding condition or blood disease? Now, again, those are the questions from the full length form that will be rolled into those two capture questions.

The study sites where the blood centers have volunteered to run the study are the New York Blood Center, Hoxworth Blood Center, Gulf Coast Regional Blood Center, Mississippi Valley Regional Blood Center, and the Red Cross, Lewis and Clark Region. The centers were selected to represent a mix of the US geographic areas.

Two hundred and fifty donors from each center who meet the definition of the frequent donor that I mentioned to you before, the sample total would be 1,250. And the final sample will include donors with the full range of the 6 month interval from their last donations. So, we are not going to just take apheresis donors. We are going to take whole blood donors also.

The process is that the donor will be enrolled by designated blood center staff, and given informed consent. The donor will fill out the abbreviated form first, return it to study personnel, then they will fill out the standard full length form. The study personnel will evaluate the two documents for discordant responses.

During the study the donor's eligibility to donate will be based on both questionnaires. So, if information shows up on one, but not the other, it would be included in the evaluation of their eligibility that day. If discordant responses are found, the donor will undergo cognitive interview either on-site right at the time, or as soon as possible, which we have said should be within three or four days by telephone. And the interviews will be audiotaped.

Each center will provide two staff members for cognitive interview training by the NCHS staff. Blood center staff will be selected based on those with skills that lend themselves to cognitive interviewing such as donor counselors that do HIV counseling, et cetera.

The data evaluation. The audiotapes will be transcribed, and the transcriptions, and if needed by NCHS, the audiotape will be evaluated by NCHS. The demographic and discordant responses will be entered into a database for evaluation.

Our timeline, May 3, we have a tentative date for scheduling the cognitive training of blood center staff. June through December we will execute the protocol at the blood centers. In January 2006, there will be transcription of the cognitive interviews. February/March 2006, NCHS will do the analysis of the cognitive interviews. And May 2006, we would hope to have submission of results to FDA.

Now, we want to focus on the fact that cognitive interviewing is really our main evaluation tool. It's understanding how people use the capture questions, and whether it works as well as the full length form.

The cost burden, the cost to AABB, there will be interviewer travel for training, interviewer training and analysis of the cognitive interviews by NCHS will be paid by AABB. They will need to have a study administrator, and do data entry. For the blood center there is study personnel time and materials. This is an unfunded study, so this is all voluntary use of funds by the centers and by AABB.

In conclusion, the abbreviated form was developed at the request of the FDA. It received BPAC endorsement in 2003. At the request of FDA the task force has developed a study for the abbreviated form. This is not intended to be an epidemiologic study. The emphasis for evaluation is on the cognitive interviews, and post-implementation evaluation will additional data.

And if I can also say that we have, and we'll hear from Mary Beth about Blood Systems, in a way we have kind of piloted the whole concept, because Blood Systems has been doing this for quite a while. So, we see from what Mary Beth will tell us, how it's working so far.

Thank you very much.

DR. ALLEN: Thank you for the overview.

A quick question. On the abbreviated questionnaire capture questions, if somebody responds yes, there has been a medical event or whatever, the process then would be to do what? Administer the full length questionnaire? Have the data collection person interview them to find out what it was, and make a decision as would be done for the full length questionnaire? How is done?

MS. KESSLER: During the study they will fill out the abbreviated form, and then fill out the full length before any evaluation is made. Let's say we were implementing, and we just had the abbreviated form. The health historian would be required to follow-up on what was the medical condition to determine eligibility, as they do with many questions in the current full length form.

You may not remember, but we have support documents for the abbreviated questionnaire including flow charts for the questions like we have for the full length questionnaire as support material for the blood center staff.

DR. ALLEN: And I thought that was all very complete. I guess my question was really much more assuming that it's accepted at some point and implemented, what would be the process once the capture question is flagged as a yes?

MS. KESSLER: There would have to be follow-up interview by the health historian to determine what was the condition, when was it, and would they be eligible or not.

DR. ALLEN: Other questions or discussion?

DR. DOPPELT: In regards to the discordant responses between one form and the other, are the cognitive interviewers just asking additional medical information? Or are they specifically asking why is it that you answered this question this way and that question that way?

MS. KESSLER: Yes, it's the latter. They will be asking what was their understanding of the one question, versus the full length question. It will be similar to the way NCHS did and does cognitive interviews for evaluating answers to all questions. It was how do people understand the questions, and what motivated them to answer one way or another.

DR. EPSTEIN: I have, Debbie, a question about the questions, the capture questions. The first is had any new medical problems or diagnoses. And I'm wondering how that language was selected, and whether the alternative of had any new health problems or medical diagnoses would be more effective. Because if you ask if someone had a medical problem, you are asking them to make a judgment about what happened to their health, whether it was medical. And health I think is a little bit broader. Now, I'm not a questionnaire expert, but the issue is how did this particular language get selected?

MS. KESSLER: I don't really that saying health problems was really discussed. It was always focused on medical problems or diagnoses. But we can take that suggestion under advisement.

DR. DI MICHELE: I just want to first of all say that this is a major undertaking, and I really do congratulate you on taking on this study. And I do certainly understand why it's important to do so. I also do not want to, in my questions that follow, do not want to do what the PPTA is going to say that we can't make the perfect be the enemy of the good. And I'm not going to suggest that as well.

But just with respect to study design, I have a couple of questions. And is this an appropriate forum to ask those questions?

DR. ALLEN: I think let's ask for clarification now, and we'll have a chance as a committee to discuss it. I would rather get through the other presentations first. George Schreiber has been asked to look at it, and he has got a brief overview. So, if we want to get into extensive discussion of study design, I would prefer that we delay that just a little bit.

DR. DI MICHELE: Okay, then I'll just ask some general questions then. When the interviewer goes to do let's say the telephone interview, how will the confidentiality of the subject be maintained?

MS. KESSLER: The donor has signed consent saying that they would be interviewed, possibly over the telephone. How the tapes would be handled, only study personnel could listen to them, and the usual confidentiality safeguards are put in place. But they are told up front.

DR. DI MICHELE: I saw that consent. But like the telephone interviewer calls and says may I speak to?

MS. KESSLER: Right, to the name of the donor.

DR. DI MICHELE: So, they will have the name of the donor.

MS. KESSLER: Oh, yes.

DR. DI MICHELE: It won't be just the study number.

MS. KESSLER: No, they will have the name of the donor.

DR. DI MICHELE: One other quick question. Will the method of recruitment allow for all age demographics to be included, so that you can look at the younger donors who is answering these questions, versus the older donors?

MS. KESSLER: Yes, we really intend to get as broad a representation of donors as we can.

MS. BASSETT: Mary Beth Bassett from Blood Systems.

I just wanted to let the committee know in answer to your question the medical capture question that Blood Systems has been using says have you had any new medical conditions or health problems?

DR. ALLEN: Other questions? Yes, Ms. Baker.

MS. BAKER: Thank you.

As a repeat donor, actually the first time I was ever confronted with the possibility of donation, I walked out after seeing the long form, and being very intimidated by it, although I have a great deal of survey experience myself, because I found it very awkward, no coherence, and not somebody who is squeamish or unexposed to the medical profession for over 20 years. But I have since become a repeat donor, and I do see that the thought of an abbreviated questionnaire would be terrific.

A couple of questions. For this particular study would eligibility require people who have telephones?

MS. KESSLER: Not necessarily, because they could be interviewed on-site if the blood center is set up to do it.

MS. BAKER: And in case you cannot interview them on-site, what is your timeframe for the telephone? Is it three days, a week?

MS. KESSLER: Three days is our intent.

MS. BAKER: But you don't have a firm cut off for a week? I'm thinking about donor recall.

MS. KESSLER: Three days is the intent, but we didn't want to be completely inflexible, so it's as soon as possible.

MS. BAKER: And primary English speakers? If English is not your primary language, have you given some thought to eligibility or non-eligibility?

MS. KESSLER: That would also be something that the local blood center would determine that they could do, some blood centers where they might have Spanish as an available donor history form, could do it in Spanish as well.

MS. BAKER: And did the committee look at low literacy? And having this go through a screening for low literacy?

MS. KESSLER: That was thought about when we developed even the full length form. And in that case, if for any reason a donor doesn't comprehend, because they can't read, because their language skills are bad, the blood center can administer the form verbally. So, the health historian can just read the questions to the donor. That can be done with the abbreviated or the full length form, and that is part of the user's guide and the process of how either form would be used.

MS. BAKER: Great, thank you.

DR. ALLEN: But what you raise is certainly a very important question in general to the whole donation process of capturing information correctly, whether it is full length or abbreviated.

MS. BAKER: Are these questionnaires that we are proposing to use, are they not standard throughout all blood collect sites throughout the US?

DR. ALLEN: No, each blood collection center that has a license, and the Red Cross for example has one license for all of its centers, each license holder has its own version, and it may or may not be part of the uniform or other. But they are all reviewed and approved by the FDA.

DR. LEITMAN: This is a suggestion. We all grapple with the best phrasing for medical capture questions, that most likely to make the donor remember something that happened in an interim period. And what we use, which is extremely broad, and just as a suggestion is since your last donation, have you been under a doctor's care? And sometimes that helps jog memory. It sounds like you have validated these two also, but if again something comes up in your next study, that tends to be more of a broad question.

MS. KESSLER: Thank you.

DR. LEW: I don't think there is one perfect one, because I can tell you in our medical interviews that we do for every single patient, we also ask have you ever seen the doctor, or had any medical care? And then later we go on for the full review of systems dot by dot. And I can you there is no perfect question. Sometimes they will say yes on one or no, and then later they are all positive. You have just got to do your best.

MS. BAKER: One more question. Can you comment on what would be the trigger for assuming this all would pass, and we have an abbreviated questionnaire, who would decide when the abbreviated questionnaire would be implemented? For example, if you don't just donate blood at one site, would it be the blood donor site? If you only went to one blood donor site, then perhaps they would have a record and a date to validate that this is the six month window. We could approach this person with the abbreviated questionnaire. But if you are not going to the same site for donation, who would determine, oh, yes, I'm within my six month window?

DR. ALLEN: Well, the six month window would have to be within a given system. And as Ms. Bassett will tell us in a little bit for example, you can go to any one of the Blood Systems sites and the information is available online. Now, if you go to another center operated by a different blood collection center, then it's a whole different system. So, it would be dependent on being within the same system.

Dr. Orton, do you want to address the question of the FDA process for approval?

DR. ORTON: We have actually a checklist. Any of the questions or information that are required by FDA, we have a checklist for. If any blood center chooses to use the current AABB donor history questionnaire, and that's in recipe that they are including that information, then we make sure that that they adhere to it. But how exactly questions are worded, if we have at FDA, required some wording, what order the questions are in, we don't give information for that.

So, our checklist is basically are the items asked, not anything to do with whether it's a compound question. Some places may have three or four questions, others have compounded to one. We do not make any comment on that.

DR. ALLEN: Other questions?

DR. SCHREIBER: Debbie, the way it's structured now with the capture questions, it's conceivable that the abbreviated one could pick up conditions that you are not interested in excluding people as donors. For example, I tried it out on a couple of people, and it picked up pap smear, mammography in those capture, and colonoscopy. But these are not questions that are on the full length.

So, are you going to do some kind of pre-screening to try to get rid of those people, and not reflux them into the cognitive interview before? It would seem that it's very non-specific, and you could wind up having a lot of people that you are going to do these interviews in that you are going to get any yield from.

MS. KESSLER: I don't think we have actually thought about that. I think that's something that we should think about before we go with the study. But that's a good suggestion.

DR. ALLEN: I'm glad you brought that up, because that was actually my major concern about the study design, was that the capture question -- the Venn diagram has a lot of areas of non-overlap, and certainly within the process of making a decision of whether or not that person is eligible for a donor, that can be screened out subsequently during the history taking process.

But the interview process, that's another one that does need to be clarified. It may be a point that we would want to discuss as a committee and make a recommendation on to the AABB, I don't know.

DR. SCHREIBER: One other question. I have to hark back to those famous four. How were they selected to start with? Where did they come from?

MS. KESSLER: They were blood donors selected by NCHS.

DR. SCHREIBER: You just randomly pulled four?

MS. KESSLER: They were people who fit the criteria. They had donated blood twice with a full length form, and the last donation was within six months.

DR. HARVATH: I'm just curious. Did you say that you would always have the donor answer the abbreviated history question first, and then the full length later? And I wondered if there was any discussion about considering randomizing of giving the -- randomizing between some people addressing the full length first, and the abbreviated second? I'm just curious about that.

MS. KESSLER: No, because we want to see if the full length captures all of those conditions. So, we didn't want to trigger somebody by seeing the full length form first, and then having that contaminate the abbreviated form.

DR. ALLEN: Okay, all of the considerations. Thank you very much. And I'm sure that we will be asking you further questions later.

Our next presentation is experience with the abbreviated donor history questionnaire at Blood Systems, Mary Beth Bassett, Vice President of Quality Assurance and Regulatory Affairs.

Agenda Item: III. Study Design for Abbreviated Uniform Donor History Questionnaire - C. Experience with an Abbreviated Donor History Questionnaire - Mary Beth Bassett, BS, Medical Technology (ASCP) and Microbiology, Vice President of Quality Assurance & Regulatory Affairs, Blood Systems, Inc.

MS. BASSETT: Thank you for inviting me to speak on Blood Systems' continued success with the abbreviated donor history questionnaire process. Just a little bit about Blood Systems. We are the parent company of United Blood Services. We operate 18 blood centers in 12 states. We collect about 1.1 million units of red blood cells, and about 120,000 units of platelet pheresis on an annual basis. We have two consolidated testing laboratories and a research institute in San Francisco, and we are the nation's second largest blood collection agency.

We believe that the use of the abbreviated questionnaire is important, and has benefitted our donors and our organization. We began implementation of the abbreviated process in August 2003, so the experience and the data that I'm going to share with you today is from that point, current.

In the interest of my time allotted, there are some slides that you will have in your handout that I'm not going to speak directly to today.

In 2003, I presented to BPAC, and that presentation included why Blood Systems had implemented the ADHQ process. I told you the reasons were to improve customer satisfaction; increase donor retention; increase our donation frequency; and increase blood availability. And those still are our reasons today.

I reported on our policy for repeat, frequent donations, and that is three allogenic donations previous, and one of those has to be in the last six months.

I have data to provide for you for three months of data related to medical history deferrals, temporary deferrals, biomarker deferrals, post-donation information, and our error rate. And that information I will update for you for current information.

And lastly, I also presented some information from a donor survey. So, today, I'm going to talk again about a donor satisfaction survey that we recently conducted. I'll compare some of the results from our abbreviated, and our full questionnaire, and provide you some new data that we have on post-donation information, and on apheresis platelet donors.

The first thing I want to talk about is the donor satisfaction survey. This was a survey that we conducted in fourth quarter of 2004. It really was a study and a research that we had done that was part of a more general survey that we had conducted about the whole donation process. But out of that survey, 750 or 42 percent of those that we talked to actually responded to the question, if you could change one thing about your donation process, what would it be?

This chart really illustrates what the answer to that question is, and it really shows the desire of donors for a more efficient, less time consuming process. You can see that 82 percent of the time the donors were asking for some kind of change in the donor process. Fifty-five percent wanted it shortened; 27 percent asked for it to be speeded up. So, this data really does point us to having a need for blood centers to find a safe way to streamline the donation process.

Now, to just confirm for ourselves that our abbreviated process was answering some of these concerns from our donors, we actually conducted another survey in March 2005, so this month. This survey though, was conducted with people that were eligible for the abbreviated questionnaire.

Our objectives were to determine if the donors are more satisfied and more likely to donate in the future if they received the abbreviated questionnaire. We completed 250 interviews for those donors who were eligible and actually received the abbreviated questionnaire, and also 250 were interviewed for those that were eligible, but received the full questionnaire. Both groups got the same interview, and the results are statistically significant.

These are the results. Donors from both of the groups were asked to rate the person that greeted them. They were asked to rate the interview process, the phlebotomist, and their overall experience. And these were asked by four different questions. A rating of 5 was considered excellent, and this chart only focuses on the excellent responses. And as you can see, the donors who actually got the abbreviated questionnaire responded had more excellent responses than those that received the full questionnaire.

This really parallels the data that I presented last time to the BPAC, and it is relatively interesting, because the only thing that really changed in the process was shortening the interview. But for donors who got the abbreviated questionnaire, they rated everything as better.

Since we were doing a survey, we wanted to try to get information about what the commitment the donors might have to future donations. And we asked the question, do you plan to donate blood in the coming months? As you imagine, these are really very experienced, repeat donors, and so most of them did indicate that they would donate again.

However, there was one thing that was relatively significant. Five percent of the donors receiving the full questionnaire were not sure that they would donate again. And for those that got the abbreviated questionnaire, only 1 percent were not sure.

Now, I would like to take us into the data that we have collected. This is an update of information from the last time I presented at BPAC. We actually analyzed data from 3 time periods, and there are 20 day time periods. The first started in October 2003, ending November 20, 2004.

The number of donors that presented and opportunities we had to review the records you can see ranged from 53,000 up to 57,000. Our donor eligibility, those that are able to have the ADHQ, remained pretty consistent at 32-33 percent. But you can see our actual usage, that means those that were actually eligible for the questionnaire and got the questionnaire increased from 48 percent to nearly 76 percent.

This table is really a summary of all of those three time periods that I just talked about. What we have here is a total number of 165,786 donors. So, these are the records we had an opportunity to review. You can see 42,644 of those are first time donors, so they are not eligible for the ADHQ. Our repeat donors were 69,000 and those were repeat donors that also were not eligible. And then those that were eligible for the abbreviated process was 53,996.

The next line down is medical history deferrals. You can see here first time donors, we had a medical history deferral rate of 5.64; repeat donors not eligible is 2.2, and medical deferrals for those that were eligible for ADHQ was 0.50. Now, this did not change from the last time I reported to you, and really isn't probably surprising to most of us.

This next line, temporary deferrals is really deferrals that are related to blood, pressure, temperature, plus, hematocrit, and really doesn't influence the abbreviated donation process, but it does continue to reflect that even though the donors are part of the abbreviated process, they have been frequently screened, and therefore they are less likely to have health issues in subsequent donations.

You can see here we had 14 percent first time donors that had temporary deferrals. We cut that in half at our eligible for ADHQ to 7 percent.

Confirmed positives shows similar results, 0.92, 0.13, down to 0.02 for those that are eligible for the abbreviated process. That is about 40 times lower for abbreviated donors.

This final row is post-donation information. You can see that that information stays pretty consistent: 0.15, 0.19, 0.15 here. So, there isn't a lot of difference related to post-donation information. Shortly, I've got some additional information around post-donation information that I will share with you.

This chart just takes a closer look at the donors that were eligible for the ADHQ. Remember, that number was 53,996. You can see that out of that, 32,000 or 60 percent actually got the abbreviated questionnaire, and 40 percent received the full questionnaire.

Again, moving from right to left, you can see that the rates of medical and temporary deferrals in the donors receiving the full questionnaire is slightly higher, but all of the numbers between both groups remain very close and very low. There is very little difference between the confirmed positive and the post-donation information.

In one of our previous discussions we had the concern that maybe the platelet pheresis were influencing some of the rates in these groups. So, we looked at that, although our size is really quite small. The number was again, 53,996, and we only had eligible for the ADHQ that donations were less than 8 weeks of 8 percent or 4,274.

So, the platelet donors are a much smaller portion of our data set. And their deferrals rates are pretty similar. We don't think there is really an impact on the overall documents or the overall rates. Again, the rates of the medical history and temporary deferrals are low. The difference in the medical history and deferral rates here though you can see is higher. There really isn't much difference going on between these.

This next information is around the post-donation information. We actually took the November 1 through the 20th timeframe and looked at all of the post-donation information that we had received. And that was a total of 120. So, this is not just for abbreviated or just for full. It's all of the post-donation information that we received in that timeframe.

What you can see here is 50 percent of the post-donation information we got from donor call back; 34 percent we got from donor deferral or information that the donor provided us on a subsequent donation; 11 percent came from telerecruitment; and 5 percent were generated from third party.

Now, we wanted to ask ourselves did the donor know this information at the time that they were giving their health history? And we found yes to that. So, there were 37 out of that 120 where the donor really knew that information. This 77, they couldn't have possibly known. They got flu or sick or some kind of an illness after the fact.

So, if you look at those 37, the majority of the cases, you can see 24 here out of the 37 are related to travel or resident risk either for malaria or new variant CJD; 4 cases were related to medicine, and 3 were related to medical condition.

Then we wanted to take the analysis one step further. We wanted to ask ourselves how many times might a donor withhold deferring information? And is either the abbreviated or the full questionnaire clearly better at discovering deferring information? And this is what we found.

Twenty-four donors out of that 37 didn't provide information on one occasion; 21 of those 24 didn't provide that information using the full questionnaire; 3 used the abbreviated questionnaire. We had 4 donors that didn't provide that information on 2 occasions; 3 of those donors were screened twice with the full questionnaire, 1 was screened once with the full questionnaire, and one with the abbreviated questionnaire.

We had 4 donors that didn't provide the information on three occasions; 3 of those were screened 3 times on the full; 1 was screened 2 times on the full; and 1 on the abbreviated.

So, this slide just shows more of the same. You can see that we have 5-6 times that donors were not providing us the information. You can see that just on this slide, the full questionnaire failed 13 times, and the abbreviated questionnaire failed 3 times. So, really what does this mean? What is this telling us?

This doesn't tell us definitively which questionnaire is better at discovering the deferring information, but it does tell us something that we all know. Not even the full questionnaire is 100 percent at getting deferring information from our donors.

This is just to provide you the error rate from the time that we implemented ADHQ until fourth quarter of 2004, and you can see the learning curve here. We had an error rate of 0.23 percent; fourth quarter of 2004, our error rate is 0.008 percent.

So, in conclusion, our experience to date is that donors want a shorter process, and they like our abbreviated questionnaire. One-third of our donors are eligible, and three-quarters of those donors are actually getting the abbreviated questionnaire.

There doesn't appear to be data that indicates the abbreviated process increases risk. It doesn't appear that platelet donors are biasing our analysis. The analysis of PDI shows that donors withhold information during both the full questionnaire and the abbreviated process. And our root cause and implementation has really reduced our error rate.

Just a thank you to my collaborators who helped me put this presentation together and really analyze the data. And my thanks to all of you for letting me share our experience.

DR. ALLEN: Thank you very much.

I've got a couple of questions, Mary Beth. First, when you talk about withholding information from either the full length or the abbreviated questionnaire, for example with variant CJD, what you are talking about is somebody who says, well, yes, if you add it all up, I was in England for a total of eight months, and I forgot about it.

MS. BASSETT: Right, it's those kinds of things that are difficult for the donors to actually remember and calculate.

DR. ALLEN: And second question, when you have got a group of donors that are eligible to receive the abbreviated questionnaire, why do only 60 or 70 percent get it, and not closer to 100 percent?

MS. BASSETT: It's because of the implementation. It's not fully implemented at all of the centers, and the rate of implementation may be more at fixed sites than it is at mobiles. So, we do have some population of our donors that are eligible, but not getting it just because we are not offering it yet.

DR. ALLEN: Okay, fine. Dr. Lew.

DR. LEW: Two questions. One is you showed kind of striking information, except I don't have the denominator, unless you mentioned it, and I didn't catch it. When you mentioned how many times the donor may not reveal deferring information, you point out more persons who took the full form in terms of absolute numbers left out information, compared to those who took the abbreviated form. But I don't know how many people got the full form, what percentage of those.

MS. BASSETT: What are you looking at?

DR. LEW: You last couple of slides, when you talk about how many times a donor may not reveal deferring information. And you say 21 of those who took the full form, versus 3 who took the abbreviated questionnaire did not reveal certain information that you got back I guess on follow-up interviews. Are you getting what I'm saying?

DR. ALLEN: There isn't a single denominator figure, if I understand.

DR. LEW: But then this is meaningless. What I'm trying to point out is that if you gave 5 million people the full form, and only 21 on follow-up made a mistake, and then you gave 5 people the abbreviated questionnaire, and 3 out of the 5 made a mistake, that is a huge difference. Without a denominator, I don't know how to interpret these two slides.

DR. ALLEN: I basically agree with you, but let me I think explain what happened in this instance. If somebody has been coming in as a repeat donor, and received the full length questionnaire let's say four or five times, and then the abbreviated questionnaire was put in, and maybe the second time they used the abbreviated questionnaire they recalled, gee, I forgot about the time I was a graduate student over in England. Did that really count? Yes, I guess it was 1980, so I forgot about it.

What they did was to then go back. All right, the information came up here at the second time of the abbreviated, so they went back and then looked at all the prior donations, and even though the questions were asked, none of those prior ones elicited it. So, it's not a formal study. It is basically an experience analysis, where suddenly you get discrepant information. And I think that --

DR. LEW: I guess I would have preferred just to say it's clear even when you use the full form people make mistakes, just as when you use an abbreviated form.

MS. BASSETT: That's correct.

DR. LEW: I know, but the way it was presented, it wasn't quite that way. The other question I had was you did show that when you did the comparison between the abbreviated and full form, there was just a slight, but statistically significant increase in medical deferrals with the full form. Did you all follow-up what those medical deferrals were?

MS. BASSETT: Dr. Kamal, can you respond to that?

DR. KAMAL: Hani(?) Kamal from Blood Systems. No, we don't have a breakdown to what questions these medical deferrals were for.

DR. DI MICHELE: Actually, my question was about the same thing, except the way I interpreted it, and this is what I'm asking for clarification on is that with the full length form, there were more initial deferrals; deferrals on the basis of how the questionnaire was answered with the full length form than with the abbreviated form. But in terms of ultimate confirmed positives, or confirmed deferrals, they were not statistically significant. Is that the way I should interpret the data on those slides?

MS. BASSETT: Yes.

DR. LEITMAN: A couple of questions. On the abbreviated form the only travel question is 13, been outside the US or Canada since your last donation. So, if the most common reason for forgetting to reveal deferrable risk was a remote travel history to UK countries prior to 1996, or residence on a military base prior to 1996, that's not going to be brought out if all subsequent histories are taken through the abbreviated donor history questionnaire. That jogging of the memory will only come up on the full donor history questionnaire.

MS. BASSETT: Are you talking about our travel capture question?

DR. LEITMAN: The abbreviated donor history travel question. Oh, I'm sorry. Never mind, it's a different one.

Also, 0.25 percent error in implementing. That's interesting. Do you know what the source is of the error?

MS. BASSETT: Yes, at that point in time when we first implemented, we were using a manual process. It was computerized, but it had a lot of manual actions to it. You had to go into the computer on one screen to try to find your three allogenic donations. And so, if you made a mistake, what you were doing was finding somebody that maybe had done an atallogous(?) and you counted those.

And also, the six month timeframe might have been missed. Many times what we saw is they missed the year. They thought it was six months, and they didn't look at the current year. So, we knew that we had an issue going on. And how we changed that, we put corrective action in place, and now our computerized system actually when they put in the donor ID, it will pop up on the screen, and it says whether or not that person is eligible or not. And then we have double blind entry to make sure we have entered it correctly as well. So, that's why our error rate has gone down.

We still have a few errors, probably only 2-3. And that's just manual entry kinds of issues, or not paying attention to the screen.

DR. EPSTEIN: Mary Beth, I would like to come back to the point that was raised earlier about medical deferrals. These are slides 12 and 13. When analyzed eligibles, either all eligibles, or those who had donated with 8 weeks.

I'm struck by the fact that the difference in rate of medical history deferrals is really quite remarkable. For the cohort overall it's about a third higher with the full length questionnaire. And when you look at frequent donors, presumably mainly platelet donors, it is four fold higher. And both those findings are statistically significant.

So, whereas you have pointed out that the confirmed positive rates aren't statistically significant, without analyzing what the deferrals were, I think it's very hard to dismiss that as an unimportant finding. It's exactly the opposite of what we had hoped for. I mean to say that those aren't significant deferrals because they are not reflected in positive infectious disease tests overlooks the fact that we may have reasons to ask those questions. We are worried about false negative infectious disease tests, as well as other conditions for which we cannot test.

So, I find it disquieting, both because the magnitude of that effect seems large, and because at least at this forum you have not been able to tell us what those deferrals were. So, in brief, I think these are very disturbing findings.

MS. BASSETT: I think one of the real markers though, and maybe one of the most important markers for that kind of thing is around your confirmed positive rates. That's really one of your main things for sensitivity.

DR. ORTON: I think one of the other pieces to what Jay is saying is because some of those medical history questions that are rolled in donor safety issues. So, heart disease, lung disease, those wouldn't be found in the marker rates at all.

So, in fact if you were as an example, drawing people who had heart disease and weren't reporting it, and they were reporting it in a full length versus an abbreviated, you might see that. I'm not saying that's what was reflected here, but I think that's also another consideration, where the marker rates are not going to capture that safety information.

MS. BASSETT: That is correct. Dr. Kamal, do you have any information to share with that, since you were part of the group that did this analysis?

DR. KAMAL: Obviously, all people who were eligible, in the prior slides it shows only 0.5 percent. So, we are dealing with a very select group, with the lowest medical deferral, temporary deferral, and confirmed positive. The group pheresis or frequent platelet donors are only 8 percent of that group. And the number of deferrals here, we are talking about 13 medical deferrals divided among the 2 groups.

We are reporting what we have seen, and as Jay mentioned, unfortunately, we were not able -- we created this number by computer queries, counting numbers only. So, whether they were part of the capture question or not, we can't tell. We have about 20 out of the 35 questions are shared between the full questionnaire as well as the abbreviated questionnaire. So, among these groups there is truly no difference, or the questionnaire is not a factor.

And the only thing, we looked at information about the capture question. And as mentioned earlier, it is a wide net catching much information that we may not be interested in. We are getting acid reflux disease. We are getting hypertension. So, our capture questions, we looked at some information, and it seems like it's not a matter donor understanding or comprehension or sharing information. As a matter of fact, yes, we are getting more information that non-deferring and deferring conditions.

DR. KUEHNERT: I'm sorry if this was covered before. I had to step out. Maybe I missed it. But just as far as the slide comparing those eligible for the abbreviated questionnaire and then comparing those who were given the abbreviated versus the full length, obviously they weren't randomized. And so, what was the determination on whether they got the abbreviated versus full? Was that it that the people who got the full length occurred before implementation? What was the determination on who got what?

MS. BASSETT: The reason that we have a difference between those that are eligible for the abbreviated, and those that received full and not is just because of the implementation. Some of our centers in our organization that haven't done full implementation. So, we just have an automatic really kind of control group, because they were eligible, but they didn't have the process in place to be able to given the abbreviated questionnaire.

DR. KUEHNERT: So, is there a difference in geographic region or demographics between the two groups?

MS. BASSETT: I don't think so.

DR. KUEHNERT: I was just trying to understand whether there might be other reasons for this difference in medical history deferral between the two groups.

MS. BASSETT: And we've got some implementation at all of our centers. It's just not full implementation at all of them.

DR. ALLEN: One might expect that in fact there is some difference. I would assume that generally in each center it's implemented at the main facility before it is on the mobile collection.

MS. BASSETT: That's absolutely correct.

DR. ALLEN: So, there may be rural/urban differences. There may be as well some geographic differences.

MS. BASSETT: Mobile versus fixed is definitely one of the rationales, because it's much easier to implement this at a fixed site, than it is at a mobile operation. So, those are some of the differences as well.

DR. LEW: Was there any effort to try to find out who actually returned when you did your comparison for full form versus abbreviated form? Because it would make sense if someone had an unpleasant experience, as you point out in your questionnaire, that they might be a little upset, so when they answer their questions, they say of course I'm not going to come back next time. Versus once they got home, they relaxed a little, that all goes away.

We all know polls aren't that accurate. So, is there any effort to try to capture who really does come back? Because that's the whole thing. We are trying to make it abbreviated so people come back. But if it really doesn't work, well, do you want to lose some important medical information?

MS. BASSETT: We are still doing some analysis on that donor retention and return rate from this data that we have.

DR. ALLEN: There is a difference between the Blood Systems history question collection process and the AABB process in that Blood Systems, Mary Beth, is all verbal questions and answers. There is no self-administered questionnaire.

MS. BASSETT: No self-administered.

DR. ALLEN: So, there is that difference in process.

Other questions or comments for Ms. Bassett? Thank you very much.

Agenda Item: Open Public Hearing

We will at this point move to our open public hearing. I'm aware of only one speaker who wishes to speak. I need to read the open public hearing announcement for general matters meetings.

Both the Food and Drug Administration, FDA, and the public believe in a transparent process for information gathering and decision-making. To insure 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's 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 relationships 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 company or a 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 financial relationships. If you choose not to address this issue of financial relationships at the beginning of your statement, it will not preclude you from speaking.

The one speaker that I have on the list for this morning is Ms. Mary Gustavson, Senior Director, Global Regulatory Policy of the Plasma Protein Therapeutics Association, PPTA. She is speaking on the AABB abbreviated questionnaire.

Ms. Gustavson.

MS. GUSTAVSON: Thank you, Dr. Allen and committee.

I am Mary Gustavson. I am a salaried employee of PPTA. Yesterday during the first open public hearing, the who, what, and why of us was presented, so I won't repeat that. It's in the written statement.

PPTA supports the implementation of an abbreviated donor history questionnaire for frequent donors. PPTA has been a participant in the AABB interorganizational task force to develop the uniform donor history questionnaire. The AABB task force developed the full length questionnaire and abbreviated questionnaire to be used in tandem to enhance the donor interview process.

PPTA formed a subgroup of the task force to develop questionnaires to be used to screen donors of source plasma. It was recognized that separate documents were needed, because of some of the differences in collection practices and donor screening requirements between whole blood donors and source plasma donors.

The PPTA subgroup used the AABB UDHQ and made modifications designed to elicit information relevant to source plasma donors. PPTA is currently reviewing FDA's comments to our submission, and plans to meet with FDA in the near future to discuss PPTA's plan for moving forward on this important initiative.

The abbreviated questionnaire is an important adjunct to the full length questionnaire for repeat and frequent donors. While no study may be perfect in design, scope, or size, PPTA encourages FDA to avoid making the perfect the enemy of the good. As all of us are aware, it has been the practice over the years to implement donor history questionnaires with no evaluation or studies. The current efforts of the blood and plasma communities to implement standardized questionnaires, including an abbreviated questionnaire for frequent donors should be encouraged and facilitated.

Thank you.

DR. ALLEN: Thank you very much.

Questions or comments for Ms. Gustavson? Thank you very much.

Are there any other speakers who wish to make comments during the open public hearing session? Hearing none, the open public hearing session is now closed. We will move to open committee discussion of the question. Dr. Schreiber, do you want to make a brief summary presentation about the abbreviated questionnaire and proposed study?

DR. SCHREIBER: I guess I was asked to make some comments. To read the questionnaires in a little bit more depth and comment. And these are strictly my thoughts related to no one else's perception of reality.

When I look at it, the problem that we have is that the previous version contained one medical history capture question. And the questions were then amended for test subjects in this one-to-one cognitive interviews to try to determine the adequacy as a substitute for the specific medical questions.

And they failed to identify this one case that we have heard about. If we had never tried those four, we probably wouldn't be here today. We would all be administering the abbreviated questionnaire.

So, the solution as we heard, was that the task force then split the question into two parts. And that is what we are now looking at is the evaluation of how those two questions are compared to the full length. And I have to say they have gone to a lot of problems, and have really spent a lot of time thinking through this, and I commend the task force for all the work they have done.

The issue is a study is proposed to "test how well the two capture questionnaires identify relevant donor eligibility information and repeat donors." And we are asked, is the design appropriate to answer this question? What does the proposed study hope to accomplish? Well, as Debbie clearly said, this is not an epidemiological study. So, are we expecting it to be big enough to be able to test these differences? My conclusion is no, and you'll see that.

It will study discordancies between the two instruments, the capture questions versus the 16 medical condition questions. And one of the things is that they are all different time periods. So, one of the questions is how well can you map one to the other in a study like this? The other thing that we have heard is that the idea is to reflux discordant donors to cognitive interviews, to try to understand why these differences exist. And as Debbie said, the emphasis for the evaluation is on cognitive interviews, and to conduct these cognitive interviews.

So, I think one of the things that we have to ask ourselves, are there going to be enough people that are discordant to be able to do enough cognitive interviews to make this important? And it's hard to look at this in terms of statistical design for a full blown epidemiological or randomized trial. And I think Debbie was very clear, that's not what the intent of the group was. And if that's the intent of what the FDA is going to require, or feels that's needed, I think that that has to be made right up front, so that they know what direction to go in.

I took in the 1,250 to try to see what I would expect as discordancies. And I said that there would probably be a 1 percent deferral rate on the full donor history questionnaire. As we saw from BCP, their rate was about 0.85, so the 1 percent is probably in the right ball park.

If that is true, then we would expect to see 13 people that were deferred. If we missed one-tenth of them by the HDAQ, we would have 1 discordant person out of the 1,250. If there are some other criteria that I'm missing, and there are other conditions in there at that are less sensitive to be screened for, maybe the discordancies would be higher, but those are probably things that we're not really interested in trying to evaluate anyway.

And that's why I was asking if there was going to be some kind of pre-screening so a lot of time isn't spent and wasted on conditions that are not trying to be judged in the two questionnaires. For example, pap smears or mammographies, colonoscopies, things that would not be deferral criteria on either of the questionnaires. But when I tried it on a small number of people, those three conditions came up repeatedly on the abbreviated questionnaire, that people answered yes to the screening questions.

So, then the question is, okay, how many do you need to have a meaningful sample size? And I just did a rough back of the envelope calculation. And I come up with about 6,200. The 6,200 would assume a deferral rate on the HDHQ of 1 percent, and that the ADHQ misses 10 percent of the deferrals. Even there, what you would find is that we are not going to get any that are deferred by the DHQ and not the ADHQ.

So, I weighted it so that we were assuming that we're not going to miss so many. And here we would we would have six, which would give us the statistical power. And kind of the rule of thumb that I'm always told by the statisticians is that you need five in a cell to be able to take a statistical difference. So, that what we are really comparing is the six versus the zero. And here, this would yield your six that you would have for your cognitive interviews.

This assumes that the HDHQ and the abbreviated one are equally sensitive. If they are equally sensitive, they would both miss the same number. And there is no pointer here, but in this case we would miss one on either, so we would have two that would be refluxed to the cognitive interviews. So, we are half as good now as we were with the four.

These are my expectations. The study is not designed to provide evidence of comparability. And as they said, really in the selection of sample size and the design, that's not what the intent of the design was. I would predict that discrepancies between the two instruments will be found. And before I saw the BPC data, I thought that the 17 questions would pick up more positives.

My other prediction is that whatever the findings are of this study, it will probably lead to another cycle of DHQ revisions. And then my question would be, are we going to be back here in a year, in 2006 designing another study, because we found that we had to do some rewording on the HDAQ because of discrepancies that were found in the limited number of subjects that were identified?

I think that the degree of discordancies will be on the order of less than a half of a percent. If you take that into account, we always look at the exceptions, but we have to look at and realize that there is a 99.5 percent concordance. If we figure that we're only going to have, as I had in my other calculations, a tenth of a percent, then we are going to be picking up 99.9 percent concordance between the two instruments. Which I think that when you are looking at this type of a survey or an instrument that has its problems that we all realize, that's a pretty good degree of concordance.

The sample size requirements to demonstrate statistical dependency or statistical discrepancy between the two instruments is on the order of 6,000. And I don't think that by any stretch of the imagination would the resources that are available to the AABB and this group, that an ambitious undertaking like that could be actually achieved. I don't think the resources are there to do that.

The ADHQ contains two general capture questions relating to time since the last donation. The DHQ instrument contains questions that have specific time periods, past 12 months, ever or current. With these differences, it's difficult to design a study to demonstrate equivalency or lack of equivalency of the two history forms.

So, I think in summary, we need to, in reviewing what's been put forward to us, keep clearly in mind of what the objectives are. And then I think in my mind, one of the most important questions is what degree of difference is willing to be accepted to be able to initiate the abbreviated donor history questionnaire in our screening population?

And I think now the community has been waiting for several years, and have been tackling this question. And if we do continue studies, at what end will be able to stop and actually go out and administer the questionnaire? And I'm not sure that we know that answer. We hear, as Jay raised from the BCP questionnaire and their administration, we find a small difference. The more people we administer the questionnaire to, the smaller that difference has to be to remain statistically significant.

So, if we administer that questionnaire or that test to 2 million, you probably have that a statistical difference of 0.01 percent would be highly significant. Here we saw that we had about three-tenths of a percent difference between the two questionnaires. Which also keep in mind that it's not the same questionnaire here, because this is a further revision.

But I think that one of the questions we just have to keep on asking ourselves is how far do we go? And where do we stop? What is FDA willing to accept in terms of an evaluation? So, that's my random thoughts on the question.

DR. ALLEN: Thank you, Dr. Schreiber. Questions or comments specifically for Dr. Schreiber?

DR. DI MICHELE: Is it possible for me to ask him some study design related questions?

DR. ALLEN: We will get into general discussion. I think right now let's focus on clarification or questions on his presentation or perspective. We're only a couple of minutes away.

DR. KUEHNERT: That was a nice presentation. I just wanted to ask you about the difference between -- you sort of distinguished between demonstrating equivalence and demonstrating discordancy. And so, what are the differences between those two concepts?

DR. SCHREIBER: Well, one, when you go into the design, you design it and do your power calculations that the two instruments are going to be the same. The other is that you design it expecting that there will be a difference. And you need bigger numbers to demonstrate equivalency than to you do to demonstrate differences.

DR. BIANCO: Celso Bianco, America's Blood Centers. This is just a comment that I think can contribute to the discussion. We are comparing the long donor history questionnaire to the abbreviated questionnaire or the other way around. But if we try to compare the abbreviated questionnaire to the abbreviated questionnaire, the long history to the long history, I'm sure that we would find the same degree of differences.

If we look at the RAD study and the surveys that were done, for instance the one that I recall very well from Alan Williams as the first author, found that 1.8 percent of the donors that received this survey revealed risks that they had not revealed at the time of the interview, at the time of donation; much higher than all the numbers that we are talking now about.

DR. LEW: I just wanted to comment that I agree with Dr. Schreiber that 17 questions to 1, you know you are going to find a difference, anyone who objectively can think about that. And to me, the bottom line is not oh look, there is a statistically significant difference, because there was, and you expect it.

But I guess getting back to the presentation, the two important distinctions that I wanted to hear was those who were missed, why were they missed specifically? Was it something very important? Versus how many more people did you get back to come to donate blood?

Those are the two important questions to me. And would the study that is being design potentially answer that? No, it won't address that. Maybe going back to it, then this study is no longer valid, I don't know. The key is that this survey was -- a lot of time and effort has been spent in developing it. It seems reasonable as it is now. I'm not the study that is going out will tell us.

DR. KATZ: I feel very strongly both ways. Celso's comment about the 1.9 percent undisclosed deferrable risk I think is very important, and at my blood center we have done some things attempting to reduce that, that we believe work, that have nothing to do with the length of the questionnaire, and that I consider to be probably more important for transfusion safety over the long haul, because they are really attempting to get at recent risks in a certain sense. But that's one aside.

There is another issue here that I think is fascinating. And that is in the ABC centers only 29 out of 75, 76, 77 whatever we are this week. There is enormous diversity already. Most of the centers are using completed unvalidated materials to screen their donors. We have recently moved to the long uniform donor history questionnaire. And our preliminary two months of data suggests very strongly that we are picking up stuff due to the nature of the questions, the ordering of the questions and time that we missed with the previously approved.

So, I want to consider in the absence of data, and recognizing that Blood Systems' abbreviated questionnaire was not designed and validated in that fashion, that it may be very difficult to prove equivalence, inferiority, superiority, but it almost certainly better than the way we have been doing things over the last long interval of time.

And I would just ask the question, is the study needed at all? And it is a question. As I disclosed, I'm one of the investigators. My center is one of them. I would love to do this study. I'm not sure at the end of it all, I did the same sample size calculations as Dr. Schreiber, so statistical significance is not going to be demonstrated with the resources available to execute this study. And a number of discordances in the sample size that we have proposed may well not give us the information we want, and as George said, we'll be back here.

DR. ALLEN: Thank you for the comments. I'm going to just make a couple of brief comments, and then ask Dr. Orton to come back and give us the FDA perspective and questions for the committee. We'll get into the broader discussion.

The abbreviated donor history questionnaire, which we all have in our packets actually has 30 questions on it, give or take one or two. I think I've got the most recent version. And it starts out, are you feeling healthy and well today? It's got the other specific questions with regard to have you taken aspirin, headache and fever, which we all recognize as being the West Nile virus question that probably is not that useful.

It's got a question for female donors about pregnancy, donation history, and 16 weeks. And then it comes to the two capture questions, any new medical problems or diagnoses, any new medical treatments? And then it gets into questions about medications, on the medical deferral list, and again, people have a chance to look at that list now, which I think is much preferable to having somebody reel off a list of questions or doing it in some other way.

It asks for travel outside the United States or Canada, the contact with blood, needle stick and HIV and AIDS questions, sexual history contact. All of that, Tab 2, ear or body piercing. It's not just the two capture questions standing out there alone. And these are donors that have donated at least a couple of times, the last time within 6 months, gone through the full questionnaire.

This is a select group of donors. I think we'll get into the discussion about the issues that were raised by Dr. Schreiber and Dr. Katz's last question of is this even needed at all, and what's the next process? We do, if we possibly can, want to move this whole process forward. And that's what our discussion today should be aimed at.

Dr. Orton, would you provide our summary, and then we'll get into the more in-depth discussion, and attempt to resolve the issue of the questions.

Agenda Item: III. Study Design for Abbreviated Uniform Donor History Questionnaire - C. FDA Perspective and Questions for the Committee

DR. ORTON: Yes, I want to stress the FDA as well wants to move forward in the appropriate path. I'm actually the liaison on the abbreviated task force, again as a regulatory liaison. We certainly have tried not to tell the task force what path they need to take. And so, any recommendations, we clearly want to hear.

If a study needs to be done, or if there is something in implementation measurements we can take that we think are adequate, we certainly want to hear them, analyze them, and then within the regulatory framework that gives us some restrictions, move forward. So, it's important to us, the opinions of the committee today.

So, the first question is does the committee agree that the proposed study design exclusive of sample size is adequate to reasonably demonstrate equivalence or lack of equivalence between the two capture questions on the abbreviated questionnaire, and the 16 specific questions on the full length questionnaire?

DR. ALLEN: Open for general discussion. Dr. DiMichele.

Agenda Item: III. Study Design for Abbreviated Uniform Donor History Questionnaire - E. Committee Discussion and Recommendations

DR. DI MICHELE: Well, if we are going to discuss the study design with respect to issues exclusive of sample size, there are two questions I had. The first was if we are using the abbreviated questionnaire as the test document, should the study design actually incorporate the use of the test document into ultimate donor deferral as part of the study measures, which it plans to do.

Because one would think -- again, in a strict study design, that you would use the approved questionnaire to actually make the deferral, and would not use the answers to the abbreviated questionnaire to help in making that deferral process, because I'm not sure that it would not confound the results of the study.

The second is that if the cognitive interview is the most important aspect of the study, and that's what I wanted to clarify with Dr. Schreiber, because I was thinking that if it is important, then although it's incredibly labor intensive, and I can see why it wouldn't be done this way, should the interview not be done with all participants, whether they have discordance or not, in order to ascertain understanding of the questions, if that's what is truly important? And I think those are my two major study design issues.

DR. ALLEN: Dr. Klein.

DR. KLEIN: Well, first of all, I would just like to remind the committee that this committee voted 10 to 2 with 1 abstention to go ahead in December. Having said that, I think, unless I'm missing something, I just heard an hour's discussion that this study wasn't designed to tell us about equivalence or non-equivalence. I think from the very beginning that's what I heard.

So, in some ways the question is not helpful to me. I thought the importance of the study was try to find out when there were discordant answers in a subjective way from experts, in doing so, what the perceptions were that made people answer differently. And I think that's all we are ever going to get out of the study.

Now, if we think that's important, obviously we should go ahead. But if we think that we'll get that subjective information and come back three months from now, and vote 11-1 with 1 abstention and still it won't go forward, then I think we are wasting a lot of time.

DR. ALLEN: Thank you.

Ms. Kessler stated this is not an epidemiologic study. Dr. Schreiber re-emphasized that point. So, to me it is much more of a descriptive study, if you will, to try to identify through the back-to-back administration of the two questionnaires with discrepancies defined by the cognitive interviews.

Is there concern that we can document that the abbreviated questionnaire is missing a significant number or significant types of medical-related issues that would otherwise defer this fairly select group? And to me, that's the real issue. And I fully appreciate we are asked to make our discussion now, absent discussion of study size.

But it seems to me that although I've got minor points about the study design in terms of the back-to-back administration of the questionnaires and that sort of thing, it seems to me that it's a useful, descriptive analysis to try to do. And I would think we ought to go ahead.

Dr. DiMichele.

DR. DI MICHELE: I was just going to ask, is there a difference between this not being an epidemiologic study, and this being a validation of an instrument study? That is what I'm thinking of it as, not an epidemiologic study, but an instrument validation study. And that is the basis of my comments. Is this not how we should proceed? Maybe we should get some clarification on that.

DR. KATZ: I think that's pretty accurate.

I just wanted to make one thing -- the study instrument, both instruments are the instruments of record. A deferrable answer on either one results in deferral of the donor. So, they are really double screened during the study. The reason that the ADHQ was first was so as not bias responses on the ADHQ by coaching off the long form. But we don't care about colonoscopy or pap smears. Those are not deferrable answers. They will come up on the ADHQ, and we'll say that's not deferrable. Those won't be discordant.

So, it's really pretty simple, and it's basically designed exclusive of sample size to see whatever discordance we dredge up, is there a way to address those by making the process better in subsequent iterations, would be my understanding.

DR. ALLEN: Yes, I think that's a very important distinction. Some of the questions as answered on the questionnaire probably would result in deferral in and of themselves. The medical screening questionnaires, they are capture questions that then result in a subsequent history interview process, that if this were implemented on a routine basis, it would still go through that same clarification process to try to elicit is it information that is important to make a decision about whether this person can donate today.

So, it's not the questionnaire itself that is going to result in the deferral, it's primarily a capture questionnaire.

DR. KLEIN: Jim, I just think that my frustration is the same one that I felt yesterday, that the question is kind of setting me up for the wrong answer. I look at the question, and we have just heard an hour's discussion saying the study wasn't designed to do that. So, how in the world can you say that the study is going to do that?

And we have all discussed what the value of the study might be or might not be. But to me, it doesn't seem to have any relevance to the question that we're being asked. If I'm missing something here, perhaps someone can explain it to me.

DR. ALLEN: The question, I agree with you, in order to reasonably demonstrate equivalence or lack of it, that would require an epidemiological study to do that. And yes, I concur with you on that point.

DR. KUEHNERT: I just wondered if in addition to this proposed study, if there is formal plans for evaluation of the questionnaire on an ongoing basis? And this question came up the last time this topic was discussed by BPAC. What are the plans for evaluation? Because the data from Blood Systems are very helpful, but they bring up in some ways, more questions than answers.

And so, is there a way to assure that there is going to be an ongoing process for evaluation that would be complementary to what this study would do, which I think will be important, but obviously will not answer everything.

DR. QUIROLO: My concern is that it's not really a study. It's a validation. It's not statistically significant. So, it's up to subjective analysis. And somebody in the FDA or somewhere is going to decide whether they think they are equivalent based on this data. If they don't, are we going to be back here again in a few months with a different questionnaire, doing the same thing?

And further, if the questions need to be changed again, well, then we have to do another study or some other kind of validation measure for each one of these revisions. I don't quite understand what the process is going to be.

DR. SCHREIBER: I think from what has already gone on, and from the numbers that we have seen, we know that the instruments are equivalent. They are equivalent at a 99.5-.7-.8 percent. So, we know that they are equivalent. How much inequivalency are we willing to accept? But when you have that small of a percentage that you are trying to ascertain a difference for, we have to remember that we are screening out all of the negatives as negatives, and there is a high degree of concordance with the instrument.

So, that what we are really trying to do is find the cases that are not picked up or picked up wrong with that, and then try to find out whether some wording can be changed in the questionnaire. And if that's what the objective is with that 1 percent deferral rate for the medical histories. And that's what was said that was trying to be evaluated. We are not going to get enough to be able to look at it in any sense, with only three or four or five cases.

DR. LEW: I would like to make it real clear. The data we saw had nothing to do with the form that is going to be tested. They use a different process, the presentation we heard. So, I like the data, because it does help us a little. It is a similar but not exact abbreviated form. So, we really can't say what this abbreviated form which we are voting for, is going for.

But the other thought to it is I guess I have to agree with Dr. Katz. If all FDA was looking for was an equivalent study, this is not it. And so, the point is moot. And do we need this study? I think we will have to think very carefully. Again, it is a different form. Maybe we need it just so we can say we have semi-validated a new form.

I still think the question that Dr. Allen brought up, is this form going to unfortunately not capture some important questions that should have deferred an important deferral question? We need to look at that issue.

And then again, it would be nice to know, since the blood banks do want people to come back or be satisfied, that those be incorporated into whatever mini-study that is done.

DR. KUEHNERT: I guess I just have to ask this question, because I think it's probably on everybody's minds. What is it going to take to actually get the abbreviated questionnaire approved? Because I think that's what people are afraid of, is holding this up. And there are many things that can be done to evaluate it once it's implemented to see exactly what the differences are. And I think clearly the impression is this is going to be better than anything we have out there.

So, I guess I'm trying to understand what are the processes for approval that we are being asked to sort of help FDA decide on? And it's sort of a black box that I think is making at least me sort of reluctant to vote any particular way.

DR. ORTON: Can I just add one thing to clarify, I think that will be helpful? As we described for the full length questionnaire, every single question had focus group and cognitive one-on-ones, every single one. We modified. We retested them with the focus groups, in some cases the cognitives.

Part of our concern was we have two new questions that have had no study. Now, I agree and understand that it's based on a somewhat validated, for lack of a better term, instrument. But that was our concern. We have two questions that have not been tested at all to determine whether -- just the fact I think that Jay brought up, should it say health included with medical, hasn't been tested at all. So, that was one of our concerns.

I think that the use of equivalents -- and after we had kind of gone through what the questions are going to be, the little red flag started going up when we say equivalents, we're going to be thinking about tens of thousands, an epidemiologic study. So, I think that this is probably an unfortunate set of wording, which is where we get into an alternative design, perhaps.

DR. KUEHNERT: Can it be changed?

DR. ALLEN: Let's come back to that point.

Dr. Katz.

DR. KATZ: Well, nobody responded to Matt when he asked about ongoing study. And I can't speak for all five of the centers that have agreed to do the study. Our plan was in fact because we were going to have two really bright people trained as cognitive interviewers, to then do cognitive interviews on every ADHQ -- presuming it's approved -- donor who in fact gave us post-donation information, or was subsequently deferred for something that we should have known.

And we were going to do that ongoing, until we either realized that's the nature of asking thousands of people questions, or we found patterns that suggested a need to suggest alternations in questions. And I suspect there will be more than one place that would use that approach.

MS. KESSLER: Well, my comment follows that. If you remember in 2003 when this was discussed, the AABB task force committed itself to post-implementation follow-up of whatever kind seems to be appropriate, including post-donation information, new health information, maybe doing cognitive interviews. So, we did commit ourselves to follow-up study.

DR. ALLEN: Let me just throw out alternative wording for the committee's consideration, and also for FDA consideration. Does it help if we modify the question to read, does the committee agree that the proposed study design, exclusive of sample size, is adequate to be reasonably certain that the abbreviated donor history questionnaire will be a satisfactory screening questionnaire for the eligible donor population? Dr. Schreiber does not like that.

DR. SCHREIBER: The other thing that I heard from Sharyn is that the concern was that all the other questions had been cognitively tested. If that is the concern that was raised with these two questions, we can very easily do cognitive testing with nine people each on each of those questions, and test and come back with the same degree of cognitive testing that went into the whole rest of the questionnaire.

So, if the FDA concern is at that level, there is no sense to go forward with something as costly and time consuming as this, where we could do in a week, the cognitive testing of those two questions. And remember that no one was tested -- the questionnaire has never been validated. The word "validation" in my mind is a misnomer here. What we did is we tested comprehension of the questionnaire, and then we call that validation. We never looked at people that had conditions, and see if those people were picked up with those particular questions.

So, just go to back again, we could do a much easier job and quicker and less strain. And if we follow this time schedule that Debbie has laid out, it's going to be next June when we have the results to look at again. And then we are going to be reviewing that and say, gee, maybe we should ask the questions a different way.

And I think Dr. Leitman raised an issue about whether the questions are worded right. Well, we could take various deviations of that now and test those in these cognitive interviews to give them the best chance to go forward. And if that's what it takes, I think that's what we should recommend.

DR. KUEHNERT: Another point I don't think has been brought up is equivalence might be bad, because we don't know -- you brought up the issue of gold standard. We don't know what the gold standard is other than confirmed positives. And Jay brought up the issues that the deferral may be picking up things that testing are not. But on the other hand, they may also be false positives, so we just don't know. So, I think that's part of our challenge here, is that there is no gold standard for comparing this.

DR. ALLEN: All right, well, either that or -- Dr. Epstein, did you hear Dr. Schreiber's last comment?

DR. EPSTEIN: I'm afraid I had other business. I got distracted.

DR. ALLEN: Okay, the committee is a bit hung up on the wording of question one, and that fact that it in fact really does presume, I think, that this is an epidemiologic study with statistical equivalents being determined. I think the committee is not certain that that is: (1) either necessary; or (2) achievable.

George, do you want to re-describe or make your comments again?

DR. EPSTEIN: I understand that point. What we can do is perhaps substitute the word "comparable" for equivalence, because equivalence is a rigorous statistical standard. Comparable sort of leaves it open to judgment, what's an acceptable difference based on the outcome of the data. It leaves you scratching your head, because you haven't pre-defined a delta that you would live with. But I think it's closer to what will really happen.

Another way of asking that question, I guess it's back to George, if you have 1,250 data elements, what is the smallest difference that could be detected with statistical significance? Because what you have said is you need a study of 6,000 to show equivalence with a delta I think you said of 0.5 percent. But the other way of looking at it is what's the power of the study as proposed, and could we live with that as a more qualitative measure?

DR. SCHREIBER: I think another issue came up concerning Sharyn's comment. These questions have not been tested cognitively. And the only test on any other portion of the abbreviated questionnaire has been the cognitive testing. And I guess my question was not dealing with the sample size issues, but could we just do cognitive testing on a variety of capture questions to the same degree that we have done on the other instrument, and live with that?

You raised the issue about the questions not being tested, the capture questions. Another issue came up while you were out of the room, whether there were better wordings. So, we could do cognitive testing a lot quicker, and it would be a lot cheaper, and probably give us the same yield in results.

DR. EPSTEIN: Well, I certainly would be in favor of cognitive testing of capture questions. I don't see how anyone could object to that. I think the real trade off here is whether something done prospectively on the model proposed adds much value compared to the kind of post-use testing that has been done by UBS. I think that it's obvious that the power of those post-market -- it's not really an approval process -- but phase IV analyses is much greater.

It's just that you run the risk that if you find significant deficiency of the abbreviated, then you have potentially compromised the safety of the blood supply. So, that's been the whole dilemma from the start. But the answer to it is not an underpowered study.

So, again, I come back to the question of how gross must the difference be for a study of 1,250 subjects to detect it? Would you have to miss 50 percent? Would you pick up a 20 percent difference? In other words, what is the actual power of the proposed study?

DR. SCHREIBER: I think the power of the proposed study is pretty close to zero. I would say that you probably would have to miss -- your estimate of 50 percent is probably more in the right ball park.

DR. ALLEN: In part, the fact is that you've doing this on a very highly screen population already, so that the probability of there being any deferrable condition is extremely low. If you were doing it in the population with a higher frequency of deferrable conditions or underlying medical conditions, your power is much greater. But that's the population you want to use this on. So, that's your conundrum of to get appropriate power, the sample size goes way up, and you are going to have lots and lots with no discrepancies at all.

DR. EPSTEIN: So I think that perhaps a better pathway would be to obtain a commitment that those implementing the ADHQ would commit to doing what UBS has committed itself to doing, which follow-up evaluations with some randomization to full length. And to continue to monitor the significance of those outcomes.

I do think that before we would sort of go live I would like to see a breakdown of the medical deferrals that occurred at UBS, so that we could get some handle on significance. It's a finite number. I think it was less than a combined total of 250. So, in some reasonable period of time we could actually cull those records and find out what it really was. And as long as we were not worried by the source of those discrepancies, I think we could go forward.

So, I'm just expressing my view, and it shouldn't unduly bias the committee, but I think the proposal to get cognitive evaluation of the capture questions is very important. And I think that if you will, transforming the study design to a post-implementation design would be superior, because then we won't have this problem of small numbers. Because if we go forward with a study that's underpowered, either we are going to kid ourselves, or get an answer we don't like, that we have to redo the study. So, there is not a lot of value.

And then I think that we have to have, if you will, some belt line that as phase IV is done, that if the missed rate of deferrable conditions is higher than X, that we'll pause and reconsider. So, I think that's a path forward.

DR. ALLEN: Do you want us to perhaps complete our discussions, take a break while we reword that question, and then come back and vote on a reworded question?

DR. EPSTEIN: Well, I think the rewording is easy enough. It's just to change equivalence to comparability. But I think it would help if we had some statistical understanding of the limited power of the study. In other words, if it's 80 percent power to pick up a 50 percent difference, then that is what you are voting on. You're asking whether a study with that power to pick up that magnitude of difference is worth doing. I don't know how readily you can do that calculation, but that would be my concern.

MS. BASSETT: I just wanted to respond to Jay's request that Blood Systems would look at that information in more detail. And we have committed to ourselves to do that even before here, so that we would have the information of what's behind those numbers of increase. So, we would be glad to do that and get that information to the FDA or whoever else that that information would need to go to.

DR. EPSTEIN: Thank you.

DR. DI MICHELE: Can I just ask Dr. Epstein, are you saying then that you would still want something prospective looking at comparability, but look at equivalence in a post sort of implementation fashion? Or are you happy with just a post-implementation study?

DR. EPSTEIN: I'm not trying to pre-judge the vote of the committee. I think the committee should decide whether the study, as proposed to demonstrate comparability is acceptable. And I think the committee has in essence proposed that there at least be cognitive testing of the capture questions. If the committee votes in the negative, that this is an adequate study for all the reasons discussed, then I'm proposing as an alternative, that the game plan would be a well designed post-implementation study with clear stopping rules.

DR. ORTON: And I think the question number three addresses that, Jay, exactly what you are saying. If your answer to one is no, we are asking for alternative --

DR. EPSTEIN: But again, question two does suggest an alternative, which is a higher powered study which is done prospectively. But then that begs the question of to what extent was the full length validated in this way. That was Lew's point earlier, I think.

So, I think we've got the right set of questions here. We just need to move off calling it equivalence, because it is clear that it's not designed as an equivalency study. So, we shouldn't be asking you whether it's an adequate equivalency study if it's not an equivalency study. And I'm suggesting that a softer term that is a little bit more qualitative enables you to vote yea or nay on the study as presented.

And if you vote it down, then your choices are to redesign a larger prospective study, because you are very worried about going live with the abbreviated. Or to say well, we have enough experience with abbreviated donor questionnaires. We think it's okay to go live, but we want belt and suspenders. And so we are going to want a well designed, post-implementation study that is statistically powered, has well defined prospective endpoints and clear stopping rules.

DR. ALLEN: Certainly, that enables you to get to the point as Blood Systems has -- Mary Beth, total number of ADHQs you had was up in the 53,000?

MS. BASSETT: In this group, 53,996 was what we had looked at here. But those were at three different time periods.

DR. ALLEN: The point is since implementation you have done hundreds of thousands of donors using the abbreviated questionnaire.

DR. EPSTEIN: Right, and I think the important point is that that yielded only a couple of hundred comparators of eligibles who then got full length or abbreviated. And with the caveat that they may not have been a random assortment either, because of this confounder about mobiles.

So, it's going to take a very large post-implementation study to really answer a non-inferiority or equivalence question. I think that that's just where we are. The question is whether the committee feels that something less statistically powered that gives you some qualitative assurance that it's not awful, it's not missing 80 percent. Maybe it's not missing 50 percent is needed before you go live. Mind you, we have already approved two of these. I mean there are two large systems doing this now.

MS. BASSETT: I just want to make one clarification too. On one of the slides I think I wasn't sure which slide he was looking at when I responded earlier, but Jay, I think the one slide you were talking about where things were twice or four times was related to different groups. One was not eligible for ADHQ. They were the repeat donors, and one was eligible.

So, what you were looking at where the pheresis donors?

DR. EPSTEIN: Well, the slides that concerned me were slides 12 and 13. Slide 12 compares the medical history deferrals for eligible donors who either receive the abbreviated or full length questionnaire. And in that data set, the rate of medical deferrals was 0.45 percent for the abbreviated, however, it was one-third larger, 0.59. And that was statistically significant. P equals 0.001.

And then if you looked at the donors whose last donation was than or equal to eight weeks, you found again for eligibles the rate of medical history deferrals was 0.22 percent for abbreviated, but it was 0.89 percent, about four fold higher for full length. And that's I think, somewhat concerning. It suggests that the people who donate frequently and get abbreviated questionnaires tell you less. And I don't exactly understand why.

And the other point that I was making is that if you look at slide 12, it took a study size of almost 54,000 in order to cull out a grand total of about 270, which were then compared with abbreviated and full length. So, it requires very large follow-up to learn anything meaningful here.

So, again, I'm just stating my bias, which is that with your own analysis of the statistical power here, there is not a lot to be gained from such a small study. I would say the only reason to do it is if we're so worried that it could be a dismal step, that you want to try to pick up a 50 or an 80 percent difference, because that you should be able to see with 1,250.

DR. KLEIN: I think this a good example of a controlled trial, because I think we said that while you were out of the room, Jay.

DR. EPSTEIN: I apologize for bringing you the long version.

DR. ALLEN: And again, I think rather than simply focusing on a fairly short study period, it will be far more meaningful to follow-up on people whose call back with deferral information, people who have a positive laboratory test after having gone through the screening questionnaire, you will have selected populations, and I think it's much more important to get detailed information on that group, rather than focusing on a small number during the study.

Dr. Epstein, I had one other question. The question as given to the committee, if you look at the bottom part of it, the question focuses on the two capture questions on the abbreviated questionnaire, and the 17 specific questions on the full length questionnaire.

It seems to me that what you really want to do is to validate the whole abbreviated questionnaire. And I just wondered if you would have objection to rather than focusing on the two capture questions, to focus on the abbreviated questionnaire versus the full length questionnaire in terms of comparability?

DR. ORTON: I just want to clarify that with the exception of those capture questions, if I'm not mistaken, Debbie, the rest of the questions are the same. So, the only things that are different are these two capture questions.

DR. ALLEN: I guess my point being that it is the entire questionnaire, however, that is important.

DR. ORTON: So, even if these questions are the same, you would like to see?

DR. ALLEN: It just seems to me that what we are doing is putting the emphasis on the abbreviated questionnaire, rather than simply on the two capture questions.

DR. EPSTEIN: I think that that's a fair point, because people may respond to the questionnaire as a whole in a different way, even though the remainder is the same. If we were to go that route the revised language would be does the committee agree that the proposed study design exclusive sample size is adequate to demonstrate comparability or lack of comparability between the abbreviated questionnaire and the full length questionnaire?

DR. ALLEN: Committee response to that?

DR. KUEHNERT: Just a point of clarification on the difference between comparability and concordancy, because I thought what George spoke on was equivalence and concordance/discordance. And now we have another word, comparability. I'm just confused about what that means.

DR. EPSTEIN: Okay, what we are saying is that the rates of deferral would be in the same ball park. And by saying comparability, we are not passing judgment on how large a difference we might accept.

DR. KUEHNERT: So, it sounds like a similar thing to discordancy.

DR. EPSTEIN: Yes, it's discordant rate without having set a delta. You haven't decided how bad does it have to be before you reject. Now, again, this is not the most rigorous way to approach a study. Usually you pre-decide how much of a difference will cause rejection of the conclusion.

But if you are looking at this more as a survey, then you are really looking qualitatively at what kind of rates are you getting, and then you make some kind of belt line decision. We might have to come back to the committee and say, do these differences cross your belt line?

DR. BIANCO: Just to express that I'm very happy with the way Dr. Epstein changed the focus very much from just this study to a post-implementation study. And what I think would be very useful to all of us in the task force and everybody involved in the process is to hear your discussion of what the post-implementation studies would be, more in the detail of this study. It seems that it's not going to fly, right?

DR. ALLEN: Okay, other comments or discussion? Does anybody on the committee have a concern with the reworded question? Or is that a satisfactory question to accomplish what we need?

DR. DI MICHELE: My only question now is whether this is a study at all?

DR. ALLEN: Well, it's a study. It's not an epidemiologic study. It's not one that has got statistical validity. It is a study. You are looking at it. You will be getting qualitative-type information that you can use to make a decision on. It's not going to be one that's got statistical rigor to it, and I'm comfortable with that.

DR. LEITMAN: The committee is not being asked when this study would be done. It might be either after implementation or before? So, what I hear is before. We are only voting on before?

DR. ALLEN: I believe that -- well, Dr. Epstein.

DR. EPSTEIN: What you are being asked is whether there should be a prospective study before FDA might act to accept the abbreviated uniform donor history questionnaire. So, it's prospective, prior to implementation.

DR. LEW: Could I make a proposal that we have the cognitive studies done on the questions that are new, and then redesign a post-implementation study? Since the full form is already being used, and the big question is, is this consolidation worthwhile, I think even though it's not equivalent, I recognize it's not equivalent, the other abbreviated form, there is enough data there that it makes me comfortable that this is going to be somewhat equivalent. Then maybe we should just go into implementation if that's possible, but design a study to see how much difference there is.

DR. LEITMAN: I get the feeling that in the UBS study, if there were some very important medical deferrals that we were not picked up on your abbreviated questionnaire, deferrals that truly impacted the safety, purity, potency of units collected, Mary Beth, you would have known. That would have come out. It would have been brought to attention of quality assurance leaders and medical directors.

So, my feeling -- there is no data -- is that these medical deferrals were I forgot I had a squamous cell cancer removed on my face four months ago or something like that. Are things that might defer the donor related either to the donor's health, or they are just in your SOP for deferral, but you don't really know what the implication is. But not the male having had sex with the male, or IV drug use are the very big ones recently that we are concerned about, because that would have come to your attention. Am I wrong in thinking that?

DR. KAMAL: No, actually all of the high risk questions are word for word on the abbreviated questionnaire as they appear on the full questionnaire. Because of time limitation we show the difference between platelet donors and red cell donors. It's basically less than eight weeks. When we looked at the full questionnaire and the abbreviated questionnaire, in donors whose last donation is greater than eight weeks, the difference was not significant.

But because asked last time around if platelet donors would have an effect or bias the information, we showed you what you asked for, for donors whose last donation was greater than eight weeks, which would be red cell mostly, the difference was not significant.

DR. LEITMAN: What I was trying to get at, what is the real concern with a post-implementation surveillance or study? And the concern seems to be the slides that were shown on the difference in medically deferring conditions that were detected by the full questionnaire versus the abbreviated questionnaire. That questionnaire is not one I have seen though. I have only seen the AABB task force questionnaire, so I can't quite comment on what the difference in questions was.

But the feeling that I get is that we shouldn't be that concerned about. Concerned enough to request a study of this kind before implementation. Just like Dr. Lew said, I feel very comfortable with cognitive testing of the two capture questions, and whatever size is deemed reasonable to validate those questions, determine if there is a problem with them. Review of that data, and then proceed with implementation and a study such as this on a sample that is large enough to get at whatever you are asking for, comparability, concordancy.

DR. ALLEN: I think the question that Dr. Epstein was raising with regard to the Blood Systems data is if you've got a medical deferral at this level with the abbreviated, and at this level with the full, there really isn't an evaluation process to look at what is in between.

Mary Beth did talk about the post-donation information that comes in. And it comes in on all the donors. It comes in at about the same level. And I think generally a lot of that has to do with travel deferrals. Am I correct? We don't have people coming in with, as best I understand it, with major issues that they failed to defer themselves on. It could happen, certainly, but I don't think that has been a major problem.

MS. BASSETT: The data that I had pulled up just for those 120 that we did, 24 of those were related to travel or resident kind of risk. And as Jim stated, the post-donation information for all the groups that we have looked at has stayed relatively consistent throughout the whole study.

DR. ALLEN: Can I see a quick show of hands by the committee of people who would like to see the question with FDA approval revised to talk about cognitive evaluation and a post-implementation evaluation study, rather than a pre-implementation study? If you would like to see it that way. Okay.

Dr. Epstein, are you comfortable with revising the question to indicate that?

DR. EPSTEIN: I think a suggestion for cognitive evaluation is sort of a stand alone issue, and trying to lump things isn't going to be helpful. But I certainly don't have a problem with revising question one, as I suggested earlier. I think we have already heard a suggestion from the committee that cognitive evaluation of revised capture questions would be appropriate. We can put it to a vote, but I think we already got that advice.

DR. KLEIN: Jim, perhaps that could be our answer to question three if we go through the questions as it's written here.

DR. ALLEN: Okay. Is there further discussion on question one as revised, or are we ready to discuss it? Basically, it's does the committee agree that the proposed study design exclusive of sample size is adequate to demonstrate comparability between the abbreviated questionnaire and the full length questionnaire?

All right, Dr. Freas, would you go ahead and call the roll, please? Excuse me, Dr. Quirolo, did you have another comment?

DR. QUIROLO: I just wondered if the way we are going to vote, if the way the hands just went up, it sounds to me like we are going to vote no on one, no on two, and propose new studies.

DR. KUEHNERT: I'm sort of equally confused. I mean you could -- never mind.

DR. DOPPELT: If I understand what Dr. Epstein was saying, the question for this is do you expect the addition of these two questions to somehow roughly result in the same kinds of answers or complete catastrophe? And if you don't expect it to be a complete catastrophe, which I don't think any of us do, then why even do it?

DR. ALLEN: Before we call the role, I think thank you for the clarifying question. I think the way I had interpreted the last exchange of discussion, I was going to vote yes on one, leaving it to the FDA to decide how to implement, whether it was pre-approval or post-approval. But I think with the revision in the question, with the understanding that the FDA has heard the committee statements about cognitive validation of the questions as written, does the committee believe that this study design is adequate?

DR. DI MICHELE: Well, I just wanted to ask another question, maybe to the blood banking industry. One of the issues of doing any kind of study has been an issue of cost. When you look at sort of the cost of this study, I suppose whether you train interviewers to do 1,200 interviews or 6,000 interviews, I'm not sure whether that cost is different. But certainly, study coordination, data accumulation, data collection, data analysis is certainly quite expensive.

On the other hand, Blood Systems has just done this post-surveillance study of 53,000 or 60,000. And I'm just trying to understand the costs of doing the kind of post-surveillance that has been done on a wide scale basis, and whether that cost is even greater than a prospective study. And if not, why that is. Why it's much more cost effective to do the same kind of data collection post-implementation for 70,000 donor than the other way around?

DR. BIANCO: Thank you for thinking about cost. There are two types of costs associated with that. One, there is the cost of doing business. There is analyzing data that is generated in the normal operations to analyze deferrals. This is something that we should all be doing. And it's hard sometimes, because our computer systems have not gotten there yet. They swallow a lot of data, but spit out very little, like any other computer system. But this is part of what we do in terms of monitoring deferrals, monitoring whatever.

The other studies, that is when we need a special informed consent, specially trained people that fall outside the operations, those are much more difficult. So, the initial, this study proposed in one is much more difficult, resources-intensive than simply a study of monitoring of data that is generated during the normal activities of the blood center.

DR. DI MICHELE: Can I ask a follow-up question? Then my other question is, the task force has also put in a lot of thinking into these two medical questions. But what we are hearing from the Blood Systems report preliminarily is what has been missed -- let's say we're concerned about implementation before a study or whatever, and it appears that the little information we do know about the medical deferrals is that they are mostly on the basis of travel.

The question is, would the task force consider, if we suggest sort of a cognitive validation of these questions, would they consider putting in a third question related to interval travel?

DR. BIANCO: That's already there and was validated by --

DR. DI MICHELE: So, it's already in the abbreviated?

DR. BIANCO: It's in both.

DR. ALLEN: Yes, it asks about travel outside the United States or Canada in there interval period.

DR. DI MICHELE: So, then really the issue of what's been missed in the abbreviated questionnaire is something on the basis of a question that is already in the abbreviated questionnaire? Is that what you are saying?

DR. ORTON: No.

DR. KATZ: The way the questions are now is in the long form you get asked about everything, every time. The ADHQ then changes that to since your last donation. Then you have the opportunity that somebody either forgot the last time they got the long form, isn't asked about past travel on the ADHQ. And I think it's something that we have concerns about.

We know that post-donation information, there are the flus and the colds and the runny noses and all that. But the killer of course is, well, I told you I went to Cancan last time, but I didn't tell you I left the resort and now I'm telling you that. And then you have to go back and trace units and do all this craziness.

DR. ORTON: Would it be helpful if I actually read you the questions that are --

DR. DI MICHELE: No, I just missed it. That's okay.

DR. ALLEN: All right, let's go back to the questions. As I am interpreting the question, we have changed the wording to comparability. It is excluding statistically significant demonstration of equivalence. The committee should vote, is this study design adequate or not, yes or no? I would think that the instruction from the committee to the FDA is this is not necessarily an endorsement of pre-implementation use of the questionnaire. That's a decision that the FDA should make.

They have heard the other discussion, that with appropriate cognitive evaluation of the questionnaires, that the committee individually -- members of the committee individually at least, believe that post-approval implementation and evaluation could be an adequate alternative. And we can discuss that further under question three. Is that adequate clarification, Matt?

DR. KUEHNERT: That last part, you are saying that the two are not mutually exclusive. I was just trying to understand if a committee member were to think well, what we really want is post-implementation, to cognitively test the capture questions the way the other ones were done, and then do post-implementation data collection, then how would we vote?

DR. ALLEN: I am going to vote yes on question one, and the statement that I as an individual will give to the FDA is this might be modified and implemented as a post-approval evaluation, rather than as a study of 1,250 in only 5 centers. But I think the basic study design is adequate. It's got the restrictions of sample size, and it doesn't have statistical power. I think it will answer the questions that need to be done. I think the suggestion of cognitive evaluation of the questions before you do that is a good thing to do under any circumstances.

DR. DI MICHELE: But, Jim, the study as it's proposed is also a prospective, pre-implementation study. So, I think given everything you have said, I would vote no on what I've heard, and then get to question three. But just as a clarification so when this vote does come up, everybody understands.

DR. ALLEN: What I would suggest, these conversations are recorded, and people are taking notes. I would suggest that with your yes or no, that you provide whatever qualifications that you want to along with your vote.

Dr. Freas, let's go ahead.

DR. FREAS: Okay, I'll go around and call the votes. Before I do so, I want to clarify there are 12 voting members at the table at this time. And at the end of the voting members' polling, we will ask for the opinions of the non-voting industry representative, and the non-voting consultant.

Calling the roll now, Dr. DiMichele.

DR. DI MICHELE: No.

DR. FREAS: Dr. Kuehnert.

DR. KUEHNERT: Yes, but I don't think it's needed.

DR. FREAS: Dr. Harvath.

DR. HARVATH: No.

DR. FREAS: Dr. Klein.

DR. KLEIN: I don't believe this study will provide any useful data to the FDA, no.

DR. FREAS: Dr. Schreiber.

DR. SCHREIBER: Ditto, no.

DR. FREAS: Dr. Lew.

DR. LEW: No.

DR. FREAS: Dr. Quirolo.

DR. QUIROLO: No.

DR. FREAS: Dr. Allen.

DR. ALLEN: Yes, but I don't believe that the study is necessary pre-implementation if the cognitive evaluation is done.

DR. FREAS: Ms. Baker.

MS. BAKER: No, ditto Dr. Allen's statements.

DR. FREAS: Dr. Manno.

DR. MANNO: No.

DR. FREAS: Dr. Doppelt.

DR. DOPPELT: Yes, and I would agree that it doesn't have to be done pre-implementation.

DR. FREAS: Dr. Whittaker.

DR. WHITTAKER: No.

DR. FREAS: Could I have the opinion of the acting industry representative?

DR. KATZ: I think that my opinion is that cognitively evaluated capture questions should be included in the ADHQ, and a post-implementation study to be agreed upon by the interested parties should be allowed by the FDA.

DR. FREAS: Thank you for your comment.

A comment from the non-voting consultant?

DR. LEITMAN: I think Dr. Katz stated it perfectly, and I would ditto his comments.

DR. FREAS: So, polling the voting members we had 3 yes votes, and 9 no votes.

DR. ALLEN: All right, we have made significant comments about the sample size. Dr. Epstein, given the results of that las vote, do you need a vote on the sample size question?

DR. EPSTEIN: I don't think so. I think that the sense of the committee is that a post-approval study would be more meaningful, and that it remains to be designed. We'll seek appropriate epidemiological, statistical, cognitive input into that in cooperation with the industry.

Again, I think on question three, if there are specific suggestions on study design, we would really like to hear that.

DR. ALLEN: All right, we are now open on question three for any additional comments, recognizing that there has already been a lot that has been given.

DR. KUEHNERT: I think what Dr. Katz expressed was what I would agree with.

DR. QUIROLO: I assume there is -- I doubt that you would need it, but some kind of a stop rule, where if you realize that there was a significant difference between the two forms, that the study would be stopped, and they would be re-analyzed?

DR. KATZ: Yes, I think that's important. I think Jay actually made the point fairly clearly that that would be required. And I'm pretty sure I don't want to try and design this study or the sample size right here, but the task force -- everybody that needs to be represented, is represented. I think we could come up with it pretty quickly.

DR. LEITMAN: I can anticipate a finding where the discrepancies were almost entirely due to the memory jogging effect of remote historical travel by the donor. So, if that shows up in a year or two from now, how important will that be? Will the fact that you got more of a memory jog effect for travel to Europe or whatever, or residence on a military base, would that derail this?

DR. ALLEN: Well, that's not a question for the committee to answer, unfortunately. I suspect that the advantages of an appropriately implemented designed and implemented abbreviated questionnaire will not be derailed by such things as that. And just as to me the way in which the information is asked and accumulated will be refined over time, I think as it has been for medications and other things. This is complex. We need better ways of identifying travel histories. And it seems to me that the best way is not to ask somebody.

I couldn't begin to tell you how many actual weeks I've sent in England. I know it's well under 90 total days, but it is an important question, and I don't think that the efforts to figure out better ways to acquire the information from donors should stop at this point.

DR. KLEIN: In reference to question three, I entirely agree with the way it was summarized by Dr. Katz, and with the stopping rule. I also think that it is important for a post-donation study, although not necessary, to evaluate the impact on donor retention. If those data could be collected at the same time, I believe they would be very valuable.

DR. ALLEN: Thank you. That's an important comment. Other comments or questions?

All right, we will take a break at this point. Why don't we try to be back in 10 minutes, and then we'll go into our last section, which is the review of the site visit report from the Laboratory of Molecular Virology.

[Brief recess.]

DR. ALLEN: We will now move on to topic IV., which is the review of the site visit report for the Laboratory of Molecular Virology. The site visit was conducted on October 29, 2004. We will have a series of five presentations by the staff of the laboratory, the first being the introduction and background by Dr. Hira Nakhasi, who is Director of the Division of Emerging and Transfusion Transmitted Diseases in the Laboratory of Molecular Virology within that division.

Agenda Item: IV. Review of Site Visit Report for the Laboratory of Molecular Virology, DETTD - A. Introduction and Background - Hira Nakhasi, PhD, Director, Division of Emerging and Transfusion Transmitted Diseases, OBRR

DR. NAKHASI: Good afternoon. Thank you, Dr. Allen.

I think the purpose of my talk here is to set the stage for the individual investigators to talk about their research projects with relation to the mission of the laboratory. And I will just give you a brief introduction, and then the people will talk about the individual projects.

In the office in the Division of Emerging Transfusion Transmitted Diseases is located in the Office of Blood Research and Review, OBRR. Dr. Epstein, as all of you know, is the office director. We have a new deputy director now, Dr. Paul Mied(?). Mary Beth was acting for the time being. She is Associate Director for Regulatory Affairs, which is her second hat here.

And also Dr. Linda Smallwood -- this is now quite a bit changed here. Dr. Linda Smallwood has already retired. And then Dr. Ed Taber is the Associate Director of Medical Affairs.

The office has three divisions, the Division of Emerging Transfusion Transmitted Diseases, the Division of Hematology, and the Division of Blood Applications. The focus today is the scientific review in the Division of Emerging Transfusion Transmitted Diseases.

The division is organized in three major laboratories relevant to the subject matter, the Laboratory of Molecular Virology, who is headed by Dr. Indira Hewlett. This is the laboratory which will be discussed today. And our science review of that laboratory will be discussed today, and Dr. Hewlett will talk a little bit more about the work that is being done in the lab in the introduction.

And then we have a Laboratory of Hepatitis and Related Emerging Agents, a Laboratory of Bacterial, Parasitic and Unconventional Agents, and we also have a product testing staff which does the lot release testing.

The responsible mission of the division is to first of all, insure the safety of the nation's blood supply by reviewing, evaluating and recommending various test kits for blood-borne viral, parasitic, bacterial and now transmissible spongiform encephalopathies or prions.

Also, the mission is to conduct research applied to insure the safety and efficacy of these products, which basically results in the development, manufacture, and understanding how these pathogens work, and then the test kits and developing of the assays.

In addition to that, as I said, our responsible is the lot-release testing of all these investigational tests and the licensed products, and develop reference materials for lot-release testing, and also develop standards for validating tests. And in addition to that, the staff does the inspection of the manufacturing facilities where a lot of these test kits are manufactured, provide the scientific and technical advise to other agencies and the government components, and also presents the issues to bodies like this and other advisory committees.

The personnel and the budget of the office in 2004, as you see, the total staff is 58. There are 11 senior investigators which are supported by biologists, staff fellows, staff scientists, regulatory scientists and administrative staff and post-doctoral fellows. The total budget is approximately $400,000, and last year there were 41 publications and some book chapters published.

The regulatory activities, obviously, the major component and last year which included all sources of applications, and with the total we had around approximately 374. And also, we did lots of lot release testing, and performed inspections and laboratory investigations.

The purpose of this whole thing is to give you an idea of what our workload, both in the research, as well as the regulatory arena. The research priorities of the division are two fold. One is to focus on the areas where we deal with the safety and efficacy of these test kits for blood-borne, viral, bacterial, parasitic agents. And understanding the pathogenesis of for example retroviruses, West Nile virus.

Other parasitic agents which are being studied are malaria, leishmania and Chagas disease, and some of these bioterrorism agents both viral and bacterial, and also hepatitis-related agents such as HCV, HBV, HAV, and the pathogenesis of the prion proteins and bacterial agents.

The second important component of the research is the assay development. I think the assay development is for new technologies for detection of retroviruses, or characterizations of new forms of variants of the new pathogens. And then lot-release as I said, development of lot-release panels and standards for existing, as well as new and emerging pathogens.

Also, looking at the new technologies which come down the pike, we are actually engaged in that, because that is one of the issues, as you heard yesterday about the critical path issues, to understand what are new technologies or new way of thinking how we can help industry stakeholders to overcome the issues which could be resolved earlier on.

For example, use of microarray DNA technology. Now we have heard nanotechnology, and there are other newer technologies where we can multiplex some of these tests, and we can test many more pathogens in one swoop, such as developing the TSE methodologies and detection of validation methods for TSE agents, and bacterial agents.

So, how does it fit in to the critical path initiatives, which you heard obviously? Whereas, you heard the nice story about the West Nile virus, how we engaged in how that helped in developing the test. Similarly, studies were done with the smallpox and blood safety issues, HIV surveillance study in Cameroon, which may hear from Dr. Hewlett.

Studies were initially to find out the sensitivity of various HBsAg tests and HBV NAT, and developing as I told you, the oligonucleotide based microarray chip, and other concepts of parasitic and viral agents. So, that's what we hear from what is going to be coming down the pike, and how we can work with those issues.

In addition to the critical path initiative, we also have a mandate about some of these counterterrorism initiatives, developing laboratory expertise in new technologies, as I mentioned earlier, nanotechnology, microarray technology, and any other new technologies that come down the pike. And then developing panels and standards, lot-release testing, and also understanding how these pathogens work.

With that, I will close my summary. After me, Dr. Indira Hewlett, she is the Chief of Laboratory of the Laboratory of Molecular Virology. She will be talking too, talking first about the laboratory, and then her own project. After that Dr. Andrew Dayton will talk about his work, and after that Dr. Subhash Dhawan will talk about his work. And last but not least, Dr. Maria Rios will talk about her work.

Thank you.

DR. ALLEN: Okay, thank you very much, Dr. Nakhasi. And he has basically introduced the staff. I would just ask the presenters please try to keep within the time limits, since we are running a little bit later, and we want adequate discussion time for the committee.

Dr. Hewlett.

Agenda Item: IV. Review of Site Visit Report for the Laboratory of Molecular Virology, DETTD - B. Overview of Laboratory and Diagnosis of Pathogenesis of HIV Variant: A Progress Report - Indira Hewlett, PhD, Chief, Laboratory of Molecular Virology, OBRR, FDA

DR. HEWLETT: Good morning. I'm going to be presenting an overview of the Laboratory of Molecular Virology. And that will be followed by just some highlights of my own research program in this lab.

Just to note that we are one of the three research laboratories in the Division of Emerging and Transfusion Transmitted Diseases. This lab was created during the CBER reorganization of 1992, and at the time it was given responsibility for all HIV tests that were based on synthetic peptides and molecular probes and recombinant DNA technology. In 1999, we acquired responsibility for all retroviral products, and from 2001, we actually added on West Nile virus, vaccinia, and software reviews for the division.

The lab is organized into five groups. Four of them conduct research, and they are actually manned by individuals who are researchers and reviewers. So, they do both research and review. And we currently have a product review section which is composed of individuals who do exclusively regulatory review of product applications.

The regulatory mission of the lab is to review and license applications for all in vitro tests, that is donor screening and diagnostic assays for HIV, HTLV and other retroviruses in blood, plasma and other body fluids, for example, oral fluid. We also review West Nile virus donor screening assays, and this has been going on since 2003, when implementation under IND took place.

We perform concept reviews for smallpox submissions. And this is only a small part of the activity, but it does exist in this laboratory. We also perform software and instrument reviews for the division, as I stated earlier.

The regulatory product area is a review of applications for convention antibody, antigen, and nucleic acid tests, rapid tests for diagnosis, patient monitoring assays, genotyping assays, and drug resistance, and home use assays, although we don't have any activity in this area at the moment.

We also perform inspections of test kit manufacturing facilities. We develop guidelines, review criteria and standards for validation of tests, and our staff are also involved in developing policies related to their use through interagency, industry and BPAC discussions.

So, to support the regulatory mission of the laboratory, we have research programs. And the research is basically focused on HIV viruses, HTLV, West Nile, and vaccinia virus. We conduct laboratory investigations of disease transmission and pathogenesis. Our staff are involved in developing and evaluating new technologies and methods to insure blood safety from transmission of HIV, HTLV, West Nile and vaccinia.

And another important aspect of our laboratory activity is standards development. So, we are constantly engaged in some type of standards development activity for licensure and surveillance of retroviral tests, and more recently for West Nile virus assays.

And just to give you some key points of each of the programs -- you will hear more about this from each of the individual principal investigators -- in my own group we are looking at genetic diversity and the impact on diagnosis and pathogenesis of some of these variants. We are developing reference reagents for HIV NAT assays. There is an individual in my own section who is working on vaccinia virus as a model for smallpox. And we have looked at viremia in vaccinated individuals, and we are starting out some work on the molecular basis for myocarditis induced by smallpox vaccination.

We have Dr. Dayton's lab that is looking at the molecular biology and gene regulation of HIV in infected primary human macrophages.

Dr. Dhawan is looking at diagnosis and pathogenesis of HIV, focusing on host and viral factors in disease transmission and progression. And he is also developing diagnostic tools for viral detection.

And Dr. Rios -- this is the youngest program in the laboratory -- has actually been highly productive since she joined the lab, and has been developing programs for diagnosis and pathogenesis of West Nile virus, looking at genetic characteristics, tropism and standards development for licensure of West Nile assays.

So, in summary, this lab has actually had product responsibility for HIV tests for a fair length of time. It's actually over a decade now. We have had new product responsibilities including West Nile virus, vaccinia, software and instrument reviews. And we have actually maintained and initiated some new research programs to support these regulatory activities.

So, that's the end of my overview. Are there any questions? Or should I move on to the next part of my presentation, which is actually my own work that is going on in my own group.

As I said in the overview, the major areas of investigation in my group are actually three major areas. That is diagnosis of genetic diversity of HIV, and its impact on diagnosis. We are also looking at pathogenesis of these variants, subtypes, HIV-2, multi-drug resistant viruses.

And we have a program on standards development. We have developed standards for HIV-1, and we are now looking at subtypes and HIV-2. These standards would be for blood screening and to monitor therapy in vaccine trials. And finally, we have a small program on vaccinia and smallpox as it relates to blood safety.

In regard to the first area with HIV variants, the study goals were to evaluate the sensitivity of existing and new blood screening assays for diverse subtypes with sera from Cameroon. We chose to work in this area because it is known to harbor diverse HIV strains. We have characterized and genotyped several variants from this area.

We are now investigating the pathogenesis of some of these major variants, and we are trying to identify samples to serve as candidate reference materials from this study. I would like to say at this point that actually NHLBI, just a few weeks ago, informed us that they are providing us with funds to actually continue this very work in Cameroon. So, we are looking forward to actually pursuing this in a more aggressive manner, and to be able to characterize these viruses completely, and of course to look at their impact on blood safety.

The conclusions from our study were that the FDA licensed tests were able to detect additional positive specimens than those identified in Cameroon. We found that EIA results from different manufacturers did not always agree with each other. The Western blot may not always resolve some of the discordant results. This is when you are looking at non-B subtypes in African sera.

We found that CRF02_AG was the most prevalent viral strain in Cameroon. This is about 70 percent of the viruses circulating in Cameroon are CRF02_AG. This is different than the picture that was seen about maybe six or seven years ago, where the discrete subtypes were actually predominant and those were subtypes A and D. So, we are seeing a very different picture emerging over time in terms of the predominant viral strain found in this particular area.

We have also identified new circulating recombinant forms. These are unique forms. There is actually one isolate of each, so that's AC, AF and GF. The good news is they were all detected by the nucleic acid tests that are licensed in the US. So, from a blood safety and diagnostic point of view, they were not an issue.

We have identified in the study, several discordant samples. And we are now analyzing these discordant samples as well as some negative samples with degenerate primers. We are also looking at PCR-RT assay. This is in collaboration with a group in the Office of Vaccines.

We are also looking forward to using some new technology to look for new viruses in this set of samples. And in fact, with the funding we are getting from NHLBI we will be initiating a collaboration with Phillipe Niambi(?) from Cameroon to acquire some very interesting samples that are cross-reactive with SIV peptides, and start looking for cross-species transmission, as well as new variants that might be evolving in this area in the course of time.

We have a proposal to look at microarrays, viral discovery arrays for identifying genotypes, subtypes in new viruses. Since we have sequence data of these recombinants, we can validate this using the sequence data that we have already acquired.

And from the pathogenesis point of view, we are looking at both tropism and apoptosis by molecular cloning of envelope genes of specific unique isolates to see if there are any differences in these properties of these variants relative to the known subtype B viruses.

In another type of variant that we are looking at are HIV drug-resistant viruses. And the goal here was to look at their replication capacity, tropism, genotype, apoptosis capabilities, and chemokine and cytokine product. It's essentially co-receptor usage as a function of virus evolution in patients who are on antiretroviral therapy.

We also wanted to examine whether there was correlation between genotype co-receptor usage and virus replication. In this study we isolated viruses from patients who were naive and patients on prolonged antiretroviral therapy. And we have found that actually they continue to use classical chemokine co-receptors CCR5 and CXCR4. So, there aren't any new receptors that are being used by viruses from patients on antiretroviral therapy.

There might potentially be a switch to co-receptors, to a specific co-receptor. In this particular case it was switched to CCR5 after they were on prolonged antiretroviral therapy. And that's not unusual, because that co-receptor usage is associated with less pathogenesis than the CXCR4 using viruses.

Tropism was not apparently related to CD4 count, viral load, genotypic mutation in the whole genes and the C2V3 region, which is actually the major region that dictates the tropism of HIV viruses. We are now planning some future studies on cytokine regulation, the ability of drug-resistant viruses to cause apoptosis, since we would expect that if they are less pathogenic than naive viruses, they could potentially be less capable of inducing apoptosis.

Of course we have all heard about this unusual virus that was isolated in New York a couple of weeks ago, and David Ho is actually working on this virus to see how different it is than the standard drug-resistant strains.

In the standards development area, we have developed a subtype panel. This has been formulated, and is available for manufacturers to use to standardize their assays for either quantitation or qualitative detection of the major HIV subtypes.

We have also developed an HIV-2 panel. It is actually in the final stages of development. We have data from several manufacturers who have validated HIV-2 nucleic acid tests, because we see this as being the next virus on the horizon. Although it's not a huge public health issue in the US per se, but there is a need to have HIV-2 NAT assays in blood screening, because we already have HIV-2 antibody testing, and this is probably where the next round of product development is going in terms of multiplex NAT.

So, that is pretty much where we are with the HIV projects. In regard to smallpox vaccination, this area was actually initiated about two years ago. The reason we started this is because there were concerns for blood donation. There was some talk about mass immunization at the time of the 9/11 incident, and concerns were raised about whether these mass vaccinees could present as blood donors.

There was concern about potential viremia in vaccinated individuals. And the concern was heightened due to lack of data on virus transmission by blood. So, studies were needed to determine the appropriate deferral period for vaccinated donors.

The second issue had to do with potential false positive reactions with sera from smallpox vaccinees similar to what was observed with flu vaccination many years ago. And this of course would be an issue, because if there was a very significant rate of false positivity, you could have a major impact on blood availability due to increased numbers of donors being deferred as a result.

The conclusions from our study, we have actually looked at plasma, and at this point we have looked at cells from vaccinated individuals. Our findings are essentially negative. We were not able to isolate vaccinia virus from both cells and plasma using actually a very well validated culture method that Dr. Srinivasan in our lab developed.

And what we have found is that plasma from vaccinees may not pose a significant risk for virus transmission. In fact, even the cells based on culture studies, seem to be lacking virus that's detectable by our technique.

In regard to availability, what we have observed is that with most donor screening assays, we did not see significant interference. We looked at 13 assays, and out of the 13 though with 4 we actually did see at certain time points post-vaccination, some false positivity. And we are now investigating that further to see in fact if they are real. So, that's where we stand with the vaccinia study.

So, in summary, our lab has been investigating the molecular epidemiology of HIV strains in a country with high viral diversity. We have actually expanded this collaboration to other areas where virus evolution is occurring, specifically in the Russian territories. This is through the Biotechnology Engagement Program. So, that project is going to be initiated in the next couple of months.

We are also looking at potentially setting something up in the southern part of the India where there seems to be some virus evolution occurring.

We are looking at the diagnostic impact of HIV variants. This is by acquiring these samples and testing all the licensed blood screening assays, as well as we hope in the future to look at the rapid tests that are used for diagnosis, because in fact those tests are used more in developing countries than here, and in those areas where we see higher viral diversity than we have seen in the US to date. So, this issue is actually more critical we think in countries outside the US. But it is an issue that needs to be looked at from the standpoint of AIDS pathogenesis and prevention.

So, we are continuing to look at virologic characteristics of drug-resistant variants. We are developing reference materials for HIV-2, as I mentioned. We are also looking at developing reference rate for CRF02_AB, which is the predominant viral subtype in some parts of the world. And we are pursuing some studies on vaccinia, specifically on identifying the -- actually to look for viremia in these vaccinated individuals by a Taqman assay.

In the area of new technology we have just been funded by the Office of Science to start some work in developing or to evaluate nanoparticle technology for its applicability for pathogen detection. We are starting by looking at HIV and potentially West Nile as the first phase of the evaluation of this technology.

So, I'll stop there. Thank you.

DR. ALLEN: Thank you, Dr. Hewlett.

Clarification questions from any of the committee members for Dr. Hewlett? All right, we'll move on to the presentation by Dr. Andrew Dayton.

Agenda Item: IV. Review of Site Visit Report for the Laboratory of Molecular Virology, DETTD - C. The Molecular Biology of HIV Infection of Primary Human Macrophages - Andrew Dayton, MD, PhD, Section Head, LMV, OBRR, FDA

DR. DAYTON: Good afternoon. I'm Andrew Dayton.

The work in my laboratory for many years has been -- at least since I have been at the FDA, has been focused largely on the molecular biology of HIV infection of primary human macrophages. As many of you know, the macrophage is a central cell in the body involved in the pathogenesis of HIV and the clinical course of AIDS. It is one of the first cell types to be infected during a transmission.

It is a conduit for bringing HIV across the blood-brain barrier into the brain. And it serves as a reservoir for virus throughout the clinical course of the disease. That's largely the reason why I have been focusing on it. Also, we can study it. It's a very interesting cell type in that you can have a lot of replication without a lot of cytopathic effect. So, you can look at what happens to these cells.

Now, most of the work comes from our key project, which has been identification and characterization of macrophage genetic pathways regulated by HIV infection and/or individual HIV genes. Mostly we have focused on the HIV Tat gene, which is known to affect a large number of cellular genes.

We have gotten into some corollary projects as a result of this. We have looked at genes that have seemed to be of interest, and they have taken us into related fields which are all relevant. In particular, we have been looking at the regulation of apoptosis in and also by macrophages. I'm going to summarize some of that data in a minute.

We have recently obtained a grant from the Office of Women's Health to study gender bias in HIV infection of common blood cell populations including macrophages. And also that grant covers the effects of antiretroviral agents in HIV infection of macrophages. I should say as a footnote here for mission relevance, one of the hats we wear is regulating HIV drug-resistant genotype assays.

We have also been looking at flavivirus replication in macrophages. We have been doing this in collaboration with other members of the laboratory, in particular Maria Rios and also Indira.

We have historically worked -- this recent collaboration with others in the lab has been on West Nile virus. We have worked on dengue in the past, and have reported dengue replicons, which may be useful for vaccine development. And we are also looking at co-infections of HIV and West Nile virus in macrophages. And in general, we are interested in the basic molecular biology of the HIV autoregulation.

So, the work on apoptosis derived from looking at genes which were regulated by HIV infection. And a very interesting finding we made is that HIV infection or Tat protein upregulates the expression of TRAIL which is TNF Related Apoptosis Inducing Ligand. It's a factor which is secreted and causes apoptosis in nearby cells.

And this is of tremendous interest, because TRAIL, which is produced by macrophages, could contribute to bystander mediated apoptosis. When we did this work it was a theoretical connection. Shortly therefore in the SKID(?) human-mouse model other investigators actually showed very convincingly that we were having a lot of apoptosis induction in that system by TRAIL and not by direct infection. So, we were very happy to see the groundbreaking work of ours turned out to have some relevance.

We also found that the HIV infection and Tat upregulate in macrophages at least, Bcl-2, which may contribute to protection of macrophages from apoptosis.

More recently, we have been looking at a very interesting finding that again came out of our microarray analysis actually, looking to see which genes were upregulated by HIV and Tat. And this is the upregulation of the IL-7 receptor alpha chain at the levels of both RNA and protein.

I'll show you data on this briefly, but what the data is going to say is that the upregulated IL-7 receptor is exposed on the exterior of the plasma membrane. It is functional, and this upregulation appears to be in a predominant population of macrophages. I won't actually show you the data for that today.

This is the key data. We won't go into all the details. But if you look at the protein data, uninfected cells have a very low level of IL-7 receptor. But if you treat with Tat protein or infect the cells, you get induction of the receptor at the level of protein. This is an inhibition of Tat by an antibody, just showing the specificity.

The same thing is seen at the level of RNA in these two portions of the gel. These are normalization controls. The critical data is right here. In uninfected cells you do have a background of RNA for the receptor. But when you infect with HIV, you get a big upregulation.

Similarly, you see the same thing with Tat, the use of normalization controls. You get a small amount of production if you don't add Tat to the cultures. But when you add Tat to the cultures, you get an induction at the level of RNA.

Then this data shows that the newly induct IL-7 receptor is functional. And it just shows that it phosphorylates or activates STAT-3. The question arose, okay, we are upregulating the receptor. But can it do anything? Maybe it's defective. But the answer is it does do something.

These are Western blots probed with antibodies to activate its STAT-3. And you can see that what these have done, these cells are grown. Then you get rid of all the cytokines, put them in serum-free medium. Then you blast them with the IL-7 cytokine to look for intracellular signaling. And when you blast uninfected cells with IL-7 for 20 minutes, you don't get any signaling. But if they have been HIV-infected, you do get signaling. These are controls.

Similarly, you see the same thing with Tat treatment. If there is no Tat treatment, you blast them with IL-7, there is no detectable signaling. If you blast them with IL-7 and they have been pre-treated with Tat, you get signaling. So, the upregulated IL-7 receptor was displayed on the surface and it's functional.

This is the data that ties this into a novel positive feedback loop. And all this says is if you dump IL-7 cytokine into these cultures, you increase the amount of HIV. And this is a dose response curve with increased amounts of IL-7. These are physiologic levels.

So, the basic observation is HIV upregulates the IL-7 receptor. And IL-7 stimulation through the IL-7 receptor upregulates HIV. So, you have a positive feedback loop. And just to lead you through that, this is what it looks like presumably. There is IL-7 cytokine floating around in the supernatant when you are in plasma.

When you infect with the HIV and starting HIV products, one of those is the Tat protein, which feedbacks back, and we believe it stimulates IL-7 receptor transcription. That gives you IL-7 receptors on the surface, which now signal from the IL-7 cytokine that is kicking around. And that signaling, we believe, goes back into the nucleus and regulates genes which influence HIV replication. It may be at the level of entry, we don't know.

So, to give you an idea of where we are going to take this, since the site visit we have actually been awarded a grant to pursue this work and others by the National Institutes of Health. It's an intramural H targeted antiviral grant. And part of that is to do mechanistic studies on HIV upregulation by the IL-7 system in these cell types. And part of it is to study the effects of Nef on primarily human monocytes and macrophages.

The grant is centered around the use of transfection in monocytes, which is a new technique that we have actually developed, and it's giving the potential to do a lot of experiments which haven't been able to be done before in macrophages. And that will enable us to do the mechanistic studies.

This is just my rogue's gallery of macrophages transfected with yellow fluorescent protein, which you can see is green. And you get all sorts of things. You nice stretched out fibroblastic macrophages. You get some classic fried egg morphology macrophages and small round ones. So, we are getting a nice transfection.

Just to remind you of the flavivirus biology that we have done. We have developed dengue virus replicons expressing exogenous genetic material, including HIV proteins. These are potentially useful for vaccine vectors. It also gives us a good background to go into some basic molecular biology of West Nile virus, if we can get outside funding for it.

And we have also been looking at replication of West Nile virus in macrophages with Indira and Maria.

Finally, I mentioned that we had a grant from the Office of Women's Health for gender bias in HIV replication in blood cells. This is looking at the effect of hormones and HIV replication, and also looking at the effect of hormones on specific drugs and their ability to inhibit HIV.

And that is pretty much it. If there are any questions, I would be glad to answer them.

DR. ALLEN: Thank you, Dr. Dayton.

Questions, clarification for Dr. Dayton from the committee?

DR. LEW: Would you venture just an hypothesis on why infection with macrophages does not always induce apoptosis as compared to the two tropic viruses that tend to infect and also kill?

DR. DAYTON: It's tremendously system-dependent. I really can't say anything other than that. You pick your experimental system and you work with it, and it just varies all over the place, so I really can't answer the question.

DR. DI MICHELE: Will you be looking at the role of Bcl-2 upregulation in these macrophages in some of the malignant complications of HIV?

DR. DAYTON: You mean how does it achieve the upregulation? Or what is the significance?

DR. DI MICHELE: The significance, and whether it actually has a role in the malignancy potential in these patients.

DR. DAYTON: We really haven't had the resources to pursue that. Of course, we would love to. Macrophages, as you probably know, are very resistant to apoptosis anyway, so there is some question about whether they really need the Bcl-2. But it was an interesting observation. The observations on TRAIL I think have much more significance than the observations on Bcl-2.

DR. ALLEN: Thank you, Dr. Dayton.

Our next presentation is by Dr. Subhash Dhawan.

Agenda Item: IV. Review of Site Visit Report for the Laboratory of Molecular Virology, DETTD - D. Viral and Host Factors in the Pathogenesis of HIV-1 Infection: An Overview - Subhash Dhawan, PhD, Section Head, LMV, OBRR, FDA

DR. DHAWAN: Well, good afternoon committee members and everyone present here. There is a lot to talk about, but the time today will permit me just to give you a flavor of what we have been doing in the past four years.

The title of my brief talk will be, "Viral and Host Factors in the Pathogenesis of HIV-1 Infection." Besides being involved heavily in the review of product applications and inspection of manufacturing facilities, my research interest has been in studying the following studies. We examined the role of HIV-Tat protein in viral pathogenesis, as Andy mentioned before. We worked to understand the involvement of host factors in HIV pathogenesis, and also developed strategies to enhance sensitivity of diagnostic assays.

These studies highly relate to our mission, because we heavily regulate vaccines in our center, and the Office of Blood primarily we regulate the diagnostic tools.

So, briefly, the Tat protein is one of the regulatory proteins of HIV-1. It's a very small protein, about 6 kilodalton. It is believed to be involved in the progression of HIV infection, the co-factor for Kaposi's sarcoma, and also is involved in the apoptosis of normal cells and in neurological disorders.

The evidence published in the literature shows that high levels of anti-Tat antibody relate to low viral load and seropositive non-progressors. So, the higher the level of anti-Tat in the body, the slower the disease progression. It's very well documented now.

So, these studies actually led us to focus our studies to develop a strategy to target the functional domains that are involved in the progression of infection by neutralizing them, and use them as a therapeutic approach, like a therapeutic vaccine. And I was just happy that I just shared with Dr. Allen the e-mail I saw.

Dr. Luke Montaya(?) recently endorsed how useful the therapeutic vaccine could be, as opposed to the preventive vaccine, which none out of 30 are showing any promising results. So, that gives me really good feedback to focus more on our studies, and confidence to believe what we are doing. I think that something could be meaningful.

So, Tat, as I mentioned, is a small protein. So, we mapped the entire Tat protein into multiple domains and figured out the two key domains representing 21-40 amino acid residue, which is cystine rich, and the other one is arginine rich domain, which ranges for 53-68 amino acids. They are key domains that made pathogenesis.

As you can see, there are multiple cytopathic effects. Notice when the cells were treated with these two synthetic peptides and then challenged with HIV, the results were quite similar to what we observed with the common and Tat proteins.

In a panel over here we demonstrated not only the recombinant Tat, but these same domains up here, they promoted the formation of new blood vessel formation in the CAM assay, like chicken ellen(?) toy(?) assay. So, these two domains represent a key domain and are the key target for neutralization in order to slow down the progression of infection.

But before we could use these domains in the form of a therapeutic vaccine, we must modify these domains, because they are pathogenic. If you don't modify and use them just as is, well they will induce immune response. Eventually you neutralize those domains. But before they could do that, they will cause more adverse effect, because they will potentially be given to infected individuals. And the cells that already are infected will produce more virus.

So, in order to do that, we modify. We may very selective changes in the side chain of cystine and arginine. And by this slight modification it completely blocked their pathogenic activity, and yet retained their immunological characteristics.

Since we last had a site visit, we raised antibody in mice, and we got very encouraging results. We used those antibodies and tested them in the infected assay, and we were able to demonstrate somewhere in the order of 40-70 percent inhibition of HIV application in monocyte by macrophages even in the absence of Tat proteins.

So, we are trying to do now is to use them in -- preferably there is a mouse model available. We would like to test it out, otherwise we are in collaboration with Dr. Chris Morty(?) in San Antonio, Texas. We are trying to test their immunological capability in rhesus.

And these are the results which I just said, in one slide. The modification that we made in the cystine site chain, and also in the arginine over here completely blocked the angiogenesis, as you see clearly in your handout.

So, a summary of the results are the cystine rich and arginine rich Tat sequences, they represent key domains in the HIV Tat protein that made pathogenesis. Synthetic Tat MPC now multiple MPCs from multiple peptide conjugates, what we did was we used these domains and made a synthetic construct, because this blood type will not be able to raise antibody as it is.

So, in order to increase the size, we used the synthetic backbone of lysine, and then put these domains on top of them, and they were able to induce effective immune response. So, antibody inhibited induced cytopathic effects. It's very interesting results.

And selective sites for modification I mentioned. They completely blocked the functional domain. And this modification retained large (?) characteristics, but completely blocked their pathogenic effects.

In another project which is quite related to the mission in the Office of Blood is develop tools to increase the sensitivity of an assay. It's very important to increase, have a diagnostic test that have high sensitivity. Typically in the EIA we use a secondary antibody reconjugated to enzyme, either HRP or alkaline phosphalyse or any other label.

But the limitation is the availability of the functional group, because these are conjugated through either amino or cystine group. And the antibody can have only X number of these functional groups. So, we now increase the binding of or conjugate more of these enzymes of the labels. I artificially created a number of primary means by using a lysine chain. In this for example we used 20.

So, for each amino group, I created 20 groups, and then conjugated HRP. And similarly by selectively cleaving the antibody for the sulfide reducing, this sulfide group, I conjugated two of those in order to increase the number of labels per antibody molecules.

And the results briefly were shown here, very encouraging. This is actually the Western blot showing the ability of these conjugates to detect even the bands that were not even visible in the conventional conjugates, we were able to detect those bands here. Now, by the way, this is a commercially available blot. We didn't create it in our laboratory. We just simply tested our conjugate that we prepared, the ones where we artificially increased the amount of label on each of the antibodies.

So, we couldn't pursue this due to the lack of funding and lack of resources. So, basically this work was done nearly two or three years ago, and basically I could not do anything more on this.

So, to summarize, the signal amplification conjugates, as just stated, increased the sensitivity of immunoassays. I plan to, given the funding situation, I plan to apply this to the same technique to develop a NAT. You can do the same thing in nucleic acids as well.

So, in the future plan we will try to immunize, depending upon the funding. We will use this Tat MPC. We will challenge them with SHIV 89.6, effectively infect the rhesus macaques and study the progression of disease. Also, for the diagnostic purposes we plan to test commercially available panels of normal serum, and whatever we require the manufacturer to do to demonstrate their efficacy of test kits by using multiple commercially available panels.

That's basically it. So, thank you very much.

DR. ALLEN: Thank you, Dr. Dhawan.

Comments or questions for Dr. Dhawan from the committee members. That's very exciting research.

DR. LEW: Using your genetically altered Tat, have you found that it can, instead of stimulating TRAIL production, it actually does not stimulate TRAIL production? Because if TRAIL is the cause of bystander cell death, that would be very, very interesting. And technically, the use of antibody to the Tat, potentially that could prevent again TRAIL upregulation and bystander cell death.

DR. DHAWAN: Oh, well, sure. This is what Andy talked about in his presentation. You're right.

DR. LEW: I didn't hear him say he was going to use those particular ones.

DR. DHAWAN: We don't know. What he has demonstrated is if he uses the antibody against Tat protein, it will neutralize Tat induced effect. And he very clearly demonstrated that. Now whether or not those domains are involved in the inhibition of TRAIL, we don't know. But we know is -- see, we are talking two different questions here. The one you are talking about, the upregulation of Tat protein that mediates other pathogenic effects primarily in the neurological disorders.

But what I am talking about is how it progresses the infection or the virus application in cells that are already infected. So, if we can block the domains that are responsible for promoting viral infection, even though the Tat may bind to the cell, because the binding site is different, blocking the functional domain that induces response in individuals' binding sites are or receptor sites, which we don't know for Tat, can block the Tat induced signaling in the infected cells.

Also, not only the infected cells, it can block the signaling in the uninfected cells, because Tat not only promotes the viral pathogenesis, it also activates normal cells and makes them more susceptible to HIV infection. It works in two ways. So, it works in endocrine(?) fashion, and it also works in a paracrine(?) fashion. Blocking this effect can block both the functions.

DR. LEW: That's why I find it interesting. Tat has many, many different functions, and that's clear. But I just think it's very interesting, we don't know how Tat works to induce TRAIL production. And just giving Tat we know can induce it without the HIV infection. You have the constructs. It might be worthwhile to just see its relationship to TRAIL.

DR. DHAWAN: Yes, that would be great. Let me clarify, the construct that I mentioned is not recombinant. It's totally synthetic. We made it in the laboratory using totally ad hoc synthetic chemistry.

DR. ALLEN: Thank you very much.

We'll move onto our last presentation for the morning, Dr. Maria Rios.

Agenda Item: IV. Review of Site Visit Report for the Laboratory of Molecular Virology, DETTD - E. West Nile Virus: Pathogenesis and Diagnostic Tools - Maria Rios, PhD, Senior Staff Fellow, LMV, OBRR, FDA

DR. RIOS: Good afternoon. What I have to talk about is the new program in the DETTD. And it is on West Nile. This program was established two years ago, and that is what we have been doing so far. I don't think I need to keep you up-to-date on what is going on, because we know that West Nile in the US has reoccurred for six consecutive years, which is unheard of in the West Nile field. And actually, from the first epidemics to the 2004, was going up and up and up, and we don't know where it's heading.

These six consecutive years of epidemics has accounted for over 16,000 human cases of disease, and 600 deaths reported to CDC. It is estimated that about over a million people have been infected with West Nile.

The transmission by blood transfusion was documented in 2002, and led to a series of actions mainly by the CDC and FDA, and jointly with the test kit manufacturers and the blot community to expedite development and implementation of screening assays for West Nile within eight months of the identification of the issue.

The research plans were to develop a standard reagent material well characterized to use with the purpose of regulatory issues in licensure of tests. We also engaged in studying genetic variation for the evolution of West Nile virus. So, why is that important? It's because the viral isolation characterization would allow us to investigate whether the tests are picking up any potential variant that may occur during this period of epidemics.

To study viral infectivity and pathogenesis, including the cellular tropism, and to perform infectivity in large animals, non-human primates that could simulate blood transfusion, address outstanding questions regarding blood transfusion issues.

So, we got the standard material in reagents. We have developed a prototype panel, and evaluated it in collaboration with other laboratories. The panel is composed of 14 members with a broad range of viral copy numbers. We used two different isolates. The first isolate was the flamingo NY99 that CDC had available at the time of test development.

But with our folks in FDA we isolated, completely characterized a human isolate from 2002. So, it is characterized genetically in the infectivity level to use for the licensure and lot release purposes. We distributed it to different laboratories, some of which were capable of performing quantitative assays to assign final copy. And these studies are under epidemiological investigation to find copy assignment.

As a result of these studies, we have seen that there is a great variability in performance when the low copy numbers panel, like 5 and 10 copies per mL are used. But in the qualitative assays we tried the ones used for blood screening perform much better than the quantitative assays. And that is where we stand regarding standard reagents right now.

To study the genetic evolution, I ready said why it's important to perform this study, because viral mutation may affect the performance of these nucleic tests. And it may also affect pathogenesis, and have an impact in vaccine development. So, it's fully aligned with the CBER main mission, not only of the OBRI, but the CBER main mission.

And we engaged in collecting samples and isolating virus for full sequention genetic analysis. And because of the limitation of material, we investigated whether if we isolate the virus from something like viracells, whether it would induce any change in the material that would interfere with what we were intending to do with the panels.

And we did investigate three pathogens, and we have seen no variation in the structure of the genes compared with the countered virus to the original virus in the plasma. This was done in one or two specimens that we had in large volume to sequence this straight from the plasma.

Right now we have sequenced all the 25 isolated, the fully structured gene. And two isolates are not fully sequenced. Now, we have three I just learned from the post-doc that is helping me with this work, that we have two human genome West Nile isolates from human plasma fully characterized, and we are planning on using these to evaluate this viability.

What are we planning to go on with? To continue this study for full genetic sequence, and try to correlate that with pathogenesis. We have a repository with the help from the blood centers, mainly Red Cross and Blood Systems Laboratory. We plan on going on with these sequencing studies to expand these studies to obtain samples from patients that had West Nile, not blood donors, because our studies have been biased so far, because all these samples that we have isolated virus and sequenced isolates based on the available screening assays.

So, what we want to do is to get non-biased samples which allow us to investigate where the isolates that we identify in these samples could not be missed by the current screening assays.

So, we are using microarray technology to expedite genotyping determination, and to identify potential point mutations. We have developed a fully structured gene microarray already, and we are planning on going on and including 11,000 base pairs of virus in the microarray for quite detection of variants. And these obviously will help us to evaluate the assays and their performance.

In terms of infectivity and pathogenesis, the outstanding question at that time was the cellular tropism for West Nile infection was unknown following the blood transfusion. We know that when the mosquito bites, the way that the virus enters the human body is through the resident monocytes of dendritic cells and so on. But we did not know which would be the cell that will be capable in circulating blood to attach the virus harboring initiative replication.

So, we engaged, based on the similarities that the resident macrophage would be the target for primary infection to study monocyte derived macrophage in collaboration with Dr. Dayton that had the system in place, to ask the question of whether that would be a target for West Nile replication.

We also engaged in looking at other human blood cells, peripheral blood mononuclear cells in general, T-cells and B-cells, and to investigate the host factors involved in pathogenesis. That includes chemokines and cytokines in human host factors. And these would be very relevant for CBER for treatment and analysis and vaccines and so on.

What we have achieved so far regarding the viral tropism is that now I have data that shows that early stage culture monocyte -- not only macrophage-derived monocytes, but early culture monocytes can be infected by West Nile. In tests in the supernatant indicates productive infection. The viral loads go as high as 10 to the 6th in the supernatant.

The value of the application persists for 27 days in 9 experiments, 9 of the 10 experiments we have performed. And actually, we have not one experiment of up to 47 days, but we have three experiments that we followed the tissue culture for 47 days. And it is still producing virus, and these virus are infectious intraviral cells.

So, we have shown that macrophage can be infected, and produce virus which are infectious. And what is interesting about this culture is that there is no morphological alteration in the monocyte during the course of infection. So, it seems that the cell does not get harmed with the viral infection, at least in a gross way of looking at it.

So, these data suggest that human monocytes in macrophage could play an initial role in replication following blood transfusion, and that would be important obviously to understand.

The last thing we are planning to address, and we have been engaged heavily in trying to gather funding, as you all heard about funding, is to perform the infectivity studies in large non-human primates. And are planning on doing it in baboons. And why is that? Because the baboons can get a large volume of plasma that would simulate the human transfusion, as opposed to small animals that would not have the capacity of taking high intake of plasma that has low viral load.

The major focus here is to look at units that have not positive IgN positive, meaning low viral load in the presence of antibody, so that we could define first of all, the minimal infectious dose with non-human primates. And then address the infectivity of human plasma in the presence and absence of antibody, which is a very critical question to address now.

We have no information. We have information that the three previous stages that defined not positive Ig negative were high tider of virus are infectious, but not the two of the IgM and IgG in low viral tider.

And I would stop here and just acknowledge the group that has been working. Dr. Hewlett's group has fully supported me in these short two years, and that's why we have achieved so much.

Thank you.

DR. ALLEN: Thank you, Dr. Rios.

I would just like to point out that when Dr. Rios came to the FDA, less than three years ago, her background was in red cell molecular biology. And I think she had intended to continue pursuing investigations in that area. West Nile virus had erupted. The FDA needed to get studies underway, and Dr. Rios volunteered. So, you made a remarkable transformation.

DR. RIOS: And actually, when I was hired it was to work with the retroviral HGIV and HIV, which was my background field that I have addressed in red cells, because I have been very interested.

One thing I didn't mention was we are looking at blood partition, and we have found that red cells carry almost a log more compared to the paired plasma. So, maybe we are not looking at the proper specimen for viral screening. And that's one of the things are addressing now, is to look what is the idea sample for screening, whether it's whole blood or plasma.

DR. ALLEN: Okay, any other questions quickly?

DR. DI MICHELE: Maria, I wanted to congratulate you and everyone else for all of the excellent work that they have presented. Just out of curiosity, your laboratory is doing this work on West Nile. What percentage of all the work on West Nile does this laboratory represent? Is there a lot of other work on West Nile virus?

DR. ALLEN: You mean at the FDA?

DR. DI MICHELE: No, elsewhere, just in general.

DR. RIOS: There are laboratories. CDC is doing a lot of work, and the blood centers are doing a lot of work. The group Mike Bush and Sue Stream(?) actually the infectivity studies which we are having a hard time in getting funded will be fully supported by the blood community providing samples from which we would isolate and characterize the virus, and try to correlate the non-human primates' response and things that we cannot address in human with that.

And then the state department of health and some other laboratories more focused on treatment have been addressing that.

DR. DI MICHELE: No, I understand the US collaborations. But elsewhere in the world, are there other laboratories that focus on this?

DR. RIOS: Russian. And actually, we are engaged in trying to get money from bioterrorism. If you are not aware of NIEAT(?) just listed West Nile as category B agent, but it can be easily manipulated in the lab and spread. And there is a program that helps us, to link us, the American government to the Russian government, mainly scientists that have been working in bioweapons to develop some of these studies. So, Russian has been working on it.

Israel has some work on it. Australia has done a lot of work on Kungi(?) virus, that is the West Nile equivalent for them. And West Nile has spread now to Mexico, Puerto Rico, and it is going downwards. I would imagine that because of bird migration, South America soon will be hit. And we know from Russian data that the birds migrate from Africa to Siberia in Russian, and you have episodes of epidemics of West Nile even in Siberia, where the climate is cold. So, we need more.

DR. ALLEN: Thank you.

DR. NAKHASI: Donna, I just wanted to make a clarification here. I think the research that Maria Rios presented today is really a critical path issue, because the issues for example, panel development, the areas of looking at the genetic variation, that type of research is not done outside. And that's the reason she is focusing on that. And also the pathogenesis issue, because it is not the issue of looking at pathogenesis, but it is the issue of looking at which component of the blood replicates, because it's important for us in blood products.

So, I think it's very important to concentrate on that whether any other labs or doing it or not. But these type of areas in which she is focusing is more of a critical path issue, which is relevant to what we are doing.

DR. RIOS: Thank you, Hira. Actually, if I had understood this question in the way now, there are very few studies in genetic variation which focus on a very small fragment of the envelope region, about 200 base pairs. But we know now that the NS5, that it's a non-instructor region. It's critical to differentiate West Nile from Japanese encephalitis. So, no, there is not much focus on that outside.

DR. KATZ: Dr. Rios, can you describe how you intend to get the unbiased sampling? Are you going to use mosquito pools or birds or clinical cases outside the blood screening, or all of the above?

DR. RIOS: I have been contacting actual just the last West Nile conference in San Jose, I have contacted some health department people. And I get some promise, not real samples. But what I would like to do is to get samples from health department laboratories that come from confirmation of disease that are not associated with the blood, and see if there is any extra variation for those.

As you all know, West Nile virus is a very unique virus in the sense that 80 percent of the infections are fully symptomatic, but 1 percent is highly fatal. It is people that develop neurological symptoms. Between 10 and 14 percent of people died. And in the last West Nile conference it was reported by the State Department that West Nile death, it's not accounted for. And these slides will be posted on the CDC Web site. Because the physicians do not follow when a donor dies, whether or not West Nile was a potential case.

DR. ALLEN: Okay, we didn't hear a discussion about the FDA flamingo farm.

Thank you all very much. Thank you, Dr. Rios. It's almost 1:00 pm. We are going to very briefly into closed session, which means committee members and senior FDA staff only please.

[Whereupon, the open session of the meeting was adjourned at 5:00 pm.]